From bc0fcb9823cd0894934cf968b525c575833d7078 Mon Sep 17 00:00:00 2001 From: Yilin Zhu Date: Sun, 12 Apr 2026 13:07:54 +0800 Subject: ipv6: xfrm6: release dst on error in xfrm6_rcv_encap() xfrm6_rcv_encap() performs an IPv6 route lookup when the skb does not already have a dst attached. ip6_route_input_lookup() returns a referenced dst entry even when the lookup resolves to an error route. If dst->error is set, xfrm6_rcv_encap() drops the skb without attaching the dst to the skb and without releasing the reference returned by the lookup. Repeated packets hitting this path therefore leak dst entries. Release the dst before jumping to the drop path. Fixes: 0146dca70b87 ("xfrm: add support for UDPv6 encapsulation of ESP") Cc: stable@kernel.org Reported-by: Yifan Wu Reported-by: Juefei Pu Co-developed-by: Yuan Tan Signed-off-by: Yuan Tan Suggested-by: Xin Liu Tested-by: Ruide Cao Signed-off-by: Yilin Zhu Signed-off-by: Ren Wei Reviewed-by: Simon Horman Signed-off-by: Steffen Klassert --- net/ipv6/xfrm6_protocol.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv6/xfrm6_protocol.c b/net/ipv6/xfrm6_protocol.c index ea2f805d3b01..9b586fcec485 100644 --- a/net/ipv6/xfrm6_protocol.c +++ b/net/ipv6/xfrm6_protocol.c @@ -88,8 +88,10 @@ int xfrm6_rcv_encap(struct sk_buff *skb, int nexthdr, __be32 spi, dst = ip6_route_input_lookup(dev_net(skb->dev), skb->dev, &fl6, skb, flags); - if (dst->error) + if (dst->error) { + dst_release(dst); goto drop; + } skb_dst_set(skb, dst); } -- cgit v1.2.3 From ec54093e6a8f87e800bb6aa15eb7fc1e33faa524 Mon Sep 17 00:00:00 2001 From: Michael Bommarito Date: Sun, 19 Apr 2026 18:35:42 -0400 Subject: xfrm: ah: account for ESN high bits in async callbacks AH allocates its temporary auth/ICV layout differently when ESN is enabled: the async ahash setup appends a 4-byte seqhi slot before the ICV or auth_data area, but the async completion callbacks still reconstruct the temporary layout as if seqhi were absent. With an async AH implementation selected, that makes AH copy or compare the wrong bytes on both the IPv4 and IPv6 paths. In UML repro on IPv4 AH with ESN and forced async hmac(sha1), ping fails with 100% packet loss, and the callback logs show the pre-fix drift: ah4 output_done: esn=1 err=0 icv_off=20 expected_off=24 ah4 input_done: esn=1 auth_off=20 expected_auth_off=24 icv_off=32 expected_icv_off=36 Reconstruct the callback-side layout the same way the setup path built it by skipping the ESN seqhi slot before locating the saved auth_data or ICV. Per RFC 4302, the ESN high-order 32 bits participate in the AH ICV computation, so the async callbacks must account for the seqhi slot. Post-fix, the same IPv4 AH+ESN+forced-async-hmac(sha1) UML repro shows the corrected offset (ah4 output_done: esn=1 err=0 icv_off=24 expected_off=24) and ping succeeds; net/ipv4/ah4.o and net/ipv6/ah6.o build clean at W=1. IPv6 AH+ESN was not exercised at runtime, and the change has not been tested against a real async hardware AH engine. Fixes: d4d573d0334d ("{IPv4,xfrm} Add ESN support for AH egress part") Fixes: d8b2a8600b0e ("{IPv4,xfrm} Add ESN support for AH ingress part") Fixes: 26dd70c3fad3 ("{IPv6,xfrm} Add ESN support for AH egress part") Fixes: 8d6da6f32557 ("{IPv6,xfrm} Add ESN support for AH ingress part") Cc: stable@vger.kernel.org Assisted-by: Codex:gpt-5-4 Assisted-by: Claude:claude-opus-4-7 Signed-off-by: Michael Bommarito Signed-off-by: Steffen Klassert --- net/ipv4/ah4.c | 14 ++++++++++++-- net/ipv6/ah6.c | 14 ++++++++++++-- 2 files changed, 24 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/ipv4/ah4.c b/net/ipv4/ah4.c index 5fb812443a08..4366cbac3f06 100644 --- a/net/ipv4/ah4.c +++ b/net/ipv4/ah4.c @@ -124,9 +124,14 @@ static void ah_output_done(void *data, int err) struct iphdr *top_iph = ip_hdr(skb); struct ip_auth_hdr *ah = ip_auth_hdr(skb); int ihl = ip_hdrlen(skb); + int seqhi_len = 0; + __be32 *seqhi; + if (x->props.flags & XFRM_STATE_ESN) + seqhi_len = sizeof(*seqhi); iph = AH_SKB_CB(skb)->tmp; - icv = ah_tmp_icv(iph, ihl); + seqhi = (__be32 *)((char *)iph + ihl); + icv = ah_tmp_icv(seqhi, seqhi_len); memcpy(ah->auth_data, icv, ahp->icv_trunc_len); top_iph->tos = iph->tos; @@ -270,12 +275,17 @@ static void ah_input_done(void *data, int err) struct ip_auth_hdr *ah = ip_auth_hdr(skb); int ihl = ip_hdrlen(skb); int ah_hlen = (ah->hdrlen + 2) << 2; + int seqhi_len = 0; + __be32 *seqhi; if (err) goto out; + if (x->props.flags & XFRM_STATE_ESN) + seqhi_len = sizeof(*seqhi); work_iph = AH_SKB_CB(skb)->tmp; - auth_data = ah_tmp_auth(work_iph, ihl); + seqhi = (__be32 *)((char *)work_iph + ihl); + auth_data = ah_tmp_auth(seqhi, seqhi_len); icv = ah_tmp_icv(auth_data, ahp->icv_trunc_len); err = crypto_memneq(icv, auth_data, ahp->icv_trunc_len) ? -EBADMSG : 0; diff --git a/net/ipv6/ah6.c b/net/ipv6/ah6.c index cb26beea4398..de1e68199a01 100644 --- a/net/ipv6/ah6.c +++ b/net/ipv6/ah6.c @@ -317,14 +317,19 @@ static void ah6_output_done(void *data, int err) struct ipv6hdr *top_iph = ipv6_hdr(skb); struct ip_auth_hdr *ah = ip_auth_hdr(skb); struct tmp_ext *iph_ext; + int seqhi_len = 0; + __be32 *seqhi; extlen = skb_network_header_len(skb) - sizeof(struct ipv6hdr); if (extlen) extlen += sizeof(*iph_ext); + if (x->props.flags & XFRM_STATE_ESN) + seqhi_len = sizeof(*seqhi); iph_base = AH_SKB_CB(skb)->tmp; iph_ext = ah_tmp_ext(iph_base); - icv = ah_tmp_icv(iph_ext, extlen); + seqhi = (__be32 *)((char *)iph_ext + extlen); + icv = ah_tmp_icv(seqhi, seqhi_len); memcpy(ah->auth_data, icv, ahp->icv_trunc_len); memcpy(top_iph, iph_base, IPV6HDR_BASELEN); @@ -471,13 +476,18 @@ static void ah6_input_done(void *data, int err) struct ip_auth_hdr *ah = ip_auth_hdr(skb); int hdr_len = skb_network_header_len(skb); int ah_hlen = ipv6_authlen(ah); + int seqhi_len = 0; + __be32 *seqhi; if (err) goto out; + if (x->props.flags & XFRM_STATE_ESN) + seqhi_len = sizeof(*seqhi); work_iph = AH_SKB_CB(skb)->tmp; auth_data = ah_tmp_auth(work_iph, hdr_len); - icv = ah_tmp_icv(auth_data, ahp->icv_trunc_len); + seqhi = (__be32 *)(auth_data + ahp->icv_trunc_len); + icv = ah_tmp_icv(seqhi, seqhi_len); err = crypto_memneq(icv, auth_data, ahp->icv_trunc_len) ? -EBADMSG : 0; if (err) -- cgit v1.2.3 From 1e8e3f449b1e73b73a843257635b9c50f0cc0f0a Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Mon, 20 Apr 2026 23:15:32 +0200 Subject: netfilter: arp_tables: fix IEEE1394 ARP payload parsing Weiming Shi says: "arp_packet_match() unconditionally parses the ARP payload assuming two hardware addresses are present (source and target). However, IPv4-over-IEEE1394 ARP (RFC 2734) omits the target hardware address field, and arp_hdr_len() already accounts for this by returning a shorter length for ARPHRD_IEEE1394 devices. As a result, on IEEE1394 interfaces arp_packet_match() advances past a nonexistent target hardware address and reads the wrong bytes for both the target device address comparison and the target IP address. This causes arptables rules to match against garbage data, leading to incorrect filtering decisions: packets that should be accepted may be dropped and vice versa. The ARP stack in net/ipv4/arp.c (arp_create and arp_process) already handles this correctly by skipping the target hardware address for ARPHRD_IEEE1394. Apply the same pattern to arp_packet_match()." Mangle the original patch to always return 0 (no match) in case user matches on the target hardware address which is never present in IEEE1394. Note that this returns 0 (no match) for either normal and inverse match because matching in the target hardware address in ARPHRD_IEEE1394 has never been supported by arptables. This is intentional, matching on the target hardware address should never evaluate true for ARPHRD_IEEE1394. Moreover, adjust arpt_mangle to drop the packet too as AI suggests: In arpt_mangle, the logic assumes a standard ARP layout. Because IEEE1394 (FireWire) omits the target hardware address, the linear pointer arithmetic miscalculates the offset for the target IP address. This causes mangling operations to write to the wrong location, leading to packet corruption. To ensure safety, this patch drops packets (NF_DROP) when mangling is requested for these fields on IEEE1394 devices, as the current implementation cannot correctly map the FireWire ARP payload. This omits both mangling target hardware and IP address. Even if IP address mangling should be possible in IEEE1394, this would require to adjust arpt_mangle offset calculation, which has never been supported. Based on patch from Weiming Shi . Fixes: 6752c8db8e0c ("firewire net, ipv4 arp: Extend hardware address and remove driver-level packet inspection.") Reported-by: Xiang Mei Signed-off-by: Pablo Neira Ayuso --- net/ipv4/netfilter/arp_tables.c | 18 +++++++++++++++--- net/ipv4/netfilter/arpt_mangle.c | 8 ++++++++ 2 files changed, 23 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c index 1cdd9c28ab2d..97ead883e4a1 100644 --- a/net/ipv4/netfilter/arp_tables.c +++ b/net/ipv4/netfilter/arp_tables.c @@ -110,13 +110,25 @@ static inline int arp_packet_match(const struct arphdr *arphdr, arpptr += dev->addr_len; memcpy(&src_ipaddr, arpptr, sizeof(u32)); arpptr += sizeof(u32); - tgt_devaddr = arpptr; - arpptr += dev->addr_len; + + if (IS_ENABLED(CONFIG_FIREWIRE_NET) && dev->type == ARPHRD_IEEE1394) { + if (unlikely(memchr_inv(arpinfo->tgt_devaddr.mask, 0, + sizeof(arpinfo->tgt_devaddr.mask)))) + return 0; + + tgt_devaddr = NULL; + } else { + tgt_devaddr = arpptr; + arpptr += dev->addr_len; + } memcpy(&tgt_ipaddr, arpptr, sizeof(u32)); if (NF_INVF(arpinfo, ARPT_INV_SRCDEVADDR, arp_devaddr_compare(&arpinfo->src_devaddr, src_devaddr, - dev->addr_len)) || + dev->addr_len))) + return 0; + + if (tgt_devaddr && NF_INVF(arpinfo, ARPT_INV_TGTDEVADDR, arp_devaddr_compare(&arpinfo->tgt_devaddr, tgt_devaddr, dev->addr_len))) diff --git a/net/ipv4/netfilter/arpt_mangle.c b/net/ipv4/netfilter/arpt_mangle.c index a4e07e5e9c11..f65dd339208e 100644 --- a/net/ipv4/netfilter/arpt_mangle.c +++ b/net/ipv4/netfilter/arpt_mangle.c @@ -40,6 +40,10 @@ target(struct sk_buff *skb, const struct xt_action_param *par) } arpptr += pln; if (mangle->flags & ARPT_MANGLE_TDEV) { + if (unlikely(IS_ENABLED(CONFIG_FIREWIRE_NET) && + skb->dev->type == ARPHRD_IEEE1394)) + return NF_DROP; + if (ARPT_DEV_ADDR_LEN_MAX < hln || (arpptr + hln > skb_tail_pointer(skb))) return NF_DROP; @@ -47,6 +51,10 @@ target(struct sk_buff *skb, const struct xt_action_param *par) } arpptr += hln; if (mangle->flags & ARPT_MANGLE_TIP) { + if (unlikely(IS_ENABLED(CONFIG_FIREWIRE_NET) && + skb->dev->type == ARPHRD_IEEE1394)) + return NF_DROP; + if (ARPT_MANGLE_ADDR_LEN_MAX < pln || (arpptr + pln > skb_tail_pointer(skb))) return NF_DROP; -- cgit v1.2.3 From f3224ee463f8f6f6ced7dcdf6081add4f8128527 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 16 Apr 2026 15:14:51 +0200 Subject: netfilter: nf_tables: use list_del_rcu for netlink hooks nft_netdev_unregister_hooks and __nft_unregister_flowtable_net_hooks need to use list_del_rcu(), this list can be walked by concurrent dumpers. Add a new helper and use it consistently. Fixes: f9a43007d3f7 ("netfilter: nf_tables: double hook unregistration in netns path") Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_tables_api.c | 44 ++++++++++++++++++------------------------- 1 file changed, 18 insertions(+), 26 deletions(-) (limited to 'net') diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index 8537b94653d3..07e151245765 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -374,6 +374,12 @@ static void nft_netdev_hook_free_rcu(struct nft_hook *hook) call_rcu(&hook->rcu, __nft_netdev_hook_free_rcu); } +static void nft_netdev_hook_unlink_free_rcu(struct nft_hook *hook) +{ + list_del_rcu(&hook->list); + nft_netdev_hook_free_rcu(hook); +} + static void nft_netdev_unregister_hooks(struct net *net, struct list_head *hook_list, bool release_netdev) @@ -384,10 +390,8 @@ static void nft_netdev_unregister_hooks(struct net *net, list_for_each_entry_safe(hook, next, hook_list, list) { list_for_each_entry(ops, &hook->ops_list, list) nf_unregister_net_hook(net, ops); - if (release_netdev) { - list_del(&hook->list); - nft_netdev_hook_free_rcu(hook); - } + if (release_netdev) + nft_netdev_hook_unlink_free_rcu(hook); } } @@ -2271,10 +2275,8 @@ void nf_tables_chain_destroy(struct nft_chain *chain) if (nft_base_chain_netdev(table->family, basechain->ops.hooknum)) { list_for_each_entry_safe(hook, next, - &basechain->hook_list, list) { - list_del_rcu(&hook->list); - nft_netdev_hook_free_rcu(hook); - } + &basechain->hook_list, list) + nft_netdev_hook_unlink_free_rcu(hook); } module_put(basechain->type->owner); if (rcu_access_pointer(basechain->stats)) { @@ -2974,6 +2976,7 @@ err_hooks: list_for_each_entry(ops, &h->ops_list, list) nf_unregister_net_hook(ctx->net, ops); } + /* hook.list is on stack, no need for list_del_rcu() */ list_del(&h->list); nft_netdev_hook_free_rcu(h); } @@ -8852,10 +8855,8 @@ static void __nft_unregister_flowtable_net_hooks(struct net *net, list_for_each_entry_safe(hook, next, hook_list, list) { list_for_each_entry(ops, &hook->ops_list, list) nft_unregister_flowtable_ops(net, flowtable, ops); - if (release_netdev) { - list_del(&hook->list); - nft_netdev_hook_free_rcu(hook); - } + if (release_netdev) + nft_netdev_hook_unlink_free_rcu(hook); } } @@ -8926,8 +8927,7 @@ err_unregister_net_hooks: nft_unregister_flowtable_ops(net, flowtable, ops); } - list_del_rcu(&hook->list); - nft_netdev_hook_free_rcu(hook); + nft_netdev_hook_unlink_free_rcu(hook); } return err; @@ -8937,10 +8937,8 @@ static void nft_hooks_destroy(struct list_head *hook_list) { struct nft_hook *hook, *next; - list_for_each_entry_safe(hook, next, hook_list, list) { - list_del_rcu(&hook->list); - nft_netdev_hook_free_rcu(hook); - } + list_for_each_entry_safe(hook, next, hook_list, list) + nft_netdev_hook_unlink_free_rcu(hook); } static int nft_flowtable_update(struct nft_ctx *ctx, const struct nlmsghdr *nlh, @@ -9028,8 +9026,7 @@ err_flowtable_update_hook: nft_unregister_flowtable_ops(ctx->net, flowtable, ops); } - list_del_rcu(&hook->list); - nft_netdev_hook_free_rcu(hook); + nft_netdev_hook_unlink_free_rcu(hook); } return err; @@ -9535,13 +9532,8 @@ err: static void nf_tables_flowtable_destroy(struct nft_flowtable *flowtable) { - struct nft_hook *hook, *next; - flowtable->data.type->free(&flowtable->data); - list_for_each_entry_safe(hook, next, &flowtable->hook_list, list) { - list_del_rcu(&hook->list); - nft_netdev_hook_free_rcu(hook); - } + nft_hooks_destroy(&flowtable->hook_list); kfree(flowtable->name); module_put(flowtable->data.type->owner); kfree(flowtable); -- cgit v1.2.3 From a6134e62dba2ea4f760b29d5226907f447c92400 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Wed, 15 Apr 2026 17:56:14 +0200 Subject: netfilter: nf_tables: join hook list via splice_list_rcu() in commit phase Publish new hooks in the list into the basechain/flowtable using splice_list_rcu() to ensure netlink dump list traversal via rcu is safe while concurrent ruleset update is going on. Fixes: 78d9f48f7f44 ("netfilter: nf_tables: add devices to existing flowtable") Fixes: b9703ed44ffb ("netfilter: nf_tables: support for adding new devices to an existing netdev chain") Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_tables_api.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index 07e151245765..ae10116af923 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -10838,8 +10838,8 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb) nft_chain_commit_update(nft_trans_container_chain(trans)); nf_tables_chain_notify(&ctx, NFT_MSG_NEWCHAIN, &nft_trans_chain_hooks(trans)); - list_splice(&nft_trans_chain_hooks(trans), - &nft_trans_basechain(trans)->hook_list); + list_splice_rcu(&nft_trans_chain_hooks(trans), + &nft_trans_basechain(trans)->hook_list); /* trans destroyed after rcu grace period */ } else { nft_chain_commit_drop_policy(nft_trans_container_chain(trans)); @@ -10968,8 +10968,8 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb) nft_trans_flowtable(trans), &nft_trans_flowtable_hooks(trans), NFT_MSG_NEWFLOWTABLE); - list_splice(&nft_trans_flowtable_hooks(trans), - &nft_trans_flowtable(trans)->hook_list); + list_splice_rcu(&nft_trans_flowtable_hooks(trans), + &nft_trans_flowtable(trans)->hook_list); } else { nft_clear(net, nft_trans_flowtable(trans)); nf_tables_flowtable_notify(&ctx, -- cgit v1.2.3 From 10f79dbd7719d1da9f5884d13060322d8729f091 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Wed, 15 Apr 2026 22:58:23 +0200 Subject: netfilter: nf_tables: add hook transactions for device deletions Restore the flag that indicates that the hook is going away, ie. NFT_HOOK_REMOVE, but add a new transaction object to track deletion of hooks without altering the basechain/flowtable hook_list during the preparation phase. The existing approach that moves the hook from the basechain/flowtable hook_list to transaction hook_list breaks netlink dump path readers of this RCU-protected list. It should be possible use an array for nft_trans_hook to store the deleted hooks to compact the representation but I am not expecting many hook object, specially now that wildcard support for devices is in place. Note that the nft_trans_chain_hooks() list contains a list of struct nft_trans_hook objects for DELCHAIN and DELFLOWTABLE commands, while this list stores struct nft_hook objects for NEWCHAIN and NEWFLOWTABLE. Note that new commands can be updated to use nft_trans_hook for consistency. This patch also adapts the event notification path to deal with the list of hook transactions. Fixes: 7d937b107108 ("netfilter: nf_tables: support for deleting devices in an existing netdev chain") Fixes: b6d9014a3335 ("netfilter: nf_tables: delete flowtable hooks via transaction list") Reported-by: Xiang Mei Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_tables_api.c | 264 ++++++++++++++++++++++++++++++++---------- 1 file changed, 204 insertions(+), 60 deletions(-) (limited to 'net') diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index ae10116af923..d20ce5c36d31 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -380,6 +380,32 @@ static void nft_netdev_hook_unlink_free_rcu(struct nft_hook *hook) nft_netdev_hook_free_rcu(hook); } +static void nft_trans_hook_destroy(struct nft_trans_hook *trans_hook) +{ + list_del(&trans_hook->list); + kfree(trans_hook); +} + +static void nft_netdev_unregister_trans_hook(struct net *net, + const struct nft_table *table, + struct list_head *hook_list) +{ + struct nft_trans_hook *trans_hook, *next; + struct nf_hook_ops *ops; + struct nft_hook *hook; + + list_for_each_entry_safe(trans_hook, next, hook_list, list) { + hook = trans_hook->hook; + + if (!(table->flags & NFT_TABLE_F_DORMANT)) { + list_for_each_entry(ops, &hook->ops_list, list) + nf_unregister_net_hook(net, ops); + } + nft_netdev_hook_unlink_free_rcu(hook); + nft_trans_hook_destroy(trans_hook); + } +} + static void nft_netdev_unregister_hooks(struct net *net, struct list_head *hook_list, bool release_netdev) @@ -1946,15 +1972,69 @@ static int nft_nla_put_hook_dev(struct sk_buff *skb, struct nft_hook *hook) return nla_put_string(skb, attr, hook->ifname); } +struct nft_hook_dump_ctx { + struct nft_hook *first; + int n; +}; + +static int nft_dump_basechain_hook_one(struct sk_buff *skb, + struct nft_hook *hook, + struct nft_hook_dump_ctx *dump_ctx) +{ + if (!dump_ctx->first) + dump_ctx->first = hook; + + if (nft_nla_put_hook_dev(skb, hook)) + return -1; + + dump_ctx->n++; + + return 0; +} + +static int nft_dump_basechain_hook_list(struct sk_buff *skb, + const struct net *net, + const struct list_head *hook_list, + struct nft_hook_dump_ctx *dump_ctx) +{ + struct nft_hook *hook; + int err; + + list_for_each_entry_rcu(hook, hook_list, list, + lockdep_commit_lock_is_held(net)) { + err = nft_dump_basechain_hook_one(skb, hook, dump_ctx); + if (err < 0) + return err; + } + + return 0; +} + +static int nft_dump_basechain_trans_hook_list(struct sk_buff *skb, + const struct list_head *trans_hook_list, + struct nft_hook_dump_ctx *dump_ctx) +{ + struct nft_trans_hook *trans_hook; + int err; + + list_for_each_entry(trans_hook, trans_hook_list, list) { + err = nft_dump_basechain_hook_one(skb, trans_hook->hook, dump_ctx); + if (err < 0) + return err; + } + + return 0; +} + static int nft_dump_basechain_hook(struct sk_buff *skb, const struct net *net, int family, const struct nft_base_chain *basechain, - const struct list_head *hook_list) + const struct list_head *hook_list, + const struct list_head *trans_hook_list) { const struct nf_hook_ops *ops = &basechain->ops; - struct nft_hook *hook, *first = NULL; + struct nft_hook_dump_ctx dump_hook_ctx = {}; struct nlattr *nest, *nest_devs; - int n = 0; nest = nla_nest_start_noflag(skb, NFTA_CHAIN_HOOK); if (nest == NULL) @@ -1969,23 +2049,23 @@ static int nft_dump_basechain_hook(struct sk_buff *skb, if (!nest_devs) goto nla_put_failure; - if (!hook_list) + if (!hook_list && !trans_hook_list) hook_list = &basechain->hook_list; - list_for_each_entry_rcu(hook, hook_list, list, - lockdep_commit_lock_is_held(net)) { - if (!first) - first = hook; - - if (nft_nla_put_hook_dev(skb, hook)) - goto nla_put_failure; - n++; + if (hook_list && + nft_dump_basechain_hook_list(skb, net, hook_list, &dump_hook_ctx)) { + goto nla_put_failure; + } else if (trans_hook_list && + nft_dump_basechain_trans_hook_list(skb, trans_hook_list, + &dump_hook_ctx)) { + goto nla_put_failure; } + nla_nest_end(skb, nest_devs); - if (n == 1 && - !hook_is_prefix(first) && - nla_put_string(skb, NFTA_HOOK_DEV, first->ifname)) + if (dump_hook_ctx.n == 1 && + !hook_is_prefix(dump_hook_ctx.first) && + nla_put_string(skb, NFTA_HOOK_DEV, dump_hook_ctx.first->ifname)) goto nla_put_failure; } nla_nest_end(skb, nest); @@ -1999,7 +2079,8 @@ static int nf_tables_fill_chain_info(struct sk_buff *skb, struct net *net, u32 portid, u32 seq, int event, u32 flags, int family, const struct nft_table *table, const struct nft_chain *chain, - const struct list_head *hook_list) + const struct list_head *hook_list, + const struct list_head *trans_hook_list) { struct nlmsghdr *nlh; @@ -2015,7 +2096,7 @@ static int nf_tables_fill_chain_info(struct sk_buff *skb, struct net *net, NFTA_CHAIN_PAD)) goto nla_put_failure; - if (!hook_list && + if (!hook_list && !trans_hook_list && (event == NFT_MSG_DELCHAIN || event == NFT_MSG_DESTROYCHAIN)) { nlmsg_end(skb, nlh); @@ -2026,7 +2107,8 @@ static int nf_tables_fill_chain_info(struct sk_buff *skb, struct net *net, const struct nft_base_chain *basechain = nft_base_chain(chain); struct nft_stats __percpu *stats; - if (nft_dump_basechain_hook(skb, net, family, basechain, hook_list)) + if (nft_dump_basechain_hook(skb, net, family, basechain, + hook_list, trans_hook_list)) goto nla_put_failure; if (nla_put_be32(skb, NFTA_CHAIN_POLICY, @@ -2062,7 +2144,8 @@ nla_put_failure: } static void nf_tables_chain_notify(const struct nft_ctx *ctx, int event, - const struct list_head *hook_list) + const struct list_head *hook_list, + const struct list_head *trans_hook_list) { struct nftables_pernet *nft_net; struct sk_buff *skb; @@ -2082,7 +2165,7 @@ static void nf_tables_chain_notify(const struct nft_ctx *ctx, int event, err = nf_tables_fill_chain_info(skb, ctx->net, ctx->portid, ctx->seq, event, flags, ctx->family, ctx->table, - ctx->chain, hook_list); + ctx->chain, hook_list, trans_hook_list); if (err < 0) { kfree_skb(skb); goto err; @@ -2128,7 +2211,7 @@ static int nf_tables_dump_chains(struct sk_buff *skb, NFT_MSG_NEWCHAIN, NLM_F_MULTI, table->family, table, - chain, NULL) < 0) + chain, NULL, NULL) < 0) goto done; nl_dump_check_consistent(cb, nlmsg_hdr(skb)); @@ -2182,7 +2265,7 @@ static int nf_tables_getchain(struct sk_buff *skb, const struct nfnl_info *info, err = nf_tables_fill_chain_info(skb2, net, NETLINK_CB(skb).portid, info->nlh->nlmsg_seq, NFT_MSG_NEWCHAIN, - 0, family, table, chain, NULL); + 0, family, table, chain, NULL, NULL); if (err < 0) goto err_fill_chain_info; @@ -2345,8 +2428,12 @@ static struct nft_hook *nft_hook_list_find(struct list_head *hook_list, list_for_each_entry(hook, hook_list, list) { if (!strncmp(hook->ifname, this->ifname, - min(hook->ifnamelen, this->ifnamelen))) + min(hook->ifnamelen, this->ifnamelen))) { + if (hook->flags & NFT_HOOK_REMOVE) + continue; + return hook; + } } return NULL; @@ -3105,6 +3192,32 @@ static int nf_tables_newchain(struct sk_buff *skb, const struct nfnl_info *info, return nf_tables_addchain(&ctx, family, policy, flags, extack); } +static int nft_trans_delhook(struct nft_hook *hook, + struct list_head *del_list) +{ + struct nft_trans_hook *trans_hook; + + trans_hook = kmalloc_obj(*trans_hook, GFP_KERNEL); + if (!trans_hook) + return -ENOMEM; + + trans_hook->hook = hook; + list_add_tail(&trans_hook->list, del_list); + hook->flags |= NFT_HOOK_REMOVE; + + return 0; +} + +static void nft_trans_delhook_abort(struct list_head *del_list) +{ + struct nft_trans_hook *trans_hook, *next; + + list_for_each_entry_safe(trans_hook, next, del_list, list) { + trans_hook->hook->flags &= ~NFT_HOOK_REMOVE; + nft_trans_hook_destroy(trans_hook); + } +} + static int nft_delchain_hook(struct nft_ctx *ctx, struct nft_base_chain *basechain, struct netlink_ext_ack *extack) @@ -3131,7 +3244,10 @@ static int nft_delchain_hook(struct nft_ctx *ctx, err = -ENOENT; goto err_chain_del_hook; } - list_move(&hook->list, &chain_del_list); + if (nft_trans_delhook(hook, &chain_del_list) < 0) { + err = -ENOMEM; + goto err_chain_del_hook; + } } trans = nft_trans_alloc_chain(ctx, NFT_MSG_DELCHAIN); @@ -3151,7 +3267,7 @@ static int nft_delchain_hook(struct nft_ctx *ctx, return 0; err_chain_del_hook: - list_splice(&chain_del_list, &basechain->hook_list); + nft_trans_delhook_abort(&chain_del_list); nft_chain_release_hook(&chain_hook); return err; @@ -8941,6 +9057,24 @@ static void nft_hooks_destroy(struct list_head *hook_list) nft_netdev_hook_unlink_free_rcu(hook); } +static void nft_flowtable_unregister_trans_hook(struct net *net, + struct nft_flowtable *flowtable, + struct list_head *hook_list) +{ + struct nft_trans_hook *trans_hook, *next; + struct nf_hook_ops *ops; + struct nft_hook *hook; + + list_for_each_entry_safe(trans_hook, next, hook_list, list) { + hook = trans_hook->hook; + list_for_each_entry(ops, &hook->ops_list, list) + nft_unregister_flowtable_ops(net, flowtable, ops); + + nft_netdev_hook_unlink_free_rcu(hook); + nft_trans_hook_destroy(trans_hook); + } +} + static int nft_flowtable_update(struct nft_ctx *ctx, const struct nlmsghdr *nlh, struct nft_flowtable *flowtable, struct netlink_ext_ack *extack) @@ -9199,7 +9333,10 @@ static int nft_delflowtable_hook(struct nft_ctx *ctx, err = -ENOENT; goto err_flowtable_del_hook; } - list_move(&hook->list, &flowtable_del_list); + if (nft_trans_delhook(hook, &flowtable_del_list) < 0) { + err = -ENOMEM; + goto err_flowtable_del_hook; + } } trans = nft_trans_alloc(ctx, NFT_MSG_DELFLOWTABLE, @@ -9220,7 +9357,7 @@ static int nft_delflowtable_hook(struct nft_ctx *ctx, return 0; err_flowtable_del_hook: - list_splice(&flowtable_del_list, &flowtable->hook_list); + nft_trans_delhook_abort(&flowtable_del_list); nft_flowtable_hook_release(&flowtable_hook); return err; @@ -9285,8 +9422,10 @@ static int nf_tables_fill_flowtable_info(struct sk_buff *skb, struct net *net, u32 portid, u32 seq, int event, u32 flags, int family, struct nft_flowtable *flowtable, - struct list_head *hook_list) + struct list_head *hook_list, + struct list_head *trans_hook_list) { + struct nft_trans_hook *trans_hook; struct nlattr *nest, *nest_devs; struct nft_hook *hook; struct nlmsghdr *nlh; @@ -9303,7 +9442,7 @@ static int nf_tables_fill_flowtable_info(struct sk_buff *skb, struct net *net, NFTA_FLOWTABLE_PAD)) goto nla_put_failure; - if (!hook_list && + if (!hook_list && !trans_hook_list && (event == NFT_MSG_DELFLOWTABLE || event == NFT_MSG_DESTROYFLOWTABLE)) { nlmsg_end(skb, nlh); @@ -9325,13 +9464,20 @@ static int nf_tables_fill_flowtable_info(struct sk_buff *skb, struct net *net, if (!nest_devs) goto nla_put_failure; - if (!hook_list) + if (!hook_list && !trans_hook_list) hook_list = &flowtable->hook_list; - list_for_each_entry_rcu(hook, hook_list, list, - lockdep_commit_lock_is_held(net)) { - if (nft_nla_put_hook_dev(skb, hook)) - goto nla_put_failure; + if (hook_list) { + list_for_each_entry_rcu(hook, hook_list, list, + lockdep_commit_lock_is_held(net)) { + if (nft_nla_put_hook_dev(skb, hook)) + goto nla_put_failure; + } + } else if (trans_hook_list) { + list_for_each_entry(trans_hook, trans_hook_list, list) { + if (nft_nla_put_hook_dev(skb, trans_hook->hook)) + goto nla_put_failure; + } } nla_nest_end(skb, nest_devs); nla_nest_end(skb, nest); @@ -9385,7 +9531,7 @@ static int nf_tables_dump_flowtable(struct sk_buff *skb, NFT_MSG_NEWFLOWTABLE, NLM_F_MULTI | NLM_F_APPEND, table->family, - flowtable, NULL) < 0) + flowtable, NULL, NULL) < 0) goto done; nl_dump_check_consistent(cb, nlmsg_hdr(skb)); @@ -9485,7 +9631,7 @@ static int nf_tables_getflowtable(struct sk_buff *skb, err = nf_tables_fill_flowtable_info(skb2, net, NETLINK_CB(skb).portid, info->nlh->nlmsg_seq, NFT_MSG_NEWFLOWTABLE, 0, family, - flowtable, NULL); + flowtable, NULL, NULL); if (err < 0) goto err_fill_flowtable_info; @@ -9498,7 +9644,9 @@ err_fill_flowtable_info: static void nf_tables_flowtable_notify(struct nft_ctx *ctx, struct nft_flowtable *flowtable, - struct list_head *hook_list, int event) + struct list_head *hook_list, + struct list_head *trans_hook_list, + int event) { struct nftables_pernet *nft_net = nft_pernet(ctx->net); struct sk_buff *skb; @@ -9518,7 +9666,8 @@ static void nf_tables_flowtable_notify(struct nft_ctx *ctx, err = nf_tables_fill_flowtable_info(skb, ctx->net, ctx->portid, ctx->seq, event, flags, - ctx->family, flowtable, hook_list); + ctx->family, flowtable, + hook_list, trans_hook_list); if (err < 0) { kfree_skb(skb); goto err; @@ -10052,9 +10201,7 @@ static void nft_commit_release(struct nft_trans *trans) break; case NFT_MSG_DELCHAIN: case NFT_MSG_DESTROYCHAIN: - if (nft_trans_chain_update(trans)) - nft_hooks_destroy(&nft_trans_chain_hooks(trans)); - else + if (!nft_trans_chain_update(trans)) nf_tables_chain_destroy(nft_trans_chain(trans)); break; case NFT_MSG_DELRULE: @@ -10075,9 +10222,7 @@ static void nft_commit_release(struct nft_trans *trans) break; case NFT_MSG_DELFLOWTABLE: case NFT_MSG_DESTROYFLOWTABLE: - if (nft_trans_flowtable_update(trans)) - nft_hooks_destroy(&nft_trans_flowtable_hooks(trans)); - else + if (!nft_trans_flowtable_update(trans)) nf_tables_flowtable_destroy(nft_trans_flowtable(trans)); break; } @@ -10837,31 +10982,28 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb) if (nft_trans_chain_update(trans)) { nft_chain_commit_update(nft_trans_container_chain(trans)); nf_tables_chain_notify(&ctx, NFT_MSG_NEWCHAIN, - &nft_trans_chain_hooks(trans)); + &nft_trans_chain_hooks(trans), NULL); list_splice_rcu(&nft_trans_chain_hooks(trans), &nft_trans_basechain(trans)->hook_list); /* trans destroyed after rcu grace period */ } else { nft_chain_commit_drop_policy(nft_trans_container_chain(trans)); nft_clear(net, nft_trans_chain(trans)); - nf_tables_chain_notify(&ctx, NFT_MSG_NEWCHAIN, NULL); + nf_tables_chain_notify(&ctx, NFT_MSG_NEWCHAIN, NULL, NULL); nft_trans_destroy(trans); } break; case NFT_MSG_DELCHAIN: case NFT_MSG_DESTROYCHAIN: if (nft_trans_chain_update(trans)) { - nf_tables_chain_notify(&ctx, NFT_MSG_DELCHAIN, + nf_tables_chain_notify(&ctx, NFT_MSG_DELCHAIN, NULL, &nft_trans_chain_hooks(trans)); - if (!(table->flags & NFT_TABLE_F_DORMANT)) { - nft_netdev_unregister_hooks(net, - &nft_trans_chain_hooks(trans), - true); - } + nft_netdev_unregister_trans_hook(net, table, + &nft_trans_chain_hooks(trans)); } else { nft_chain_del(nft_trans_chain(trans)); nf_tables_chain_notify(&ctx, NFT_MSG_DELCHAIN, - NULL); + NULL, NULL); nf_tables_unregister_hook(ctx.net, ctx.table, nft_trans_chain(trans)); } @@ -10967,6 +11109,7 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb) nf_tables_flowtable_notify(&ctx, nft_trans_flowtable(trans), &nft_trans_flowtable_hooks(trans), + NULL, NFT_MSG_NEWFLOWTABLE); list_splice_rcu(&nft_trans_flowtable_hooks(trans), &nft_trans_flowtable(trans)->hook_list); @@ -10975,6 +11118,7 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb) nf_tables_flowtable_notify(&ctx, nft_trans_flowtable(trans), NULL, + NULL, NFT_MSG_NEWFLOWTABLE); } nft_trans_destroy(trans); @@ -10984,16 +11128,18 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb) if (nft_trans_flowtable_update(trans)) { nf_tables_flowtable_notify(&ctx, nft_trans_flowtable(trans), + NULL, &nft_trans_flowtable_hooks(trans), trans->msg_type); - nft_unregister_flowtable_net_hooks(net, - nft_trans_flowtable(trans), - &nft_trans_flowtable_hooks(trans)); + nft_flowtable_unregister_trans_hook(net, + nft_trans_flowtable(trans), + &nft_trans_flowtable_hooks(trans)); } else { list_del_rcu(&nft_trans_flowtable(trans)->list); nf_tables_flowtable_notify(&ctx, nft_trans_flowtable(trans), NULL, + NULL, trans->msg_type); nft_unregister_flowtable_net_hooks(net, nft_trans_flowtable(trans), @@ -11157,8 +11303,7 @@ static int __nf_tables_abort(struct net *net, enum nfnl_abort_action action) case NFT_MSG_DELCHAIN: case NFT_MSG_DESTROYCHAIN: if (nft_trans_chain_update(trans)) { - list_splice(&nft_trans_chain_hooks(trans), - &nft_trans_basechain(trans)->hook_list); + nft_trans_delhook_abort(&nft_trans_chain_hooks(trans)); } else { nft_use_inc_restore(&table->use); nft_clear(trans->net, nft_trans_chain(trans)); @@ -11272,8 +11417,7 @@ static int __nf_tables_abort(struct net *net, enum nfnl_abort_action action) case NFT_MSG_DELFLOWTABLE: case NFT_MSG_DESTROYFLOWTABLE: if (nft_trans_flowtable_update(trans)) { - list_splice(&nft_trans_flowtable_hooks(trans), - &nft_trans_flowtable(trans)->hook_list); + nft_trans_delhook_abort(&nft_trans_flowtable_hooks(trans)); } else { nft_use_inc_restore(&table->use); nft_clear(trans->net, nft_trans_flowtable(trans)); -- cgit v1.2.3 From 846c76ecc02973b05ae909dd4248c11bfa277fc1 Mon Sep 17 00:00:00 2001 From: KaFai Wan Date: Tue, 21 Apr 2026 23:58:01 +0800 Subject: bpf: Reject TCP_NODELAY in TCP header option callbacks A BPF_SOCK_OPS program can enable BPF_SOCK_OPS_WRITE_HDR_OPT_CB_FLAG and then call bpf_setsockopt(TCP_NODELAY) from BPF_SOCK_OPS_HDR_OPT_LEN_CB or BPF_SOCK_OPS_WRITE_HDR_OPT_CB. In these callbacks, bpf_setsockopt(TCP_NODELAY) can reach __tcp_sock_set_nodelay(), which can call tcp_push_pending_frames(). >From BPF_SOCK_OPS_HDR_OPT_LEN_CB, tcp_push_pending_frames() can call tcp_current_mss(), which calls tcp_established_options() and re-enters bpf_skops_hdr_opt_len(). BPF_SOCK_OPS_HDR_OPT_LEN_CB -> bpf_setsockopt(TCP_NODELAY) -> tcp_push_pending_frames() -> tcp_current_mss() -> tcp_established_options() -> bpf_skops_hdr_opt_len() -> BPF_SOCK_OPS_HDR_OPT_LEN_CB >From BPF_SOCK_OPS_WRITE_HDR_OPT_CB, tcp_push_pending_frames() can call tcp_write_xmit(), which calls tcp_transmit_skb(). That path recomputes header option length through tcp_established_options() and bpf_skops_hdr_opt_len() before re-entering bpf_skops_write_hdr_opt(). BPF_SOCK_OPS_WRITE_HDR_OPT_CB -> bpf_setsockopt(TCP_NODELAY) -> tcp_push_pending_frames() -> tcp_write_xmit() -> tcp_transmit_skb() -> tcp_established_options() -> bpf_skops_hdr_opt_len() -> bpf_skops_write_hdr_opt() -> BPF_SOCK_OPS_WRITE_HDR_OPT_CB This leads to unbounded recursion and can overflow the kernel stack. Reject TCP_NODELAY with -EOPNOTSUPP in bpf_sock_ops_setsockopt() when bpf_setsockopt() is called from BPF_SOCK_OPS_HDR_OPT_LEN_CB or BPF_SOCK_OPS_WRITE_HDR_OPT_CB. Fixes: 7e41df5dbba2 ("bpf: Add a few optnames to bpf_setsockopt") Closes: https://lore.kernel.org/bpf/d1d523c9-6901-4454-a183-94462b8f3e4e@std.uestc.edu.cn/ Reported-by: Quan Sun <2022090917019@std.uestc.edu.cn> Reported-by: Yinhao Hu Reported-by: Kaiyan Mei Signed-off-by: KaFai Wan Signed-off-by: Martin KaFai Lau Reviewed-by: Jiayuan Chen Link: https://patch.msgid.link/20260421155804.135786-2-kafai.wan@linux.dev --- net/core/filter.c | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'net') diff --git a/net/core/filter.c b/net/core/filter.c index 5fa9189eb772..96849f4c1fbc 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -5833,6 +5833,12 @@ BPF_CALL_5(bpf_sock_ops_setsockopt, struct bpf_sock_ops_kern *, bpf_sock, if (!is_locked_tcp_sock_ops(bpf_sock)) return -EOPNOTSUPP; + /* TCP_NODELAY triggers tcp_push_pending_frames() and re-enters these callbacks. */ + if ((bpf_sock->op == BPF_SOCK_OPS_HDR_OPT_LEN_CB || + bpf_sock->op == BPF_SOCK_OPS_WRITE_HDR_OPT_CB) && + level == SOL_TCP && optname == TCP_NODELAY) + return -EOPNOTSUPP; + return _bpf_setsockopt(bpf_sock->sk, level, optname, optval, optlen); } -- cgit v1.2.3 From 54377fcab51f6f1f8807827d3751be42279e1a6a Mon Sep 17 00:00:00 2001 From: KaFai Wan Date: Tue, 21 Apr 2026 23:58:02 +0800 Subject: bpf: Reject TCP_NODELAY in bpf-tcp-cc A BPF TCP congestion control program can call bpf_setsockopt() from its callbacks. In current kernels, if it calls bpf_setsockopt(TCP_NODELAY) from cwnd_event_tx_start(), the call can re-enter the TCP transmit path before the outer tcp_transmit_skb() has completed and advanced the send head. This can re-trigger CA_EVENT_TX_START and lead to unbounded recursion: tcp_transmit_skb() -> tcp_event_data_sent() -> tcp_ca_event(sk, CA_EVENT_TX_START) -> cwnd_event_tx_start() -> bpf_setsockopt(TCP_NODELAY) -> tcp_push_pending_frames() -> tcp_write_xmit() -> tcp_transmit_skb() This leads to unbounded recursion and can overflow the kernel stack. Reject TCP_NODELAY with -EOPNOTSUPP for bpf-tcp-cc by introducing a dedicated setsockopt proto for BPF_PROG_TYPE_STRUCT_OPS TCP congestion control programs. To keep it simple, all tcp-cc ops is rejected for TCP_NODELAY. Fixes: 7e41df5dbba2 ("bpf: Add a few optnames to bpf_setsockopt") Suggested-by: Martin KaFai Lau Signed-off-by: KaFai Wan Signed-off-by: Martin KaFai Lau Reviewed-by: Jiayuan Chen Link: https://patch.msgid.link/20260421155804.135786-3-kafai.wan@linux.dev --- net/core/filter.c | 24 ++++++++++++++++++++++++ net/ipv4/bpf_tcp_ca.c | 2 +- 2 files changed, 25 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/core/filter.c b/net/core/filter.c index 96849f4c1fbc..2914f5330310 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -5688,6 +5688,30 @@ const struct bpf_func_proto bpf_sk_getsockopt_proto = { .arg5_type = ARG_CONST_SIZE, }; +BPF_CALL_5(bpf_sk_setsockopt_nodelay, struct sock *, sk, int, level, + int, optname, char *, optval, int, optlen) +{ + /* + * TCP_NODELAY triggers tcp_push_pending_frames() and re-enters + * CA_EVENT_TX_START in bpf_tcp_cc. + */ + if (level == SOL_TCP && optname == TCP_NODELAY) + return -EOPNOTSUPP; + + return _bpf_setsockopt(sk, level, optname, optval, optlen); +} + +const struct bpf_func_proto bpf_sk_setsockopt_nodelay_proto = { + .func = bpf_sk_setsockopt_nodelay, + .gpl_only = false, + .ret_type = RET_INTEGER, + .arg1_type = ARG_PTR_TO_BTF_ID_SOCK_COMMON, + .arg2_type = ARG_ANYTHING, + .arg3_type = ARG_ANYTHING, + .arg4_type = ARG_PTR_TO_MEM | MEM_RDONLY, + .arg5_type = ARG_CONST_SIZE, +}; + BPF_CALL_5(bpf_unlocked_sk_setsockopt, struct sock *, sk, int, level, int, optname, char *, optval, int, optlen) { diff --git a/net/ipv4/bpf_tcp_ca.c b/net/ipv4/bpf_tcp_ca.c index 008edc7f6688..791e15063237 100644 --- a/net/ipv4/bpf_tcp_ca.c +++ b/net/ipv4/bpf_tcp_ca.c @@ -168,7 +168,7 @@ bpf_tcp_ca_get_func_proto(enum bpf_func_id func_id, */ if (prog_ops_moff(prog) != offsetof(struct tcp_congestion_ops, release)) - return &bpf_sk_setsockopt_proto; + return &bpf_sk_setsockopt_nodelay_proto; return NULL; case BPF_FUNC_getsockopt: /* Since get/setsockopt is usually expected to -- cgit v1.2.3 From 375e4e33c18dfa05c5dfd5f3dfffeb29343dd4c7 Mon Sep 17 00:00:00 2001 From: Weiming Shi Date: Tue, 21 Apr 2026 23:54:12 -0700 Subject: bpf: Fix NULL pointer dereference in bpf_sk_storage_clone and diag paths bpf_selem_unlink_nofail() sets SDATA(selem)->smap to NULL before removing the selem from the storage hlist. A concurrent RCU reader in bpf_sk_storage_clone() can observe the selem still on the list with smap already NULL, causing a NULL pointer dereference. general protection fault, probably for non-canonical address 0xdffffc000000000a: KASAN: null-ptr-deref in range [0x0000000000000050-0x0000000000000057] RIP: 0010:bpf_sk_storage_clone+0x1cd/0xaa0 net/core/bpf_sk_storage.c:174 Call Trace: sk_clone+0xfed/0x1980 net/core/sock.c:2591 inet_csk_clone_lock+0x30/0x760 net/ipv4/inet_connection_sock.c:1222 tcp_create_openreq_child+0x35/0x2680 net/ipv4/tcp_minisocks.c:571 tcp_v4_syn_recv_sock+0x123/0xf90 net/ipv4/tcp_ipv4.c:1729 tcp_check_req+0x8e1/0x2580 include/net/tcp.h:855 tcp_v4_rcv+0x1845/0x3b80 net/ipv4/tcp_ipv4.c:2347 Add a NULL check for smap in bpf_sk_storage_clone(). bpf_sk_storage_diag_put_all() has the same issue. Add a NULL check and pass the validated smap directly to diag_get(), which is refactored to take smap as a parameter instead of reading it internally. bpf_sk_storage_diag_put() uses diag->maps[i] which is always valid under its refcount, so diag->maps[i] is passed directly to diag_get(). Fixes: 5d800f87d0a5 ("bpf: Support lockless unlink when freeing map or local storage") Reported-by: Xiang Mei Acked-by: Amery Hung Signed-off-by: Weiming Shi Signed-off-by: Martin KaFai Lau Link: https://patch.msgid.link/20260422065411.1007737-2-bestswngs@gmail.com --- net/core/bpf_sk_storage.c | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/net/core/bpf_sk_storage.c b/net/core/bpf_sk_storage.c index 14eb7812bda4..dc3e8fce8809 100644 --- a/net/core/bpf_sk_storage.c +++ b/net/core/bpf_sk_storage.c @@ -172,7 +172,7 @@ int bpf_sk_storage_clone(const struct sock *sk, struct sock *newsk) struct bpf_map *map; smap = rcu_dereference(SDATA(selem)->smap); - if (!(smap->map.map_flags & BPF_F_CLONE)) + if (!smap || !(smap->map.map_flags & BPF_F_CLONE)) continue; /* Note that for lockless listeners adding new element @@ -531,10 +531,10 @@ err_free: } EXPORT_SYMBOL_GPL(bpf_sk_storage_diag_alloc); -static int diag_get(struct bpf_local_storage_data *sdata, struct sk_buff *skb) +static int diag_get(struct bpf_local_storage_map *smap, + struct bpf_local_storage_data *sdata, struct sk_buff *skb) { struct nlattr *nla_stg, *nla_value; - struct bpf_local_storage_map *smap; /* It cannot exceed max nlattr's payload */ BUILD_BUG_ON(U16_MAX - NLA_HDRLEN < BPF_LOCAL_STORAGE_MAX_VALUE_SIZE); @@ -543,7 +543,6 @@ static int diag_get(struct bpf_local_storage_data *sdata, struct sk_buff *skb) if (!nla_stg) return -EMSGSIZE; - smap = rcu_dereference(sdata->smap); if (nla_put_u32(skb, SK_DIAG_BPF_STORAGE_MAP_ID, smap->map.id)) goto errout; @@ -596,9 +595,11 @@ static int bpf_sk_storage_diag_put_all(struct sock *sk, struct sk_buff *skb, saved_len = skb->len; hlist_for_each_entry_rcu(selem, &sk_storage->list, snode) { smap = rcu_dereference(SDATA(selem)->smap); + if (!smap) + continue; diag_size += nla_value_size(smap->map.value_size); - if (nla_stgs && diag_get(SDATA(selem), skb)) + if (nla_stgs && diag_get(smap, SDATA(selem), skb)) /* Continue to learn diag_size */ err = -EMSGSIZE; } @@ -665,7 +666,7 @@ int bpf_sk_storage_diag_put(struct bpf_sk_storage_diag *diag, diag_size += nla_value_size(diag->maps[i]->value_size); - if (nla_stgs && diag_get(sdata, skb)) + if (nla_stgs && diag_get((struct bpf_local_storage_map *)diag->maps[i], sdata, skb)) /* Continue to learn diag_size */ err = -EMSGSIZE; } -- cgit v1.2.3 From 6451d58a355642b612f2bf948ad39108c998ac2a Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Mon, 20 Apr 2026 19:48:41 +0000 Subject: sockmap: Fix sk_psock_drop() race vs sock_map_{unhash,close,destroy}(). syzbot reported a splat in sock_map_destroy() [0], where psock was NULL even though sk->sk_prot still pointed to tcp_bpf_prots[][]. The stack trace shows how badly the path was excercised, see inet_release() calls tcp_close(), not sock_map_close() yet, but finally reaching sock_map_destroy(). The root cause is a lack of synchronisation. Even if sk_psock_get() fails to bump psock->refcnt, it does not guarantee that sk_psock_drop() has finished, and thus sk->sk_prot might not have been restored to the original one. Commit 4b4647add7d3 ("sock_map: avoid race between sock_map_close and sk_psock_put") attempted to address this, but it was insufficient for two reasons. It did not cover sock_map_unhash() and sock_map_destroy(), and it missed the corner case where sk_psock() is NULL. On non-x86 platforms, sk_psock_restore_proto(sk, psock) and rcu_assign_sk_user_data(sk, NULL) can be reordered because there is no address dependency between sk->sk_prot and sk->sk_user_data. sk_psock_get() returning NULL implies nothing about sk->sk_prot. Let's simply retry sk_psock_get() in the unlikely case. Note that we cannot avoid loop even if we added memory barrier in sk_psock_drop() and sock_map_psock_get_checked(). Also note that sock_map_destroy() cannot be called from softirq while sock_map_close() has also been running. It is because sock_map_destroy() requires SOCK_DEAD, so sock_map_destroy() cannot happen until sock_map_close() has finished the saved_close() (which is tcp_close()). [0]: WARNING: CPU: 1 PID: 8459 at net/core/sock_map.c:1667 sock_map_destroy+0x28b/0x2b0 net/core/sock_map.c:1667 Modules linked in: CPU: 1 UID: 0 PID: 8459 Comm: syz.0.1109 Not tainted syzkaller #0 PREEMPT_{RT,(full)} Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/12/2025 RIP: 0010:sock_map_destroy+0x28b/0x2b0 net/core/sock_map.c:1667 Code: 8b 36 49 83 c6 38 4c 89 f0 48 c1 e8 03 42 80 3c 38 00 74 08 4c 89 f7 e8 93 62 22 f9 4d 8b 3e e9 79 ff ff ff e8 a6 2b c3 f8 90 <0f> 0b 90 eb 9c e8 9b 2b c3 f8 4c 89 e7 be 03 00 00 00 e8 0e 4e bc RSP: 0018:ffffc9000d067be8 EFLAGS: 00010293 RAX: ffffffff88fb30aa RBX: ffff888024832000 RCX: ffff888024283b80 RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000000 RBP: 0000000000000001 R08: 0000000000000000 R09: 0000000000000000 R10: dffffc0000000000 R11: ffffed100862e946 R12: dffffc0000000000 R13: ffff888024832000 R14: ffffffff995b2208 R15: ffffffff88fb2e20 FS: 0000555579a7d500(0000) GS:ffff8881269c2000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00002000000048c0 CR3: 000000003713a000 CR4: 00000000003526f0 Call Trace: inet_csk_destroy_sock+0x166/0x3a0 net/ipv4/inet_connection_sock.c:1294 __tcp_close+0xcc1/0xfd0 net/ipv4/tcp.c:3262 tcp_close+0x28/0x110 net/ipv4/tcp.c:3274 inet_release+0x144/0x190 net/ipv4/af_inet.c:435 __sock_release net/socket.c:649 [inline] sock_close+0xc0/0x240 net/socket.c:1439 __fput+0x45b/0xa80 fs/file_table.c:468 task_work_run+0x1d4/0x260 kernel/task_work.c:227 resume_user_mode_work include/linux/resume_user_mode.h:50 [inline] exit_to_user_mode_loop+0xec/0x110 kernel/entry/common.c:43 exit_to_user_mode_prepare include/linux/irq-entry-common.h:225 [inline] syscall_exit_to_user_mode_work include/linux/entry-common.h:175 [inline] syscall_exit_to_user_mode include/linux/entry-common.h:210 [inline] do_syscall_64+0x2bd/0x3b0 arch/x86/entry/syscall_64.c:100 entry_SYSCALL_64_after_hwframe+0x77/0x7f RIP: 0033:0x7f265847ebe9 Code: ff ff c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 40 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 a8 ff ff ff f7 d8 64 89 01 48 RSP: 002b:00007ffd158dfbd8 EFLAGS: 00000246 ORIG_RAX: 00000000000001b4 RAX: 0000000000000000 RBX: 000000000002ddb0 RCX: 00007f265847ebe9 RDX: 0000000000000000 RSI: 000000000000001e RDI: 0000000000000003 RBP: 00007f26586a7da0 R08: 0000000000000001 R09: 0000000e158dfecf R10: 0000001b30a20000 R11: 0000000000000246 R12: 00007f26586a5fac R13: 00007f26586a5fa0 R14: ffffffffffffffff R15: 00007ffd158dfcf0 Fixes: 1aa12bdf1bfb ("bpf: sockmap, add sock close() hook to remove socks") Fixes: b05545e15e1f ("bpf: sockmap, fix transition through disconnect without close") Fixes: d8616ee2affc ("bpf, sockmap: Fix sk->sk_forward_alloc warn_on in sk_stream_kill_queues") Reported-by: syzbot+b0842d38af58376d1fdc@syzkaller.appspotmail.com Closes: https://lore.kernel.org/bpf/69cec5ef.050a0220.2dbe29.0009.GAE@google.com/ Signed-off-by: Kuniyuki Iwashima Signed-off-by: Martin KaFai Lau Reviewed-by: Jiayuan Chen Link: https://patch.msgid.link/20260420194846.1089595-1-kuniyu@google.com --- net/core/sock_map.c | 39 +++++++++++++++++++++++++-------------- 1 file changed, 25 insertions(+), 14 deletions(-) (limited to 'net') diff --git a/net/core/sock_map.c b/net/core/sock_map.c index 02a68be3002a..99e3789492a0 100644 --- a/net/core/sock_map.c +++ b/net/core/sock_map.c @@ -1630,18 +1630,23 @@ void sock_map_unhash(struct sock *sk) void (*saved_unhash)(struct sock *sk); struct sk_psock *psock; +retry: rcu_read_lock(); psock = sk_psock(sk); if (unlikely(!psock)) { rcu_read_unlock(); saved_unhash = READ_ONCE(sk->sk_prot)->unhash; + if (unlikely(saved_unhash == sock_map_unhash)) + goto retry; } else { saved_unhash = psock->saved_unhash; sock_map_remove_links(sk, psock); rcu_read_unlock(); + + if (WARN_ON_ONCE(saved_unhash == sock_map_unhash)) + return; } - if (WARN_ON_ONCE(saved_unhash == sock_map_unhash)) - return; + if (saved_unhash) saved_unhash(sk); } @@ -1652,20 +1657,25 @@ void sock_map_destroy(struct sock *sk) void (*saved_destroy)(struct sock *sk); struct sk_psock *psock; +retry: rcu_read_lock(); psock = sk_psock_get(sk); if (unlikely(!psock)) { rcu_read_unlock(); saved_destroy = READ_ONCE(sk->sk_prot)->destroy; + if (unlikely(saved_destroy == sock_map_destroy)) + goto retry; } else { saved_destroy = psock->saved_destroy; sock_map_remove_links(sk, psock); rcu_read_unlock(); sk_psock_stop(psock); sk_psock_put(sk, psock); + + if (WARN_ON_ONCE(saved_destroy == sock_map_destroy)) + return; } - if (WARN_ON_ONCE(saved_destroy == sock_map_destroy)) - return; + if (saved_destroy) saved_destroy(sk); } @@ -1676,32 +1686,33 @@ void sock_map_close(struct sock *sk, long timeout) void (*saved_close)(struct sock *sk, long timeout); struct sk_psock *psock; +retry: lock_sock(sk); rcu_read_lock(); - psock = sk_psock(sk); + psock = sk_psock_get(sk); if (likely(psock)) { saved_close = psock->saved_close; sock_map_remove_links(sk, psock); - psock = sk_psock_get(sk); - if (unlikely(!psock)) - goto no_psock; rcu_read_unlock(); sk_psock_stop(psock); release_sock(sk); cancel_delayed_work_sync(&psock->work); sk_psock_put(sk, psock); + + /* Make sure we do not recurse. This is a bug. + * Leak the socket instead of crashing on a stack overflow. + */ + if (WARN_ON_ONCE(saved_close == sock_map_close)) + return; } else { saved_close = READ_ONCE(sk->sk_prot)->close; -no_psock: rcu_read_unlock(); release_sock(sk); + + if (unlikely(saved_close == sock_map_close)) + goto retry; } - /* Make sure we do not recurse. This is a bug. - * Leak the socket instead of crashing on a stack overflow. - */ - if (WARN_ON_ONCE(saved_close == sock_map_close)) - return; saved_close(sk, timeout); } EXPORT_SYMBOL_GPL(sock_map_close); -- cgit v1.2.3 From 1081de1accb2b224516cca7071122c59532d0b22 Mon Sep 17 00:00:00 2001 From: Weiming Shi Date: Thu, 23 Apr 2026 11:38:32 -0700 Subject: bpf: Fix NULL pointer dereference in bpf_skb_fib_lookup() When tot_len is not provided by the user, bpf_skb_fib_lookup() resolves the FIB result's output device via dev_get_by_index_rcu() to check skb forwardability and fill in mtu_result. The returned pointer is dereferenced without a NULL check. If the device is concurrently unregistered, dev_get_by_index_rcu() returns NULL and is_skb_forwardable() crashes at dev->flags: KASAN: null-ptr-deref in range [0x00000000000000b0-0x00000000000000b7] Call Trace: is_skb_forwardable (include/linux/netdevice.h:4365) bpf_skb_fib_lookup (net/core/filter.c:6446) bpf_prog_test_run_skb (net/bpf/test_run.c) __sys_bpf (kernel/bpf/syscall.c) Add the missing NULL check, returning -ENODEV to be consistent with how bpf_ipv4_fib_lookup() and bpf_ipv6_fib_lookup() handle the same condition. Fixes: 4f74fede40df ("bpf: Add mtu checking to FIB forwarding helper") Reported-by: Xiang Mei Signed-off-by: Weiming Shi Signed-off-by: Martin KaFai Lau Acked-by: Paul Chaignon Link: https://patch.msgid.link/20260423183831.1325480-2-bestswngs@gmail.com --- net/core/filter.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'net') diff --git a/net/core/filter.c b/net/core/filter.c index 2914f5330310..bc96c18df4e0 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -6473,6 +6473,8 @@ BPF_CALL_4(bpf_skb_fib_lookup, struct sk_buff *, skb, * against MTU of FIB lookup resulting net_device */ dev = dev_get_by_index_rcu(net, params->ifindex); + if (unlikely(!dev)) + return -ENODEV; if (!is_skb_forwardable(dev, skb)) rc = BPF_FIB_LKUP_RET_FRAG_NEEDED; -- cgit v1.2.3 From 4b2b4d7d4e203c92db8966b163edfacb1f0e1e29 Mon Sep 17 00:00:00 2001 From: Jiexun Wang Date: Fri, 17 Apr 2026 20:25:06 +0800 Subject: netfilter: xt_policy: fix strict mode inbound policy matching match_policy_in() walks sec_path entries from the last transform to the first one, but strict policy matching needs to consume info->pol[] in the same forward order as the rule layout. Derive the strict-match policy position from the number of transforms already consumed so that multi-element inbound rules are matched consistently. Fixes: c4b885139203 ("[NETFILTER]: x_tables: replace IPv4/IPv6 policy match by address family independant version") Reported-by: Yuan Tan Reported-by: Yifan Wu Reported-by: Juefei Pu Reported-by: Xin Liu Signed-off-by: Jiexun Wang Signed-off-by: Ren Wei Acked-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- net/netfilter/xt_policy.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/netfilter/xt_policy.c b/net/netfilter/xt_policy.c index cb6e8279010a..b5fa65558318 100644 --- a/net/netfilter/xt_policy.c +++ b/net/netfilter/xt_policy.c @@ -63,7 +63,7 @@ match_policy_in(const struct sk_buff *skb, const struct xt_policy_info *info, return 0; for (i = sp->len - 1; i >= 0; i--) { - pos = strict ? i - sp->len + 1 : 0; + pos = strict ? sp->len - i - 1 : 0; if (pos >= info->len) return 0; e = &info->pol[pos]; -- cgit v1.2.3 From fe11e5c40817b84abaa5d83bfb6586d8412bfd07 Mon Sep 17 00:00:00 2001 From: Kai Ma Date: Wed, 22 Apr 2026 22:54:18 +0800 Subject: netfilter: reject zero shift in nft_bitwise Reject zero shift operands for nft_bitwise left and right shift expressions during initialization. The carry propagation logic computes the carry from the adjacent 32-bit word using BITS_PER_TYPE(u32) - shift. A zero shift operand turns this into a 32-bit shift, which is undefined behaviour. Reject zero shift operands in the control plane, alongside the existing check for values greater than or equal to 32, so malformed rules never reach the packet path. Fixes: 567d746b55bc ("netfilter: bitwise: add support for shifts.") Cc: stable@kernel.org Reported-by: Yuan Tan Reported-by: Yifan Wu Reported-by: Juefei Pu Reported-by: Xin Liu Signed-off-by: Kai Ma Signed-off-by: Ren Wei Reviewed-by: Fernando Fernandez Mancera Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nft_bitwise.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/netfilter/nft_bitwise.c b/net/netfilter/nft_bitwise.c index 13808e9cd999..94dccdcfa06b 100644 --- a/net/netfilter/nft_bitwise.c +++ b/net/netfilter/nft_bitwise.c @@ -196,7 +196,8 @@ static int nft_bitwise_init_shift(struct nft_bitwise *priv, if (err < 0) return err; - if (priv->data.data[0] >= BITS_PER_TYPE(u32)) { + if (!priv->data.data[0] || + priv->data.data[0] >= BITS_PER_TYPE(u32)) { nft_data_release(&priv->data, desc.type); return -EINVAL; } -- cgit v1.2.3 From 8cf6809cddcbe301aedfc6b51bcd4944d45795f6 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 23 Apr 2026 02:19:11 +0200 Subject: netfilter: nf_conntrack_sip: don't use simple_strtoul MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Replace unsafe port parsing in epaddr_len(), ct_sip_parse_header_uri(), and ct_sip_parse_request() with a new sip_parse_port() helper that validates each digit against the buffer limit, eliminating the use of simple_strtoul() which assumes NUL-terminated strings. The previous code dereferenced pointers without bounds checks after sip_parse_addr() and relied on simple_strtoul() on non-NUL-terminated skb data. A port that reaches the buffer limit without a trailing character is also rejected as malformed. Also get rid of all simple_strtoul() usage in conntrack, prefer a stricter version instead. There are intentional changes: - Bail out if number is > UINT_MAX and indicate a failure, same for too long sequences. While we do accept 05535 as port 5535, we will not accept e.g. 'sip:10.0.0.1:005060'. While its syntactically valid under RFC 3261, we should restrict this to not waste cycles when presented with malformed packets with 64k '0' characters. - Force base 10 in ct_sip_parse_numerical_param(). This is used to fetch 'expire=' and 'rports='; both are expected to use base-10. - In nf_nat_sip.c, only accept the parsed value if its within the 1k-64k range. - epaddr_len now returns 0 if the port is invalid, as it already does for invalid ip addresses. This is intentional. nf_conntrack_sip performs lots of guesswork to find the right parts of the message to parse. Being stricter could break existing setups. Connection tracking helpers are designed to allow traffic to pass, not to block it. Based on an earlier patch from Jenny Guanni Qu . Fixes: 05e3ced297fe ("[NETFILTER]: nf_conntrack_sip: introduce SIP-URI parsing helper") Reported-by: Klaudia Kloc Reported-by: Dawid Moczadło Reported-by: Jenny Guanni Qu . Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_conntrack_sip.c | 152 ++++++++++++++++++++++++++++++--------- net/netfilter/nf_nat_sip.c | 1 + 2 files changed, 119 insertions(+), 34 deletions(-) (limited to 'net') diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c index 182cfb119448..1eb55907d470 100644 --- a/net/netfilter/nf_conntrack_sip.c +++ b/net/netfilter/nf_conntrack_sip.c @@ -181,6 +181,57 @@ static int sip_parse_addr(const struct nf_conn *ct, const char *cp, return 1; } +/* Parse optional port number after IP address. + * Returns false on malformed input, true otherwise. + * If port is non-NULL, stores parsed port in network byte order. + * If no port is present, sets *port to default SIP port. + */ +static bool sip_parse_port(const char *dptr, const char **endp, + const char *limit, __be16 *port) +{ + unsigned int p = 0; + int len = 0; + + if (dptr >= limit) + return false; + + if (*dptr != ':') { + if (port) + *port = htons(SIP_PORT); + if (endp) + *endp = dptr; + return true; + } + + dptr++; /* skip ':' */ + + while (dptr < limit && isdigit(*dptr)) { + p = p * 10 + (*dptr - '0'); + dptr++; + len++; + if (len > 5) /* max "65535" */ + return false; + } + + if (len == 0) + return false; + + /* reached limit while parsing port */ + if (dptr >= limit) + return false; + + if (p < 1024 || p > 65535) + return false; + + if (port) + *port = htons(p); + + if (endp) + *endp = dptr; + + return true; +} + /* skip ip address. returns its length. */ static int epaddr_len(const struct nf_conn *ct, const char *dptr, const char *limit, int *shift) @@ -193,11 +244,8 @@ static int epaddr_len(const struct nf_conn *ct, const char *dptr, return 0; } - /* Port number */ - if (*dptr == ':') { - dptr++; - dptr += digits_len(ct, dptr, limit, shift); - } + if (!sip_parse_port(dptr, &dptr, limit, NULL)) + return 0; return dptr - aux; } @@ -228,6 +276,51 @@ static int skp_epaddr_len(const struct nf_conn *ct, const char *dptr, return epaddr_len(ct, dptr, limit, shift); } +/* simple_strtoul stops after first non-number character. + * But as we're not dealing with c-strings, we can't rely on + * hitting \r,\n,\0 etc. before moving past end of buffer. + * + * This is a variant of simple_strtoul, but doesn't require + * a c-string. + * + * If value exceeds UINT_MAX, 0 is returned. + */ +static unsigned int sip_strtouint(const char *cp, unsigned int len, char **endp) +{ + const unsigned int max = sizeof("4294967295"); + unsigned int olen = len; + const char *s = cp; + u64 result = 0; + + if (len > max) + len = max; + + while (olen > 0 && isdigit(*s)) { + unsigned int value; + + if (len == 0) + goto err; + + value = *s - '0'; + result = result * 10 + value; + + if (result > UINT_MAX) + goto err; + s++; + len--; + olen--; + } + + if (endp) + *endp = (char *)s; + + return result; +err: + if (endp) + *endp = (char *)cp; + return 0; +} + /* Parse a SIP request line of the form: * * Request-Line = Method SP Request-URI SP SIP-Version CRLF @@ -241,7 +334,6 @@ int ct_sip_parse_request(const struct nf_conn *ct, { const char *start = dptr, *limit = dptr + datalen, *end; unsigned int mlen; - unsigned int p; int shift = 0; /* Skip method and following whitespace */ @@ -267,14 +359,8 @@ int ct_sip_parse_request(const struct nf_conn *ct, if (!sip_parse_addr(ct, dptr, &end, addr, limit, true)) return -1; - if (end < limit && *end == ':') { - end++; - p = simple_strtoul(end, (char **)&end, 10); - if (p < 1024 || p > 65535) - return -1; - *port = htons(p); - } else - *port = htons(SIP_PORT); + if (!sip_parse_port(end, &end, limit, port)) + return -1; if (end == dptr) return 0; @@ -509,7 +595,6 @@ int ct_sip_parse_header_uri(const struct nf_conn *ct, const char *dptr, union nf_inet_addr *addr, __be16 *port) { const char *c, *limit = dptr + datalen; - unsigned int p; int ret; ret = ct_sip_walk_headers(ct, dptr, dataoff ? *dataoff : 0, datalen, @@ -520,14 +605,8 @@ int ct_sip_parse_header_uri(const struct nf_conn *ct, const char *dptr, if (!sip_parse_addr(ct, dptr + *matchoff, &c, addr, limit, true)) return -1; - if (*c == ':') { - c++; - p = simple_strtoul(c, (char **)&c, 10); - if (p < 1024 || p > 65535) - return -1; - *port = htons(p); - } else - *port = htons(SIP_PORT); + if (!sip_parse_port(c, &c, limit, port)) + return -1; if (dataoff) *dataoff = c - dptr; @@ -609,7 +688,7 @@ int ct_sip_parse_numerical_param(const struct nf_conn *ct, const char *dptr, return 0; start += strlen(name); - *val = simple_strtoul(start, &end, 0); + *val = sip_strtouint(start, limit - start, (char **)&end); if (start == end) return -1; if (matchoff && matchlen) { @@ -1064,6 +1143,8 @@ static int process_sdp(struct sk_buff *skb, unsigned int protoff, mediaoff = sdpoff; for (i = 0; i < ARRAY_SIZE(sdp_media_types); ) { + char *end; + if (ct_sip_get_sdp_header(ct, *dptr, mediaoff, *datalen, SDP_HDR_MEDIA, SDP_HDR_UNSPEC, &mediaoff, &medialen) <= 0) @@ -1079,8 +1160,8 @@ static int process_sdp(struct sk_buff *skb, unsigned int protoff, mediaoff += t->len; medialen -= t->len; - port = simple_strtoul(*dptr + mediaoff, NULL, 10); - if (port == 0) + port = sip_strtouint(*dptr + mediaoff, *datalen - mediaoff, (char **)&end); + if (port == 0 || *dptr + mediaoff == end) continue; if (port < 1024 || port > 65535) { nf_ct_helper_log(skb, ct, "wrong port %u", port); @@ -1254,7 +1335,7 @@ static int process_register_request(struct sk_buff *skb, unsigned int protoff, */ if (ct_sip_get_header(ct, *dptr, 0, *datalen, SIP_HDR_EXPIRES, &matchoff, &matchlen) > 0) - expires = simple_strtoul(*dptr + matchoff, NULL, 10); + expires = sip_strtouint(*dptr + matchoff, *datalen - matchoff, NULL); ret = ct_sip_parse_header_uri(ct, *dptr, NULL, *datalen, SIP_HDR_CONTACT, NULL, @@ -1358,7 +1439,7 @@ static int process_register_response(struct sk_buff *skb, unsigned int protoff, if (ct_sip_get_header(ct, *dptr, 0, *datalen, SIP_HDR_EXPIRES, &matchoff, &matchlen) > 0) - expires = simple_strtoul(*dptr + matchoff, NULL, 10); + expires = sip_strtouint(*dptr + matchoff, *datalen - matchoff, NULL); while (1) { unsigned int c_expires = expires; @@ -1418,10 +1499,12 @@ static int process_sip_response(struct sk_buff *skb, unsigned int protoff, struct nf_conn *ct = nf_ct_get(skb, &ctinfo); unsigned int matchoff, matchlen, matchend; unsigned int code, cseq, i; + char *end; if (*datalen < strlen("SIP/2.0 200")) return NF_ACCEPT; - code = simple_strtoul(*dptr + strlen("SIP/2.0 "), NULL, 10); + code = sip_strtouint(*dptr + strlen("SIP/2.0 "), + *datalen - strlen("SIP/2.0 "), NULL); if (!code) { nf_ct_helper_log(skb, ct, "cannot get code"); return NF_DROP; @@ -1432,8 +1515,8 @@ static int process_sip_response(struct sk_buff *skb, unsigned int protoff, nf_ct_helper_log(skb, ct, "cannot parse cseq"); return NF_DROP; } - cseq = simple_strtoul(*dptr + matchoff, NULL, 10); - if (!cseq && *(*dptr + matchoff) != '0') { + cseq = sip_strtouint(*dptr + matchoff, *datalen - matchoff, (char **)&end); + if (*dptr + matchoff == end) { nf_ct_helper_log(skb, ct, "cannot get cseq"); return NF_DROP; } @@ -1482,6 +1565,7 @@ static int process_sip_request(struct sk_buff *skb, unsigned int protoff, for (i = 0; i < ARRAY_SIZE(sip_handlers); i++) { const struct sip_handler *handler; + char *end; handler = &sip_handlers[i]; if (handler->request == NULL) @@ -1498,8 +1582,8 @@ static int process_sip_request(struct sk_buff *skb, unsigned int protoff, nf_ct_helper_log(skb, ct, "cannot parse cseq"); return NF_DROP; } - cseq = simple_strtoul(*dptr + matchoff, NULL, 10); - if (!cseq && *(*dptr + matchoff) != '0') { + cseq = sip_strtouint(*dptr + matchoff, *datalen - matchoff, (char **)&end); + if (*dptr + matchoff == end) { nf_ct_helper_log(skb, ct, "cannot get cseq"); return NF_DROP; } @@ -1575,7 +1659,7 @@ static int sip_help_tcp(struct sk_buff *skb, unsigned int protoff, &matchoff, &matchlen) <= 0) break; - clen = simple_strtoul(dptr + matchoff, (char **)&end, 10); + clen = sip_strtouint(dptr + matchoff, datalen - matchoff, (char **)&end); if (dptr + matchoff == end) break; diff --git a/net/netfilter/nf_nat_sip.c b/net/netfilter/nf_nat_sip.c index c845b6d1a2bd..9fbfc6bff0c2 100644 --- a/net/netfilter/nf_nat_sip.c +++ b/net/netfilter/nf_nat_sip.c @@ -246,6 +246,7 @@ static unsigned int nf_nat_sip(struct sk_buff *skb, unsigned int protoff, if (ct_sip_parse_numerical_param(ct, *dptr, matchend, *datalen, "rport=", &poff, &plen, &n) > 0 && + n >= 1024 && n <= 65535 && htons(n) == ct->tuplehash[dir].tuple.dst.u.udp.port && htons(n) != ct->tuplehash[!dir].tuple.src.u.udp.port) { __be16 p = ct->tuplehash[!dir].tuple.src.u.udp.port; -- cgit v1.2.3 From b5c111f4967ba4fdecdd318923ec7b081e9ef95f Mon Sep 17 00:00:00 2001 From: Amery Hung Date: Thu, 23 Apr 2026 15:23:55 -0700 Subject: bpf: Fix sk_local_storage diag dumping uninitialized special fields Call check_and_init_map_value() after the copy_map_value() to zero out special field regions. diag_get() copies sk_local_storage map values into a netlink message using copy_map_value{_locked}(), which intentionally skip special fields. However, the destination buffer from nla_reserve_64bit() is not zeroed and the skipped regions contain uninitialized skb data can be sent to userspace. Fixes: 1ed4d92458a9 ("bpf: INET_DIAG support in bpf_sk_storage") Signed-off-by: Amery Hung Signed-off-by: Martin KaFai Lau Link: https://patch.msgid.link/20260423222356.155387-1-ameryhung@gmail.com --- net/core/bpf_sk_storage.c | 1 + 1 file changed, 1 insertion(+) (limited to 'net') diff --git a/net/core/bpf_sk_storage.c b/net/core/bpf_sk_storage.c index dc3e8fce8809..ecd659f79fd4 100644 --- a/net/core/bpf_sk_storage.c +++ b/net/core/bpf_sk_storage.c @@ -557,6 +557,7 @@ static int diag_get(struct bpf_local_storage_map *smap, sdata->data, true); else copy_map_value(&smap->map, nla_data(nla_value), sdata->data); + check_and_init_map_value(&smap->map, nla_data(nla_value)); nla_nest_end(skb, nla_stg); return 0; -- cgit v1.2.3 From 711a9c018ad252b2807f85d44e1267b595644f9b Mon Sep 17 00:00:00 2001 From: Rio Liu Date: Wed, 15 Apr 2026 16:57:13 +0000 Subject: wifi: mac80211: skip ieee80211_verify_sta_ht_mcs_support check in non-strict mode Some Xfinity XB8 firmware advertises >1 spatial stream MCS indexes in their basic HT-MCS set. On cards with lower spatial streams, the check would fail, and we'd be stuck with no HT when in fact work fine with its own supported rate. This change makes it so the check is only performed in strict mode. Fixes: 574faa0e936d ("wifi: mac80211: add HT and VHT basic set verification") Signed-off-by: Rio Liu Link: https://patch.msgid.link/99Mv9QEceyPrQhSP52MtAVmz0_kWJmzqotJjD9YW6LGLqk-AZloAueUyHCURilFkuqOh6Ecv8i2KKdSE1ujP3AnbU5QEouVisT1w_V3xdfc=@r26.me Signed-off-by: Johannes Berg --- net/mac80211/mlme.c | 9 +++++++++ 1 file changed, 9 insertions(+) (limited to 'net') diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c index 160ae65a5c64..298ebff6bbf8 100644 --- a/net/mac80211/mlme.c +++ b/net/mac80211/mlme.c @@ -437,6 +437,15 @@ ieee80211_verify_sta_ht_mcs_support(struct ieee80211_sub_if_data *sdata, memcpy(&sta_ht_cap, &sband->ht_cap, sizeof(sta_ht_cap)); ieee80211_apply_htcap_overrides(sdata, &sta_ht_cap); + /* + * Some Xfinity XB8 firmware advertises >1 spatial stream MCS indexes in + * their basic HT-MCS set. On cards with lower spatial streams, the check + * would fail, and we'd be stuck with no HT when it in fact work fine with + * its own supported rate. So check it only in strict mode. + */ + if (!ieee80211_hw_check(&sdata->local->hw, STRICT)) + return true; + /* * P802.11REVme/D7.0 - 6.5.4.2.4 * ... -- cgit v1.2.3 From 381cd547bc6e35a610c5dfebe554d891eea40f03 Mon Sep 17 00:00:00 2001 From: Michael Bommarito Date: Tue, 21 Apr 2026 18:45:52 -0400 Subject: wifi: nl80211: require admin perm on SET_PMK / DEL_PMK NL80211_CMD_SET_PMK and NL80211_CMD_DEL_PMK manage the offloaded 4-way-handshake PMK state used by drivers advertising NL80211_EXT_FEATURE_4WAY_HANDSHAKE_STA_1X. The only in-tree driver that wires up both ->set_pmk / ->del_pmk and advertises the feature today is brcmfmac, so the practical reach of this patch is narrow. Both ops were introduced without a .flags gate, so the generic netlink layer dispatches them to an unprivileged caller instead of rejecting with -EPERM at the permission check. Every other connection-state op in the adjacent block (CONNECT, ASSOCIATE, AUTHENTICATE, SET_KEY, ...) carries GENL_UNS_ADMIN_PERM; SET_PMK / DEL_PMK were introduced without the flag in 2017 and left unchanged by later refactors. Johannes checked the original Intel submission history and confirmed there is no admin check in any prior revision either, so this seems likely to be a simple oversight rather than an intentional carve-out. Require GENL_UNS_ADMIN_PERM so the genl layer performs the same capable(CAP_NET_ADMIN) check as its siblings. wpa_supplicant already needs CAP_NET_ADMIN for every other nl80211 op it issues, so supplicant operation is unaffected. The worst case the missing gate enables today is an unprivileged local process on a multi-user system invalidating the offloaded PMK state of another user's 4-way-handshake session, forcing a full EAP re-auth on the next reconnect. Verified in UML: an unprivileged probe (uid=1000) sees SET_MULTICAST_TO_UNICAST (sibling op with GENL_UNS_ADMIN_PERM) return -EPERM on both pre- and post-fix kernels, while SET_PMK / DEL_PMK return -ENODEV from nl80211_pre_doit()'s wdev lookup pre- fix (proving dispatch crossed the genl permission check) and -EPERM post-fix (rejected at the genl layer as intended). Suggested-by: Johannes Berg Fixes: 3a00df5707b6 ("cfg80211: support 4-way handshake offloading for 802.1X") Assisted-by: Claude:claude-opus-4-7 Signed-off-by: Michael Bommarito Acked-by: Arend van Spriel Link: https://patch.msgid.link/20260421224552.4044147-1-michael.bommarito@gmail.com Signed-off-by: Johannes Berg --- net/wireless/nl80211.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'net') diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index f334cdef8958..67088804dcc7 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -19828,6 +19828,7 @@ static const struct genl_small_ops nl80211_small_ops[] = { .cmd = NL80211_CMD_SET_PMK, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_set_pmk, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = IFLAGS(NL80211_FLAG_NEED_NETDEV_UP | NL80211_FLAG_CLEAR_SKB), }, @@ -19835,6 +19836,7 @@ static const struct genl_small_ops nl80211_small_ops[] = { .cmd = NL80211_CMD_DEL_PMK, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_del_pmk, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = IFLAGS(NL80211_FLAG_NEED_NETDEV_UP), }, { -- cgit v1.2.3 From 9b55d5c1f5e481e391957f9096d798ca331c461b Mon Sep 17 00:00:00 2001 From: Michael Bommarito Date: Tue, 21 Apr 2026 20:06:51 -0400 Subject: wifi: mac80211: check ieee80211_rx_data_set_link return in pubsta MLO path __ieee80211_rx_handle_packet() resolves the link via ieee80211_rx_data_set_link() on the pubsta->mlo path but ignores the helper's return value. Inside the helper, rx->link = rcu_dereference(rx->sdata->link[link_id]); can leave rx->link NULL if link_id references a slot already cleared by ieee80211_vif_set_links() during station-initiated ML reconfiguration (see mlme.c's ieee80211_ml_reconfiguration(), which invalidates sdata->link[] before the matching ieee80211_sta_remove_link() loop walks the link-sta hash). RX dispatch still resolves a link_sta from the hash and then drops into ieee80211_prepare_and_rx_handle(), which dereferences link->conf->addr. Every other user site of ieee80211_rx_data_set_link() checks the return and bails on failure; only this branch did not. Mirror the safe pattern. Fixes: e66b7920aa5a ("wifi: mac80211: fix initialization of rx->link and rx->link_sta") Assisted-by: Claude:claude-opus-4-7 Signed-off-by: Michael Bommarito Link: https://patch.msgid.link/20260422000651.4184602-1-michael.bommarito@gmail.com Signed-off-by: Johannes Berg --- net/mac80211/rx.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c index 3e5d1c47a5b0..5a92413a911f 100644 --- a/net/mac80211/rx.c +++ b/net/mac80211/rx.c @@ -5380,7 +5380,9 @@ static void __ieee80211_rx_handle_packet(struct ieee80211_hw *hw, if (!link_sta) goto out; - ieee80211_rx_data_set_link(&rx, link_sta->link_id); + if (!ieee80211_rx_data_set_link(&rx, + link_sta->link_id)) + goto out; } if (ieee80211_prepare_and_rx_handle(&rx, skb, true)) -- cgit v1.2.3 From 7a5b81e0c87a075afd572f659d8eb68c9c4cd2ba Mon Sep 17 00:00:00 2001 From: Catherine Date: Fri, 24 Apr 2026 21:14:36 +0800 Subject: wifi: mac80211: drop stray 'static' from fast-RX rx_result ieee80211_invoke_fast_rx() is documented as safe for parallel RX, but its per-invocation rx_result is declared static. Concurrent callers then share one instance and can overwrite each other's result between ieee80211_rx_mesh_data() and the switch on res. That can make a packet that was queued or consumed by ieee80211_rx_mesh_data() fall through into ieee80211_rx_8023(), or make a packet that should continue return as queued. Make res an automatic variable so each invocation keeps its own result. Fixes: 3468e1e0c639 ("wifi: mac80211: add mesh fast-rx support") Cc: stable@vger.kernel.org Signed-off-by: Catherine Link: https://patch.msgid.link/20260424131435.83212-2-enderaoelyther@gmail.com Signed-off-by: Johannes Berg --- net/mac80211/rx.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c index 5a92413a911f..d18e962126ce 100644 --- a/net/mac80211/rx.c +++ b/net/mac80211/rx.c @@ -4971,7 +4971,7 @@ static bool ieee80211_invoke_fast_rx(struct ieee80211_rx_data *rx, struct sk_buff *skb = rx->skb; struct ieee80211_hdr *hdr = (void *)skb->data; struct ieee80211_rx_status *status = IEEE80211_SKB_RXCB(skb); - static ieee80211_rx_result res; + ieee80211_rx_result res; int orig_len = skb->len; int hdrlen = ieee80211_hdrlen(hdr->frame_control); int snap_offs = hdrlen; -- cgit v1.2.3 From 732b463449fd0ef90acd13cda68eab1c91adb00c Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Fri, 17 Apr 2026 20:19:39 -0700 Subject: net/sched: netem: fix probability gaps in 4-state loss model The 4-state Markov chain in loss_4state() has gaps at the boundaries between transition probability ranges. The comparisons use: if (rnd < a4) else if (a4 < rnd && rnd < a1 + a4) When rnd equals a boundary value exactly, neither branch matches and no state transition occurs. The redundant lower-bound check (a4 < rnd) is already implied by being in the else branch. Remove the unnecessary lower-bound comparisons so the ranges are contiguous and every random value produces a transition, matching the GI (General and Intuitive) loss model specification. This bug goes back to original implementation of this model. Fixes: 661b79725fea ("netem: revised correlated loss generator") Signed-off-by: Stephen Hemminger Reviewed-by: Simon Horman Link: https://patch.msgid.link/20260418032027.900913-2-stephen@networkplumber.org Signed-off-by: Jakub Kicinski --- net/sched/sch_netem.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c index 20df1c08b1e9..8ee72cac1faf 100644 --- a/net/sched/sch_netem.c +++ b/net/sched/sch_netem.c @@ -227,10 +227,10 @@ static bool loss_4state(struct netem_sched_data *q) if (rnd < clg->a4) { clg->state = LOST_IN_GAP_PERIOD; return true; - } else if (clg->a4 < rnd && rnd < clg->a1 + clg->a4) { + } else if (rnd < clg->a1 + clg->a4) { clg->state = LOST_IN_BURST_PERIOD; return true; - } else if (clg->a1 + clg->a4 < rnd) { + } else { clg->state = TX_IN_GAP_PERIOD; } @@ -247,9 +247,9 @@ static bool loss_4state(struct netem_sched_data *q) case LOST_IN_BURST_PERIOD: if (rnd < clg->a3) clg->state = TX_IN_BURST_PERIOD; - else if (clg->a3 < rnd && rnd < clg->a2 + clg->a3) { + else if (rnd < clg->a2 + clg->a3) { clg->state = TX_IN_GAP_PERIOD; - } else if (clg->a2 + clg->a3 < rnd) { + } else { clg->state = LOST_IN_BURST_PERIOD; return true; } -- cgit v1.2.3 From 4185701fcce6b426b6c3630b25330dddd9c47b0d Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Fri, 17 Apr 2026 20:19:40 -0700 Subject: net/sched: netem: fix queue limit check to include reordered packets The queue limit check in netem_enqueue() uses q->t_len which only counts packets in the internal tfifo. Packets placed in sch->q by the reorder path (__qdisc_enqueue_head) are not counted, allowing the total queue occupancy to exceed sch->limit under reordering. Include sch->q.qlen in the limit check. Fixes: f8d4bc455047 ("net/sched: netem: account for backlog updates from child qdisc") Signed-off-by: Stephen Hemminger Reviewed-by: Simon Horman Link: https://patch.msgid.link/20260418032027.900913-3-stephen@networkplumber.org Signed-off-by: Jakub Kicinski --- net/sched/sch_netem.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c index 8ee72cac1faf..d400a730eadd 100644 --- a/net/sched/sch_netem.c +++ b/net/sched/sch_netem.c @@ -524,7 +524,7 @@ static int netem_enqueue(struct sk_buff *skb, struct Qdisc *sch, 1 << get_random_u32_below(8); } - if (unlikely(q->t_len >= sch->limit)) { + if (unlikely(sch->q.qlen >= sch->limit)) { /* re-link segs, so that qdisc_drop_all() frees them all */ skb->next = segs; qdisc_drop_all(skb, sch, to_free); -- cgit v1.2.3 From 986afaf809940577224a99c3a08d97a15eb37e93 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Fri, 17 Apr 2026 20:19:41 -0700 Subject: net/sched: netem: only reseed PRNG when seed is explicitly provided netem_change() unconditionally reseeds the PRNG on every tc change command. If TCA_NETEM_PRNG_SEED is not specified, a new random seed is generated, destroying reproducibility for users who set a deterministic seed on a previous change. Move the initial random seed generation to netem_init() and only reseed in netem_change() when TCA_NETEM_PRNG_SEED is explicitly provided by the user. Fixes: 4072d97ddc44 ("netem: add prng attribute to netem_sched_data") Signed-off-by: Stephen Hemminger Reviewed-by: Simon Horman Link: https://patch.msgid.link/20260418032027.900913-4-stephen@networkplumber.org Signed-off-by: Jakub Kicinski --- net/sched/sch_netem.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c index d400a730eadd..556f9747f0e7 100644 --- a/net/sched/sch_netem.c +++ b/net/sched/sch_netem.c @@ -1112,11 +1112,10 @@ static int netem_change(struct Qdisc *sch, struct nlattr *opt, /* capping jitter to the range acceptable by tabledist() */ q->jitter = min_t(s64, abs(q->jitter), INT_MAX); - if (tb[TCA_NETEM_PRNG_SEED]) + if (tb[TCA_NETEM_PRNG_SEED]) { q->prng.seed = nla_get_u64(tb[TCA_NETEM_PRNG_SEED]); - else - q->prng.seed = get_random_u64(); - prandom_seed_state(&q->prng.prng_state, q->prng.seed); + prandom_seed_state(&q->prng.prng_state, q->prng.seed); + } unlock: sch_tree_unlock(sch); @@ -1139,6 +1138,9 @@ static int netem_init(struct Qdisc *sch, struct nlattr *opt, return -EINVAL; q->loss_model = CLG_RANDOM; + q->prng.seed = get_random_u64(); + prandom_seed_state(&q->prng.prng_state, q->prng.seed); + ret = netem_change(sch, opt, extack); if (ret) pr_info("netem: change failed\n"); -- cgit v1.2.3 From 01801c359a74737b9b1aa28568b60374d857241a Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Fri, 17 Apr 2026 20:19:42 -0700 Subject: net/sched: netem: validate slot configuration Reject slot configurations that have no defensible meaning: - negative min_delay or max_delay - min_delay greater than max_delay - negative dist_delay or dist_jitter - negative max_packets or max_bytes Negative or out-of-order delays underflow in get_slot_next(), producing garbage intervals. Negative limits trip the per-slot accounting (packets_left/bytes_left <= 0) on the first packet of every slot, defeating the rate-limiting half of the slot feature. Note that dist_jitter has been silently coerced to its absolute value by get_slot() since the feature was introduced; rejecting negatives here converts that silent coercion into -EINVAL. The abs() can be removed in a follow-up. Fixes: 836af83b54e3 ("netem: support delivering packets in delayed time slots") Signed-off-by: Stephen Hemminger Reviewed-by: Simon Horman Link: https://patch.msgid.link/20260418032027.900913-5-stephen@networkplumber.org Signed-off-by: Jakub Kicinski --- net/sched/sch_netem.c | 29 +++++++++++++++++++++++++++++ 1 file changed, 29 insertions(+) (limited to 'net') diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c index 556f9747f0e7..640b51be807a 100644 --- a/net/sched/sch_netem.c +++ b/net/sched/sch_netem.c @@ -827,6 +827,29 @@ static int get_dist_table(struct disttable **tbl, const struct nlattr *attr) return 0; } +static int validate_slot(const struct nlattr *attr, struct netlink_ext_ack *extack) +{ + const struct tc_netem_slot *c = nla_data(attr); + + if (c->min_delay < 0 || c->max_delay < 0) { + NL_SET_ERR_MSG_ATTR(extack, attr, "negative slot delay"); + return -EINVAL; + } + if (c->min_delay > c->max_delay) { + NL_SET_ERR_MSG_ATTR(extack, attr, "slot min delay greater than max delay"); + return -EINVAL; + } + if (c->dist_delay < 0 || c->dist_jitter < 0) { + NL_SET_ERR_MSG_ATTR(extack, attr, "negative dist delay"); + return -EINVAL; + } + if (c->max_packets < 0 || c->max_bytes < 0) { + NL_SET_ERR_MSG_ATTR(extack, attr, "negative slot limit"); + return -EINVAL; + } + return 0; +} + static void get_slot(struct netem_sched_data *q, const struct nlattr *attr) { const struct tc_netem_slot *c = nla_data(attr); @@ -1040,6 +1063,12 @@ static int netem_change(struct Qdisc *sch, struct nlattr *opt, goto table_free; } + if (tb[TCA_NETEM_SLOT]) { + ret = validate_slot(tb[TCA_NETEM_SLOT], extack); + if (ret) + goto table_free; + } + sch_tree_lock(sch); /* backup q->clg and q->loss_model */ old_clg = q->clg; -- cgit v1.2.3 From 51e94e1e2fef351c74d69eb53666df808d26af95 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Fri, 17 Apr 2026 20:19:43 -0700 Subject: net/sched: netem: fix slot delay calculation overflow get_slot_next() computes a random delay between min_delay and max_delay using: get_random_u32() * (max_delay - min_delay) >> 32 This overflows signed 64-bit arithmetic when the delay range exceeds approximately 2.1 seconds (2^31 nanoseconds), producing a negative result that effectively disables slot-based pacing. This is a realistic configuration for WAN emulation (e.g., slot 1s 5s). Use mul_u64_u32_shr() which handles the widening multiply without overflow. Fixes: 0a9fe5c375b5 ("netem: slotting with non-uniform distribution") Signed-off-by: Stephen Hemminger Reviewed-by: Simon Horman Link: https://patch.msgid.link/20260418032027.900913-6-stephen@networkplumber.org Signed-off-by: Jakub Kicinski --- net/sched/sch_netem.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c index 640b51be807a..475c14b3dbdb 100644 --- a/net/sched/sch_netem.c +++ b/net/sched/sch_netem.c @@ -659,9 +659,8 @@ static void get_slot_next(struct netem_sched_data *q, u64 now) if (!q->slot_dist) next_delay = q->slot_config.min_delay + - (get_random_u32() * - (q->slot_config.max_delay - - q->slot_config.min_delay) >> 32); + mul_u64_u32_shr(q->slot_config.max_delay - q->slot_config.min_delay, + get_random_u32(), 32); else next_delay = tabledist(q->slot_config.dist_delay, (s32)(q->slot_config.dist_jitter), -- cgit v1.2.3 From 90be9fedb218ee95a1cf59050d1306fbfb0e8b87 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Fri, 17 Apr 2026 20:19:44 -0700 Subject: net/sched: netem: check for negative latency and jitter Reject requests with negative latency or jitter. A negative value added to current timestamp (u64) wraps to an enormous time_to_send, disabling dequeue. The original UAPI used u32 for these values; the conversion to 64-bit time values via TCA_NETEM_LATENCY64 and TCA_NETEM_JITTER64 allowed signed values to reach the kernel without validation. Jitter is already silently clamped by an abs() in netem_change(); that abs() can be removed in a follow-up once this rejection is in place. Fixes: 99803171ef04 ("netem: add uapi to express delay and jitter in nanoseconds") Signed-off-by: Stephen Hemminger Reviewed-by: Simon Horman Link: https://patch.msgid.link/20260418032027.900913-7-stephen@networkplumber.org Signed-off-by: Jakub Kicinski --- net/sched/sch_netem.c | 22 ++++++++++++++++++++++ 1 file changed, 22 insertions(+) (limited to 'net') diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c index 475c14b3dbdb..bc18e1976b6e 100644 --- a/net/sched/sch_netem.c +++ b/net/sched/sch_netem.c @@ -826,6 +826,16 @@ static int get_dist_table(struct disttable **tbl, const struct nlattr *attr) return 0; } +static int validate_time(const struct nlattr *attr, const char *name, + struct netlink_ext_ack *extack) +{ + if (nla_get_s64(attr) < 0) { + NL_SET_ERR_MSG_ATTR_FMT(extack, attr, "negative %s", name); + return -EINVAL; + } + return 0; +} + static int validate_slot(const struct nlattr *attr, struct netlink_ext_ack *extack) { const struct tc_netem_slot *c = nla_data(attr); @@ -1068,6 +1078,18 @@ static int netem_change(struct Qdisc *sch, struct nlattr *opt, goto table_free; } + if (tb[TCA_NETEM_LATENCY64]) { + ret = validate_time(tb[TCA_NETEM_LATENCY64], "latency", extack); + if (ret) + goto table_free; + } + + if (tb[TCA_NETEM_JITTER64]) { + ret = validate_time(tb[TCA_NETEM_JITTER64], "jitter", extack); + if (ret) + goto table_free; + } + sch_tree_lock(sch); /* backup q->clg and q->loss_model */ old_clg = q->clg; -- cgit v1.2.3 From d3aeb889dcbd78e95f500d383799a23d949796e0 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 23 Apr 2026 06:28:39 +0000 Subject: net/sched: sch_choke: annotate data-races in choke_dump_stats() choke_dump_stats() only runs with RTNL held. It reads fields that can be changed in qdisc fast path. Add READ_ONCE()/WRITE_ONCE() annotations. Fixes: edb09eb17ed8 ("net: sched: do not acquire qdisc spinlock in qdisc/class stats dump") Signed-off-by: Eric Dumazet Reviewed-by: Jamal Hadi Salim Link: https://patch.msgid.link/20260423062839.2524324-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/sched/sch_choke.c | 26 ++++++++++++++++---------- 1 file changed, 16 insertions(+), 10 deletions(-) (limited to 'net') diff --git a/net/sched/sch_choke.c b/net/sched/sch_choke.c index 94df8e741a97..2875bcdb18a4 100644 --- a/net/sched/sch_choke.c +++ b/net/sched/sch_choke.c @@ -229,7 +229,7 @@ static int choke_enqueue(struct sk_buff *skb, struct Qdisc *sch, /* Draw a packet at random from queue and compare flow */ if (choke_match_random(q, skb, &idx)) { - q->stats.matched++; + WRITE_ONCE(q->stats.matched, q->stats.matched + 1); choke_drop_by_idx(sch, idx, to_free); goto congestion_drop; } @@ -241,11 +241,13 @@ static int choke_enqueue(struct sk_buff *skb, struct Qdisc *sch, qdisc_qstats_overlimit(sch); if (use_harddrop(q) || !use_ecn(q) || !INET_ECN_set_ce(skb)) { - q->stats.forced_drop++; + WRITE_ONCE(q->stats.forced_drop, + q->stats.forced_drop + 1); goto congestion_drop; } - q->stats.forced_mark++; + WRITE_ONCE(q->stats.forced_mark, + q->stats.forced_mark + 1); } else if (++q->vars.qcount) { if (red_mark_probability(p, &q->vars, q->vars.qavg)) { q->vars.qcount = 0; @@ -253,11 +255,13 @@ static int choke_enqueue(struct sk_buff *skb, struct Qdisc *sch, qdisc_qstats_overlimit(sch); if (!use_ecn(q) || !INET_ECN_set_ce(skb)) { - q->stats.prob_drop++; + WRITE_ONCE(q->stats.prob_drop, + q->stats.prob_drop + 1); goto congestion_drop; } - q->stats.prob_mark++; + WRITE_ONCE(q->stats.prob_mark, + q->stats.prob_mark + 1); } } else q->vars.qR = red_random(p); @@ -272,7 +276,7 @@ static int choke_enqueue(struct sk_buff *skb, struct Qdisc *sch, return NET_XMIT_SUCCESS; } - q->stats.pdrop++; + WRITE_ONCE(q->stats.pdrop, q->stats.pdrop + 1); return qdisc_drop(skb, sch, to_free); congestion_drop: @@ -461,10 +465,12 @@ static int choke_dump_stats(struct Qdisc *sch, struct gnet_dump *d) { struct choke_sched_data *q = qdisc_priv(sch); struct tc_choke_xstats st = { - .early = q->stats.prob_drop + q->stats.forced_drop, - .marked = q->stats.prob_mark + q->stats.forced_mark, - .pdrop = q->stats.pdrop, - .matched = q->stats.matched, + .early = READ_ONCE(q->stats.prob_drop) + + READ_ONCE(q->stats.forced_drop), + .marked = READ_ONCE(q->stats.prob_mark) + + READ_ONCE(q->stats.forced_mark), + .pdrop = READ_ONCE(q->stats.pdrop), + .matched = READ_ONCE(q->stats.matched), }; return gnet_stats_copy_app(d, &st, sizeof(st)); -- cgit v1.2.3 From 59b145771c7982cfe9020d4e9e22da92d6b5ae31 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 23 Apr 2026 06:35:27 +0000 Subject: net/sched: sch_fq_pie: annotate data-races in fq_pie_dump_stats() fq_codel_dump_stats() acquires the qdisc spinlock a bit too late. Move this acquisition before we fill tc_fq_pie_xstats with live data. Alternative would be to add READ_ONCE() and WRITE_ONCE() annotations, but the spinlock is needed anyway to scan q->new_flows and q->old_flows. Fixes: ec97ecf1ebe4 ("net: sched: add Flow Queue PIE packet scheduler") Signed-off-by: Eric Dumazet Reviewed-by: Jamal Hadi Salim Link: https://patch.msgid.link/20260423063527.2568262-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/sched/sch_fq_pie.c | 19 ++++++++++--------- 1 file changed, 10 insertions(+), 9 deletions(-) (limited to 'net') diff --git a/net/sched/sch_fq_pie.c b/net/sched/sch_fq_pie.c index 154c70f489f2..7becbf5362b3 100644 --- a/net/sched/sch_fq_pie.c +++ b/net/sched/sch_fq_pie.c @@ -509,18 +509,19 @@ nla_put_failure: static int fq_pie_dump_stats(struct Qdisc *sch, struct gnet_dump *d) { struct fq_pie_sched_data *q = qdisc_priv(sch); - struct tc_fq_pie_xstats st = { - .packets_in = q->stats.packets_in, - .overlimit = q->stats.overlimit, - .overmemory = q->overmemory, - .dropped = q->stats.dropped, - .ecn_mark = q->stats.ecn_mark, - .new_flow_count = q->new_flow_count, - .memory_usage = q->memory_usage, - }; + struct tc_fq_pie_xstats st = { 0 }; struct list_head *pos; sch_tree_lock(sch); + + st.packets_in = q->stats.packets_in; + st.overlimit = q->stats.overlimit; + st.overmemory = q->overmemory; + st.dropped = q->stats.dropped; + st.ecn_mark = q->stats.ecn_mark; + st.new_flow_count = q->new_flow_count; + st.memory_usage = q->memory_usage; + list_for_each(pos, &q->new_flows) st.new_flows_len++; -- cgit v1.2.3 From 9e6bf146b55999a095bb14f73a843942456d1adc Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Tue, 21 Apr 2026 15:16:33 +0200 Subject: ipv6: rpl: reserve mac_len headroom when recompressed SRH grows ipv6_rpl_srh_rcv() decompresses an RFC 6554 Source Routing Header, swaps the next segment into ipv6_hdr->daddr, recompresses, then pulls the old header and pushes the new one plus the IPv6 header back. The recompressed header can be larger than the received one when the swap reduces the common-prefix length the segments share with daddr (CmprI=0, CmprE>0, seg[0][0] != daddr[0] gives the maximum +8 bytes). pskb_expand_head() was gated on segments_left == 0, so on earlier segments the push consumed unchecked headroom. Once skb_push() leaves fewer than skb->mac_len bytes in front of data, skb_mac_header_rebuild()'s call to: skb_set_mac_header(skb, -skb->mac_len); will store (data - head) - mac_len into the u16 mac_header field, which wraps to ~65530, and the following memmove() writes mac_len bytes ~64KiB past skb->head. A single AF_INET6/SOCK_RAW/IPV6_HDRINCL packet over lo with a two segment type-3 SRH (CmprI=0, CmprE=15) reaches headroom 8 after one pass; KASAN reports a 14-byte OOB write in ipv6_rthdr_rcv. Fix this by expanding the head whenever the remaining room is less than the push size plus mac_len, and request that much extra so the rebuilt MAC header fits afterwards. Fixes: 8610c7c6e3bd ("net: ipv6: add support for rpl sr exthdr") Cc: stable Reported-by: Anthropic Signed-off-by: Greg Kroah-Hartman Link: https://patch.msgid.link/2026042133-gout-unvented-1bd9@gregkh Signed-off-by: Jakub Kicinski --- net/ipv6/exthdrs.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/ipv6/exthdrs.c b/net/ipv6/exthdrs.c index 95558fd6f447..03cbce842c1a 100644 --- a/net/ipv6/exthdrs.c +++ b/net/ipv6/exthdrs.c @@ -491,6 +491,7 @@ static int ipv6_rpl_srh_rcv(struct sk_buff *skb) struct net *net = dev_net(skb->dev); struct inet6_dev *idev; struct ipv6hdr *oldhdr; + unsigned int chdr_len; unsigned char *buf; int accept_rpl_seg; int i, err; @@ -592,8 +593,10 @@ looped_back: skb_pull(skb, ((hdr->hdrlen + 1) << 3)); skb_postpull_rcsum(skb, oldhdr, sizeof(struct ipv6hdr) + ((hdr->hdrlen + 1) << 3)); - if (unlikely(!hdr->segments_left)) { - if (pskb_expand_head(skb, sizeof(struct ipv6hdr) + ((chdr->hdrlen + 1) << 3), 0, + chdr_len = sizeof(struct ipv6hdr) + ((chdr->hdrlen + 1) << 3); + if (unlikely(!hdr->segments_left || + skb_headroom(skb) < chdr_len + skb->mac_len)) { + if (pskb_expand_head(skb, chdr_len + skb->mac_len, 0, GFP_ATOMIC)) { __IP6_INC_STATS(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_OUTDISCARDS); kfree_skb(skb); @@ -603,7 +606,7 @@ looped_back: oldhdr = ipv6_hdr(skb); } - skb_push(skb, ((chdr->hdrlen + 1) << 3) + sizeof(struct ipv6hdr)); + skb_push(skb, chdr_len); skb_reset_network_header(skb); skb_mac_header_rebuild(skb); skb_set_transport_header(skb, sizeof(struct ipv6hdr)); -- cgit v1.2.3 From 3d07ca5c0fae311226f737963984bd94bb159a87 Mon Sep 17 00:00:00 2001 From: Weiming Shi Date: Thu, 23 Apr 2026 00:19:58 +0800 Subject: net/sched: taprio: fix NULL pointer dereference in class dump When a TAPRIO child qdisc is deleted via RTM_DELQDISC, taprio_graft() is called with new == NULL and stores NULL into q->qdiscs[cl - 1]. Subsequent RTM_GETTCLASS dump operations walk all classes via taprio_walk() and call taprio_dump_class(), which calls taprio_leaf() returning the NULL pointer, then dereferences it to read child->handle, causing a kernel NULL pointer dereference. The bug is reachable with namespace-scoped CAP_NET_ADMIN on any kernel with CONFIG_NET_SCH_TAPRIO enabled. On systems with unprivileged user namespaces enabled, an unprivileged local user can trigger a kernel panic by creating a taprio qdisc inside a new network namespace, grafting an explicit child qdisc, deleting it, and requesting a class dump. The RTM_GETTCLASS dump itself requires no capability. Oops: general protection fault, probably for non-canonical address 0xdffffc0000000007: 0000 [#1] SMP KASAN NOPTI KASAN: null-ptr-deref in range [0x0000000000000038-0x000000000000003f] RIP: 0010:taprio_dump_class (net/sched/sch_taprio.c:2478) Call Trace: tc_fill_tclass (net/sched/sch_api.c:1966) qdisc_class_dump (net/sched/sch_api.c:2326) taprio_walk (net/sched/sch_taprio.c:2514) tc_dump_tclass_qdisc (net/sched/sch_api.c:2352) tc_dump_tclass_root (net/sched/sch_api.c:2370) tc_dump_tclass (net/sched/sch_api.c:2431) rtnl_dumpit (net/core/rtnetlink.c:6864) netlink_dump (net/netlink/af_netlink.c:2325) rtnetlink_rcv_msg (net/core/rtnetlink.c:6959) netlink_rcv_skb (net/netlink/af_netlink.c:2550) Fix this by substituting &noop_qdisc when new is NULL in taprio_graft(), a common pattern used by other qdiscs (e.g., multiq_graft()) to ensure the q->qdiscs[] slots are never NULL. This makes control-plane dump paths safe without requiring individual NULL checks. Since the data-plane paths (taprio_enqueue and taprio_dequeue_from_txq) previously had explicit NULL guards that would drop/skip the packet cleanly, update those checks to test for &noop_qdisc instead. Without this, packets would reach taprio_enqueue_one() which increments the root qdisc's qlen and backlog before calling the child's enqueue; noop_qdisc drops the packet but those counters are never rolled back, permanently inflating the root qdisc's statistics. After this change *old can be a valid qdisc, NULL, or &noop_qdisc. Only call qdisc_put(*old) in the first case to avoid decreasing noop_qdisc's refcount, which was never increased. Fixes: 665338b2a7a0 ("net/sched: taprio: dump class stats for the actual q->qdiscs[]") Reported-by: Xiang Mei Signed-off-by: Weiming Shi Acked-by: Jamal Hadi Salim Tested-by: Weiming Shi Link: https://patch.msgid.link/20260422161958.2517539-3-bestswngs@gmail.com Signed-off-by: Jakub Kicinski --- net/sched/sch_taprio.c | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/sched/sch_taprio.c b/net/sched/sch_taprio.c index a47a09d76400..45245157e00a 100644 --- a/net/sched/sch_taprio.c +++ b/net/sched/sch_taprio.c @@ -634,7 +634,7 @@ static int taprio_enqueue(struct sk_buff *skb, struct Qdisc *sch, queue = skb_get_queue_mapping(skb); child = q->qdiscs[queue]; - if (unlikely(!child)) + if (unlikely(child == &noop_qdisc)) return qdisc_drop(skb, sch, to_free); if (taprio_skb_exceeds_queue_max_sdu(sch, skb)) { @@ -717,7 +717,7 @@ static struct sk_buff *taprio_dequeue_from_txq(struct Qdisc *sch, int txq, int len; u8 tc; - if (unlikely(!child)) + if (unlikely(child == &noop_qdisc)) return NULL; if (TXTIME_ASSIST_IS_ENABLED(q->flags)) @@ -2184,6 +2184,9 @@ static int taprio_graft(struct Qdisc *sch, unsigned long cl, if (!dev_queue) return -EINVAL; + if (!new) + new = &noop_qdisc; + if (dev->flags & IFF_UP) dev_deactivate(dev, false); @@ -2197,14 +2200,14 @@ static int taprio_graft(struct Qdisc *sch, unsigned long cl, *old = q->qdiscs[cl - 1]; if (FULL_OFFLOAD_IS_ENABLED(q->flags)) { WARN_ON_ONCE(dev_graft_qdisc(dev_queue, new) != *old); - if (new) + if (new != &noop_qdisc) qdisc_refcount_inc(new); - if (*old) + if (*old && *old != &noop_qdisc) qdisc_put(*old); } q->qdiscs[cl - 1] = new; - if (new) + if (new != &noop_qdisc) new->flags |= TCQ_F_ONETXQUEUE | TCQ_F_NOPARENT; if (dev->flags & IFF_UP) -- cgit v1.2.3 From 5b0c911bcdbd982f7748d11c0b39ec5808eae2de Mon Sep 17 00:00:00 2001 From: Morduan Zang Date: Thu, 23 Apr 2026 09:05:57 +0800 Subject: net: phonet: do not BUG_ON() in pn_socket_autobind() on failed bind MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit syzbot reported a kernel BUG triggered from pn_socket_sendmsg() via pn_socket_autobind(): kernel BUG at net/phonet/socket.c:213! RIP: 0010:pn_socket_autobind net/phonet/socket.c:213 [inline] RIP: 0010:pn_socket_sendmsg+0x240/0x250 net/phonet/socket.c:421 Call Trace: sock_sendmsg_nosec+0x112/0x150 net/socket.c:797 __sock_sendmsg net/socket.c:812 [inline] __sys_sendto+0x402/0x590 net/socket.c:2280 ... pn_socket_autobind() calls pn_socket_bind() with port 0 and, on -EINVAL, assumes the socket was already bound and asserts that the port is non-zero: err = pn_socket_bind(sock, ..., sizeof(struct sockaddr_pn)); if (err != -EINVAL) return err; BUG_ON(!pn_port(pn_sk(sock->sk)->sobject)); return 0; /* socket was already bound */ However pn_socket_bind() also returns -EINVAL when sk->sk_state is not TCP_CLOSE, even when the socket has never been bound and pn_port() is still 0. In that case the BUG_ON() fires and panics the kernel from a user-triggerable path. Treat the "bind returned -EINVAL but pn_port() is still 0" case as a regular error and propagate -EINVAL to the caller instead of crashing. Existing callers already translate a non-zero return from pn_socket_autobind() into -ENOBUFS/-EAGAIN, so returning -EINVAL here only changes behaviour from panic to a normal errno. Fixes: ba113a94b750 ("Phonet: common socket glue") Reported-by: syzbot+706f5eb79044e686c794@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=706f5eb79044e686c794 Suggested-by: Remi Denis-Courmont Signed-off-by: Morduan Zang Signed-off-by: zhanjun Acked-by: Rémi Denis-Courmont Link: https://patch.msgid.link/87A8960A2045AF3C+20260423010557.138124-1-zhangdandan@uniontech.com Signed-off-by: Jakub Kicinski --- net/phonet/socket.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/phonet/socket.c b/net/phonet/socket.c index c4af26357144..631a99cdbd00 100644 --- a/net/phonet/socket.c +++ b/net/phonet/socket.c @@ -208,9 +208,15 @@ static int pn_socket_autobind(struct socket *sock) sa.spn_family = AF_PHONET; err = pn_socket_bind(sock, (struct sockaddr_unsized *)&sa, sizeof(struct sockaddr_pn)); - if (err != -EINVAL) + /* + * pn_socket_bind() also returns -EINVAL when sk_state != TCP_CLOSE + * without a prior bind, so -EINVAL alone is not sufficient to infer + * that the socket was already bound. Only treat it as "already + * bound" when the port is non-zero; otherwise propagate the error + * instead of crashing the kernel. + */ + if (err != -EINVAL || unlikely(!pn_port(pn_sk(sock->sk)->sobject))) return err; - BUG_ON(!pn_port(pn_sk(sock->sk)->sobject)); return 0; /* socket was already bound */ } -- cgit v1.2.3 From b3b6babf47517fde6b6de2493dea28e8831b9347 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Thu, 23 Apr 2026 05:34:54 +0000 Subject: ipmr: Free mr_table after RCU grace period. With CONFIG_IP_MROUTE_MULTIPLE_TABLES=n, ipmr_fib_lookup() does not check if net->ipv4.mrt is NULL. Since default_device_exit_batch() is called after ->exit_rtnl(), a device could receive IGMP packets and access net->ipv4.mrt during/after ipmr_rules_exit_rtnl(). If ipmr_rules_exit_rtnl() had already cleared it and freed the memory, the access would trigger null-ptr-deref or use-after-free. Let's fix it by using RCU helper and free mrt after RCU grace period. In addition, check_net(net) is added to mroute_clean_tables() and ipmr_cache_unresolved() to synchronise via mfc_unres_lock. This prevents ipmr_cache_unresolved() from putting skb into c->_c.mfc_un.unres.unresolved after mroute_clean_tables() purges it. For the same reason, timer_shutdown_sync() is moved after mroute_clean_tables(). Since rhltable_destroy() holds mutex internally, rcu_work is used, and it is placed as the first member because rcu_head must be placed within <4K offset. mr_table is alraedy 3864 bytes without rcu_work. Note that IP6MR is not yet converted to ->exit_rtnl(), so this change is not needed for now but will be. Fixes: b22b01867406 ("ipmr: Convert ipmr_net_exit_batch() to ->exit_rtnl().") Signed-off-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20260423053456.4097409-1-kuniyu@google.com Signed-off-by: Jakub Kicinski --- net/ipv4/ipmr.c | 108 +++++++++++++++++++++++++++------------------------ net/ipv4/ipmr_base.c | 16 ++++++++ 2 files changed, 74 insertions(+), 50 deletions(-) (limited to 'net') diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c index 8a08d09b4c30..2058ca860294 100644 --- a/net/ipv4/ipmr.c +++ b/net/ipv4/ipmr.c @@ -151,16 +151,6 @@ static struct mr_table *__ipmr_get_table(struct net *net, u32 id) return NULL; } -static struct mr_table *ipmr_get_table(struct net *net, u32 id) -{ - struct mr_table *mrt; - - rcu_read_lock(); - mrt = __ipmr_get_table(net, id); - rcu_read_unlock(); - return mrt; -} - static int ipmr_fib_lookup(struct net *net, struct flowi4 *flp4, struct mr_table **mrt) { @@ -293,7 +283,7 @@ static void __net_exit ipmr_rules_exit_rtnl(struct net *net, struct mr_table *mrt, *next; list_for_each_entry_safe(mrt, next, &net->ipv4.mr_tables, list) { - list_del(&mrt->list); + list_del_rcu(&mrt->list); ipmr_free_table(mrt, dev_kill_list); } } @@ -315,28 +305,30 @@ bool ipmr_rule_default(const struct fib_rule *rule) } EXPORT_SYMBOL(ipmr_rule_default); #else -#define ipmr_for_each_table(mrt, net) \ - for (mrt = net->ipv4.mrt; mrt; mrt = NULL) - static struct mr_table *ipmr_mr_table_iter(struct net *net, struct mr_table *mrt) { if (!mrt) - return net->ipv4.mrt; + return rcu_dereference(net->ipv4.mrt); return NULL; } -static struct mr_table *ipmr_get_table(struct net *net, u32 id) +static struct mr_table *__ipmr_get_table(struct net *net, u32 id) { - return net->ipv4.mrt; + return rcu_dereference_check(net->ipv4.mrt, + lockdep_rtnl_is_held() || + !rcu_access_pointer(net->ipv4.mrt)); } -#define __ipmr_get_table ipmr_get_table +#define ipmr_for_each_table(mrt, net) \ + for (mrt = __ipmr_get_table(net, 0); mrt; mrt = NULL) static int ipmr_fib_lookup(struct net *net, struct flowi4 *flp4, struct mr_table **mrt) { - *mrt = net->ipv4.mrt; + *mrt = rcu_dereference(net->ipv4.mrt); + if (!*mrt) + return -EAGAIN; return 0; } @@ -347,7 +339,8 @@ static int __net_init ipmr_rules_init(struct net *net) mrt = ipmr_new_table(net, RT_TABLE_DEFAULT); if (IS_ERR(mrt)) return PTR_ERR(mrt); - net->ipv4.mrt = mrt; + + rcu_assign_pointer(net->ipv4.mrt, mrt); return 0; } @@ -358,9 +351,10 @@ static void __net_exit ipmr_rules_exit(struct net *net) static void __net_exit ipmr_rules_exit_rtnl(struct net *net, struct list_head *dev_kill_list) { - ipmr_free_table(net->ipv4.mrt, dev_kill_list); + struct mr_table *mrt = rcu_dereference_protected(net->ipv4.mrt, 1); - net->ipv4.mrt = NULL; + RCU_INIT_POINTER(net->ipv4.mrt, NULL); + ipmr_free_table(mrt, dev_kill_list); } static int ipmr_rules_dump(struct net *net, struct notifier_block *nb, @@ -381,6 +375,17 @@ bool ipmr_rule_default(const struct fib_rule *rule) EXPORT_SYMBOL(ipmr_rule_default); #endif +static struct mr_table *ipmr_get_table(struct net *net, u32 id) +{ + struct mr_table *mrt; + + rcu_read_lock(); + mrt = __ipmr_get_table(net, id); + rcu_read_unlock(); + + return mrt; +} + static inline int ipmr_hash_cmp(struct rhashtable_compare_arg *arg, const void *ptr) { @@ -441,12 +446,11 @@ static void ipmr_free_table(struct mr_table *mrt, struct list_head *dev_kill_lis WARN_ON_ONCE(!mr_can_free_table(net)); - timer_shutdown_sync(&mrt->ipmr_expire_timer); mroute_clean_tables(mrt, MRT_FLUSH_VIFS | MRT_FLUSH_VIFS_STATIC | MRT_FLUSH_MFC | MRT_FLUSH_MFC_STATIC, &ipmr_dev_kill_list); - rhltable_destroy(&mrt->mfc_hash); - kfree(mrt); + timer_shutdown_sync(&mrt->ipmr_expire_timer); + mr_table_free(mrt); WARN_ON_ONCE(!net_initialized(net) && !list_empty(&ipmr_dev_kill_list)); list_splice(&ipmr_dev_kill_list, dev_kill_list); @@ -1135,12 +1139,19 @@ static int ipmr_cache_report(const struct mr_table *mrt, static int ipmr_cache_unresolved(struct mr_table *mrt, vifi_t vifi, struct sk_buff *skb, struct net_device *dev) { + struct net *net = read_pnet(&mrt->net); const struct iphdr *iph = ip_hdr(skb); - struct mfc_cache *c; + struct mfc_cache *c = NULL; bool found = false; int err; spin_lock_bh(&mfc_unres_lock); + + if (!check_net(net)) { + err = -EINVAL; + goto err; + } + list_for_each_entry(c, &mrt->mfc_unres_queue, _c.list) { if (c->mfc_mcastgrp == iph->daddr && c->mfc_origin == iph->saddr) { @@ -1153,10 +1164,8 @@ static int ipmr_cache_unresolved(struct mr_table *mrt, vifi_t vifi, /* Create a new entry if allowable */ c = ipmr_cache_alloc_unres(); if (!c) { - spin_unlock_bh(&mfc_unres_lock); - - kfree_skb(skb); - return -ENOBUFS; + err = -ENOBUFS; + goto err; } /* Fill in the new cache entry */ @@ -1166,17 +1175,8 @@ static int ipmr_cache_unresolved(struct mr_table *mrt, vifi_t vifi, /* Reflect first query at mrouted. */ err = ipmr_cache_report(mrt, skb, vifi, IGMPMSG_NOCACHE); - - if (err < 0) { - /* If the report failed throw the cache entry - out - Brad Parker - */ - spin_unlock_bh(&mfc_unres_lock); - - ipmr_cache_free(c); - kfree_skb(skb); - return err; - } + if (err < 0) + goto err; atomic_inc(&mrt->cache_resolve_queue_len); list_add(&c->_c.list, &mrt->mfc_unres_queue); @@ -1189,18 +1189,26 @@ static int ipmr_cache_unresolved(struct mr_table *mrt, vifi_t vifi, /* See if we can append the packet */ if (c->_c.mfc_un.unres.unresolved.qlen > 3) { - kfree_skb(skb); + c = NULL; err = -ENOBUFS; - } else { - if (dev) { - skb->dev = dev; - skb->skb_iif = dev->ifindex; - } - skb_queue_tail(&c->_c.mfc_un.unres.unresolved, skb); - err = 0; + goto err; + } + + if (dev) { + skb->dev = dev; + skb->skb_iif = dev->ifindex; } + skb_queue_tail(&c->_c.mfc_un.unres.unresolved, skb); + spin_unlock_bh(&mfc_unres_lock); + return 0; + +err: + spin_unlock_bh(&mfc_unres_lock); + if (c) + ipmr_cache_free(c); + kfree_skb(skb); return err; } @@ -1346,7 +1354,7 @@ static void mroute_clean_tables(struct mr_table *mrt, int flags, } if (flags & MRT_FLUSH_MFC) { - if (atomic_read(&mrt->cache_resolve_queue_len) != 0) { + if (atomic_read(&mrt->cache_resolve_queue_len) != 0 || !check_net(net)) { spin_lock_bh(&mfc_unres_lock); list_for_each_entry_safe(c, tmp, &mrt->mfc_unres_queue, list) { list_del(&c->list); diff --git a/net/ipv4/ipmr_base.c b/net/ipv4/ipmr_base.c index 37a3c144276c..3930d612c3de 100644 --- a/net/ipv4/ipmr_base.c +++ b/net/ipv4/ipmr_base.c @@ -28,6 +28,20 @@ void vif_device_init(struct vif_device *v, v->link = dev->ifindex; } +static void __mr_free_table(struct work_struct *work) +{ + struct mr_table *mrt = container_of(to_rcu_work(work), + struct mr_table, work); + + rhltable_destroy(&mrt->mfc_hash); + kfree(mrt); +} + +void mr_table_free(struct mr_table *mrt) +{ + queue_rcu_work(system_unbound_wq, &mrt->work); +} + struct mr_table * mr_table_alloc(struct net *net, u32 id, struct mr_table_ops *ops, @@ -50,6 +64,8 @@ mr_table_alloc(struct net *net, u32 id, kfree(mrt); return ERR_PTR(err); } + + INIT_RCU_WORK(&mrt->work, __mr_free_table); INIT_LIST_HEAD(&mrt->mfc_cache_list); INIT_LIST_HEAD(&mrt->mfc_unres_queue); -- cgit v1.2.3 From 4438113be604ee67a7bf4f81da6e1cca41332ce4 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Fri, 24 Apr 2026 16:58:38 +0200 Subject: neigh: let neigh_xmit take skb ownership neigh_xmit always releases the skb, except when no neighbour table is found. But even the first added user of neigh_xmit (mpls) relied on neigh_xmit to release the skb (or queue it for tx). sashiko reported: If neigh_xmit() is called with an uninitialized neighbor table (for example, NEIGH_ND_TABLE when IPv6 is disabled), it returns -EAFNOSUPPORT and bypasses its internal out_kfree_skb error path. Because the return value of neigh_xmit() is ignored here, does this leak the SKB? Assume full ownership and remove the last code path that doesn't xmit or free skb. Fixes: 4fd3d7d9e868 ("neigh: Add helper function neigh_xmit") Signed-off-by: Florian Westphal Reviewed-by: Kuniyuki Iwashima Reviewed-by: Ido Schimmel Link: https://patch.msgid.link/20260424145843.74055-1-fw@strlen.de Signed-off-by: Jakub Kicinski --- net/core/neighbour.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/core/neighbour.c b/net/core/neighbour.c index 9e12524b67fa..5d9216016507 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -3210,8 +3210,10 @@ int neigh_xmit(int index, struct net_device *dev, rcu_read_lock(); tbl = rcu_dereference(neigh_tables[index]); - if (!tbl) - goto out_unlock; + if (!tbl) { + rcu_read_unlock(); + goto out_kfree_skb; + } if (index == NEIGH_ARP_TABLE) { u32 key = *((u32 *)addr); @@ -3227,7 +3229,6 @@ int neigh_xmit(int index, struct net_device *dev, goto out_kfree_skb; } err = READ_ONCE(neigh->output)(neigh, skb); -out_unlock: rcu_read_unlock(); } else if (index == NEIGH_LINK_TABLE) { @@ -3237,11 +3238,10 @@ out_unlock: goto out_kfree_skb; err = dev_queue_xmit(skb); } -out: return err; out_kfree_skb: kfree_skb(skb); - goto out; + return err; } EXPORT_SYMBOL(neigh_xmit); -- cgit v1.2.3 From 2b9f6f7065d4cfb65ba19126e0b35ac4544c3f3a Mon Sep 17 00:00:00 2001 From: Altan Hacigumus Date: Thu, 23 Apr 2026 18:46:38 -0700 Subject: tcp: make probe0 timer handle expired user timeout tcp_clamp_probe0_to_user_timeout() computes remaining time in jiffies using subtraction with an unsigned lvalue. If elapsed probing time exceeds the configured TCP_USER_TIMEOUT, the underflow yields a large value. This ends up re-arming the probe timer for a full backoff interval instead of expiring immediately, delaying connection teardown beyond the configured timeout. Fix this by preventing underflow so user-set timeout expiration is handled correctly without extending the probe timer. Fixes: 344db93ae3ee ("tcp: make TCP_USER_TIMEOUT accurate for zero window probes") Link: https://lore.kernel.org/r/20260414013634.43997-1-ahacigu.linux@gmail.com Signed-off-by: Altan Hacigumus Reviewed-by: Eric Dumazet Link: https://patch.msgid.link/20260424014639.54110-1-ahacigu.linux@gmail.com Signed-off-by: Jakub Kicinski --- net/ipv4/tcp_timer.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c index 8d791a954cd6..322db13333c7 100644 --- a/net/ipv4/tcp_timer.c +++ b/net/ipv4/tcp_timer.c @@ -50,7 +50,8 @@ static u32 tcp_clamp_rto_to_user_timeout(const struct sock *sk) u32 tcp_clamp_probe0_to_user_timeout(const struct sock *sk, u32 when) { const struct inet_connection_sock *icsk = inet_csk(sk); - u32 remaining, user_timeout; + u32 user_timeout; + s32 remaining; s32 elapsed; user_timeout = READ_ONCE(icsk->icsk_user_timeout); @@ -61,7 +62,7 @@ u32 tcp_clamp_probe0_to_user_timeout(const struct sock *sk, u32 when) if (unlikely(elapsed < 0)) elapsed = 0; remaining = msecs_to_jiffies(user_timeout) - elapsed; - remaining = max_t(u32, remaining, TCP_TIMEOUT_MIN); + remaining = max_t(int, remaining, TCP_TIMEOUT_MIN); return min_t(u32, remaining, when); } -- cgit v1.2.3 From 3bc179bc7146c26c9dff75d2943d10528274e301 Mon Sep 17 00:00:00 2001 From: Breno Leitao Date: Fri, 24 Apr 2026 08:31:16 -0700 Subject: netpoll: fix IPv6 local-address corruption netpoll_setup() decides whether to auto-populate the local source address by testing np->local_ip.ip, which only inspects the first 4 bytes of the union inet_addr storage. For an IPv6 netpoll whose caller-supplied local address has a zero high-32 bits (::1, ::, IPv4-mapped ::ffff:a.b.c.d, etc.), this misdetects the address as unset (which they are not, but the first 4 bytes are empty), calls netpoll_take_ipv6() and overwrites it with whatever matching link-local/global address the device happens to expose first. Introduce a helper netpoll_local_ip_unset() that picks the correct family-aware test (ipv6_addr_any() for IPv6, !.ip for IPv4) and use it from netpoll_setup(). Reproducer is something like: echo "::2" > local_ip echo 1 > enabled cat local_ip # before this fix: 2001:db8::1 (caller-supplied ::2 was clobbered) # after this fix: ::2 Fixes: b7394d2429c1 ("netpoll: prepare for ipv6") Signed-off-by: Breno Leitao Link: https://patch.msgid.link/20260424-netpoll_fix-v1-1-3a55348c625f@debian.org Signed-off-by: Jakub Kicinski --- net/core/netpoll.c | 19 ++++++++++++++++++- 1 file changed, 18 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/core/netpoll.c b/net/core/netpoll.c index cd74beffd209..4381e0fc25bf 100644 --- a/net/core/netpoll.c +++ b/net/core/netpoll.c @@ -704,6 +704,23 @@ static int netpoll_take_ipv4(struct netpoll *np, struct net_device *ndev) return 0; } +/* + * Test whether the caller left np->local_ip unset, so that + * netpoll_setup() should auto-populate it from the egress device. + * + * np->local_ip is a union of __be32 (IPv4) and struct in6_addr (IPv6), + * so an IPv6 address whose first 4 bytes are zero (e.g. ::1, ::2, + * IPv4-mapped ::ffff:a.b.c.d) must not be tested via the IPv4 arm — + * doing so would misclassify a caller-supplied address as unset and + * silently overwrite it with whatever address the device exposes. + */ +static bool netpoll_local_ip_unset(const struct netpoll *np) +{ + if (np->ipv6) + return ipv6_addr_any(&np->local_ip.in6); + return !np->local_ip.ip; +} + int netpoll_setup(struct netpoll *np) { struct net *net = current->nsproxy->net_ns; @@ -747,7 +764,7 @@ int netpoll_setup(struct netpoll *np) rtnl_lock(); } - if (!np->local_ip.ip) { + if (netpoll_local_ip_unset(np)) { if (!np->ipv6) { err = netpoll_take_ipv4(np, ndev); if (err) -- cgit v1.2.3 From fa90a3145c0340c3f624206a81637c542254ea1d Mon Sep 17 00:00:00 2001 From: Cosmin Ratiu Date: Wed, 22 Apr 2026 17:06:48 +0300 Subject: xfrm: Don't clobber inner headers when already set On VXLAN over IPsec egress, xfrm{4,6}_transport_output() blindly overwrite inner_transport_header (== the inner TCP header saved in VXLAN iptunnel_handle_offloads() -> skb_reset_inner_headers()) with the current transport_header (== the VXLAN outer UDP header set by udp_tunnel_xmit_skb()). This was a latent bug, harmless until commit [1] added a doff validation check in qdisc_pkt_len_segs_init() for encapsulated GSO packets. With the wrong inner_transport_header set by xfrm, qdisc_pkt_len_segs_init() interprets inner_transport_header as a TCP header, reads doff=0 from the upper byte of the VNI and drops the packet with DROP_REASON_SKB_BAD_GSO. Besides the use in GSO to determine the header size of segmented packets, inner_transport_header might be used by drivers to set up inner checksum offloading by pointing the HW to the inner transport header. A quick browse through available drivers shows that mlx5 uses skb->csum_start specifically for this scenario, while others either don't support VXLAN over IPsec crypto offload (ixgbe) or the HW is capable of parsing the packets itself (nfp, Chelsio). But in all cases, it is more correct to let the inner_transport_header point to the innermost header instead of overwriting it in xfrm. So fix this by guarding all four inner header save sites in xfrm_output.c (xfrm{4,6}_transport_output, xfrm{4,6}_tunnel_encap_add) with a check for skb->inner_protocol. When inner_protocol is set, a tunnel layer (VXLAN, Geneve, GRE, etc.) has already saved the correct inner header offsets and they must not be overwritten. When inner_protocol is zero, no prior tunnel encapsulation exists and xfrm must save the inner headers itself. The tunnel mode checks are only added for completion, since they aren't strictly required, as xfrm_output() forces software GSO in tunnel mode before encap. This makes the previously added test pass: # ./tools/testing/selftests/drivers/net/hw/ipsec_vxlan.py TAP version 13 1..4 ok 1 ipsec_vxlan.test_vxlan_ipsec_crypto_offload.outer_v4_inner_v4 ok 2 ipsec_vxlan.test_vxlan_ipsec_crypto_offload.outer_v4_inner_v6 ok 3 ipsec_vxlan.test_vxlan_ipsec_crypto_offload.outer_v6_inner_v4 ok 4 ipsec_vxlan.test_vxlan_ipsec_crypto_offload.outer_v6_inner_v6 # Totals: pass:4 fail:0 xfail:0 xpass:0 skip:0 error:0 [1] commit 7fb4c1967011 ("net: pull headers in qdisc_pkt_len_segs_init()") Fixes: f1bd7d659ef0 ("xfrm: Add encapsulation header offsets while SKB is not encrypted") Signed-off-by: Cosmin Ratiu Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_output.c | 20 ++++++++++++++------ 1 file changed, 14 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/net/xfrm/xfrm_output.c b/net/xfrm/xfrm_output.c index a9652b422f51..cc35c2fcbbe0 100644 --- a/net/xfrm/xfrm_output.c +++ b/net/xfrm/xfrm_output.c @@ -66,7 +66,9 @@ static int xfrm4_transport_output(struct xfrm_state *x, struct sk_buff *skb) struct iphdr *iph = ip_hdr(skb); int ihl = iph->ihl * 4; - skb_set_inner_transport_header(skb, skb_transport_offset(skb)); + if (!skb->inner_protocol) + skb_set_inner_transport_header(skb, + skb_transport_offset(skb)); skb_set_network_header(skb, -x->props.header_len); skb->mac_header = skb->network_header + @@ -167,7 +169,9 @@ static int xfrm6_transport_output(struct xfrm_state *x, struct sk_buff *skb) int hdr_len; iph = ipv6_hdr(skb); - skb_set_inner_transport_header(skb, skb_transport_offset(skb)); + if (!skb->inner_protocol) + skb_set_inner_transport_header(skb, + skb_transport_offset(skb)); hdr_len = xfrm6_hdr_offset(x, skb, &prevhdr); if (hdr_len < 0) @@ -276,8 +280,10 @@ static int xfrm4_tunnel_encap_add(struct xfrm_state *x, struct sk_buff *skb) struct iphdr *top_iph; int flags; - skb_set_inner_network_header(skb, skb_network_offset(skb)); - skb_set_inner_transport_header(skb, skb_transport_offset(skb)); + if (!skb->inner_protocol) { + skb_set_inner_network_header(skb, skb_network_offset(skb)); + skb_set_inner_transport_header(skb, skb_transport_offset(skb)); + } skb_set_network_header(skb, -x->props.header_len); skb->mac_header = skb->network_header + @@ -321,8 +327,10 @@ static int xfrm6_tunnel_encap_add(struct xfrm_state *x, struct sk_buff *skb) struct ipv6hdr *top_iph; int dsfield; - skb_set_inner_network_header(skb, skb_network_offset(skb)); - skb_set_inner_transport_header(skb, skb_transport_offset(skb)); + if (!skb->inner_protocol) { + skb_set_inner_network_header(skb, skb_network_offset(skb)); + skb_set_inner_transport_header(skb, skb_transport_offset(skb)); + } skb_set_network_header(skb, -x->props.header_len); skb->mac_header = skb->network_header + -- cgit v1.2.3 From f9c52a6ba9780bd27e0bf4c044fd91c13c778b6e Mon Sep 17 00:00:00 2001 From: Andrea Mayer Date: Tue, 21 Apr 2026 11:47:35 +0200 Subject: net: ipv6: fix NOREF dst use in seg6 and rpl lwtunnels seg6_input_core() and rpl_input() call ip6_route_input() which sets a NOREF dst on the skb, then pass it to dst_cache_set_ip6() invoking dst_hold() unconditionally. On PREEMPT_RT, ksoftirqd is preemptible and a higher-priority task can release the underlying pcpu_rt between the lookup and the caching through a concurrent FIB lookup on a shared nexthop. Simplified race sequence: ksoftirqd/X higher-prio task (same CPU X) ----------- -------------------------------- seg6_input_core(,skb)/rpl_input(skb) dst_cache_get() -> miss ip6_route_input(skb) -> ip6_pol_route(,skb,flags) [RT6_LOOKUP_F_DST_NOREF in flags] -> FIB lookup resolves fib6_nh [nhid=N route] -> rt6_make_pcpu_route() [creates pcpu_rt, refcount=1] pcpu_rt->sernum = fib6_sernum [fib6_sernum=W] -> cmpxchg(fib6_nh.rt6i_pcpu, NULL, pcpu_rt) [slot was empty, store succeeds] -> skb_dst_set_noref(skb, dst) [dst is pcpu_rt, refcount still 1] rt_genid_bump_ipv6() -> bumps fib6_sernum [fib6_sernum from W to Z] ip6_route_output() -> ip6_pol_route() -> FIB lookup resolves fib6_nh [nhid=N] -> rt6_get_pcpu_route() pcpu_rt->sernum != fib6_sernum [W <> Z, stale] -> prev = xchg(rt6i_pcpu, NULL) -> dst_release(prev) [prev is pcpu_rt, refcount 1->0, dead] dst = skb_dst(skb) [dst is the dead pcpu_rt] dst_cache_set_ip6(dst) -> dst_hold() on dead dst -> WARN / use-after-free For the race to occur, ksoftirqd must be preemptible (PREEMPT_RT without PREEMPT_RT_NEEDS_BH_LOCK) and a concurrent task must be able to release the pcpu_rt. Shared nexthop objects provide such a path, as two routes pointing to the same nhid share the same fib6_nh and its rt6i_pcpu entry. Fix seg6_input_core() and rpl_input() by calling skb_dst_force() after ip6_route_input() to force the NOREF dst into a refcounted one before caching. The output path is not affected as ip6_route_output() already returns a refcounted dst. Fixes: af4a2209b134 ("ipv6: sr: use dst_cache in seg6_input") Fixes: a7a29f9c361f ("net: ipv6: add rpl sr tunnel") Cc: stable@vger.kernel.org Signed-off-by: Andrea Mayer Reviewed-by: Simon Horman Reviewed-by: Justin Iurman Link: https://patch.msgid.link/20260421094735.20997-1-andrea.mayer@uniroma2.it Signed-off-by: Paolo Abeni --- net/ipv6/rpl_iptunnel.c | 9 +++++++++ net/ipv6/seg6_iptunnel.c | 9 +++++++++ 2 files changed, 18 insertions(+) (limited to 'net') diff --git a/net/ipv6/rpl_iptunnel.c b/net/ipv6/rpl_iptunnel.c index c7942cf65567..4e10adcd70e8 100644 --- a/net/ipv6/rpl_iptunnel.c +++ b/net/ipv6/rpl_iptunnel.c @@ -287,7 +287,16 @@ static int rpl_input(struct sk_buff *skb) if (!dst) { ip6_route_input(skb); + + /* ip6_route_input() sets a NOREF dst; force a refcount on it + * before caching or further use. + */ + skb_dst_force(skb); dst = skb_dst(skb); + if (unlikely(!dst)) { + err = -ENETUNREACH; + goto drop; + } /* cache only if we don't create a dst reference loop */ if (!dst->error && lwtst != dst->lwtstate) { diff --git a/net/ipv6/seg6_iptunnel.c b/net/ipv6/seg6_iptunnel.c index 9b64343ebad6..4c45c0a77d75 100644 --- a/net/ipv6/seg6_iptunnel.c +++ b/net/ipv6/seg6_iptunnel.c @@ -515,7 +515,16 @@ static int seg6_input_core(struct net *net, struct sock *sk, if (!dst) { ip6_route_input(skb); + + /* ip6_route_input() sets a NOREF dst; force a refcount on it + * before caching or further use. + */ + skb_dst_force(skb); dst = skb_dst(skb); + if (unlikely(!dst)) { + err = -ENETUNREACH; + goto drop; + } /* cache only if we don't create a dst reference loop */ if (!dst->error && lwtst != dst->lwtstate) { -- cgit v1.2.3 From 4ca07b9239bd0478ae586632a2ed72be37ed8407 Mon Sep 17 00:00:00 2001 From: "William A. Kennington III" Date: Thu, 23 Apr 2026 00:46:52 -0700 Subject: net: mctp i2c: check length before marking flow active Currently, mctp_i2c_get_tx_flow_state() is called before the packet length sanity check. This function marks a new flow as active in the MCTP core. If the sanity check fails, mctp_i2c_xmit() returns early without calling mctp_i2c_lock_nest(). This results in a mismatched locking state: the flow is active, but the I2C bus lock was never acquired for it. When the flow is later released, mctp_i2c_release_flow() will see the active state and queue an unlock marker. The TX thread will then decrement midev->i2c_lock_count from 0, causing it to underflow to -1. This underflow permanently breaks the driver's locking logic, allowing future transmissions to occur without holding the I2C bus lock, leading to bus collisions and potential hardware hangs. Move the mctp_i2c_get_tx_flow_state() call to after the length sanity check to ensure we only transition the flow state if we are actually going to proceed with the transmission and locking. Fixes: f5b8abf9fc3d ("mctp i2c: MCTP I2C binding driver") Signed-off-by: William A. Kennington III Acked-by: Jeremy Kerr Link: https://patch.msgid.link/20260423074741.201460-1-william@wkennington.com Signed-off-by: Paolo Abeni --- net/sched/cls_flower.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/sched/cls_flower.c b/net/sched/cls_flower.c index 88f8a32fab2b..b9672ea05747 100644 --- a/net/sched/cls_flower.c +++ b/net/sched/cls_flower.c @@ -556,6 +556,7 @@ static int __fl_delete(struct tcf_proto *tp, struct cls_fl_filter *f, struct netlink_ext_ack *extack) { struct cls_fl_head *head = fl_head_dereference(tp); + struct fl_flow_mask *mask; *last = false; @@ -572,11 +573,12 @@ static int __fl_delete(struct tcf_proto *tp, struct cls_fl_filter *f, list_del_rcu(&f->list); spin_unlock(&tp->lock); - *last = fl_mask_put(head, f->mask); + mask = f->mask; if (!tc_skip_hw(f->flags)) fl_hw_destroy_filter(tp, f, rtnl_held, extack); tcf_unbind_filter(tp, &f->res); __fl_put(f); + *last = fl_mask_put(head, mask); return 0; } -- cgit v1.2.3 From b89769f936a8fa9e66de72ddc1b71a9745a488e6 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Mon, 27 Apr 2026 12:06:06 -0700 Subject: net: psp: check for device unregister when creating assoc psp_assoc_device_get_locked() obtains a psp_dev reference via psp_dev_get_for_sock() (which uses psp_dev_tryget() under RCU); it then acquires psd->lock and drops the reference. Before the lock is taken, psp_dev_unregister() can run to completion: take psd->lock, clear out state, unlock, drop the registration reference. The expectation is that the lock prevents device unregistration, but much like with netdevs special care has to be taken when "upgrading" a reference to a locked device. Add the missing check if device is still alive. psp_dev_is_registered() exists already but had no callers, which makes me wonder if I either forgot to add this or lost the check during refactoring... Reported-by: Yiming Qian Fixes: 6b46ca260e22 ("net: psp: add socket security association code") Reviewed-by: Willem de Bruijn Link: https://patch.msgid.link/20260427190606.366101-1-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/psp/psp_nl.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/psp/psp_nl.c b/net/psp/psp_nl.c index 6afd7707ec12..0cc744a6e1c9 100644 --- a/net/psp/psp_nl.c +++ b/net/psp/psp_nl.c @@ -305,8 +305,13 @@ int psp_assoc_device_get_locked(const struct genl_split_ops *ops, psd = psp_dev_get_for_sock(socket->sk); if (psd) { - err = psp_dev_check_access(psd, genl_info_net(info)); - if (err) { + /* Extra care needed here, psp_dev_get_for_sock() only gives + * us access to struct psp_dev's memory, which is quite weak. + */ + mutex_lock(&psd->lock); + if (!psp_dev_is_registered(psd) || + psp_dev_check_access(psd, genl_info_net(info))) { + mutex_unlock(&psd->lock); psp_dev_put(psd); psd = NULL; } @@ -319,7 +324,6 @@ int psp_assoc_device_get_locked(const struct genl_split_ops *ops, id = info->attrs[PSP_A_ASSOC_DEV_ID]; if (psd) { - mutex_lock(&psd->lock); if (id && psd->id != nla_get_u32(id)) { mutex_unlock(&psd->lock); NL_SET_ERR_MSG_ATTR(info->extack, id, -- cgit v1.2.3 From b718342a7fbaa2dff5fefc31988c07af8c6cbc21 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Mon, 27 Apr 2026 12:58:56 -0700 Subject: net: psp: require admin permission for dev-set and key-rotate The dev-set and key-rotate netlink operations modify shared device state (PSP version configuration and cryptographic key material, respectively) but do not require CAP_NET_ADMIN. The only access control is psp_dev_check_access() which merely verifies netns membership. Fixes: 00c94ca2b99e ("psp: base PSP device support") Reviewed-by: Daniel Zahka Reviewed-by: Willem de Bruijn Link: https://patch.msgid.link/20260427195856.401223-1-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/psp/psp-nl-gen.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/psp/psp-nl-gen.c b/net/psp/psp-nl-gen.c index 22a48d0fa378..953309952cef 100644 --- a/net/psp/psp-nl-gen.c +++ b/net/psp/psp-nl-gen.c @@ -76,7 +76,7 @@ static const struct genl_split_ops psp_nl_ops[] = { .post_doit = psp_device_unlock, .policy = psp_dev_set_nl_policy, .maxattr = PSP_A_DEV_PSP_VERSIONS_ENA, - .flags = GENL_CMD_CAP_DO, + .flags = GENL_ADMIN_PERM | GENL_CMD_CAP_DO, }, { .cmd = PSP_CMD_KEY_ROTATE, @@ -85,7 +85,7 @@ static const struct genl_split_ops psp_nl_ops[] = { .post_doit = psp_device_unlock, .policy = psp_key_rotate_nl_policy, .maxattr = PSP_A_DEV_ID, - .flags = GENL_CMD_CAP_DO, + .flags = GENL_ADMIN_PERM | GENL_CMD_CAP_DO, }, { .cmd = PSP_CMD_RX_ASSOC, -- cgit v1.2.3 From 576a5d2bad4814c881a829576b1261b9b8159d2b Mon Sep 17 00:00:00 2001 From: Xin Long Date: Sun, 26 Apr 2026 10:46:40 -0400 Subject: netfilter: skip recording stale or retransmitted INIT An INIT whose init_tag matches the peer's vtag does not provide new state information. It indicates either: - a stale INIT (after INIT-ACK has already been seen on the same side), or - a retransmitted INIT (after INIT has already been recorded on the same side). In both cases, the INIT must not update ct->proto.sctp.init[] state, since it does not advance the handshake tracking and may otherwise corrupt INIT/INIT-ACK validation logic. Allow INIT processing only when the conntrack entry is newly created (SCTP_CONNTRACK_NONE), or when the init_tag differs from the stored peer vtag. Note it skips the check for the ct with old_state SCTP_CONNTRACK_NONE in nf_conntrack_sctp_packet(), as it is just created in sctp_new() where it set ct->proto.sctp.vtag[IP_CT_DIR_REPLY] = ih->init_tag. Fixes: 9fb9cbb1082d ("[NETFILTER]: Add nf_conntrack subsystem.") Signed-off-by: Xin Long Reviewed-by: Marcelo Ricardo Leitner Acked-by: Florian Westphal Link: https://patch.msgid.link/ee56c3e416452b2a40589a2a85245ac2ad5e9f4b.1777214801.git.lucien.xin@gmail.com Signed-off-by: Jakub Kicinski --- net/netfilter/nf_conntrack_proto_sctp.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/netfilter/nf_conntrack_proto_sctp.c b/net/netfilter/nf_conntrack_proto_sctp.c index 645d2c43ebf7..7e10fa65cbdd 100644 --- a/net/netfilter/nf_conntrack_proto_sctp.c +++ b/net/netfilter/nf_conntrack_proto_sctp.c @@ -466,9 +466,13 @@ int nf_conntrack_sctp_packet(struct nf_conn *ct, if (!ih) goto out_unlock; - if (ct->proto.sctp.init[dir] && ct->proto.sctp.init[!dir]) - ct->proto.sctp.init[!dir] = 0; - ct->proto.sctp.init[dir] = 1; + /* Do not record INIT matching peer vtag (stale or retransmitted INIT). */ + if (old_state == SCTP_CONNTRACK_NONE || + ct->proto.sctp.vtag[!dir] != ih->init_tag) { + if (ct->proto.sctp.init[dir] && ct->proto.sctp.init[!dir]) + ct->proto.sctp.init[!dir] = 0; + ct->proto.sctp.init[dir] = 1; + } pr_debug("Setting vtag %x for dir %d\n", ih->init_tag, !dir); ct->proto.sctp.vtag[!dir] = ih->init_tag; -- cgit v1.2.3 From 8a92cb475ca90d84db769e4d4383e631ace0d6e5 Mon Sep 17 00:00:00 2001 From: Xin Long Date: Sun, 26 Apr 2026 10:46:41 -0400 Subject: sctp: discard stale INIT after handshake completion MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit After an association reaches ESTABLISHED, the peer’s init_tag is already known from the handshake. Any subsequent INIT with the same init_tag is not a valid restart, but a delayed or duplicate INIT. Drop such INIT chunks in sctp_sf_do_unexpected_init() instead of processing them as new association attempts. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: Xin Long Acked-by: Marcelo Ricardo Leitner Link: https://patch.msgid.link/5788c76c1ee122a3ed00189e88dcf9df1fba226c.1777214801.git.lucien.xin@gmail.com Signed-off-by: Jakub Kicinski --- net/sctp/sm_statefuns.c | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'net') diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c index 7b823d759141..8e89a870780c 100644 --- a/net/sctp/sm_statefuns.c +++ b/net/sctp/sm_statefuns.c @@ -1556,6 +1556,12 @@ static enum sctp_disposition sctp_sf_do_unexpected_init( /* Tag the variable length parameters. */ chunk->param_hdr.v = skb_pull(chunk->skb, sizeof(struct sctp_inithdr)); + if (asoc->state >= SCTP_STATE_ESTABLISHED) { + /* Discard INIT matching peer vtag after handshake completion (stale INIT). */ + if (ntohl(chunk->subh.init_hdr->init_tag) == asoc->peer.i.init_tag) + return sctp_sf_pdiscard(net, ep, asoc, type, arg, commands); + } + /* Verify the INIT chunk before processing it. */ err_chunk = NULL; if (!sctp_verify_init(net, ep, asoc, chunk->chunk_hdr->type, -- cgit v1.2.3 From 44967ac3785ebef6442377708925181d4a0eb1c8 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 27 Apr 2026 08:36:02 +0000 Subject: net/sched: sch_cake: annotate data-races in cake_dump_stats() (I) MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit cake_dump_stats() runs without qdisc spinlock being held. In this first patch, I add READ_ONCE()/WRITE_ONCE() annotations for the following fields: - way_hits - way_misses - way_collisions - sparse_flow_count - decaying_flow_count Other annotations are added in following patches, to ease code review. Fixes: 046f6fd5daef ("sched: Add Common Applications Kept Enhanced (cake) qdisc") Signed-off-by: Eric Dumazet Acked-by: "Toke Høiland-Jørgensen" Link: https://patch.msgid.link/20260427083606.459355-2-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/sched/sch_cake.c | 42 +++++++++++++++++++++--------------------- 1 file changed, 21 insertions(+), 21 deletions(-) (limited to 'net') diff --git a/net/sched/sch_cake.c b/net/sched/sch_cake.c index 02e1fa4577ae..bcc601fc486b 100644 --- a/net/sched/sch_cake.c +++ b/net/sched/sch_cake.c @@ -813,7 +813,7 @@ skip_hash: i++, k = (k + 1) % CAKE_SET_WAYS) { if (q->tags[outer_hash + k] == flow_hash) { if (i) - q->way_hits++; + WRITE_ONCE(q->way_hits, q->way_hits + 1); if (!q->flows[outer_hash + k].set) { /* need to increment host refcnts */ @@ -831,7 +831,7 @@ skip_hash: for (i = 0; i < CAKE_SET_WAYS; i++, k = (k + 1) % CAKE_SET_WAYS) { if (!q->flows[outer_hash + k].set) { - q->way_misses++; + WRITE_ONCE(q->way_misses, q->way_misses + 1); allocate_src = cake_dsrc(flow_mode); allocate_dst = cake_ddst(flow_mode); goto found; @@ -841,7 +841,7 @@ skip_hash: /* With no empty queues, default to the original * queue, accept the collision, update the host tags. */ - q->way_collisions++; + WRITE_ONCE(q->way_collisions, q->way_collisions + 1); allocate_src = cake_dsrc(flow_mode); allocate_dst = cake_ddst(flow_mode); @@ -1917,11 +1917,11 @@ static s32 cake_enqueue(struct sk_buff *skb, struct Qdisc *sch, if (!flow->set) { list_add_tail(&flow->flowchain, &b->new_flows); } else { - b->decaying_flow_count--; + WRITE_ONCE(b->decaying_flow_count, b->decaying_flow_count - 1); list_move_tail(&flow->flowchain, &b->new_flows); } flow->set = CAKE_SET_SPARSE; - b->sparse_flow_count++; + WRITE_ONCE(b->sparse_flow_count, b->sparse_flow_count + 1); flow->deficit = cake_get_flow_quantum(b, flow, q->config->flow_mode); } else if (flow->set == CAKE_SET_SPARSE_WAIT) { @@ -1929,7 +1929,7 @@ static s32 cake_enqueue(struct sk_buff *skb, struct Qdisc *sch, * in the bulk rotation. */ flow->set = CAKE_SET_BULK; - b->sparse_flow_count--; + WRITE_ONCE(b->sparse_flow_count, b->sparse_flow_count - 1); b->bulk_flow_count++; cake_inc_srchost_bulk_flow_count(b, flow, q->config->flow_mode); @@ -2149,7 +2149,7 @@ retry: */ if (flow->set == CAKE_SET_SPARSE) { if (flow->head) { - b->sparse_flow_count--; + WRITE_ONCE(b->sparse_flow_count, b->sparse_flow_count - 1); b->bulk_flow_count++; cake_inc_srchost_bulk_flow_count(b, flow, q->config->flow_mode); @@ -2192,27 +2192,27 @@ retry: cake_dec_srchost_bulk_flow_count(b, flow, q->config->flow_mode); cake_dec_dsthost_bulk_flow_count(b, flow, q->config->flow_mode); - b->decaying_flow_count++; + WRITE_ONCE(b->decaying_flow_count, b->decaying_flow_count + 1); } else if (flow->set == CAKE_SET_SPARSE || flow->set == CAKE_SET_SPARSE_WAIT) { - b->sparse_flow_count--; - b->decaying_flow_count++; + WRITE_ONCE(b->sparse_flow_count, b->sparse_flow_count - 1); + WRITE_ONCE(b->decaying_flow_count, b->decaying_flow_count + 1); } flow->set = CAKE_SET_DECAYING; } else { /* remove empty queue from the flowchain */ list_del_init(&flow->flowchain); if (flow->set == CAKE_SET_SPARSE || - flow->set == CAKE_SET_SPARSE_WAIT) - b->sparse_flow_count--; - else if (flow->set == CAKE_SET_BULK) { + flow->set == CAKE_SET_SPARSE_WAIT) { + WRITE_ONCE(b->sparse_flow_count, b->sparse_flow_count - 1); + } else if (flow->set == CAKE_SET_BULK) { b->bulk_flow_count--; cake_dec_srchost_bulk_flow_count(b, flow, q->config->flow_mode); cake_dec_dsthost_bulk_flow_count(b, flow, q->config->flow_mode); - } else - b->decaying_flow_count--; - + } else { + WRITE_ONCE(b->decaying_flow_count, b->decaying_flow_count - 1); + } flow->set = CAKE_SET_NONE; } goto begin; @@ -3050,12 +3050,12 @@ static int cake_dump_stats(struct Qdisc *sch, struct gnet_dump *d) PUT_TSTAT_U32(BASE_DELAY_US, ktime_to_us(ns_to_ktime(b->base_delay))); - PUT_TSTAT_U32(WAY_INDIRECT_HITS, b->way_hits); - PUT_TSTAT_U32(WAY_MISSES, b->way_misses); - PUT_TSTAT_U32(WAY_COLLISIONS, b->way_collisions); + PUT_TSTAT_U32(WAY_INDIRECT_HITS, READ_ONCE(b->way_hits)); + PUT_TSTAT_U32(WAY_MISSES, READ_ONCE(b->way_misses)); + PUT_TSTAT_U32(WAY_COLLISIONS, READ_ONCE(b->way_collisions)); - PUT_TSTAT_U32(SPARSE_FLOWS, b->sparse_flow_count + - b->decaying_flow_count); + PUT_TSTAT_U32(SPARSE_FLOWS, READ_ONCE(b->sparse_flow_count) + + READ_ONCE(b->decaying_flow_count)); PUT_TSTAT_U32(BULK_FLOWS, b->bulk_flow_count); PUT_TSTAT_U32(UNRESPONSIVE_FLOWS, b->unresponsive_flow_count); PUT_TSTAT_U32(MAX_SKBLEN, b->max_skblen); -- cgit v1.2.3 From 91a96427b93b9ba27413077b7e825d2fefbfa134 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 27 Apr 2026 08:36:03 +0000 Subject: net/sched: sch_cake: annotate data-races in cake_dump_stats() (II) MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit cake_dump_stats() runs without qdisc spinlock being held. In this second patch, I add READ_ONCE()/WRITE_ONCE() annotations for the following fields: - bulk_flow_count - unresponsive_flow_count - max_skblen - flow_quantum Other annotations are added in following patches, to ease code review. Fixes: 046f6fd5daef ("sched: Add Common Applications Kept Enhanced (cake) qdisc") Signed-off-by: Eric Dumazet Acked-by: "Toke Høiland-Jørgensen" Link: https://patch.msgid.link/20260427083606.459355-3-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/sched/sch_cake.c | 34 +++++++++++++++++++--------------- 1 file changed, 19 insertions(+), 15 deletions(-) (limited to 'net') diff --git a/net/sched/sch_cake.c b/net/sched/sch_cake.c index bcc601fc486b..d7465ee4c550 100644 --- a/net/sched/sch_cake.c +++ b/net/sched/sch_cake.c @@ -1590,7 +1590,8 @@ static unsigned int cake_drop(struct Qdisc *sch, struct sk_buff **to_free) } if (cobalt_queue_full(&flow->cvars, &b->cparams, now)) - b->unresponsive_flow_count++; + WRITE_ONCE(b->unresponsive_flow_count, + b->unresponsive_flow_count + 1); len = qdisc_pkt_len(skb); q->buffer_used -= skb->truesize; @@ -1795,7 +1796,7 @@ static s32 cake_enqueue(struct sk_buff *skb, struct Qdisc *sch, } if (unlikely(len > b->max_skblen)) - b->max_skblen = len; + WRITE_ONCE(b->max_skblen, len); if (qdisc_pkt_segs(skb) > 1 && q->config->rate_flags & CAKE_FLAG_SPLIT_GSO) { struct sk_buff *segs, *nskb; @@ -1930,7 +1931,7 @@ static s32 cake_enqueue(struct sk_buff *skb, struct Qdisc *sch, */ flow->set = CAKE_SET_BULK; WRITE_ONCE(b->sparse_flow_count, b->sparse_flow_count - 1); - b->bulk_flow_count++; + WRITE_ONCE(b->bulk_flow_count, b->bulk_flow_count + 1); cake_inc_srchost_bulk_flow_count(b, flow, q->config->flow_mode); cake_inc_dsthost_bulk_flow_count(b, flow, q->config->flow_mode); @@ -2150,7 +2151,7 @@ retry: if (flow->set == CAKE_SET_SPARSE) { if (flow->head) { WRITE_ONCE(b->sparse_flow_count, b->sparse_flow_count - 1); - b->bulk_flow_count++; + WRITE_ONCE(b->bulk_flow_count, b->bulk_flow_count + 1); cake_inc_srchost_bulk_flow_count(b, flow, q->config->flow_mode); cake_inc_dsthost_bulk_flow_count(b, flow, q->config->flow_mode); @@ -2177,7 +2178,8 @@ retry: if (!skb) { /* this queue was actually empty */ if (cobalt_queue_empty(&flow->cvars, &b->cparams, now)) - b->unresponsive_flow_count--; + WRITE_ONCE(b->unresponsive_flow_count, + b->unresponsive_flow_count - 1); if (flow->cvars.p_drop || flow->cvars.count || ktime_before(now, flow->cvars.drop_next)) { @@ -2187,7 +2189,7 @@ retry: list_move_tail(&flow->flowchain, &b->decaying_flows); if (flow->set == CAKE_SET_BULK) { - b->bulk_flow_count--; + WRITE_ONCE(b->bulk_flow_count, b->bulk_flow_count - 1); cake_dec_srchost_bulk_flow_count(b, flow, q->config->flow_mode); cake_dec_dsthost_bulk_flow_count(b, flow, q->config->flow_mode); @@ -2206,7 +2208,7 @@ retry: flow->set == CAKE_SET_SPARSE_WAIT) { WRITE_ONCE(b->sparse_flow_count, b->sparse_flow_count - 1); } else if (flow->set == CAKE_SET_BULK) { - b->bulk_flow_count--; + WRITE_ONCE(b->bulk_flow_count, b->bulk_flow_count - 1); cake_dec_srchost_bulk_flow_count(b, flow, q->config->flow_mode); cake_dec_dsthost_bulk_flow_count(b, flow, q->config->flow_mode); @@ -2329,9 +2331,9 @@ static void cake_set_rate(struct cake_tin_data *b, u64 rate, u32 mtu, u8 rate_shft = 0; u64 rate_ns = 0; - b->flow_quantum = 1514; if (rate) { - b->flow_quantum = max(min(rate >> 12, 1514ULL), 300ULL); + WRITE_ONCE(b->flow_quantum, + max(min(rate >> 12, 1514ULL), 300ULL)); rate_shft = 34; rate_ns = ((u64)NSEC_PER_SEC) << rate_shft; rate_ns = div64_u64(rate_ns, max(MIN_RATE, rate)); @@ -2339,8 +2341,10 @@ static void cake_set_rate(struct cake_tin_data *b, u64 rate, u32 mtu, rate_ns >>= 1; rate_shft--; } - } /* else unlimited, ie. zero delay */ - + } else { + /* else unlimited, ie. zero delay */ + WRITE_ONCE(b->flow_quantum, 1514); + } b->tin_rate_bps = rate; b->tin_rate_ns = rate_ns; b->tin_rate_shft = rate_shft; @@ -3056,11 +3060,11 @@ static int cake_dump_stats(struct Qdisc *sch, struct gnet_dump *d) PUT_TSTAT_U32(SPARSE_FLOWS, READ_ONCE(b->sparse_flow_count) + READ_ONCE(b->decaying_flow_count)); - PUT_TSTAT_U32(BULK_FLOWS, b->bulk_flow_count); - PUT_TSTAT_U32(UNRESPONSIVE_FLOWS, b->unresponsive_flow_count); - PUT_TSTAT_U32(MAX_SKBLEN, b->max_skblen); + PUT_TSTAT_U32(BULK_FLOWS, READ_ONCE(b->bulk_flow_count)); + PUT_TSTAT_U32(UNRESPONSIVE_FLOWS, READ_ONCE(b->unresponsive_flow_count)); + PUT_TSTAT_U32(MAX_SKBLEN, READ_ONCE(b->max_skblen)); - PUT_TSTAT_U32(FLOW_QUANTUM, b->flow_quantum); + PUT_TSTAT_U32(FLOW_QUANTUM, READ_ONCE(b->flow_quantum)); nla_nest_end(d->skb, ts); } -- cgit v1.2.3 From 276a98a434964088fccd4745db5b34d6e831e358 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 27 Apr 2026 08:36:04 +0000 Subject: net/sched: sch_cake: annotate data-races in cake_dump_stats() (III) MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit cake_dump_stats() runs without qdisc spinlock being held. In this third patch, I add READ_ONCE()/WRITE_ONCE() annotations for the following fields: - packets - tin_dropped - tin_ecn_mark - ack_drops - peak_delay - avge_delay - base_delay Other annotations are added in following patches, to ease code review. Fixes: 046f6fd5daef ("sched: Add Common Applications Kept Enhanced (cake) qdisc") Signed-off-by: Eric Dumazet Acked-by: "Toke Høiland-Jørgensen" Link: https://patch.msgid.link/20260427083606.459355-4-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/sched/sch_cake.c | 38 ++++++++++++++++++++------------------ 1 file changed, 20 insertions(+), 18 deletions(-) (limited to 'net') diff --git a/net/sched/sch_cake.c b/net/sched/sch_cake.c index d7465ee4c550..c5aae31565e9 100644 --- a/net/sched/sch_cake.c +++ b/net/sched/sch_cake.c @@ -1600,7 +1600,7 @@ static unsigned int cake_drop(struct Qdisc *sch, struct sk_buff **to_free) sch->qstats.backlog -= len; flow->dropped++; - b->tin_dropped++; + WRITE_ONCE(b->tin_dropped, b->tin_dropped + 1); if (q->config->rate_flags & CAKE_FLAG_INGRESS) cake_advance_shaper(q, b, skb, now, true); @@ -1820,7 +1820,7 @@ static s32 cake_enqueue(struct sk_buff *skb, struct Qdisc *sch, numsegs++; slen += segs->len; q->buffer_used += segs->truesize; - b->packets++; + WRITE_ONCE(b->packets, b->packets + 1); } /* stats */ @@ -1844,7 +1844,7 @@ static s32 cake_enqueue(struct sk_buff *skb, struct Qdisc *sch, ack = cake_ack_filter(q, flow); if (ack) { - b->ack_drops++; + WRITE_ONCE(b->ack_drops, b->ack_drops + 1); sch->qstats.drops++; ack_pkt_len = qdisc_pkt_len(ack); b->bytes += ack_pkt_len; @@ -1860,7 +1860,7 @@ static s32 cake_enqueue(struct sk_buff *skb, struct Qdisc *sch, } /* stats */ - b->packets++; + WRITE_ONCE(b->packets, b->packets + 1); b->bytes += len - ack_pkt_len; b->backlogs[idx] += len - ack_pkt_len; b->tin_backlog += len - ack_pkt_len; @@ -2236,7 +2236,7 @@ retry: b->tin_deficit -= len; } flow->dropped++; - b->tin_dropped++; + WRITE_ONCE(b->tin_dropped, b->tin_dropped + 1); qdisc_tree_reduce_backlog(sch, 1, qdisc_pkt_len(skb)); qdisc_qstats_drop(sch); qdisc_dequeue_drop(sch, skb, reason); @@ -2244,17 +2244,19 @@ retry: goto retry; } - b->tin_ecn_mark += !!flow->cvars.ecn_marked; + WRITE_ONCE(b->tin_ecn_mark, b->tin_ecn_mark + !!flow->cvars.ecn_marked); qdisc_bstats_update(sch, skb); WRITE_ONCE(q->last_active, now); /* collect delay stats */ delay = ktime_to_ns(ktime_sub(now, cobalt_get_enqueue_time(skb))); - b->avge_delay = cake_ewma(b->avge_delay, delay, 8); - b->peak_delay = cake_ewma(b->peak_delay, delay, - delay > b->peak_delay ? 2 : 8); - b->base_delay = cake_ewma(b->base_delay, delay, - delay < b->base_delay ? 2 : 8); + WRITE_ONCE(b->avge_delay, cake_ewma(b->avge_delay, delay, 8)); + WRITE_ONCE(b->peak_delay, + cake_ewma(b->peak_delay, delay, + delay > b->peak_delay ? 2 : 8)); + WRITE_ONCE(b->base_delay, + cake_ewma(b->base_delay, delay, + delay < b->base_delay ? 2 : 8)); len = cake_advance_shaper(q, b, skb, now, false); flow->deficit -= len; @@ -3042,17 +3044,17 @@ static int cake_dump_stats(struct Qdisc *sch, struct gnet_dump *d) PUT_TSTAT_U32(INTERVAL_US, ktime_to_us(ns_to_ktime(b->cparams.interval))); - PUT_TSTAT_U32(SENT_PACKETS, b->packets); - PUT_TSTAT_U32(DROPPED_PACKETS, b->tin_dropped); - PUT_TSTAT_U32(ECN_MARKED_PACKETS, b->tin_ecn_mark); - PUT_TSTAT_U32(ACKS_DROPPED_PACKETS, b->ack_drops); + PUT_TSTAT_U32(SENT_PACKETS, READ_ONCE(b->packets)); + PUT_TSTAT_U32(DROPPED_PACKETS, READ_ONCE(b->tin_dropped)); + PUT_TSTAT_U32(ECN_MARKED_PACKETS, READ_ONCE(b->tin_ecn_mark)); + PUT_TSTAT_U32(ACKS_DROPPED_PACKETS, READ_ONCE(b->ack_drops)); PUT_TSTAT_U32(PEAK_DELAY_US, - ktime_to_us(ns_to_ktime(b->peak_delay))); + ktime_to_us(ns_to_ktime(READ_ONCE(b->peak_delay)))); PUT_TSTAT_U32(AVG_DELAY_US, - ktime_to_us(ns_to_ktime(b->avge_delay))); + ktime_to_us(ns_to_ktime(READ_ONCE(b->avge_delay)))); PUT_TSTAT_U32(BASE_DELAY_US, - ktime_to_us(ns_to_ktime(b->base_delay))); + ktime_to_us(ns_to_ktime(READ_ONCE(b->base_delay)))); PUT_TSTAT_U32(WAY_INDIRECT_HITS, READ_ONCE(b->way_hits)); PUT_TSTAT_U32(WAY_MISSES, READ_ONCE(b->way_misses)); -- cgit v1.2.3 From 8fab48d87745a6ab1cec594b8d5865d9ae2db879 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 27 Apr 2026 08:36:05 +0000 Subject: net/sched: sch_cake: annotate data-races in cake_dump_stats() (IV) MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit cake_dump_stats() runs without qdisc spinlock being held. In this fourth patch, I add READ_ONCE()/WRITE_ONCE() annotations for the following fields: - avg_peak_bandwidth - buffer_limit - buffer_max_used - avg_netoff - max_netlen - max_adjlen - min_netlen - min_adjlen - active_queues - tin_rate_bps - bytes - tin_backlog Other annotations are added in following patch, to ease code review. Fixes: 046f6fd5daef ("sched: Add Common Applications Kept Enhanced (cake) qdisc") Signed-off-by: Eric Dumazet Acked-by: Toke Høiland-Jørgensen Link: https://patch.msgid.link/20260427083606.459355-5-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/sched/sch_cake.c | 90 +++++++++++++++++++++++++++------------------------- 1 file changed, 46 insertions(+), 44 deletions(-) (limited to 'net') diff --git a/net/sched/sch_cake.c b/net/sched/sch_cake.c index c5aae31565e9..975f5d6d6982 100644 --- a/net/sched/sch_cake.c +++ b/net/sched/sch_cake.c @@ -1379,9 +1379,9 @@ static u32 cake_calc_overhead(struct cake_sched_data *qd, u32 len, u32 off) len -= off; if (qd->max_netlen < len) - qd->max_netlen = len; + WRITE_ONCE(qd->max_netlen, len); if (qd->min_netlen > len) - qd->min_netlen = len; + WRITE_ONCE(qd->min_netlen, len); len += q->rate_overhead; @@ -1401,9 +1401,9 @@ static u32 cake_calc_overhead(struct cake_sched_data *qd, u32 len, u32 off) } if (qd->max_adjlen < len) - qd->max_adjlen = len; + WRITE_ONCE(qd->max_adjlen, len); if (qd->min_adjlen > len) - qd->min_adjlen = len; + WRITE_ONCE(qd->min_adjlen, len); return len; } @@ -1416,7 +1416,7 @@ static u32 cake_overhead(struct cake_sched_data *q, const struct sk_buff *skb) u16 segs = qdisc_pkt_segs(skb); u32 len = qdisc_pkt_len(skb); - q->avg_netoff = cake_ewma(q->avg_netoff, off << 16, 8); + WRITE_ONCE(q->avg_netoff, cake_ewma(q->avg_netoff, off << 16, 8)); if (segs == 1) return cake_calc_overhead(q, len, off); @@ -1596,7 +1596,7 @@ static unsigned int cake_drop(struct Qdisc *sch, struct sk_buff **to_free) len = qdisc_pkt_len(skb); q->buffer_used -= skb->truesize; b->backlogs[idx] -= len; - b->tin_backlog -= len; + WRITE_ONCE(b->tin_backlog, b->tin_backlog - len); sch->qstats.backlog -= len; flow->dropped++; @@ -1824,11 +1824,11 @@ static s32 cake_enqueue(struct sk_buff *skb, struct Qdisc *sch, } /* stats */ - b->bytes += slen; b->backlogs[idx] += slen; - b->tin_backlog += slen; sch->qstats.backlog += slen; q->avg_window_bytes += slen; + WRITE_ONCE(b->bytes, b->bytes + slen); + WRITE_ONCE(b->tin_backlog, b->tin_backlog + slen); qdisc_tree_reduce_backlog(sch, 1-numsegs, len-slen); consume_skb(skb); @@ -1847,7 +1847,7 @@ static s32 cake_enqueue(struct sk_buff *skb, struct Qdisc *sch, WRITE_ONCE(b->ack_drops, b->ack_drops + 1); sch->qstats.drops++; ack_pkt_len = qdisc_pkt_len(ack); - b->bytes += ack_pkt_len; + WRITE_ONCE(b->bytes, b->bytes + ack_pkt_len); q->buffer_used += skb->truesize - ack->truesize; if (q->config->rate_flags & CAKE_FLAG_INGRESS) cake_advance_shaper(q, b, ack, now, true); @@ -1861,11 +1861,11 @@ static s32 cake_enqueue(struct sk_buff *skb, struct Qdisc *sch, /* stats */ WRITE_ONCE(b->packets, b->packets + 1); - b->bytes += len - ack_pkt_len; b->backlogs[idx] += len - ack_pkt_len; - b->tin_backlog += len - ack_pkt_len; sch->qstats.backlog += len - ack_pkt_len; q->avg_window_bytes += len - ack_pkt_len; + WRITE_ONCE(b->bytes, b->bytes + len - ack_pkt_len); + WRITE_ONCE(b->tin_backlog, b->tin_backlog + len - ack_pkt_len); } if (q->overflow_timeout) @@ -1895,9 +1895,9 @@ static s32 cake_enqueue(struct sk_buff *skb, struct Qdisc *sch, u64 b = q->avg_window_bytes * (u64)NSEC_PER_SEC; b = div64_u64(b, window_interval); - q->avg_peak_bandwidth = - cake_ewma(q->avg_peak_bandwidth, b, - b > q->avg_peak_bandwidth ? 2 : 8); + WRITE_ONCE(q->avg_peak_bandwidth, + cake_ewma(q->avg_peak_bandwidth, b, + b > q->avg_peak_bandwidth ? 2 : 8)); q->avg_window_bytes = 0; q->avg_window_begin = now; @@ -1938,7 +1938,7 @@ static s32 cake_enqueue(struct sk_buff *skb, struct Qdisc *sch, } if (q->buffer_used > q->buffer_max_used) - q->buffer_max_used = q->buffer_used; + WRITE_ONCE(q->buffer_max_used, q->buffer_used); if (q->buffer_used <= q->buffer_limit) return NET_XMIT_SUCCESS; @@ -1978,7 +1978,7 @@ static struct sk_buff *cake_dequeue_one(struct Qdisc *sch) skb = dequeue_head(flow); len = qdisc_pkt_len(skb); b->backlogs[q->cur_flow] -= len; - b->tin_backlog -= len; + WRITE_ONCE(b->tin_backlog, b->tin_backlog - len); sch->qstats.backlog -= len; q->buffer_used -= skb->truesize; sch->q.qlen--; @@ -2043,7 +2043,7 @@ static struct sk_buff *cake_dequeue(struct Qdisc *sch) cake_configure_rates(sch, new_rate, true); q->last_checked_active = now; - q->active_queues = num_active_qs; + WRITE_ONCE(q->active_queues, num_active_qs); } begin: @@ -2347,7 +2347,7 @@ static void cake_set_rate(struct cake_tin_data *b, u64 rate, u32 mtu, /* else unlimited, ie. zero delay */ WRITE_ONCE(b->flow_quantum, 1514); } - b->tin_rate_bps = rate; + WRITE_ONCE(b->tin_rate_bps, rate); b->tin_rate_ns = rate_ns; b->tin_rate_shft = rate_shft; @@ -2617,25 +2617,27 @@ static void cake_reconfigure(struct Qdisc *sch) { struct cake_sched_data *qd = qdisc_priv(sch); struct cake_sched_config *q = qd->config; + u32 buffer_limit; cake_configure_rates(sch, qd->config->rate_bps, false); if (q->buffer_config_limit) { - qd->buffer_limit = q->buffer_config_limit; + buffer_limit = q->buffer_config_limit; } else if (q->rate_bps) { u64 t = q->rate_bps * q->interval; do_div(t, USEC_PER_SEC / 4); - qd->buffer_limit = max_t(u32, t, 4U << 20); + buffer_limit = max_t(u32, t, 4U << 20); } else { - qd->buffer_limit = ~0; + buffer_limit = ~0; } sch->flags &= ~TCQ_F_CAN_BYPASS; - qd->buffer_limit = min(qd->buffer_limit, - max(sch->limit * psched_mtu(qdisc_dev(sch)), - q->buffer_config_limit)); + WRITE_ONCE(qd->buffer_limit, + min(buffer_limit, + max(sch->limit * psched_mtu(qdisc_dev(sch)), + q->buffer_config_limit))); } static int cake_config_change(struct cake_sched_config *q, struct nlattr *opt, @@ -2780,10 +2782,10 @@ static int cake_change(struct Qdisc *sch, struct nlattr *opt, return ret; if (overhead_changed) { - qd->max_netlen = 0; - qd->max_adjlen = 0; - qd->min_netlen = ~0; - qd->min_adjlen = ~0; + WRITE_ONCE(qd->max_netlen, 0); + WRITE_ONCE(qd->max_adjlen, 0); + WRITE_ONCE(qd->min_netlen, ~0); + WRITE_ONCE(qd->min_adjlen, ~0); } if (qd->tins) { @@ -3001,15 +3003,15 @@ static int cake_dump_stats(struct Qdisc *sch, struct gnet_dump *d) goto nla_put_failure; \ } while (0) - PUT_STAT_U64(CAPACITY_ESTIMATE64, q->avg_peak_bandwidth); - PUT_STAT_U32(MEMORY_LIMIT, q->buffer_limit); - PUT_STAT_U32(MEMORY_USED, q->buffer_max_used); - PUT_STAT_U32(AVG_NETOFF, ((q->avg_netoff + 0x8000) >> 16)); - PUT_STAT_U32(MAX_NETLEN, q->max_netlen); - PUT_STAT_U32(MAX_ADJLEN, q->max_adjlen); - PUT_STAT_U32(MIN_NETLEN, q->min_netlen); - PUT_STAT_U32(MIN_ADJLEN, q->min_adjlen); - PUT_STAT_U32(ACTIVE_QUEUES, q->active_queues); + PUT_STAT_U64(CAPACITY_ESTIMATE64, READ_ONCE(q->avg_peak_bandwidth)); + PUT_STAT_U32(MEMORY_LIMIT, READ_ONCE(q->buffer_limit)); + PUT_STAT_U32(MEMORY_USED, READ_ONCE(q->buffer_max_used)); + PUT_STAT_U32(AVG_NETOFF, ((READ_ONCE(q->avg_netoff) + 0x8000) >> 16)); + PUT_STAT_U32(MAX_NETLEN, READ_ONCE(q->max_netlen)); + PUT_STAT_U32(MAX_ADJLEN, READ_ONCE(q->max_adjlen)); + PUT_STAT_U32(MIN_NETLEN, READ_ONCE(q->min_netlen)); + PUT_STAT_U32(MIN_ADJLEN, READ_ONCE(q->min_adjlen)); + PUT_STAT_U32(ACTIVE_QUEUES, READ_ONCE(q->active_queues)); #undef PUT_STAT_U32 #undef PUT_STAT_U64 @@ -3035,9 +3037,9 @@ static int cake_dump_stats(struct Qdisc *sch, struct gnet_dump *d) if (!ts) goto nla_put_failure; - PUT_TSTAT_U64(THRESHOLD_RATE64, b->tin_rate_bps); - PUT_TSTAT_U64(SENT_BYTES64, b->bytes); - PUT_TSTAT_U32(BACKLOG_BYTES, b->tin_backlog); + PUT_TSTAT_U64(THRESHOLD_RATE64, READ_ONCE(b->tin_rate_bps)); + PUT_TSTAT_U64(SENT_BYTES64, READ_ONCE(b->bytes)); + PUT_TSTAT_U32(BACKLOG_BYTES, READ_ONCE(b->tin_backlog)); PUT_TSTAT_U32(TARGET_US, ktime_to_us(ns_to_ktime(b->cparams.target))); @@ -3304,10 +3306,10 @@ static int cake_mq_change(struct Qdisc *sch, struct nlattr *opt, struct cake_sched_data *qd = qdisc_priv(chld); if (overhead_changed) { - qd->max_netlen = 0; - qd->max_adjlen = 0; - qd->min_netlen = ~0; - qd->min_adjlen = ~0; + WRITE_ONCE(qd->max_netlen, 0); + WRITE_ONCE(qd->max_adjlen, 0); + WRITE_ONCE(qd->min_netlen, ~0); + WRITE_ONCE(qd->min_adjlen, ~0); } if (qd->tins) { -- cgit v1.2.3 From a6c95b833dc17e84d16a8ac0f40fd0931616a52d Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 27 Apr 2026 08:36:06 +0000 Subject: net/sched: sch_cake: annotate data-races in cake_dump_stats() (V) MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit cake_dump_stats() runs without qdisc spinlock being held. In this final patch, I add READ_ONCE()/WRITE_ONCE() annotations for cparams.target and cparams.interval. Fixes: 046f6fd5daef ("sched: Add Common Applications Kept Enhanced (cake) qdisc") Signed-off-by: Eric Dumazet Acked-by: "Toke Høiland-Jørgensen" Link: https://patch.msgid.link/20260427083606.459355-6-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/sched/sch_cake.c | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/net/sched/sch_cake.c b/net/sched/sch_cake.c index 975f5d6d6982..13c6d1869a14 100644 --- a/net/sched/sch_cake.c +++ b/net/sched/sch_cake.c @@ -2356,10 +2356,11 @@ static void cake_set_rate(struct cake_tin_data *b, u64 rate, u32 mtu, byte_target_ns = (byte_target * rate_ns) >> rate_shft; - b->cparams.target = max((byte_target_ns * 3) / 2, target_ns); - b->cparams.interval = max(rtt_est_ns + - b->cparams.target - target_ns, - b->cparams.target * 2); + WRITE_ONCE(b->cparams.target, + max((byte_target_ns * 3) / 2, target_ns)); + WRITE_ONCE(b->cparams.interval, + max(rtt_est_ns + b->cparams.target - target_ns, + b->cparams.target * 2)); b->cparams.mtu_time = byte_target_ns; b->cparams.p_inc = 1 << 24; /* 1/256 */ b->cparams.p_dec = 1 << 20; /* 1/4096 */ @@ -3042,9 +3043,9 @@ static int cake_dump_stats(struct Qdisc *sch, struct gnet_dump *d) PUT_TSTAT_U32(BACKLOG_BYTES, READ_ONCE(b->tin_backlog)); PUT_TSTAT_U32(TARGET_US, - ktime_to_us(ns_to_ktime(b->cparams.target))); + ktime_to_us(ns_to_ktime(READ_ONCE(b->cparams.target)))); PUT_TSTAT_U32(INTERVAL_US, - ktime_to_us(ns_to_ktime(b->cparams.interval))); + ktime_to_us(ns_to_ktime(READ_ONCE(b->cparams.interval)))); PUT_TSTAT_U32(SENT_PACKETS, READ_ONCE(b->packets)); PUT_TSTAT_U32(DROPPED_PACKETS, READ_ONCE(b->tin_dropped)); -- cgit v1.2.3 From 5f95c21fc23a7ef22b4d27d1ed9bb55557ffb926 Mon Sep 17 00:00:00 2001 From: Gang Yan Date: Mon, 27 Apr 2026 21:54:33 +0200 Subject: mptcp: sockopt: set timestamp flags on subflow socket, not msk Both mptcp_setsockopt_sol_socket_tstamp() and mptcp_setsockopt_sol_socket_timestamping() iterate over subflows, acquire the subflow socket lock, but then erroneously pass the MPTCP msk socket to sock_set_timestamp() / sock_set_timestamping() instead of the subflow ssk. As a result, the timestamp flags are set on the wrong socket and have no effect on the actual subflows. Pass ssk instead of sk to both helpers. Fixes: 9061f24bf82e ("mptcp: sockopt: propagate timestamp request to subflows") Cc: stable@vger.kernel.org Signed-off-by: Gang Yan Reviewed-by: Matthieu Baerts (NGI0) Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20260427-net-mptcp-misc-fixes-7-1-rc2-v1-1-7432b7f279fa@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/sockopt.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/mptcp/sockopt.c b/net/mptcp/sockopt.c index de90a2897d2d..79db15903e7a 100644 --- a/net/mptcp/sockopt.c +++ b/net/mptcp/sockopt.c @@ -161,7 +161,7 @@ static int mptcp_setsockopt_sol_socket_tstamp(struct mptcp_sock *msk, int optnam struct sock *ssk = mptcp_subflow_tcp_sock(subflow); bool slow = lock_sock_fast(ssk); - sock_set_timestamp(sk, optname, !!val); + sock_set_timestamp(ssk, optname, !!val); unlock_sock_fast(ssk, slow); } @@ -237,7 +237,7 @@ static int mptcp_setsockopt_sol_socket_timestamping(struct mptcp_sock *msk, struct sock *ssk = mptcp_subflow_tcp_sock(subflow); bool slow = lock_sock_fast(ssk); - sock_set_timestamping(sk, optname, timestamping); + sock_set_timestamping(ssk, optname, timestamping); unlock_sock_fast(ssk, slow); } -- cgit v1.2.3 From b5c52908d52c6c8eb8933264aa6087a0600fd892 Mon Sep 17 00:00:00 2001 From: Gang Yan Date: Mon, 27 Apr 2026 21:54:34 +0200 Subject: mptcp: fix scheduling with atomic in timestamp sockopt Using lock_sock_fast() (atomic context) around sock_set_timestamp() and sock_set_timestamping() is unsafe, as both helpers can sleep. Replace lock_sock_fast() with sleepable lock_sock()/release_sock() to avoid scheduling while atomic panic. Fixes: 9061f24bf82e ("mptcp: sockopt: propagate timestamp request to subflows") Cc: stable@vger.kernel.org Reported-by: Sashiko Closes: https://sashiko.dev/#/patchset/20260420093343.16443-1-gang.yan@linux.dev Signed-off-by: Gang Yan Reviewed-by: Matthieu Baerts (NGI0) Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20260427-net-mptcp-misc-fixes-7-1-rc2-v1-2-7432b7f279fa@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/sockopt.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/mptcp/sockopt.c b/net/mptcp/sockopt.c index 79db15903e7a..0efe40be2fde 100644 --- a/net/mptcp/sockopt.c +++ b/net/mptcp/sockopt.c @@ -159,10 +159,10 @@ static int mptcp_setsockopt_sol_socket_tstamp(struct mptcp_sock *msk, int optnam lock_sock(sk); mptcp_for_each_subflow(msk, subflow) { struct sock *ssk = mptcp_subflow_tcp_sock(subflow); - bool slow = lock_sock_fast(ssk); + lock_sock(ssk); sock_set_timestamp(ssk, optname, !!val); - unlock_sock_fast(ssk, slow); + release_sock(ssk); } release_sock(sk); @@ -235,10 +235,10 @@ static int mptcp_setsockopt_sol_socket_timestamping(struct mptcp_sock *msk, mptcp_for_each_subflow(msk, subflow) { struct sock *ssk = mptcp_subflow_tcp_sock(subflow); - bool slow = lock_sock_fast(ssk); + lock_sock(ssk); sock_set_timestamping(ssk, optname, timestamping); - unlock_sock_fast(ssk, slow); + release_sock(ssk); } release_sock(sk); -- cgit v1.2.3 From f14d6e9c3678a067f304abba561e0c5446c7e845 Mon Sep 17 00:00:00 2001 From: "Matthieu Baerts (NGI0)" Date: Mon, 27 Apr 2026 21:54:35 +0200 Subject: mptcp: fastclose msk when linger time is 0 The SO_LINGER socket option has been supported for a while with MPTCP sockets [1], but it didn't cause the equivalent of a TCP reset as expected when enabled and its time was set to 0. This was causing some behavioural differences with TCP where some connections were not promptly stopped as expected. To fix that, an extra condition is checked at close() time before sending an MP_FASTCLOSE, the MPTCP equivalent of a TCP reset. Note that backporting up to [1] will be difficult as more changes are needed to be able to send MP_FASTCLOSE. It seems better to stop at [2], which was supposed to already imitate TCP. Validated with MPTCP packetdrill tests [3]. Fixes: 268b12387460 ("mptcp: setsockopt: support SO_LINGER") [1] Fixes: d21f83485518 ("mptcp: use fastclose on more edge scenarios") [2] Cc: stable@vger.kernel.org Reported-by: Lance Tuller Closes: https://github.com/lance0/xfr/pull/67 Link: https://github.com/multipath-tcp/packetdrill/pull/196 [3] Reviewed-by: Mat Martineau Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20260427-net-mptcp-misc-fixes-7-1-rc2-v1-3-7432b7f279fa@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/protocol.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index 718e910ff23f..4546a8b09884 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -3302,7 +3302,8 @@ bool __mptcp_close(struct sock *sk, long timeout) goto cleanup; } - if (mptcp_data_avail(msk) || timeout < 0) { + if (mptcp_data_avail(msk) || timeout < 0 || + (sock_flag(sk, SOCK_LINGER) && !sk->sk_lingertime)) { /* If the msk has read data, or the caller explicitly ask it, * do the MPTCP equivalent of TCP reset, aka MPTCP fastclose */ -- cgit v1.2.3 From 1774d3cf3cf17baaf30c095606cda496268283b3 Mon Sep 17 00:00:00 2001 From: "Matthieu Baerts (NGI0)" Date: Mon, 27 Apr 2026 21:54:36 +0200 Subject: mptcp: pm: kernel: reset fullmesh counter after flush This variable counts how many MPTCP endpoints have a 'fullmesh' flag set. After having flushed all MPTCP endpoints, it is then needed to reset this counter. Without this reset, this counter exposed to the userspace is wrong, but also non-fullmesh endpoints added after the flush will not be taken into account to create subflows in reaction to ADD_ADDRs. Fixes: f88191c7f361 ("mptcp: pm: in-kernel: record fullmesh endp nb") Cc: stable@vger.kernel.org Reported-by: Sashiko Closes: https://sashiko.dev/#/patchset/20260422-mptcp-inc-limits-v6-0-903181771530%40kernel.org?part=15 Reviewed-by: Mat Martineau Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20260427-net-mptcp-misc-fixes-7-1-rc2-v1-4-7432b7f279fa@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/pm_kernel.c | 1 + 1 file changed, 1 insertion(+) (limited to 'net') diff --git a/net/mptcp/pm_kernel.c b/net/mptcp/pm_kernel.c index 0ebf43be9939..c9f1e5af3cd3 100644 --- a/net/mptcp/pm_kernel.c +++ b/net/mptcp/pm_kernel.c @@ -1278,6 +1278,7 @@ static void __reset_counters(struct pm_nl_pernet *pernet) WRITE_ONCE(pernet->endp_signal_max, 0); WRITE_ONCE(pernet->endp_subflow_max, 0); WRITE_ONCE(pernet->endp_laminar_max, 0); + WRITE_ONCE(pernet->endp_fullmesh_max, 0); pernet->endpoints = 0; } -- cgit v1.2.3 From 28465227c80fe417b4013c432be1f3737cb9f9a3 Mon Sep 17 00:00:00 2001 From: Ruijie Li Date: Wed, 29 Apr 2026 00:41:43 +0800 Subject: xfrm: provide message size for XFRM_MSG_MAPPING The compat 64=>32 translation path handles XFRM_MSG_MAPPING, but xfrm_msg_min[] does not provide the native payload size for this message type. Add the missing XFRM_MSG_MAPPING entry so compat translation can size and translate mapping notifications correctly. Fixes: 5461fc0c8d9f ("xfrm/compat: Add 64=>32-bit messages translator") Cc: stable@kernel.org Reported-by: Yuan Tan Reported-by: Yifan Wu Reported-by: Juefei Pu Reported-by: Xin Liu Signed-off-by: Ruijie Li Signed-off-by: Ren Wei Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_user.c | 1 + 1 file changed, 1 insertion(+) (limited to 'net') diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index d56450f61669..38a90e5ee3d9 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -3323,6 +3323,7 @@ const int xfrm_msg_min[XFRM_NR_MSGTYPES] = { [XFRM_MSG_GETSADINFO - XFRM_MSG_BASE] = sizeof(u32), [XFRM_MSG_NEWSPDINFO - XFRM_MSG_BASE] = sizeof(u32), [XFRM_MSG_GETSPDINFO - XFRM_MSG_BASE] = sizeof(u32), + [XFRM_MSG_MAPPING - XFRM_MSG_BASE] = XMSGSIZE(xfrm_user_mapping), [XFRM_MSG_SETDEFAULT - XFRM_MSG_BASE] = XMSGSIZE(xfrm_userpolicy_default), [XFRM_MSG_GETDEFAULT - XFRM_MSG_BASE] = XMSGSIZE(xfrm_userpolicy_default), }; -- cgit v1.2.3 From 14acf9652e5690de3c7486c6db5fb8dafd0a32a3 Mon Sep 17 00:00:00 2001 From: Michal Kosiorek Date: Wed, 29 Apr 2026 10:54:51 +0200 Subject: xfrm: defensively unhash xfrm_state lists in __xfrm_state_delete KASAN reproduces a slab-use-after-free in __xfrm_state_delete()'s hlist_del_rcu calls under syzkaller load on linux-6.12.y stable (reproduced on 6.12.47, also reachable via the same code path on torvalds/master and on the ipsec tree). Nine unique signatures cluster in the xfrm_state lifecycle, the load-bearing one being: BUG: KASAN: slab-use-after-free in __hlist_del include/linux/list.h:990 [inline] BUG: KASAN: slab-use-after-free in hlist_del_rcu include/linux/rculist.h:516 [inline] BUG: KASAN: slab-use-after-free in __xfrm_state_delete net/xfrm/xfrm_state.c Write of size 8 at addr ffff8881198bcb70 by task kworker/u8:9/435 Workqueue: netns cleanup_net Call Trace: __hlist_del / hlist_del_rcu __xfrm_state_delete xfrm_state_delete xfrm_state_flush xfrm_state_fini ops_exit_list cleanup_net The other observed signatures hit the same slab object from __xfrm_state_lookup, xfrm_alloc_spi, __xfrm_state_insert and an OOB write variant of __xfrm_state_delete, all on the byseq/byspi hash chains. __xfrm_state_delete() guards its byseq and byspi unhashes with value-based predicates: if (x->km.seq) hlist_del_rcu(&x->byseq); if (x->id.spi) hlist_del_rcu(&x->byspi); while everywhere else in the file (e.g. state_cache, state_cache_input) the safer hlist_unhashed() check is used. xfrm_alloc_spi() sets x->id.spi = newspi inside xfrm_state_lock and then immediately inserts into byspi, but a path that observes x->id.spi != 0 outside of xfrm_state_lock can still skip-or-hit the byspi unhash inconsistently with whether x is actually on the list. The same holds for x->km.seq versus byseq, and the bydst/bysrc unhashes have no predicate at all, so a second __xfrm_state_delete() on the same object writes through LIST_POISON pprev. The defensive change here: - Use hlist_del_init_rcu() instead of hlist_del_rcu() on bydst, bysrc, byseq and byspi so a second deletion is a no-op rather than a write through LIST_POISON pprev. The byseq/byspi nodes are already initialised in xfrm_state_alloc(). - Test hlist_unhashed() rather than the value predicate for byseq/byspi, so the unhash decision tracks list state rather than mutable scalar fields. Empirical verification: applied this patch on top of v6.12.47, rebuilt, and re-ran the same syzkaller harness for 1h16m on a previously-crashy configuration that produced ~100 hits each of slab-use-after-free Read in xfrm_alloc_spi / Read in __xfrm_state_lookup / Write in __xfrm_state_delete. After the patch, 7.1M execs across 32 VMs at ~1550 exec/sec produced zero xfrm_state UAF/OOB hits. /proc/slabinfo confirms the xfrm_state slab is actively allocated and freed during the run (~143 KiB resident), so the fuzzer is still exercising those code paths -- they just no longer crash. Reproduction: - Linux 6.12.47 x86_64 + KASAN_GENERIC + KASAN_INLINE + KCOV - syzkaller @ 746545b8b1e4c3a128db8652b340d3df90ce61db - 32 QEMU/KVM VMs x 2 vCPU on AWS c5.metal bare metal - 9 unique signatures collected in ~9h, all within xfrm_state lifecycle Fixes: fe9f1d8779cb ("xfrm: add state hashtable keyed by seq") Fixes: 7b4dc3600e48 ("[XFRM]: Do not add a state whose SPI is zero to the SPI hash.") Reported-by: Michal Kosiorek Tested-by: Michal Kosiorek Cc: stable@vger.kernel.org Signed-off-by: Michal Kosiorek Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_state.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c index 1748d374abca..686014d39429 100644 --- a/net/xfrm/xfrm_state.c +++ b/net/xfrm/xfrm_state.c @@ -818,17 +818,17 @@ int __xfrm_state_delete(struct xfrm_state *x) spin_lock(&net->xfrm.xfrm_state_lock); list_del(&x->km.all); - hlist_del_rcu(&x->bydst); - hlist_del_rcu(&x->bysrc); - if (x->km.seq) - hlist_del_rcu(&x->byseq); + hlist_del_init_rcu(&x->bydst); + hlist_del_init_rcu(&x->bysrc); + if (!hlist_unhashed(&x->byseq)) + hlist_del_init_rcu(&x->byseq); if (!hlist_unhashed(&x->state_cache)) hlist_del_rcu(&x->state_cache); if (!hlist_unhashed(&x->state_cache_input)) hlist_del_rcu(&x->state_cache_input); - if (x->id.spi) - hlist_del_rcu(&x->byspi); + if (!hlist_unhashed(&x->byspi)) + hlist_del_init_rcu(&x->byspi); net->xfrm.state_num--; xfrm_nat_keepalive_state_updated(x); spin_unlock(&net->xfrm.xfrm_state_lock); -- cgit v1.2.3 From 1049970d7583194eedc30e45a3c898b2cb1c30ba Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Mon, 27 Apr 2026 14:34:45 +0200 Subject: netfilter: replace skb_try_make_writable() by skb_ensure_writable() skb_try_make_writable() only works on clones and uncloned packets might have their network header in paged fragments. nft_fwd needs to work for the ingress and egress hooks, but the egress hook where skb->data points to the mac header, use skb_network_offset() to include the mac header. The flowtable is fine since it already uses the transport offset. Fixes: d32de98ea70f ("netfilter: nft_fwd_netdev: allow to forward packets via neighbour layer") Fixes: 7d2086871762 ("netfilter: nf_flow_table: move ipv4 offload hook code to nf_flow_table") Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_flow_table_ip.c | 4 ++-- net/netfilter/nft_fwd_netdev.c | 5 +++-- 2 files changed, 5 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/netfilter/nf_flow_table_ip.c b/net/netfilter/nf_flow_table_ip.c index fd56d663cb5b..dbd7644fdbeb 100644 --- a/net/netfilter/nf_flow_table_ip.c +++ b/net/netfilter/nf_flow_table_ip.c @@ -524,7 +524,7 @@ static int nf_flow_offload_forward(struct nf_flowtable_ctx *ctx, return 0; } - if (skb_try_make_writable(skb, thoff + ctx->hdrsize)) + if (skb_ensure_writable(skb, thoff + ctx->hdrsize)) return -1; flow_offload_refresh(flow_table, flow, false); @@ -1037,7 +1037,7 @@ static int nf_flow_offload_ipv6_forward(struct nf_flowtable_ctx *ctx, return 0; } - if (skb_try_make_writable(skb, thoff + ctx->hdrsize)) + if (skb_ensure_writable(skb, thoff + ctx->hdrsize)) return -1; flow_offload_refresh(flow_table, flow, false); diff --git a/net/netfilter/nft_fwd_netdev.c b/net/netfilter/nft_fwd_netdev.c index 4bce36c3a6a0..2cc809303ce8 100644 --- a/net/netfilter/nft_fwd_netdev.c +++ b/net/netfilter/nft_fwd_netdev.c @@ -100,6 +100,7 @@ static void nft_fwd_neigh_eval(const struct nft_expr *expr, int oif = regs->data[priv->sreg_dev]; unsigned int verdict = NF_STOLEN; struct sk_buff *skb = pkt->skb; + int nhoff = skb_network_offset(skb); struct net_device *dev; int neigh_table; @@ -111,7 +112,7 @@ static void nft_fwd_neigh_eval(const struct nft_expr *expr, verdict = NFT_BREAK; goto out; } - if (skb_try_make_writable(skb, sizeof(*iph))) { + if (skb_ensure_writable(skb, nhoff + sizeof(*iph))) { verdict = NF_DROP; goto out; } @@ -132,7 +133,7 @@ static void nft_fwd_neigh_eval(const struct nft_expr *expr, verdict = NFT_BREAK; goto out; } - if (skb_try_make_writable(skb, sizeof(*ip6h))) { + if (skb_ensure_writable(skb, nhoff + sizeof(*ip6h))) { verdict = NF_DROP; goto out; } -- cgit v1.2.3 From 0a0b35f0bf10b4c2be607465f5c9c12c8681305b Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Mon, 27 Apr 2026 14:34:48 +0200 Subject: netfilter: nft_fwd_netdev: add device and headroom validate with neigh forwarding The ttl field has been decremented already and evaluation of this rule would proceed, just drop this packet instead if there is no destination device to forwards this packet. This is exactly what nf_dup already does in this case. Moreover, check for headroom and call skb_expand_head() like in the IP output path to ensure there is sufficient headroom when forwarding this via neigh_xmit(). Fixes: d32de98ea70f ("netfilter: nft_fwd_netdev: allow to forward packets via neighbour layer") Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nft_fwd_netdev.c | 16 ++++++++++++++-- 1 file changed, 14 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/netfilter/nft_fwd_netdev.c b/net/netfilter/nft_fwd_netdev.c index 2cc809303ce8..605b1d42abce 100644 --- a/net/netfilter/nft_fwd_netdev.c +++ b/net/netfilter/nft_fwd_netdev.c @@ -102,6 +102,7 @@ static void nft_fwd_neigh_eval(const struct nft_expr *expr, struct sk_buff *skb = pkt->skb; int nhoff = skb_network_offset(skb); struct net_device *dev; + unsigned int hh_len; int neigh_table; switch (priv->nfproto) { @@ -153,8 +154,19 @@ static void nft_fwd_neigh_eval(const struct nft_expr *expr, } dev = dev_get_by_index_rcu(nft_net(pkt), oif); - if (dev == NULL) - return; + if (dev == NULL) { + verdict = NF_DROP; + goto out; + } + + hh_len = LL_RESERVED_SPACE(dev); + if (unlikely(skb_headroom(skb) < hh_len && dev->header_ops)) { + skb = skb_expand_head(skb, hh_len); + if (!skb) { + verdict = NF_STOLEN; + goto out; + } + } skb->dev = dev; skb_clear_tstamp(skb); -- cgit v1.2.3 From 1d47b55b36d2ec73fe6901212c8b28a593c3b27c Mon Sep 17 00:00:00 2001 From: Weiming Shi Date: Mon, 27 Apr 2026 14:34:50 +0200 Subject: netfilter: nft_fwd_netdev: use recursion counter in neigh egress path nft_fwd_neigh can be used in egress chains (NF_NETDEV_EGRESS). When the forwarding rule targets the same device or two devices forward to each other, neigh_xmit() triggers dev_queue_xmit() which re-enters nf_hook_egress(), causing infinite recursion and stack overflow. Move the nf_get_nf_dup_skb_recursion() accessor and NF_RECURSION_LIMIT to the shared header nf_dup_netdev.h as a static inline, so that nft_fwd_netdev can use the recursion counter directly without exported function call overhead. Guard neigh_xmit() with the same recursion limit already used in nf_do_netdev_egress(). [ Updated to cache the nf_get_nf_dup_skb_recursion pointer. --pablo ] Fixes: f87b9464d152 ("netfilter: nft_fwd_netdev: Support egress hook") Reported-by: Xiang Mei Signed-off-by: Weiming Shi Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_dup_netdev.c | 16 ---------------- net/netfilter/nft_fwd_netdev.c | 8 ++++++++ 2 files changed, 8 insertions(+), 16 deletions(-) (limited to 'net') diff --git a/net/netfilter/nf_dup_netdev.c b/net/netfilter/nf_dup_netdev.c index e348fb90b8dc..3b0a70e154cd 100644 --- a/net/netfilter/nf_dup_netdev.c +++ b/net/netfilter/nf_dup_netdev.c @@ -13,22 +13,6 @@ #include #include -#define NF_RECURSION_LIMIT 2 - -#ifndef CONFIG_PREEMPT_RT -static u8 *nf_get_nf_dup_skb_recursion(void) -{ - return this_cpu_ptr(&softnet_data.xmit.nf_dup_skb_recursion); -} -#else - -static u8 *nf_get_nf_dup_skb_recursion(void) -{ - return ¤t->net_xmit.nf_dup_skb_recursion; -} - -#endif - static void nf_do_netdev_egress(struct sk_buff *skb, struct net_device *dev, enum nf_dev_hooks hook) { diff --git a/net/netfilter/nft_fwd_netdev.c b/net/netfilter/nft_fwd_netdev.c index 605b1d42abce..b9e88d7cf308 100644 --- a/net/netfilter/nft_fwd_netdev.c +++ b/net/netfilter/nft_fwd_netdev.c @@ -95,6 +95,7 @@ static void nft_fwd_neigh_eval(const struct nft_expr *expr, struct nft_regs *regs, const struct nft_pktinfo *pkt) { + u8 *nf_dup_skb_recursion = nf_get_nf_dup_skb_recursion(); struct nft_fwd_neigh *priv = nft_expr_priv(expr); void *addr = ®s->data[priv->sreg_addr]; int oif = regs->data[priv->sreg_dev]; @@ -153,6 +154,11 @@ static void nft_fwd_neigh_eval(const struct nft_expr *expr, goto out; } + if (*nf_dup_skb_recursion > NF_RECURSION_LIMIT) { + verdict = NF_DROP; + goto out; + } + dev = dev_get_by_index_rcu(nft_net(pkt), oif); if (dev == NULL) { verdict = NF_DROP; @@ -170,7 +176,9 @@ static void nft_fwd_neigh_eval(const struct nft_expr *expr, skb->dev = dev; skb_clear_tstamp(skb); + (*nf_dup_skb_recursion)++; neigh_xmit(neigh_table, dev, addr, skb); + (*nf_dup_skb_recursion)--; out: regs->verdict.code = verdict; } -- cgit v1.2.3 From 735a309b4bfb9e1e26636ff4a3e8a146f53c54f9 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Mon, 27 Apr 2026 19:53:20 -0700 Subject: net: add net_iov_init() and use it to initialize ->page_type Commit db359fccf212 ("mm: introduce a new page type for page pool in page type") added a page_type field to struct net_iov at the same offset as struct page::page_type, so that page_pool_set_pp_info() can call __SetPageNetpp() uniformly on both pages and net_iovs. The page-type API requires the field to hold the UINT_MAX "no type" sentinel before a type can be set; for real struct page that invariant is established by the page allocator on free. struct net_iov is not allocated through the page allocator, so the field is left as zero (io_uring zcrx, which uses __GFP_ZERO) or as slab garbage (devmem, which uses kvmalloc_objs() without zeroing). When the page pool then calls page_pool_set_pp_info() on a freshly-bound niov, __SetPageNetpp()'s VM_BUG_ON_PAGE(page->page_type != UINT_MAX) fires and the kernel BUGs. Triggered in selftests by io_uring zcrx setup through the fbnic queue restart path: kernel BUG at ./include/linux/page-flags.h:1062! RIP: 0010:page_pool_set_pp_info (./include/linux/page-flags.h:1062 net/core/page_pool.c:716) Call Trace: net_mp_niov_set_page_pool (net/core/page_pool.c:1360) io_pp_zc_alloc_netmems (io_uring/zcrx.c:1089 io_uring/zcrx.c:1110) fbnic_fill_bdq (./include/net/page_pool/helpers.h:160 drivers/net/ethernet/meta/fbnic/fbnic_txrx.c:906) __fbnic_nv_restart (drivers/net/ethernet/meta/fbnic/fbnic_txrx.c:2470 drivers/net/ethernet/meta/fbnic/fbnic_txrx.c:2874) fbnic_queue_start (drivers/net/ethernet/meta/fbnic/fbnic_txrx.c:2903) netdev_rx_queue_reconfig (net/core/netdev_rx_queue.c:137) __netif_mp_open_rxq (net/core/netdev_rx_queue.c:234) io_register_zcrx (io_uring/zcrx.c:818 io_uring/zcrx.c:903) __io_uring_register (io_uring/register.c:931) __do_sys_io_uring_register (io_uring/register.c:1029) do_syscall_64 (arch/x86/entry/syscall_64.c:63 arch/x86/entry/syscall_64.c:94) The same path is reachable through devmem dmabuf binding via netdev_nl_bind_rx_doit() -> net_devmem_bind_dmabuf_to_queue(). Add a net_iov_init() helper that stamps ->owner, ->type and the ->page_type sentinel, and use it from both the devmem and io_uring zcrx niov init loops. Fixes: db359fccf212 ("mm: introduce a new page type for page pool in page type") Acked-by: Vlastimil Babka (SUSE) Acked-by: Byungchul Park Reviewed-by: Jens Axboe Acked-by: Pavel Begunkov Link: https://patch.msgid.link/20260428025320.853452-1-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/core/devmem.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'net') diff --git a/net/core/devmem.c b/net/core/devmem.c index cde4c89bc146..468344739db2 100644 --- a/net/core/devmem.c +++ b/net/core/devmem.c @@ -297,8 +297,7 @@ net_devmem_bind_dmabuf(struct net_device *dev, for (i = 0; i < owner->area.num_niovs; i++) { niov = &owner->area.niovs[i]; - niov->type = NET_IOV_DMABUF; - niov->owner = &owner->area; + net_iov_init(niov, &owner->area, NET_IOV_DMABUF); page_pool_set_dma_addr_netmem(net_iov_to_netmem(niov), net_devmem_get_dma_addr(niov)); if (direction == DMA_TO_DEVICE) -- cgit v1.2.3 From b31681206e3f527970a7c7ed807fbf6a028fc25b Mon Sep 17 00:00:00 2001 From: Hamza Mahfooz Date: Tue, 28 Apr 2026 08:53:39 -0400 Subject: hv_sock: fix ARM64 support VMBUS ring buffers must be page aligned. Therefore, the current value of 24K presents a challenge on ARM64 kernels (with 64K pages). So, use VMBUS_RING_SIZE() to ensure they are always aligned and large enough to hold all of the relevant data. Cc: stable@vger.kernel.org Fixes: 77ffe33363c0 ("hv_sock: use HV_HYP_PAGE_SIZE for Hyper-V communication") Tested-by: Dexuan Cui Reviewed-by: Dexuan Cui Signed-off-by: Hamza Mahfooz Acked-by: Stefano Garzarella Link: https://patch.msgid.link/20260428125339.13963-1-hamzamahfooz@linux.microsoft.com Signed-off-by: Jakub Kicinski --- net/vmw_vsock/hyperv_transport.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/vmw_vsock/hyperv_transport.c b/net/vmw_vsock/hyperv_transport.c index f862988c1e86..7a8963595bf9 100644 --- a/net/vmw_vsock/hyperv_transport.c +++ b/net/vmw_vsock/hyperv_transport.c @@ -375,10 +375,10 @@ static void hvs_open_connection(struct vmbus_channel *chan) } else { sndbuf = max_t(int, sk->sk_sndbuf, RINGBUFFER_HVS_SND_SIZE); sndbuf = min_t(int, sndbuf, RINGBUFFER_HVS_MAX_SIZE); - sndbuf = ALIGN(sndbuf, HV_HYP_PAGE_SIZE); + sndbuf = VMBUS_RING_SIZE(sndbuf); rcvbuf = max_t(int, sk->sk_rcvbuf, RINGBUFFER_HVS_RCV_SIZE); rcvbuf = min_t(int, rcvbuf, RINGBUFFER_HVS_MAX_SIZE); - rcvbuf = ALIGN(rcvbuf, HV_HYP_PAGE_SIZE); + rcvbuf = VMBUS_RING_SIZE(rcvbuf); } chan->max_pkt_size = HVS_MAX_PKT_SIZE; -- cgit v1.2.3 From 5ef343614db766acdc01c56d66e780a1b43c6ac6 Mon Sep 17 00:00:00 2001 From: Hasan Basbunar Date: Tue, 28 Apr 2026 19:07:39 +0200 Subject: page_pool: fix memory-provider leak in page_pool_create_percpu() error path When page_pool_create_percpu() fails on page_pool_list(), it falls through to its err_uninit: label, which calls page_pool_uninit(). At that point page_pool_init() has already taken two references when the user requested PP_FLAG_ALLOW_UNREADABLE_NETMEM: pool->mp_ops->init(pool) static_branch_inc(&page_pool_mem_providers); Neither is undone by page_pool_uninit(); both are only undone by __page_pool_destroy() (success-side teardown). The error path therefore leaks the per-provider reference taken by mp_ops->init (io_zcrx_ifq->refs in the io_uring zcrx provider, the dmabuf binding refcount in the devmem provider) plus one increment of the page_pool_mem_providers static branch on every failure of xa_alloc_cyclic() inside page_pool_list(). The leaked io_zcrx_ifq->refs in turn pins everything io_zcrx_ifq_free() would release on cleanup: ifq->user (uid), ifq->mm_account (mmdrop), ifq->dev (device refcount), ifq->netdev_tracker (netdev refcount), and the rbuf region. The leaked static branch increment forces all subsequent page_pool_alloc_netmems() and page_pool_return_page() callers to take the slow mp_ops branch for the lifetime of the kernel. Reachable via the io_uring zcrx path: io_uring_register(IORING_REGISTER_ZCRX_IFQ) /* CAP_NET_ADMIN */ -> __io_uring_register -> io_register_zcrx -> zcrx_register_netdev -> netif_mp_open_rxq -> driver ndo_queue_mem_alloc -> page_pool_create_percpu -> page_pool_init succeeds (mp_ops->init runs, branch++) -> page_pool_list fails (xa_alloc_cyclic -ENOMEM) -> goto err_uninit <-- leak The same shape applies to the devmem dmabuf provider via mp_dmabuf_devmem_init()/mp_dmabuf_devmem_destroy(). Restore the cleanup symmetry by moving the mp_ops->destroy() and static_branch_dec() calls out of __page_pool_destroy() and into page_pool_uninit(), so page_pool_uninit() is again the strict inverse of page_pool_init(). page_pool_uninit() has only two callers (the err_uninit: path and __page_pool_destroy()), so this preserves the single-call invariant on the success path while fixing the err path. The error path of page_pool_init() itself still skips the mp_ops cleanup correctly: mp_ops->init is the last action that takes a reference before page_pool_init() returns 0, so when it returns an error neither the refcount nor the static branch has been touched. Triggering the bug requires xa_alloc_cyclic() to fail with -ENOMEM, which under normal GFP_KERNEL retry behaviour is rare. It is deterministic under CONFIG_FAULT_INJECTION with fail_page_alloc / xa fault injection, or under sustained memory pressure. The leak is silent: there is no warning, and the released kernel build continues running with a permanently-incremented static branch. Fixes: 0f9214046893 ("memory-provider: dmabuf devmem memory provider") Signed-off-by: Hasan Basbunar Link: https://patch.msgid.link/20260428170739.34881-1-basbunarhasan@gmail.com Signed-off-by: Jakub Kicinski --- net/core/page_pool.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/core/page_pool.c b/net/core/page_pool.c index 877bbf7a1938..6e576dec80db 100644 --- a/net/core/page_pool.c +++ b/net/core/page_pool.c @@ -327,6 +327,11 @@ static void page_pool_uninit(struct page_pool *pool) if (!pool->system) free_percpu(pool->recycle_stats); #endif + + if (pool->mp_ops) { + pool->mp_ops->destroy(pool); + static_branch_dec(&page_pool_mem_providers); + } } /** @@ -1146,11 +1151,6 @@ static void __page_pool_destroy(struct page_pool *pool) page_pool_unlist(pool); page_pool_uninit(pool); - if (pool->mp_ops) { - pool->mp_ops->destroy(pool); - static_branch_dec(&page_pool_mem_providers); - } - kfree(pool); } -- cgit v1.2.3 From 6813985ca456d1f5677ad9554f55805cbf27e16f Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Tue, 28 Apr 2026 17:35:18 +0200 Subject: netfilter: x_tables: add .check_hooks to matches and targets Add a new .check_hooks interface for checking if the match/target is used from the validate hook according to its configuration. Move existing conditional hook check based on the match/target configuration from .checkentry to .check_hooks for the following matches/targets: - addrtype - devgroup - physdev - policy - set - TCPMSS - SET This is a preparation patch to fix nft_compat, not functional changes are intended. Based on patch from Florian Westphal. Signed-off-by: Pablo Neira Ayuso --- net/netfilter/x_tables.c | 79 ++++++++++++++++++++++++++++++++++++++++----- net/netfilter/xt_TCPMSS.c | 33 ++++++++++--------- net/netfilter/xt_addrtype.c | 25 ++++++++++---- net/netfilter/xt_devgroup.c | 18 ++++++++--- net/netfilter/xt_physdev.c | 20 +++++++++--- net/netfilter/xt_policy.c | 24 +++++++++++--- net/netfilter/xt_set.c | 39 ++++++++++++++-------- 7 files changed, 179 insertions(+), 59 deletions(-) (limited to 'net') diff --git a/net/netfilter/x_tables.c b/net/netfilter/x_tables.c index 9f837fb5ceb4..2c67c2e6b132 100644 --- a/net/netfilter/x_tables.c +++ b/net/netfilter/x_tables.c @@ -477,11 +477,9 @@ int xt_check_proc_name(const char *name, unsigned int size) } EXPORT_SYMBOL(xt_check_proc_name); -int xt_check_match(struct xt_mtchk_param *par, - unsigned int size, u16 proto, bool inv_proto) +static int xt_check_match_common(struct xt_mtchk_param *par, + unsigned int size, u16 proto, bool inv_proto) { - int ret; - if (XT_ALIGN(par->match->matchsize) != size && par->match->matchsize != -1) { /* @@ -530,6 +528,14 @@ int xt_check_match(struct xt_mtchk_param *par, par->match->proto); return -EINVAL; } + + return 0; +} + +static int xt_checkentry_match(struct xt_mtchk_param *par) +{ + int ret; + if (par->match->checkentry != NULL) { ret = par->match->checkentry(par); if (ret < 0) @@ -538,8 +544,34 @@ int xt_check_match(struct xt_mtchk_param *par, /* Flag up potential errors. */ return -EIO; } + + return 0; +} + +int xt_check_hooks_match(struct xt_mtchk_param *par) +{ + if (par->match->check_hooks != NULL) + return par->match->check_hooks(par); + return 0; } +EXPORT_SYMBOL_GPL(xt_check_hooks_match); + +int xt_check_match(struct xt_mtchk_param *par, + unsigned int size, u16 proto, bool inv_proto) +{ + int ret; + + ret = xt_check_match_common(par, size, proto, inv_proto); + if (ret < 0) + return ret; + + ret = xt_check_hooks_match(par); + if (ret < 0) + return ret; + + return xt_checkentry_match(par); +} EXPORT_SYMBOL_GPL(xt_check_match); /** xt_check_entry_match - check that matches end before start of target @@ -1012,11 +1044,9 @@ bool xt_find_jump_offset(const unsigned int *offsets, } EXPORT_SYMBOL(xt_find_jump_offset); -int xt_check_target(struct xt_tgchk_param *par, - unsigned int size, u16 proto, bool inv_proto) +static int xt_check_target_common(struct xt_tgchk_param *par, + unsigned int size, u16 proto, bool inv_proto) { - int ret; - if (XT_ALIGN(par->target->targetsize) != size) { pr_err_ratelimited("%s_tables: %s.%u target: invalid size %u (kernel) != (user) %u\n", xt_prefix[par->family], par->target->name, @@ -1061,6 +1091,23 @@ int xt_check_target(struct xt_tgchk_param *par, par->target->proto); return -EINVAL; } + + return 0; +} + +int xt_check_hooks_target(struct xt_tgchk_param *par) +{ + if (par->target->check_hooks != NULL) + return par->target->check_hooks(par); + + return 0; +} +EXPORT_SYMBOL_GPL(xt_check_hooks_target); + +static int xt_checkentry_target(struct xt_tgchk_param *par) +{ + int ret; + if (par->target->checkentry != NULL) { ret = par->target->checkentry(par); if (ret < 0) @@ -1071,6 +1118,22 @@ int xt_check_target(struct xt_tgchk_param *par, } return 0; } + +int xt_check_target(struct xt_tgchk_param *par, + unsigned int size, u16 proto, bool inv_proto) +{ + int ret; + + ret = xt_check_target_common(par, size, proto, inv_proto); + if (ret < 0) + return ret; + + ret = xt_check_hooks_target(par); + if (ret < 0) + return ret; + + return xt_checkentry_target(par); +} EXPORT_SYMBOL_GPL(xt_check_target); /** diff --git a/net/netfilter/xt_TCPMSS.c b/net/netfilter/xt_TCPMSS.c index 116a885adb3c..80e1634bc51f 100644 --- a/net/netfilter/xt_TCPMSS.c +++ b/net/netfilter/xt_TCPMSS.c @@ -247,6 +247,21 @@ tcpmss_tg6(struct sk_buff *skb, const struct xt_action_param *par) } #endif +static int tcpmss_tg4_check_hooks(const struct xt_tgchk_param *par) +{ + const struct xt_tcpmss_info *info = par->targinfo; + + if (info->mss == XT_TCPMSS_CLAMP_PMTU && + (par->hook_mask & ~((1 << NF_INET_FORWARD) | + (1 << NF_INET_LOCAL_OUT) | + (1 << NF_INET_POST_ROUTING))) != 0) { + pr_info_ratelimited("path-MTU clamping only supported in FORWARD, OUTPUT and POSTROUTING hooks\n"); + return -EINVAL; + } + + return 0; +} + /* Must specify -p tcp --syn */ static inline bool find_syn_match(const struct xt_entry_match *m) { @@ -262,17 +277,9 @@ static inline bool find_syn_match(const struct xt_entry_match *m) static int tcpmss_tg4_check(const struct xt_tgchk_param *par) { - const struct xt_tcpmss_info *info = par->targinfo; const struct ipt_entry *e = par->entryinfo; const struct xt_entry_match *ematch; - if (info->mss == XT_TCPMSS_CLAMP_PMTU && - (par->hook_mask & ~((1 << NF_INET_FORWARD) | - (1 << NF_INET_LOCAL_OUT) | - (1 << NF_INET_POST_ROUTING))) != 0) { - pr_info_ratelimited("path-MTU clamping only supported in FORWARD, OUTPUT and POSTROUTING hooks\n"); - return -EINVAL; - } if (par->nft_compat) return 0; @@ -286,17 +293,9 @@ static int tcpmss_tg4_check(const struct xt_tgchk_param *par) #if IS_ENABLED(CONFIG_IP6_NF_IPTABLES) static int tcpmss_tg6_check(const struct xt_tgchk_param *par) { - const struct xt_tcpmss_info *info = par->targinfo; const struct ip6t_entry *e = par->entryinfo; const struct xt_entry_match *ematch; - if (info->mss == XT_TCPMSS_CLAMP_PMTU && - (par->hook_mask & ~((1 << NF_INET_FORWARD) | - (1 << NF_INET_LOCAL_OUT) | - (1 << NF_INET_POST_ROUTING))) != 0) { - pr_info_ratelimited("path-MTU clamping only supported in FORWARD, OUTPUT and POSTROUTING hooks\n"); - return -EINVAL; - } if (par->nft_compat) return 0; @@ -312,6 +311,7 @@ static struct xt_target tcpmss_tg_reg[] __read_mostly = { { .family = NFPROTO_IPV4, .name = "TCPMSS", + .check_hooks = tcpmss_tg4_check_hooks, .checkentry = tcpmss_tg4_check, .target = tcpmss_tg4, .targetsize = sizeof(struct xt_tcpmss_info), @@ -322,6 +322,7 @@ static struct xt_target tcpmss_tg_reg[] __read_mostly = { { .family = NFPROTO_IPV6, .name = "TCPMSS", + .check_hooks = tcpmss_tg4_check_hooks, .checkentry = tcpmss_tg6_check, .target = tcpmss_tg6, .targetsize = sizeof(struct xt_tcpmss_info), diff --git a/net/netfilter/xt_addrtype.c b/net/netfilter/xt_addrtype.c index a77088943107..913dbe3aa5e2 100644 --- a/net/netfilter/xt_addrtype.c +++ b/net/netfilter/xt_addrtype.c @@ -153,14 +153,10 @@ addrtype_mt_v1(const struct sk_buff *skb, struct xt_action_param *par) return ret; } -static int addrtype_mt_checkentry_v1(const struct xt_mtchk_param *par) +static int addrtype_mt_check_hooks(const struct xt_mtchk_param *par) { - const char *errmsg = "both incoming and outgoing interface limitation cannot be selected"; struct xt_addrtype_info_v1 *info = par->matchinfo; - - if (info->flags & XT_ADDRTYPE_LIMIT_IFACE_IN && - info->flags & XT_ADDRTYPE_LIMIT_IFACE_OUT) - goto err; + const char *errmsg; if (par->hook_mask & ((1 << NF_INET_PRE_ROUTING) | (1 << NF_INET_LOCAL_IN)) && @@ -176,6 +172,21 @@ static int addrtype_mt_checkentry_v1(const struct xt_mtchk_param *par) goto err; } + return 0; +err: + pr_info_ratelimited("%s\n", errmsg); + return -EINVAL; +} + +static int addrtype_mt_checkentry_v1(const struct xt_mtchk_param *par) +{ + const char *errmsg = "both incoming and outgoing interface limitation cannot be selected"; + struct xt_addrtype_info_v1 *info = par->matchinfo; + + if (info->flags & XT_ADDRTYPE_LIMIT_IFACE_IN && + info->flags & XT_ADDRTYPE_LIMIT_IFACE_OUT) + goto err; + #if IS_ENABLED(CONFIG_IP6_NF_IPTABLES) if (par->family == NFPROTO_IPV6) { if ((info->source | info->dest) & XT_ADDRTYPE_BLACKHOLE) { @@ -211,6 +222,7 @@ static struct xt_match addrtype_mt_reg[] __read_mostly = { .family = NFPROTO_IPV4, .revision = 1, .match = addrtype_mt_v1, + .check_hooks = addrtype_mt_check_hooks, .checkentry = addrtype_mt_checkentry_v1, .matchsize = sizeof(struct xt_addrtype_info_v1), .me = THIS_MODULE @@ -221,6 +233,7 @@ static struct xt_match addrtype_mt_reg[] __read_mostly = { .family = NFPROTO_IPV6, .revision = 1, .match = addrtype_mt_v1, + .check_hooks = addrtype_mt_check_hooks, .checkentry = addrtype_mt_checkentry_v1, .matchsize = sizeof(struct xt_addrtype_info_v1), .me = THIS_MODULE diff --git a/net/netfilter/xt_devgroup.c b/net/netfilter/xt_devgroup.c index 9520dd00070b..6d1a44ab5eee 100644 --- a/net/netfilter/xt_devgroup.c +++ b/net/netfilter/xt_devgroup.c @@ -33,14 +33,10 @@ static bool devgroup_mt(const struct sk_buff *skb, struct xt_action_param *par) return true; } -static int devgroup_mt_checkentry(const struct xt_mtchk_param *par) +static int devgroup_mt_check_hooks(const struct xt_mtchk_param *par) { const struct xt_devgroup_info *info = par->matchinfo; - if (info->flags & ~(XT_DEVGROUP_MATCH_SRC | XT_DEVGROUP_INVERT_SRC | - XT_DEVGROUP_MATCH_DST | XT_DEVGROUP_INVERT_DST)) - return -EINVAL; - if (info->flags & XT_DEVGROUP_MATCH_SRC && par->hook_mask & ~((1 << NF_INET_PRE_ROUTING) | (1 << NF_INET_LOCAL_IN) | @@ -56,9 +52,21 @@ static int devgroup_mt_checkentry(const struct xt_mtchk_param *par) return 0; } +static int devgroup_mt_checkentry(const struct xt_mtchk_param *par) +{ + const struct xt_devgroup_info *info = par->matchinfo; + + if (info->flags & ~(XT_DEVGROUP_MATCH_SRC | XT_DEVGROUP_INVERT_SRC | + XT_DEVGROUP_MATCH_DST | XT_DEVGROUP_INVERT_DST)) + return -EINVAL; + + return 0; +} + static struct xt_match devgroup_mt_reg __read_mostly = { .name = "devgroup", .match = devgroup_mt, + .check_hooks = devgroup_mt_check_hooks, .checkentry = devgroup_mt_checkentry, .matchsize = sizeof(struct xt_devgroup_info), .family = NFPROTO_UNSPEC, diff --git a/net/netfilter/xt_physdev.c b/net/netfilter/xt_physdev.c index d2b0b52434fa..dd98f758176c 100644 --- a/net/netfilter/xt_physdev.c +++ b/net/netfilter/xt_physdev.c @@ -91,14 +91,10 @@ match_outdev: return (!!ret ^ !(info->invert & XT_PHYSDEV_OP_OUT)); } -static int physdev_mt_check(const struct xt_mtchk_param *par) +static int physdev_mt_check_hooks(const struct xt_mtchk_param *par) { const struct xt_physdev_info *info = par->matchinfo; - static bool brnf_probed __read_mostly; - if (!(info->bitmask & XT_PHYSDEV_OP_MASK) || - info->bitmask & ~XT_PHYSDEV_OP_MASK) - return -EINVAL; if (info->bitmask & (XT_PHYSDEV_OP_OUT | XT_PHYSDEV_OP_ISOUT) && (!(info->bitmask & XT_PHYSDEV_OP_BRIDGED) || info->invert & XT_PHYSDEV_OP_BRIDGED) && @@ -107,6 +103,18 @@ static int physdev_mt_check(const struct xt_mtchk_param *par) return -EINVAL; } + return 0; +} + +static int physdev_mt_check(const struct xt_mtchk_param *par) +{ + const struct xt_physdev_info *info = par->matchinfo; + static bool brnf_probed __read_mostly; + + if (!(info->bitmask & XT_PHYSDEV_OP_MASK) || + info->bitmask & ~XT_PHYSDEV_OP_MASK) + return -EINVAL; + #define X(memb) strnlen(info->memb, sizeof(info->memb)) >= sizeof(info->memb) if (info->bitmask & XT_PHYSDEV_OP_IN) { if (info->physindev[0] == '\0') @@ -141,6 +149,7 @@ static struct xt_match physdev_mt_reg[] __read_mostly = { { .name = "physdev", .family = NFPROTO_IPV4, + .check_hooks = physdev_mt_check_hooks, .checkentry = physdev_mt_check, .match = physdev_mt, .matchsize = sizeof(struct xt_physdev_info), @@ -149,6 +158,7 @@ static struct xt_match physdev_mt_reg[] __read_mostly = { { .name = "physdev", .family = NFPROTO_IPV6, + .check_hooks = physdev_mt_check_hooks, .checkentry = physdev_mt_check, .match = physdev_mt, .matchsize = sizeof(struct xt_physdev_info), diff --git a/net/netfilter/xt_policy.c b/net/netfilter/xt_policy.c index b5fa65558318..ff54e3a8581e 100644 --- a/net/netfilter/xt_policy.c +++ b/net/netfilter/xt_policy.c @@ -126,13 +126,10 @@ policy_mt(const struct sk_buff *skb, struct xt_action_param *par) return ret; } -static int policy_mt_check(const struct xt_mtchk_param *par) +static int policy_mt_check_hooks(const struct xt_mtchk_param *par) { const struct xt_policy_info *info = par->matchinfo; - const char *errmsg = "neither incoming nor outgoing policy selected"; - - if (!(info->flags & (XT_POLICY_MATCH_IN|XT_POLICY_MATCH_OUT))) - goto err; + const char *errmsg; if (par->hook_mask & ((1 << NF_INET_PRE_ROUTING) | (1 << NF_INET_LOCAL_IN)) && info->flags & XT_POLICY_MATCH_OUT) { @@ -144,6 +141,21 @@ static int policy_mt_check(const struct xt_mtchk_param *par) errmsg = "input policy not valid in POSTROUTING and OUTPUT"; goto err; } + + return 0; +err: + pr_info_ratelimited("%s\n", errmsg); + return -EINVAL; +} + +static int policy_mt_check(const struct xt_mtchk_param *par) +{ + const struct xt_policy_info *info = par->matchinfo; + const char *errmsg = "neither incoming nor outgoing policy selected"; + + if (!(info->flags & (XT_POLICY_MATCH_IN|XT_POLICY_MATCH_OUT))) + goto err; + if (info->len > XT_POLICY_MAX_ELEM) { errmsg = "too many policy elements"; goto err; @@ -158,6 +170,7 @@ static struct xt_match policy_mt_reg[] __read_mostly = { { .name = "policy", .family = NFPROTO_IPV4, + .check_hooks = policy_mt_check_hooks, .checkentry = policy_mt_check, .match = policy_mt, .matchsize = sizeof(struct xt_policy_info), @@ -166,6 +179,7 @@ static struct xt_match policy_mt_reg[] __read_mostly = { { .name = "policy", .family = NFPROTO_IPV6, + .check_hooks = policy_mt_check_hooks, .checkentry = policy_mt_check, .match = policy_mt, .matchsize = sizeof(struct xt_policy_info), diff --git a/net/netfilter/xt_set.c b/net/netfilter/xt_set.c index 731bc2cafae4..4ae04bba9358 100644 --- a/net/netfilter/xt_set.c +++ b/net/netfilter/xt_set.c @@ -430,6 +430,29 @@ set_target_v3(struct sk_buff *skb, const struct xt_action_param *par) return XT_CONTINUE; } +static int +set_target_v3_check_hooks(const struct xt_tgchk_param *par) +{ + const struct xt_set_info_target_v3 *info = par->targinfo; + + if (info->map_set.index != IPSET_INVALID_ID) { + if (strncmp(par->table, "mangle", 7)) { + pr_info_ratelimited("--map-set only usable from mangle table\n"); + return -EINVAL; + } + if (((info->flags & IPSET_FLAG_MAP_SKBPRIO) | + (info->flags & IPSET_FLAG_MAP_SKBQUEUE)) && + (par->hook_mask & ~(1 << NF_INET_FORWARD | + 1 << NF_INET_LOCAL_OUT | + 1 << NF_INET_POST_ROUTING))) { + pr_info_ratelimited("mapping of prio or/and queue is allowed only from OUTPUT/FORWARD/POSTROUTING chains\n"); + return -EINVAL; + } + } + + return 0; +} + static int set_target_v3_checkentry(const struct xt_tgchk_param *par) { @@ -459,20 +482,6 @@ set_target_v3_checkentry(const struct xt_tgchk_param *par) } if (info->map_set.index != IPSET_INVALID_ID) { - if (strncmp(par->table, "mangle", 7)) { - pr_info_ratelimited("--map-set only usable from mangle table\n"); - ret = -EINVAL; - goto cleanup_del; - } - if (((info->flags & IPSET_FLAG_MAP_SKBPRIO) | - (info->flags & IPSET_FLAG_MAP_SKBQUEUE)) && - (par->hook_mask & ~(1 << NF_INET_FORWARD | - 1 << NF_INET_LOCAL_OUT | - 1 << NF_INET_POST_ROUTING))) { - pr_info_ratelimited("mapping of prio or/and queue is allowed only from OUTPUT/FORWARD/POSTROUTING chains\n"); - ret = -EINVAL; - goto cleanup_del; - } index = ip_set_nfnl_get_byindex(par->net, info->map_set.index); if (index == IPSET_INVALID_ID) { @@ -672,6 +681,7 @@ static struct xt_target set_targets[] __read_mostly = { .family = NFPROTO_IPV4, .target = set_target_v3, .targetsize = sizeof(struct xt_set_info_target_v3), + .check_hooks = set_target_v3_check_hooks, .checkentry = set_target_v3_checkentry, .destroy = set_target_v3_destroy, .me = THIS_MODULE @@ -682,6 +692,7 @@ static struct xt_target set_targets[] __read_mostly = { .family = NFPROTO_IPV6, .target = set_target_v3, .targetsize = sizeof(struct xt_set_info_target_v3), + .check_hooks = set_target_v3_check_hooks, .checkentry = set_target_v3_checkentry, .destroy = set_target_v3_destroy, .me = THIS_MODULE -- cgit v1.2.3 From 2f768d638d977eff824f64dcc9639e3fea32da8f Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Tue, 28 Apr 2026 19:04:07 +0200 Subject: netfilter: nft_compat: run xt_check_hooks_{match,target}() from .validate Several matches and one target check that the hook is correct from checkentry(), however, the basechain is only available from nft_table_validate(). This patch uses xt_check_hooks_{match,target}() from the nft_compat expression .validate path. This patch sets the table in the nft_ctx struct in nft_table_validate() which is required by this patch. Based on patch from Florian Westphal. Fixes: 0ca743a55991 ("netfilter: nf_tables: add compatibility layer for x_tables") Reported-by: Xiang Mei Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_tables_api.c | 1 + net/netfilter/nft_compat.c | 45 +++++++++++++++++++++++++++++++++---------- 2 files changed, 36 insertions(+), 10 deletions(-) (limited to 'net') diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index d20ce5c36d31..38e33c66c618 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -4205,6 +4205,7 @@ static int nft_table_validate(struct net *net, const struct nft_table *table) struct nft_chain *chain; struct nft_ctx ctx = { .net = net, + .table = (struct nft_table *)table, .family = table->family, }; int err = 0; diff --git a/net/netfilter/nft_compat.c b/net/netfilter/nft_compat.c index decc725a33c2..0caa9304d2d0 100644 --- a/net/netfilter/nft_compat.c +++ b/net/netfilter/nft_compat.c @@ -261,10 +261,10 @@ nft_target_init(const struct nft_ctx *ctx, const struct nft_expr *expr, return ret; } - nft_target_set_tgchk_param(&par, ctx, target, info, &e, proto, inv); - nft_compat_wait_for_destructors(ctx->net); + nft_target_set_tgchk_param(&par, ctx, target, info, &e, proto, inv); + ret = xt_check_target(&par, size, proto, inv); if (ret < 0) { if (ret == -ENOENT) { @@ -353,8 +353,6 @@ nla_put_failure: static int nft_target_validate(const struct nft_ctx *ctx, const struct nft_expr *expr) { - struct xt_target *target = expr->ops->data; - unsigned int hook_mask = 0; int ret; if (ctx->family != NFPROTO_IPV4 && @@ -377,11 +375,21 @@ static int nft_target_validate(const struct nft_ctx *ctx, const struct nft_base_chain *basechain = nft_base_chain(ctx->chain); const struct nf_hook_ops *ops = &basechain->ops; + unsigned int hook_mask = 1 << ops->hooknum; + struct xt_target *target = expr->ops->data; + void *info = nft_expr_priv(expr); + struct xt_tgchk_param par; + union nft_entry e = {}; - hook_mask = 1 << ops->hooknum; if (target->hooks && !(hook_mask & target->hooks)) return -EINVAL; + nft_target_set_tgchk_param(&par, ctx, target, info, &e, 0, false); + + ret = xt_check_hooks_target(&par); + if (ret < 0) + return ret; + ret = nft_compat_chain_validate_dependency(ctx, target->table); if (ret < 0) return ret; @@ -515,10 +523,10 @@ __nft_match_init(const struct nft_ctx *ctx, const struct nft_expr *expr, return ret; } - nft_match_set_mtchk_param(&par, ctx, match, info, &e, proto, inv); - nft_compat_wait_for_destructors(ctx->net); + nft_match_set_mtchk_param(&par, ctx, match, info, &e, proto, inv); + return xt_check_match(&par, size, proto, inv); } @@ -614,8 +622,6 @@ static int nft_match_large_dump(struct sk_buff *skb, static int nft_match_validate(const struct nft_ctx *ctx, const struct nft_expr *expr) { - struct xt_match *match = expr->ops->data; - unsigned int hook_mask = 0; int ret; if (ctx->family != NFPROTO_IPV4 && @@ -638,11 +644,30 @@ static int nft_match_validate(const struct nft_ctx *ctx, const struct nft_base_chain *basechain = nft_base_chain(ctx->chain); const struct nf_hook_ops *ops = &basechain->ops; + unsigned int hook_mask = 1 << ops->hooknum; + struct xt_match *match = expr->ops->data; + size_t size = XT_ALIGN(match->matchsize); + struct xt_mtchk_param par; + union nft_entry e = {}; + void *info; - hook_mask = 1 << ops->hooknum; if (match->hooks && !(hook_mask & match->hooks)) return -EINVAL; + if (NFT_EXPR_SIZE(size) > NFT_MATCH_LARGE_THRESH) { + struct nft_xt_match_priv *priv = nft_expr_priv(expr); + + info = priv->info; + } else { + info = nft_expr_priv(expr); + } + + nft_match_set_mtchk_param(&par, ctx, match, info, &e, 0, false); + + ret = xt_check_hooks_match(&par); + if (ret < 0) + return ret; + ret = nft_compat_chain_validate_dependency(ctx, match->table); if (ret < 0) return ret; -- cgit v1.2.3 From 8bedb6c46945752a688d9b0cf2021e0e68b1876c Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Tue, 28 Apr 2026 19:37:57 +0200 Subject: netfilter: xt_CT: fix usersize for v1 and v2 revision While resurrecting the conntrack-tool test cases I found following bug: In: iptables -I OUTPUT -t raw -p 13 -j CT --timeout test-generic Out: [0:0] -A OUTPUT -p 13 -j CT --timeout test Data after first four bytes of the timeout policy name is never copied to userspace because its treated as kernel-only. Fixes: ec2318904965 ("xtables: extend matches and targets with .usersize") Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- net/netfilter/xt_CT.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/netfilter/xt_CT.c b/net/netfilter/xt_CT.c index 498f5871c84a..d2aeacf94230 100644 --- a/net/netfilter/xt_CT.c +++ b/net/netfilter/xt_CT.c @@ -354,7 +354,7 @@ static struct xt_target xt_ct_tg_reg[] __read_mostly = { .family = NFPROTO_IPV4, .revision = 1, .targetsize = sizeof(struct xt_ct_target_info_v1), - .usersize = offsetof(struct xt_ct_target_info, ct), + .usersize = offsetof(struct xt_ct_target_info_v1, ct), .checkentry = xt_ct_tg_check_v1, .destroy = xt_ct_tg_destroy_v1, .target = xt_ct_target_v1, @@ -366,7 +366,7 @@ static struct xt_target xt_ct_tg_reg[] __read_mostly = { .family = NFPROTO_IPV4, .revision = 2, .targetsize = sizeof(struct xt_ct_target_info_v1), - .usersize = offsetof(struct xt_ct_target_info, ct), + .usersize = offsetof(struct xt_ct_target_info_v1, ct), .checkentry = xt_ct_tg_check_v2, .destroy = xt_ct_tg_destroy_v1, .target = xt_ct_target_v1, @@ -398,7 +398,7 @@ static struct xt_target xt_ct_tg_reg[] __read_mostly = { .family = NFPROTO_IPV6, .revision = 1, .targetsize = sizeof(struct xt_ct_target_info_v1), - .usersize = offsetof(struct xt_ct_target_info, ct), + .usersize = offsetof(struct xt_ct_target_info_v1, ct), .checkentry = xt_ct_tg_check_v1, .destroy = xt_ct_tg_destroy_v1, .target = xt_ct_target_v1, @@ -410,7 +410,7 @@ static struct xt_target xt_ct_tg_reg[] __read_mostly = { .family = NFPROTO_IPV6, .revision = 2, .targetsize = sizeof(struct xt_ct_target_info_v1), - .usersize = offsetof(struct xt_ct_target_info, ct), + .usersize = offsetof(struct xt_ct_target_info_v1, ct), .checkentry = xt_ct_tg_check_v2, .destroy = xt_ct_tg_destroy_v1, .target = xt_ct_target_v1, -- cgit v1.2.3 From 63bac027860308d1344f761cb47aabb3b30973fd Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Wed, 29 Apr 2026 08:21:35 +0200 Subject: netfilter: nf_tables: fix netdev hook allocation memleak with dormant tables sashiko says: could the related code in __nf_tables_abort() leak the struct nft_hook objects when the table is dormant? In __nf_tables_abort(), when rolling back a NEWCHAIN transaction that updates hooks, the code conditionally unregisters and frees the hooks only if the table is not dormant [..] if (!(table->flags & NFT_TABLE_F_DORMANT)) { nft_netdev_unregister_hooks(net, &nft_trans_chain_hooks(trans), true); } ... nft_trans_destroy(trans); Unfortunately netdev family mixes hook registration and allocation. Push table struct down and only check for the flag to unregister. Fixes: 216e7bf7402c ("netfilter: nf_tables: skip netdev hook unregistration if table is dormant") Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_tables_api.c | 34 ++++++++++++++++++++-------------- 1 file changed, 20 insertions(+), 14 deletions(-) (limited to 'net') diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index 38e33c66c618..87387adbca65 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -407,6 +407,7 @@ static void nft_netdev_unregister_trans_hook(struct net *net, } static void nft_netdev_unregister_hooks(struct net *net, + const struct nft_table *table, struct list_head *hook_list, bool release_netdev) { @@ -414,8 +415,10 @@ static void nft_netdev_unregister_hooks(struct net *net, struct nf_hook_ops *ops; list_for_each_entry_safe(hook, next, hook_list, list) { - list_for_each_entry(ops, &hook->ops_list, list) - nf_unregister_net_hook(net, ops); + if (!(table->flags & NFT_TABLE_F_DORMANT)) { + list_for_each_entry(ops, &hook->ops_list, list) + nf_unregister_net_hook(net, ops); + } if (release_netdev) nft_netdev_hook_unlink_free_rcu(hook); } @@ -452,20 +455,25 @@ static void __nf_tables_unregister_hook(struct net *net, struct nft_base_chain *basechain; const struct nf_hook_ops *ops; - if (table->flags & NFT_TABLE_F_DORMANT || - !nft_is_base_chain(chain)) + if (!nft_is_base_chain(chain)) return; basechain = nft_base_chain(chain); ops = &basechain->ops; + /* must also be called for dormant tables */ + if (nft_base_chain_netdev(table->family, basechain->ops.hooknum)) { + nft_netdev_unregister_hooks(net, table, &basechain->hook_list, + release_netdev); + return; + } + + if (table->flags & NFT_TABLE_F_DORMANT) + return; + if (basechain->type->ops_unregister) return basechain->type->ops_unregister(net, ops); - if (nft_base_chain_netdev(table->family, basechain->ops.hooknum)) - nft_netdev_unregister_hooks(net, &basechain->hook_list, - release_netdev); - else - nf_unregister_net_hook(net, &basechain->ops); + nf_unregister_net_hook(net, &basechain->ops); } static void nf_tables_unregister_hook(struct net *net, @@ -11282,11 +11290,9 @@ static int __nf_tables_abort(struct net *net, enum nfnl_abort_action action) break; case NFT_MSG_NEWCHAIN: if (nft_trans_chain_update(trans)) { - if (!(table->flags & NFT_TABLE_F_DORMANT)) { - nft_netdev_unregister_hooks(net, - &nft_trans_chain_hooks(trans), - true); - } + nft_netdev_unregister_hooks(net, table, + &nft_trans_chain_hooks(trans), + true); free_percpu(nft_trans_chain_stats(trans)); kfree(nft_trans_chain_name(trans)); nft_trans_destroy(trans); -- cgit v1.2.3 From 58689498ca3384851145a754dbb1d8ed1cf9fb54 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 28 Apr 2026 16:15:59 -0700 Subject: net: tls: fix strparser anchor skb leak on offload RX setup failure When tls_set_device_offload_rx() fails at tls_dev_add(), the error path calls tls_sw_free_resources_rx() to clean up the SW context that was initialized by tls_set_sw_offload(). This function calls tls_sw_release_resources_rx() (which stops the strparser via tls_strp_stop()) and tls_sw_free_ctx_rx() (which kfrees the context), but never frees the anchor skb that was allocated by alloc_skb(0) in tls_strp_init(). Note that tls_sw_free_resources_rx() is exclusively used for this "failed to start offload" code path, there's no other caller. The leak did not exist before commit 84c61fe1a75b ("tls: rx: do not use the standard strparser"), because the standard strparser doesn't try to pre-allocate an skb. The normal close path in tls_sk_proto_close() handles cleanup by calling tls_sw_strparser_done() (which calls tls_strp_done()) after dropping the socket lock, because tls_strp_done() does cancel_work_sync() and the strparser work handler takes the socket lock. Fixes: 84c61fe1a75b ("tls: rx: do not use the standard strparser") Signed-off-by: Jakub Kicinski Reviewed-by: Vadim Fedorenko Link: https://patch.msgid.link/20260428231559.1358502-1-kuba@kernel.org Signed-off-by: Paolo Abeni --- net/tls/tls.h | 1 + net/tls/tls_strp.c | 6 ++++++ net/tls/tls_sw.c | 4 ++++ 3 files changed, 11 insertions(+) (limited to 'net') diff --git a/net/tls/tls.h b/net/tls/tls.h index e8f81a006520..12f44cb649c9 100644 --- a/net/tls/tls.h +++ b/net/tls/tls.h @@ -188,6 +188,7 @@ int tls_strp_dev_init(void); void tls_strp_dev_exit(void); void tls_strp_done(struct tls_strparser *strp); +void __tls_strp_done(struct tls_strparser *strp); void tls_strp_stop(struct tls_strparser *strp); int tls_strp_init(struct tls_strparser *strp, struct sock *sk); void tls_strp_data_ready(struct tls_strparser *strp); diff --git a/net/tls/tls_strp.c b/net/tls/tls_strp.c index 98e12f0ff57e..c72e88317627 100644 --- a/net/tls/tls_strp.c +++ b/net/tls/tls_strp.c @@ -624,6 +624,12 @@ void tls_strp_done(struct tls_strparser *strp) WARN_ON(!strp->stopped); cancel_work_sync(&strp->work); + __tls_strp_done(strp); +} + +/* For setup error paths where the strparser was initialized but never armed. */ +void __tls_strp_done(struct tls_strparser *strp) +{ tls_strp_anchor_free(strp); } diff --git a/net/tls/tls_sw.c b/net/tls/tls_sw.c index 94d2ae0daa8c..798243eabb1f 100644 --- a/net/tls/tls_sw.c +++ b/net/tls/tls_sw.c @@ -2624,8 +2624,12 @@ void tls_sw_free_ctx_rx(struct tls_context *tls_ctx) void tls_sw_free_resources_rx(struct sock *sk) { struct tls_context *tls_ctx = tls_get_ctx(sk); + struct tls_sw_context_rx *ctx; + + ctx = tls_sw_ctx_rx(tls_ctx); tls_sw_release_resources_rx(sk); + __tls_strp_done(&ctx->strp); tls_sw_free_ctx_rx(tls_ctx); } -- cgit v1.2.3 From 1e01abec856593e02cd69fd95b784c10dd46880c Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Wed, 29 Apr 2026 09:39:11 +0200 Subject: net/sched: cls_flower: revert unintended changes While applying the blamed commit 4ca07b9239bd ("net: mctp i2c: check length before marking flow active"), I unintentionally included unrelated and unacceptable changes. Revert them. Fixes: 4ca07b9239bd ("net: mctp i2c: check length before marking flow active") Reported-by: Jeremy Kerr Closes: https://lore.kernel.org/netdev/bd8704fe0bd53e278add5cde4873256656623e2e.camel@codeconstruct.com.au/ Signed-off-by: Paolo Abeni Link: https://patch.msgid.link/043026a53ff84da88b17648c4b0d17f0331749cb.1777447863.git.pabeni@redhat.com Signed-off-by: Paolo Abeni --- net/sched/cls_flower.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) (limited to 'net') diff --git a/net/sched/cls_flower.c b/net/sched/cls_flower.c index b9672ea05747..88f8a32fab2b 100644 --- a/net/sched/cls_flower.c +++ b/net/sched/cls_flower.c @@ -556,7 +556,6 @@ static int __fl_delete(struct tcf_proto *tp, struct cls_fl_filter *f, struct netlink_ext_ack *extack) { struct cls_fl_head *head = fl_head_dereference(tp); - struct fl_flow_mask *mask; *last = false; @@ -573,12 +572,11 @@ static int __fl_delete(struct tcf_proto *tp, struct cls_fl_filter *f, list_del_rcu(&f->list); spin_unlock(&tp->lock); - mask = f->mask; + *last = fl_mask_put(head, f->mask); if (!tc_skip_hw(f->flags)) fl_hw_destroy_filter(tp, f, rtnl_held, extack); tcf_unbind_filter(tp, &f->res); __fl_put(f); - *last = fl_mask_put(head, mask); return 0; } -- cgit v1.2.3 From 0bf00859d7a5ab685901c36f29df063b825cfaaa Mon Sep 17 00:00:00 2001 From: Fernando Fernandez Mancera Date: Tue, 28 Apr 2026 12:25:46 +0200 Subject: netfilter: nf_socket: skip socket lookup for non-first fragments Both nft_socket and xt_socket relies on L4 headers to perform socket lookup in the slow path. For fragmented packets, while the IP protocol remains constant across all fragments, only the first fragment contains the actual L4 header. As the expression/match could be attached to a chain with a priority lower than -400, it could bypass defragmentation. Add a check for fragmentation in the lookup functions directly so the problem is handled for both nft_socket and xt_socket at the same time. In addition, future users of the functions would not need to care about this. Fixes: 902d6a4c2a4f ("netfilter: nf_defrag: Skip defrag if NOTRACK is set") Fixes: 554ced0a6e29 ("netfilter: nf_tables: add support for native socket matching") Signed-off-by: Fernando Fernandez Mancera Signed-off-by: Pablo Neira Ayuso --- net/ipv4/netfilter/nf_socket_ipv4.c | 3 +++ net/ipv6/netfilter/nf_socket_ipv6.c | 5 +++-- 2 files changed, 6 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/ipv4/netfilter/nf_socket_ipv4.c b/net/ipv4/netfilter/nf_socket_ipv4.c index 5080fa5fbf6a..f9c6755f5ec5 100644 --- a/net/ipv4/netfilter/nf_socket_ipv4.c +++ b/net/ipv4/netfilter/nf_socket_ipv4.c @@ -94,6 +94,9 @@ struct sock *nf_sk_lookup_slow_v4(struct net *net, const struct sk_buff *skb, #endif int doff = 0; + if (ntohs(iph->frag_off) & IP_OFFSET) + return NULL; + if (iph->protocol == IPPROTO_UDP || iph->protocol == IPPROTO_TCP) { struct tcphdr _hdr; struct udphdr *hp; diff --git a/net/ipv6/netfilter/nf_socket_ipv6.c b/net/ipv6/netfilter/nf_socket_ipv6.c index ced8bd44828e..893f2aeb4711 100644 --- a/net/ipv6/netfilter/nf_socket_ipv6.c +++ b/net/ipv6/netfilter/nf_socket_ipv6.c @@ -100,6 +100,7 @@ struct sock *nf_sk_lookup_slow_v6(struct net *net, const struct sk_buff *skb, const struct in6_addr *daddr = NULL, *saddr = NULL; struct ipv6hdr *iph = ipv6_hdr(skb), ipv6_var; struct sk_buff *data_skb = NULL; + unsigned short fragoff = 0; int doff = 0; int thoff = 0, tproto; #if IS_ENABLED(CONFIG_NF_CONNTRACK) @@ -107,8 +108,8 @@ struct sock *nf_sk_lookup_slow_v6(struct net *net, const struct sk_buff *skb, struct nf_conn const *ct; #endif - tproto = ipv6_find_hdr(skb, &thoff, -1, NULL, NULL); - if (tproto < 0) { + tproto = ipv6_find_hdr(skb, &thoff, -1, &fragoff, NULL); + if (tproto < 0 || fragoff) { pr_debug("unable to find transport header in IPv6 packet, dropping\n"); return NULL; } -- cgit v1.2.3 From 009d203e56dbe8db2589455b9e3644955f30313a Mon Sep 17 00:00:00 2001 From: Fernando Fernandez Mancera Date: Tue, 28 Apr 2026 12:25:47 +0200 Subject: netfilter: nf_tables: skip L4 header parsing for non-first fragments The tproxy, osf and exthdr (SCTP) expressions rely on the presence of transport layer headers to perform socket lookups, fingerprint matching, or chunk extraction. For fragmented packets, while the IP protocol remains constant across all fragments, only the first fragment contains the actual L4 header. The expressions could be attached to a chain with a priority lower than -400, bypassing defragmentation. Or could be used in stateless environments where defragmentation is not happening at all. This could result in garbage data being used for the matching. Add a check for pkt->fragoff so only unfragmented packets or the first fragment is processed. Fixes: 133dc203d77d ("netfilter: nft_exthdr: Support SCTP chunks") Fixes: 4ed8eb6570a4 ("netfilter: nf_tables: Add native tproxy support") Fixes: b96af92d6eaf ("netfilter: nf_tables: implement Passive OS fingerprint module in nft_osf") Signed-off-by: Fernando Fernandez Mancera Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_tables_core.c | 2 +- net/netfilter/nft_exthdr.c | 2 +- net/netfilter/nft_osf.c | 2 +- net/netfilter/nft_tproxy.c | 8 ++++---- 4 files changed, 7 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/net/netfilter/nf_tables_core.c b/net/netfilter/nf_tables_core.c index 5ddd5b6e135f..8ab186f86dd4 100644 --- a/net/netfilter/nf_tables_core.c +++ b/net/netfilter/nf_tables_core.c @@ -153,7 +153,7 @@ static bool nft_payload_fast_eval(const struct nft_expr *expr, if (priv->base == NFT_PAYLOAD_NETWORK_HEADER) ptr = skb_network_header(skb) + pkt->nhoff; else { - if (!(pkt->flags & NFT_PKTINFO_L4PROTO)) + if (!(pkt->flags & NFT_PKTINFO_L4PROTO) || pkt->fragoff) return false; ptr = skb->data + nft_thoff(pkt); } diff --git a/net/netfilter/nft_exthdr.c b/net/netfilter/nft_exthdr.c index 0407d6f708ae..e6a07c0df207 100644 --- a/net/netfilter/nft_exthdr.c +++ b/net/netfilter/nft_exthdr.c @@ -376,7 +376,7 @@ static void nft_exthdr_sctp_eval(const struct nft_expr *expr, const struct sctp_chunkhdr *sch; struct sctp_chunkhdr _sch; - if (pkt->tprot != IPPROTO_SCTP) + if (pkt->tprot != IPPROTO_SCTP || pkt->fragoff) goto err; do { diff --git a/net/netfilter/nft_osf.c b/net/netfilter/nft_osf.c index c02d5cb52143..45fe56da5044 100644 --- a/net/netfilter/nft_osf.c +++ b/net/netfilter/nft_osf.c @@ -33,7 +33,7 @@ static void nft_osf_eval(const struct nft_expr *expr, struct nft_regs *regs, return; } - if (pkt->tprot != IPPROTO_TCP) { + if (pkt->tprot != IPPROTO_TCP || pkt->fragoff) { regs->verdict.code = NFT_BREAK; return; } diff --git a/net/netfilter/nft_tproxy.c b/net/netfilter/nft_tproxy.c index f2101af8c867..89be443734f6 100644 --- a/net/netfilter/nft_tproxy.c +++ b/net/netfilter/nft_tproxy.c @@ -30,8 +30,8 @@ static void nft_tproxy_eval_v4(const struct nft_expr *expr, __be16 tport = 0; struct sock *sk; - if (pkt->tprot != IPPROTO_TCP && - pkt->tprot != IPPROTO_UDP) { + if ((pkt->tprot != IPPROTO_TCP && + pkt->tprot != IPPROTO_UDP) || pkt->fragoff) { regs->verdict.code = NFT_BREAK; return; } @@ -97,8 +97,8 @@ static void nft_tproxy_eval_v6(const struct nft_expr *expr, memset(&taddr, 0, sizeof(taddr)); - if (pkt->tprot != IPPROTO_TCP && - pkt->tprot != IPPROTO_UDP) { + if ((pkt->tprot != IPPROTO_TCP && + pkt->tprot != IPPROTO_UDP) || pkt->fragoff) { regs->verdict.code = NFT_BREAK; return; } -- cgit v1.2.3 From 952e121c96137c73bd3e59bb20a93ef659376947 Mon Sep 17 00:00:00 2001 From: Fernando Fernandez Mancera Date: Tue, 28 Apr 2026 12:25:48 +0200 Subject: netfilter: xtables: fix L4 header parsing for non-first fragments Multiple targets and matches relies on L4 header to operate. For fragmented packets, every fragment carries the transport protocol identifier, but only the first fragment contains the L4 header. As the 'raw' table can be configured to run at priority -450 (before defragmentation at -400), the target/match can be reached before reassembly. In this case, non-first fragments have their payload incorrectly parsed as a TCP/UDP header. This would be of course a misconfiguration scenario. In most of the cases this just lead to a unreliable behavior for fragmented traffic. Add a fragment check to ensure target/match only evaluates unfragmented packets or the first fragment in the stream. Fixes: 902d6a4c2a4f ("netfilter: nf_defrag: Skip defrag if NOTRACK is set") Signed-off-by: Fernando Fernandez Mancera Signed-off-by: Pablo Neira Ayuso --- net/netfilter/xt_TPROXY.c | 11 +++++++++-- net/netfilter/xt_ecn.c | 4 ++++ net/netfilter/xt_hashlimit.c | 4 +++- net/netfilter/xt_osf.c | 3 +++ net/netfilter/xt_tcpmss.c | 4 ++++ 5 files changed, 23 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/netfilter/xt_TPROXY.c b/net/netfilter/xt_TPROXY.c index e4bea1d346cf..5f60e7298a1e 100644 --- a/net/netfilter/xt_TPROXY.c +++ b/net/netfilter/xt_TPROXY.c @@ -86,6 +86,9 @@ tproxy_tg4_v0(struct sk_buff *skb, const struct xt_action_param *par) { const struct xt_tproxy_target_info *tgi = par->targinfo; + if (par->fragoff) + return NF_DROP; + return tproxy_tg4(xt_net(par), skb, tgi->laddr, tgi->lport, tgi->mark_mask, tgi->mark_value); } @@ -95,6 +98,9 @@ tproxy_tg4_v1(struct sk_buff *skb, const struct xt_action_param *par) { const struct xt_tproxy_target_info_v1 *tgi = par->targinfo; + if (par->fragoff) + return NF_DROP; + return tproxy_tg4(xt_net(par), skb, tgi->laddr.ip, tgi->lport, tgi->mark_mask, tgi->mark_value); } @@ -106,6 +112,7 @@ tproxy_tg6_v1(struct sk_buff *skb, const struct xt_action_param *par) { const struct ipv6hdr *iph = ipv6_hdr(skb); const struct xt_tproxy_target_info_v1 *tgi = par->targinfo; + unsigned short fragoff = 0; struct udphdr _hdr, *hp; struct sock *sk; const struct in6_addr *laddr; @@ -113,8 +120,8 @@ tproxy_tg6_v1(struct sk_buff *skb, const struct xt_action_param *par) int thoff = 0; int tproto; - tproto = ipv6_find_hdr(skb, &thoff, -1, NULL, NULL); - if (tproto < 0) + tproto = ipv6_find_hdr(skb, &thoff, -1, &fragoff, NULL); + if (tproto < 0 || fragoff) return NF_DROP; hp = skb_header_pointer(skb, thoff, sizeof(_hdr), &_hdr); diff --git a/net/netfilter/xt_ecn.c b/net/netfilter/xt_ecn.c index b96e8203ac54..a8503f5d26bf 100644 --- a/net/netfilter/xt_ecn.c +++ b/net/netfilter/xt_ecn.c @@ -30,6 +30,10 @@ static bool match_tcp(const struct sk_buff *skb, struct xt_action_param *par) struct tcphdr _tcph; const struct tcphdr *th; + /* this is fine for IPv6 as ecn_mt_check6() enforces -p tcp */ + if (par->fragoff) + return false; + /* In practice, TCP match does this, so can't fail. But let's * be good citizens. */ diff --git a/net/netfilter/xt_hashlimit.c b/net/netfilter/xt_hashlimit.c index 3bd127bfc114..2704b4b60d1e 100644 --- a/net/netfilter/xt_hashlimit.c +++ b/net/netfilter/xt_hashlimit.c @@ -658,6 +658,8 @@ hashlimit_init_dst(const struct xt_hashlimit_htable *hinfo, if (!(hinfo->cfg.mode & (XT_HASHLIMIT_HASH_DPT | XT_HASHLIMIT_HASH_SPT))) return 0; + if (ntohs(ip_hdr(skb)->frag_off) & IP_OFFSET) + return -1; nexthdr = ip_hdr(skb)->protocol; break; #if IS_ENABLED(CONFIG_IP6_NF_IPTABLES) @@ -681,7 +683,7 @@ hashlimit_init_dst(const struct xt_hashlimit_htable *hinfo, return 0; nexthdr = ipv6_hdr(skb)->nexthdr; protoff = ipv6_skip_exthdr(skb, sizeof(struct ipv6hdr), &nexthdr, &frag_off); - if ((int)protoff < 0) + if ((int)protoff < 0 || ntohs(frag_off) & IP6_OFFSET) return -1; break; } diff --git a/net/netfilter/xt_osf.c b/net/netfilter/xt_osf.c index dc9485854002..e8807caede68 100644 --- a/net/netfilter/xt_osf.c +++ b/net/netfilter/xt_osf.c @@ -27,6 +27,9 @@ static bool xt_osf_match_packet(const struct sk_buff *skb, struct xt_action_param *p) { + if (p->fragoff) + return false; + return nf_osf_match(skb, xt_family(p), xt_hooknum(p), xt_in(p), xt_out(p), p->matchinfo, xt_net(p), nf_osf_fingers); } diff --git a/net/netfilter/xt_tcpmss.c b/net/netfilter/xt_tcpmss.c index 0d32d4841cb3..b9da8269161d 100644 --- a/net/netfilter/xt_tcpmss.c +++ b/net/netfilter/xt_tcpmss.c @@ -32,6 +32,10 @@ tcpmss_mt(const struct sk_buff *skb, struct xt_action_param *par) u8 _opt[15 * 4 - sizeof(_tcph)]; unsigned int i, optlen; + /* this is fine for IPv6 as xt_tcpmss enforces -p tcp */ + if (par->fragoff) + return false; + /* If we don't have the whole header, drop packet. */ th = skb_header_pointer(skb, par->thoff, sizeof(_tcph), &_tcph); if (th == NULL) -- cgit v1.2.3 From ef4f741e8627512cb8c82f59a1fc7aacd854aadf Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Thu, 30 Apr 2026 16:49:48 +0200 Subject: netfilter: flowtable: ensure sufficient headroom in xmit path Check for headroom and call skb_expand_head() like in the IP output path to ensure there is sufficient headroom for the mac header when forwarding this packet as suggested by sashiko. Fixes: b5964aac51e0 ("netfilter: flowtable: consolidate xmit path") Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_flow_table_ip.c | 13 +++++++++++-- 1 file changed, 11 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/netfilter/nf_flow_table_ip.c b/net/netfilter/nf_flow_table_ip.c index dbd7644fdbeb..8d5fb7e940a1 100644 --- a/net/netfilter/nf_flow_table_ip.c +++ b/net/netfilter/nf_flow_table_ip.c @@ -471,8 +471,17 @@ struct nf_flow_xmit { static unsigned int nf_flow_queue_xmit(struct net *net, struct sk_buff *skb, struct nf_flow_xmit *xmit) { - skb->dev = xmit->outdev; - dev_hard_header(skb, skb->dev, ntohs(skb->protocol), + struct net_device *dev = xmit->outdev; + unsigned int hh_len = LL_RESERVED_SPACE(dev); + + if (unlikely(skb_headroom(skb) < hh_len && dev->header_ops)) { + skb = skb_expand_head(skb, hh_len); + if (!skb) + return NF_STOLEN; + } + + skb->dev = dev; + dev_hard_header(skb, dev, ntohs(skb->protocol), xmit->dest, xmit->source, skb->len); dev_queue_xmit(skb); -- cgit v1.2.3 From 18ed60e33e6c77d62409c1343dec1c61bae3d2e7 Mon Sep 17 00:00:00 2001 From: Jeremy Kerr Date: Wed, 29 Apr 2026 16:21:41 +0800 Subject: net: mctp: test: use a zeroed struct sockaddr_mctp Invalid sockaddr padding will cause bind() to fail; ensure we have a zeroed address in the testcase. Fixes: 0d8647bc74cb ("net: mctp: don't require a route for null-EID ingress") Signed-off-by: Jeremy Kerr Reviewed-by: Simon Horman Link: https://patch.msgid.link/20260429-dev-mctp-test-fixes-v1-1-1127b7425809@codeconstruct.com.au Signed-off-by: Jakub Kicinski --- net/mctp/test/route-test.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/mctp/test/route-test.c b/net/mctp/test/route-test.c index e1033643fab0..e4b230ef6099 100644 --- a/net/mctp/test/route-test.c +++ b/net/mctp/test/route-test.c @@ -920,9 +920,9 @@ static void mctp_test_route_input_cloned_frag(struct kunit *test) static void mctp_test_route_input_null_eid(struct kunit *test) { struct mctp_hdr hdr = RX_HDR(1, 10, 0, FL_S | FL_E | FL_TO); + struct sockaddr_mctp addr = { 0 }; struct sk_buff *skb_pkt, *skb_sk; struct mctp_test_dev *dev; - struct sockaddr_mctp addr; struct socket *sock; u8 type = 0; int rc; -- cgit v1.2.3 From 76872971064133474d9b891da05db8f7586fcc11 Mon Sep 17 00:00:00 2001 From: Jeremy Kerr Date: Wed, 29 Apr 2026 16:21:42 +0800 Subject: net: mctp: test: Use dev_direct_xmit for TX to our test device In our test cases, we typically feed a packet sequence into the routing code, then inspect the device's TXed skbs to assert specific behaviours. Using dev_queue_xmit() for our TX path introduces a fair bit of complexity between the test packet sequence and the test device's ndo_start_xmit callback; which may mean that the skbs have not hit the device at the point we're inspecting the TXed skb list. Use dev_direct_xmit instead, as we want a direct a path as possible here, and the test dev does not need any queueing, scheduling or flow control. Fixes: 6ab578739a4c ("net: mctp: test: move TX packetqueue from dst to dev") Reported-by: kernel test robot Closes: https://lore.kernel.org/oe-lkp/202604281320.525eee17-lkp@intel.com Signed-off-by: Jeremy Kerr Reviewed-by: Simon Horman Link: https://patch.msgid.link/20260429-dev-mctp-test-fixes-v1-2-1127b7425809@codeconstruct.com.au Signed-off-by: Jakub Kicinski --- net/mctp/test/utils.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/mctp/test/utils.c b/net/mctp/test/utils.c index c3987d5ade7a..6eef8d485c25 100644 --- a/net/mctp/test/utils.c +++ b/net/mctp/test/utils.c @@ -116,7 +116,7 @@ void mctp_test_destroy_dev(struct mctp_test_dev *dev) static int mctp_test_dst_output(struct mctp_dst *dst, struct sk_buff *skb) { skb->dev = dst->dev->dev; - dev_queue_xmit(skb); + dev_direct_xmit(skb, 0); return 0; } -- cgit v1.2.3 From a177ae30f78688f75ef9c6277a152c5d6979b10e Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Thu, 30 Apr 2026 16:49:51 +0200 Subject: netfilter: flowtable: fix inline vlan encapsulation in xmit path Several issues in the inline vlan support: - The layer 2 encapsulation representation in the tuple takes encap[0] as the outer header and encap[1] as the inner header as seen from the ingress path. Reverse the encap loop to push first the inner then the outer vlan header. - Postpone pushing the layer 2 header once destination device is known. This allows to calculate the needed hearoom via LL_RESERVED_SPACE to accommodate the layer 2 headers. - Add and use nf_flow_vlan_push() as suggested by Eric Woudstra, this is a simplified version of skb_vlan_push() for egress path only. Fixes: c653d5a78f34 ("netfilter: flowtable: inline vlan encapsulation in xmit path") Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_flow_table_ip.c | 110 ++++++++++++++++++++++++++------------- 1 file changed, 73 insertions(+), 37 deletions(-) (limited to 'net') diff --git a/net/netfilter/nf_flow_table_ip.c b/net/netfilter/nf_flow_table_ip.c index 8d5fb7e940a1..0ce3c209050c 100644 --- a/net/netfilter/nf_flow_table_ip.c +++ b/net/netfilter/nf_flow_table_ip.c @@ -462,32 +462,6 @@ static void nf_flow_encap_pop(struct nf_flowtable_ctx *ctx, nf_flow_ip_tunnel_pop(ctx, skb); } -struct nf_flow_xmit { - const void *dest; - const void *source; - struct net_device *outdev; -}; - -static unsigned int nf_flow_queue_xmit(struct net *net, struct sk_buff *skb, - struct nf_flow_xmit *xmit) -{ - struct net_device *dev = xmit->outdev; - unsigned int hh_len = LL_RESERVED_SPACE(dev); - - if (unlikely(skb_headroom(skb) < hh_len && dev->header_ops)) { - skb = skb_expand_head(skb, hh_len); - if (!skb) - return NF_STOLEN; - } - - skb->dev = dev; - dev_hard_header(skb, dev, ntohs(skb->protocol), - xmit->dest, xmit->source, skb->len); - dev_queue_xmit(skb); - - return NF_STOLEN; -} - static struct flow_offload_tuple_rhash * nf_flow_offload_lookup(struct nf_flowtable_ctx *ctx, struct nf_flowtable *flow_table, struct sk_buff *skb) @@ -553,6 +527,32 @@ static int nf_flow_offload_forward(struct nf_flowtable_ctx *ctx, return 1; } +/* Similar to skb_vlan_push. */ +static int nf_flow_vlan_push(struct sk_buff *skb, __be16 proto, u16 id, + u32 needed_headroom) +{ + if (skb_vlan_tag_present(skb)) { + struct vlan_hdr *vhdr; + + if (skb_cow_head(skb, needed_headroom + VLAN_HLEN)) + return -1; + + __skb_push(skb, VLAN_HLEN); + if (skb_mac_header_was_set(skb)) + skb->mac_header -= VLAN_HLEN; + + vhdr = (struct vlan_hdr *)skb->data; + skb->network_header -= VLAN_HLEN; + vhdr->h_vlan_TCI = htons(skb_vlan_tag_get(skb)); + vhdr->h_vlan_encapsulated_proto = skb->protocol; + skb->protocol = skb->vlan_proto; + skb_postpush_rcsum(skb, skb->data, VLAN_HLEN); + } + __vlan_hwaccel_put_tag(skb, proto, id); + + return 0; +} + static int nf_flow_pppoe_push(struct sk_buff *skb, u16 id) { int data_len = skb->len + sizeof(__be16); @@ -739,17 +739,19 @@ static int nf_flow_tunnel_v6_push(struct net *net, struct sk_buff *skb, } static int nf_flow_encap_push(struct sk_buff *skb, - struct flow_offload_tuple *tuple) + struct flow_offload_tuple *tuple, + struct net_device *outdev) { + u32 needed_headroom = LL_RESERVED_SPACE(outdev); int i; - for (i = 0; i < tuple->encap_num; i++) { + for (i = tuple->encap_num - 1; i >= 0; i--) { switch (tuple->encap[i].proto) { case htons(ETH_P_8021Q): case htons(ETH_P_8021AD): - skb_reset_mac_header(skb); - if (skb_vlan_push(skb, tuple->encap[i].proto, - tuple->encap[i].id) < 0) + if (nf_flow_vlan_push(skb, tuple->encap[i].proto, + tuple->encap[i].id, + needed_headroom) < 0) return -1; break; case htons(ETH_P_PPP_SES): @@ -762,6 +764,44 @@ static int nf_flow_encap_push(struct sk_buff *skb, return 0; } +struct nf_flow_xmit { + const void *dest; + const void *source; + struct net_device *outdev; + struct flow_offload_tuple *tuple; +}; + +static void __nf_flow_queue_xmit(struct net *net, struct sk_buff *skb, + struct nf_flow_xmit *xmit) +{ + struct net_device *dev = xmit->outdev; + unsigned int hh_len = LL_RESERVED_SPACE(dev); + + if (unlikely(skb_headroom(skb) < hh_len && dev->header_ops)) { + skb = skb_expand_head(skb, hh_len); + if (!skb) + return; + } + + skb->dev = dev; + dev_hard_header(skb, dev, ntohs(skb->protocol), + xmit->dest, xmit->source, skb->len); + dev_queue_xmit(skb); +} + +static unsigned int nf_flow_queue_xmit(struct net *net, struct sk_buff *skb, + struct nf_flow_xmit *xmit) +{ + if (xmit->tuple->encap_num) { + if (nf_flow_encap_push(skb, xmit->tuple, xmit->outdev) < 0) + return NF_DROP; + } + + __nf_flow_queue_xmit(net, skb, xmit); + + return NF_STOLEN; +} + unsigned int nf_flow_offload_ip_hook(void *priv, struct sk_buff *skb, const struct nf_hook_state *state) @@ -806,9 +846,6 @@ nf_flow_offload_ip_hook(void *priv, struct sk_buff *skb, if (nf_flow_tunnel_v4_push(state->net, skb, other_tuple, &ip_daddr) < 0) return NF_DROP; - if (nf_flow_encap_push(skb, other_tuple) < 0) - return NF_DROP; - switch (tuplehash->tuple.xmit_type) { case FLOW_OFFLOAD_XMIT_NEIGH: rt = dst_rtable(tuplehash->tuple.dst_cache); @@ -838,6 +875,7 @@ nf_flow_offload_ip_hook(void *priv, struct sk_buff *skb, WARN_ON_ONCE(1); return NF_DROP; } + xmit.tuple = other_tuple; return nf_flow_queue_xmit(state->net, skb, &xmit); } @@ -1128,9 +1166,6 @@ nf_flow_offload_ipv6_hook(void *priv, struct sk_buff *skb, &ip6_daddr, encap_limit) < 0) return NF_DROP; - if (nf_flow_encap_push(skb, other_tuple) < 0) - return NF_DROP; - switch (tuplehash->tuple.xmit_type) { case FLOW_OFFLOAD_XMIT_NEIGH: rt = dst_rt6_info(tuplehash->tuple.dst_cache); @@ -1160,6 +1195,7 @@ nf_flow_offload_ipv6_hook(void *priv, struct sk_buff *skb, WARN_ON_ONCE(1); return NF_DROP; } + xmit.tuple = other_tuple; return nf_flow_queue_xmit(state->net, skb, &xmit); } -- cgit v1.2.3 From 69c54f80f4a7072b51b5b5939185ca5e572be982 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Thu, 30 Apr 2026 16:49:53 +0200 Subject: netfilter: flowtable: fix inline pppoe encapsulation in xmit path Address two issues in the inline pppoe encapsulation: - Add needs_gso_segment flag to segment PPPoE packets in software given that there is no GSO support for this. - Use FLOW_OFFLOAD_XMIT_DIRECT since neighbour cache is not available in point-to-point device, use the hardware address that is obtained via flowtable path discovery (ie. fill_forward_path). Fixes: 18d27bed0880 ("netfilter: flowtable: inline pppoe encapsulation in xmit path") Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_flow_table_core.c | 1 + net/netfilter/nf_flow_table_ip.c | 42 +++++++++++++++++++++++++++++++++++--- net/netfilter/nf_flow_table_path.c | 7 ++++++- 3 files changed, 46 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/netfilter/nf_flow_table_core.c b/net/netfilter/nf_flow_table_core.c index 2c4140e6f53c..785d8c244a77 100644 --- a/net/netfilter/nf_flow_table_core.c +++ b/net/netfilter/nf_flow_table_core.c @@ -122,6 +122,7 @@ static int flow_offload_fill_route(struct flow_offload *flow, flow_tuple->tun = route->tuple[dir].in.tun; flow_tuple->encap_num = route->tuple[dir].in.num_encaps; + flow_tuple->needs_gso_segment = route->tuple[dir].out.needs_gso_segment; flow_tuple->tun_num = route->tuple[dir].in.num_tuns; switch (route->tuple[dir].xmit_type) { diff --git a/net/netfilter/nf_flow_table_ip.c b/net/netfilter/nf_flow_table_ip.c index 0ce3c209050c..2eba64eb393a 100644 --- a/net/netfilter/nf_flow_table_ip.c +++ b/net/netfilter/nf_flow_table_ip.c @@ -553,7 +553,8 @@ static int nf_flow_vlan_push(struct sk_buff *skb, __be16 proto, u16 id, return 0; } -static int nf_flow_pppoe_push(struct sk_buff *skb, u16 id) +static int nf_flow_pppoe_push(struct sk_buff *skb, u16 id, + u32 needed_headroom) { int data_len = skb->len + sizeof(__be16); struct ppp_hdr { @@ -562,7 +563,7 @@ static int nf_flow_pppoe_push(struct sk_buff *skb, u16 id) } *ph; __be16 proto; - if (skb_cow_head(skb, PPPOE_SES_HLEN)) + if (skb_cow_head(skb, needed_headroom + PPPOE_SES_HLEN)) return -1; switch (skb->protocol) { @@ -755,7 +756,8 @@ static int nf_flow_encap_push(struct sk_buff *skb, return -1; break; case htons(ETH_P_PPP_SES): - if (nf_flow_pppoe_push(skb, tuple->encap[i].id) < 0) + if (nf_flow_pppoe_push(skb, tuple->encap[i].id, + needed_headroom) < 0) return -1; break; } @@ -769,6 +771,7 @@ struct nf_flow_xmit { const void *source; struct net_device *outdev; struct flow_offload_tuple *tuple; + bool needs_gso_segment; }; static void __nf_flow_queue_xmit(struct net *net, struct sk_buff *skb, @@ -789,10 +792,41 @@ static void __nf_flow_queue_xmit(struct net *net, struct sk_buff *skb, dev_queue_xmit(skb); } +static unsigned int nf_flow_encap_gso_xmit(struct net *net, struct sk_buff *skb, + struct nf_flow_xmit *xmit) +{ + struct sk_buff *segs, *nskb; + + segs = skb_gso_segment(skb, 0); + if (IS_ERR(segs)) + return NF_DROP; + + if (segs) + consume_skb(skb); + else + segs = skb; + + skb_list_walk_safe(segs, segs, nskb) { + skb_mark_not_on_list(segs); + + if (nf_flow_encap_push(segs, xmit->tuple, xmit->outdev) < 0) { + kfree_skb(segs); + kfree_skb_list(nskb); + return NF_STOLEN; + } + __nf_flow_queue_xmit(net, segs, xmit); + } + + return NF_STOLEN; +} + static unsigned int nf_flow_queue_xmit(struct net *net, struct sk_buff *skb, struct nf_flow_xmit *xmit) { if (xmit->tuple->encap_num) { + if (skb_is_gso(skb) && xmit->needs_gso_segment) + return nf_flow_encap_gso_xmit(net, skb, xmit); + if (nf_flow_encap_push(skb, xmit->tuple, xmit->outdev) < 0) return NF_DROP; } @@ -876,6 +910,7 @@ nf_flow_offload_ip_hook(void *priv, struct sk_buff *skb, return NF_DROP; } xmit.tuple = other_tuple; + xmit.needs_gso_segment = tuplehash->tuple.needs_gso_segment; return nf_flow_queue_xmit(state->net, skb, &xmit); } @@ -1196,6 +1231,7 @@ nf_flow_offload_ipv6_hook(void *priv, struct sk_buff *skb, return NF_DROP; } xmit.tuple = other_tuple; + xmit.needs_gso_segment = tuplehash->tuple.needs_gso_segment; return nf_flow_queue_xmit(state->net, skb, &xmit); } diff --git a/net/netfilter/nf_flow_table_path.c b/net/netfilter/nf_flow_table_path.c index 6bb9579dcc2a..9e88ea6a2eef 100644 --- a/net/netfilter/nf_flow_table_path.c +++ b/net/netfilter/nf_flow_table_path.c @@ -86,6 +86,7 @@ struct nft_forward_info { u8 ingress_vlans; u8 h_source[ETH_ALEN]; u8 h_dest[ETH_ALEN]; + bool needs_gso_segment; enum flow_offload_xmit_type xmit_type; }; @@ -138,8 +139,11 @@ static void nft_dev_path_info(const struct net_device_path_stack *stack, path->encap.proto; info->num_encaps++; } - if (path->type == DEV_PATH_PPPOE) + if (path->type == DEV_PATH_PPPOE) { memcpy(info->h_dest, path->encap.h_dest, ETH_ALEN); + info->xmit_type = FLOW_OFFLOAD_XMIT_DIRECT; + info->needs_gso_segment = 1; + } break; case DEV_PATH_BRIDGE: if (is_zero_ether_addr(info->h_source)) @@ -279,6 +283,7 @@ static void nft_dev_forward_path(const struct nft_pktinfo *pkt, memcpy(route->tuple[dir].out.h_dest, info.h_dest, ETH_ALEN); route->tuple[dir].xmit_type = info.xmit_type; } + route->tuple[dir].out.needs_gso_segment = info.needs_gso_segment; } int nft_flow_route(const struct nft_pktinfo *pkt, const struct nf_conn *ct, -- cgit v1.2.3 From 3744b0964d5267c0b651bcd8f8c25db6bf4ccbac Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Wed, 29 Apr 2026 17:46:48 +0200 Subject: ipv6: Implement limits on extension header parsing ipv6_{skip_exthdr,find_hdr}() and ip6_{tnl_parse_tlv_enc_lim, protocol_deliver_rcu}() iterate over IPv6 extension headers until they find a non-extension-header protocol or run out of packet data. The loops have no iteration counter, relying solely on the packet length to bound them. For a crafted packet with 8-byte extension headers filling a 64KB jumbogram, this means a worst case of up to ~8k iterations with a skb_header_pointer call each. ipv6_skip_exthdr(), for example, is used where it parses the inner quoted packet inside an incoming ICMPv6 error: - icmpv6_rcv - checksum validation - case ICMPV6_DEST_UNREACH - icmpv6_notify - pskb_may_pull() <- pull inner IPv6 header - ipv6_skip_exthdr() <- iterates here - pskb_may_pull() - ipprot->err_handler() <- sk lookup The per-iteration cost of ipv6_skip_exthdr itself is generally light, but skb_header_pointer becomes more costly on reassembled packets: the first ~1232 bytes of the inner packet are in the skb's linear area, but the remaining ~63KB are in the frag_list where skb_copy_bits is needed to read data. Initially, the idea was to add a configurable limit via a new sysctl knob with default 8, in line with knobs from commit 47d3d7ac656a ("ipv6: Implement limits on Hop-by-Hop and Destination options"), but two reasons eventually argued against it: - It adds to UAPI that needs to be maintained forever, and upcoming work is restricting extension header ordering anyway, leaving little reason for another sysctl knob - exthdrs_core.c is always built-in even when CONFIG_IPV6=n, where struct net has no .ipv6 member, so the read site would need an ifdef'd fallback to a constant anyway Therefore, just use a constant (IP6_MAX_EXT_HDRS_CNT). All four extension header walking functions are now bound by this limit. Note that the check in ip6_protocol_deliver_rcu() happens right before the goto resubmit, such that we don't have to have a test for ipv6_ext_hdr() in the fast-path. There's an ongoing IETF draft-iurman-6man-eh-occurrences to enforce IPv6 extension headers ordering and occurrence. The latter also discusses security implications. As per RFC8200 section 4.1, the occurrence rules for extension headers provide a practical upper bound which is 8. In order to be conservative, let's define IP6_MAX_EXT_HDRS_CNT as 12 to leave enough room for quirky setups. In the unlikely event that this is still not enough, then we might need to reconsider a sysctl. Signed-off-by: Daniel Borkmann Reviewed-by: Ido Schimmel Reviewed-by: Eric Dumazet Reviewed-by: Justin Iurman Link: https://patch.msgid.link/20260429154648.809751-1-daniel@iogearbox.net Signed-off-by: Jakub Kicinski --- net/ipv6/exthdrs_core.c | 7 +++++++ net/ipv6/ip6_input.c | 5 +++++ net/ipv6/ip6_tunnel.c | 4 ++++ 3 files changed, 16 insertions(+) (limited to 'net') diff --git a/net/ipv6/exthdrs_core.c b/net/ipv6/exthdrs_core.c index 49e31e4ae7b7..9d06d487e8b1 100644 --- a/net/ipv6/exthdrs_core.c +++ b/net/ipv6/exthdrs_core.c @@ -73,6 +73,7 @@ int ipv6_skip_exthdr(const struct sk_buff *skb, int start, u8 *nexthdrp, __be16 *frag_offp) { u8 nexthdr = *nexthdrp; + int exthdr_cnt = 0; *frag_offp = 0; @@ -82,6 +83,8 @@ int ipv6_skip_exthdr(const struct sk_buff *skb, int start, u8 *nexthdrp, if (nexthdr == NEXTHDR_NONE) return -1; + if (unlikely(exthdr_cnt++ >= IP6_MAX_EXT_HDRS_CNT)) + return -1; hp = skb_header_pointer(skb, start, sizeof(_hdr), &_hdr); if (!hp) return -1; @@ -190,6 +193,7 @@ int ipv6_find_hdr(const struct sk_buff *skb, unsigned int *offset, { unsigned int start = skb_network_offset(skb) + sizeof(struct ipv6hdr); u8 nexthdr = ipv6_hdr(skb)->nexthdr; + int exthdr_cnt = 0; bool found; if (fragoff) @@ -216,6 +220,9 @@ int ipv6_find_hdr(const struct sk_buff *skb, unsigned int *offset, return -ENOENT; } + if (unlikely(exthdr_cnt++ >= IP6_MAX_EXT_HDRS_CNT)) + return -EBADMSG; + hp = skb_header_pointer(skb, start, sizeof(_hdr), &_hdr); if (!hp) return -EBADMSG; diff --git a/net/ipv6/ip6_input.c b/net/ipv6/ip6_input.c index 967b07aeb683..8972863c93ee 100644 --- a/net/ipv6/ip6_input.c +++ b/net/ipv6/ip6_input.c @@ -403,6 +403,7 @@ INDIRECT_CALLABLE_DECLARE(int tcp_v6_rcv(struct sk_buff *)); void ip6_protocol_deliver_rcu(struct net *net, struct sk_buff *skb, int nexthdr, bool have_final) { + int exthdr_cnt = IP6CB(skb)->flags & IP6SKB_HOPBYHOP ? 1 : 0; const struct inet6_protocol *ipprot; struct inet6_dev *idev; unsigned int nhoff; @@ -487,6 +488,10 @@ resubmit_final: nexthdr = ret; goto resubmit_final; } else { + if (unlikely(exthdr_cnt++ >= IP6_MAX_EXT_HDRS_CNT)) { + SKB_DR_SET(reason, IPV6_TOO_MANY_EXTHDRS); + goto discard; + } goto resubmit; } } else if (ret == 0) { diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c index c468c83af0f2..9d1037ac082f 100644 --- a/net/ipv6/ip6_tunnel.c +++ b/net/ipv6/ip6_tunnel.c @@ -399,11 +399,15 @@ __u16 ip6_tnl_parse_tlv_enc_lim(struct sk_buff *skb, __u8 *raw) unsigned int nhoff = raw - skb->data; unsigned int off = nhoff + sizeof(*ipv6h); u8 nexthdr = ipv6h->nexthdr; + int exthdr_cnt = 0; while (ipv6_ext_hdr(nexthdr) && nexthdr != NEXTHDR_NONE) { struct ipv6_opt_hdr *hdr; u16 optlen; + if (unlikely(exthdr_cnt++ >= IP6_MAX_EXT_HDRS_CNT)) + break; + if (!pskb_may_pull(skb, off + sizeof(*hdr))) break; -- cgit v1.2.3 From baa3c65435fb3f450b262672bc06db887a92d397 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Thu, 30 Apr 2026 21:55:01 +0200 Subject: netfilter: flowtable: use skb_pull_rcsum() to pop vlan/pppoe header This adjusts the checksum, if required, after pulling the layer 2 header, either the pppoe header or the inner vlan header in the double-tagged vlan packets. Fixes: 4cd91f7c290f ("netfilter: flowtable: add vlan support") Fixes: 72efd585f714 ("netfilter: flowtable: add pppoe support") Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_flow_table_ip.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/netfilter/nf_flow_table_ip.c b/net/netfilter/nf_flow_table_ip.c index 2eba64eb393a..9c05a50d6013 100644 --- a/net/netfilter/nf_flow_table_ip.c +++ b/net/netfilter/nf_flow_table_ip.c @@ -445,13 +445,13 @@ static void nf_flow_encap_pop(struct nf_flowtable_ctx *ctx, switch (skb->protocol) { case htons(ETH_P_8021Q): vlan_hdr = (struct vlan_hdr *)skb->data; - __skb_pull(skb, VLAN_HLEN); + skb_pull_rcsum(skb, VLAN_HLEN); vlan_set_encap_proto(skb, vlan_hdr); skb_reset_network_header(skb); break; case htons(ETH_P_PPP_SES): skb->protocol = __nf_flow_pppoe_proto(skb); - skb_pull(skb, PPPOE_SES_HLEN); + skb_pull_rcsum(skb, PPPOE_SES_HLEN); skb_reset_network_header(skb); break; } -- cgit v1.2.3 From 4f34002e2e37639133693c13a2a9977fab86880d Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 30 Apr 2026 07:06:11 +0000 Subject: ipmr: prevent info-leak in pmr_cache_report() Yiming Qian reported: ipmr_cache_report()` allocates a report skb with `alloc_skb(128, GFP_ATOMIC)` and appends a `struct igmphdr` using `skb_put()`. In the non-`IGMPMSG_WHOLEPKT` path it initializes only: - `igmp->type` - `igmp->code` but does not initialize: - `igmp->csum` - `igmp->group` Later, `igmpmsg_netlink_event()` copies the bytes after `sizeof(struct igmpmsg)` into the `IPMRA_CREPORT_PKT` netlink attribute and emits `RTM_NEWCACHEREPORT` on `RTNLGRP_IPV4_MROUTE_R`. As a result, 6 bytes of stale heap data from the skb head are disclosed to userspace. Let's use skb_put_zero() instead of skb_put() to fix this bug. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Reported-by: Yiming Qian Signed-off-by: Eric Dumazet Reviewed-by: Ido Schimmel Link: https://patch.msgid.link/20260430070611.4004529-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/ipv4/ipmr.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c index 2058ca860294..05fb6eefe0be 100644 --- a/net/ipv4/ipmr.c +++ b/net/ipv4/ipmr.c @@ -1112,11 +1112,12 @@ static int ipmr_cache_report(const struct mr_table *mrt, msg->im_vif_hi = vifi >> 8; ipv4_pktinfo_prepare(mroute_sk, pkt, false); memcpy(skb->cb, pkt->cb, sizeof(skb->cb)); - /* Add our header */ - igmp = skb_put(skb, sizeof(struct igmphdr)); + /* Add our header. + * Note that code, csum and group fields are cleared. + */ + igmp = skb_put_zero(skb, sizeof(struct igmphdr)); igmp->type = assert; msg->im_msgtype = assert; - igmp->code = 0; ip_hdr(skb)->tot_len = htons(skb->len); /* Fix the length */ skb->transport_header = skb->network_header; } -- cgit v1.2.3 From 4b9e327991815e128ad3af75c3a04630a63ce3e0 Mon Sep 17 00:00:00 2001 From: Kai Zen Date: Thu, 30 Apr 2026 18:26:48 +0300 Subject: net: rtnetlink: zero ifla_vf_broadcast to avoid stack infoleak in rtnl_fill_vfinfo rtnl_fill_vfinfo() declares struct ifla_vf_broadcast on the stack without initialisation: struct ifla_vf_broadcast vf_broadcast; The struct contains a single fixed 32-byte field: /* include/uapi/linux/if_link.h */ struct ifla_vf_broadcast { __u8 broadcast[32]; }; The function then copies dev->broadcast into it using dev->addr_len as the length: memcpy(vf_broadcast.broadcast, dev->broadcast, dev->addr_len); On Ethernet devices (the overwhelming majority of SR-IOV NICs) dev->addr_len is 6, so only the first 6 bytes of broadcast[] are written. The remaining 26 bytes retain whatever was previously on the kernel stack. The full struct is then handed to userspace via: nla_put(skb, IFLA_VF_BROADCAST, sizeof(vf_broadcast), &vf_broadcast) leaking up to 26 bytes of uninitialised kernel stack per VF per RTM_GETLINK request, repeatable. The other vf_* structs in the same function are explicitly zeroed for exactly this reason - see the memset() calls for ivi, vf_vlan_info, node_guid and port_guid a few lines above. vf_broadcast was simply missed when it was added. Reachability: any unprivileged local process can open AF_NETLINK / NETLINK_ROUTE without capabilities and send RTM_GETLINK with an IFLA_EXT_MASK attribute carrying RTEXT_FILTER_VF. The kernel walks each VF and emits IFLA_VF_BROADCAST, leaking 26 bytes of stack per VF per request. Stack residue at this call site can include return addresses and transient sensitive data; KASAN with stack instrumentation, or KMSAN, will flag the nla_put() when reproduced. Zero the on-stack struct before the partial memcpy, matching the existing pattern used for the other vf_* structs in the same function. Fixes: 75345f888f70 ("ipoib: show VF broadcast address") Cc: stable@vger.kernel.org Signed-off-by: Kai Zen Link: https://patch.msgid.link/3c506e8f936e52b57620269b55c348af05d413a2.1777557228.git.kai.aizen.dev@gmail.com Signed-off-by: Jakub Kicinski --- net/core/rtnetlink.c | 1 + 1 file changed, 1 insertion(+) (limited to 'net') diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index b613bb6e07df..df042da422ef 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -1572,6 +1572,7 @@ static noinline_for_stack int rtnl_fill_vfinfo(struct sk_buff *skb, port_guid.vf = ivi.vf; memcpy(vf_mac.mac, ivi.mac, sizeof(ivi.mac)); + memset(&vf_broadcast, 0, sizeof(vf_broadcast)); memcpy(vf_broadcast.broadcast, dev->broadcast, dev->addr_len); vf_vlan.vlan = ivi.vlan; vf_vlan.qos = ivi.qos; -- cgit v1.2.3 From c6bebaa744f7579eb72800a262fbfeb93e40db04 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 30 Apr 2026 16:48:36 +0000 Subject: ipv4: igmp: annotate data-races in igmp_heard_query() Multiple cpus can run igmp_heard_query() concurrently. Add missing READ_ONCE()/WRITE_ONCE() over following in_dev fields. - mr_qrv - mr_qi - mr_qri - mr_v1_seen - mr_v2_seen Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Reported-by: syzbot+ae9a171f239b14485310@syzkaller.appspotmail.com Closes: https://lore.kernel.org/netdev/69f38675.050a0220.3cbe47.0002.GAE@google.com Signed-off-by: Eric Dumazet Link: https://patch.msgid.link/20260430164836.872079-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/ipv4/igmp.c | 58 +++++++++++++++++++++++++++++++++++---------------------- 1 file changed, 36 insertions(+), 22 deletions(-) (limited to 'net') diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c index a674fb44ec25..a9ad39064f3b 100644 --- a/net/ipv4/igmp.c +++ b/net/ipv4/igmp.c @@ -122,16 +122,29 @@ * contradict to specs provided this delay is small enough. */ -#define IGMP_V1_SEEN(in_dev) \ - (IPV4_DEVCONF_ALL_RO(dev_net(in_dev->dev), FORCE_IGMP_VERSION) == 1 || \ - IN_DEV_CONF_GET((in_dev), FORCE_IGMP_VERSION) == 1 || \ - ((in_dev)->mr_v1_seen && \ - time_before(jiffies, (in_dev)->mr_v1_seen))) -#define IGMP_V2_SEEN(in_dev) \ - (IPV4_DEVCONF_ALL_RO(dev_net(in_dev->dev), FORCE_IGMP_VERSION) == 2 || \ - IN_DEV_CONF_GET((in_dev), FORCE_IGMP_VERSION) == 2 || \ - ((in_dev)->mr_v2_seen && \ - time_before(jiffies, (in_dev)->mr_v2_seen))) +static bool IGMP_V1_SEEN(const struct in_device *in_dev) +{ + unsigned long seen; + + if (IPV4_DEVCONF_ALL_RO(dev_net(in_dev->dev), FORCE_IGMP_VERSION) == 1) + return true; + if (IN_DEV_CONF_GET((in_dev), FORCE_IGMP_VERSION) == 1) + return true; + seen = READ_ONCE(in_dev->mr_v1_seen); + return seen && time_before(jiffies, seen); +} + +static bool IGMP_V2_SEEN(const struct in_device *in_dev) +{ + unsigned long seen; + + if (IPV4_DEVCONF_ALL_RO(dev_net(in_dev->dev), FORCE_IGMP_VERSION) == 2) + return true; + if (IN_DEV_CONF_GET((in_dev), FORCE_IGMP_VERSION) == 2) + return true; + seen = READ_ONCE(in_dev->mr_v2_seen); + return seen && time_before(jiffies, seen); +} static int unsolicited_report_interval(struct in_device *in_dev) { @@ -954,23 +967,21 @@ static bool igmp_heard_query(struct in_device *in_dev, struct sk_buff *skb, int max_delay; int mark = 0; struct net *net = dev_net(in_dev->dev); - + unsigned long seen; if (len == 8) { + seen = jiffies + READ_ONCE(in_dev->mr_qrv) * READ_ONCE(in_dev->mr_qi) + + READ_ONCE(in_dev->mr_qri); if (ih->code == 0) { /* Alas, old v1 router presents here. */ max_delay = IGMP_QUERY_RESPONSE_INTERVAL; - in_dev->mr_v1_seen = jiffies + - (in_dev->mr_qrv * in_dev->mr_qi) + - in_dev->mr_qri; + WRITE_ONCE(in_dev->mr_v1_seen, seen); group = 0; } else { /* v2 router present */ max_delay = ih->code*(HZ/IGMP_TIMER_SCALE); - in_dev->mr_v2_seen = jiffies + - (in_dev->mr_qrv * in_dev->mr_qi) + - in_dev->mr_qri; + WRITE_ONCE(in_dev->mr_v2_seen, seen); } /* cancel the interface change timer */ WRITE_ONCE(in_dev->mr_ifc_count, 0); @@ -995,6 +1006,8 @@ static bool igmp_heard_query(struct in_device *in_dev, struct sk_buff *skb, if (!max_delay) max_delay = 1; /* can't mod w/ 0 */ } else { /* v3 */ + unsigned long mr_qi; + if (!pskb_may_pull(skb, sizeof(struct igmpv3_query))) return true; @@ -1015,15 +1028,16 @@ static bool igmp_heard_query(struct in_device *in_dev, struct sk_buff *skb, * received value was zero, use the default or statically * configured value. */ - in_dev->mr_qrv = ih3->qrv ?: READ_ONCE(net->ipv4.sysctl_igmp_qrv); - in_dev->mr_qi = IGMPV3_QQIC(ih3->qqic)*HZ ?: IGMP_QUERY_INTERVAL; - + WRITE_ONCE(in_dev->mr_qrv, + ih3->qrv ?: READ_ONCE(net->ipv4.sysctl_igmp_qrv)); + mr_qi = IGMPV3_QQIC(ih3->qqic)*HZ ?: IGMP_QUERY_INTERVAL; + WRITE_ONCE(in_dev->mr_qi, mr_qi); /* RFC3376, 8.3. Query Response Interval: * The number of seconds represented by the [Query Response * Interval] must be less than the [Query Interval]. */ - if (in_dev->mr_qri >= in_dev->mr_qi) - in_dev->mr_qri = (in_dev->mr_qi/HZ - 1)*HZ; + if (READ_ONCE(in_dev->mr_qri) >= mr_qi) + WRITE_ONCE(in_dev->mr_qri, (mr_qi/HZ - 1) * HZ); if (!group) { /* general query */ if (ih3->nsrcs) -- cgit v1.2.3 From 6d4106e8df94c0c52cf3ca6a6a0d01567fb3844e Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 30 Apr 2026 08:00:56 +0000 Subject: net/sched: sch_pie: annotate more data-races in pie_dump_stats() My prior patch missed few READ_ONCE()/WRITE_ONCE() annotations. Fixes: 5154561d9b11 ("net/sched: sch_pie: annotate data-races in pie_dump_stats()") Signed-off-by: Eric Dumazet Link: https://patch.msgid.link/20260430080056.35104-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/sched/sch_pie.c | 14 ++++++-------- 1 file changed, 6 insertions(+), 8 deletions(-) (limited to 'net') diff --git a/net/sched/sch_pie.c b/net/sched/sch_pie.c index fb53fbf0e328..b41f2def2e2c 100644 --- a/net/sched/sch_pie.c +++ b/net/sched/sch_pie.c @@ -219,16 +219,14 @@ void pie_process_dequeue(struct sk_buff *skb, struct pie_params *params, * packet timestamp. */ if (!params->dq_rate_estimator) { - vars->qdelay = now - pie_get_enqueue_time(skb); + WRITE_ONCE(vars->qdelay, + backlog ? now - pie_get_enqueue_time(skb) : 0); if (vars->dq_tstamp != DTIME_INVALID) dtime = now - vars->dq_tstamp; vars->dq_tstamp = now; - if (backlog == 0) - vars->qdelay = 0; - if (dtime == 0) return; @@ -376,7 +374,7 @@ void pie_calculate_probability(struct pie_params *params, struct pie_vars *vars, if (qdelay > (PSCHED_NS2TICKS(250 * NSEC_PER_MSEC))) delta += MAX_PROB / (100 / 2); - vars->prob += delta; + WRITE_ONCE(vars->prob, vars->prob + delta); if (delta > 0) { /* prevent overflow */ @@ -401,7 +399,7 @@ void pie_calculate_probability(struct pie_params *params, struct pie_vars *vars, if (qdelay == 0 && qdelay_old == 0 && update_prob) /* Reduce drop probability to 98.4% */ - vars->prob -= vars->prob / 64; + WRITE_ONCE(vars->prob, vars->prob - vars->prob / 64); WRITE_ONCE(vars->qdelay, qdelay); vars->backlog_old = backlog; @@ -501,7 +499,7 @@ static int pie_dump_stats(struct Qdisc *sch, struct gnet_dump *d) { struct pie_sched_data *q = qdisc_priv(sch); struct tc_pie_xstats st = { - .prob = q->vars.prob << BITS_PER_BYTE, + .prob = READ_ONCE(q->vars.prob) << BITS_PER_BYTE, .delay = ((u32)PSCHED_TICKS2NS(READ_ONCE(q->vars.qdelay))) / NSEC_PER_USEC, .packets_in = READ_ONCE(q->stats.packets_in), @@ -512,7 +510,7 @@ static int pie_dump_stats(struct Qdisc *sch, struct gnet_dump *d) }; /* avg_dq_rate is only valid if dq_rate_estimator is enabled */ - st.dq_rate_estimating = q->params.dq_rate_estimator; + st.dq_rate_estimating = READ_ONCE(q->params.dq_rate_estimator); /* unscale and return dq_rate in bytes per sec */ if (st.dq_rate_estimating) -- cgit v1.2.3 From 4bc852006b62eae8ea77e797192d089367e854ff Mon Sep 17 00:00:00 2001 From: Sagarika Sharma Date: Thu, 30 Apr 2026 20:09:00 +0000 Subject: ipv6: update route serial number on NETDEV_CHANGE When using IPv6 ECMP routes, if a netdev listed as a nexthop experiences a carrier change event (e.g., a bond device generating a NETDEV_CHANGE event after its slaves go linkdown), established connections utilizing that nexthop fail to fail over to other available nexthops. Instead, these connections stall or drop. This happens because the IPv6 FIB code does not invalidate the socket's cached destination when a NETDEV_CHANGE event occurs. While fib6_ifdown() correctly marks the nexthop with RTNH_F_LINKDOWN, it leaves the route's serial number unchanged. As a result, sockets with a previously cached dst do not realize the route is no longer viable and continue to try using the non-functional nexthop. This behavior contrasts with IPv4, which actively flushes cached destinations on a NETDEV_CHANGE event (see fib_netdev_event() in net/ipv4/fib_frontend.c). Fix this by updating the route serial number in fib6_ifdown() when setting RTNH_F_LINKDOWN. This invalidates stale cached destinations, forcing sockets to perform a new route lookup and fail over to a functioning nexthop. Fixes: 51ebd3181572 ("ipv6: add support of equal cost multipath (ECMP)") Signed-off-by: Sagarika Sharma Reviewed-by: Kuniyuki Iwashima Reviewed-by: Ido Schimmel Reviewed-by: Eric Dumazet Link: https://patch.msgid.link/20260430200909.527827-2-sharmasagarika@google.com Signed-off-by: Jakub Kicinski --- net/ipv6/route.c | 1 + 1 file changed, 1 insertion(+) (limited to 'net') diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 19eb6b702227..0dc0316530ca 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -4995,6 +4995,7 @@ static int fib6_ifdown(struct fib6_info *rt, void *p_arg) rt->fib6_flags & (RTF_LOCAL | RTF_ANYCAST)) break; rt->fib6_nh->fib_nh_flags |= RTNH_F_LINKDOWN; + fib6_update_sernum(net, rt); rt6_multipath_rebalance(rt); break; } -- cgit v1.2.3 From 458d5615272d3de535748342eb68ca492343048c Mon Sep 17 00:00:00 2001 From: Jamal Hadi Salim Date: Thu, 30 Apr 2026 11:29:55 -0400 Subject: net/sched: sch_red: Replace direct dequeue call with peek and qdisc_dequeue_peeked When red qdisc has children (eg qfq qdisc) whose peek() callback is qdisc_peek_dequeued(), we could get a kernel panic. When the parent of such qdiscs (eg illustrated in patch #3 as tbf) wants to retrieve an skb from its child (red in this case), it will do the following: 1a. do a peek() - and when sensing there's an skb the child can offer, then - the child in this case(red) calls its child's (qfq) peek. qfq does the right thing and will return the gso_skb queue packet. Note: if there wasnt a gso_skb entry then qfq will store it there. 1b. invoke a dequeue() on the child (red). And herein lies the problem. - red will call the child's dequeue() which will essentially just try to grab something of qfq's queue. [ 78.667668][ T363] KASAN: null-ptr-deref in range [0x0000000000000048-0x000000000000004f] [ 78.667927][ T363] CPU: 1 UID: 0 PID: 363 Comm: ping Not tainted 7.1.0-rc1-00033-g46f74a3f7d57-dirty #790 PREEMPT(full) [ 78.668263][ T363] Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011 [ 78.668486][ T363] RIP: 0010:qfq_dequeue+0x446/0xc90 [sch_qfq] [ 78.668718][ T363] Code: 54 c0 e8 dd 90 00 f1 48 c7 c7 e0 03 54 c0 48 89 de e8 ce 90 00 f1 48 8d 7b 48 b8 ff ff 37 00 48 89 fa 48 c1 e0 2a 48 c1 ea 03 <80> 3c 02 00 74 05 e8 ef a1 e1 f1 48 8b 7b 48 48 8d 54 24 58 48 8d [ 78.669312][ T363] RSP: 0018:ffff88810de573e0 EFLAGS: 00010216 [ 78.669533][ T363] RAX: dffffc0000000000 RBX: 0000000000000000 RCX: 0000000000000000 [ 78.669790][ T363] RDX: 0000000000000009 RSI: 0000000000000004 RDI: 0000000000000048 [ 78.670044][ T363] RBP: ffff888110dc4000 R08: ffffffffb1b0885a R09: fffffbfff6ba9078 [ 78.670297][ T363] R10: 0000000000000003 R11: ffff888110e31c80 R12: 0000001880000000 [ 78.670560][ T363] R13: ffff888110dc4150 R14: ffff888110dc42b8 R15: 0000000000000200 [ 78.670814][ T363] FS: 00007f66a8f09c40(0000) GS:ffff888163428000(0000) knlGS:0000000000000000 [ 78.671110][ T363] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 78.671324][ T363] CR2: 000055db4c6a30a8 CR3: 000000010da67000 CR4: 0000000000750ef0 [ 78.671585][ T363] PKRU: 55555554 [ 78.671713][ T363] Call Trace: [ 78.671843][ T363] [ 78.671936][ T363] ? __pfx_qfq_dequeue+0x10/0x10 [sch_qfq] [ 78.672148][ T363] ? __pfx__printk+0x10/0x10 [ 78.672322][ T363] ? srso_alias_return_thunk+0x5/0xfbef5 [ 78.672496][ T363] ? lockdep_hardirqs_on_prepare+0xa8/0x1a0 [ 78.672706][ T363] ? srso_alias_return_thunk+0x5/0xfbef5 [ 78.672875][ T363] ? trace_hardirqs_on+0x19/0x1a0 [ 78.673047][ T363] red_dequeue+0x65/0x270 [sch_red] [ 78.673217][ T363] ? srso_alias_return_thunk+0x5/0xfbef5 [ 78.673385][ T363] tbf_dequeue.cold+0xb0/0x70c [sch_tbf] [ 78.673566][ T363] __qdisc_run+0x169/0x1900 The right thing to do in #1b is to grab the skb off gso_skb queue. This patchset fixes that issue by changing #1b to use qdisc_dequeue_peeked() method instead. Fixes: 77be155cba4e ("pkt_sched: Add peek emulation for non-work-conserving qdiscs.") Reported-by: Manas Reported-by: Rakshit Awasthi Signed-off-by: Jamal Hadi Salim Reviewed-by: Eric Dumazet Link: https://patch.msgid.link/20260430152957.194015-2-jhs@mojatatu.com Signed-off-by: Jakub Kicinski --- net/sched/sch_red.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/sched/sch_red.c b/net/sched/sch_red.c index 432b8a3000a5..4d0e44a2e7c6 100644 --- a/net/sched/sch_red.c +++ b/net/sched/sch_red.c @@ -162,7 +162,7 @@ static struct sk_buff *red_dequeue(struct Qdisc *sch) struct red_sched_data *q = qdisc_priv(sch); struct Qdisc *child = q->qdisc; - skb = child->dequeue(child); + skb = qdisc_dequeue_peeked(child); if (skb) { qdisc_bstats_update(sch, skb); qdisc_qstats_backlog_dec(sch, skb); -- cgit v1.2.3 From 1b9bc71153b01dbde8045b9edede4240f4f5520e Mon Sep 17 00:00:00 2001 From: Victor Nogueria Date: Thu, 30 Apr 2026 11:29:56 -0400 Subject: net/sched: sch_sfb: Replace direct dequeue call with peek and qdisc_dequeue_peeked When sfb has children (eg qfq qdisc) whose peek() callback is qdisc_peek_dequeued(), we could get a kernel panic. When the parent of such qdiscs (eg illustrated in patch #3 as tbf) wants to retrieve an skb from its child (sfb in this case), it will do the following: 1a. do a peek() - and when sensing there's an skb the child can offer, then - the child in this case(sfb) calls its child's (qfq) peek. qfq does the right thing and will return the gso_skb queue packet. Note: if there wasnt a gso_skb entry then qfq will store it there. 1b. invoke a dequeue() on the child (sfb). And herein lies the problem. - sfb will call the child's dequeue() which will essentially just try to grab something of qfq's queue. [ 127.594489][ T453] KASAN: null-ptr-deref in range [0x0000000000000048-0x000000000000004f] [ 127.594741][ T453] CPU: 2 UID: 0 PID: 453 Comm: ping Not tainted 7.1.0-rc1-00035-gac961974495b-dirty #793 PREEMPT(full) [ 127.595059][ T453] Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011 [ 127.595254][ T453] RIP: 0010:qfq_dequeue+0x35c/0x1650 [sch_qfq] [ 127.595461][ T453] Code: 00 fc ff df 80 3c 02 00 0f 85 17 0e 00 00 4c 8d 73 48 48 89 9d b8 02 00 00 48 b8 00 00 00 00 00 fc ff df 4c 89 f2 48 c1 ea 03 <80> 3c 02 00 0f 85 76 0c 00 00 48 b8 00 00 00 00 00 fc ff df 4c 8b [ 127.596081][ T453] RSP: 0018:ffff88810e5af440 EFLAGS: 00010216 [ 127.596337][ T453] RAX: dffffc0000000000 RBX: 0000000000000000 RCX: dffffc0000000000 [ 127.596623][ T453] RDX: 0000000000000009 RSI: 0000001880000000 RDI: ffff888104fd82b0 [ 127.596917][ T453] RBP: ffff888104fd8000 R08: ffff888104fd8280 R09: 1ffff110211893a3 [ 127.597165][ T453] R10: 1ffff110211893a6 R11: 1ffff110211893a7 R12: 0000001880000000 [ 127.597404][ T453] R13: ffff888104fd82b8 R14: 0000000000000048 R15: 0000000040000000 [ 127.597644][ T453] FS: 00007fc380cbfc40(0000) GS:ffff88816f2a8000(0000) knlGS:0000000000000000 [ 127.597956][ T453] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 127.598160][ T453] CR2: 00005610aa9890a8 CR3: 000000010369e000 CR4: 0000000000750ef0 [ 127.598390][ T453] PKRU: 55555554 [ 127.598509][ T453] Call Trace: [ 127.598629][ T453] [ 127.598718][ T453] ? mark_held_locks+0x40/0x70 [ 127.598890][ T453] ? srso_alias_return_thunk+0x5/0xfbef5 [ 127.599053][ T453] sfb_dequeue+0x88/0x4d0 [ 127.599174][ T453] ? ktime_get+0x137/0x230 [ 127.599328][ T453] ? srso_alias_return_thunk+0x5/0xfbef5 [ 127.599480][ T453] ? qdisc_peek_dequeued+0x7b/0x350 [sch_qfq] [ 127.599670][ T453] ? srso_alias_return_thunk+0x5/0xfbef5 [ 127.599831][ T453] tbf_dequeue+0x6b1/0x1098 [sch_tbf] [ 127.599988][ T453] __qdisc_run+0x169/0x1900 The right thing to do in #1b is to grab the skb off gso_skb queue. This patchset fixes that issue by changing #1b to use qdisc_dequeue_peeked() method instead. Fixes: e13e02a3c68d ("net_sched: SFB flow scheduler") Signed-off-by: Victor Nogueria Reviewed-by: Eric Dumazet Link: https://patch.msgid.link/20260430152957.194015-3-jhs@mojatatu.com Signed-off-by: Jakub Kicinski --- net/sched/sch_sfb.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/sched/sch_sfb.c b/net/sched/sch_sfb.c index bd5ef561030f..d3ee8e5479b3 100644 --- a/net/sched/sch_sfb.c +++ b/net/sched/sch_sfb.c @@ -441,7 +441,7 @@ static struct sk_buff *sfb_dequeue(struct Qdisc *sch) struct Qdisc *child = q->qdisc; struct sk_buff *skb; - skb = child->dequeue(q->qdisc); + skb = qdisc_dequeue_peeked(child); if (skb) { qdisc_bstats_update(sch, skb); -- cgit v1.2.3 From 1d324c2f43f70c965f25c58cc3611c779adbe47e Mon Sep 17 00:00:00 2001 From: Maoyi Xie Date: Thu, 30 Apr 2026 18:33:18 +0800 Subject: ip6_gre: Use cached t->net in ip6erspan_changelink(). After commit 5e72ce3e3980 ("net: ipv6: Use link netns in newlink() of rtnl_link_ops"), ip6erspan_newlink() correctly resolves the per-netns ip6gre hash via link_net. ip6erspan_changelink() was not converted in that series and still uses dev_net(dev), which diverges from the device's creation netns after IFLA_NET_NS_FD migration. This re-inserts the tunnel into the wrong per-netns hash. The original netns keeps a stale entry. When that netns is later destroyed, ip6gre_exit_rtnl_net() walks the stale entry, producing a slab-use-after-free reported by KASAN, followed by a kernel BUG at net/core/dev.c (LIST_POISON1) in unregister_netdevice_many_notify(). Reachable from an unprivileged user namespace (unshare --user --map-root-user --net). ip6gre_changelink() earlier in the same file already uses the cached t->net; only ip6erspan_changelink() has the wrong shape. Fixes: 2d665034f239 ("net: ip6_gre: Fix ip6erspan hlen calculation") Cc: stable@vger.kernel.org # v5.15+ Signed-off-by: Maoyi Xie Reviewed-by: Eric Dumazet Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20260430103318.3206018-1-maoyi.xie@ntu.edu.sg Signed-off-by: Jakub Kicinski --- net/ipv6/ip6_gre.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/ipv6/ip6_gre.c b/net/ipv6/ip6_gre.c index 63fc8556b475..365b4059eb20 100644 --- a/net/ipv6/ip6_gre.c +++ b/net/ipv6/ip6_gre.c @@ -2262,10 +2262,11 @@ static int ip6erspan_changelink(struct net_device *dev, struct nlattr *tb[], struct nlattr *data[], struct netlink_ext_ack *extack) { - struct ip6gre_net *ign = net_generic(dev_net(dev), ip6gre_net_id); + struct ip6_tnl *t = netdev_priv(dev); struct __ip6_tnl_parm p; - struct ip6_tnl *t; + struct ip6gre_net *ign; + ign = net_generic(t->net, ip6gre_net_id); t = ip6gre_changelink_common(dev, tb, data, &p, extack); if (IS_ERR(t)) return PTR_ERR(t); -- cgit v1.2.3 From 0799e5943611006b346b8813c7daf7dd5aa26bfd Mon Sep 17 00:00:00 2001 From: Lyes Bourennani Date: Wed, 22 Apr 2026 00:20:22 +0200 Subject: batman-adv: fix integer overflow on buff_pos Fixing an integer overflow present in batadv_iv_ogm_send_to_if. The size check is done using the int type in batadv_iv_ogm_aggr_packet whereas the buff_pos variable uses the s16 type. This could lead to an out-of-bound read. Cc: stable@vger.kernel.org Fixes: c6c8fea29769 ("net: Add batman-adv meshing protocol") Signed-off-by: Lyes Bourennani Signed-off-by: Alexis Pinson Signed-off-by: Sven Eckelmann --- net/batman-adv/bat_iv_ogm.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/batman-adv/bat_iv_ogm.c b/net/batman-adv/bat_iv_ogm.c index f28e9cbf8ad5..618d1889c04e 100644 --- a/net/batman-adv/bat_iv_ogm.c +++ b/net/batman-adv/bat_iv_ogm.c @@ -335,7 +335,7 @@ static void batadv_iv_ogm_send_to_if(struct batadv_forw_packet *forw_packet, struct batadv_priv *bat_priv = netdev_priv(hard_iface->mesh_iface); const char *fwd_str; u8 packet_num; - s16 buff_pos; + int buff_pos; struct batadv_ogm_packet *batadv_ogm_packet; struct sk_buff *skb; u8 *packet_pos; -- cgit v1.2.3 From 3243543592425beec83d453793e9d27caa0d8e66 Mon Sep 17 00:00:00 2001 From: Jiexun Wang Date: Mon, 27 Apr 2026 14:43:33 +0800 Subject: batman-adv: reject new tp_meter sessions during teardown Prevent tp_meter from starting new sender or receiver sessions after mesh_state has left BATADV_MESH_ACTIVE. Fixes: 33a3bb4a3345 ("batman-adv: throughput meter implementation") Cc: stable@kernel.org Reported-by: Yuan Tan Reported-by: Yifan Wu Reported-by: Juefei Pu Reported-by: Xin Liu Co-developed-by: Luxing Yin Signed-off-by: Luxing Yin Signed-off-by: Jiexun Wang Signed-off-by: Ren Wei Signed-off-by: Sven Eckelmann --- net/batman-adv/tp_meter.c | 17 ++++++++++++++++- 1 file changed, 16 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/batman-adv/tp_meter.c b/net/batman-adv/tp_meter.c index 2e42f6b348c8..d9a80e459c2e 100644 --- a/net/batman-adv/tp_meter.c +++ b/net/batman-adv/tp_meter.c @@ -947,6 +947,13 @@ void batadv_tp_start(struct batadv_priv *bat_priv, const u8 *dst, /* look for an already existing test towards this node */ spin_lock_bh(&bat_priv->tp_list_lock); + if (atomic_read(&bat_priv->mesh_state) != BATADV_MESH_ACTIVE) { + spin_unlock_bh(&bat_priv->tp_list_lock); + batadv_tp_batctl_error_notify(BATADV_TP_REASON_DST_UNREACHABLE, + dst, bat_priv, session_cookie); + return; + } + tp_vars = batadv_tp_list_find(bat_priv, dst); if (tp_vars) { spin_unlock_bh(&bat_priv->tp_list_lock); @@ -1329,9 +1336,12 @@ static struct batadv_tp_vars * batadv_tp_init_recv(struct batadv_priv *bat_priv, const struct batadv_icmp_tp_packet *icmp) { - struct batadv_tp_vars *tp_vars; + struct batadv_tp_vars *tp_vars = NULL; spin_lock_bh(&bat_priv->tp_list_lock); + if (atomic_read(&bat_priv->mesh_state) != BATADV_MESH_ACTIVE) + goto out_unlock; + tp_vars = batadv_tp_list_find_session(bat_priv, icmp->orig, icmp->session); if (tp_vars) @@ -1464,6 +1474,9 @@ void batadv_tp_meter_recv(struct batadv_priv *bat_priv, struct sk_buff *skb) { struct batadv_icmp_tp_packet *icmp; + if (atomic_read(&bat_priv->mesh_state) != BATADV_MESH_ACTIVE) + goto out; + icmp = (struct batadv_icmp_tp_packet *)skb->data; switch (icmp->subtype) { @@ -1478,6 +1491,8 @@ void batadv_tp_meter_recv(struct batadv_priv *bat_priv, struct sk_buff *skb) "Received unknown TP Metric packet type %u\n", icmp->subtype); } + +out: consume_skb(skb); } -- cgit v1.2.3 From 3d3cf6a7314aca4df0a6dde28ce784a2a30d0166 Mon Sep 17 00:00:00 2001 From: Jiexun Wang Date: Mon, 27 Apr 2026 14:43:34 +0800 Subject: batman-adv: stop tp_meter sessions during mesh teardown TP meter sessions remain linked on bat_priv->tp_list after the netlink request has already finished. When the mesh interface is removed, batadv_mesh_free() currently tears down the mesh without first draining these sessions. A running sender thread or a late incoming tp_meter packet can then keep processing against a mesh instance which is already shutting down. Synchronize tp_meter with the mesh lifetime by stopping all active sessions from batadv_mesh_free() and waiting for sender threads to exit before teardown continues. Fixes: 33a3bb4a3345 ("batman-adv: throughput meter implementation") Cc: stable@kernel.org Reported-by: Yuan Tan Reported-by: Yifan Wu Reported-by: Juefei Pu Reported-by: Xin Liu Co-developed-by: Luxing Yin Signed-off-by: Luxing Yin Signed-off-by: Jiexun Wang Signed-off-by: Ren Wei Signed-off-by: Sven Eckelmann --- net/batman-adv/main.c | 1 + net/batman-adv/tp_meter.c | 94 ++++++++++++++++++++++++++++++++++++++--------- net/batman-adv/tp_meter.h | 1 + net/batman-adv/types.h | 4 ++ 4 files changed, 82 insertions(+), 18 deletions(-) (limited to 'net') diff --git a/net/batman-adv/main.c b/net/batman-adv/main.c index 3a35aadd8b41..a4d33ee0fda5 100644 --- a/net/batman-adv/main.c +++ b/net/batman-adv/main.c @@ -249,6 +249,7 @@ void batadv_mesh_free(struct net_device *mesh_iface) atomic_set(&bat_priv->mesh_state, BATADV_MESH_DEACTIVATING); batadv_purge_outstanding_packets(bat_priv, NULL); + batadv_tp_stop_all(bat_priv); batadv_gw_node_free(bat_priv); diff --git a/net/batman-adv/tp_meter.c b/net/batman-adv/tp_meter.c index d9a80e459c2e..58ca59a2799e 100644 --- a/net/batman-adv/tp_meter.c +++ b/net/batman-adv/tp_meter.c @@ -12,6 +12,7 @@ #include #include #include +#include #include #include #include @@ -365,23 +366,38 @@ static void batadv_tp_vars_put(struct batadv_tp_vars *tp_vars) } /** - * batadv_tp_sender_cleanup() - cleanup sender data and drop and timer - * @bat_priv: the bat priv with all the mesh interface information - * @tp_vars: the private data of the current TP meter session to cleanup + * batadv_tp_list_detach() - remove tp session from mesh session list once + * @tp_vars: the private data of the current TP meter session */ -static void batadv_tp_sender_cleanup(struct batadv_priv *bat_priv, - struct batadv_tp_vars *tp_vars) +static void batadv_tp_list_detach(struct batadv_tp_vars *tp_vars) { - cancel_delayed_work(&tp_vars->finish_work); + bool detached = false; spin_lock_bh(&tp_vars->bat_priv->tp_list_lock); - hlist_del_rcu(&tp_vars->list); + if (!hlist_unhashed(&tp_vars->list)) { + hlist_del_init_rcu(&tp_vars->list); + detached = true; + } spin_unlock_bh(&tp_vars->bat_priv->tp_list_lock); + if (!detached) + return; + + atomic_dec(&tp_vars->bat_priv->tp_num); + /* drop list reference */ batadv_tp_vars_put(tp_vars); +} - atomic_dec(&tp_vars->bat_priv->tp_num); +/** + * batadv_tp_sender_cleanup() - cleanup sender data and drop and timer + * @tp_vars: the private data of the current TP meter session to cleanup + */ +static void batadv_tp_sender_cleanup(struct batadv_tp_vars *tp_vars) +{ + cancel_delayed_work_sync(&tp_vars->finish_work); + + batadv_tp_list_detach(tp_vars); /* kill the timer and remove its reference */ timer_delete_sync(&tp_vars->timer); @@ -886,7 +902,8 @@ out: batadv_orig_node_put(orig_node); batadv_tp_sender_end(bat_priv, tp_vars); - batadv_tp_sender_cleanup(bat_priv, tp_vars); + batadv_tp_sender_cleanup(tp_vars); + complete(&tp_vars->finished); batadv_tp_vars_put(tp_vars); @@ -918,7 +935,8 @@ static void batadv_tp_start_kthread(struct batadv_tp_vars *tp_vars) batadv_tp_vars_put(tp_vars); /* cleanup of failed tp meter variables */ - batadv_tp_sender_cleanup(bat_priv, tp_vars); + batadv_tp_sender_cleanup(tp_vars); + complete(&tp_vars->finished); return; } @@ -1024,6 +1042,7 @@ void batadv_tp_start(struct batadv_priv *bat_priv, const u8 *dst, tp_vars->start_time = jiffies; init_waitqueue_head(&tp_vars->more_bytes); + init_completion(&tp_vars->finished); spin_lock_init(&tp_vars->unacked_lock); INIT_LIST_HEAD(&tp_vars->unacked_list); @@ -1126,14 +1145,7 @@ static void batadv_tp_receiver_shutdown(struct timer_list *t) "Shutting down for inactivity (more than %dms) from %pM\n", BATADV_TP_RECV_TIMEOUT, tp_vars->other_end); - spin_lock_bh(&tp_vars->bat_priv->tp_list_lock); - hlist_del_rcu(&tp_vars->list); - spin_unlock_bh(&tp_vars->bat_priv->tp_list_lock); - - /* drop list reference */ - batadv_tp_vars_put(tp_vars); - - atomic_dec(&bat_priv->tp_num); + batadv_tp_list_detach(tp_vars); spin_lock_bh(&tp_vars->unacked_lock); list_for_each_entry_safe(un, safe, &tp_vars->unacked_list, list) { @@ -1496,6 +1508,52 @@ out: consume_skb(skb); } +/** + * batadv_tp_stop_all() - stop all currently running tp meter sessions + * @bat_priv: the bat priv with all the mesh interface information + */ +void batadv_tp_stop_all(struct batadv_priv *bat_priv) +{ + struct batadv_tp_vars *tp_vars[BATADV_TP_MAX_NUM]; + struct batadv_tp_vars *tp_var; + size_t count = 0; + size_t i; + + spin_lock_bh(&bat_priv->tp_list_lock); + hlist_for_each_entry(tp_var, &bat_priv->tp_list, list) { + if (WARN_ON_ONCE(count >= BATADV_TP_MAX_NUM)) + break; + + if (!kref_get_unless_zero(&tp_var->refcount)) + continue; + + tp_vars[count++] = tp_var; + } + spin_unlock_bh(&bat_priv->tp_list_lock); + + for (i = 0; i < count; i++) { + tp_var = tp_vars[i]; + + switch (tp_var->role) { + case BATADV_TP_SENDER: + batadv_tp_sender_shutdown(tp_var, + BATADV_TP_REASON_CANCEL); + wake_up(&tp_var->more_bytes); + wait_for_completion(&tp_var->finished); + break; + case BATADV_TP_RECEIVER: + batadv_tp_list_detach(tp_var); + if (timer_shutdown_sync(&tp_var->timer)) + batadv_tp_vars_put(tp_var); + break; + } + + batadv_tp_vars_put(tp_var); + } + + synchronize_net(); +} + /** * batadv_tp_meter_init() - initialize global tp_meter structures */ diff --git a/net/batman-adv/tp_meter.h b/net/batman-adv/tp_meter.h index f0046d366eac..4e97cd10cd02 100644 --- a/net/batman-adv/tp_meter.h +++ b/net/batman-adv/tp_meter.h @@ -17,6 +17,7 @@ void batadv_tp_start(struct batadv_priv *bat_priv, const u8 *dst, u32 test_length, u32 *cookie); void batadv_tp_stop(struct batadv_priv *bat_priv, const u8 *dst, u8 return_value); +void batadv_tp_stop_all(struct batadv_priv *bat_priv); void batadv_tp_meter_recv(struct batadv_priv *bat_priv, struct sk_buff *skb); #endif /* _NET_BATMAN_ADV_TP_METER_H_ */ diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h index 8fc5fe0e9b05..daa06f421154 100644 --- a/net/batman-adv/types.h +++ b/net/batman-adv/types.h @@ -14,6 +14,7 @@ #include #include #include +#include #include #include #include @@ -1328,6 +1329,9 @@ struct batadv_tp_vars { /** @finish_work: work item for the finishing procedure */ struct delayed_work finish_work; + /** @finished: completion signaled when a sender thread exits */ + struct completion finished; + /** @test_length: test length in milliseconds */ u32 test_length; -- cgit v1.2.3 From 046111a1a35a1720748f254377d3d1663664ea61 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 30 Apr 2026 06:16:09 +0000 Subject: net/sched: sch_cake: annotate data-races in cake_dump_class_stats (I) MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit cake_dump_class_stats() runs without qdisc spinlock being held. In this first patch, I add READ_ONCE()/WRITE_ONCE() annotations for: - flow->head - flow->dropped - b->backlogs[] Fixes: 046f6fd5daef ("sched: Add Common Applications Kept Enhanced (cake) qdisc") Signed-off-by: Eric Dumazet Acked-by: Toke Høiland-Jørgensen Reviewed-by: Simon Horman Link: https://patch.msgid.link/20260430061610.3503483-2-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/sched/sch_cake.c | 24 ++++++++++++------------ 1 file changed, 12 insertions(+), 12 deletions(-) (limited to 'net') diff --git a/net/sched/sch_cake.c b/net/sched/sch_cake.c index 13c6d1869a14..806eb73d6a05 100644 --- a/net/sched/sch_cake.c +++ b/net/sched/sch_cake.c @@ -914,7 +914,7 @@ static struct sk_buff *dequeue_head(struct cake_flow *flow) struct sk_buff *skb = flow->head; if (skb) { - flow->head = skb->next; + WRITE_ONCE(flow->head, skb->next); skb_mark_not_on_list(skb); } @@ -926,7 +926,7 @@ static struct sk_buff *dequeue_head(struct cake_flow *flow) static void flow_queue_add(struct cake_flow *flow, struct sk_buff *skb) { if (!flow->head) - flow->head = skb; + WRITE_ONCE(flow->head, skb); else flow->tail->next = skb; flow->tail = skb; @@ -1357,7 +1357,7 @@ found: if (elig_ack_prev) elig_ack_prev->next = elig_ack->next; else - flow->head = elig_ack->next; + WRITE_ONCE(flow->head, elig_ack->next); skb_mark_not_on_list(elig_ack); @@ -1595,11 +1595,11 @@ static unsigned int cake_drop(struct Qdisc *sch, struct sk_buff **to_free) len = qdisc_pkt_len(skb); q->buffer_used -= skb->truesize; - b->backlogs[idx] -= len; WRITE_ONCE(b->tin_backlog, b->tin_backlog - len); + WRITE_ONCE(b->backlogs[idx], b->backlogs[idx] - len); sch->qstats.backlog -= len; - flow->dropped++; + WRITE_ONCE(flow->dropped, flow->dropped + 1); WRITE_ONCE(b->tin_dropped, b->tin_dropped + 1); if (q->config->rate_flags & CAKE_FLAG_INGRESS) @@ -1824,11 +1824,11 @@ static s32 cake_enqueue(struct sk_buff *skb, struct Qdisc *sch, } /* stats */ - b->backlogs[idx] += slen; sch->qstats.backlog += slen; q->avg_window_bytes += slen; WRITE_ONCE(b->bytes, b->bytes + slen); WRITE_ONCE(b->tin_backlog, b->tin_backlog + slen); + WRITE_ONCE(b->backlogs[idx], b->backlogs[idx] + slen); qdisc_tree_reduce_backlog(sch, 1-numsegs, len-slen); consume_skb(skb); @@ -1861,11 +1861,11 @@ static s32 cake_enqueue(struct sk_buff *skb, struct Qdisc *sch, /* stats */ WRITE_ONCE(b->packets, b->packets + 1); - b->backlogs[idx] += len - ack_pkt_len; sch->qstats.backlog += len - ack_pkt_len; q->avg_window_bytes += len - ack_pkt_len; WRITE_ONCE(b->bytes, b->bytes + len - ack_pkt_len); WRITE_ONCE(b->tin_backlog, b->tin_backlog + len - ack_pkt_len); + WRITE_ONCE(b->backlogs[idx], b->backlogs[idx] + len - ack_pkt_len); } if (q->overflow_timeout) @@ -1977,7 +1977,7 @@ static struct sk_buff *cake_dequeue_one(struct Qdisc *sch) if (flow->head) { skb = dequeue_head(flow); len = qdisc_pkt_len(skb); - b->backlogs[q->cur_flow] -= len; + WRITE_ONCE(b->backlogs[q->cur_flow], b->backlogs[q->cur_flow] - len); WRITE_ONCE(b->tin_backlog, b->tin_backlog - len); sch->qstats.backlog -= len; q->buffer_used -= skb->truesize; @@ -2235,7 +2235,7 @@ retry: flow->deficit -= len; b->tin_deficit -= len; } - flow->dropped++; + WRITE_ONCE(flow->dropped, flow->dropped + 1); WRITE_ONCE(b->tin_dropped, b->tin_dropped + 1); qdisc_tree_reduce_backlog(sch, 1, qdisc_pkt_len(skb)); qdisc_qstats_drop(sch); @@ -3137,7 +3137,7 @@ static int cake_dump_class_stats(struct Qdisc *sch, unsigned long cl, flow = &b->flows[idx % CAKE_QUEUES]; - if (flow->head) { + if (READ_ONCE(flow->head)) { sch_tree_lock(sch); skb = flow->head; while (skb) { @@ -3146,8 +3146,8 @@ static int cake_dump_class_stats(struct Qdisc *sch, unsigned long cl, } sch_tree_unlock(sch); } - qs.backlog = b->backlogs[idx % CAKE_QUEUES]; - qs.drops = flow->dropped; + qs.backlog = READ_ONCE(b->backlogs[idx % CAKE_QUEUES]); + qs.drops = READ_ONCE(flow->dropped); } if (gnet_stats_copy_queue(d, NULL, &qs, qs.qlen) < 0) return -1; -- cgit v1.2.3 From 67dc6c56b871617deac85b9f72500b69b1fdf835 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 30 Apr 2026 06:16:10 +0000 Subject: net/sched: sch_cake: annotate data-races in cake_dump_class_stats (II) MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit cake_dump_class_stats() runs without qdisc spinlock being held. In this second patch, I add READ_ONCE()/WRITE_ONCE() annotations for: - flow->deficit - flow->cvars.dropping - flow->cvars.count - flow->cvars.p_drop - flow->cvars.blue_timer - flow->cvars.drop_next Fixes: 046f6fd5daef ("sched: Add Common Applications Kept Enhanced (cake) qdisc") Signed-off-by: Eric Dumazet Acked-by: Toke Høiland-Jørgensen Reviewed-by: Simon Horman Link: https://patch.msgid.link/20260430061610.3503483-3-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/sched/sch_cake.c | 131 ++++++++++++++++++++++++++++----------------------- 1 file changed, 71 insertions(+), 60 deletions(-) (limited to 'net') diff --git a/net/sched/sch_cake.c b/net/sched/sch_cake.c index 806eb73d6a05..5862933be8d7 100644 --- a/net/sched/sch_cake.c +++ b/net/sched/sch_cake.c @@ -399,14 +399,14 @@ static void cake_configure_rates(struct Qdisc *sch, u64 rate, bool rate_adjust); * Here, invsqrt is a fixed point number (< 1.0), 32bit mantissa, aka Q0.32 */ -static void cobalt_newton_step(struct cobalt_vars *vars) +static void cobalt_newton_step(struct cobalt_vars *vars, u32 count) { u32 invsqrt, invsqrt2; u64 val; invsqrt = vars->rec_inv_sqrt; invsqrt2 = ((u64)invsqrt * invsqrt) >> 32; - val = (3LL << 32) - ((u64)vars->count * invsqrt2); + val = (3LL << 32) - ((u64)count * invsqrt2); val >>= 2; /* avoid overflow in following multiply */ val = (val * invsqrt) >> (32 - 2 + 1); @@ -414,12 +414,12 @@ static void cobalt_newton_step(struct cobalt_vars *vars) vars->rec_inv_sqrt = val; } -static void cobalt_invsqrt(struct cobalt_vars *vars) +static void cobalt_invsqrt(struct cobalt_vars *vars, u32 count) { - if (vars->count < REC_INV_SQRT_CACHE) - vars->rec_inv_sqrt = inv_sqrt_cache[vars->count]; + if (count < REC_INV_SQRT_CACHE) + vars->rec_inv_sqrt = inv_sqrt_cache[count]; else - cobalt_newton_step(vars); + cobalt_newton_step(vars, count); } static void cobalt_vars_init(struct cobalt_vars *vars) @@ -449,16 +449,19 @@ static bool cobalt_queue_full(struct cobalt_vars *vars, bool up = false; if (ktime_to_ns(ktime_sub(now, vars->blue_timer)) > p->target) { - up = !vars->p_drop; - vars->p_drop += p->p_inc; - if (vars->p_drop < p->p_inc) - vars->p_drop = ~0; - vars->blue_timer = now; - } - vars->dropping = true; - vars->drop_next = now; + u32 p_drop = vars->p_drop; + + up = !p_drop; + p_drop += p->p_inc; + if (p_drop < p->p_inc) + p_drop = ~0; + WRITE_ONCE(vars->p_drop, p_drop); + WRITE_ONCE(vars->blue_timer, now); + } + WRITE_ONCE(vars->dropping, true); + WRITE_ONCE(vars->drop_next, now); if (!vars->count) - vars->count = 1; + WRITE_ONCE(vars->count, 1); return up; } @@ -475,20 +478,20 @@ static bool cobalt_queue_empty(struct cobalt_vars *vars, if (vars->p_drop && ktime_to_ns(ktime_sub(now, vars->blue_timer)) > p->target) { if (vars->p_drop < p->p_dec) - vars->p_drop = 0; + WRITE_ONCE(vars->p_drop, 0); else - vars->p_drop -= p->p_dec; - vars->blue_timer = now; + WRITE_ONCE(vars->p_drop, vars->p_drop - p->p_dec); + WRITE_ONCE(vars->blue_timer, now); down = !vars->p_drop; } - vars->dropping = false; + WRITE_ONCE(vars->dropping, false); if (vars->count && ktime_to_ns(ktime_sub(now, vars->drop_next)) >= 0) { - vars->count--; - cobalt_invsqrt(vars); - vars->drop_next = cobalt_control(vars->drop_next, - p->interval, - vars->rec_inv_sqrt); + WRITE_ONCE(vars->count, vars->count - 1); + cobalt_invsqrt(vars, vars->count); + WRITE_ONCE(vars->drop_next, + cobalt_control(vars->drop_next, p->interval, + vars->rec_inv_sqrt)); } return down; @@ -507,6 +510,7 @@ static enum qdisc_drop_reason cobalt_should_drop(struct cobalt_vars *vars, bool next_due, over_target; ktime_t schedule; u64 sojourn; + u32 count; /* The 'schedule' variable records, in its sign, whether 'now' is before or * after 'drop_next'. This allows 'drop_next' to be updated before the next @@ -528,21 +532,22 @@ static enum qdisc_drop_reason cobalt_should_drop(struct cobalt_vars *vars, over_target = sojourn > p->target && sojourn > p->mtu_time * bulk_flows * 2 && sojourn > p->mtu_time * 4; - next_due = vars->count && ktime_to_ns(schedule) >= 0; + count = vars->count; + next_due = count && ktime_to_ns(schedule) >= 0; vars->ecn_marked = false; if (over_target) { if (!vars->dropping) { - vars->dropping = true; - vars->drop_next = cobalt_control(now, - p->interval, - vars->rec_inv_sqrt); + WRITE_ONCE(vars->dropping, true); + WRITE_ONCE(vars->drop_next, + cobalt_control(now, p->interval, + vars->rec_inv_sqrt)); } - if (!vars->count) - vars->count = 1; + if (!count) + count = 1; } else if (vars->dropping) { - vars->dropping = false; + WRITE_ONCE(vars->dropping, false); } if (next_due && vars->dropping) { @@ -550,23 +555,23 @@ static enum qdisc_drop_reason cobalt_should_drop(struct cobalt_vars *vars, if (!(vars->ecn_marked = INET_ECN_set_ce(skb))) reason = QDISC_DROP_CONGESTED; - vars->count++; - if (!vars->count) - vars->count--; - cobalt_invsqrt(vars); - vars->drop_next = cobalt_control(vars->drop_next, - p->interval, - vars->rec_inv_sqrt); + count++; + if (!count) + count--; + cobalt_invsqrt(vars, count); + WRITE_ONCE(vars->drop_next, + cobalt_control(vars->drop_next, p->interval, + vars->rec_inv_sqrt)); schedule = ktime_sub(now, vars->drop_next); } else { while (next_due) { - vars->count--; - cobalt_invsqrt(vars); - vars->drop_next = cobalt_control(vars->drop_next, - p->interval, - vars->rec_inv_sqrt); + count--; + cobalt_invsqrt(vars, count); + WRITE_ONCE(vars->drop_next, + cobalt_control(vars->drop_next, p->interval, + vars->rec_inv_sqrt)); schedule = ktime_sub(now, vars->drop_next); - next_due = vars->count && ktime_to_ns(schedule) >= 0; + next_due = count && ktime_to_ns(schedule) >= 0; } } @@ -575,11 +580,12 @@ static enum qdisc_drop_reason cobalt_should_drop(struct cobalt_vars *vars, get_random_u32() < vars->p_drop) reason = QDISC_DROP_FLOOD_PROTECTION; + WRITE_ONCE(vars->count, count); /* Overload the drop_next field as an activity timeout */ - if (!vars->count) - vars->drop_next = ktime_add_ns(now, p->interval); + if (!count) + WRITE_ONCE(vars->drop_next, ktime_add_ns(now, p->interval)); else if (ktime_to_ns(schedule) > 0 && reason == QDISC_DROP_UNSPEC) - vars->drop_next = now; + WRITE_ONCE(vars->drop_next, now); return reason; } @@ -1924,7 +1930,7 @@ static s32 cake_enqueue(struct sk_buff *skb, struct Qdisc *sch, flow->set = CAKE_SET_SPARSE; WRITE_ONCE(b->sparse_flow_count, b->sparse_flow_count + 1); - flow->deficit = cake_get_flow_quantum(b, flow, q->config->flow_mode); + WRITE_ONCE(flow->deficit, cake_get_flow_quantum(b, flow, q->config->flow_mode)); } else if (flow->set == CAKE_SET_SPARSE_WAIT) { /* this flow was empty, accounted as a sparse flow, but actually * in the bulk rotation. @@ -2166,7 +2172,8 @@ retry: } } - flow->deficit += cake_get_flow_quantum(b, flow, q->config->flow_mode); + WRITE_ONCE(flow->deficit, + flow->deficit + cake_get_flow_quantum(b, flow, q->config->flow_mode)); list_move_tail(&flow->flowchain, &b->old_flows); goto retry; @@ -2232,7 +2239,7 @@ retry: if (q->config->rate_flags & CAKE_FLAG_INGRESS) { len = cake_advance_shaper(q, b, skb, now, true); - flow->deficit -= len; + WRITE_ONCE(flow->deficit, flow->deficit - len); b->tin_deficit -= len; } WRITE_ONCE(flow->dropped, flow->dropped + 1); @@ -2259,7 +2266,7 @@ retry: delay < b->base_delay ? 2 : 8)); len = cake_advance_shaper(q, b, skb, now, false); - flow->deficit -= len; + WRITE_ONCE(flow->deficit, flow->deficit - len); b->tin_deficit -= len; if (ktime_after(q->time_next_packet, now) && sch->q.qlen) { @@ -3153,6 +3160,8 @@ static int cake_dump_class_stats(struct Qdisc *sch, unsigned long cl, return -1; if (flow) { ktime_t now = ktime_get(); + bool dropping; + u32 p_drop; stats = nla_nest_start_noflag(d->skb, TCA_STATS_APP); if (!stats) @@ -3167,21 +3176,23 @@ static int cake_dump_class_stats(struct Qdisc *sch, unsigned long cl, goto nla_put_failure; \ } while (0) - PUT_STAT_S32(DEFICIT, flow->deficit); - PUT_STAT_U32(DROPPING, flow->cvars.dropping); - PUT_STAT_U32(COBALT_COUNT, flow->cvars.count); - PUT_STAT_U32(P_DROP, flow->cvars.p_drop); - if (flow->cvars.p_drop) { + PUT_STAT_S32(DEFICIT, READ_ONCE(flow->deficit)); + dropping = READ_ONCE(flow->cvars.dropping); + PUT_STAT_U32(DROPPING, dropping); + PUT_STAT_U32(COBALT_COUNT, READ_ONCE(flow->cvars.count)); + p_drop = READ_ONCE(flow->cvars.p_drop); + PUT_STAT_U32(P_DROP, p_drop); + if (p_drop) { PUT_STAT_S32(BLUE_TIMER_US, ktime_to_us( ktime_sub(now, - flow->cvars.blue_timer))); + READ_ONCE(flow->cvars.blue_timer)))); } - if (flow->cvars.dropping) { + if (dropping) { PUT_STAT_S32(DROP_NEXT_US, ktime_to_us( ktime_sub(now, - flow->cvars.drop_next))); + READ_ONCE(flow->cvars.drop_next)))); } if (nla_nest_end(d->skb, stats) < 0) -- cgit v1.2.3 From 7e7be31bfdb066c1c780dcd6b1224078fc54063f Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Wed, 29 Apr 2026 15:29:38 -0700 Subject: net: tls: fix silent data drop under pipe back-pressure tls_sw_splice_read() uses len when advancing rxm->offset / rxm->full_len after skb_splice_bits(), rather than copied (the actual number of bytes successfully spliced into the pipe). When the destination pipe cannot accept all the requested bytes, splice_to_pipe() returns fewer bytes than len, and 'len - copied' of data is effectively skipped over. Fixes: e062fe99cccd ("tls: splice_read: fix accessing pre-processed records") Link: https://patch.msgid.link/20260429222944.2139041-2-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/tls/tls_sw.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/tls/tls_sw.c b/net/tls/tls_sw.c index 798243eabb1f..2590e855f6a5 100644 --- a/net/tls/tls_sw.c +++ b/net/tls/tls_sw.c @@ -2317,9 +2317,9 @@ ssize_t tls_sw_splice_read(struct socket *sock, loff_t *ppos, if (copied < 0) goto splice_requeue; - if (chunk < rxm->full_len) { - rxm->offset += len; - rxm->full_len -= len; + if (copied < rxm->full_len) { + rxm->offset += copied; + rxm->full_len -= copied; goto splice_requeue; } -- cgit v1.2.3 From 0cfff13c94cb5fa818bb374945ff280e08dc1bb9 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Mon, 4 May 2026 08:54:27 +0200 Subject: wifi: mac80211: tests: mark HT check strict The HT check now only applies in strict mode since APs were found to be broken. Mark it as such. Fixes: 711a9c018ad2 ("wifi: mac80211: skip ieee80211_verify_sta_ht_mcs_support check in non-strict mode") Signed-off-by: Johannes Berg --- net/mac80211/tests/chan-mode.c | 1 + 1 file changed, 1 insertion(+) (limited to 'net') diff --git a/net/mac80211/tests/chan-mode.c b/net/mac80211/tests/chan-mode.c index adc069065e73..fa370831d617 100644 --- a/net/mac80211/tests/chan-mode.c +++ b/net/mac80211/tests/chan-mode.c @@ -65,6 +65,7 @@ static const struct determine_chan_mode_case { .ht_capa_mask = { .mcs.rx_mask[0] = 0xf7, }, + .strict = true, }, { .desc = "Masking out a RX rate in VHT capabilities", .conn_mode = IEEE80211_CONN_MODE_EHT, -- cgit v1.2.3 From afbd961305eb483515650ccfcb7743608e7add78 Mon Sep 17 00:00:00 2001 From: Julian Anastasov Date: Thu, 30 Apr 2026 10:44:13 +0300 Subject: ipvs: fixes for the new ip_vs_status info Sashiko reports some problems for the recently added /proc/net/ip_vs_status: * ip_vs_status_show() as a table reader may run long after the conn_tab and svc_table table are released. While ip_vs_conn_flush() properly changes the conn_tab_changes counter when conn_tab is removed, ip_vs_del_service() and ip_vs_flush() were missing such change for the svc_table_changes counter. As result, readers like ip_vs_dst_event() and ip_vs_status_show() may continue to use a freed table after a cond_resched_rcu() call. * While counting the buckets in ip_vs_status_show() make sure we traverse only the needed number of entries in the chain. This also prevents possible overflow of the 'count' variable. * Add check for 'loops' to prevent infinite loops while restarting the traversal on table change. * While IP_VS_CONN_TAB_MAX_BITS is 20 on 32-bit platforms and there is no risk to overflow when multiplying the number of conn_tab buckets to 100, prefer the div_u64() helper to make the following dividing safer. * Use 0440 permissions for ip_vs_status to restrict the info only to root due to the exported information for hash distribution. Link: https://sashiko.dev/#/patchset/20260410112352.23599-1-fw%40strlen.de Fixes: 9a9ccef907a7 ("ipvs: add ip_vs_status info") Signed-off-by: Julian Anastasov Signed-off-by: Pablo Neira Ayuso --- net/netfilter/ipvs/ip_vs_ctl.c | 51 +++++++++++++++++++++++++++++------------- 1 file changed, 36 insertions(+), 15 deletions(-) (limited to 'net') diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c index 6632daa87ded..27e50afe9a54 100644 --- a/net/netfilter/ipvs/ip_vs_ctl.c +++ b/net/netfilter/ipvs/ip_vs_ctl.c @@ -2032,6 +2032,9 @@ static int ip_vs_del_service(struct ip_vs_service *svc) cancel_delayed_work_sync(&ipvs->svc_resize_work); if (t) { rcu_assign_pointer(ipvs->svc_table, NULL); + /* Inform readers that table is removed */ + smp_mb__before_atomic(); + atomic_inc(&ipvs->svc_table_changes); while (1) { p = rcu_dereference_protected(t->new_tbl, 1); call_rcu(&t->rcu_head, ip_vs_rht_rcu_free); @@ -2078,6 +2081,9 @@ static int ip_vs_flush(struct netns_ipvs *ipvs, bool cleanup) t = rcu_dereference_protected(ipvs->svc_table, 1); if (t) { rcu_assign_pointer(ipvs->svc_table, NULL); + /* Inform readers that table is removed */ + smp_mb__before_atomic(); + atomic_inc(&ipvs->svc_table_changes); while (1) { p = rcu_dereference_protected(t->new_tbl, 1); call_rcu(&t->rcu_head, ip_vs_rht_rcu_free); @@ -3004,7 +3010,8 @@ static int ip_vs_status_show(struct seq_file *seq, void *v) int old_gen, new_gen; u32 counts[8]; u32 bucket; - int count; + u32 count; + int loops; u32 sum1; u32 sum; int i; @@ -3020,6 +3027,7 @@ static int ip_vs_status_show(struct seq_file *seq, void *v) if (!atomic_read(&ipvs->conn_count)) goto after_conns; old_gen = atomic_read(&ipvs->conn_tab_changes); + loops = 0; repeat_conn: smp_rmb(); /* ipvs->conn_tab and conn_tab_changes */ @@ -3032,8 +3040,11 @@ repeat_conn: resched_score++; ip_vs_rht_walk_bucket_rcu(t, bucket, head) { count = 0; - hlist_bl_for_each_entry_rcu(hn, e, head, node) + hlist_bl_for_each_entry_rcu(hn, e, head, node) { count++; + if (count >= ARRAY_SIZE(counts) - 1) + break; + } } resched_score += count; if (resched_score >= 100) { @@ -3042,37 +3053,41 @@ repeat_conn: new_gen = atomic_read(&ipvs->conn_tab_changes); /* New table installed ? */ if (old_gen != new_gen) { + /* Too many changes? */ + if (++loops >= 5) + goto after_conns; old_gen = new_gen; goto repeat_conn; } } - counts[min(count, (int)ARRAY_SIZE(counts) - 1)]++; + counts[count]++; } } for (sum = 0, i = 0; i < ARRAY_SIZE(counts); i++) sum += counts[i]; sum1 = sum - counts[0]; - seq_printf(seq, "Conn buckets empty:\t%u (%lu%%)\n", - counts[0], (unsigned long)counts[0] * 100 / max(sum, 1U)); + seq_printf(seq, "Conn buckets empty:\t%u (%llu%%)\n", + counts[0], div_u64((u64)counts[0] * 100U, max(sum, 1U))); for (i = 1; i < ARRAY_SIZE(counts); i++) { if (!counts[i]) continue; - seq_printf(seq, "Conn buckets len-%d:\t%u (%lu%%)\n", + seq_printf(seq, "Conn buckets len-%d:\t%u (%llu%%)\n", i, counts[i], - (unsigned long)counts[i] * 100 / max(sum1, 1U)); + div_u64((u64)counts[i] * 100U, max(sum1, 1U))); } after_conns: t = rcu_dereference(ipvs->svc_table); count = ip_vs_get_num_services(ipvs); - seq_printf(seq, "Services:\t%d\n", count); + seq_printf(seq, "Services:\t%u\n", count); seq_printf(seq, "Service buckets:\t%d (%d bits, lfactor %d)\n", t ? t->size : 0, t ? t->bits : 0, t ? t->lfactor : 0); if (!count) goto after_svc; old_gen = atomic_read(&ipvs->svc_table_changes); + loops = 0; repeat_svc: smp_rmb(); /* ipvs->svc_table and svc_table_changes */ @@ -3086,8 +3101,11 @@ repeat_svc: ip_vs_rht_walk_bucket_rcu(t, bucket, head) { count = 0; hlist_bl_for_each_entry_rcu(svc, e, head, - s_list) + s_list) { count++; + if (count >= ARRAY_SIZE(counts) - 1) + break; + } } resched_score += count; if (resched_score >= 100) { @@ -3096,24 +3114,27 @@ repeat_svc: new_gen = atomic_read(&ipvs->svc_table_changes); /* New table installed ? */ if (old_gen != new_gen) { + /* Too many changes? */ + if (++loops >= 5) + goto after_svc; old_gen = new_gen; goto repeat_svc; } } - counts[min(count, (int)ARRAY_SIZE(counts) - 1)]++; + counts[count]++; } } for (sum = 0, i = 0; i < ARRAY_SIZE(counts); i++) sum += counts[i]; sum1 = sum - counts[0]; - seq_printf(seq, "Service buckets empty:\t%u (%lu%%)\n", - counts[0], (unsigned long)counts[0] * 100 / max(sum, 1U)); + seq_printf(seq, "Service buckets empty:\t%u (%llu%%)\n", + counts[0], div_u64((u64)counts[0] * 100U, max(sum, 1U))); for (i = 1; i < ARRAY_SIZE(counts); i++) { if (!counts[i]) continue; - seq_printf(seq, "Service buckets len-%d:\t%u (%lu%%)\n", + seq_printf(seq, "Service buckets len-%d:\t%u (%llu%%)\n", i, counts[i], - (unsigned long)counts[i] * 100 / max(sum1, 1U)); + div_u64((u64)counts[i] * 100U, max(sum1, 1U))); } after_svc: @@ -5039,7 +5060,7 @@ int __net_init ip_vs_control_net_init(struct netns_ipvs *ipvs) ipvs->net->proc_net, ip_vs_stats_percpu_show, NULL)) goto err_percpu; - if (!proc_create_net_single("ip_vs_status", 0, ipvs->net->proc_net, + if (!proc_create_net_single("ip_vs_status", 0440, ipvs->net->proc_net, ip_vs_status_show, NULL)) goto err_status; #endif -- cgit v1.2.3 From f2da9a96abb4b7a64626e931cedd85f05d5498ca Mon Sep 17 00:00:00 2001 From: Julian Anastasov Date: Thu, 30 Apr 2026 10:44:14 +0300 Subject: ipvs: fix races around the conn_lfactor and svc_lfactor sysctl vars Sashiko warns that the new sysctls vars can be changed after the hash tables are destroyed and their respective resizing works canceled, leading to mod_delayed_work() being called for canceled works. Solve this in different ways. conn_tab can be present even without services and is destroyed only on netns exit, so use disable_delayed_work_sync() to disable the work instead of adding more synchronization mechanisms. As for the svc_table, it is destroyed when the services are deleted, so we must be sure that netns exit is not called yet (the check for 'enable') and the work is not canceled by checking all under same mutex lock. Also, use WRITE_ONCE when updating the sysctl vars as we already read them with READ_ONCE. Link: https://sashiko.dev/#/patchset/20260410112352.23599-1-fw%40strlen.de Fixes: 8d7de5477e47 ("ipvs: add conn_lfactor and svc_lfactor sysctl vars") Signed-off-by: Julian Anastasov Signed-off-by: Pablo Neira Ayuso --- net/netfilter/ipvs/ip_vs_conn.c | 2 +- net/netfilter/ipvs/ip_vs_ctl.c | 12 +++++++++--- 2 files changed, 10 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/netfilter/ipvs/ip_vs_conn.c b/net/netfilter/ipvs/ip_vs_conn.c index 2082bfb2d93c..84a4921a7865 100644 --- a/net/netfilter/ipvs/ip_vs_conn.c +++ b/net/netfilter/ipvs/ip_vs_conn.c @@ -1835,7 +1835,7 @@ static void ip_vs_conn_flush(struct netns_ipvs *ipvs) if (!rcu_dereference_protected(ipvs->conn_tab, 1)) return; - cancel_delayed_work_sync(&ipvs->conn_resize_work); + disable_delayed_work_sync(&ipvs->conn_resize_work); if (!atomic_read(&ipvs->conn_count)) goto unreg; diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c index 27e50afe9a54..caec516856e9 100644 --- a/net/netfilter/ipvs/ip_vs_ctl.c +++ b/net/netfilter/ipvs/ip_vs_ctl.c @@ -2469,7 +2469,7 @@ static int ipvs_proc_conn_lfactor(const struct ctl_table *table, int write, if (val < -8 || val > 8) { ret = -EINVAL; } else { - *valp = val; + WRITE_ONCE(*valp, val); if (rcu_access_pointer(ipvs->conn_tab)) mod_delayed_work(system_unbound_wq, &ipvs->conn_resize_work, 0); @@ -2496,10 +2496,16 @@ static int ipvs_proc_svc_lfactor(const struct ctl_table *table, int write, if (val < -8 || val > 8) { ret = -EINVAL; } else { - *valp = val; - if (rcu_access_pointer(ipvs->svc_table)) + mutex_lock(&ipvs->service_mutex); + WRITE_ONCE(*valp, val); + /* Make sure the services are present */ + if (rcu_access_pointer(ipvs->svc_table) && + READ_ONCE(ipvs->enable) && + !test_bit(IP_VS_WORK_SVC_NORESIZE, + &ipvs->work_flags)) mod_delayed_work(system_unbound_wq, &ipvs->svc_resize_work, 0); + mutex_unlock(&ipvs->service_mutex); } } return ret; -- cgit v1.2.3 From d493d9de1c21313cf62be0f6e1a4d48385fa7beb Mon Sep 17 00:00:00 2001 From: Julian Anastasov Date: Thu, 30 Apr 2026 10:44:15 +0300 Subject: ipvs: fix the spin_lock usage for RT build syzbot reports for sleeping function called from invalid context [1]. The recently added code for resizable hash tables uses hlist_bl bit locks in combination with spin_lock for the connection fields (cp->lock). Fix the following problems: * avoid using spin_lock(&cp->lock) under locked bit lock because it sleeps on PREEMPT_RT * as the recent changes call ip_vs_conn_hash() only for newly allocated connection, the spin_lock can be removed there because the connection is still not linked to table and does not need cp->lock protection. * the lock can be removed also from ip_vs_conn_unlink() where we are the last connection user. * the last place that is fixed is ip_vs_conn_fill_cport() where now the cp->lock is locked before the other locks to ensure other packets do not modify the cp->flags in non-atomic way. Here we make sure cport and flags are changed only once if two or more packets race to fill the cport. Also, we fill cport early, so that if we race with resizing there will be valid cport key for the hashing. Add a warning if too many hash table changes occur for our RCU read-side critical section which is error condition but minor because the connection still can expire gracefully. Still, restore the cport to 0 to allow retransmitted packet to properly fill the cport. Problems reported by Sashiko. [1]: BUG: sleeping function called from invalid context at kernel/locking/spinlock_rt.c:48 in_atomic(): 1, irqs_disabled(): 0, non_block: 0, pid: 16, name: ktimers/0 preempt_count: 2, expected: 0 RCU nest depth: 3, expected: 3 8 locks held by ktimers/0/16: #0: ffffffff8de5f260 (local_bh){.+.+}-{1:3}, at: __local_bh_disable_ip+0x3c/0x420 kernel/softirq.c:163 #1: ffffffff8dfc80c0 (rcu_read_lock){....}-{1:3}, at: __local_bh_disable_ip+0x3c/0x420 kernel/softirq.c:163 #2: ffff8880b8826360 (&base->expiry_lock){+...}-{3:3}, at: spin_lock include/linux/spinlock_rt.h:45 [inline] #2: ffff8880b8826360 (&base->expiry_lock){+...}-{3:3}, at: timer_base_lock_expiry kernel/time/timer.c:1502 [inline] #2: ffff8880b8826360 (&base->expiry_lock){+...}-{3:3}, at: __run_timer_base+0x120/0x9f0 kernel/time/timer.c:2384 #3: ffffffff8dfc80c0 (rcu_read_lock){....}-{1:3}, at: rcu_lock_acquire include/linux/rcupdate.h:300 [inline] #3: ffffffff8dfc80c0 (rcu_read_lock){....}-{1:3}, at: rcu_read_lock include/linux/rcupdate.h:838 [inline] #3: ffffffff8dfc80c0 (rcu_read_lock){....}-{1:3}, at: __rt_spin_lock kernel/locking/spinlock_rt.c:50 [inline] #3: ffffffff8dfc80c0 (rcu_read_lock){....}-{1:3}, at: rt_spin_lock+0x1e0/0x400 kernel/locking/spinlock_rt.c:57 #4: ffffc90000157a80 ((&cp->timer)){+...}-{0:0}, at: call_timer_fn+0xd4/0x5e0 kernel/time/timer.c:1745 #5: ffffffff8dfc80c0 (rcu_read_lock){....}-{1:3}, at: rcu_lock_acquire include/linux/rcupdate.h:300 [inline] #5: ffffffff8dfc80c0 (rcu_read_lock){....}-{1:3}, at: rcu_read_lock include/linux/rcupdate.h:838 [inline] #5: ffffffff8dfc80c0 (rcu_read_lock){....}-{1:3}, at: ip_vs_conn_unlink net/netfilter/ipvs/ip_vs_conn.c:315 [inline] #5: ffffffff8dfc80c0 (rcu_read_lock){....}-{1:3}, at: ip_vs_conn_expire+0x257/0x2390 net/netfilter/ipvs/ip_vs_conn.c:1260 #6: ffffffff8de5f260 (local_bh){.+.+}-{1:3}, at: __local_bh_disable_ip+0x3c/0x420 kernel/softirq.c:163 #7: ffff888068d4c3f0 (&cp->lock#2){+...}-{3:3}, at: spin_lock include/linux/spinlock_rt.h:45 [inline] #7: ffff888068d4c3f0 (&cp->lock#2){+...}-{3:3}, at: ip_vs_conn_unlink net/netfilter/ipvs/ip_vs_conn.c:324 [inline] #7: ffff888068d4c3f0 (&cp->lock#2){+...}-{3:3}, at: ip_vs_conn_expire+0xd4a/0x2390 net/netfilter/ipvs/ip_vs_conn.c:1260 Preemption disabled at: [] bit_spin_lock include/linux/bit_spinlock.h:38 [inline] [] hlist_bl_lock+0x18/0x110 include/linux/list_bl.h:149 CPU: 0 UID: 0 PID: 16 Comm: ktimers/0 Tainted: G W L syzkaller #0 PREEMPT_{RT,(full)} Tainted: [W]=WARN, [L]=SOFTLOCKUP Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 03/18/2026 Call Trace: dump_stack_lvl+0xe8/0x150 lib/dump_stack.c:120 __might_resched+0x329/0x480 kernel/sched/core.c:9162 __rt_spin_lock kernel/locking/spinlock_rt.c:48 [inline] rt_spin_lock+0xc2/0x400 kernel/locking/spinlock_rt.c:57 spin_lock include/linux/spinlock_rt.h:45 [inline] ip_vs_conn_unlink net/netfilter/ipvs/ip_vs_conn.c:324 [inline] ip_vs_conn_expire+0xd4a/0x2390 net/netfilter/ipvs/ip_vs_conn.c:1260 call_timer_fn+0x192/0x5e0 kernel/time/timer.c:1748 expire_timers kernel/time/timer.c:1799 [inline] __run_timers kernel/time/timer.c:2374 [inline] __run_timer_base+0x6a3/0x9f0 kernel/time/timer.c:2386 run_timer_base kernel/time/timer.c:2395 [inline] run_timer_softirq+0xb7/0x170 kernel/time/timer.c:2405 handle_softirqs+0x1de/0x6d0 kernel/softirq.c:622 __do_softirq kernel/softirq.c:656 [inline] run_ktimerd+0x69/0x100 kernel/softirq.c:1151 smpboot_thread_fn+0x541/0xa50 kernel/smpboot.c:160 kthread+0x388/0x470 kernel/kthread.c:436 ret_from_fork+0x514/0xb70 arch/x86/kernel/process.c:158 ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:245 Reported-by: syzbot+504e778ddaecd36fdd17@syzkaller.appspotmail.com Link: https://sashiko.dev/#/patchset/20260415200216.79699-1-ja%40ssi.bg Link: https://sashiko.dev/#/patchset/20260420165539.85174-4-ja%40ssi.bg Link: https://sashiko.dev/#/patchset/20260422135823.50489-4-ja%40ssi.bg Fixes: 2fa7cc9c7025 ("ipvs: switch to per-net connection table") Signed-off-by: Julian Anastasov Signed-off-by: Pablo Neira Ayuso --- net/netfilter/ipvs/ip_vs_conn.c | 74 +++++++++++++++++++++++------------------ 1 file changed, 41 insertions(+), 33 deletions(-) (limited to 'net') diff --git a/net/netfilter/ipvs/ip_vs_conn.c b/net/netfilter/ipvs/ip_vs_conn.c index 84a4921a7865..9ea6b4fa78bf 100644 --- a/net/netfilter/ipvs/ip_vs_conn.c +++ b/net/netfilter/ipvs/ip_vs_conn.c @@ -267,27 +267,20 @@ static inline int ip_vs_conn_hash(struct ip_vs_conn *cp) hash_key2 = hash_key; use2 = false; } + conn_tab_lock(t, cp, hash_key, hash_key2, use2, true /* new_hash */, &head, &head2); - spin_lock(&cp->lock); - - if (!(cp->flags & IP_VS_CONN_F_HASHED)) { - cp->flags |= IP_VS_CONN_F_HASHED; - WRITE_ONCE(cp->hn0.hash_key, hash_key); - WRITE_ONCE(cp->hn1.hash_key, hash_key2); - refcount_inc(&cp->refcnt); - hlist_bl_add_head_rcu(&cp->hn0.node, head); - if (use2) - hlist_bl_add_head_rcu(&cp->hn1.node, head2); - ret = 1; - } else { - pr_err("%s(): request for already hashed, called from %pS\n", - __func__, __builtin_return_address(0)); - ret = 0; - } - spin_unlock(&cp->lock); + cp->flags |= IP_VS_CONN_F_HASHED; + WRITE_ONCE(cp->hn0.hash_key, hash_key); + WRITE_ONCE(cp->hn1.hash_key, hash_key2); + refcount_inc(&cp->refcnt); + hlist_bl_add_head_rcu(&cp->hn0.node, head); + if (use2) + hlist_bl_add_head_rcu(&cp->hn1.node, head2); + conn_tab_unlock(head, head2); + ret = 1; /* Schedule resizing if load increases */ if (atomic_read(&ipvs->conn_count) > t->u_thresh && @@ -321,7 +314,6 @@ static inline bool ip_vs_conn_unlink(struct ip_vs_conn *cp) conn_tab_lock(t, cp, hash_key, hash_key2, use2, false /* new_hash */, &head, &head2); - spin_lock(&cp->lock); if (cp->flags & IP_VS_CONN_F_HASHED) { /* Decrease refcnt and unlink conn only if we are last user */ @@ -334,7 +326,6 @@ static inline bool ip_vs_conn_unlink(struct ip_vs_conn *cp) } } - spin_unlock(&cp->lock); conn_tab_unlock(head, head2); rcu_read_unlock(); @@ -637,6 +628,7 @@ void ip_vs_conn_fill_cport(struct ip_vs_conn *cp, __be16 cport) struct ip_vs_conn_hnode *hn; u32 hash_key, hash_key_new; struct ip_vs_conn_param p; + bool by_me = false; int ntbl; int dir; @@ -664,8 +656,16 @@ retry: t = rcu_dereference(t->new_tbl); ntbl++; /* We are lost? */ - if (ntbl >= 2) + if (ntbl >= 2) { + spin_lock_bh(&cp->lock); + if (cp->flags & IP_VS_CONN_F_NO_CPORT && by_me) + cp->cport = 0; + /* hn1 will be rehashed on next packet */ + spin_unlock_bh(&cp->lock); + IP_VS_ERR_RL("%s(): Too many ht changes for dir %d\n", + __func__, dir); return; + } } /* Rehashing during resize? Use the recent table for adds */ @@ -683,10 +683,13 @@ retry: if (head > head2 && t == t2) swap(head, head2); + /* Protect the cp->flags modification */ + spin_lock_bh(&cp->lock); + /* Lock seqcount only for the old bucket, even if we are on new table * because it affects the del operation, not the adding. */ - spin_lock_bh(&t->lock[hash_key & t->lock_mask].l); + spin_lock(&t->lock[hash_key & t->lock_mask].l); preempt_disable_nested(); write_seqcount_begin(&t->seqc[hash_key & t->seqc_mask]); @@ -704,14 +707,23 @@ retry: hlist_bl_unlock(head); write_seqcount_end(&t->seqc[hash_key & t->seqc_mask]); preempt_enable_nested(); - spin_unlock_bh(&t->lock[hash_key & t->lock_mask].l); + spin_unlock(&t->lock[hash_key & t->lock_mask].l); + spin_unlock_bh(&cp->lock); hash_key = hash_key_new; goto retry; } - spin_lock(&cp->lock); - if ((cp->flags & IP_VS_CONN_F_NO_CPORT) && - (cp->flags & IP_VS_CONN_F_HASHED)) { + /* Fill cport once, even if multiple packets try to do it */ + if (cp->flags & IP_VS_CONN_F_NO_CPORT && (!cp->cport || by_me)) { + /* If we race with resizing make sure cport is set for dir 1 */ + if (!cp->cport) { + cp->cport = cport; + by_me = true; + } + if (!dir) { + atomic_dec(&ipvs->no_cport_conns[af_id]); + cp->flags &= ~IP_VS_CONN_F_NO_CPORT; + } /* We do not recalc hash_key_r under lock, we assume the * parameters in cp do not change, i.e. cport is * the only possible change. @@ -726,21 +738,17 @@ retry: hlist_bl_del_rcu(&hn->node); hlist_bl_add_head_rcu(&hn->node, head_new); } - if (!dir) { - atomic_dec(&ipvs->no_cport_conns[af_id]); - cp->flags &= ~IP_VS_CONN_F_NO_CPORT; - cp->cport = cport; - } } - spin_unlock(&cp->lock); if (head != head2) hlist_bl_unlock(head2); hlist_bl_unlock(head); write_seqcount_end(&t->seqc[hash_key & t->seqc_mask]); preempt_enable_nested(); - spin_unlock_bh(&t->lock[hash_key & t->lock_mask].l); - if (dir--) + spin_unlock(&t->lock[hash_key & t->lock_mask].l); + + spin_unlock_bh(&cp->lock); + if (dir-- && by_me) goto next_dir; } -- cgit v1.2.3 From fbe1e01e818ee6db86ff947599bf0bea96de7e71 Mon Sep 17 00:00:00 2001 From: Julian Anastasov Date: Thu, 30 Apr 2026 10:44:16 +0300 Subject: ipvs: do not leak dest after get from dest trash Sashiko warns about leaked dest if ip_vs_start_estimator() fails in ip_vs_add_dest(). Add ip_vs_trash_put_dest() to put back the dest into dest trash. Link: https://sashiko.dev/#/patchset/20260428175725.72050-1-ja%40ssi.bg Fixes: 705dd3444081 ("ipvs: use kthreads for stats estimation") Signed-off-by: Julian Anastasov Signed-off-by: Pablo Neira Ayuso --- net/netfilter/ipvs/ip_vs_ctl.c | 37 ++++++++++++++++++++++++------------- 1 file changed, 24 insertions(+), 13 deletions(-) (limited to 'net') diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c index caec516856e9..d81077c2457a 100644 --- a/net/netfilter/ipvs/ip_vs_ctl.c +++ b/net/netfilter/ipvs/ip_vs_ctl.c @@ -1102,6 +1102,24 @@ out: return dest; } +/* Put destination in trash */ +static void ip_vs_trash_put_dest(struct netns_ipvs *ipvs, + struct ip_vs_dest *dest, unsigned long istart, + bool cleanup) +{ + spin_lock_bh(&ipvs->dest_trash_lock); + IP_VS_DBG_BUF(3, "Moving dest %s:%u into trash, dest->refcnt=%d\n", + IP_VS_DBG_ADDR(dest->af, &dest->addr), ntohs(dest->port), + refcount_read(&dest->refcnt)); + if (list_empty(&ipvs->dest_trash) && !cleanup) + mod_timer(&ipvs->dest_trash_timer, + jiffies + (IP_VS_DEST_TRASH_PERIOD >> 1)); + /* dest lives in trash with reference */ + list_add(&dest->t_list, &ipvs->dest_trash); + dest->idle_start = istart; + spin_unlock_bh(&ipvs->dest_trash_lock); +} + static void ip_vs_dest_rcu_free(struct rcu_head *head) { struct ip_vs_dest *dest; @@ -1461,9 +1479,12 @@ ip_vs_add_dest(struct ip_vs_service *svc, struct ip_vs_dest_user_kern *udest) ntohs(dest->vport)); ret = ip_vs_start_estimator(svc->ipvs, &dest->stats); + /* On error put back dest into the trash */ if (ret < 0) - return ret; - __ip_vs_update_dest(svc, dest, udest, 1); + ip_vs_trash_put_dest(svc->ipvs, dest, dest->idle_start, + false); + else + __ip_vs_update_dest(svc, dest, udest, 1); } else { /* * Allocate and initialize the dest structure @@ -1533,17 +1554,7 @@ static void __ip_vs_del_dest(struct netns_ipvs *ipvs, struct ip_vs_dest *dest, */ ip_vs_rs_unhash(dest); - spin_lock_bh(&ipvs->dest_trash_lock); - IP_VS_DBG_BUF(3, "Moving dest %s:%u into trash, dest->refcnt=%d\n", - IP_VS_DBG_ADDR(dest->af, &dest->addr), ntohs(dest->port), - refcount_read(&dest->refcnt)); - if (list_empty(&ipvs->dest_trash) && !cleanup) - mod_timer(&ipvs->dest_trash_timer, - jiffies + (IP_VS_DEST_TRASH_PERIOD >> 1)); - /* dest lives in trash with reference */ - list_add(&dest->t_list, &ipvs->dest_trash); - dest->idle_start = 0; - spin_unlock_bh(&ipvs->dest_trash_lock); + ip_vs_trash_put_dest(ipvs, dest, 0, cleanup); /* Queue up delayed work to expire all no destination connections. * No-op when CONFIG_SYSCTL is disabled. -- cgit v1.2.3 From 2fd109238925d53c44ea409df0558844af7877b8 Mon Sep 17 00:00:00 2001 From: Julian Anastasov Date: Thu, 30 Apr 2026 10:44:17 +0300 Subject: ipvs: fix races around est_mutex and est_cpulist Sashiko reports for races and possible crash around the usage of est_cpulist_valid and sysctl_est_cpulist. The problem is that we do not lock est_mutex in some places which can lead to wrong write ordering and as result problems when calling cpumask_weight() and cpumask_empty(). Fix them by moving the est_max_threads read/write under locked est_mutex. Do the same for one ip_vs_est_reload_start() call to protect the cpumask_empty() usage of sysctl_est_cpulist. To remove the chance of deadlock while stopping the estimation kthreads, keep the data structure for kthread 0 even after last estimator is removed and do not hold mutexes while stopping this task. Now we will use a new flag 'needed' to know when kthread 0 should run. The kthreads above 0 do not use mutexes, so stop them under est_mutex because their kthread data still can be destroyed if they do not serve estimators. Now all kthreads will be started by the est_reload_work to properly serialize the stop/start for kthread 0. Reduce the use of service_mutex in ip_vs_est_calc_phase() because under est_mutex we can safely walk est_kt_arr to stop the kthreads above slot 0. As ip_vs_stop_estimator() for tot_stats should be called under service_mutex, do it early in the netns exit path in ip_vs_flush() to avoid locking the mutex again later. It still should be called in ip_vs_control_net_cleanup_sysctl() when we are called during netns init error. Use -2 for ktid as indicator if estimator was already stopped. Finally, fix use-after-free for kd->est_row in ip_vs_est_calc_phase(). est->ktrow should simply switch to a delay value while estimator is linked to est_temp_list. Link: https://sashiko.dev/#/patchset/20260331165015.2777765-1-longman%40redhat.com Link: https://sashiko.dev/#/patchset/20260420171308.87192-1-ja%40ssi.bg Link: https://sashiko.dev/#/patchset/20260422125123.40658-1-ja%40ssi.bg Link: https://sashiko.dev/#/patchset/20260424175858.54752-1-ja%40ssi.bg Link: https://sashiko.dev/#/patchset/20260425103918.7447-1-ja%40ssi.bg Fixes: f0be83d54217 ("ipvs: add est_cpulist and est_nice sysctl vars") Signed-off-by: Julian Anastasov Signed-off-by: Pablo Neira Ayuso --- net/netfilter/ipvs/ip_vs_ctl.c | 51 +++++++++++++++++++++----- net/netfilter/ipvs/ip_vs_est.c | 83 ++++++++++++++++++++++++------------------ 2 files changed, 90 insertions(+), 44 deletions(-) (limited to 'net') diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c index d81077c2457a..5c9f8e0e238f 100644 --- a/net/netfilter/ipvs/ip_vs_ctl.c +++ b/net/netfilter/ipvs/ip_vs_ctl.c @@ -261,12 +261,28 @@ static void est_reload_work_handler(struct work_struct *work) if (!kd) continue; /* New config ? Stop kthread tasks */ - if (genid != genid_done) - ip_vs_est_kthread_stop(kd); + if (genid != genid_done) { + if (!id) { + /* Only we can stop kt 0 but not under mutex */ + mutex_unlock(&ipvs->est_mutex); + ip_vs_est_kthread_stop(kd); + mutex_lock(&ipvs->est_mutex); + if (!READ_ONCE(ipvs->enable)) + goto unlock; + /* kd for kt 0 is never destroyed */ + } else { + ip_vs_est_kthread_stop(kd); + } + } if (!kd->task && !ip_vs_est_stopped(ipvs)) { + bool start; + /* Do not start kthreads above 0 in calc phase */ - if ((!id || !ipvs->est_calc_phase) && - ip_vs_est_kthread_start(ipvs, kd) < 0) + if (id) + start = !ipvs->est_calc_phase; + else + start = kd->needed; + if (start && ip_vs_est_kthread_start(ipvs, kd) < 0) repeat = true; } } @@ -1823,11 +1839,16 @@ ip_vs_add_service(struct netns_ipvs *ipvs, struct ip_vs_service_user_kern *u, *svc_p = svc; if (!READ_ONCE(ipvs->enable)) { + mutex_lock(&ipvs->est_mutex); + /* Now there is a service - full throttle */ WRITE_ONCE(ipvs->enable, 1); + ipvs->est_max_threads = ip_vs_est_max_threads(ipvs); + /* Start estimation for first time */ - ip_vs_est_reload_start(ipvs); + ip_vs_est_reload_start(ipvs, true); + mutex_unlock(&ipvs->est_mutex); } return 0; @@ -2103,6 +2124,11 @@ static int ip_vs_flush(struct netns_ipvs *ipvs, bool cleanup) t = p; } } + /* Stop the tot_stats estimator early under service_mutex + * to avoid locking it again later. + */ + if (cleanup) + ip_vs_stop_estimator_tot_stats(ipvs); return 0; } @@ -2348,7 +2374,7 @@ static int ipvs_proc_est_cpumask_set(const struct ctl_table *table, /* est_max_threads may depend on cpulist size */ ipvs->est_max_threads = ip_vs_est_max_threads(ipvs); ipvs->est_calc_phase = 1; - ip_vs_est_reload_start(ipvs); + ip_vs_est_reload_start(ipvs, true); unlock: mutex_unlock(&ipvs->est_mutex); @@ -2428,7 +2454,7 @@ static int ipvs_proc_est_nice(const struct ctl_table *table, int write, mutex_lock(&ipvs->est_mutex); if (*valp != val) { *valp = val; - ip_vs_est_reload_start(ipvs); + ip_vs_est_reload_start(ipvs, true); } mutex_unlock(&ipvs->est_mutex); } @@ -2455,7 +2481,7 @@ static int ipvs_proc_run_estimation(const struct ctl_table *table, int write, mutex_lock(&ipvs->est_mutex); if (*valp != val) { *valp = val; - ip_vs_est_reload_start(ipvs); + ip_vs_est_reload_start(ipvs, true); } mutex_unlock(&ipvs->est_mutex); } @@ -5005,7 +5031,14 @@ static void __net_exit ip_vs_control_net_cleanup_sysctl(struct netns_ipvs *ipvs) cancel_delayed_work_sync(&ipvs->defense_work); cancel_work_sync(&ipvs->defense_work.work); unregister_net_sysctl_table(ipvs->sysctl_hdr); - ip_vs_stop_estimator(ipvs, &ipvs->tot_stats->s); + if (ipvs->tot_stats->s.est.ktid != -2) { + /* Not stopped yet? This happens only on netns init error and + * we even do not need to lock the service_mutex for this case. + */ + mutex_lock(&ipvs->service_mutex); + ip_vs_stop_estimator(ipvs, &ipvs->tot_stats->s); + mutex_unlock(&ipvs->service_mutex); + } if (ipvs->est_cpulist_valid) free_cpumask_var(ipvs->sysctl_est_cpulist); diff --git a/net/netfilter/ipvs/ip_vs_est.c b/net/netfilter/ipvs/ip_vs_est.c index 433ba3cab58c..ab09f5182951 100644 --- a/net/netfilter/ipvs/ip_vs_est.c +++ b/net/netfilter/ipvs/ip_vs_est.c @@ -68,6 +68,11 @@ and the limit of estimators per kthread - est_add_ktid: ktid where to add new ests, can point to empty slot where we should add kt data + - data protected by service_mutex: est_temp_list, est_add_ktid, + est_kt_count(R/W), est_kt_arr(R/W), est_genid_done, kd->needed(R/W) + - data protected by est_mutex: est_genid, est_max_threads, sysctl_est_cpulist, + est_cpulist_valid, sysctl_est_nice, est_stopped, sysctl_run_estimation, + est_kt_count(R), est_kt_arr(R), kd->needed(R), kd->task (id > 0) */ static struct lock_class_key __ipvs_est_key; @@ -227,14 +232,17 @@ static int ip_vs_estimation_kthread(void *data) } /* Schedule stop/start for kthread tasks */ -void ip_vs_est_reload_start(struct netns_ipvs *ipvs) +void ip_vs_est_reload_start(struct netns_ipvs *ipvs, bool restart) { + lockdep_assert_held(&ipvs->est_mutex); + /* Ignore reloads before first service is added */ if (!READ_ONCE(ipvs->enable)) return; ip_vs_est_stopped_recalc(ipvs); - /* Bump the kthread configuration genid */ - atomic_inc(&ipvs->est_genid); + /* Bump the kthread configuration genid if stopping is requested */ + if (restart) + atomic_inc(&ipvs->est_genid); queue_delayed_work(system_long_wq, &ipvs->est_reload_work, 0); } @@ -304,12 +312,17 @@ static int ip_vs_est_add_kthread(struct netns_ipvs *ipvs) void *arr = NULL; int i; - if ((unsigned long)ipvs->est_kt_count >= ipvs->est_max_threads && - READ_ONCE(ipvs->enable) && ipvs->est_max_threads) - return -EINVAL; - mutex_lock(&ipvs->est_mutex); + /* Allow kt 0 data to be created before the services are added + * and limit the kthreads when services are present. + */ + if ((unsigned long)ipvs->est_kt_count >= ipvs->est_max_threads && + READ_ONCE(ipvs->enable) && ipvs->est_max_threads) { + ret = -EINVAL; + goto out; + } + for (i = 0; i < id; i++) { if (!ipvs->est_kt_arr[i]) break; @@ -333,6 +346,7 @@ static int ip_vs_est_add_kthread(struct netns_ipvs *ipvs) kd->est_timer = jiffies; kd->id = id; ip_vs_est_set_params(ipvs, kd); + kd->needed = 1; /* Pre-allocate stats used in calc phase */ if (!id && !kd->calc_stats) { @@ -341,12 +355,8 @@ static int ip_vs_est_add_kthread(struct netns_ipvs *ipvs) goto out; } - /* Start kthread tasks only when services are present */ - if (READ_ONCE(ipvs->enable) && !ip_vs_est_stopped(ipvs)) { - ret = ip_vs_est_kthread_start(ipvs, kd); - if (ret < 0) - goto out; - } + /* Request kthread to be started */ + ip_vs_est_reload_start(ipvs, false); if (arr) ipvs->est_kt_count++; @@ -482,12 +492,11 @@ out: /* Start estimation for stats */ int ip_vs_start_estimator(struct netns_ipvs *ipvs, struct ip_vs_stats *stats) { + struct ip_vs_est_kt_data *kd = ipvs->est_kt_count > 0 ? + ipvs->est_kt_arr[0] : NULL; struct ip_vs_estimator *est = &stats->est; int ret; - if (!ipvs->est_max_threads && READ_ONCE(ipvs->enable)) - ipvs->est_max_threads = ip_vs_est_max_threads(ipvs); - est->ktid = -1; est->ktrow = IPVS_EST_NTICKS - 1; /* Initial delay */ @@ -496,8 +505,15 @@ int ip_vs_start_estimator(struct netns_ipvs *ipvs, struct ip_vs_stats *stats) * will not allocate much memory, just for kt 0. */ ret = 0; - if (!ipvs->est_kt_count || !ipvs->est_kt_arr[0]) + if (!kd) { ret = ip_vs_est_add_kthread(ipvs); + } else if (!kd->needed) { + mutex_lock(&ipvs->est_mutex); + /* We have job for the kt 0 task */ + kd->needed = 1; + ip_vs_est_reload_start(ipvs, true); + mutex_unlock(&ipvs->est_mutex); + } if (ret >= 0) hlist_add_head(&est->list, &ipvs->est_temp_list); else @@ -578,16 +594,14 @@ void ip_vs_stop_estimator(struct netns_ipvs *ipvs, struct ip_vs_stats *stats) } end_kt0: - /* kt 0 is freed after all other kthreads and chains are empty */ + /* kt 0 task is stopped after all other kt slots and chains are empty */ if (ipvs->est_kt_count == 1 && hlist_empty(&ipvs->est_temp_list)) { kd = ipvs->est_kt_arr[0]; - if (!kd || !kd->est_count) { + if (kd && !kd->est_count) { mutex_lock(&ipvs->est_mutex); - if (kd) { - ip_vs_est_kthread_destroy(kd); - ipvs->est_kt_arr[0] = NULL; - } - ipvs->est_kt_count--; + /* Keep the kt0 data but request kthread_stop */ + kd->needed = 0; + ip_vs_est_reload_start(ipvs, true); mutex_unlock(&ipvs->est_mutex); ipvs->est_add_ktid = 0; } @@ -647,9 +661,9 @@ static int ip_vs_est_calc_limits(struct netns_ipvs *ipvs, int *chain_max) u64 val; INIT_HLIST_HEAD(&chain); - mutex_lock(&ipvs->service_mutex); + mutex_lock(&ipvs->est_mutex); kd = ipvs->est_kt_arr[0]; - mutex_unlock(&ipvs->service_mutex); + mutex_unlock(&ipvs->est_mutex); s = kd ? kd->calc_stats : NULL; if (!s) goto out; @@ -748,16 +762,16 @@ static void ip_vs_est_calc_phase(struct netns_ipvs *ipvs) if (!ip_vs_est_calc_limits(ipvs, &chain_max)) return; - mutex_lock(&ipvs->service_mutex); - /* Stop all other tasks, so that we can immediately move the * estimators to est_temp_list without RCU grace period */ mutex_lock(&ipvs->est_mutex); for (id = 1; id < ipvs->est_kt_count; id++) { /* netns clean up started, abort */ - if (!READ_ONCE(ipvs->enable)) - goto unlock2; + if (kthread_should_stop() || !READ_ONCE(ipvs->enable)) { + mutex_unlock(&ipvs->est_mutex); + return; + } kd = ipvs->est_kt_arr[id]; if (!kd) continue; @@ -765,9 +779,11 @@ static void ip_vs_est_calc_phase(struct netns_ipvs *ipvs) } mutex_unlock(&ipvs->est_mutex); + mutex_lock(&ipvs->service_mutex); + /* Move all estimators to est_temp_list but carefully, * all estimators and kthread data can be released while - * we reschedule. Even for kthread 0. + * we reschedule. */ step = 0; @@ -849,9 +865,7 @@ walk_chain: ip_vs_stop_estimator(ipvs, stats); /* Tasks are stopped, move without RCU grace period */ est->ktid = -1; - est->ktrow = row - kd->est_row; - if (est->ktrow < 0) - est->ktrow += IPVS_EST_NTICKS; + est->ktrow = delay; hlist_add_head(&est->list, &ipvs->est_temp_list); /* kd freed ? */ if (last) @@ -889,7 +903,6 @@ end_dequeue: if (genid == atomic_read(&ipvs->est_genid)) ipvs->est_calc_phase = 0; -unlock2: mutex_unlock(&ipvs->est_mutex); unlock: -- cgit v1.2.3 From 4ee52b7021a7cb9356f8b9aff5631c68512a9e1b Mon Sep 17 00:00:00 2001 From: Julian Anastasov Date: Thu, 30 Apr 2026 10:44:18 +0300 Subject: ipvs: fix shift-out-of-bounds in ip_vs_rht_desired_size Calling roundup_pow_of_two() with 0 has undefined result: UBSAN: shift-out-of-bounds in ./include/linux/log2.h:57:13 shift exponent 64 is too large for 64-bit type 'unsigned long' CPU: 1 UID: 0 PID: 77 Comm: kworker/u8:4 Not tainted syzkaller #0 PREEMPT(full) Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 04/18/2026 Workqueue: events_unbound conn_resize_work_handler Call Trace: dump_stack_lvl+0xe8/0x150 lib/dump_stack.c:120 ubsan_epilogue+0xa/0x30 lib/ubsan.c:233 __ubsan_handle_shift_out_of_bounds+0x385/0x410 lib/ubsan.c:494 __roundup_pow_of_two include/linux/log2.h:57 [inline] ip_vs_rht_desired_size+0x2cf/0x410 net/netfilter/ipvs/ip_vs_core.c:240 ip_vs_conn_desired_size net/netfilter/ipvs/ip_vs_conn.c:765 [inline] conn_resize_work_handler+0x1b6/0x14c0 net/netfilter/ipvs/ip_vs_conn.c:822 process_one_work kernel/workqueue.c:3302 [inline] process_scheduled_works+0xb5d/0x1860 kernel/workqueue.c:3385 worker_thread+0xa53/0xfc0 kernel/workqueue.c:3466 kthread+0x388/0x470 kernel/kthread.c:436 ret_from_fork+0x514/0xb70 arch/x86/kernel/process.c:158 ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:245 Reported-by: syzbot+217f1db9c791e27fe54a@syzkaller.appspotmail.com Fixes: b655388111cf ("ipvs: add resizable hash tables") Signed-off-by: Julian Anastasov Signed-off-by: Pablo Neira Ayuso --- net/netfilter/ipvs/ip_vs_core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/netfilter/ipvs/ip_vs_core.c b/net/netfilter/ipvs/ip_vs_core.c index f5b7a2047291..d40b404c1bf6 100644 --- a/net/netfilter/ipvs/ip_vs_core.c +++ b/net/netfilter/ipvs/ip_vs_core.c @@ -237,7 +237,7 @@ int ip_vs_rht_desired_size(struct netns_ipvs *ipvs, struct ip_vs_rht *t, int n, { if (!t) return 1 << min_bits; - n = roundup_pow_of_two(n); + n = n > 0 ? roundup_pow_of_two(n) : 1; if (lfactor < 0) { int factor = min(-lfactor, max_bits); -- cgit v1.2.3 From aa6065206987278291c09d0c6aebed687114c925 Mon Sep 17 00:00:00 2001 From: Waiman Long Date: Thu, 30 Apr 2026 10:44:19 +0300 Subject: ipvs: Guard access of HK_TYPE_KTHREAD cpumask with RCU The ip_vs_ctl.c file and the associated ip_vs.h file are the only places in the kernel where HK_TYPE_KTHREAD cpumask is being retrieved and used. Now that HK_TYPE_KTHREAD/HK_TYPE_DOMAIN cpumask can be changed at run time. We need to use RCU to guard access to this cpumask to avoid a potential UAF problem as the returned cpumask may be freed before it is being used. We can replace HK_TYPE_KTHREAD by HK_TYPE_DOMAIN as they are aliases of each other, but keeping the HK_TYPE_KTHREAD name can highlight the fact that it is the kthread initiated by ipvs that is being controlled. Fixes: 03ff73510169 ("cpuset: Update HK_TYPE_DOMAIN cpumask from cpuset") Signed-off-by: Waiman Long Signed-off-by: Julian Anastasov Signed-off-by: Pablo Neira Ayuso --- net/netfilter/ipvs/ip_vs_ctl.c | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c index 5c9f8e0e238f..c7c7f6a7a9f6 100644 --- a/net/netfilter/ipvs/ip_vs_ctl.c +++ b/net/netfilter/ipvs/ip_vs_ctl.c @@ -2394,11 +2394,14 @@ static int ipvs_proc_est_cpumask_get(const struct ctl_table *table, mutex_lock(&ipvs->est_mutex); - if (ipvs->est_cpulist_valid) - mask = *valp; - else - mask = (struct cpumask *)housekeeping_cpumask(HK_TYPE_KTHREAD); - ret = scnprintf(buffer, size, "%*pbl\n", cpumask_pr_args(mask)); + /* HK_TYPE_KTHREAD cpumask needs RCU protection */ + scoped_guard(rcu) { + if (ipvs->est_cpulist_valid) + mask = *valp; + else + mask = (struct cpumask *)housekeeping_cpumask(HK_TYPE_KTHREAD); + ret = scnprintf(buffer, size, "%*pbl\n", cpumask_pr_args(mask)); + } mutex_unlock(&ipvs->est_mutex); -- cgit v1.2.3 From d82ba05263c69fa2437fe93e4e561cc40f4c03af Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Fri, 1 May 2026 07:39:41 +0000 Subject: af_unix: Set gc_in_progress to true in unix_gc(). Igor Ushakov reported that unix_gc() could run with gc_in_progress being false if the work is scheduled while running: Thread 1 Thread 2 Thread 3 -------- -------- -------- unix_schedule_gc() unix_schedule_gc() `- if (!gc_in_progress) `- if (!gc_in_progress) |- gc_in_progress = true | `- queue_work() | unix_gc() <----------------/ | | |- gc_in_progress = true ... `- queue_work() | | `- gc_in_progress = false | | unix_gc() <---------------------------------------------' | ... /* gc_in_progress == false */ | `- gc_in_progress = false unix_peek_fpl() relies on gc_in_progress not to confuse GC by MSG_PEEK. Let's set gc_in_progress to true in unix_gc(). Fixes: 8b90a9f819dc ("af_unix: Run GC on only one CPU.") Reported-by: Igor Ushakov Signed-off-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20260501073945.1884564-1-kuniyu@google.com Signed-off-by: Jakub Kicinski --- net/unix/garbage.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/unix/garbage.c b/net/unix/garbage.c index a7967a345827..0783555e2526 100644 --- a/net/unix/garbage.c +++ b/net/unix/garbage.c @@ -607,6 +607,8 @@ static void unix_gc(struct work_struct *work) struct sk_buff_head hitlist; struct sk_buff *skb; + WRITE_ONCE(gc_in_progress, true); + spin_lock(&unix_gc_lock); if (unix_graph_state == UNIX_GRAPH_NOT_CYCLIC) { @@ -649,10 +651,8 @@ void unix_schedule_gc(struct user_struct *user) READ_ONCE(user->unix_inflight) < UNIX_INFLIGHT_SANE_USER) return; - if (!READ_ONCE(gc_in_progress)) { - WRITE_ONCE(gc_in_progress, true); + if (!READ_ONCE(gc_in_progress)) queue_work(system_dfl_wq, &unix_gc_work); - } if (user && READ_ONCE(unix_graph_cyclic_sccs)) flush_work(&unix_gc_work); -- cgit v1.2.3 From 76b93a8107574006b25495664304ea9237494d70 Mon Sep 17 00:00:00 2001 From: Breno Leitao Date: Fri, 1 May 2026 02:58:41 -0700 Subject: netpoll: pass buffer size to egress_dev() to avoid MAC truncation egress_dev() formats np->dev_mac via snprintf() but receives buf as a bare char *, so it cannot derive the buffer size from the pointer. The size argument was hardcoded to MAC_ADDR_STR_LEN (3 * ETH_ALEN - 1 = 17), which is silly wrong in two ways: 1) misleading kernel log output on the MAC-selected target path (np->dev_name[0] == '\0'); for example "aa:bb:cc:dd:ee:ff doesn't exist, aborting" was logged as "aa:bb:cc:dd:ee:f doesn't exist, aborting". 2) the second argument of snprintf is the size of the buffer, not the size of what you want to write. Add a bufsz parameter to egress_dev() and pass sizeof(buf) from each caller, matching the standard snprintf() idiom and removing the hardcoded size from the helper. Every caller already declares "char buf[MAC_ADDR_STR_LEN + 1]" so the formatted MAC continues to fit. Tested by booting with netconsole=6665@/aa:bb:cc:dd:ee:ff,6666@10.0.0.1/00:11:22:33:44:55 on a kernel without a matching device. Pre-fix dmesg shows "aa:bb:cc:dd:ee:f doesn't exist, aborting"; post-fix shows the full "aa:bb:cc:dd:ee:ff doesn't exist, aborting". Fixes: f8a10bed32f5 ("netconsole: allow selection of egress interface via MAC address") Cc: stable@vger.kernel.org Signed-off-by: Breno Leitao Link: https://patch.msgid.link/20260501-netpoll_snprintf_fix-v1-1-84b0566e6597@debian.org Signed-off-by: Jakub Kicinski --- net/core/netpoll.c | 23 +++++++++++++---------- 1 file changed, 13 insertions(+), 10 deletions(-) (limited to 'net') diff --git a/net/core/netpoll.c b/net/core/netpoll.c index 4381e0fc25bf..84faace50ac2 100644 --- a/net/core/netpoll.c +++ b/net/core/netpoll.c @@ -608,14 +608,16 @@ EXPORT_SYMBOL_GPL(__netpoll_setup); /* * Returns a pointer to a string representation of the identifier used * to select the egress interface for the given netpoll instance. buf - * must be a buffer of length at least MAC_ADDR_STR_LEN + 1. + * is used to format np->dev_mac when np->dev_name is empty; bufsz must + * be at least MAC_ADDR_STR_LEN + 1 to fit the formatted MAC address + * and its NUL terminator. */ -static char *egress_dev(struct netpoll *np, char *buf) +static char *egress_dev(struct netpoll *np, char *buf, size_t bufsz) { if (np->dev_name[0]) return np->dev_name; - snprintf(buf, MAC_ADDR_STR_LEN, "%pM", np->dev_mac); + snprintf(buf, bufsz, "%pM", np->dev_mac); return buf; } @@ -645,7 +647,7 @@ static int netpoll_take_ipv6(struct netpoll *np, struct net_device *ndev) if (!IS_ENABLED(CONFIG_IPV6)) { np_err(np, "IPv6 is not supported %s, aborting\n", - egress_dev(np, buf)); + egress_dev(np, buf, sizeof(buf))); return -EINVAL; } @@ -667,7 +669,7 @@ static int netpoll_take_ipv6(struct netpoll *np, struct net_device *ndev) } if (err) { np_err(np, "no IPv6 address for %s, aborting\n", - egress_dev(np, buf)); + egress_dev(np, buf, sizeof(buf))); return err; } @@ -687,14 +689,14 @@ static int netpoll_take_ipv4(struct netpoll *np, struct net_device *ndev) in_dev = __in_dev_get_rtnl(ndev); if (!in_dev) { np_err(np, "no IP address for %s, aborting\n", - egress_dev(np, buf)); + egress_dev(np, buf, sizeof(buf))); return -EDESTADDRREQ; } ifa = rtnl_dereference(in_dev->ifa_list); if (!ifa) { np_err(np, "no IP address for %s, aborting\n", - egress_dev(np, buf)); + egress_dev(np, buf, sizeof(buf))); return -EDESTADDRREQ; } @@ -736,7 +738,8 @@ int netpoll_setup(struct netpoll *np) ndev = dev_getbyhwaddr(net, ARPHRD_ETHER, np->dev_mac); if (!ndev) { - np_err(np, "%s doesn't exist, aborting\n", egress_dev(np, buf)); + np_err(np, "%s doesn't exist, aborting\n", + egress_dev(np, buf, sizeof(buf))); err = -ENODEV; goto unlock; } @@ -744,14 +747,14 @@ int netpoll_setup(struct netpoll *np) if (netdev_master_upper_dev_get(ndev)) { np_err(np, "%s is a slave device, aborting\n", - egress_dev(np, buf)); + egress_dev(np, buf, sizeof(buf))); err = -EBUSY; goto put; } if (!netif_running(ndev)) { np_info(np, "device %s not up yet, forcing it\n", - egress_dev(np, buf)); + egress_dev(np, buf, sizeof(buf))); err = dev_open(ndev, NULL); if (err) { -- cgit v1.2.3 From 059b7dbd20a6f0c539a45ddff1573cb8946685b5 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 30 Apr 2026 12:26:52 +0000 Subject: vsock/virtio: fix potential unbounded skb queue MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit virtio_transport_inc_rx_pkt() checks vvs->rx_bytes + len > vvs->buf_alloc. virtio_transport_recv_enqueue() skips coalescing for packets with VIRTIO_VSOCK_SEQ_EOM. If fed with packets with len == 0 and VIRTIO_VSOCK_SEQ_EOM, a very large number of packets can be queued because vvs->rx_bytes stays at 0. Fix this by estimating the skb metadata size: (Number of skbs in the queue) * SKB_TRUESIZE(0) Fixes: 077706165717 ("virtio/vsock: don't use skbuff state to account credit") Signed-off-by: Eric Dumazet Cc: Arseniy Krasnov Cc: Stefan Hajnoczi Cc: Stefano Garzarella Cc: "Michael S. Tsirkin" Cc: Jason Wang Cc: Xuan Zhuo Cc: "Eugenio Pérez" Cc: virtualization@lists.linux.dev Link: https://patch.msgid.link/20260430122653.554058-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/vmw_vsock/virtio_transport_common.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/vmw_vsock/virtio_transport_common.c b/net/vmw_vsock/virtio_transport_common.c index 416d533f493d..9b8014516f4f 100644 --- a/net/vmw_vsock/virtio_transport_common.c +++ b/net/vmw_vsock/virtio_transport_common.c @@ -447,7 +447,9 @@ static int virtio_transport_send_pkt_info(struct vsock_sock *vsk, static bool virtio_transport_inc_rx_pkt(struct virtio_vsock_sock *vvs, u32 len) { - if (vvs->buf_used + len > vvs->buf_alloc) + u64 skb_overhead = (skb_queue_len(&vvs->rx_queue) + 1) * SKB_TRUESIZE(0); + + if (skb_overhead + vvs->buf_used + len > vvs->buf_alloc) return false; vvs->rx_bytes += len; -- cgit v1.2.3 From c4a99a921949cddc590b22bb14eeb23dffcc3ba6 Mon Sep 17 00:00:00 2001 From: Shardul Bankar Date: Fri, 1 May 2026 21:35:34 +0200 Subject: mptcp: use MPJoinSynAckHMacFailure for SynAck HMAC failure In subflow_finish_connect(), HMAC validation of the server's HMAC in SYN/ACK + MP_JOIN increments MPTCP_MIB_JOINACKMAC ("HMAC was wrong on ACK + MP_JOIN") on failure. The function processes the SYN/ACK, not the ACK; the matching MPTCP_MIB_JOINSYNACKMAC counter ("HMAC was wrong on SYN/ACK + MP_JOIN") exists but is not incremented anywhere in the tree. The mirror site on the server, subflow_syn_recv_sock(), already uses JOINACKMAC correctly for ACK HMAC failure. Use JOINSYNACKMAC at the SYN/ACK validation site so each counter reflects the packet whose HMAC actually failed. Suggested-by: Matthieu Baerts (NGI0) Fixes: fc518953bc9c ("mptcp: add and use MIB counter infrastructure") Cc: stable@vger.kernel.org Signed-off-by: Shardul Bankar Reviewed-by: Matthieu Baerts (NGI0) Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20260501-net-mptcp-misc-fixes-7-1-rc3-v1-1-b70118df778e@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/subflow.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c index e2cb9d23e4a0..bda6862264ca 100644 --- a/net/mptcp/subflow.c +++ b/net/mptcp/subflow.c @@ -581,7 +581,7 @@ static void subflow_finish_connect(struct sock *sk, const struct sk_buff *skb) subflow->backup); if (!subflow_thmac_valid(subflow)) { - MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_JOINACKMAC); + MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_JOINSYNACKMAC); subflow->reset_reason = MPTCP_RST_EMPTCP; goto do_reset; } -- cgit v1.2.3 From a6da02d4c00fdda2417e42ad2b762a9209e6cc49 Mon Sep 17 00:00:00 2001 From: Shardul Bankar Date: Fri, 1 May 2026 21:35:35 +0200 Subject: mptcp: use MPTCP_RST_EMPTCP for ACK HMAC validation failure When HMAC validation fails on a received ACK + MP_JOIN in subflow_syn_recv_sock(), the subflow is reset with reason MPTCP_RST_EPROHIBIT ("Administratively prohibited"). This is incorrect: HMAC validation failure is an MPTCP protocol-level error, not an administrative policy denial. The mirror site on the client, in subflow_finish_connect(), already uses MPTCP_RST_EMPTCP ("MPTCP-specific error") for the same kind of HMAC failure on the SYN/ACK + MP_JOIN. Use the same reason on the server side for symmetry and accuracy. Suggested-by: Matthieu Baerts (NGI0) Fixes: 443041deb5ef ("mptcp: fix NULL pointer in can_accept_new_subflow") Cc: stable@vger.kernel.org Signed-off-by: Shardul Bankar Reviewed-by: Matthieu Baerts (NGI0) Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20260501-net-mptcp-misc-fixes-7-1-rc3-v1-2-b70118df778e@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/subflow.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c index bda6862264ca..d562e149606f 100644 --- a/net/mptcp/subflow.c +++ b/net/mptcp/subflow.c @@ -908,7 +908,7 @@ create_child: if (!subflow_hmac_valid(subflow_req, &mp_opt)) { SUBFLOW_REQ_INC_STATS(req, MPTCP_MIB_JOINACKMAC); - subflow_add_reset_reason(skb, MPTCP_RST_EPROHIBIT); + subflow_add_reset_reason(skb, MPTCP_RST_EMPTCP); goto dispose_child; } -- cgit v1.2.3 From 6254a16d6f0c672e3809ca5d7c9a28a55d71f764 Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Fri, 1 May 2026 21:35:36 +0200 Subject: mptcp: fix rx timestamp corruption on fastopen The skb cb offset containing the timestamp presence flag is cleared before loading such information. Cache such value before MPTCP CB initialization. Fixes: 36b122baf6a8 ("mptcp: add subflow_v(4,6)_send_synack()") Cc: stable@vger.kernel.org Signed-off-by: Paolo Abeni Reviewed-by: Matthieu Baerts (NGI0) Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20260501-net-mptcp-misc-fixes-7-1-rc3-v1-3-b70118df778e@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/fastopen.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/mptcp/fastopen.c b/net/mptcp/fastopen.c index 82ec15bcfd7f..082c46c0f50e 100644 --- a/net/mptcp/fastopen.c +++ b/net/mptcp/fastopen.c @@ -12,6 +12,7 @@ void mptcp_fastopen_subflow_synack_set_params(struct mptcp_subflow_context *subf struct sock *sk, *ssk; struct sk_buff *skb; struct tcp_sock *tp; + bool has_rxtstamp; /* on early fallback the subflow context is deleted by * subflow_syn_recv_sock() @@ -40,12 +41,13 @@ void mptcp_fastopen_subflow_synack_set_params(struct mptcp_subflow_context *subf */ tp->copied_seq += skb->len; subflow->ssn_offset += skb->len; + has_rxtstamp = TCP_SKB_CB(skb)->has_rxtstamp; /* Only the sequence delta is relevant */ MPTCP_SKB_CB(skb)->map_seq = -skb->len; MPTCP_SKB_CB(skb)->end_seq = 0; MPTCP_SKB_CB(skb)->offset = 0; - MPTCP_SKB_CB(skb)->has_rxtstamp = TCP_SKB_CB(skb)->has_rxtstamp; + MPTCP_SKB_CB(skb)->has_rxtstamp = has_rxtstamp; MPTCP_SKB_CB(skb)->cant_coalesce = 1; mptcp_data_lock(sk); -- cgit v1.2.3 From 70ece9d7021c54cf40c72b31b066e9088f5f75f5 Mon Sep 17 00:00:00 2001 From: "Matthieu Baerts (NGI0)" Date: Fri, 1 May 2026 21:35:37 +0200 Subject: mptcp: sockopt: increase seq in mptcp_setsockopt_all_sf mptcp_setsockopt_all_sf() was missing a call to sockopt_seq_inc(). This is required not to cause missing synchronization for newer subflows created later on. This helper is called each time a socket option is set on subflows, and future ones will need to inherit this option after their creation. Fixes: 51c5fd09e1b4 ("mptcp: add TCP_MAXSEG sockopt support") Cc: stable@vger.kernel.org Suggested-by: Paolo Abeni Reviewed-by: Mat Martineau Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20260501-net-mptcp-misc-fixes-7-1-rc3-v1-4-b70118df778e@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/sockopt.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'net') diff --git a/net/mptcp/sockopt.c b/net/mptcp/sockopt.c index 0efe40be2fde..1cf608e7357b 100644 --- a/net/mptcp/sockopt.c +++ b/net/mptcp/sockopt.c @@ -812,6 +812,10 @@ static int mptcp_setsockopt_all_sf(struct mptcp_sock *msk, int level, if (ret) break; } + + if (!ret) + sockopt_seq_inc(msk); + return ret; } -- cgit v1.2.3 From ac0841d7d202073415c808bda7848502163b87dd Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Sat, 2 May 2026 12:41:02 +0000 Subject: net: prevent possible UAF in rtnl_prop_list_size() I was mistaken by synchronize_rcu() [1] call in netdev_name_node_alt_destroy(), giving a false sense of RCU safety at delete times. We have to use list_del_rcu() to not confuse potential readers in rtnl_prop_list_size(). [1] This synchronize_rcu() call was later removed in commit 723de3ebef03 ("net: free altname using an RCU callback"). Fixes: 9f30831390ed ("net: add rcu safety to rtnl_prop_list_size()") Signed-off-by: Eric Dumazet Link: https://patch.msgid.link/20260502124102.499204-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/core/dev.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/core/dev.c b/net/core/dev.c index 06c195906231..8bfa8313ef62 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -371,7 +371,7 @@ static void netdev_name_node_alt_free(struct rcu_head *head) static void __netdev_name_node_alt_destroy(struct netdev_name_node *name_node) { netdev_name_node_del(name_node); - list_del(&name_node->list); + list_del_rcu(&name_node->list); call_rcu(&name_node->rcu, netdev_name_node_alt_free); } -- cgit v1.2.3 From 30cb24f97d44f6b81c14b85c5323de62eef1fb7f Mon Sep 17 00:00:00 2001 From: David Carlier Date: Sat, 2 May 2026 15:19:45 +0100 Subject: psp: strip variable-length PSP header in psp_dev_rcv() psp_dev_rcv() unconditionally removes a fixed PSP_ENCAP_HLEN, even when psph->hdrlen indicates that the PSP header carries optional fields. A frame whose PSP header advertises a non-zero VC or any extension would therefore be silently mis-decapsulated: option bytes would spill into the inner packet head and downstream parsing would fail on a corrupted skb. Compute the full PSP header length from psph->hdrlen, pull the optional bytes into the linear region, and strip the whole header when decapsulating. Optional fields (VC, ...) are still ignored, just discarded with the rest of the header instead of leaking. crypt_offset and the VIRT flag are intentionally not validated here - callers know their device's PSP implementation and can decide. Both in-tree callers gate on hardware-validated PSP, so this is a correctness fix rather than a reachable corruption path under current configurations. Fixes: 0eddb8023cee ("psp: provide decapsulation and receive helper for drivers") Reviewed-by: Willem de Bruijn Reviewed-by: Daniel Zahka Cc: stable@vger.kernel.org Signed-off-by: David Carlier Link: https://patch.msgid.link/20260502141945.14484-1-devnexen@gmail.com Signed-off-by: Jakub Kicinski --- net/psp/psp_main.c | 42 +++++++++++++++++++++++++++++++----------- 1 file changed, 31 insertions(+), 11 deletions(-) (limited to 'net') diff --git a/net/psp/psp_main.c b/net/psp/psp_main.c index 9508b6c38003..e45549f08eef 100644 --- a/net/psp/psp_main.c +++ b/net/psp/psp_main.c @@ -263,15 +263,16 @@ EXPORT_SYMBOL(psp_dev_encapsulate); /* Receive handler for PSP packets. * - * Presently it accepts only already-authenticated packets and does not - * support optional fields, such as virtualization cookies. The caller should - * ensure that skb->data is pointing to the mac header, and that skb->mac_len - * is set. This function does not currently adjust skb->csum (CHECKSUM_COMPLETE - * is not supported). + * Accepts only already-authenticated packets. The full PSP header is + * stripped according to psph->hdrlen; any optional fields it advertises + * (virtualization cookies, etc.) are ignored and discarded along with the + * rest of the header. The caller should ensure that skb->data is pointing + * to the mac header, and that skb->mac_len is set. This function does not + * currently adjust skb->csum (CHECKSUM_COMPLETE is not supported). */ int psp_dev_rcv(struct sk_buff *skb, u16 dev_id, u8 generation, bool strip_icv) { - int l2_hlen = 0, l3_hlen, encap; + int l2_hlen = 0, l3_hlen, encap, psp_hlen; struct psp_skb_ext *pse; struct psphdr *psph; struct ethhdr *eth; @@ -312,18 +313,36 @@ int psp_dev_rcv(struct sk_buff *skb, u16 dev_id, u8 generation, bool strip_icv) if (unlikely(uh->dest != htons(PSP_DEFAULT_UDP_PORT))) return -EINVAL; - pse = skb_ext_add(skb, SKB_EXT_PSP); - if (!pse) + psph = (struct psphdr *)(skb->data + l2_hlen + l3_hlen + + sizeof(struct udphdr)); + + /* Strip the full PSP header per psph->hdrlen; VC/options are pulled + * into the linear region only so they can be discarded with the + * rest of the header. + */ + psp_hlen = (psph->hdrlen + 1) * 8; + + if (unlikely(psp_hlen < sizeof(struct psphdr))) + return -EINVAL; + + if (psp_hlen > sizeof(struct psphdr) && + !pskb_may_pull(skb, l2_hlen + l3_hlen + + sizeof(struct udphdr) + psp_hlen)) return -EINVAL; psph = (struct psphdr *)(skb->data + l2_hlen + l3_hlen + sizeof(struct udphdr)); + + pse = skb_ext_add(skb, SKB_EXT_PSP); + if (!pse) + return -EINVAL; + pse->spi = psph->spi; pse->dev_id = dev_id; pse->generation = generation; pse->version = FIELD_GET(PSPHDR_VERFL_VERSION, psph->verfl); - encap = PSP_ENCAP_HLEN; + encap = sizeof(struct udphdr) + psp_hlen; encap += strip_icv ? PSP_TRL_SIZE : 0; if (proto == htons(ETH_P_IP)) { @@ -340,8 +359,9 @@ int psp_dev_rcv(struct sk_buff *skb, u16 dev_id, u8 generation, bool strip_icv) ipv6h->payload_len = htons(ntohs(ipv6h->payload_len) - encap); } - memmove(skb->data + PSP_ENCAP_HLEN, skb->data, l2_hlen + l3_hlen); - skb_pull(skb, PSP_ENCAP_HLEN); + memmove(skb->data + sizeof(struct udphdr) + psp_hlen, + skb->data, l2_hlen + l3_hlen); + skb_pull(skb, sizeof(struct udphdr) + psp_hlen); if (strip_icv) pskb_trim(skb, skb->len - PSP_TRL_SIZE); -- cgit v1.2.3 From f4c50a4034e62ab75f1d5cdd191dd5f9c77fdff4 Mon Sep 17 00:00:00 2001 From: Kuan-Ting Chen Date: Mon, 4 May 2026 23:27:12 +0800 Subject: xfrm: esp: avoid in-place decrypt on shared skb frags MSG_SPLICE_PAGES can attach pages from a pipe directly to an skb. TCP marks such skbs with SKBFL_SHARED_FRAG after skb_splice_from_iter(), so later paths that may modify packet data can first make a private copy. The IPv4/IPv6 datagram append paths did not set this flag when splicing pages into UDP skbs. That leaves an ESP-in-UDP packet made from shared pipe pages looking like an ordinary uncloned nonlinear skb. ESP input then takes the no-COW fast path for uncloned skbs without a frag_list and decrypts in place over data that is not owned privately by the skb. Mark IPv4/IPv6 datagram splice frags with SKBFL_SHARED_FRAG, matching TCP. Also make ESP input fall back to skb_cow_data() when the flag is present, so ESP does not decrypt externally backed frags in place. Private nonlinear skb frags still use the existing fast path. This intentionally does not change ESP output. In esp_output_head(), the path that appends the ESP trailer to existing skb tailroom without calling skb_cow_data() is not reachable for nonlinear skbs: skb_tailroom() returns zero when skb->data_len is nonzero, while ESP tailen is positive. Thus ESP output will either use the separate destination-frag path or fall back to skb_cow_data(). Fixes: cac2661c53f3 ("esp4: Avoid skb_cow_data whenever possible") Fixes: 03e2a30f6a27 ("esp6: Avoid skb_cow_data whenever possible") Fixes: 7da0dde68486 ("ip, udp: Support MSG_SPLICE_PAGES") Fixes: 6d8192bd69bb ("ip6, udp6: Support MSG_SPLICE_PAGES") Reported-by: Hyunwoo Kim Reported-by: Kuan-Ting Chen Tested-by: Hyunwoo Kim Cc: stable@vger.kernel.org Signed-off-by: Kuan-Ting Chen Signed-off-by: Steffen Klassert --- net/ipv4/esp4.c | 3 ++- net/ipv4/ip_output.c | 2 ++ net/ipv6/esp6.c | 3 ++- net/ipv6/ip6_output.c | 2 ++ 4 files changed, 8 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/ipv4/esp4.c b/net/ipv4/esp4.c index 6dfc0bcdef65..6a5febbdbee4 100644 --- a/net/ipv4/esp4.c +++ b/net/ipv4/esp4.c @@ -873,7 +873,8 @@ static int esp_input(struct xfrm_state *x, struct sk_buff *skb) nfrags = 1; goto skip_cow; - } else if (!skb_has_frag_list(skb)) { + } else if (!skb_has_frag_list(skb) && + !skb_has_shared_frag(skb)) { nfrags = skb_shinfo(skb)->nr_frags; nfrags++; diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index e4790cc7b5c2..5bcd73cbdb41 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -1233,6 +1233,8 @@ alloc_new_skb: if (err < 0) goto error; copy = err; + if (!(flags & MSG_NO_SHARED_FRAGS)) + skb_shinfo(skb)->flags |= SKBFL_SHARED_FRAG; wmem_alloc_delta += copy; } else if (!zc) { int i = skb_shinfo(skb)->nr_frags; diff --git a/net/ipv6/esp6.c b/net/ipv6/esp6.c index 9f75313734f8..9c06c5a1419d 100644 --- a/net/ipv6/esp6.c +++ b/net/ipv6/esp6.c @@ -915,7 +915,8 @@ static int esp6_input(struct xfrm_state *x, struct sk_buff *skb) nfrags = 1; goto skip_cow; - } else if (!skb_has_frag_list(skb)) { + } else if (!skb_has_frag_list(skb) && + !skb_has_shared_frag(skb)) { nfrags = skb_shinfo(skb)->nr_frags; nfrags++; diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index 7e92909ab5be..1f2a33fbed6e 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -1794,6 +1794,8 @@ alloc_new_skb: if (err < 0) goto error; copy = err; + if (!(flags & MSG_NO_SHARED_FRAGS)) + skb_shinfo(skb)->flags |= SKBFL_SHARED_FRAG; wmem_alloc_delta += copy; } else if (!zc) { int i = skb_shinfo(skb)->nr_frags; -- cgit v1.2.3 From f4268b466190dae95a7585f69b4f1f8ad097632c Mon Sep 17 00:00:00 2001 From: Lee Jones Date: Wed, 29 Apr 2026 13:40:41 +0000 Subject: nfc: llcp: Fix use-after-free in llcp_sock_release() llcp_sock_release() unconditionally unlinks the socket from the local sockets list. However, if the socket is still in connecting state, it is on the connecting list. Fix this by checking the socket state and unlinking from the correct list. Fixes: b4011239a08e ("NFC: llcp: Fix non blocking sockets connections") Signed-off-by: Lee Jones Link: https://patch.msgid.link/20260429134115.3558604-1-lee@kernel.org Signed-off-by: David Heidelberg --- net/nfc/llcp_sock.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'net') diff --git a/net/nfc/llcp_sock.c b/net/nfc/llcp_sock.c index f1be1e84f665..feab29fc62f4 100644 --- a/net/nfc/llcp_sock.c +++ b/net/nfc/llcp_sock.c @@ -633,6 +633,8 @@ static int llcp_sock_release(struct socket *sock) if (sock->type == SOCK_RAW) nfc_llcp_sock_unlink(&local->raw_sockets, sk); + else if (sk->sk_state == LLCP_CONNECTING) + nfc_llcp_sock_unlink(&local->connecting_sockets, sk); else nfc_llcp_sock_unlink(&local->sockets, sk); -- cgit v1.2.3 From b493ea2765cc17cb8aa7e7544a4b6dcb05b6ed77 Mon Sep 17 00:00:00 2001 From: Lee Jones Date: Wed, 29 Apr 2026 13:40:42 +0000 Subject: nfc: llcp: Fix use-after-free race in nfc_llcp_recv_cc() A race condition exists in the NFC LLCP connection state machine where the connection acceptance packet (CC) can be processed concurrently with socket release. This can lead to a use-after-free of the socket object. When nfc_llcp_recv_cc() moves the socket from the connecting_sockets list to the sockets list, it does so without holding the socket lock. If llcp_sock_release() is executing concurrently, it might have already unlinked the socket and dropped its references, which can result in nfc_llcp_recv_cc() linking a freed socket into the live list. Fix this by holding lock_sock() during the state transition and list movement in nfc_llcp_recv_cc(). After acquiring the lock, check if the socket is still hashed to ensure it hasn't already been unlinked and marked for destruction by the release path. This aligns the locking pattern with recv_hdlc() and recv_disc(). Fixes: a69f32af86e3 ("NFC: Socket linked list") Signed-off-by: Lee Jones Reviewed-by: Simon Horman Link: https://patch.msgid.link/20260429134115.3558604-2-lee@kernel.org Signed-off-by: David Heidelberg --- net/nfc/llcp_core.c | 11 +++++++++++ 1 file changed, 11 insertions(+) (limited to 'net') diff --git a/net/nfc/llcp_core.c b/net/nfc/llcp_core.c index db5bc6a878dd..dc65c719f35f 100644 --- a/net/nfc/llcp_core.c +++ b/net/nfc/llcp_core.c @@ -1218,6 +1218,15 @@ static void nfc_llcp_recv_cc(struct nfc_llcp_local *local, sk = &llcp_sock->sk; + lock_sock(sk); + + /* Check if socket was destroyed whilst waiting for the lock */ + if (!sk_hashed(sk)) { + release_sock(sk); + nfc_llcp_sock_put(llcp_sock); + return; + } + /* Unlink from connecting and link to the client array */ nfc_llcp_sock_unlink(&local->connecting_sockets, sk); nfc_llcp_sock_link(&local->sockets, sk); @@ -1229,6 +1238,8 @@ static void nfc_llcp_recv_cc(struct nfc_llcp_local *local, sk->sk_state = LLCP_CONNECTED; sk->sk_state_change(sk); + release_sock(sk); + nfc_llcp_sock_put(llcp_sock); } -- cgit v1.2.3 From 83861c48ba122f85cc8384780764b3a791341678 Mon Sep 17 00:00:00 2001 From: Ilya Maximets Date: Thu, 30 Apr 2026 23:32:50 +0200 Subject: openvswitch: vport: fix race between tunnel creation and linking When a tunnel vport is created it first creates the tunnel device, e.g., with geneve_dev_create_fb(), then it calls ovs_netdev_link() to take a reference and link it to the device that represents openvswitch datapath. The creation of the device is happening under RTNL, but then RTNL is released and re-acquired to find the device by name. It is technically possible for the tunnel device to be re-named or deleted within that window while RTNL is not held, and some other device created in its place. This will cause a non-tunnel device to be referenced in the vport and tunnel-specific functions used on it, e.g. vxlan_get_options() that directly casts the private netdev data into a struct vxlan_dev causing an invalid memory access: BUG: KASAN: slab-use-after-free in vxlan_get_options+0x323/0x3a0 vxlan_get_options+0x323/0x3a0 ovs_vport_cmd_new+0x6e3/0xd30 Fix that by taking a reference to the just created device before releasing RTNL. This ensures that the device in the vport is always the one that was just created. The search by name is only needed for a standard vport-netdev that links pre-existing devices, so that functionality and device type checks are moved to netdev_create(). It is also awkward that ovs_netdev_link() takes ownership of the vport and destroys it on failure. It doesn't know the type of the port it is dealing with, so we need to pass down the indicator that it's a tunnel, so the link can be properly deleted on failure. It's possible to refactor the logic to make the ovs_netdev_link() do only the linking part and let the callers perform a proper destruction, but it will be much more code for each legacy tunnel port type, so it is not worth it for the bug fix. Fixes: 614732eaa12d ("openvswitch: Use regular VXLAN net_device device") Reported-by: Yuan Tan Reported-by: Yifan Wu Reported-by: Juefei Pu Reported-by: Xin Liu Reported-by: Yang Yang Signed-off-by: Ilya Maximets Acked-by: Eelco Chaudron Link: https://patch.msgid.link/20260430213349.407991-1-i.maximets@ovn.org Signed-off-by: Paolo Abeni --- net/openvswitch/vport-geneve.c | 5 +++- net/openvswitch/vport-gre.c | 5 +++- net/openvswitch/vport-netdev.c | 58 +++++++++++++++++++++++++----------------- net/openvswitch/vport-netdev.h | 2 +- net/openvswitch/vport-vxlan.c | 5 +++- 5 files changed, 48 insertions(+), 27 deletions(-) (limited to 'net') diff --git a/net/openvswitch/vport-geneve.c b/net/openvswitch/vport-geneve.c index b10e1602c6b1..cb5ea4424ffc 100644 --- a/net/openvswitch/vport-geneve.c +++ b/net/openvswitch/vport-geneve.c @@ -97,6 +97,9 @@ static struct vport *geneve_tnl_create(const struct vport_parms *parms) goto error; } + vport->dev = dev; + netdev_hold(vport->dev, &vport->dev_tracker, GFP_KERNEL); + rtnl_unlock(); return vport; error: @@ -111,7 +114,7 @@ static struct vport *geneve_create(const struct vport_parms *parms) if (IS_ERR(vport)) return vport; - return ovs_netdev_link(vport, parms->name); + return ovs_netdev_link(vport, true); } static struct vport_ops ovs_geneve_vport_ops = { diff --git a/net/openvswitch/vport-gre.c b/net/openvswitch/vport-gre.c index 4014c9b5eb79..6cb5a697b396 100644 --- a/net/openvswitch/vport-gre.c +++ b/net/openvswitch/vport-gre.c @@ -63,6 +63,9 @@ static struct vport *gre_tnl_create(const struct vport_parms *parms) return ERR_PTR(err); } + vport->dev = dev; + netdev_hold(vport->dev, &vport->dev_tracker, GFP_KERNEL); + rtnl_unlock(); return vport; } @@ -75,7 +78,7 @@ static struct vport *gre_create(const struct vport_parms *parms) if (IS_ERR(vport)) return vport; - return ovs_netdev_link(vport, parms->name); + return ovs_netdev_link(vport, true); } static struct vport_ops ovs_gre_vport_ops = { diff --git a/net/openvswitch/vport-netdev.c b/net/openvswitch/vport-netdev.c index 12055af832dc..a92ca8b37f96 100644 --- a/net/openvswitch/vport-netdev.c +++ b/net/openvswitch/vport-netdev.c @@ -73,37 +73,21 @@ static struct net_device *get_dpdev(const struct datapath *dp) return local->dev; } -struct vport *ovs_netdev_link(struct vport *vport, const char *name) +struct vport *ovs_netdev_link(struct vport *vport, bool tunnel) { int err; - vport->dev = dev_get_by_name(ovs_dp_get_net(vport->dp), name); - if (!vport->dev) { + if (WARN_ON_ONCE(!vport->dev)) { err = -ENODEV; goto error_free_vport; } - /* Ensure that the device exists and that the provided - * name is not one of its aliases. - */ - if (strcmp(name, ovs_vport_name(vport))) { - err = -ENODEV; - goto error_put; - } - netdev_tracker_alloc(vport->dev, &vport->dev_tracker, GFP_KERNEL); - if (vport->dev->flags & IFF_LOOPBACK || - (vport->dev->type != ARPHRD_ETHER && - vport->dev->type != ARPHRD_NONE) || - ovs_is_internal_dev(vport->dev)) { - err = -EINVAL; - goto error_put; - } rtnl_lock(); err = netdev_master_upper_dev_link(vport->dev, get_dpdev(vport->dp), NULL, NULL, NULL); if (err) - goto error_unlock; + goto error_put_unlock; err = netdev_rx_handler_register(vport->dev, netdev_frame_hook, vport); @@ -119,10 +103,11 @@ struct vport *ovs_netdev_link(struct vport *vport, const char *name) error_master_upper_dev_unlink: netdev_upper_dev_unlink(vport->dev, get_dpdev(vport->dp)); -error_unlock: - rtnl_unlock(); -error_put: +error_put_unlock: + if (tunnel && vport->dev->reg_state == NETREG_REGISTERED) + rtnl_delete_link(vport->dev, 0, NULL); netdev_put(vport->dev, &vport->dev_tracker); + rtnl_unlock(); error_free_vport: ovs_vport_free(vport); return ERR_PTR(err); @@ -132,12 +117,39 @@ EXPORT_SYMBOL_GPL(ovs_netdev_link); static struct vport *netdev_create(const struct vport_parms *parms) { struct vport *vport; + int err; vport = ovs_vport_alloc(0, &ovs_netdev_vport_ops, parms); if (IS_ERR(vport)) return vport; - return ovs_netdev_link(vport, parms->name); + vport->dev = dev_get_by_name(ovs_dp_get_net(vport->dp), parms->name); + if (!vport->dev) { + err = -ENODEV; + goto error_free_vport; + } + netdev_tracker_alloc(vport->dev, &vport->dev_tracker, GFP_KERNEL); + + /* Ensure that the provided name is not an alias. */ + if (strcmp(parms->name, ovs_vport_name(vport))) { + err = -ENODEV; + goto error_put; + } + + if (vport->dev->flags & IFF_LOOPBACK || + (vport->dev->type != ARPHRD_ETHER && + vport->dev->type != ARPHRD_NONE) || + ovs_is_internal_dev(vport->dev)) { + err = -EINVAL; + goto error_put; + } + + return ovs_netdev_link(vport, false); +error_put: + netdev_put(vport->dev, &vport->dev_tracker); +error_free_vport: + ovs_vport_free(vport); + return ERR_PTR(err); } static void vport_netdev_free(struct rcu_head *rcu) diff --git a/net/openvswitch/vport-netdev.h b/net/openvswitch/vport-netdev.h index c5d83a43bfc4..6c0d7366f986 100644 --- a/net/openvswitch/vport-netdev.h +++ b/net/openvswitch/vport-netdev.h @@ -13,7 +13,7 @@ struct vport *ovs_netdev_get_vport(struct net_device *dev); -struct vport *ovs_netdev_link(struct vport *vport, const char *name); +struct vport *ovs_netdev_link(struct vport *vport, bool tunnel); void ovs_netdev_detach_dev(struct vport *); int __init ovs_netdev_init(void); diff --git a/net/openvswitch/vport-vxlan.c b/net/openvswitch/vport-vxlan.c index 0b881b043bcf..c1b37b50d29e 100644 --- a/net/openvswitch/vport-vxlan.c +++ b/net/openvswitch/vport-vxlan.c @@ -126,6 +126,9 @@ static struct vport *vxlan_tnl_create(const struct vport_parms *parms) goto error; } + vport->dev = dev; + netdev_hold(vport->dev, &vport->dev_tracker, GFP_KERNEL); + rtnl_unlock(); return vport; error: @@ -140,7 +143,7 @@ static struct vport *vxlan_create(const struct vport_parms *parms) if (IS_ERR(vport)) return vport; - return ovs_netdev_link(vport, parms->name); + return ovs_netdev_link(vport, true); } static struct vport_ops ovs_vxlan_netdev_vport_ops = { -- cgit v1.2.3 From aa69918bd418e700309fdd08509dba324fb24296 Mon Sep 17 00:00:00 2001 From: Ilya Maximets Date: Fri, 1 May 2026 01:38:37 +0200 Subject: openvswitch: vport: fix self-deadlock on release of tunnel ports vports are used concurrently and protected by RCU, so netdev_put() must happen after the RCU grace period. So, either in an RCU call or after the synchronize_net(). The rtnl_delete_link() must happen under RTNL and so can't be executed in RCU context. Calling synchronize_net() while holding RTNL is not a good idea for performance and system stability under load in general, so calling netdev_put() in RCU call is the right solution here. However, when the device is deleted, rtnl_unlock() will call netdev_run_todo() and block until all the references are gone. In the current code this means that we never reach the call_rcu() and the vport is never freed and the reference is never released, causing a self-deadlock on device removal. Fix that by moving the rcu_call() before the rtnl_unlock(), so the scheduled RCU callback will be executed when synchronize_net() is called from the rtnl_unlock()->netdev_run_todo() while the RTNL itself is already released. Fixes: 6931d21f87bc ("openvswitch: defer tunnel netdev_put to RCU release") Cc: stable@vger.kernel.org Acked-by: Eelco Chaudron Signed-off-by: Ilya Maximets Acked-by: Aaron Conole Link: https://patch.msgid.link/20260430233848.440994-2-i.maximets@ovn.org Signed-off-by: Paolo Abeni --- net/openvswitch/vport-netdev.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/openvswitch/vport-netdev.c b/net/openvswitch/vport-netdev.c index a92ca8b37f96..c42642075685 100644 --- a/net/openvswitch/vport-netdev.c +++ b/net/openvswitch/vport-netdev.c @@ -208,9 +208,13 @@ void ovs_netdev_tunnel_destroy(struct vport *vport) */ if (vport->dev->reg_state == NETREG_REGISTERED) rtnl_delete_link(vport->dev, 0, NULL); - rtnl_unlock(); + /* We can't put the device reference yet, since it can still be in + * use, but rtnl_unlock()->netdev_run_todo() will block until all + * the references are released, so the RCU call must be before it. + */ call_rcu(&vport->rcu, vport_netdev_free); + rtnl_unlock(); } EXPORT_SYMBOL_GPL(ovs_netdev_tunnel_destroy); -- cgit v1.2.3 From 44b550d88b267320459d518c0743a241ab2108fa Mon Sep 17 00:00:00 2001 From: Nan Li Date: Fri, 1 May 2026 09:08:44 +0800 Subject: net/rds: handle zerocopy send cleanup before the message is queued A zerocopy send can fail after user pages have been pinned but before the message is attached to the sending socket. The purge path currently infers zerocopy state from rm->m_rs, so an unqueued message can be cleaned up as if it owned normal payload pages. However, zerocopy ownership is really determined by the presence of op_mmp_znotifier, regardless of whether the message has reached the socket queue. Capture op_mmp_znotifier up front in rds_message_purge() and use it as the cleanup discriminator. If the message is already associated with a socket, keep the existing completion path. Otherwise, drop the pinned page accounting directly and release the notifier before putting the payload pages. This keeps early send failure cleanup consistent with the zerocopy lifetime rules without changing the normal queued completion path. Fixes: 0cebaccef3ac ("rds: zerocopy Tx support.") Cc: stable@kernel.org Reported-by: Yuan Tan Reported-by: Yifan Wu Reported-by: Juefei Pu Reported-by: Xin Liu Co-developed-by: Xiao Liu Signed-off-by: Xiao Liu Signed-off-by: Nan Li Signed-off-by: Ren Wei Reviewed-by: Allison Henderson Link: https://patch.msgid.link/d2ea98a6313d5467bac00f7c9fef8c7acddb9258.1777550074.git.tonanli66@gmail.com Signed-off-by: Paolo Abeni --- net/rds/message.c | 20 +++++++++++++++----- 1 file changed, 15 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/rds/message.c b/net/rds/message.c index eaa6f22601a4..25fedcb3cd00 100644 --- a/net/rds/message.c +++ b/net/rds/message.c @@ -131,24 +131,34 @@ static void rds_rm_zerocopy_callback(struct rds_sock *rs, */ static void rds_message_purge(struct rds_message *rm) { + struct rds_znotifier *znotifier; unsigned long i, flags; - bool zcopy = false; + bool zcopy; if (unlikely(test_bit(RDS_MSG_PAGEVEC, &rm->m_flags))) return; spin_lock_irqsave(&rm->m_rs_lock, flags); + znotifier = rm->data.op_mmp_znotifier; + rm->data.op_mmp_znotifier = NULL; + zcopy = !!znotifier; + if (rm->m_rs) { struct rds_sock *rs = rm->m_rs; - if (rm->data.op_mmp_znotifier) { - zcopy = true; - rds_rm_zerocopy_callback(rs, rm->data.op_mmp_znotifier); + if (znotifier) { + rds_rm_zerocopy_callback(rs, znotifier); rds_wake_sk_sleep(rs); - rm->data.op_mmp_znotifier = NULL; } sock_put(rds_rs_to_sk(rs)); rm->m_rs = NULL; + } else if (znotifier) { + /* + * Zerocopy can fail before the message is queued on the + * socket, so there is no rs to carry the notification. + */ + mm_unaccount_pinned_pages(&znotifier->z_mmp); + kfree(rds_info_from_znotifier(znotifier)); } spin_unlock_irqrestore(&rm->m_rs_lock, flags); -- cgit v1.2.3 From ac8eb3e18f41e2cc8492cc1d358bcb786c850270 Mon Sep 17 00:00:00 2001 From: Benjamin Berg Date: Tue, 5 May 2026 15:15:40 +0200 Subject: wifi: mac80211: use safe list iteration in radar detect work The call to ieee80211_dfs_cac_cancel can cause the iterated chanctx to be freed and removed from the list. Guard against this to avoid a slab-use-after-free error. Cc: stable@vger.kernel.org Fixes: bca8bc0399ac ("wifi: mac80211: handle ieee80211_radar_detected() for MLO") Signed-off-by: Benjamin Berg Link: https://patch.msgid.link/20260505151539.236d63a1b736.I35dbb9e96a2d4a480be208770fdd99ba3b817b79@changeid Signed-off-by: Johannes Berg --- net/mac80211/util.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/mac80211/util.c b/net/mac80211/util.c index b093bc203c81..2529b01e2cd5 100644 --- a/net/mac80211/util.c +++ b/net/mac80211/util.c @@ -3700,11 +3700,11 @@ void ieee80211_dfs_radar_detected_work(struct wiphy *wiphy, struct ieee80211_local *local = container_of(work, struct ieee80211_local, radar_detected_work); struct cfg80211_chan_def chandef; - struct ieee80211_chanctx *ctx; + struct ieee80211_chanctx *ctx, *tmp; lockdep_assert_wiphy(local->hw.wiphy); - list_for_each_entry(ctx, &local->chanctx_list, list) { + list_for_each_entry_safe(ctx, tmp, &local->chanctx_list, list) { if (ctx->replace_state == IEEE80211_CHANCTX_REPLACES_OTHER) continue; -- cgit v1.2.3 From 901a7d9e2f280a9e76e6c58406a519cb11ad5ff8 Mon Sep 17 00:00:00 2001 From: Alyssa Ross Date: Sun, 3 May 2026 21:25:16 +0200 Subject: ipv6: default IPV6_SIT to m This basically defaulted to m until recently, since IPV6 defaulted to m. Since IPV6 was changed to a boolean with a default of y, IPV6_SIT started defaulting to built-in as well. This results in a surprise sit0 device by default for defconfig (and defconfig-derived config) users at boot. For me, this broke an (admittedly non-robust) script. Preserve the behaviour of most configs by avoiding building this module, that's probably overall seldom used compared to IPv6 as a whole, into the kernel. Fixes: 309b905deee59 ("ipv6: convert CONFIG_IPV6 to built-in only and clean up Kconfigs") Signed-off-by: Alyssa Ross Reviewed-by: Fernando Fernandez Mancera Link: https://patch.msgid.link/20260503192515.290900-2-hi@alyssa.is Signed-off-by: Jakub Kicinski --- net/ipv6/Kconfig | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/ipv6/Kconfig b/net/ipv6/Kconfig index c024aa77f25b..c3806c6ac96f 100644 --- a/net/ipv6/Kconfig +++ b/net/ipv6/Kconfig @@ -164,7 +164,7 @@ config IPV6_SIT select INET_TUNNEL select NET_IP_TUNNEL select IPV6_NDISC_NODETYPE - default y + default m help Tunneling means encapsulating data of one protocol type within another protocol and sending it over a channel that understands the @@ -172,7 +172,7 @@ config IPV6_SIT into IPv4 packets. This is useful if you want to connect two IPv6 networks over an IPv4-only path. - Saying M here will produce a module called sit. If unsure, say Y. + Saying M here will produce a module called sit. If unsure, say M. config IPV6_SIT_6RD bool "IPv6: IPv6 Rapid Deployment (6RD)" -- cgit v1.2.3 From 5ad509c1fdad4bf0993b72d1b3d462f036d8a0d8 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Mon, 4 May 2026 06:43:13 +0000 Subject: ipv6: Fix null-ptr-deref in fib6_mtu(). syzbot reported null-ptr-deref in fib6_mtu(). [0] When res->f6i->fib6_pmtu is 0 in fib6_mtu(), it fetches MTU from __in6_dev_get(nh->fib_nh_dev)->cnf.mtu6. However, __in6_dev_get() could return NULL when the device is being unregistered. Let's return 0 MTU if __in6_dev_get() returns NULL in fib6_mtu(). [0]: Oops: general protection fault, probably for non-canonical address 0xdffffc00000000bc: 0000 [#1] SMP KASAN NOPTI KASAN: null-ptr-deref in range [0x00000000000005e0-0x00000000000005e7] CPU: 0 UID: 0 PID: 7890 Comm: syz.2.502 Tainted: G L syzkaller #0 PREEMPT(full) Tainted: [L]=SOFTLOCKUP Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.3-debian-1.16.3-2 04/01/2014 RIP: 0010:fib6_mtu net/ipv6/route.c:1648 [inline] RIP: 0010:rt6_insert_exception+0x9eb/0x10a0 net/ipv6/route.c:1753 Code: 3b 14 cf f7 45 85 f6 0f 85 1d 02 00 00 e8 7d 19 cf f7 48 8d bb e0 05 00 00 48 b8 00 00 00 00 00 fc ff df 48 89 fa 48 c1 ea 03 <0f> b6 14 02 48 89 f8 83 e0 07 83 c0 03 38 d0 7c 08 84 d2 0f 85 89 RSP: 0000:ffffc9000610f120 EFLAGS: 00010202 RAX: dffffc0000000000 RBX: 0000000000000000 RCX: ffffc9000c001000 RDX: 00000000000000bc RSI: ffffffff8a38bc83 RDI: 00000000000005e0 RBP: ffff888052f06000 R08: 0000000000000005 R09: 0000000000000000 R10: 0000000000000001 R11: 0000000000000000 R12: ffff888042d16c00 R13: ffff888042d16cc8 R14: 0000000000000001 R15: 0000000000000500 FS: 0000000000000000(0000) GS:ffff88809717d000(0063) knlGS:00000000f540db40 CS: 0010 DS: 002b ES: 002b CR0: 0000000080050033 CR2: 00000000f73c6d50 CR3: 000000006eff0000 CR4: 0000000000352ef0 Call Trace: __ip6_rt_update_pmtu+0x555/0xd60 net/ipv6/route.c:2982 ip6_update_pmtu+0x34f/0x3b0 net/ipv6/route.c:3014 icmpv6_err+0x2a2/0x3f0 net/ipv6/icmp.c:82 icmpv6_notify+0x35e/0x820 net/ipv6/icmp.c:1087 icmpv6_rcv+0x10bf/0x1ae0 net/ipv6/icmp.c:1228 ip6_protocol_deliver_rcu+0xf97/0x1500 net/ipv6/ip6_input.c:478 ip6_input_finish+0x1e4/0x4a0 net/ipv6/ip6_input.c:529 NF_HOOK include/linux/netfilter.h:318 [inline] NF_HOOK include/linux/netfilter.h:312 [inline] ip6_input+0x105/0x2f0 net/ipv6/ip6_input.c:540 ip6_mc_input+0x513/0xf50 net/ipv6/ip6_input.c:630 dst_input include/net/dst.h:480 [inline] ip6_rcv_finish net/ipv6/ip6_input.c:119 [inline] NF_HOOK include/linux/netfilter.h:318 [inline] NF_HOOK include/linux/netfilter.h:312 [inline] ipv6_rcv+0x34c/0x3d0 net/ipv6/ip6_input.c:351 __netif_receive_skb_one_core+0x12d/0x1e0 net/core/dev.c:6202 __netif_receive_skb+0x1f/0x120 net/core/dev.c:6315 netif_receive_skb_internal net/core/dev.c:6401 [inline] netif_receive_skb+0x13b/0x7f0 net/core/dev.c:6460 tun_rx_batched.isra.0+0x3f6/0x750 drivers/net/tun.c:1511 tun_get_user+0x1e31/0x3c20 drivers/net/tun.c:1955 tun_chr_write_iter+0xdc/0x200 drivers/net/tun.c:2001 new_sync_write fs/read_write.c:595 [inline] vfs_write+0x6ac/0x1070 fs/read_write.c:688 ksys_write+0x12a/0x250 fs/read_write.c:740 do_syscall_32_irqs_on arch/x86/entry/syscall_32.c:83 [inline] do_int80_emulation+0x141/0x700 arch/x86/entry/syscall_32.c:172 asm_int80_emulation+0x1a/0x20 arch/x86/include/asm/idtentry.h:621 RIP: 0023:0xf715616b Code: 57 56 53 8b 44 24 14 f6 00 08 75 23 8b 44 24 18 8b 5c 24 1c 8b 4c 24 20 8b 54 24 24 8b 74 24 28 8b 7c 24 2c 8b 6c 24 30 cd 80 <5b> 5e 5f 5d c3 5b 5e 5f 5d e9 f7 a1 ff ff 66 90 66 90 66 90 90 53 RSP: 002b:00000000f540d44c EFLAGS: 00000246 ORIG_RAX: 0000000000000004 RAX: ffffffffffffffda RBX: 00000000000000c8 RCX: 0000000080000640 RDX: 000000000000007a RSI: 0000000000000000 RDI: 0000000000000000 RBP: 0000000000000000 R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000000 R11: 0000000000000292 R12: 0000000000000000 R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000000 Fixes: dcd1f572954f ("net/ipv6: Remove fib6_idev") Reported-by: syzbot+01f005f9c6387ca6f6dd@syzkaller.appspotmail.com Closes: https://lore.kernel.org/netdev/69f83f22.170a0220.13cc2.0004.GAE@google.com/ Signed-off-by: Kuniyuki Iwashima Reviewed-by: Ido Schimmel Link: https://patch.msgid.link/20260504064316.3820775-1-kuniyu@google.com Signed-off-by: Jakub Kicinski --- net/ipv6/route.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'net') diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 0dc0316530ca..e3d355d1fbd6 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -1645,6 +1645,10 @@ static unsigned int fib6_mtu(const struct fib6_result *res) rcu_read_lock(); idev = __in6_dev_get(dev); + if (!idev) { + rcu_read_unlock(); + return 0; + } mtu = READ_ONCE(idev->cnf.mtu6); rcu_read_unlock(); } -- cgit v1.2.3 From f83e07b29246f468bc7c99f98ca1897843fa8167 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 4 May 2026 16:38:42 +0000 Subject: net/sched: sch_fq_codel: annotate data-races from fq_codel_dump_class_stats() fq_codel_dump_class_stats() acquires qdisc spinlock only when requested to follow flow->head chain. As we did in sch_cake recently, add the missing READ_ONCE()/WRITE_ONCE() annotations. Fixes: edb09eb17ed8 ("net: sched: do not acquire qdisc spinlock in qdisc/class stats dump") Signed-off-by: Eric Dumazet Reviewed-by: Jamal Hadi Salim Link: https://patch.msgid.link/20260504163842.1162001-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/sched/sch_fq_codel.c | 39 ++++++++++++++++++++------------------- 1 file changed, 20 insertions(+), 19 deletions(-) (limited to 'net') diff --git a/net/sched/sch_fq_codel.c b/net/sched/sch_fq_codel.c index 0664b2f2d6f2..24db54684e8a 100644 --- a/net/sched/sch_fq_codel.c +++ b/net/sched/sch_fq_codel.c @@ -117,7 +117,7 @@ static inline struct sk_buff *dequeue_head(struct fq_codel_flow *flow) { struct sk_buff *skb = flow->head; - flow->head = skb->next; + WRITE_ONCE(flow->head, skb->next); skb_mark_not_on_list(skb); return skb; } @@ -127,7 +127,7 @@ static inline void flow_queue_add(struct fq_codel_flow *flow, struct sk_buff *skb) { if (flow->head == NULL) - flow->head = skb; + WRITE_ONCE(flow->head, skb); else flow->tail->next = skb; flow->tail = skb; @@ -173,8 +173,8 @@ static unsigned int fq_codel_drop(struct Qdisc *sch, unsigned int max_packets, } while (++i < max_packets && len < threshold); /* Tell codel to increase its signal strength also */ - flow->cvars.count += i; - q->backlogs[idx] -= len; + WRITE_ONCE(flow->cvars.count, flow->cvars.count + i); + WRITE_ONCE(q->backlogs[idx], q->backlogs[idx] - len); q->memory_usage -= mem; sch->qstats.drops += i; sch->qstats.backlog -= len; @@ -204,13 +204,13 @@ static int fq_codel_enqueue(struct sk_buff *skb, struct Qdisc *sch, codel_set_enqueue_time(skb); flow = &q->flows[idx]; flow_queue_add(flow, skb); - q->backlogs[idx] += qdisc_pkt_len(skb); + WRITE_ONCE(q->backlogs[idx], q->backlogs[idx] + qdisc_pkt_len(skb)); qdisc_qstats_backlog_inc(sch, skb); if (list_empty(&flow->flowchain)) { list_add_tail(&flow->flowchain, &q->new_flows); q->new_flow_count++; - flow->deficit = q->quantum; + WRITE_ONCE(flow->deficit, q->quantum); } get_codel_cb(skb)->mem_usage = skb->truesize; q->memory_usage += get_codel_cb(skb)->mem_usage; @@ -263,7 +263,8 @@ static struct sk_buff *dequeue_func(struct codel_vars *vars, void *ctx) flow = container_of(vars, struct fq_codel_flow, cvars); if (flow->head) { skb = dequeue_head(flow); - q->backlogs[flow - q->flows] -= qdisc_pkt_len(skb); + WRITE_ONCE(q->backlogs[flow - q->flows], + q->backlogs[flow - q->flows] - qdisc_pkt_len(skb)); q->memory_usage -= get_codel_cb(skb)->mem_usage; sch->q.qlen--; sch->qstats.backlog -= qdisc_pkt_len(skb); @@ -296,7 +297,7 @@ begin: flow = list_first_entry(head, struct fq_codel_flow, flowchain); if (flow->deficit <= 0) { - flow->deficit += q->quantum; + WRITE_ONCE(flow->deficit, flow->deficit + q->quantum); list_move_tail(&flow->flowchain, &q->old_flows); goto begin; } @@ -314,7 +315,7 @@ begin: goto begin; } qdisc_bstats_update(sch, skb); - flow->deficit -= qdisc_pkt_len(skb); + WRITE_ONCE(flow->deficit, flow->deficit - qdisc_pkt_len(skb)); if (q->cstats.drop_count) { qdisc_tree_reduce_backlog(sch, q->cstats.drop_count, @@ -328,7 +329,7 @@ begin: static void fq_codel_flow_purge(struct fq_codel_flow *flow) { rtnl_kfree_skbs(flow->head, flow->tail); - flow->head = NULL; + WRITE_ONCE(flow->head, NULL); } static void fq_codel_reset(struct Qdisc *sch) @@ -656,21 +657,21 @@ static int fq_codel_dump_class_stats(struct Qdisc *sch, unsigned long cl, memset(&xstats, 0, sizeof(xstats)); xstats.type = TCA_FQ_CODEL_XSTATS_CLASS; - xstats.class_stats.deficit = flow->deficit; + xstats.class_stats.deficit = READ_ONCE(flow->deficit); xstats.class_stats.ldelay = - codel_time_to_us(flow->cvars.ldelay); - xstats.class_stats.count = flow->cvars.count; - xstats.class_stats.lastcount = flow->cvars.lastcount; - xstats.class_stats.dropping = flow->cvars.dropping; - if (flow->cvars.dropping) { - codel_tdiff_t delta = flow->cvars.drop_next - + codel_time_to_us(READ_ONCE(flow->cvars.ldelay)); + xstats.class_stats.count = READ_ONCE(flow->cvars.count); + xstats.class_stats.lastcount = READ_ONCE(flow->cvars.lastcount); + xstats.class_stats.dropping = READ_ONCE(flow->cvars.dropping); + if (xstats.class_stats.dropping) { + codel_tdiff_t delta = READ_ONCE(flow->cvars.drop_next) - codel_get_time(); xstats.class_stats.drop_next = (delta >= 0) ? codel_time_to_us(delta) : -codel_time_to_us(-delta); } - if (flow->head) { + if (READ_ONCE(flow->head)) { sch_tree_lock(sch); skb = flow->head; while (skb) { @@ -679,7 +680,7 @@ static int fq_codel_dump_class_stats(struct Qdisc *sch, unsigned long cl, } sch_tree_unlock(sch); } - qs.backlog = q->backlogs[idx]; + qs.backlog = READ_ONCE(q->backlogs[idx]); qs.drops = 0; } if (gnet_stats_copy_queue(d, NULL, &qs, qs.qlen) < 0) -- cgit v1.2.3 From d73a9a63f9f7f7c17637731fd28daf3665992d1e Mon Sep 17 00:00:00 2001 From: Jason Xing Date: Sat, 2 May 2026 23:07:15 +0300 Subject: xsk: reject sw-csum UMEM binding to IFF_TX_SKB_NO_LINEAR devices skb_checksum_help() is a common helper that writes the folded 16-bit checksum back via skb->data + csum_start + csum_offset, i.e. it relies on the skb's linear head and fails (with WARN_ONCE and -EINVAL) when skb_headlen() is 0. AF_XDP generic xmit takes two very different paths depending on the netdev. Drivers that advertise IFF_TX_SKB_NO_LINEAR (e.g. virtio_net) skip the "copy payload into a linear head" step on purpose as a performance optimisation: xsk_build_skb_zerocopy() only attaches UMEM pages as frags and never calls skb_put(), so skb_headlen() stays 0 for the whole skb. For these skbs there is simply no linear area for skb_checksum_help() to write the csum into - the sw-csum fallback is structurally inapplicable. The patch tries to catch this and reject the combination with error at setup time. Rejecting at bind() converts this silent per-packet failure into a synchronous, actionable -EOPNOTSUPP at setup time. HW csum and launch_time metadata on IFF_TX_SKB_NO_LINEAR drivers are unaffected because they do not call skb_checksum_help(). Without the patch, every descriptor carrying 'XDP_TX_METADATA | XDP_TXMD_FLAGS_CHECKSUM' produces: 1) a WARN_ONCE "offset (N) >= skb_headlen() (0)" from skb_checksum_help(), 2) sendmsg() returning -EINVAL without consuming the descriptor (invalid_descs is not incremented), 3) a wedged TX ring: __xsk_generic_xmit() does not advance the consumer on non-EOVERFLOW errors, so the next sendmsg() re-reads the same descriptor and re-hits the same WARN until the socket is closed. Closes: https://lore.kernel.org/all/20260419045822.843BFC2BCAF@smtp.kernel.org/#t Acked-by: Stanislav Fomichev Signed-off-by: Jason Xing Acked-by: Stanislav Fomichev Signed-off-by: Jason Xing Reviewed-by: Alexander Lobakin Fixes: 30c3055f9c0d ("xsk: wrap generic metadata handling onto separate function") Link: https://patch.msgid.link/20260502200722.53960-2-kerneljasonxing@gmail.com Signed-off-by: Jakub Kicinski --- net/xdp/xsk_buff_pool.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'net') diff --git a/net/xdp/xsk_buff_pool.c b/net/xdp/xsk_buff_pool.c index cd7bc50872f6..d981cfdd8535 100644 --- a/net/xdp/xsk_buff_pool.c +++ b/net/xdp/xsk_buff_pool.c @@ -175,6 +175,9 @@ int xp_assign_dev(struct xsk_buff_pool *pool, if (force_zc && force_copy) return -EINVAL; + if (pool->tx_sw_csum && (netdev->priv_flags & IFF_TX_SKB_NO_LINEAR)) + return -EOPNOTSUPP; + if (xsk_get_pool_from_qid(netdev, queue_id)) return -EBUSY; -- cgit v1.2.3 From 0bb7a9caf5c1d6e25ba376ea6b39261ad28550f4 Mon Sep 17 00:00:00 2001 From: Jason Xing Date: Sat, 2 May 2026 23:07:16 +0300 Subject: xsk: free the skb when hitting the upper bound MAX_SKB_FRAGS Fix it by explicitly adding kfree_skb() before returning back to its caller. How to reproduce it in virtio_net: 1. the current skb is the first one (which means xs->skb is NULL) and hit the limit MAX_SKB_FRAGS. 2. xsk_build_skb_zerocopy() returns -EOVERFLOW. 3. the caller xsk_build_skb() clears skb by using 'skb = NULL;'. This is why bug can be triggered. 4. there is no chance to free this skb anymore. Note that if in this case the xs->skb is not NULL, xsk_build_skb() will call xsk_drop_skb(xs->skb) to do the right thing. Fixes: cf24f5a5feea ("xsk: add support for AF_XDP multi-buffer on Tx path") Acked-by: Stanislav Fomichev Signed-off-by: Jason Xing Reviewed-by: Alexander Lobakin Link: https://patch.msgid.link/20260502200722.53960-3-kerneljasonxing@gmail.com Signed-off-by: Jakub Kicinski --- net/xdp/xsk.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/xdp/xsk.c b/net/xdp/xsk.c index 887abed25466..d706b1e0bf60 100644 --- a/net/xdp/xsk.c +++ b/net/xdp/xsk.c @@ -856,8 +856,11 @@ static struct sk_buff *xsk_build_skb_zerocopy(struct xdp_sock *xs, addr = buffer - pool->addrs; for (copied = 0, i = skb_shinfo(skb)->nr_frags; copied < len; i++) { - if (unlikely(i >= MAX_SKB_FRAGS)) + if (unlikely(i >= MAX_SKB_FRAGS)) { + if (!xs->skb) + kfree_skb(skb); return ERR_PTR(-EOVERFLOW); + } page = pool->umem->pgs[addr >> PAGE_SHIFT]; get_page(page); -- cgit v1.2.3 From 8cd3c1c6e7d9a1f0954159ec5f2fdaa7f6a48bd8 Mon Sep 17 00:00:00 2001 From: Jason Xing Date: Sat, 2 May 2026 23:07:17 +0300 Subject: xsk: handle NULL dereference of the skb without frags issue When a first descriptor (xs->skb == NULL) triggers -EOVERFLOW in xsk_build_skb_zerocopy() (e.g., MAX_SKB_FRAGS exceeded), the free_err -EOVERFLOW handler unconditionally dereferences xs->skb via xsk_inc_num_desc(xs->skb) and xsk_drop_skb(xs->skb), causing a NULL pointer dereference. Fix this by guarding the existing xsk_inc_num_desc()/xsk_drop_skb() calls with an xs->skb check (for the continuation case), and add an else branch for the first-descriptor case that manually cancels the one reserved CQ slot and increments invalid_descs by one to account for the single invalid descriptor. Fixes: cf24f5a5feea ("xsk: add support for AF_XDP multi-buffer on Tx path") Acked-by: Stanislav Fomichev Signed-off-by: Jason Xing Reviewed-by: Alexander Lobakin Link: https://patch.msgid.link/20260502200722.53960-4-kerneljasonxing@gmail.com Signed-off-by: Jakub Kicinski --- net/xdp/xsk.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/xdp/xsk.c b/net/xdp/xsk.c index d706b1e0bf60..06ee260f3afc 100644 --- a/net/xdp/xsk.c +++ b/net/xdp/xsk.c @@ -976,9 +976,14 @@ free_err: kfree_skb(skb); if (err == -EOVERFLOW) { - /* Drop the packet */ - xsk_inc_num_desc(xs->skb); - xsk_drop_skb(xs->skb); + if (xs->skb) { + /* Drop the packet */ + xsk_inc_num_desc(xs->skb); + xsk_drop_skb(xs->skb); + } else { + xsk_cq_cancel_locked(xs->pool, 1); + xs->tx->invalid_descs++; + } xskq_cons_release(xs->tx); } else { /* Let application retry */ -- cgit v1.2.3 From 0f3776583d282550dbafe6082a914efcf9094d59 Mon Sep 17 00:00:00 2001 From: Jason Xing Date: Sat, 2 May 2026 23:07:18 +0300 Subject: xsk: fix use-after-free of xs->skb in xsk_build_skb() free_err path When xsk_build_skb() processes multi-buffer packets in copy mode, the first descriptor stores data into the skb linear area without adding any frags, so nr_frags stays at 0. The caller then sets xs->skb = skb to accumulate subsequent descriptors. If a continuation descriptor fails (e.g. alloc_page returns NULL with -EAGAIN), we jump to free_err where the condition: if (skb && !skb_shinfo(skb)->nr_frags) kfree_skb(skb); evaluates to true because nr_frags is still 0 (the first descriptor used the linear area, not frags). This frees the skb while xs->skb still points to it, creating a dangling pointer. On the next transmit attempt or socket close, xs->skb is dereferenced, causing a use-after-free or double-free. Fix by using a !xs->skb check to handle first frag situation, ensuring we only free skbs that were freshly allocated in this call (xs->skb is NULL) and never free an in-progress multi-buffer skb that the caller still references. Closes: https://lore.kernel.org/all/20260415082654.21026-4-kerneljasonxing@gmail.com/ Fixes: 6b9c129c2f93 ("xsk: remove @first_frag from xsk_build_skb()") Acked-by: Stanislav Fomichev Signed-off-by: Jason Xing Reviewed-by: Alexander Lobakin Link: https://patch.msgid.link/20260502200722.53960-5-kerneljasonxing@gmail.com Signed-off-by: Jakub Kicinski --- net/xdp/xsk.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/xdp/xsk.c b/net/xdp/xsk.c index 06ee260f3afc..55378c3855d5 100644 --- a/net/xdp/xsk.c +++ b/net/xdp/xsk.c @@ -972,7 +972,7 @@ static struct sk_buff *xsk_build_skb(struct xdp_sock *xs, return skb; free_err: - if (skb && !skb_shinfo(skb)->nr_frags) + if (skb && !xs->skb) kfree_skb(skb); if (err == -EOVERFLOW) { -- cgit v1.2.3 From 3dec153ae484e3b2ddac841156e197ba54c8df94 Mon Sep 17 00:00:00 2001 From: Jason Xing Date: Sat, 2 May 2026 23:07:19 +0300 Subject: xsk: prevent CQ desync when freeing half-built skbs in xsk_build_skb() Once xsk_skb_init_misc() has been called on an skb, its destructor is set to xsk_destruct_skb(), which submits the descriptor address(es) to the completion queue and advances the CQ producer. If such an skb is subsequently freed via kfree_skb() along an error path - before the skb has ever been handed to the driver - the destructor still runs and submits a bogus, half-initialized address to the CQ. Postpone the init phase when we believe the allocation of first frag is successfully completed. Before this init, skb can be safely freed by kfree_skb(). Closes: https://lore.kernel.org/all/20260419045822.843BFC2BCAF@smtp.kernel.org/ Fixes: c30d084960cf ("xsk: avoid overwriting skb fields for multi-buffer traffic") Acked-by: Stanislav Fomichev Signed-off-by: Jason Xing Reviewed-by: Alexander Lobakin Link: https://patch.msgid.link/20260502200722.53960-6-kerneljasonxing@gmail.com Signed-off-by: Jakub Kicinski --- net/xdp/xsk.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/xdp/xsk.c b/net/xdp/xsk.c index 55378c3855d5..af3c5752bb63 100644 --- a/net/xdp/xsk.c +++ b/net/xdp/xsk.c @@ -819,8 +819,6 @@ static struct sk_buff *xsk_build_skb_zerocopy(struct xdp_sock *xs, return ERR_PTR(err); skb_reserve(skb, hr); - - xsk_skb_init_misc(skb, xs, desc->addr); if (desc->options & XDP_TX_METADATA) { err = xsk_skb_metadata(skb, buffer, desc, pool, hr); if (unlikely(err)) @@ -917,7 +915,6 @@ static struct sk_buff *xsk_build_skb(struct xdp_sock *xs, if (unlikely(err)) goto free_err; - xsk_skb_init_misc(skb, xs, desc->addr); if (desc->options & XDP_TX_METADATA) { err = xsk_skb_metadata(skb, buffer, desc, xs->pool, hr); @@ -967,6 +964,8 @@ static struct sk_buff *xsk_build_skb(struct xdp_sock *xs, } } + if (!xs->skb) + xsk_skb_init_misc(skb, xs, desc->addr); xsk_inc_num_desc(skb); return skb; -- cgit v1.2.3 From 8c2cff50afdd2b53c7cc2ca2297301c0ffd3e802 Mon Sep 17 00:00:00 2001 From: Jason Xing Date: Sat, 2 May 2026 23:07:20 +0300 Subject: xsk: avoid skb leak in XDP_TX_METADATA case Fix it by explicitly adding kfree_skb() before returning back to its caller. How to reproduce it in virtio_net: 1. the current skb is the first one (which means no frag and xs->skb is NULL) and users enable metadata feature. 2. xsk_skb_metadata() returns a error code. 3. the caller xsk_build_skb() clears skb by using 'skb = NULL;'. 4. there is no chance to free this skb anymore. Closes: https://lore.kernel.org/all/20260415085204.3F87AC19424@smtp.kernel.org/ Fixes: 30c3055f9c0d ("xsk: wrap generic metadata handling onto separate function") Acked-by: Stanislav Fomichev Signed-off-by: Jason Xing Reviewed-by: Alexander Lobakin Link: https://patch.msgid.link/20260502200722.53960-7-kerneljasonxing@gmail.com Signed-off-by: Jakub Kicinski --- net/xdp/xsk.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/xdp/xsk.c b/net/xdp/xsk.c index af3c5752bb63..770ba4695a9d 100644 --- a/net/xdp/xsk.c +++ b/net/xdp/xsk.c @@ -821,8 +821,10 @@ static struct sk_buff *xsk_build_skb_zerocopy(struct xdp_sock *xs, skb_reserve(skb, hr); if (desc->options & XDP_TX_METADATA) { err = xsk_skb_metadata(skb, buffer, desc, pool, hr); - if (unlikely(err)) + if (unlikely(err)) { + kfree_skb(skb); return ERR_PTR(err); + } } } else { struct xsk_addrs *xsk_addr; -- cgit v1.2.3 From e0f229025a8e774a695017a376c4a01279c0e66e Mon Sep 17 00:00:00 2001 From: Jason Xing Date: Sat, 2 May 2026 23:07:21 +0300 Subject: xsk: fix xsk_addrs slab leak on multi-buffer error path When xsk_build_skb() / xsk_build_skb_zerocopy() sees the first continuation descriptor, it promotes destructor_arg from an inlined address to a freshly allocated xsk_addrs (num_descs = 1). The counter is bumped to >= 2 only at the very end of a successful build (by calling xsk_inc_num_desc()). If the build fails in between (e.g. alloc_page() returns NULL with -EAGAIN, or the MAX_SKB_FRAGS overflow hits), we jump to free_err, skip calling xsk_inc_num_desc() to increment num_descs and leave the half-built skb attached to xs->skb for the app to retry. The skb now has 1) destructor_arg = a real xsk_addrs pointer, 2) num_descs = 1 If the app never retries and just close()s the socket, xsk_release() calls xsk_drop_skb() -> xsk_consume_skb(), which decides whether to free xsk_addrs by testing num_descs > 1: if (unlikely(num_descs > 1)) kmem_cache_free(xsk_tx_generic_cache, destructor_arg); Because num_descs is exactly 1 the branch is skipped and the xsk_addrs object is leaked to the xsk_tx_generic_cache slab. Fix it by directly testing if destructor_arg is still addr. Or else it is modified and used to store the newly allocated memory from xsk_tx_generic_cache regardless of increment of num_desc, which we need to handle. Closes: https://lore.kernel.org/all/20260419045824.D9E5EC2BCAF@smtp.kernel.org/ Fixes: 0ebc27a4c67d ("xsk: avoid data corruption on cq descriptor number") Acked-by: Stanislav Fomichev Signed-off-by: Jason Xing Reviewed-by: Alexander Lobakin Link: https://patch.msgid.link/20260502200722.53960-8-kerneljasonxing@gmail.com Signed-off-by: Jakub Kicinski --- net/xdp/xsk.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/xdp/xsk.c b/net/xdp/xsk.c index 770ba4695a9d..079abd4bcb69 100644 --- a/net/xdp/xsk.c +++ b/net/xdp/xsk.c @@ -685,7 +685,7 @@ static void xsk_cq_submit_addr_locked(struct xsk_buff_pool *pool, spin_lock_irqsave(&pool->cq_prod_lock, flags); idx = xskq_get_prod(pool->cq); - if (unlikely(num_descs > 1)) { + if (unlikely(!xsk_skb_destructor_is_addr(skb))) { xsk_addr = (struct xsk_addrs *)skb_shinfo(skb)->destructor_arg; for (i = 0; i < num_descs; i++) { @@ -740,7 +740,7 @@ static void xsk_consume_skb(struct sk_buff *skb) u32 num_descs = xsk_get_num_desc(skb); struct xsk_addrs *xsk_addr; - if (unlikely(num_descs > 1)) { + if (unlikely(!xsk_skb_destructor_is_addr(skb))) { xsk_addr = (struct xsk_addrs *)skb_shinfo(skb)->destructor_arg; kmem_cache_free(xsk_tx_generic_cache, xsk_addr); } -- cgit v1.2.3 From 203cee647f551abc87b992045cd920b117ff990a Mon Sep 17 00:00:00 2001 From: Jason Xing Date: Sat, 2 May 2026 23:07:22 +0300 Subject: xsk: fix u64 descriptor address truncation on 32-bit architectures In copy mode TX, xsk_skb_destructor_set_addr() stores the 64-bit descriptor address into skb_shinfo(skb)->destructor_arg (void *) via a uintptr_t cast: skb_shinfo(skb)->destructor_arg = (void *)((uintptr_t)addr | 0x1UL); On 32-bit architectures uintptr_t is 32 bits, so the upper 32 bits of the descriptor address are silently dropped. In XDP_ZEROCOPY unaligned mode the chunk offset is encoded in bits 48-63 of the descriptor address (XSK_UNALIGNED_BUF_OFFSET_SHIFT = 48), meaning the offset is lost entirely. The completion queue then returns a truncated address to userspace, making buffer recycling impossible. Fix this by handling the 32-bit case directly in xsk_skb_destructor_set_addr(): when !CONFIG_64BIT, allocate an xsk_addrs struct (the same path already used for multi-descriptor SKBs) to store the full u64 address. The existing tagged-pointer logic in xsk_skb_destructor_is_addr() stays unchanged: slab pointers returned from kmem_cache_zalloc() are always word-aligned and therefore have bit 0 clear, which correctly identifies them as a struct pointer rather than an inline tagged address on every architecture. Factor the shared kmem_cache_zalloc + destructor_arg assignment into __xsk_addrs_alloc() and add a wrapper xsk_addrs_alloc() that handles the inline-to-list upgrade (is_addr check + get_addr + num_descs = 1). The three former open-coded kmem_cache_zalloc call sites now reduce to a single call each. Propagate the -ENOMEM from xsk_skb_destructor_set_addr() through xsk_skb_init_misc() so the caller can clean up the skb via kfree_skb() before skb->destructor is installed. The overhead is one extra kmem_cache_zalloc per first descriptor on 32-bit only; 64-bit builds are completely unchanged. Closes: https://lore.kernel.org/all/20260419045824.D9E5EC2BCAF@smtp.kernel.org/ Fixes: 0ebc27a4c67d ("xsk: avoid data corruption on cq descriptor number") Signed-off-by: Jason Xing Acked-by: Stanislav Fomichev Reviewed-by: Alexander Lobakin Link: https://patch.msgid.link/20260502200722.53960-9-kerneljasonxing@gmail.com Signed-off-by: Jakub Kicinski --- net/xdp/xsk.c | 88 +++++++++++++++++++++++++++++++++++++---------------------- 1 file changed, 56 insertions(+), 32 deletions(-) (limited to 'net') diff --git a/net/xdp/xsk.c b/net/xdp/xsk.c index 079abd4bcb69..5e5786cd9af5 100644 --- a/net/xdp/xsk.c +++ b/net/xdp/xsk.c @@ -646,9 +646,42 @@ static u64 xsk_skb_destructor_get_addr(struct sk_buff *skb) return (u64)((uintptr_t)skb_shinfo(skb)->destructor_arg & ~0x1UL); } -static void xsk_skb_destructor_set_addr(struct sk_buff *skb, u64 addr) +static struct xsk_addrs *__xsk_addrs_alloc(struct sk_buff *skb, u64 addr) { - skb_shinfo(skb)->destructor_arg = (void *)((uintptr_t)addr | 0x1UL); + struct xsk_addrs *xsk_addr; + + xsk_addr = kmem_cache_zalloc(xsk_tx_generic_cache, GFP_KERNEL); + if (unlikely(!xsk_addr)) + return NULL; + + xsk_addr->addrs[0] = addr; + skb_shinfo(skb)->destructor_arg = (void *)xsk_addr; + return xsk_addr; +} + +static struct xsk_addrs *xsk_addrs_alloc(struct sk_buff *skb) +{ + struct xsk_addrs *xsk_addr; + + if (!xsk_skb_destructor_is_addr(skb)) + return (struct xsk_addrs *)skb_shinfo(skb)->destructor_arg; + + xsk_addr = __xsk_addrs_alloc(skb, xsk_skb_destructor_get_addr(skb)); + if (likely(xsk_addr)) + xsk_addr->num_descs = 1; + return xsk_addr; +} + +static int xsk_skb_destructor_set_addr(struct sk_buff *skb, u64 addr) +{ + if (IS_ENABLED(CONFIG_64BIT)) { + skb_shinfo(skb)->destructor_arg = (void *)((uintptr_t)addr | 0x1UL); + return 0; + } + + if (unlikely(!__xsk_addrs_alloc(skb, addr))) + return -ENOMEM; + return 0; } static void xsk_inc_num_desc(struct sk_buff *skb) @@ -724,14 +757,20 @@ void xsk_destruct_skb(struct sk_buff *skb) sock_wfree(skb); } -static void xsk_skb_init_misc(struct sk_buff *skb, struct xdp_sock *xs, - u64 addr) +static int xsk_skb_init_misc(struct sk_buff *skb, struct xdp_sock *xs, + u64 addr) { + int err; + + err = xsk_skb_destructor_set_addr(skb, addr); + if (unlikely(err)) + return err; + skb->dev = xs->dev; skb->priority = READ_ONCE(xs->sk.sk_priority); skb->mark = READ_ONCE(xs->sk.sk_mark); skb->destructor = xsk_destruct_skb; - xsk_skb_destructor_set_addr(skb, addr); + return 0; } static void xsk_consume_skb(struct sk_buff *skb) @@ -829,18 +868,9 @@ static struct sk_buff *xsk_build_skb_zerocopy(struct xdp_sock *xs, } else { struct xsk_addrs *xsk_addr; - if (xsk_skb_destructor_is_addr(skb)) { - xsk_addr = kmem_cache_zalloc(xsk_tx_generic_cache, - GFP_KERNEL); - if (!xsk_addr) - return ERR_PTR(-ENOMEM); - - xsk_addr->num_descs = 1; - xsk_addr->addrs[0] = xsk_skb_destructor_get_addr(skb); - skb_shinfo(skb)->destructor_arg = (void *)xsk_addr; - } else { - xsk_addr = (struct xsk_addrs *)skb_shinfo(skb)->destructor_arg; - } + xsk_addr = xsk_addrs_alloc(skb); + if (!xsk_addr) + return ERR_PTR(-ENOMEM); /* in case of -EOVERFLOW that could happen below, * xsk_consume_skb() will release this node as whole skb @@ -929,19 +959,10 @@ static struct sk_buff *xsk_build_skb(struct xdp_sock *xs, struct page *page; u8 *vaddr; - if (xsk_skb_destructor_is_addr(skb)) { - xsk_addr = kmem_cache_zalloc(xsk_tx_generic_cache, - GFP_KERNEL); - if (!xsk_addr) { - err = -ENOMEM; - goto free_err; - } - - xsk_addr->num_descs = 1; - xsk_addr->addrs[0] = xsk_skb_destructor_get_addr(skb); - skb_shinfo(skb)->destructor_arg = (void *)xsk_addr; - } else { - xsk_addr = (struct xsk_addrs *)skb_shinfo(skb)->destructor_arg; + xsk_addr = xsk_addrs_alloc(skb); + if (!xsk_addr) { + err = -ENOMEM; + goto free_err; } if (unlikely(nr_frags == (MAX_SKB_FRAGS - 1) && xp_mb_desc(desc))) { @@ -966,8 +987,11 @@ static struct sk_buff *xsk_build_skb(struct xdp_sock *xs, } } - if (!xs->skb) - xsk_skb_init_misc(skb, xs, desc->addr); + if (!xs->skb) { + err = xsk_skb_init_misc(skb, xs, desc->addr); + if (unlikely(err)) + goto free_err; + } xsk_inc_num_desc(skb); return skb; -- cgit v1.2.3 From 283fc9e44ff5b5ac967439b4951b80bd4299f4e4 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Tue, 5 May 2026 15:15:34 +0200 Subject: wifi: mac80211: remove station if connection prep fails If connection preparation fails for MLO connections, then the interface is completely reset to non-MLD. In this case, we must not keep the station since it's related to the link of the vif being removed. Delete an existing station. Any "new_sta" is already being removed, so that doesn't need changes. This fixes a use-after-free/double-free in debugfs if that's enabled, because a vif going from MLD (and to MLD, but that's not relevant here) recreates its entire debugfs. Cc: stable@vger.kernel.org Fixes: 81151ce462e5 ("wifi: mac80211: support MLO authentication/association with one link") Reviewed-by: Miriam Rachel Korenblit Link: https://patch.msgid.link/20260505151533.c4e52deb06ad.Iafe56cec7de8512626169496b134bce3a6c17010@changeid Signed-off-by: Johannes Berg --- net/mac80211/mlme.c | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c index 298ebff6bbf8..0a0f27836d57 100644 --- a/net/mac80211/mlme.c +++ b/net/mac80211/mlme.c @@ -9149,7 +9149,7 @@ static int ieee80211_prep_connection(struct ieee80211_sub_if_data *sdata, struct ieee80211_bss *bss = (void *)cbss->priv; struct sta_info *new_sta = NULL; struct ieee80211_link_data *link; - bool have_sta = false; + struct sta_info *have_sta = NULL; bool mlo; int err; u16 new_links; @@ -9168,11 +9168,8 @@ static int ieee80211_prep_connection(struct ieee80211_sub_if_data *sdata, mlo = false; } - if (assoc) { - rcu_read_lock(); + if (assoc) have_sta = sta_info_get(sdata, ap_mld_addr); - rcu_read_unlock(); - } if (mlo && !have_sta && WARN_ON(sdata->vif.valid_links || sdata->vif.active_links)) @@ -9336,6 +9333,8 @@ static int ieee80211_prep_connection(struct ieee80211_sub_if_data *sdata, out_release_chan: ieee80211_link_release_channel(link); out_err: + if (mlo && have_sta) + WARN_ON(__sta_info_destroy(have_sta)); ieee80211_vif_set_links(sdata, 0, 0); return err; } -- cgit v1.2.3 From 0f3c0a197309717d74729568f88957d448847937 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Tue, 5 May 2026 13:38:37 +0200 Subject: wifi: nl80211: fix NL80211_PMSR_FTM_REQ_ATTR_FTMS_PER_BURST usage This is documented as a u8 and has a policy of NLA_U8, but uses nla_get_u32() which means it's completely broken on big-endian. Fix it to use nla_get_u8(). Fixes: 9bb7e0f24e7e ("cfg80211: add peer measurement with FTM initiator API") Link: https://patch.msgid.link/20260505113837.260159-2-johannes@sipsolutions.net Signed-off-by: Johannes Berg --- net/wireless/pmsr.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/wireless/pmsr.c b/net/wireless/pmsr.c index 4c8ea0583f94..d6cd0de64d1f 100644 --- a/net/wireless/pmsr.c +++ b/net/wireless/pmsr.c @@ -88,7 +88,7 @@ static int pmsr_parse_ftm(struct cfg80211_registered_device *rdev, out->ftm.ftms_per_burst = 0; if (tb[NL80211_PMSR_FTM_REQ_ATTR_FTMS_PER_BURST]) out->ftm.ftms_per_burst = - nla_get_u32(tb[NL80211_PMSR_FTM_REQ_ATTR_FTMS_PER_BURST]); + nla_get_u8(tb[NL80211_PMSR_FTM_REQ_ATTR_FTMS_PER_BURST]); if (capa->ftm.max_ftms_per_burst && (out->ftm.ftms_per_burst > capa->ftm.max_ftms_per_burst || -- cgit v1.2.3 From 15994bb0cbb8fc4879da7552ddd08c1896261c39 Mon Sep 17 00:00:00 2001 From: Maoyi Xie Date: Wed, 6 May 2026 14:48:53 +0800 Subject: wifi: nl80211: require CAP_NET_ADMIN over the target netns in SET_WIPHY_NETNS NL80211_CMD_SET_WIPHY_NETNS dispatches with GENL_UNS_ADMIN_PERM, which verifies that the caller has CAP_NET_ADMIN for the source netns. It doesn't verify that the caller has CAP_NET_ADMIN over the target netns selected by NL80211_ATTR_NETNS_FD or NL80211_ATTR_PID. This diverges from the convention enforced in net/core/rtnetlink.c::rtnl_get_net_ns_capable(): /* For now, the caller is required to have CAP_NET_ADMIN in * the user namespace owning the target net ns. */ if (!sk_ns_capable(sk, net->user_ns, CAP_NET_ADMIN)) return ERR_PTR(-EACCES); A user with CAP_NET_ADMIN in their own user namespace can therefore push a wiphy into an arbitrary netns (including init_net) over which they have no privilege. Mirror the rtnetlink convention by requiring CAP_NET_ADMIN in the target netns before calling cfg80211_switch_netns(). Signed-off-by: Maoyi Xie Link: https://patch.msgid.link/20260506064854.2207105-2-maoyixie.tju@gmail.com Signed-off-by: Johannes Berg --- net/wireless/nl80211.c | 13 +++++++++++++ 1 file changed, 13 insertions(+) (limited to 'net') diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index 67088804dcc7..db546dd93d08 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -13867,6 +13867,19 @@ static int nl80211_wiphy_netns(struct sk_buff *skb, struct genl_info *info) if (IS_ERR(net)) return PTR_ERR(net); + /* + * The caller already has CAP_NET_ADMIN over the source netns + * (enforced by GENL_UNS_ADMIN_PERM on the genl op). Mirror the + * convention used by net/core/rtnetlink.c::rtnl_get_net_ns_capable() + * and require CAP_NET_ADMIN over the target netns as well, so that + * a caller that is privileged in their own user namespace cannot + * push a wiphy into a netns where they have no privilege. + */ + if (!ns_capable(net->user_ns, CAP_NET_ADMIN)) { + put_net(net); + return -EPERM; + } + err = 0; /* check if anything to do */ -- cgit v1.2.3 From 79240f3f6d766b342b57c32397d643e1cfa26b81 Mon Sep 17 00:00:00 2001 From: Maoyi Xie Date: Wed, 6 May 2026 14:48:54 +0800 Subject: wifi: nl80211: re-check wiphy netns in nl80211_prepare_wdev_dump() continuation NL80211_CMD_GET_SCAN is implemented as a multi-call dumpit. The first invocation of nl80211_prepare_wdev_dump() validates the requested wdev against the caller's netns via __cfg80211_wdev_from_attrs(). Subsequent invocations look up the same wiphy by its global index and do not check that the wiphy is still in the caller's netns. Add the same filter to the continuation path. If the wiphy's netns no longer matches the caller's, return -ENODEV and the netlink dump machinery terminates the walk cleanly. Signed-off-by: Maoyi Xie Link: https://patch.msgid.link/20260506064854.2207105-3-maoyixie.tju@gmail.com Signed-off-by: Johannes Berg --- net/wireless/nl80211.c | 12 ++++++++++++ 1 file changed, 12 insertions(+) (limited to 'net') diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index db546dd93d08..7db9cd433801 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -1276,6 +1276,18 @@ static int nl80211_prepare_wdev_dump(struct netlink_callback *cb, rtnl_unlock(); return -ENODEV; } + + /* + * The first invocation validated the wdev's netns against + * the caller via __cfg80211_wdev_from_attrs(). The wiphy + * may have moved netns between dumpit invocations (via + * NL80211_CMD_SET_WIPHY_NETNS), so re-check here. + */ + if (!net_eq(wiphy_net(wiphy), sock_net(cb->skb->sk))) { + rtnl_unlock(); + return -ENODEV; + } + *rdev = wiphy_to_rdev(wiphy); *wdev = NULL; -- cgit v1.2.3 From b819db93d73f4593636299e229914052b89e3ef2 Mon Sep 17 00:00:00 2001 From: Pauli Virtanen Date: Sun, 12 Apr 2026 21:47:42 +0300 Subject: Bluetooth: SCO: fix sleeping under spinlock in sco_conn_ready sco_conn_ready calls sleeping functions under conn->lock spinlock. The critical section can be reduced: conn->hcon is modified only with hdev->lock held. It is guaranteed to be held in sco_conn_ready, so conn->lock is not needed to guard it. Move taking conn->lock after lock_sock(parent). This also follows the lock ordering lock_sock() > conn->lock elsewhere in the file. Fixes: 27c24fda62b60 ("Bluetooth: switch to lock_sock in SCO") Signed-off-by: Pauli Virtanen Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/sco.c | 20 +++++++++----------- 1 file changed, 9 insertions(+), 11 deletions(-) (limited to 'net') diff --git a/net/bluetooth/sco.c b/net/bluetooth/sco.c index 18826d4b9c0b..3a5479538e85 100644 --- a/net/bluetooth/sco.c +++ b/net/bluetooth/sco.c @@ -1377,26 +1377,24 @@ static void sco_conn_ready(struct sco_conn *conn) sk->sk_state_change(sk); release_sock(sk); } else { - sco_conn_lock(conn); - - if (!conn->hcon) { - sco_conn_unlock(conn); + if (!conn->hcon) return; - } + + lockdep_assert_held(&conn->hcon->hdev->lock); parent = sco_get_sock_listen(&conn->hcon->src); - if (!parent) { - sco_conn_unlock(conn); + if (!parent) return; - } lock_sock(parent); + sco_conn_lock(conn); + sk = sco_sock_alloc(sock_net(parent), NULL, BTPROTO_SCO, GFP_ATOMIC, 0); if (!sk) { - release_sock(parent); sco_conn_unlock(conn); + release_sock(parent); return; } @@ -1417,9 +1415,9 @@ static void sco_conn_ready(struct sco_conn *conn) /* Wake up parent */ parent->sk_data_ready(parent); - release_sock(parent); - sco_conn_unlock(conn); + + release_sock(parent); } } -- cgit v1.2.3 From 0beddb0c380bed5f5b8e61ddbe14635bb73d0b41 Mon Sep 17 00:00:00 2001 From: David Carlier Date: Sun, 12 Apr 2026 21:29:16 +0100 Subject: Bluetooth: hci_conn: fix potential UAF in create_big_sync Add hci_conn_valid() check in create_big_sync() to detect stale connections before proceeding with BIG creation. Handle the resulting -ECANCELED in create_big_complete() and re-validate the connection under hci_dev_lock() before dereferencing, matching the pattern used by create_le_conn_complete() and create_pa_complete(). Keep the hci_conn object alive across the async boundary by taking a reference via hci_conn_get() when queueing create_big_sync(), and dropping it in the completion callback. The refcount and the lock are complementary: the refcount keeps the object allocated, while hci_dev_lock() serializes hci_conn_hash_del()'s list_del_rcu() on hdev->conn_hash, as required by hci_conn_del(). hci_conn_put() is called outside hci_dev_unlock() so the final put (which resolves to kfree() via bt_link_release) does not run under hdev->lock, though the release path would be safe either way. Without this, create_big_complete() would unconditionally dereference the conn pointer on error, causing a use-after-free via hci_connect_cfm() and hci_conn_del(). Fixes: eca0ae4aea66 ("Bluetooth: Add initial implementation of BIS connections") Cc: stable@vger.kernel.org Co-developed-by: Luiz Augusto von Dentz Signed-off-by: Luiz Augusto von Dentz Signed-off-by: David Carlier Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/hci_conn.c | 19 ++++++++++++++++++- 1 file changed, 18 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c index 3a0592599086..96e345fcf303 100644 --- a/net/bluetooth/hci_conn.c +++ b/net/bluetooth/hci_conn.c @@ -2130,6 +2130,9 @@ static int create_big_sync(struct hci_dev *hdev, void *data) u32 flags = 0; int err; + if (!hci_conn_valid(hdev, conn)) + return -ECANCELED; + if (qos->bcast.out.phys == BIT(1)) flags |= MGMT_ADV_FLAG_SEC_2M; @@ -2204,11 +2207,24 @@ static void create_big_complete(struct hci_dev *hdev, void *data, int err) bt_dev_dbg(hdev, "conn %p", conn); + if (err == -ECANCELED) + goto done; + + hci_dev_lock(hdev); + + if (!hci_conn_valid(hdev, conn)) + goto unlock; + if (err) { bt_dev_err(hdev, "Unable to create BIG: %d", err); hci_connect_cfm(conn, err); hci_conn_del(conn); } + +unlock: + hci_dev_unlock(hdev); +done: + hci_conn_put(conn); } struct hci_conn *hci_bind_bis(struct hci_dev *hdev, bdaddr_t *dst, __u8 sid, @@ -2336,10 +2352,11 @@ struct hci_conn *hci_connect_bis(struct hci_dev *hdev, bdaddr_t *dst, BT_BOUND, &data); /* Queue start periodic advertising and create BIG */ - err = hci_cmd_sync_queue(hdev, create_big_sync, conn, + err = hci_cmd_sync_queue(hdev, create_big_sync, hci_conn_get(conn), create_big_complete); if (err < 0) { hci_conn_drop(conn); + hci_conn_put(conn); return ERR_PTR(err); } -- cgit v1.2.3 From 5ddb8014261137cadaf83ab5617a588d80a22586 Mon Sep 17 00:00:00 2001 From: Luiz Augusto von Dentz Date: Fri, 10 Apr 2026 15:29:52 -0400 Subject: Bluetooth: hci_event: Fix OOB read and infinite loop in hci_le_create_big_complete_evt hci_le_create_big_complete_evt() iterates over BT_BOUND connections for a BIG handle using a while loop, accessing ev->bis_handle[i++] on each iteration. However, there is no check that i stays within ev->num_bis before the array access. When a controller sends a LE_Create_BIG_Complete event with fewer bis_handle entries than there are BT_BOUND connections for that BIG, or with num_bis=0, the loop reads beyond the valid bis_handle[] flex array into adjacent heap memory. Since the out-of-bounds values typically exceed HCI_CONN_HANDLE_MAX (0x0EFF), hci_conn_set_handle() rejects them and the connection remains in BT_BOUND state. The same connection is then found again by hci_conn_hash_lookup_big_state(), creating an infinite loop with hci_dev_lock held. Fix this by terminating the BIG if in case not all BIS could be setup properly. Fixes: a0bfde167b50 ("Bluetooth: ISO: Add support for connecting multiple BISes") Cc: stable@vger.kernel.org Signed-off-by: ZhiTao Ou Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/hci_event.c | 27 +++++++++++++++++++++++++-- 1 file changed, 25 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c index b2ee6b6a0f56..1b3b9131affa 100644 --- a/net/bluetooth/hci_event.c +++ b/net/bluetooth/hci_event.c @@ -7118,9 +7118,29 @@ static void hci_le_create_big_complete_evt(struct hci_dev *hdev, void *data, continue; } + if (ev->num_bis <= i) { + bt_dev_err(hdev, + "Not enough BIS handles for BIG 0x%2.2x", + ev->handle); + ev->status = HCI_ERROR_UNSPECIFIED; + hci_connect_cfm(conn, ev->status); + hci_conn_del(conn); + continue; + } + if (hci_conn_set_handle(conn, - __le16_to_cpu(ev->bis_handle[i++]))) + __le16_to_cpu(ev->bis_handle[i++]))) { + bt_dev_err(hdev, + "Failed to set BIS handle for BIG 0x%2.2x", + ev->handle); + /* Force error so BIG gets terminated as not all BIS + * could be connected. + */ + ev->status = HCI_ERROR_UNSPECIFIED; + hci_connect_cfm(conn, ev->status); + hci_conn_del(conn); continue; + } conn->state = BT_CONNECTED; set_bit(HCI_CONN_BIG_CREATED, &conn->flags); @@ -7129,7 +7149,10 @@ static void hci_le_create_big_complete_evt(struct hci_dev *hdev, void *data, hci_iso_setup_path(conn); } - if (!ev->status && !i) + /* If there is an unexpected error or if no BISes have been connected + * for the BIG, terminate it. + */ + if (ev->status == HCI_ERROR_UNSPECIFIED || (!ev->status && !i)) /* If no BISes have been connected for the BIG, * terminate. This is in case all bound connections * have been closed before the BIG creation -- cgit v1.2.3 From 72b8deccff17a7644e0367e1aaf1a36cfb014324 Mon Sep 17 00:00:00 2001 From: Dudu Lu Date: Wed, 15 Apr 2026 17:39:53 +0800 Subject: Bluetooth: bnep: fix incorrect length parsing in bnep_rx_frame() extension handling In bnep_rx_frame(), the BNEP_FILTER_NET_TYPE_SET and BNEP_FILTER_MULTI_ADDR_SET extension header parsing has two bugs: 1) The 2-byte length field is read with *(u16 *)(skb->data + 1), which performs a native-endian read. The BNEP protocol specifies this field in big-endian (network byte order), and the same file correctly uses get_unaligned_be16() for the identical fields in bnep_ctrl_set_netfilter() and bnep_ctrl_set_mcfilter(). 2) The length is multiplied by 2, but unlike BNEP_SETUP_CONN_REQ where the length byte counts UUID pairs (requiring * 2 for two UUIDs per entry), the filter extension length field already represents the total data size in bytes. This is confirmed by bnep_ctrl_set_netfilter() which reads the same field as a byte count and divides by 4 to get the number of filter entries. The bogus * 2 means skb_pull advances twice as far as it should, either dropping valid data from the next header or causing the pull to fail entirely when the doubled length exceeds the remaining skb. Fix by splitting the pull into two steps: first use skb_pull_data() to safely pull and validate the 3-byte fixed header (ctrl type + length), then pull the variable-length data using the properly decoded length. Fixes: bf8b9a9cb77b ("Bluetooth: bnep: Add support to extended headers of control frames") Signed-off-by: Dudu Lu Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/bnep/core.c | 13 ++++++++++--- 1 file changed, 10 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/bluetooth/bnep/core.c b/net/bluetooth/bnep/core.c index d44987d4515c..853c8d7644b5 100644 --- a/net/bluetooth/bnep/core.c +++ b/net/bluetooth/bnep/core.c @@ -330,11 +330,18 @@ static int bnep_rx_frame(struct bnep_session *s, struct sk_buff *skb) goto badframe; break; case BNEP_FILTER_MULTI_ADDR_SET: - case BNEP_FILTER_NET_TYPE_SET: - /* Pull: ctrl type (1 b), len (2 b), data (len bytes) */ - if (!skb_pull(skb, 3 + *(u16 *)(skb->data + 1) * 2)) + case BNEP_FILTER_NET_TYPE_SET: { + u8 *hdr; + + /* Pull ctrl type (1 b) + len (2 b) */ + hdr = skb_pull_data(skb, 3); + if (!hdr) + goto badframe; + /* Pull data (len bytes); length is big-endian */ + if (!skb_pull(skb, get_unaligned_be16(&hdr[1]))) goto badframe; break; + } default: kfree_skb(skb); return 0; -- cgit v1.2.3 From 4f42363c814f28fe3f59847c35acf1ed033bedd4 Mon Sep 17 00:00:00 2001 From: Dudu Lu Date: Wed, 15 Apr 2026 18:43:55 +0800 Subject: Bluetooth: l2cap: fix MPS check in l2cap_ecred_reconf_req The L2CAP specification states that if more than one channel is being reconfigured, the MPS shall not be decreased. The current check has two issues: 1) The comparison uses >= (greater-than-or-equal), which incorrectly rejects reconfiguration requests where the MPS stays the same. Since the spec says MPS "shall be greater than or equal to the current MPS", only a strict decrease (remote_mps > mps) should be rejected. Keeping the same MPS is valid. 2) The multi-channel guard uses `&& i` (loop index) to approximate "more than one channel", but this incorrectly allows MPS decrease for the first channel (i==0) even when multiple channels are being reconfigured. Replace with `&& num_scid > 1` which correctly checks whether the request covers more than one channel. Fixes: 7accb1c4321a ("Bluetooth: L2CAP: Fix invalid response to L2CAP_ECRED_RECONF_REQ") Signed-off-by: Dudu Lu Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/l2cap_core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c index 77dec104a9c3..b15374b951fa 100644 --- a/net/bluetooth/l2cap_core.c +++ b/net/bluetooth/l2cap_core.c @@ -5428,7 +5428,7 @@ static inline int l2cap_ecred_reconf_req(struct l2cap_conn *conn, * configured, the MPS field may be less than the current MPS * of that channel. */ - if (chan[i]->remote_mps >= mps && i) { + if (chan[i]->remote_mps > mps && num_scid > 1) { BT_ERR("chan %p decreased MPS %u -> %u", chan[i], chan[i]->remote_mps, mps); result = L2CAP_RECONF_INVALID_MPS; -- cgit v1.2.3 From 91b5a598b5285da794b72619f31777b62dd336f8 Mon Sep 17 00:00:00 2001 From: Mikhail Gavrilov Date: Wed, 15 Apr 2026 02:52:37 +0500 Subject: Bluetooth: l2cap: defer conn param update to avoid conn->lock/hdev->lock inversion When a BLE peripheral sends an L2CAP Connection Parameter Update Request the processing path is: process_pending_rx() [takes conn->lock] l2cap_le_sig_channel() l2cap_conn_param_update_req() hci_le_conn_update() [takes hdev->lock] Meanwhile other code paths take the locks in the opposite order: l2cap_chan_connect() [takes hdev->lock] ... mutex_lock(&conn->lock) l2cap_conn_ready() [hdev->lock via hci_cb_list_lock] ... mutex_lock(&conn->lock) This is a classic AB/BA deadlock which lockdep reports as a circular locking dependency when connecting a BLE MIDI keyboard (Carry-On FC-49). Fix this by making hci_le_conn_update() defer the HCI command through hci_cmd_sync_queue() so it no longer needs to take hdev->lock in the caller context. The sync callback uses __hci_cmd_sync_status_sk() to wait for the HCI_EV_LE_CONN_UPDATE_COMPLETE event, then updates the stored connection parameters (hci_conn_params) and notifies userspace (mgmt_new_conn_param) only after the controller has confirmed the update. A reference on hci_conn is held via hci_conn_get()/hci_conn_put() for the lifetime of the queued work to prevent use-after-free, and hci_conn_valid() is checked before proceeding in case the connection was removed while the work was pending. The hci_dev_lock is held across hci_conn_valid() and all conn field accesses to prevent a concurrent disconnect from invalidating the connection mid-use. Fixes: f044eb0524a0 ("Bluetooth: Store latency and supervision timeout in connection params") Signed-off-by: Mikhail Gavrilov Reviewed-by: Paul Menzel Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/hci_conn.c | 105 +++++++++++++++++++++++++++++++++++++-------- net/bluetooth/l2cap_core.c | 12 +----- 2 files changed, 88 insertions(+), 29 deletions(-) (limited to 'net') diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c index 96e345fcf303..17b46ad6a349 100644 --- a/net/bluetooth/hci_conn.c +++ b/net/bluetooth/hci_conn.c @@ -480,40 +480,107 @@ bool hci_setup_sync(struct hci_conn *conn, __u16 handle) return hci_setup_sync_conn(conn, handle); } -u8 hci_le_conn_update(struct hci_conn *conn, u16 min, u16 max, u16 latency, - u16 to_multiplier) +struct le_conn_update_data { + struct hci_conn *conn; + u16 min; + u16 max; + u16 latency; + u16 to_multiplier; +}; + +static int le_conn_update_sync(struct hci_dev *hdev, void *data) { - struct hci_dev *hdev = conn->hdev; + struct le_conn_update_data *d = data; + struct hci_conn *conn = d->conn; struct hci_conn_params *params; struct hci_cp_le_conn_update cp; + u16 timeout; + u8 store_hint; + int err; + /* Verify connection is still alive and read conn fields under + * the same lock to prevent a concurrent disconnect from freeing + * or reusing the connection while we build the HCI command. + */ hci_dev_lock(hdev); - params = hci_conn_params_lookup(hdev, &conn->dst, conn->dst_type); - if (params) { - params->conn_min_interval = min; - params->conn_max_interval = max; - params->conn_latency = latency; - params->supervision_timeout = to_multiplier; + if (!hci_conn_valid(hdev, conn)) { + hci_dev_unlock(hdev); + return -ECANCELED; } - hci_dev_unlock(hdev); - memset(&cp, 0, sizeof(cp)); cp.handle = cpu_to_le16(conn->handle); - cp.conn_interval_min = cpu_to_le16(min); - cp.conn_interval_max = cpu_to_le16(max); - cp.conn_latency = cpu_to_le16(latency); - cp.supervision_timeout = cpu_to_le16(to_multiplier); + cp.conn_interval_min = cpu_to_le16(d->min); + cp.conn_interval_max = cpu_to_le16(d->max); + cp.conn_latency = cpu_to_le16(d->latency); + cp.supervision_timeout = cpu_to_le16(d->to_multiplier); cp.min_ce_len = cpu_to_le16(0x0000); cp.max_ce_len = cpu_to_le16(0x0000); + timeout = conn->conn_timeout; - hci_send_cmd(hdev, HCI_OP_LE_CONN_UPDATE, sizeof(cp), &cp); + hci_dev_unlock(hdev); - if (params) - return 0x01; + err = __hci_cmd_sync_status_sk(hdev, HCI_OP_LE_CONN_UPDATE, + sizeof(cp), &cp, + HCI_EV_LE_CONN_UPDATE_COMPLETE, + timeout, NULL); + if (err) + return err; - return 0x00; + /* Update stored connection parameters after the controller has + * confirmed the update via the LE Connection Update Complete event. + */ + hci_dev_lock(hdev); + + params = hci_conn_params_lookup(hdev, &conn->dst, conn->dst_type); + if (params) { + params->conn_min_interval = d->min; + params->conn_max_interval = d->max; + params->conn_latency = d->latency; + params->supervision_timeout = d->to_multiplier; + store_hint = 0x01; + } else { + store_hint = 0x00; + } + + hci_dev_unlock(hdev); + + mgmt_new_conn_param(hdev, &conn->dst, conn->dst_type, store_hint, + d->min, d->max, d->latency, d->to_multiplier); + + return 0; +} + +static void le_conn_update_complete(struct hci_dev *hdev, void *data, int err) +{ + struct le_conn_update_data *d = data; + + hci_conn_put(d->conn); + kfree(d); +} + +void hci_le_conn_update(struct hci_conn *conn, u16 min, u16 max, u16 latency, + u16 to_multiplier) +{ + struct le_conn_update_data *d; + + d = kzalloc_obj(*d); + if (!d) + return; + + hci_conn_get(conn); + d->conn = conn; + d->min = min; + d->max = max; + d->latency = latency; + d->to_multiplier = to_multiplier; + + if (hci_cmd_sync_queue(conn->hdev, le_conn_update_sync, d, + le_conn_update_complete) < 0) { + hci_conn_put(conn); + kfree(d); + } } void hci_le_start_enc(struct hci_conn *conn, __le16 ediv, __le64 rand, diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c index b15374b951fa..7701528f1167 100644 --- a/net/bluetooth/l2cap_core.c +++ b/net/bluetooth/l2cap_core.c @@ -4706,16 +4706,8 @@ static inline int l2cap_conn_param_update_req(struct l2cap_conn *conn, l2cap_send_cmd(conn, cmd->ident, L2CAP_CONN_PARAM_UPDATE_RSP, sizeof(rsp), &rsp); - if (!err) { - u8 store_hint; - - store_hint = hci_le_conn_update(hcon, min, max, latency, - to_multiplier); - mgmt_new_conn_param(hcon->hdev, &hcon->dst, hcon->dst_type, - store_hint, min, max, latency, - to_multiplier); - - } + if (!err) + hci_le_conn_update(hcon, min, max, latency, to_multiplier); return 0; } -- cgit v1.2.3 From 2ff1a41a912de8517b4482e946dd951b7d80edbf Mon Sep 17 00:00:00 2001 From: Siwei Zhang Date: Wed, 15 Apr 2026 16:51:36 -0400 Subject: Bluetooth: L2CAP: Fix null-ptr-deref in l2cap_sock_state_change_cb() Add the same NULL guard already present in l2cap_sock_resume_cb() and l2cap_sock_ready_cb(). Fixes: 89bc500e41fc ("Bluetooth: Add state tracking to struct l2cap_chan") Cc: stable@kernel.org Signed-off-by: Siwei Zhang Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/l2cap_sock.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'net') diff --git a/net/bluetooth/l2cap_sock.c b/net/bluetooth/l2cap_sock.c index 71e8c1b45bce..fb3cb70a5a39 100644 --- a/net/bluetooth/l2cap_sock.c +++ b/net/bluetooth/l2cap_sock.c @@ -1657,6 +1657,9 @@ static void l2cap_sock_state_change_cb(struct l2cap_chan *chan, int state, { struct sock *sk = chan->data; + if (!sk) + return; + sk->sk_state = state; if (err) -- cgit v1.2.3 From 78a88d43dab8d23aeef934ed8ce34d40e6b3d613 Mon Sep 17 00:00:00 2001 From: Siwei Zhang Date: Wed, 15 Apr 2026 16:53:36 -0400 Subject: Bluetooth: L2CAP: Fix null-ptr-deref in l2cap_sock_get_sndtimeo_cb() Add the same NULL guard already present in l2cap_sock_resume_cb() and l2cap_sock_ready_cb(). Fixes: 8d836d71e222 ("Bluetooth: Access sk_sndtimeo indirectly in l2cap_core.c") Cc: stable@kernel.org Signed-off-by: Siwei Zhang Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/l2cap_sock.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'net') diff --git a/net/bluetooth/l2cap_sock.c b/net/bluetooth/l2cap_sock.c index fb3cb70a5a39..879c9f90269a 100644 --- a/net/bluetooth/l2cap_sock.c +++ b/net/bluetooth/l2cap_sock.c @@ -1761,6 +1761,9 @@ static long l2cap_sock_get_sndtimeo_cb(struct l2cap_chan *chan) { struct sock *sk = chan->data; + if (!sk) + return 0; + return READ_ONCE(sk->sk_sndtimeo); } -- cgit v1.2.3 From 0a120d96166301d7a95be75b52f843837dbd1219 Mon Sep 17 00:00:00 2001 From: Siwei Zhang Date: Wed, 15 Apr 2026 16:49:59 -0400 Subject: Bluetooth: L2CAP: Fix null-ptr-deref in l2cap_sock_new_connection_cb() Add the same NULL guard already present in l2cap_sock_resume_cb() and l2cap_sock_ready_cb(). Fixes: 80808e431e1e ("Bluetooth: Add l2cap_chan_ops abstraction") Cc: stable@kernel.org Signed-off-by: Siwei Zhang Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/l2cap_sock.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'net') diff --git a/net/bluetooth/l2cap_sock.c b/net/bluetooth/l2cap_sock.c index 879c9f90269a..cf590a67d364 100644 --- a/net/bluetooth/l2cap_sock.c +++ b/net/bluetooth/l2cap_sock.c @@ -1498,6 +1498,9 @@ static struct l2cap_chan *l2cap_sock_new_connection_cb(struct l2cap_chan *chan) { struct sock *sk, *parent = chan->data; + if (!parent) + return NULL; + lock_sock(parent); /* Check for backlog size */ -- cgit v1.2.3 From 4e37f6452d586b95c346a9abdd2fb80b67794f39 Mon Sep 17 00:00:00 2001 From: Pauli Virtanen Date: Sat, 18 Apr 2026 18:41:12 +0300 Subject: Bluetooth: SCO: hold sk properly in sco_conn_ready sk deref in sco_conn_ready must be done either under conn->lock, or holding a refcount, to avoid concurrent close. conn->sk and parent sk is currently accessed without either, and without checking parent->sk_state: [Task 1] [Task 2] sco_sock_release sco_conn_ready sk = conn->sk lock_sock(sk) conn->sk = NULL lock_sock(sk) release_sock(sk) sco_sock_kill(sk) UAF on sk deref and similarly for access to sco_get_sock_listen() return value. Fix possible UAF by holding sk refcount in sco_conn_ready() and making sco_get_sock_listen() increase refcount. Also recheck after lock_sock that the socket is still valid. Adjust conn->sk locking so it's protected also by lock_sock() of the associated socket if any. Fixes: 27c24fda62b60 ("Bluetooth: switch to lock_sock in SCO") Signed-off-by: Pauli Virtanen Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/sco.c | 44 ++++++++++++++++++++++++++++++++------------ 1 file changed, 32 insertions(+), 12 deletions(-) (limited to 'net') diff --git a/net/bluetooth/sco.c b/net/bluetooth/sco.c index 3a5479538e85..eba44525d41d 100644 --- a/net/bluetooth/sco.c +++ b/net/bluetooth/sco.c @@ -472,9 +472,13 @@ static struct sock *sco_get_sock_listen(bdaddr_t *src) sk1 = sk; } + sk = sk ? sk : sk1; + if (sk) + sock_hold(sk); + read_unlock(&sco_sk_list.lock); - return sk ? sk : sk1; + return sk; } static void sco_sock_destruct(struct sock *sk) @@ -515,11 +519,13 @@ static void sco_sock_kill(struct sock *sk) BT_DBG("sk %p state %d", sk, sk->sk_state); /* Sock is dead, so set conn->sk to NULL to avoid possible UAF */ + lock_sock(sk); if (sco_pi(sk)->conn) { sco_conn_lock(sco_pi(sk)->conn); sco_pi(sk)->conn->sk = NULL; sco_conn_unlock(sco_pi(sk)->conn); } + release_sock(sk); /* Kill poor orphan */ bt_sock_unlink(&sco_sk_list, sk); @@ -1365,17 +1371,28 @@ static int sco_sock_release(struct socket *sock) static void sco_conn_ready(struct sco_conn *conn) { - struct sock *parent; - struct sock *sk = conn->sk; + struct sock *parent, *sk; + + sco_conn_lock(conn); + sk = sco_sock_hold(conn); + sco_conn_unlock(conn); BT_DBG("conn %p", conn); if (sk) { lock_sock(sk); - sco_sock_clear_timer(sk); - sk->sk_state = BT_CONNECTED; - sk->sk_state_change(sk); + + /* conn->sk may have become NULL if racing with sk close, but + * due to held hdev->lock, it can't become different sk. + */ + if (conn->sk) { + sco_sock_clear_timer(sk); + sk->sk_state = BT_CONNECTED; + sk->sk_state_change(sk); + } + release_sock(sk); + sock_put(sk); } else { if (!conn->hcon) return; @@ -1390,13 +1407,15 @@ static void sco_conn_ready(struct sco_conn *conn) sco_conn_lock(conn); + /* hdev->lock guarantees conn->sk == NULL still here */ + + if (parent->sk_state != BT_LISTEN) + goto release; + sk = sco_sock_alloc(sock_net(parent), NULL, BTPROTO_SCO, GFP_ATOMIC, 0); - if (!sk) { - sco_conn_unlock(conn); - release_sock(parent); - return; - } + if (!sk) + goto release; sco_sock_init(sk, parent); @@ -1415,9 +1434,10 @@ static void sco_conn_ready(struct sco_conn *conn) /* Wake up parent */ parent->sk_data_ready(parent); +release: sco_conn_unlock(conn); - release_sock(parent); + sock_put(parent); } } -- cgit v1.2.3 From ca40d481079c05c6891a14a798c79596fd2d5f0c Mon Sep 17 00:00:00 2001 From: SeungJu Cheon Date: Tue, 21 Apr 2026 11:51:21 +0900 Subject: Bluetooth: ISO: Fix data-race on dst in iso_sock_connect() iso_sock_connect() copies the destination address into iso_pi(sk)->dst under lock_sock, then releases the lock and reads it back with bacmp() to decide between the CIS and BIS connect paths: lock_sock(sk); bacpy(&iso_pi(sk)->dst, &sa->iso_bdaddr); iso_pi(sk)->dst_type = sa->iso_bdaddr_type; release_sock(sk); if (bacmp(&iso_pi(sk)->dst, BDADDR_ANY)) // <- no lock held This read after release_sock() races with any concurrent write to iso_pi(sk)->dst on the same socket. Fix by reading the destination address directly from the local sockaddr argument (sa->iso_bdaddr) instead of iso_pi(sk)->dst. Since sa is a function-local argument, reading it requires no locking and avoids the race. This patch addresses only the bacmp() race in iso_sock_connect(); other unprotected iso_pi(sk) accesses are fixed separately in the next patch. KCSAN report: BUG: KCSAN: data-race in memcmp+0x39/0xb0 race at unknown origin, with read to 0xffff8f96ea66dde3 of 1 bytes by task 549 on cpu 1: memcmp+0x39/0xb0 iso_sock_connect+0x275/0xb40 __sys_connect_file+0xbd/0xe0 __sys_connect+0xe0/0x110 __x64_sys_connect+0x40/0x50 x64_sys_call+0xcad/0x1c60 do_syscall_64+0x133/0x590 entry_SYSCALL_64_after_hwframe+0x77/0x7f value changed: 0x00 -> 0xee Reported by Kernel Concurrency Sanitizer on: CPU: 1 UID: 0 PID: 549 Comm: iso_race_combin Not tainted 7.0.0-08391-g1d51b370a0f8 #40 PREEMPT(lazy) Fixes: ccf74f2390d6 ("Bluetooth: Add BTPROTO_ISO socket type") Signed-off-by: SeungJu Cheon Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/iso.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/bluetooth/iso.c b/net/bluetooth/iso.c index be145e2736b7..290a1b9a9daa 100644 --- a/net/bluetooth/iso.c +++ b/net/bluetooth/iso.c @@ -1193,7 +1193,7 @@ static int iso_sock_connect(struct socket *sock, struct sockaddr_unsized *addr, release_sock(sk); - if (bacmp(&iso_pi(sk)->dst, BDADDR_ANY)) + if (bacmp(&sa->iso_bdaddr, BDADDR_ANY)) err = iso_connect_cis(sk); else err = iso_connect_bis(sk); -- cgit v1.2.3 From f958c7805b18e9d69f6b322b231ecee46ec6f331 Mon Sep 17 00:00:00 2001 From: SeungJu Cheon Date: Tue, 21 Apr 2026 11:51:22 +0900 Subject: Bluetooth: ISO: Fix data-race on iso_pi(sk) in socket and HCI event paths Several iso_pi(sk) fields (qos, qos_user_set, bc_sid, base, base_len, sync_handle, bc_num_bis) are written under lock_sock in iso_sock_setsockopt() and iso_sock_bind(), but read and written under hci_dev_lock only in two other paths: - iso_connect_bis() / iso_connect_cis(), invoked from connect(2), read qos/base/bc_sid and reset qos to default_qos on the qos_user_set validation failure -- all without lock_sock. - iso_connect_ind(), invoked from hci_rx_work, writes sync_handle, bc_sid, qos.bcast.encryption, bc_num_bis, base and base_len on PA_SYNC_ESTABLISHED / PAST_RECEIVED / BIG_INFO_ADV_REPORT / PER_ADV_REPORT events. The BIG_INFO handler additionally passes &iso_pi(sk)->qos together with sync_handle / bc_num_bis / bc_bis to hci_conn_big_create_sync() while setsockopt may be mutating them. Acquire lock_sock around the affected accesses in both paths. The locking order hci_dev_lock -> lock_sock matches the existing iso_conn_big_sync() precedent, whose comment documents the same requirement for hci_conn_big_create_sync(). The HCI connect/bind helpers do not wait for command completion -- they enqueue work via hci_cmd_sync_queue{,_once}() / hci_le_create_cis_pending() and return -- so the added hold time is comparable to iso_conn_big_sync(). KCSAN report: BUG: KCSAN: data-race in iso_connect_cis / iso_sock_setsockopt read to 0xffffa3ae8ce3cdc8 of 1 bytes by task 335 on cpu 0: iso_connect_cis+0x49f/0xa20 iso_sock_connect+0x60e/0xb40 __sys_connect_file+0xbd/0xe0 __sys_connect+0xe0/0x110 __x64_sys_connect+0x40/0x50 x64_sys_call+0xcad/0x1c60 do_syscall_64+0x133/0x590 entry_SYSCALL_64_after_hwframe+0x77/0x7f write to 0xffffa3ae8ce3cdc8 of 60 bytes by task 334 on cpu 1: iso_sock_setsockopt+0x69a/0x930 do_sock_setsockopt+0xc3/0x170 __sys_setsockopt+0xd1/0x130 __x64_sys_setsockopt+0x64/0x80 x64_sys_call+0x1547/0x1c60 do_syscall_64+0x133/0x590 entry_SYSCALL_64_after_hwframe+0x77/0x7f Reported by Kernel Concurrency Sanitizer on: CPU: 1 UID: 0 PID: 334 Comm: iso_setup_race Not tainted 7.0.0-10949-g8541d8f725c6 #44 PREEMPT(lazy) The iso_connect_ind() races were found by inspection. Fixes: ccf74f2390d6 ("Bluetooth: Add BTPROTO_ISO socket type") Signed-off-by: SeungJu Cheon Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/iso.c | 54 +++++++++++++++++++++++++++++------------------------ 1 file changed, 30 insertions(+), 24 deletions(-) (limited to 'net') diff --git a/net/bluetooth/iso.c b/net/bluetooth/iso.c index 290a1b9a9daa..7cb2864fe872 100644 --- a/net/bluetooth/iso.c +++ b/net/bluetooth/iso.c @@ -347,6 +347,7 @@ static int iso_connect_bis(struct sock *sk) return -EHOSTUNREACH; hci_dev_lock(hdev); + lock_sock(sk); if (!bis_capable(hdev)) { err = -EOPNOTSUPP; @@ -399,13 +400,9 @@ static int iso_connect_bis(struct sock *sk) goto unlock; } - lock_sock(sk); - err = iso_chan_add(conn, sk, NULL); - if (err) { - release_sock(sk); + if (err) goto unlock; - } /* Update source addr of the socket */ bacpy(&iso_pi(sk)->src, &hcon->src); @@ -421,9 +418,8 @@ static int iso_connect_bis(struct sock *sk) iso_sock_set_timer(sk, READ_ONCE(sk->sk_sndtimeo)); } - release_sock(sk); - unlock: + release_sock(sk); hci_dev_unlock(hdev); hci_dev_put(hdev); return err; @@ -444,6 +440,7 @@ static int iso_connect_cis(struct sock *sk) return -EHOSTUNREACH; hci_dev_lock(hdev); + lock_sock(sk); if (!cis_central_capable(hdev)) { err = -EOPNOTSUPP; @@ -498,13 +495,9 @@ static int iso_connect_cis(struct sock *sk) goto unlock; } - lock_sock(sk); - err = iso_chan_add(conn, sk, NULL); - if (err) { - release_sock(sk); + if (err) goto unlock; - } /* Update source addr of the socket */ bacpy(&iso_pi(sk)->src, &hcon->src); @@ -520,9 +513,8 @@ static int iso_connect_cis(struct sock *sk) iso_sock_set_timer(sk, READ_ONCE(sk->sk_sndtimeo)); } - release_sock(sk); - unlock: + release_sock(sk); hci_dev_unlock(hdev); hci_dev_put(hdev); return err; @@ -2256,8 +2248,10 @@ int iso_connect_ind(struct hci_dev *hdev, bdaddr_t *bdaddr, __u8 *flags) sk = iso_get_sock(hdev, &hdev->bdaddr, bdaddr, BT_LISTEN, iso_match_sid, ev1); if (sk && !ev1->status) { + lock_sock(sk); iso_pi(sk)->sync_handle = le16_to_cpu(ev1->handle); iso_pi(sk)->bc_sid = ev1->sid; + release_sock(sk); } goto done; @@ -2268,8 +2262,10 @@ int iso_connect_ind(struct hci_dev *hdev, bdaddr_t *bdaddr, __u8 *flags) sk = iso_get_sock(hdev, &hdev->bdaddr, bdaddr, BT_LISTEN, iso_match_sid_past, ev1a); if (sk && !ev1a->status) { + lock_sock(sk); iso_pi(sk)->sync_handle = le16_to_cpu(ev1a->sync_handle); iso_pi(sk)->bc_sid = ev1a->sid; + release_sock(sk); } goto done; @@ -2296,27 +2292,35 @@ int iso_connect_ind(struct hci_dev *hdev, bdaddr_t *bdaddr, __u8 *flags) ev2); if (sk) { - int err; - struct hci_conn *hcon = iso_pi(sk)->conn->hcon; + int err = 0; + bool big_sync; + struct hci_conn *hcon; + lock_sock(sk); + + hcon = iso_pi(sk)->conn->hcon; iso_pi(sk)->qos.bcast.encryption = ev2->encryption; if (ev2->num_bis < iso_pi(sk)->bc_num_bis) iso_pi(sk)->bc_num_bis = ev2->num_bis; - if (!test_bit(BT_SK_DEFER_SETUP, &bt_sk(sk)->flags) && - !test_and_set_bit(BT_SK_BIG_SYNC, &iso_pi(sk)->flags)) { + big_sync = !test_bit(BT_SK_DEFER_SETUP, &bt_sk(sk)->flags) && + !test_and_set_bit(BT_SK_BIG_SYNC, &iso_pi(sk)->flags); + + if (big_sync) err = hci_conn_big_create_sync(hdev, hcon, &iso_pi(sk)->qos, iso_pi(sk)->sync_handle, iso_pi(sk)->bc_num_bis, iso_pi(sk)->bc_bis); - if (err) { - bt_dev_err(hdev, "hci_le_big_create_sync: %d", - err); - sock_put(sk); - sk = NULL; - } + + release_sock(sk); + + if (big_sync && err) { + bt_dev_err(hdev, "hci_le_big_create_sync: %d", + err); + sock_put(sk); + sk = NULL; } } @@ -2370,8 +2374,10 @@ int iso_connect_ind(struct hci_dev *hdev, bdaddr_t *bdaddr, __u8 *flags) if (!base || base_len > BASE_MAX_LENGTH) goto done; + lock_sock(sk); memcpy(iso_pi(sk)->base, base, base_len); iso_pi(sk)->base_len = base_len; + release_sock(sk); } else { /* This is a PA data fragment. Keep pa_data_len set to 0 * until all data has been reassembled. -- cgit v1.2.3 From 8f59d17b18a78fdfdbb67d693b3d3eb03db184e0 Mon Sep 17 00:00:00 2001 From: Pengpeng Hou Date: Thu, 23 Apr 2026 23:31:00 +0800 Subject: Bluetooth: RFCOMM: pull credit byte with skb_pull_data() rfcomm_recv_data() treats the first payload byte as a credit field when the UIH frame carries PF and credit-based flow control is enabled. After the header has been stripped, the PF/CFC path consumes that byte with a direct skb->data dereference followed by skb_pull(). A malformed short frame can reach this path without a byte available. Use skb_pull_data() so the length check and pull happen together before the returned credit byte is consumed. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: Pengpeng Hou Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/rfcomm/core.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/bluetooth/rfcomm/core.c b/net/bluetooth/rfcomm/core.c index 611a9a94151e..d11bd5337d57 100644 --- a/net/bluetooth/rfcomm/core.c +++ b/net/bluetooth/rfcomm/core.c @@ -1715,9 +1715,12 @@ static int rfcomm_recv_data(struct rfcomm_session *s, u8 dlci, int pf, struct sk } if (pf && d->cfc) { - u8 credits = *(u8 *) skb->data; skb_pull(skb, 1); + u8 *credits = skb_pull_data(skb, 1); - d->tx_credits += credits; + if (!credits) + goto drop; + + d->tx_credits += *credits; if (d->tx_credits) clear_bit(RFCOMM_TX_THROTTLED, &d->flags); } -- cgit v1.2.3 From 72d97cae2a83cecf6f47208646675ecd066d0a3e Mon Sep 17 00:00:00 2001 From: Jann Horn Date: Wed, 29 Apr 2026 15:40:46 +0200 Subject: Bluetooth: hci_event: fix memset typo hci_le_big_sync_established_evt() currently does: conn->num_bis = 0; memset(conn->bis, 0, sizeof(conn->num_bis)); sizeof(conn->num_bis) is wrong - it would make sense to either use conn->num_bis (before setting that to 0) or sizeof(conn->bis). Fix it by using sizeof(conn->bis), the least intrusive change. Luckily, nothing actually depends on this memset() working properly: Nothing seems to ever read from conn->bis beyond conn->num_bis, and when conn->num_bis is increased, the corresponding elements of conn->bis are initialized. So I think this line could also just be removed. This is a purely theoretical fix and should have no impact on actual behavior. Fixes: 42ecf1947135 ("Bluetooth: ISO: Do not emit LE BIG Create Sync if previous is pending") Signed-off-by: Jann Horn Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/hci_event.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c index 1b3b9131affa..eea2f810aafa 100644 --- a/net/bluetooth/hci_event.c +++ b/net/bluetooth/hci_event.c @@ -7191,7 +7191,7 @@ static void hci_le_big_sync_established_evt(struct hci_dev *hdev, void *data, clear_bit(HCI_CONN_CREATE_BIG_SYNC, &conn->flags); conn->num_bis = 0; - memset(conn->bis, 0, sizeof(conn->num_bis)); + memset(conn->bis, 0, sizeof(conn->bis)); for (i = 0; i < ev->num_bis; i++) { u16 handle = le16_to_cpu(ev->bis[i]); -- cgit v1.2.3 From c5d415596cb6fbdf6334b06cc87a1a5a268d8725 Mon Sep 17 00:00:00 2001 From: Michael Bommarito Date: Sat, 2 May 2026 12:43:03 -0400 Subject: Bluetooth: HIDP: serialise l2cap_unregister_user via hidp_session_sem Commit dbf666e4fc9b ("Bluetooth: HIDP: Fix possible UAF") made hidp_session_remove() drop the L2CAP reference and set session->conn = NULL once the session is considered removed, and added a bare if (session->conn) guard around the kthread-exit l2cap_unregister_user() call in hidp_session_thread(). The sibling ioctl site in hidp_connection_del() still reads session->conn unlocked and unguarded, and the kthread-exit guard itself is a lockless double-read. hidp_session_find() drops hidp_session_sem before returning, so hidp_session_remove() can null session->conn between the lookup and the call in hidp_connection_del(). Worse, since commit 752a6c9596dd ("Bluetooth: L2CAP: Fix use-after-free in l2cap_unregister_user") takes mutex_lock(&conn->lock) inside l2cap_unregister_user(), a stale non-NULL snapshot also UAFs on conn->lock. v1 only added an if (session->conn) guard at the ioctl site, which doesn't address either race; Luiz suggested snapshotting session->conn under the sem and clearing it before the call. Taking hidp_session_sem across l2cap_unregister_user() would be wrong: l2cap_conn_del() already establishes the lock order conn->lock -> hidp_session_sem via l2cap_unregister_all_users() -> user->remove == hidp_session_remove(), so taking hidp_session_sem before conn->lock would AB/BA deadlock. Factor a helper hidp_session_unregister_conn() that under down_write(&hidp_session_sem) snapshots session->conn and clears the member, then outside the sem calls l2cap_unregister_user() and l2cap_conn_put() on the snapshot. Call it from both hidp_connection_del() and hidp_session_thread()'s exit path. At most one consumer wins the write-sem; later callers observe session->conn == NULL and skip the unregister and put, so the reference hidp_session_new() took via l2cap_conn_get() is consumed exactly once. session_free() already tolerates a NULL session->conn. Fixes: dbf666e4fc9b ("Bluetooth: HIDP: Fix possible UAF") Suggested-by: Luiz Augusto von Dentz Link: https://lore.kernel.org/all/20260422011437.176643-1-michael.bommarito@gmail.com/ Signed-off-by: Michael Bommarito Assisted-by: Claude:claude-opus-4-7 Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/hidp/core.c | 27 ++++++++++++++++++++++++--- 1 file changed, 24 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/bluetooth/hidp/core.c b/net/bluetooth/hidp/core.c index 7bcf8c5ceaee..976f91eeb745 100644 --- a/net/bluetooth/hidp/core.c +++ b/net/bluetooth/hidp/core.c @@ -1035,6 +1035,28 @@ static struct hidp_session *hidp_session_find(const bdaddr_t *bdaddr) return session; } +/* + * Consume session->conn: clear the member under hidp_session_sem, then + * l2cap_unregister_user() and l2cap_conn_put() the snapshot outside the + * sem. At most one caller wins; later callers see NULL and skip. The + * reference is the one hidp_session_new() took via l2cap_conn_get(). + */ +static void hidp_session_unregister_conn(struct hidp_session *session) +{ + struct l2cap_conn *conn; + + down_write(&hidp_session_sem); + conn = session->conn; + if (conn) + session->conn = NULL; + up_write(&hidp_session_sem); + + if (conn) { + l2cap_unregister_user(conn, &session->user); + l2cap_conn_put(conn); + } +} + /* * Start session synchronously * This starts a session thread and waits until initialization @@ -1311,8 +1333,7 @@ static int hidp_session_thread(void *arg) * Instead, this call has the same semantics as if user-space tried to * delete the session. */ - if (session->conn) - l2cap_unregister_user(session->conn, &session->user); + hidp_session_unregister_conn(session); hidp_session_put(session); @@ -1418,7 +1439,7 @@ int hidp_connection_del(struct hidp_conndel_req *req) HIDP_CTRL_VIRTUAL_CABLE_UNPLUG, NULL, 0); else - l2cap_unregister_user(session->conn, &session->user); + hidp_session_unregister_conn(session); hidp_session_put(session); -- cgit v1.2.3 From 7aaa8f5e45a92678256c1e17f1fa2c2f45c61dd1 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 5 May 2026 13:00:56 +0000 Subject: ipv6: fix potential UAF caused by ip6_forward_proxy_check() ip6_forward_proxy_check() calls pskb_may_pull() which might re-allocate skb->head. Reload ipv6_hdr() after the pskb_may_pull() call to avoid using the freed memory. Fixes: e21e0b5f19ac ("[IPV6] NDISC: Handle NDP messages to proxied addresses.") Reported-by: Damiano Melotti Signed-off-by: Eric Dumazet Reviewed-by: David Ahern Reviewed-by: Ido Schimmel Link: https://patch.msgid.link/20260505130056.2927197-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/ipv6/ip6_output.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'net') diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index 1f2a33fbed6e..c14adcdd4396 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -468,6 +468,7 @@ static int ip6_forward_proxy_check(struct sk_buff *skb) default: break; } + hdr = ipv6_hdr(skb); } /* @@ -582,6 +583,8 @@ int ip6_forward(struct sk_buff *skb) if (READ_ONCE(net->ipv6.devconf_all->proxy_ndp) && pneigh_lookup(&nd_tbl, net, &hdr->daddr, skb->dev)) { int proxied = ip6_forward_proxy_check(skb); + + hdr = ipv6_hdr(skb); if (proxied > 0) { /* It's tempting to decrease the hop limit * here by 1, as we do at the end of the -- cgit v1.2.3 From 67ef49047d312be692c8c439145f4514174e517f Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 5 May 2026 13:32:33 +0000 Subject: inetpeer: add a missing read_seqretry() in inet_getpeer() When performing a lockless lookup over the inet_peer rbtree, if a matching node is found, inet_getpeer() returns it immediately without validating the seqlock sequence. This missing check introduces a race condition: Trigger Path: When a host receives an incoming fragmented IPv4 packet, ip4_frag_init() (in net/ipv4/ip_fragment.c) calls inet_getpeer_v4() to track the peer. The Race: If the packet is from a new source IP, CPU A acquires the write_seqlock, allocates a new inet_peer node (p), sets its IP address (daddr), and links it to the rbtree (rb_link_node). Uninitialized Access: Due to the lack of memory barriers between rb_link_node and the initialization of the rest of the struct (like refcount_set(&p->refcnt, 1)), CPU A can make the node visible to readers before its refcnt is initialized. This is especially true on weakly-ordered architectures like ARM64 where the CPU can reorder the memory stores. Lockless Reader: Concurrently, CPU B processes a second fragmented packet from the same source IP. CPU B does a lockless lookup, finds the newly inserted node, and returns it immediately. Use-After-Free (UAF): CPU B reads p->refcnt as uninitialized garbage (left over from previous kmalloc-128/192 allocations). If the garbage is > 0, refcount_inc_not_zero(&p->refcnt) succeeds. CPU A then executes refcount_set(&p->refcnt, 1), overwriting CPU B's increment. When CPU B finishes with the fragment queue, it calls inet_putpeer(), which drops the refcount to 0 and frees the node via RCU. The node is now freed but remains linked in the rbtree, resulting in a Use-After-Free in the rbtree. Fixes: b145425f269a ("inetpeer: remove AVL implementation in favor of RB tree") Reported-by: Damiano Melotti Signed-off-by: Eric Dumazet Link: https://patch.msgid.link/20260505133233.3039575-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/ipv4/inetpeer.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv4/inetpeer.c b/net/ipv4/inetpeer.c index d8083b9033c2..5b957a831e7c 100644 --- a/net/ipv4/inetpeer.c +++ b/net/ipv4/inetpeer.c @@ -179,7 +179,8 @@ struct inet_peer *inet_getpeer(struct inet_peer_base *base, seq = read_seqbegin(&base->lock); p = lookup(daddr, base, seq, NULL, &gc_cnt, &parent, &pp); - if (p) + /* Make sure tree was not modified during our lookup. */ + if (p && !read_seqretry(&base->lock, seq)) return p; /* retry an exact lookup, taking the lock before. -- cgit v1.2.3 From 770b136ff9bf3e319d19875da59c4f7f4853da3a Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 5 May 2026 09:11:33 +0000 Subject: net/sched: sch_sfq: annotate data-races from sfq_dump_class_stats() sfq_dump_class_stats() runs locklessly, add needed READ_ONCE() and WRITE_ONCE() annotations. Fixes: edb09eb17ed8 ("net: sched: do not acquire qdisc spinlock in qdisc/class stats dump") Signed-off-by: Eric Dumazet Link: https://patch.msgid.link/20260505091133.2452510-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/sched/sch_sfq.c | 48 +++++++++++++++++++++++++----------------------- 1 file changed, 25 insertions(+), 23 deletions(-) (limited to 'net') diff --git a/net/sched/sch_sfq.c b/net/sched/sch_sfq.c index c3f3181dba54..f39822babf88 100644 --- a/net/sched/sch_sfq.c +++ b/net/sched/sch_sfq.c @@ -225,7 +225,8 @@ static inline void sfq_dec(struct sfq_sched_data *q, sfq_index x) sfq_unlink(q, x, n, p); - d = q->slots[x].qlen--; + d = q->slots[x].qlen; + WRITE_ONCE(q->slots[x].qlen, d - 1); if (n == p && q->cur_depth == d) q->cur_depth--; sfq_link(q, x); @@ -238,7 +239,8 @@ static inline void sfq_inc(struct sfq_sched_data *q, sfq_index x) sfq_unlink(q, x, n, p); - d = ++q->slots[x].qlen; + d = q->slots[x].qlen + 1; + WRITE_ONCE(q->slots[x].qlen, d); if (q->cur_depth < d) q->cur_depth = d; sfq_link(q, x); @@ -298,7 +300,7 @@ static unsigned int sfq_drop(struct Qdisc *sch, struct sk_buff **to_free) drop: skb = q->headdrop ? slot_dequeue_head(slot) : slot_dequeue_tail(slot); len = qdisc_pkt_len(skb); - slot->backlog -= len; + WRITE_ONCE(slot->backlog, slot->backlog - len); sfq_dec(q, x); sch->q.qlen--; qdisc_qstats_backlog_dec(sch, skb); @@ -314,7 +316,7 @@ drop: q->tail = NULL; /* no more active slots */ else q->tail->next = slot->next; - q->ht[slot->hash] = SFQ_EMPTY_SLOT; + WRITE_ONCE(q->ht[slot->hash], SFQ_EMPTY_SLOT); goto drop; } @@ -364,10 +366,10 @@ sfq_enqueue(struct sk_buff *skb, struct Qdisc *sch, struct sk_buff **to_free) x = q->dep[0].next; /* get a free slot */ if (x >= SFQ_MAX_FLOWS) return qdisc_drop_reason(skb, sch, to_free, QDISC_DROP_MAXFLOWS); - q->ht[hash] = x; + WRITE_ONCE(q->ht[hash], x); slot = &q->slots[x]; slot->hash = hash; - slot->backlog = 0; /* should already be 0 anyway... */ + WRITE_ONCE(slot->backlog, 0); /* should already be 0 anyway... */ red_set_vars(&slot->vars); goto enqueue; } @@ -426,7 +428,7 @@ congestion_drop: head = slot_dequeue_head(slot); delta = qdisc_pkt_len(head) - qdisc_pkt_len(skb); sch->qstats.backlog -= delta; - slot->backlog -= delta; + WRITE_ONCE(slot->backlog, slot->backlog - delta); qdisc_drop_reason(head, sch, to_free, QDISC_DROP_FLOW_LIMIT); slot_queue_add(slot, skb); @@ -436,7 +438,7 @@ congestion_drop: enqueue: qdisc_qstats_backlog_inc(sch, skb); - slot->backlog += qdisc_pkt_len(skb); + WRITE_ONCE(slot->backlog, slot->backlog + qdisc_pkt_len(skb)); slot_queue_add(slot, skb); sfq_inc(q, x); if (slot->qlen == 1) { /* The flow is new */ @@ -452,7 +454,7 @@ enqueue: */ q->tail = slot; /* We could use a bigger initial quantum for new flows */ - slot->allot = q->quantum; + WRITE_ONCE(slot->allot, q->quantum); } if (++sch->q.qlen <= q->limit) return NET_XMIT_SUCCESS; @@ -489,7 +491,7 @@ next_slot: slot = &q->slots[a]; if (slot->allot <= 0) { q->tail = slot; - slot->allot += q->quantum; + WRITE_ONCE(slot->allot, slot->allot + q->quantum); goto next_slot; } skb = slot_dequeue_head(slot); @@ -497,10 +499,10 @@ next_slot: qdisc_bstats_update(sch, skb); sch->q.qlen--; qdisc_qstats_backlog_dec(sch, skb); - slot->backlog -= qdisc_pkt_len(skb); + WRITE_ONCE(slot->backlog, slot->backlog - qdisc_pkt_len(skb)); /* Is the slot empty? */ if (slot->qlen == 0) { - q->ht[slot->hash] = SFQ_EMPTY_SLOT; + WRITE_ONCE(q->ht[slot->hash], SFQ_EMPTY_SLOT); next_a = slot->next; if (a == next_a) { q->tail = NULL; /* no more active slots */ @@ -508,7 +510,7 @@ next_slot: } q->tail->next = next_a; } else { - slot->allot -= qdisc_pkt_len(skb); + WRITE_ONCE(slot->allot, slot->allot - qdisc_pkt_len(skb)); } return skb; } @@ -549,9 +551,9 @@ static void sfq_rehash(struct Qdisc *sch) sfq_dec(q, i); __skb_queue_tail(&list, skb); } - slot->backlog = 0; + WRITE_ONCE(slot->backlog, 0); red_set_vars(&slot->vars); - q->ht[slot->hash] = SFQ_EMPTY_SLOT; + WRITE_ONCE(q->ht[slot->hash], SFQ_EMPTY_SLOT); } q->tail = NULL; @@ -570,7 +572,7 @@ drop: dropped++; continue; } - q->ht[hash] = x; + WRITE_ONCE(q->ht[hash], x); slot = &q->slots[x]; slot->hash = hash; } @@ -581,7 +583,7 @@ drop: slot->vars.qavg = red_calc_qavg(q->red_parms, &slot->vars, slot->backlog); - slot->backlog += qdisc_pkt_len(skb); + WRITE_ONCE(slot->backlog, slot->backlog + qdisc_pkt_len(skb)); sfq_inc(q, x); if (slot->qlen == 1) { /* The flow is new */ if (q->tail == NULL) { /* It is the first flow */ @@ -591,7 +593,7 @@ drop: q->tail->next = x; } q->tail = slot; - slot->allot = q->quantum; + WRITE_ONCE(slot->allot, q->quantum); } } sch->q.qlen -= dropped; @@ -905,16 +907,16 @@ static int sfq_dump_class_stats(struct Qdisc *sch, unsigned long cl, struct gnet_dump *d) { struct sfq_sched_data *q = qdisc_priv(sch); - sfq_index idx = q->ht[cl - 1]; + sfq_index idx = READ_ONCE(q->ht[cl - 1]); struct gnet_stats_queue qs = { 0 }; struct tc_sfq_xstats xstats = { 0 }; if (idx != SFQ_EMPTY_SLOT) { const struct sfq_slot *slot = &q->slots[idx]; - xstats.allot = slot->allot; - qs.qlen = slot->qlen; - qs.backlog = slot->backlog; + xstats.allot = READ_ONCE(slot->allot); + qs.qlen = READ_ONCE(slot->qlen); + qs.backlog = READ_ONCE(slot->backlog); } if (gnet_stats_copy_queue(d, NULL, &qs, qs.qlen) < 0) return -1; @@ -930,7 +932,7 @@ static void sfq_walk(struct Qdisc *sch, struct qdisc_walker *arg) return; for (i = 0; i < q->divisor; i++) { - if (q->ht[i] == SFQ_EMPTY_SLOT) { + if (READ_ONCE(q->ht[i]) == SFQ_EMPTY_SLOT) { arg->count++; continue; } -- cgit v1.2.3 From c8f7244c8cccaaed4e6c9fe4b8a07e101d0423e5 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 5 May 2026 15:39:27 +0000 Subject: tcp: tcp_child_process() related UAF tcp_child_process( .. child ...) currently calls sock_put(child). Unfortunately @child (named @nsk in callers) can be used after this point to send a RST packet. To fix this UAF, I remove the sock_put() from tcp_child_process() and let the callers handle this after it is safe. Remove @rsk variable in tcp_v4_do_rcv() and change tcp_v6_do_rcv() so that both functions look the same. Fixes: cfb6eeb4c860 ("[TCP]: MD5 Signature Option (RFC2385) support.") Reported-by: Damiano Melotti Signed-off-by: Eric Dumazet Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20260505153927.3435532-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/ipv4/tcp_ipv4.c | 14 ++++++-------- net/ipv4/tcp_minisocks.c | 2 +- net/ipv6/tcp_ipv6.c | 13 ++++++++----- 3 files changed, 15 insertions(+), 14 deletions(-) (limited to 'net') diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 8fc24c3743c5..c0526cc03980 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -1827,7 +1827,6 @@ INDIRECT_CALLABLE_DECLARE(struct dst_entry *ipv4_dst_check(struct dst_entry *, int tcp_v4_do_rcv(struct sock *sk, struct sk_buff *skb) { enum skb_drop_reason reason; - struct sock *rsk; reason = psp_sk_rx_policy_check(sk, skb); if (reason) @@ -1863,24 +1862,21 @@ int tcp_v4_do_rcv(struct sock *sk, struct sk_buff *skb) return 0; if (nsk != sk) { reason = tcp_child_process(sk, nsk, skb); - if (reason) { - rsk = nsk; + sock_put(nsk); + if (reason) goto reset; - } return 0; } } else sock_rps_save_rxhash(sk, skb); reason = tcp_rcv_state_process(sk, skb); - if (reason) { - rsk = sk; + if (reason) goto reset; - } return 0; reset: - tcp_v4_send_reset(rsk, skb, sk_rst_convert_drop_reason(reason)); + tcp_v4_send_reset(sk, skb, sk_rst_convert_drop_reason(reason)); discard: sk_skb_reason_drop(sk, skb, reason); /* Be careful here. If this function gets more complicated and @@ -2193,8 +2189,10 @@ lookup: rst_reason = sk_rst_convert_drop_reason(drop_reason); tcp_v4_send_reset(nsk, skb, rst_reason); + sock_put(nsk); goto discard_and_relse; } + sock_put(nsk); sock_put(sk); return 0; } diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index 199f0b579e89..e6092c3ac840 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -1012,6 +1012,6 @@ enum skb_drop_reason tcp_child_process(struct sock *parent, struct sock *child, } bh_unlock_sock(child); - sock_put(child); + return reason; } diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 2c3f7a739709..51583aef0643 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -1617,12 +1617,13 @@ int tcp_v6_do_rcv(struct sock *sk, struct sk_buff *skb) if (sk->sk_state == TCP_LISTEN) { struct sock *nsk = tcp_v6_cookie_check(sk, skb); + if (!nsk) + return 0; if (nsk != sk) { - if (nsk) { - reason = tcp_child_process(sk, nsk, skb); - if (reason) - goto reset; - } + reason = tcp_child_process(sk, nsk, skb); + sock_put(nsk); + if (reason) + goto reset; return 0; } } else @@ -1827,8 +1828,10 @@ lookup: rst_reason = sk_rst_convert_drop_reason(drop_reason); tcp_v6_send_reset(nsk, skb, rst_reason); + sock_put(nsk); goto discard_and_relse; } + sock_put(nsk); sock_put(sk); return 0; } -- cgit v1.2.3 From b12014d2d36eaed4e4bec5f1ac7e91110eeb100d Mon Sep 17 00:00:00 2001 From: "Matthieu Baerts (NGI0)" Date: Tue, 5 May 2026 17:00:49 +0200 Subject: mptcp: pm: kernel: correctly retransmit ADD_ADDR ID 0 When adding the ADD_ADDR to the list, the address including the IP, port and ID are copied. On the other hand, when the endpoint corresponds to the one from the initial subflow, the ID is set to 0, as specified by the MPTCP protocol. The issue is that the ID was reset after having copied the ID in the ADD_ADDR entry. So the retransmission was done, but using a different ID than the initial one. Fixes: 8b8ed1b429f8 ("mptcp: pm: reuse ID 0 after delete and re-add") Cc: stable@vger.kernel.org Reviewed-by: Mat Martineau Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20260505-net-mptcp-pm-fixes-7-1-rc3-v1-1-fca8091060a4@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/pm_kernel.c | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/mptcp/pm_kernel.c b/net/mptcp/pm_kernel.c index c9f1e5af3cd3..fc818b63752e 100644 --- a/net/mptcp/pm_kernel.c +++ b/net/mptcp/pm_kernel.c @@ -347,6 +347,8 @@ static void mptcp_pm_create_subflow_or_signal_addr(struct mptcp_sock *msk) /* check first for announce */ if (msk->pm.add_addr_signaled < endp_signal_max) { + u8 endp_id; + /* due to racing events on both ends we can reach here while * previous add address is still running: if we invoke now * mptcp_pm_announce_addr(), that will fail and the @@ -360,19 +362,20 @@ static void mptcp_pm_create_subflow_or_signal_addr(struct mptcp_sock *msk) if (!select_signal_address(pernet, msk, &local)) goto subflow; + /* Special case for ID0: set the correct ID */ + endp_id = local.addr.id; + if (endp_id == msk->mpc_endpoint_id) + local.addr.id = 0; + /* If the alloc fails, we are on memory pressure, not worth * continuing, and trying to create subflows. */ if (!mptcp_pm_alloc_anno_list(msk, &local.addr)) return; - __clear_bit(local.addr.id, msk->pm.id_avail_bitmap); + __clear_bit(endp_id, msk->pm.id_avail_bitmap); msk->pm.add_addr_signaled++; - /* Special case for ID0: set the correct ID */ - if (local.addr.id == msk->mpc_endpoint_id) - local.addr.id = 0; - mptcp_pm_announce_addr(msk, &local.addr, false); mptcp_pm_addr_send_ack(msk); -- cgit v1.2.3 From 03f324f3f1f7619a47b9c91282cb12775ab0a2f1 Mon Sep 17 00:00:00 2001 From: "Matthieu Baerts (NGI0)" Date: Tue, 5 May 2026 17:00:50 +0200 Subject: mptcp: pm: ADD_ADDR rtx: allow ID 0 ADD_ADDR can be sent for the ID 0, which corresponds to the local address and port linked to the initial subflow. Indeed, this address could be removed, and re-added later on, e.g. what is done in the "delete re-add signal" MPTCP Join selftests. So no reason to ignore it. Fixes: 00cfd77b9063 ("mptcp: retransmit ADD_ADDR when timeout") Cc: stable@vger.kernel.org Reviewed-by: Mat Martineau Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20260505-net-mptcp-pm-fixes-7-1-rc3-v1-2-fca8091060a4@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/pm.c | 3 --- 1 file changed, 3 deletions(-) (limited to 'net') diff --git a/net/mptcp/pm.c b/net/mptcp/pm.c index 57a456690406..5056eb8db24e 100644 --- a/net/mptcp/pm.c +++ b/net/mptcp/pm.c @@ -337,9 +337,6 @@ static void mptcp_pm_add_timer(struct timer_list *timer) if (inet_sk_state_load(sk) == TCP_CLOSE) return; - if (!entry->addr.id) - return; - if (mptcp_pm_should_add_signal_addr(msk)) { sk_reset_timer(sk, timer, jiffies + TCP_RTO_MAX / 8); goto out; -- cgit v1.2.3 From 5cd6e0ad79d2615264f63929f8b457ad97ae550d Mon Sep 17 00:00:00 2001 From: "Matthieu Baerts (NGI0)" Date: Tue, 5 May 2026 17:00:51 +0200 Subject: mptcp: pm: ADD_ADDR rtx: fix potential data-race This mptcp_pm_add_timer() helper is executed as a timer callback in softirq context. To avoid any data races, the socket lock needs to be held with bh_lock_sock(). If the socket is in use, retry again soon after, similar to what is done with the keepalive timer. Fixes: 00cfd77b9063 ("mptcp: retransmit ADD_ADDR when timeout") Cc: stable@vger.kernel.org Reviewed-by: Mat Martineau Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20260505-net-mptcp-pm-fixes-7-1-rc3-v1-3-fca8091060a4@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/pm.c | 8 ++++++++ 1 file changed, 8 insertions(+) (limited to 'net') diff --git a/net/mptcp/pm.c b/net/mptcp/pm.c index 5056eb8db24e..3912128d9b86 100644 --- a/net/mptcp/pm.c +++ b/net/mptcp/pm.c @@ -337,6 +337,13 @@ static void mptcp_pm_add_timer(struct timer_list *timer) if (inet_sk_state_load(sk) == TCP_CLOSE) return; + bh_lock_sock(sk); + if (sock_owned_by_user(sk)) { + /* Try again later. */ + sk_reset_timer(sk, timer, jiffies + HZ / 20); + goto out; + } + if (mptcp_pm_should_add_signal_addr(msk)) { sk_reset_timer(sk, timer, jiffies + TCP_RTO_MAX / 8); goto out; @@ -365,6 +372,7 @@ static void mptcp_pm_add_timer(struct timer_list *timer) mptcp_pm_subflow_established(msk); out: + bh_unlock_sock(sk); __sock_put(sk); } -- cgit v1.2.3 From 9634cb35af17019baec21ca648516ce376fa10e6 Mon Sep 17 00:00:00 2001 From: "Matthieu Baerts (NGI0)" Date: Tue, 5 May 2026 17:00:52 +0200 Subject: mptcp: pm: ADD_ADDR rtx: always decrease sk refcount When an ADD_ADDR is retransmitted, the sk is held in sk_reset_timer(). It should then be released in all cases at the end. Some (unlikely) checks were returning directly instead of calling sock_put() to decrease the refcount. Jump to a new 'exit' label to call __sock_put() (which will become sock_put() in the next commit) to fix this potential leak. While at it, drop the '!msk' check which cannot happen because it is never reset, and explicitly mark the remaining one as "unlikely". Fixes: 00cfd77b9063 ("mptcp: retransmit ADD_ADDR when timeout") Cc: stable@vger.kernel.org Reviewed-by: Mat Martineau Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20260505-net-mptcp-pm-fixes-7-1-rc3-v1-4-fca8091060a4@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/pm.c | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/mptcp/pm.c b/net/mptcp/pm.c index 3912128d9b86..2a01bf1b5bfd 100644 --- a/net/mptcp/pm.c +++ b/net/mptcp/pm.c @@ -331,11 +331,8 @@ static void mptcp_pm_add_timer(struct timer_list *timer) pr_debug("msk=%p\n", msk); - if (!msk) - return; - - if (inet_sk_state_load(sk) == TCP_CLOSE) - return; + if (unlikely(inet_sk_state_load(sk) == TCP_CLOSE)) + goto exit; bh_lock_sock(sk); if (sock_owned_by_user(sk)) { @@ -373,6 +370,7 @@ static void mptcp_pm_add_timer(struct timer_list *timer) out: bh_unlock_sock(sk); +exit: __sock_put(sk); } -- cgit v1.2.3 From b7b9a461569734d33d3259d58d2507adfac107ed Mon Sep 17 00:00:00 2001 From: "Matthieu Baerts (NGI0)" Date: Tue, 5 May 2026 17:00:53 +0200 Subject: mptcp: pm: ADD_ADDR rtx: free sk if last When an ADD_ADDR is retransmitted, the sk is held in sk_reset_timer(), and released at the end. If at that moment, it was the last reference being held, the sk would not be freed. sock_put() should then be called instead of __sock_put(). But that's not enough: if it is the last reference, sock_put() will call sk_free(), which will end up calling sk_stop_timer_sync() on the same timer, and waiting indefinitely to finish. So it is needed to mark that the timer is done at the end of the timer handler when it has not been rescheduled, not to call sk_stop_timer_sync() on "itself". Fixes: 00cfd77b9063 ("mptcp: retransmit ADD_ADDR when timeout") Cc: stable@vger.kernel.org Reviewed-by: Mat Martineau Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20260505-net-mptcp-pm-fixes-7-1-rc3-v1-5-fca8091060a4@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/pm.c | 28 ++++++++++++++++++---------- 1 file changed, 18 insertions(+), 10 deletions(-) (limited to 'net') diff --git a/net/mptcp/pm.c b/net/mptcp/pm.c index 2a01bf1b5bfd..8899327e59a1 100644 --- a/net/mptcp/pm.c +++ b/net/mptcp/pm.c @@ -16,6 +16,7 @@ struct mptcp_pm_add_entry { struct list_head list; struct mptcp_addr_info addr; u8 retrans_times; + bool timer_done; struct timer_list add_timer; struct mptcp_sock *sock; struct rcu_head rcu; @@ -327,22 +328,22 @@ static void mptcp_pm_add_timer(struct timer_list *timer) add_timer); struct mptcp_sock *msk = entry->sock; struct sock *sk = (struct sock *)msk; - unsigned int timeout; + unsigned int timeout = 0; pr_debug("msk=%p\n", msk); + bh_lock_sock(sk); if (unlikely(inet_sk_state_load(sk) == TCP_CLOSE)) - goto exit; + goto out; - bh_lock_sock(sk); if (sock_owned_by_user(sk)) { /* Try again later. */ - sk_reset_timer(sk, timer, jiffies + HZ / 20); + timeout = HZ / 20; goto out; } if (mptcp_pm_should_add_signal_addr(msk)) { - sk_reset_timer(sk, timer, jiffies + TCP_RTO_MAX / 8); + timeout = TCP_RTO_MAX / 8; goto out; } @@ -360,8 +361,9 @@ static void mptcp_pm_add_timer(struct timer_list *timer) } if (entry->retrans_times < ADD_ADDR_RETRANS_MAX) - sk_reset_timer(sk, timer, - jiffies + (timeout << entry->retrans_times)); + timeout <<= entry->retrans_times; + else + timeout = 0; spin_unlock_bh(&msk->pm.lock); @@ -369,9 +371,13 @@ static void mptcp_pm_add_timer(struct timer_list *timer) mptcp_pm_subflow_established(msk); out: + if (timeout) + sk_reset_timer(sk, timer, jiffies + timeout); + else + /* if sock_put calls sk_free: avoid waiting for this timer */ + entry->timer_done = true; bh_unlock_sock(sk); -exit: - __sock_put(sk); + sock_put(sk); } struct mptcp_pm_add_entry * @@ -434,6 +440,7 @@ bool mptcp_pm_alloc_anno_list(struct mptcp_sock *msk, timer_setup(&add_entry->add_timer, mptcp_pm_add_timer, 0); reset_timer: + add_entry->timer_done = false; timeout = mptcp_adjust_add_addr_timeout(msk); if (timeout) sk_reset_timer(sk, &add_entry->add_timer, jiffies + timeout); @@ -454,7 +461,8 @@ static void mptcp_pm_free_anno_list(struct mptcp_sock *msk) spin_unlock_bh(&msk->pm.lock); list_for_each_entry_safe(entry, tmp, &free_list, list) { - sk_stop_timer_sync(sk, &entry->add_timer); + if (!entry->timer_done) + sk_stop_timer_sync(sk, &entry->add_timer); kfree_rcu(entry, rcu); } } -- cgit v1.2.3 From 3cf12492891c4b5ff54dda404a2de4ec54c9e1b5 Mon Sep 17 00:00:00 2001 From: "Matthieu Baerts (NGI0)" Date: Tue, 5 May 2026 17:00:54 +0200 Subject: mptcp: pm: ADD_ADDR rtx: resched blocked ADD_ADDR quicker When an ADD_ADDR needs to be retransmitted and another one has already been prepared -- e.g. multiple ADD_ADDRs have been sent in a row and need to be retransmitted later -- this additional retransmission will need to wait. In this case, the timer was reset to TCP_RTO_MAX / 8, which is ~15 seconds. This delay is unnecessary long: it should just be rescheduled at the next opportunity, e.g. after the retransmission timeout. Without this modification, some issues can be seen from time to time in the selftests when multiple ADD_ADDRs are sent, and the host takes time to process them, e.g. the "signal addresses, ADD_ADDR timeout" MPTCP Join selftest, especially with a debug kernel config. Note that on older kernels, 'timeout' is not available. It should be enough to replace it by one second (HZ). Fixes: 00cfd77b9063 ("mptcp: retransmit ADD_ADDR when timeout") Cc: stable@vger.kernel.org Reviewed-by: Mat Martineau Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20260505-net-mptcp-pm-fixes-7-1-rc3-v1-6-fca8091060a4@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/pm.c | 7 +------ 1 file changed, 1 insertion(+), 6 deletions(-) (limited to 'net') diff --git a/net/mptcp/pm.c b/net/mptcp/pm.c index 8899327e59a1..29d1bb6a69cf 100644 --- a/net/mptcp/pm.c +++ b/net/mptcp/pm.c @@ -342,13 +342,8 @@ static void mptcp_pm_add_timer(struct timer_list *timer) goto out; } - if (mptcp_pm_should_add_signal_addr(msk)) { - timeout = TCP_RTO_MAX / 8; - goto out; - } - timeout = mptcp_adjust_add_addr_timeout(msk); - if (!timeout) + if (!timeout || mptcp_pm_should_add_signal_addr(msk)) goto out; spin_lock_bh(&msk->pm.lock); -- cgit v1.2.3 From c6d395e2de1306b5fef0344a3c3835fbbfaa18be Mon Sep 17 00:00:00 2001 From: "Matthieu Baerts (NGI0)" Date: Tue, 5 May 2026 17:00:55 +0200 Subject: mptcp: pm: ADD_ADDR rtx: skip inactive subflows When looking at the maximum RTO amongst the subflows, inactive subflows were taken into account: that includes stale ones, and the initial one if it has been already been closed. Unusable subflows are now simply skipped. Stale ones are used as an alternative: if there are only stale ones, to take their maximum RTO and avoid to eventually fallback to net.mptcp.add_addr_timeout, which is set to 2 minutes by default. Fixes: 30549eebc4d8 ("mptcp: make ADD_ADDR retransmission timeout adaptive") Cc: stable@vger.kernel.org Reviewed-by: Mat Martineau Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20260505-net-mptcp-pm-fixes-7-1-rc3-v1-7-fca8091060a4@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/pm.c | 18 ++++++++++++++---- 1 file changed, 14 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/mptcp/pm.c b/net/mptcp/pm.c index 29d1bb6a69cf..8a5dba7fe66e 100644 --- a/net/mptcp/pm.c +++ b/net/mptcp/pm.c @@ -306,18 +306,28 @@ static unsigned int mptcp_adjust_add_addr_timeout(struct mptcp_sock *msk) const struct net *net = sock_net((struct sock *)msk); unsigned int rto = mptcp_get_add_addr_timeout(net); struct mptcp_subflow_context *subflow; - unsigned int max = 0; + unsigned int max = 0, max_stale = 0; mptcp_for_each_subflow(msk, subflow) { struct sock *ssk = mptcp_subflow_tcp_sock(subflow); struct inet_connection_sock *icsk = inet_csk(ssk); - if (icsk->icsk_rto > max) + if (!__mptcp_subflow_active(subflow)) + continue; + + if (unlikely(subflow->stale)) { + if (icsk->icsk_rto > max_stale) + max_stale = icsk->icsk_rto; + } else if (icsk->icsk_rto > max) { max = icsk->icsk_rto; + } } - if (max && max < rto) - rto = max; + if (max) + return min(max, rto); + + if (max_stale) + return min(max_stale, rto); return rto; } -- cgit v1.2.3 From 62a9b19dce77e72426f049fb99b9d1d032b9a8ea Mon Sep 17 00:00:00 2001 From: "Matthieu Baerts (NGI0)" Date: Tue, 5 May 2026 17:00:56 +0200 Subject: mptcp: pm: ADD_ADDR rtx: return early if no retrans No need to iterate over all subflows if there is no retransmission needed. Exit early in this case then. Fixes: 30549eebc4d8 ("mptcp: make ADD_ADDR retransmission timeout adaptive") Cc: stable@vger.kernel.org Reviewed-by: Mat Martineau Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20260505-net-mptcp-pm-fixes-7-1-rc3-v1-8-fca8091060a4@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/pm.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'net') diff --git a/net/mptcp/pm.c b/net/mptcp/pm.c index 8a5dba7fe66e..4a6e5ab30d80 100644 --- a/net/mptcp/pm.c +++ b/net/mptcp/pm.c @@ -308,6 +308,9 @@ static unsigned int mptcp_adjust_add_addr_timeout(struct mptcp_sock *msk) struct mptcp_subflow_context *subflow; unsigned int max = 0, max_stale = 0; + if (!rto) + return 0; + mptcp_for_each_subflow(msk, subflow) { struct sock *ssk = mptcp_subflow_tcp_sock(subflow); struct inet_connection_sock *icsk = inet_csk(ssk); -- cgit v1.2.3 From 166b78344031bf7ac9f55cb5282776cfd85f220e Mon Sep 17 00:00:00 2001 From: "Matthieu Baerts (NGI0)" Date: Tue, 5 May 2026 17:00:57 +0200 Subject: mptcp: pm: prio: skip closed subflows When sending an MP_PRIO, closed subflows need to be skipped. This fixes the case where the initial subflow got closed, re-opened later, then an MP_PRIO is needed for the same local address. Note that explicit MP_PRIO cannot be sent during the 3WHS, so it is fine to use __mptcp_subflow_active(). Fixes: 067065422fcd ("mptcp: add the outgoing MP_PRIO support") Cc: stable@vger.kernel.org Fixes: b29fcfb54cd7 ("mptcp: full disconnect implementation") Reviewed-by: Mat Martineau Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20260505-net-mptcp-pm-fixes-7-1-rc3-v1-9-fca8091060a4@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/pm.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'net') diff --git a/net/mptcp/pm.c b/net/mptcp/pm.c index 4a6e5ab30d80..3c152bf66cd5 100644 --- a/net/mptcp/pm.c +++ b/net/mptcp/pm.c @@ -284,6 +284,9 @@ int mptcp_pm_mp_prio_send_ack(struct mptcp_sock *msk, struct sock *ssk = mptcp_subflow_tcp_sock(subflow); struct mptcp_addr_info local, remote; + if (!__mptcp_subflow_active(subflow)) + continue; + mptcp_local_address((struct sock_common *)ssk, &local); if (!mptcp_addresses_equal(&local, addr, addr->port)) continue; -- cgit v1.2.3 From 7e2a4f7ca0952820731ef7bdadfc9a9e9d3571b4 Mon Sep 17 00:00:00 2001 From: Maoyi Xie Date: Mon, 4 May 2026 22:27:36 +0800 Subject: xfrm: route MIGRATE notifications to caller's netns xfrm_send_migrate() in net/xfrm/xfrm_user.c and pfkey_send_migrate() in net/key/af_key.c both hardcode &init_net for the multicast that announces a successful XFRM_MSG_MIGRATE / SADB_X_MIGRATE. XFRM_MSG_MIGRATE arrives on a per-netns NETLINK_XFRM socket, and the rest of the xfrm/af_key netlink path was made netns-aware in 2008. The other 14 multicast paths in xfrm_user.c route their event using xs_net(x), xp_net(xp) or sock_net(skb->sk); only the migrate path was missed. Two consequences of the init_net hardcoding: 1. The notification (selector, old/new endpoint addresses, and the km_address) is delivered to listeners on init_net's XFRMNLGRP_MIGRATE / pfkey BROADCAST_ALL groups rather than on the issuing netns. An IKE daemon running in init_net therefore receives migration notifications originating from any other netns on the host. 2. An IKE daemon running inside a non-init netns and subscribed to its own XFRMNLGRP_MIGRATE / pfkey groups never receives the notification of its own migration. IKEv2 MOBIKE / address-update handling inside a netns is silently broken. Thread struct net through km_migrate() and the xfrm_mgr.migrate function pointer, drop the &init_net override in xfrm_send_migrate() and pfkey_send_migrate(), and pass the caller's net (already in scope in xfrm_migrate() via sock_net(skb->sk)) all the way down. struct xfrm_mgr is in-tree only and not exported as a stable API, so the function-pointer signature change is internal. pfkey_broadcast() is already netns-aware via net_generic(net, pfkey_net_id) since the pernet conversion. The five other pfkey_broadcast() callers in af_key.c already pass xs_net(x), sock_net(sk) or a per-netns net, so this only removes the &init_net outlier. Fixes: 5c79de6e79cd ("[XFRM]: User interface for handling XFRM_MSG_MIGRATE") Cc: stable@vger.kernel.org # v5.15+ Signed-off-by: Maoyi Xie Signed-off-by: Steffen Klassert --- net/key/af_key.c | 6 +++--- net/xfrm/xfrm_policy.c | 2 +- net/xfrm/xfrm_state.c | 4 ++-- net/xfrm/xfrm_user.c | 5 ++--- 4 files changed, 8 insertions(+), 9 deletions(-) (limited to 'net') diff --git a/net/key/af_key.c b/net/key/af_key.c index a166a88d8788..9cffeef18cd9 100644 --- a/net/key/af_key.c +++ b/net/key/af_key.c @@ -3564,7 +3564,7 @@ static int set_ipsecrequest(struct sk_buff *skb, #ifdef CONFIG_NET_KEY_MIGRATE static int pfkey_send_migrate(const struct xfrm_selector *sel, u8 dir, u8 type, const struct xfrm_migrate *m, int num_bundles, - const struct xfrm_kmaddress *k, + const struct xfrm_kmaddress *k, struct net *net, const struct xfrm_encap_tmpl *encap) { int i; @@ -3669,7 +3669,7 @@ static int pfkey_send_migrate(const struct xfrm_selector *sel, u8 dir, u8 type, } /* broadcast migrate message to sockets */ - pfkey_broadcast(skb, GFP_ATOMIC, BROADCAST_ALL, NULL, &init_net); + pfkey_broadcast(skb, GFP_ATOMIC, BROADCAST_ALL, NULL, net); return 0; @@ -3680,7 +3680,7 @@ err: #else static int pfkey_send_migrate(const struct xfrm_selector *sel, u8 dir, u8 type, const struct xfrm_migrate *m, int num_bundles, - const struct xfrm_kmaddress *k, + const struct xfrm_kmaddress *k, struct net *net, const struct xfrm_encap_tmpl *encap) { return -ENOPROTOOPT; diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index c944327ce66c..59968dcbafe1 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -4703,7 +4703,7 @@ int xfrm_migrate(const struct xfrm_selector *sel, u8 dir, u8 type, } /* Stage 5 - announce */ - km_migrate(sel, dir, type, m, num_migrate, k, encap); + km_migrate(sel, dir, type, m, num_migrate, k, net, encap); xfrm_pol_put(pol); diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c index 686014d39429..395d82411a87 100644 --- a/net/xfrm/xfrm_state.c +++ b/net/xfrm/xfrm_state.c @@ -2837,7 +2837,7 @@ EXPORT_SYMBOL(km_policy_expired); #ifdef CONFIG_XFRM_MIGRATE int km_migrate(const struct xfrm_selector *sel, u8 dir, u8 type, const struct xfrm_migrate *m, int num_migrate, - const struct xfrm_kmaddress *k, + const struct xfrm_kmaddress *k, struct net *net, const struct xfrm_encap_tmpl *encap) { int err = -EINVAL; @@ -2848,7 +2848,7 @@ int km_migrate(const struct xfrm_selector *sel, u8 dir, u8 type, list_for_each_entry_rcu(km, &xfrm_km_list, list) { if (km->migrate) { ret = km->migrate(sel, dir, type, m, num_migrate, k, - encap); + net, encap); if (!ret) err = ret; } diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index 38a90e5ee3d9..71a4b7278eba 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -3271,10 +3271,9 @@ out_cancel: static int xfrm_send_migrate(const struct xfrm_selector *sel, u8 dir, u8 type, const struct xfrm_migrate *m, int num_migrate, - const struct xfrm_kmaddress *k, + const struct xfrm_kmaddress *k, struct net *net, const struct xfrm_encap_tmpl *encap) { - struct net *net = &init_net; struct sk_buff *skb; int err; @@ -3292,7 +3291,7 @@ static int xfrm_send_migrate(const struct xfrm_selector *sel, u8 dir, u8 type, #else static int xfrm_send_migrate(const struct xfrm_selector *sel, u8 dir, u8 type, const struct xfrm_migrate *m, int num_migrate, - const struct xfrm_kmaddress *k, + const struct xfrm_kmaddress *k, struct net *net, const struct xfrm_encap_tmpl *encap) { return -ENOPROTOOPT; -- cgit v1.2.3 From f040e590c035bfd9553fe79ee9585caf1b14d67b Mon Sep 17 00:00:00 2001 From: Ashutosh Desai Date: Tue, 5 May 2026 17:07:12 +0000 Subject: nfc: hci: fix out-of-bounds read in HCP header parsing Both nfc_hci_recv_from_llc() and nci_hci_data_received_cb() read packet->header from skb->data at function entry without first checking that the buffer holds at least one byte. A malicious NFC peer can send a 0-byte HCP frame that passes through the SHDLC layer and reaches these functions, causing an out-of-bounds heap read of packet->header. The same 0-byte frame, if queued as a non-final fragment, also causes the reassembly loop to underflow msg_len to UINT_MAX, triggering skb_over_panic() when the reassembled skb is written. Fix this by adding a pskb_may_pull() check at the entry of each function before packet->header is first accessed. The existing pskb_may_pull() checks before the reassembled hcp_skb is cast to struct hcp_packet remain in place to guard the 2-byte HCP message header. Fixes: 8b8d2e08bf0d ("NFC: HCI support") Fixes: 11f54f228643 ("NFC: nci: Add HCI over NCI protocol support") Cc: stable@vger.kernel.org Reviewed-by: Simon Horman Signed-off-by: Ashutosh Desai Link: https://patch.msgid.link/20260505170712.96560-1-ashutoshdesai993@gmail.com Signed-off-by: David Heidelberg --- net/nfc/hci/core.c | 10 ++++++++++ net/nfc/nci/hci.c | 10 ++++++++++ 2 files changed, 20 insertions(+) (limited to 'net') diff --git a/net/nfc/hci/core.c b/net/nfc/hci/core.c index 0d33c81a15fe..ba6f0310ffd7 100644 --- a/net/nfc/hci/core.c +++ b/net/nfc/hci/core.c @@ -861,6 +861,11 @@ static void nfc_hci_recv_from_llc(struct nfc_hci_dev *hdev, struct sk_buff *skb) struct sk_buff *frag_skb; int msg_len; + if (!pskb_may_pull(skb, NFC_HCI_HCP_PACKET_HEADER_LEN)) { + kfree_skb(skb); + return; + } + packet = (struct hcp_packet *)skb->data; if ((packet->header & ~NFC_HCI_FRAGMENT) == 0) { skb_queue_tail(&hdev->rx_hcp_frags, skb); @@ -904,6 +909,11 @@ static void nfc_hci_recv_from_llc(struct nfc_hci_dev *hdev, struct sk_buff *skb) * unblock waiting cmd context. Otherwise, enqueue to dispatch * in separate context where handler can also execute command. */ + if (!pskb_may_pull(hcp_skb, NFC_HCI_HCP_HEADER_LEN)) { + kfree_skb(hcp_skb); + return; + } + packet = (struct hcp_packet *)hcp_skb->data; type = HCP_MSG_GET_TYPE(packet->message.header); if (type == NFC_HCI_HCP_RESPONSE) { diff --git a/net/nfc/nci/hci.c b/net/nfc/nci/hci.c index 40ae8e5a7ec7..c03e8a0bd3bd 100644 --- a/net/nfc/nci/hci.c +++ b/net/nfc/nci/hci.c @@ -439,6 +439,11 @@ void nci_hci_data_received_cb(void *context, return; } + if (!pskb_may_pull(skb, NCI_HCI_HCP_PACKET_HEADER_LEN)) { + kfree_skb(skb); + return; + } + packet = (struct nci_hcp_packet *)skb->data; if ((packet->header & ~NCI_HCI_FRAGMENT) == 0) { skb_queue_tail(&ndev->hci_dev->rx_hcp_frags, skb); @@ -482,6 +487,11 @@ void nci_hci_data_received_cb(void *context, * unblock waiting cmd context. Otherwise, enqueue to dispatch * in separate context where handler can also execute command. */ + if (!pskb_may_pull(hcp_skb, NCI_HCI_HCP_HEADER_LEN)) { + kfree_skb(hcp_skb); + return; + } + packet = (struct nci_hcp_packet *)hcp_skb->data; type = NCI_HCP_MSG_GET_TYPE(packet->message.header); if (type == NCI_HCI_HCP_RESPONSE) { -- cgit v1.2.3 From d119775f2bad827edc28071c061fdd4a91f889a5 Mon Sep 17 00:00:00 2001 From: Jiexun Wang Date: Wed, 6 May 2026 22:08:23 +0800 Subject: af_unix: Reject SIOCATMARK on non-stream sockets SIOCATMARK reports whether the receive queue is at the urgent mark for MSG_OOB. In AF_UNIX, MSG_OOB is supported only for SOCK_STREAM sockets. SOCK_DGRAM and SOCK_SEQPACKET reject MSG_OOB in sendmsg() and recvmsg(), so they should not support SIOCATMARK either. Return -EOPNOTSUPP for non-stream sockets before checking the receive queue. Fixes: 314001f0bf92 ("af_unix: Add OOB support") Cc: stable@kernel.org Reported-by: Yuan Tan Reported-by: Yifan Wu Reported-by: Juefei Pu Reported-by: Xin Liu Suggested-by: Kuniyuki Iwashima Signed-off-by: Jiexun Wang Signed-off-by: Ren Wei Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20260506140825.2987635-1-n05ec@lzu.edu.cn Signed-off-by: Jakub Kicinski --- net/unix/af_unix.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'net') diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index e2d787ca3e74..1cbf36ea043b 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -3323,6 +3323,9 @@ static int unix_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg) struct sk_buff *skb; int answ = 0; + if (sk->sk_type != SOCK_STREAM) + return -EOPNOTSUPP; + mutex_lock(&u->iolock); skb = skb_peek(&sk->sk_receive_queue); -- cgit v1.2.3 From 9032f7676935a13fd402608223d326c5f62da9c0 Mon Sep 17 00:00:00 2001 From: "D. Wythe" Date: Wed, 6 May 2026 09:41:05 +0800 Subject: net/smc: fix missing sk_err when TCP handshake fails In smc_connect_work(), when the underlying TCP handshake fails, the error code (rc) must be propagated to sk_err to ensure userspace can correctly retrieve the error status via SO_ERROR. Currently, the code only handles a restricted set of error codes (e.g., EPIPE, ECONNREFUSED). If other errors occurs, such as EHOSTUNREACH, sk_err remains unset (zero). This affects applications that rely on SO_ERROR to determine connect outcome. For example, higher versions of Go's netpoller treats SO_ERROR == 0 combined with a failed getpeername() as a spurious wakeup and re-enters epoll_wait(). Under ET mode, no further edge will be generated since the socket is already in a terminal state, causing the connect to hang indefinitely or until a user-specified timeout, if one is set. Fixes: 50717a37db03 ("net/smc: nonblocking connect rework") Signed-off-by: D. Wythe Reviewed-by: Dust Li Link: https://patch.msgid.link/20260506014105.27093-1-alibuda@linux.alibaba.com Signed-off-by: Jakub Kicinski --- net/smc/af_smc.c | 8 ++------ 1 file changed, 2 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/net/smc/af_smc.c b/net/smc/af_smc.c index 1a565095376a..185dbed7de5d 100644 --- a/net/smc/af_smc.c +++ b/net/smc/af_smc.c @@ -1628,12 +1628,8 @@ static void smc_connect_work(struct work_struct *work) lock_sock(&smc->sk); if (rc != 0 || smc->sk.sk_err) { smc->sk.sk_state = SMC_CLOSED; - if (rc == -EPIPE || rc == -EAGAIN) - smc->sk.sk_err = EPIPE; - else if (rc == -ECONNREFUSED) - smc->sk.sk_err = ECONNREFUSED; - else if (signal_pending(current)) - smc->sk.sk_err = -sock_intr_errno(timeo); + if (!smc->sk.sk_err) + smc->sk.sk_err = (rc == -EAGAIN) ? EPIPE : -rc; sock_put(&smc->sk); /* passive closing */ goto out; } -- cgit v1.2.3 From 019c892e46544af0ae94ec833f79aa903c837666 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Wed, 6 May 2026 06:59:53 +0000 Subject: ipmr: Call ipmr_fib_lookup() under RCU. Yi Lai reported RCU splat in reg_vif_xmit() below. [0] When CONFIG_IP_MROUTE_MULTIPLE_TABLES=n, ipmr_fib_lookup() uses rcu_dereference() without explicit rcu_read_lock(). Although rcu_read_lock_bh() is already held by the caller __dev_queue_xmit(), lockdep requires explicit rcu_read_lock() for rcu_dereference(). Let's move up rcu_read_lock() in reg_vif_xmit() to cover ipmr_fib_lookup(). [0]: WARNING: suspicious RCU usage 7.1.0-rc2-next-20260504-9d0d467c3572 #1 Not tainted ----------------------------- net/ipv4/ipmr.c:329 suspicious rcu_dereference_check() usage! other info that might help us debug this: rcu_scheduler_active = 2, debug_locks = 1 2 locks held by syz.2.17/1779: #0: ffffffff87896440 (rcu_read_lock_bh){....}-{1:3}, at: local_bh_disable include/linux/bottom_half.h:20 [inline] #0: ffffffff87896440 (rcu_read_lock_bh){....}-{1:3}, at: rcu_read_lock_bh include/linux/rcupdate.h:891 [inline] #0: ffffffff87896440 (rcu_read_lock_bh){....}-{1:3}, at: __dev_queue_xmit+0x239/0x4140 net/core/dev.c:4792 #1: ffff88801a199d18 (_xmit_PIMREG#2){+...}-{3:3}, at: spin_lock include/linux/spinlock.h:342 [inline] #1: ffff88801a199d18 (_xmit_PIMREG#2){+...}-{3:3}, at: __netif_tx_lock include/linux/netdevice.h:4795 [inline] #1: ffff88801a199d18 (_xmit_PIMREG#2){+...}-{3:3}, at: __dev_queue_xmit+0x1d5d/0x4140 net/core/dev.c:4865 stack backtrace: CPU: 1 UID: 0 PID: 1779 Comm: syz.2.17 Not tainted 7.1.0-rc2-next-20260504-9d0d467c3572 #1 PREEMPT(lazy) Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.16.0-0-gd239552ce722-prebuilt.qemu.org 04/01/2014 Call Trace: __dump_stack lib/dump_stack.c:94 [inline] dump_stack_lvl+0x121/0x150 lib/dump_stack.c:120 dump_stack+0x19/0x20 lib/dump_stack.c:129 lockdep_rcu_suspicious+0x15b/0x1f0 kernel/locking/lockdep.c:6878 ipmr_fib_lookup net/ipv4/ipmr.c:329 [inline] reg_vif_xmit+0x2ee/0x3c0 net/ipv4/ipmr.c:540 __netdev_start_xmit include/linux/netdevice.h:5382 [inline] netdev_start_xmit include/linux/netdevice.h:5391 [inline] xmit_one net/core/dev.c:3889 [inline] dev_hard_start_xmit+0x170/0x700 net/core/dev.c:3905 __dev_queue_xmit+0x1df1/0x4140 net/core/dev.c:4871 dev_queue_xmit include/linux/netdevice.h:3423 [inline] packet_xmit+0x252/0x370 net/packet/af_packet.c:276 packet_snd net/packet/af_packet.c:3082 [inline] packet_sendmsg+0x39ad/0x5650 net/packet/af_packet.c:3114 sock_sendmsg_nosec net/socket.c:797 [inline] __sock_sendmsg net/socket.c:812 [inline] ____sys_sendmsg+0xa21/0xba0 net/socket.c:2716 ___sys_sendmsg+0x121/0x1c0 net/socket.c:2770 __sys_sendmsg+0x177/0x220 net/socket.c:2802 __do_sys_sendmsg net/socket.c:2807 [inline] __se_sys_sendmsg net/socket.c:2805 [inline] __x64_sys_sendmsg+0x80/0xc0 net/socket.c:2805 x64_sys_call+0x1d9c/0x21c0 arch/x86/include/generated/asm/syscalls_64.h:47 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline] do_syscall_64+0xc1/0x1020 arch/x86/entry/syscall_64.c:94 entry_SYSCALL_64_after_hwframe+0x76/0x7e RIP: 0033:0x7f37e563ee5d Code: ff c3 66 2e 0f 1f 84 00 00 00 00 00 90 f3 0f 1e fa 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d 93 af 1b 00 f7 d8 64 89 01 48 RSP: 002b:00007ffe5caa7fa8 EFLAGS: 00000246 ORIG_RAX: 000000000000002e RAX: ffffffffffffffda RBX: 00000000005c5fa0 RCX: 00007f37e563ee5d RDX: 0000000000000000 RSI: 00002000000012c0 RDI: 0000000000000004 RBP: 00000000005c5fa0 R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000 R13: 0000000000000000 R14: 00000000005c5fac R15: 00000000005c5fa0 Fixes: b3b6babf4751 ("ipmr: Free mr_table after RCU grace period.") Reported-by: syzkaller Reported-by: Yi Lai Closes: https://lore.kernel.org/netdev/afrY34dLXNUboevf@ly-workstation/ Signed-off-by: Kuniyuki Iwashima Reviewed-by: Eric Dumazet Reviewed-by: Ido Schimmel Link: https://patch.msgid.link/20260506065955.1695753-1-kuniyu@google.com Signed-off-by: Jakub Kicinski --- net/ipv4/ipmr.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c index 05fb6eefe0be..2628cd3a93a6 100644 --- a/net/ipv4/ipmr.c +++ b/net/ipv4/ipmr.c @@ -537,15 +537,16 @@ static netdev_tx_t reg_vif_xmit(struct sk_buff *skb, struct net_device *dev) }; int err; + rcu_read_lock(); err = ipmr_fib_lookup(net, &fl4, &mrt); if (err < 0) { + rcu_read_unlock(); kfree_skb(skb); return err; } DEV_STATS_ADD(dev, tx_bytes, skb->len); DEV_STATS_INC(dev, tx_packets); - rcu_read_lock(); /* Pairs with WRITE_ONCE() in vif_add() and vif_delete() */ ipmr_cache_report(mrt, skb, READ_ONCE(mrt->mroute_reg_vif_num), -- cgit v1.2.3 From ecddc523cfdb85b3e132f13e293224ebfdfab564 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Wed, 6 May 2026 07:04:42 +0000 Subject: tcp: Fix dst leak in tcp_v6_connect(). If a socket is bound to a wildcard address, tcp_v[46]_connect() updates it with a non-wildcard address based on the route lookup. After bhash2 was introduced in the cited commit, we must call inet_bhash2_update_saddr() to update the bhash2 entry as well. If inet_bhash2_update_saddr() fails, we must release the refcount for dst by ip_route_connect() or ip6_dst_lookup_flow(). While tcp_v4_connect() calls ip_rt_put() in the error path, tcp_v6_connect() does not call dst_release(). Let's call dst_release() when inet_bhash2_update_saddr() fails in tcp_v6_connect(). Fixes: 28044fc1d495 ("net: Add a bhash2 table hashed by port and address") Reported-by: Damiano Melotti Signed-off-by: Kuniyuki Iwashima Reviewed-by: Eric Dumazet Link: https://patch.msgid.link/20260506070443.1699879-1-kuniyu@google.com Signed-off-by: Jakub Kicinski --- net/ipv6/tcp_ipv6.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 51583aef0643..d13d49bfef19 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -288,8 +288,10 @@ static int tcp_v6_connect(struct sock *sk, struct sockaddr_unsized *uaddr, saddr = &fl6->saddr; err = inet_bhash2_update_saddr(sk, saddr, AF_INET6); - if (err) + if (err) { + dst_release(dst); goto failure; + } } /* set the source address */ -- cgit v1.2.3 From 8e72510db9fa2d41f2b06d5c01fe9020e076fee4 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Wed, 6 May 2026 12:07:13 +0200 Subject: netfilter: x_tables: allow initial table replace without emitting audit log message At the moment we emit the audit log a bit too early, which makes it necessary to also emit an unregister log in case we have to unwind errors after possible hook register failure. Followup patch will be slightly simpler if we can delay the register message until after the hooks have been wired up. Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- net/netfilter/x_tables.c | 29 ++++++++++++++++++++--------- 1 file changed, 20 insertions(+), 9 deletions(-) (limited to 'net') diff --git a/net/netfilter/x_tables.c b/net/netfilter/x_tables.c index 2c67c2e6b132..bb0cb3959551 100644 --- a/net/netfilter/x_tables.c +++ b/net/netfilter/x_tables.c @@ -1472,11 +1472,9 @@ struct xt_counters *xt_counters_alloc(unsigned int counters) } EXPORT_SYMBOL(xt_counters_alloc); -struct xt_table_info * -xt_replace_table(struct xt_table *table, - unsigned int num_counters, - struct xt_table_info *newinfo, - int *error) +static struct xt_table_info * +do_replace_table(struct xt_table *table, unsigned int num_counters, + struct xt_table_info *newinfo, int *error) { struct xt_table_info *private; unsigned int cpu; @@ -1531,10 +1529,23 @@ xt_replace_table(struct xt_table *table, } } - audit_log_nfcfg(table->name, table->af, private->number, - !private->number ? AUDIT_XT_OP_REGISTER : - AUDIT_XT_OP_REPLACE, - GFP_KERNEL); + return private; +} + +struct xt_table_info * +xt_replace_table(struct xt_table *table, unsigned int num_counters, + struct xt_table_info *newinfo, + int *error) +{ + struct xt_table_info *private; + + private = do_replace_table(table, num_counters, newinfo, error); + if (private) + audit_log_nfcfg(table->name, table->af, private->number, + !private->number ? AUDIT_XT_OP_REGISTER : + AUDIT_XT_OP_REPLACE, + GFP_KERNEL); + return private; } EXPORT_SYMBOL_GPL(xt_replace_table); -- cgit v1.2.3 From b62eb8dcf2c47d4d676a434efbd57c4f776f7829 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Wed, 6 May 2026 12:07:14 +0200 Subject: netfilter: x_tables: allocate hook ops while under mutex arp/ip(6)t_register_table() add the table to the per-netns list via xt_register_table() before allocating the per-netns hook ops copy via kmemdup_array(). This leaves a window where the table is visible in the list with ops=NULL. If the pernet exit happens runs concurrently the pre_exit callback finds the table via xt_find_table() and passes the NULL ops pointer to nf_unregister_net_hooks(), causing a NULL dereference: general protection fault in nf_unregister_net_hooks+0xbc/0x150 RIP: nf_unregister_net_hooks (net/netfilter/core.c:613) Call Trace: ipt_unregister_table_pre_exit iptable_mangle_net_pre_exit ops_pre_exit_list cleanup_net Fix by moving the ops allocation into the xtables core so the table is never in the list without valid ops. Also ensure the table is no longer processing packets before its torn down on error unwind. nf_register_net_hooks might have published at least one hook; call synchronize_rcu() if there was an error. audit log register message gets deferred until all operations have passed, this avoids need to emit another ureg message in case of error unwinding. Based on earlier patch by Tristan Madani. Fixes: f9006acc8dfe5 ("netfilter: arp_tables: pass table pointer via nf_hook_ops") Fixes: ee177a54413a ("netfilter: ip6_tables: pass table pointer via nf_hook_ops") Fixes: ae689334225f ("netfilter: ip_tables: pass table pointer via nf_hook_ops") Link: https://lore.kernel.org/netfilter-devel/20260429175613.1459342-1-tristmd@gmail.com/ Signed-off-by: Tristan Madani Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- net/ipv4/netfilter/arp_tables.c | 35 ++++------------------------- net/ipv4/netfilter/ip_tables.c | 41 ++++----------------------------- net/ipv6/netfilter/ip6_tables.c | 38 ++++--------------------------- net/netfilter/x_tables.c | 50 ++++++++++++++++++++++++++++++++++------- 4 files changed, 54 insertions(+), 110 deletions(-) (limited to 'net') diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c index 97ead883e4a1..c02e46a0271a 100644 --- a/net/ipv4/netfilter/arp_tables.c +++ b/net/ipv4/netfilter/arp_tables.c @@ -1522,13 +1522,11 @@ int arpt_register_table(struct net *net, const struct arpt_replace *repl, const struct nf_hook_ops *template_ops) { - struct nf_hook_ops *ops; - unsigned int num_ops; - int ret, i; - struct xt_table_info *newinfo; struct xt_table_info bootstrap = {0}; - void *loc_cpu_entry; + struct xt_table_info *newinfo; struct xt_table *new_table; + void *loc_cpu_entry; + int ret; newinfo = xt_alloc_table_info(repl->size); if (!newinfo) @@ -1543,7 +1541,7 @@ int arpt_register_table(struct net *net, return ret; } - new_table = xt_register_table(net, table, &bootstrap, newinfo); + new_table = xt_register_table(net, table, template_ops, &bootstrap, newinfo); if (IS_ERR(new_table)) { struct arpt_entry *iter; @@ -1553,31 +1551,6 @@ int arpt_register_table(struct net *net, return PTR_ERR(new_table); } - num_ops = hweight32(table->valid_hooks); - if (num_ops == 0) { - ret = -EINVAL; - goto out_free; - } - - ops = kmemdup_array(template_ops, num_ops, sizeof(*ops), GFP_KERNEL); - if (!ops) { - ret = -ENOMEM; - goto out_free; - } - - for (i = 0; i < num_ops; i++) - ops[i].priv = new_table; - - new_table->ops = ops; - - ret = nf_register_net_hooks(net, ops, num_ops); - if (ret != 0) - goto out_free; - - return ret; - -out_free: - __arpt_unregister_table(net, new_table); return ret; } diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c index 23c8deff8095..488c5945ebb2 100644 --- a/net/ipv4/netfilter/ip_tables.c +++ b/net/ipv4/netfilter/ip_tables.c @@ -1724,13 +1724,11 @@ int ipt_register_table(struct net *net, const struct xt_table *table, const struct ipt_replace *repl, const struct nf_hook_ops *template_ops) { - struct nf_hook_ops *ops; - unsigned int num_ops; - int ret, i; - struct xt_table_info *newinfo; struct xt_table_info bootstrap = {0}; - void *loc_cpu_entry; + struct xt_table_info *newinfo; struct xt_table *new_table; + void *loc_cpu_entry; + int ret; newinfo = xt_alloc_table_info(repl->size); if (!newinfo) @@ -1745,7 +1743,7 @@ int ipt_register_table(struct net *net, const struct xt_table *table, return ret; } - new_table = xt_register_table(net, table, &bootstrap, newinfo); + new_table = xt_register_table(net, table, template_ops, &bootstrap, newinfo); if (IS_ERR(new_table)) { struct ipt_entry *iter; @@ -1755,37 +1753,6 @@ int ipt_register_table(struct net *net, const struct xt_table *table, return PTR_ERR(new_table); } - /* No template? No need to do anything. This is used by 'nat' table, it registers - * with the nat core instead of the netfilter core. - */ - if (!template_ops) - return 0; - - num_ops = hweight32(table->valid_hooks); - if (num_ops == 0) { - ret = -EINVAL; - goto out_free; - } - - ops = kmemdup_array(template_ops, num_ops, sizeof(*ops), GFP_KERNEL); - if (!ops) { - ret = -ENOMEM; - goto out_free; - } - - for (i = 0; i < num_ops; i++) - ops[i].priv = new_table; - - new_table->ops = ops; - - ret = nf_register_net_hooks(net, ops, num_ops); - if (ret != 0) - goto out_free; - - return ret; - -out_free: - __ipt_unregister_table(net, new_table); return ret; } diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c index d585ac3c1113..dbe7c7acd702 100644 --- a/net/ipv6/netfilter/ip6_tables.c +++ b/net/ipv6/netfilter/ip6_tables.c @@ -1733,13 +1733,11 @@ int ip6t_register_table(struct net *net, const struct xt_table *table, const struct ip6t_replace *repl, const struct nf_hook_ops *template_ops) { - struct nf_hook_ops *ops; - unsigned int num_ops; - int ret, i; - struct xt_table_info *newinfo; struct xt_table_info bootstrap = {0}; - void *loc_cpu_entry; + struct xt_table_info *newinfo; struct xt_table *new_table; + void *loc_cpu_entry; + int ret; newinfo = xt_alloc_table_info(repl->size); if (!newinfo) @@ -1754,7 +1752,7 @@ int ip6t_register_table(struct net *net, const struct xt_table *table, return ret; } - new_table = xt_register_table(net, table, &bootstrap, newinfo); + new_table = xt_register_table(net, table, template_ops, &bootstrap, newinfo); if (IS_ERR(new_table)) { struct ip6t_entry *iter; @@ -1764,34 +1762,6 @@ int ip6t_register_table(struct net *net, const struct xt_table *table, return PTR_ERR(new_table); } - if (!template_ops) - return 0; - - num_ops = hweight32(table->valid_hooks); - if (num_ops == 0) { - ret = -EINVAL; - goto out_free; - } - - ops = kmemdup_array(template_ops, num_ops, sizeof(*ops), GFP_KERNEL); - if (!ops) { - ret = -ENOMEM; - goto out_free; - } - - for (i = 0; i < num_ops; i++) - ops[i].priv = new_table; - - new_table->ops = ops; - - ret = nf_register_net_hooks(net, ops, num_ops); - if (ret != 0) - goto out_free; - - return ret; - -out_free: - __ip6t_unregister_table(net, new_table); return ret; } diff --git a/net/netfilter/x_tables.c b/net/netfilter/x_tables.c index bb0cb3959551..06f27bea9eed 100644 --- a/net/netfilter/x_tables.c +++ b/net/netfilter/x_tables.c @@ -1542,7 +1542,6 @@ xt_replace_table(struct xt_table *table, unsigned int num_counters, private = do_replace_table(table, num_counters, newinfo, error); if (private) audit_log_nfcfg(table->name, table->af, private->number, - !private->number ? AUDIT_XT_OP_REGISTER : AUDIT_XT_OP_REPLACE, GFP_KERNEL); @@ -1552,20 +1551,32 @@ EXPORT_SYMBOL_GPL(xt_replace_table); struct xt_table *xt_register_table(struct net *net, const struct xt_table *input_table, + const struct nf_hook_ops *template_ops, struct xt_table_info *bootstrap, struct xt_table_info *newinfo) { struct xt_pernet *xt_net = net_generic(net, xt_pernet_id); + struct xt_table *t, *table = NULL; + struct nf_hook_ops *ops = NULL; struct xt_table_info *private; - struct xt_table *t, *table; - int ret; + unsigned int num_ops; + int ret = -EINVAL; + + num_ops = hweight32(input_table->valid_hooks); + if (num_ops == 0) + goto out; + + ret = -ENOMEM; + if (template_ops) { + ops = kmemdup_array(template_ops, num_ops, sizeof(*ops), GFP_KERNEL); + if (!ops) + goto out; + } /* Don't add one object to multiple lists. */ table = kmemdup(input_table, sizeof(struct xt_table), GFP_KERNEL); - if (!table) { - ret = -ENOMEM; + if (!table) goto out; - } mutex_lock(&xt[table->af].mutex); /* Don't autoload: we'd eat our tail... */ @@ -1579,7 +1590,7 @@ struct xt_table *xt_register_table(struct net *net, /* Simplifies replace_table code. */ table->private = bootstrap; - if (!xt_replace_table(table, 0, newinfo, &ret)) + if (!do_replace_table(table, 0, newinfo, &ret)) goto unlock; private = table->private; @@ -1588,14 +1599,37 @@ struct xt_table *xt_register_table(struct net *net, /* save number of initial entries */ private->initial_entries = private->number; + if (ops) { + int i; + + for (i = 0; i < num_ops; i++) + ops[i].priv = table; + + ret = nf_register_net_hooks(net, ops, num_ops); + if (ret != 0) { + mutex_unlock(&xt[table->af].mutex); + /* nf_register_net_hooks() might have published a + * base chain before internal error unwind. + */ + synchronize_rcu(); + goto out; + } + + table->ops = ops; + } + + audit_log_nfcfg(table->name, table->af, private->number, + AUDIT_XT_OP_REGISTER, GFP_KERNEL); + list_add(&table->list, &xt_net->tables[table->af]); mutex_unlock(&xt[table->af].mutex); return table; unlock: mutex_unlock(&xt[table->af].mutex); - kfree(table); out: + kfree(table); + kfree(ops); return ERR_PTR(ret); } EXPORT_SYMBOL_GPL(xt_register_table); -- cgit v1.2.3 From 527d6931473b75d90e38942aae6537d1a527f1fd Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Wed, 6 May 2026 12:07:15 +0200 Subject: netfilter: x_tables: add and use xt_unregister_table_pre_exit Remove the copypasted variants of _pre_exit and add one single function in the xtables core. ebtables is not compatible with x_tables and therefore unchanged. This is a preparation patch to reduce noise in the followup bug fixes. Reviewed-by: Tristan Madani Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- net/ipv4/netfilter/arp_tables.c | 9 --------- net/ipv4/netfilter/arptable_filter.c | 2 +- net/ipv4/netfilter/ip_tables.c | 9 --------- net/ipv4/netfilter/iptable_filter.c | 2 +- net/ipv4/netfilter/iptable_mangle.c | 2 +- net/ipv4/netfilter/iptable_nat.c | 1 + net/ipv4/netfilter/iptable_raw.c | 2 +- net/ipv4/netfilter/iptable_security.c | 2 +- net/ipv6/netfilter/ip6_tables.c | 9 --------- net/ipv6/netfilter/ip6table_filter.c | 2 +- net/ipv6/netfilter/ip6table_mangle.c | 2 +- net/ipv6/netfilter/ip6table_nat.c | 1 + net/ipv6/netfilter/ip6table_raw.c | 2 +- net/ipv6/netfilter/ip6table_security.c | 2 +- net/netfilter/x_tables.c | 29 +++++++++++++++++++++++++++++ 15 files changed, 40 insertions(+), 36 deletions(-) (limited to 'net') diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c index c02e46a0271a..bd348b7bad2c 100644 --- a/net/ipv4/netfilter/arp_tables.c +++ b/net/ipv4/netfilter/arp_tables.c @@ -1554,15 +1554,6 @@ int arpt_register_table(struct net *net, return ret; } -void arpt_unregister_table_pre_exit(struct net *net, const char *name) -{ - struct xt_table *table = xt_find_table(net, NFPROTO_ARP, name); - - if (table) - nf_unregister_net_hooks(net, table->ops, hweight32(table->valid_hooks)); -} -EXPORT_SYMBOL(arpt_unregister_table_pre_exit); - void arpt_unregister_table(struct net *net, const char *name) { struct xt_table *table = xt_find_table(net, NFPROTO_ARP, name); diff --git a/net/ipv4/netfilter/arptable_filter.c b/net/ipv4/netfilter/arptable_filter.c index 78cd5ee24448..393d9a8c7739 100644 --- a/net/ipv4/netfilter/arptable_filter.c +++ b/net/ipv4/netfilter/arptable_filter.c @@ -43,7 +43,7 @@ static int arptable_filter_table_init(struct net *net) static void __net_exit arptable_filter_net_pre_exit(struct net *net) { - arpt_unregister_table_pre_exit(net, "filter"); + xt_unregister_table_pre_exit(net, NFPROTO_ARP, "filter"); } static void __net_exit arptable_filter_net_exit(struct net *net) diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c index 488c5945ebb2..864489928fb5 100644 --- a/net/ipv4/netfilter/ip_tables.c +++ b/net/ipv4/netfilter/ip_tables.c @@ -1756,14 +1756,6 @@ int ipt_register_table(struct net *net, const struct xt_table *table, return ret; } -void ipt_unregister_table_pre_exit(struct net *net, const char *name) -{ - struct xt_table *table = xt_find_table(net, NFPROTO_IPV4, name); - - if (table) - nf_unregister_net_hooks(net, table->ops, hweight32(table->valid_hooks)); -} - void ipt_unregister_table_exit(struct net *net, const char *name) { struct xt_table *table = xt_find_table(net, NFPROTO_IPV4, name); @@ -1854,7 +1846,6 @@ static void __exit ip_tables_fini(void) } EXPORT_SYMBOL(ipt_register_table); -EXPORT_SYMBOL(ipt_unregister_table_pre_exit); EXPORT_SYMBOL(ipt_unregister_table_exit); EXPORT_SYMBOL(ipt_do_table); module_init(ip_tables_init); diff --git a/net/ipv4/netfilter/iptable_filter.c b/net/ipv4/netfilter/iptable_filter.c index 3ab908b74795..b2fbd9651d61 100644 --- a/net/ipv4/netfilter/iptable_filter.c +++ b/net/ipv4/netfilter/iptable_filter.c @@ -61,7 +61,7 @@ static int __net_init iptable_filter_net_init(struct net *net) static void __net_exit iptable_filter_net_pre_exit(struct net *net) { - ipt_unregister_table_pre_exit(net, "filter"); + xt_unregister_table_pre_exit(net, NFPROTO_IPV4, "filter"); } static void __net_exit iptable_filter_net_exit(struct net *net) diff --git a/net/ipv4/netfilter/iptable_mangle.c b/net/ipv4/netfilter/iptable_mangle.c index 385d945d8ebe..a99e61996197 100644 --- a/net/ipv4/netfilter/iptable_mangle.c +++ b/net/ipv4/netfilter/iptable_mangle.c @@ -96,7 +96,7 @@ static int iptable_mangle_table_init(struct net *net) static void __net_exit iptable_mangle_net_pre_exit(struct net *net) { - ipt_unregister_table_pre_exit(net, "mangle"); + xt_unregister_table_pre_exit(net, NFPROTO_IPV4, "mangle"); } static void __net_exit iptable_mangle_net_exit(struct net *net) diff --git a/net/ipv4/netfilter/iptable_nat.c b/net/ipv4/netfilter/iptable_nat.c index 625a1ca13b1b..8fc4912e790d 100644 --- a/net/ipv4/netfilter/iptable_nat.c +++ b/net/ipv4/netfilter/iptable_nat.c @@ -129,6 +129,7 @@ static int iptable_nat_table_init(struct net *net) static void __net_exit iptable_nat_net_pre_exit(struct net *net) { ipt_nat_unregister_lookups(net); + xt_unregister_table_pre_exit(net, NFPROTO_IPV4, "nat"); } static void __net_exit iptable_nat_net_exit(struct net *net) diff --git a/net/ipv4/netfilter/iptable_raw.c b/net/ipv4/netfilter/iptable_raw.c index 0e7f53964d0a..42511721e538 100644 --- a/net/ipv4/netfilter/iptable_raw.c +++ b/net/ipv4/netfilter/iptable_raw.c @@ -53,7 +53,7 @@ static int iptable_raw_table_init(struct net *net) static void __net_exit iptable_raw_net_pre_exit(struct net *net) { - ipt_unregister_table_pre_exit(net, "raw"); + xt_unregister_table_pre_exit(net, NFPROTO_IPV4, "raw"); } static void __net_exit iptable_raw_net_exit(struct net *net) diff --git a/net/ipv4/netfilter/iptable_security.c b/net/ipv4/netfilter/iptable_security.c index d885443cb267..4646bf6d7d2b 100644 --- a/net/ipv4/netfilter/iptable_security.c +++ b/net/ipv4/netfilter/iptable_security.c @@ -50,7 +50,7 @@ static int iptable_security_table_init(struct net *net) static void __net_exit iptable_security_net_pre_exit(struct net *net) { - ipt_unregister_table_pre_exit(net, "security"); + xt_unregister_table_pre_exit(net, NFPROTO_IPV4, "security"); } static void __net_exit iptable_security_net_exit(struct net *net) diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c index dbe7c7acd702..edf50bc7787e 100644 --- a/net/ipv6/netfilter/ip6_tables.c +++ b/net/ipv6/netfilter/ip6_tables.c @@ -1765,14 +1765,6 @@ int ip6t_register_table(struct net *net, const struct xt_table *table, return ret; } -void ip6t_unregister_table_pre_exit(struct net *net, const char *name) -{ - struct xt_table *table = xt_find_table(net, NFPROTO_IPV6, name); - - if (table) - nf_unregister_net_hooks(net, table->ops, hweight32(table->valid_hooks)); -} - void ip6t_unregister_table_exit(struct net *net, const char *name) { struct xt_table *table = xt_find_table(net, NFPROTO_IPV6, name); @@ -1864,7 +1856,6 @@ static void __exit ip6_tables_fini(void) } EXPORT_SYMBOL(ip6t_register_table); -EXPORT_SYMBOL(ip6t_unregister_table_pre_exit); EXPORT_SYMBOL(ip6t_unregister_table_exit); EXPORT_SYMBOL(ip6t_do_table); diff --git a/net/ipv6/netfilter/ip6table_filter.c b/net/ipv6/netfilter/ip6table_filter.c index e8992693e14a..f05a9e4b2c67 100644 --- a/net/ipv6/netfilter/ip6table_filter.c +++ b/net/ipv6/netfilter/ip6table_filter.c @@ -60,7 +60,7 @@ static int __net_init ip6table_filter_net_init(struct net *net) static void __net_exit ip6table_filter_net_pre_exit(struct net *net) { - ip6t_unregister_table_pre_exit(net, "filter"); + xt_unregister_table_pre_exit(net, NFPROTO_IPV6, "filter"); } static void __net_exit ip6table_filter_net_exit(struct net *net) diff --git a/net/ipv6/netfilter/ip6table_mangle.c b/net/ipv6/netfilter/ip6table_mangle.c index 8dd4cd0c47bd..afa4a5703e43 100644 --- a/net/ipv6/netfilter/ip6table_mangle.c +++ b/net/ipv6/netfilter/ip6table_mangle.c @@ -89,7 +89,7 @@ static int ip6table_mangle_table_init(struct net *net) static void __net_exit ip6table_mangle_net_pre_exit(struct net *net) { - ip6t_unregister_table_pre_exit(net, "mangle"); + xt_unregister_table_pre_exit(net, NFPROTO_IPV6, "mangle"); } static void __net_exit ip6table_mangle_net_exit(struct net *net) diff --git a/net/ipv6/netfilter/ip6table_nat.c b/net/ipv6/netfilter/ip6table_nat.c index 5be723232df8..bb8aa3fc42b4 100644 --- a/net/ipv6/netfilter/ip6table_nat.c +++ b/net/ipv6/netfilter/ip6table_nat.c @@ -131,6 +131,7 @@ static int ip6table_nat_table_init(struct net *net) static void __net_exit ip6table_nat_net_pre_exit(struct net *net) { ip6t_nat_unregister_lookups(net); + xt_unregister_table_pre_exit(net, NFPROTO_IPV6, "nat"); } static void __net_exit ip6table_nat_net_exit(struct net *net) diff --git a/net/ipv6/netfilter/ip6table_raw.c b/net/ipv6/netfilter/ip6table_raw.c index fc9f6754028f..32d2da81c52a 100644 --- a/net/ipv6/netfilter/ip6table_raw.c +++ b/net/ipv6/netfilter/ip6table_raw.c @@ -52,7 +52,7 @@ static int ip6table_raw_table_init(struct net *net) static void __net_exit ip6table_raw_net_pre_exit(struct net *net) { - ip6t_unregister_table_pre_exit(net, "raw"); + xt_unregister_table_pre_exit(net, NFPROTO_IPV6, "raw"); } static void __net_exit ip6table_raw_net_exit(struct net *net) diff --git a/net/ipv6/netfilter/ip6table_security.c b/net/ipv6/netfilter/ip6table_security.c index 4df14a9bae78..3dfd8d6ea4b9 100644 --- a/net/ipv6/netfilter/ip6table_security.c +++ b/net/ipv6/netfilter/ip6table_security.c @@ -49,7 +49,7 @@ static int ip6table_security_table_init(struct net *net) static void __net_exit ip6table_security_net_pre_exit(struct net *net) { - ip6t_unregister_table_pre_exit(net, "security"); + xt_unregister_table_pre_exit(net, NFPROTO_IPV6, "security"); } static void __net_exit ip6table_security_net_exit(struct net *net) diff --git a/net/netfilter/x_tables.c b/net/netfilter/x_tables.c index 06f27bea9eed..9c1e896c7b03 100644 --- a/net/netfilter/x_tables.c +++ b/net/netfilter/x_tables.c @@ -1650,6 +1650,35 @@ void *xt_unregister_table(struct xt_table *table) return private; } EXPORT_SYMBOL_GPL(xt_unregister_table); + +/** + * xt_unregister_table_pre_exit - pre-shutdown unregister of a table + * @net: network namespace + * @af: address family (e.g., NFPROTO_IPV4, NFPROTO_IPV6) + * @name: name of the table to unregister + * + * Unregisters the specified netfilter table from the given network namespace + * and also unregisters the hooks from netfilter core: no new packets will be + * processed. + */ +void xt_unregister_table_pre_exit(struct net *net, u8 af, const char *name) +{ + struct xt_pernet *xt_net = net_generic(net, xt_pernet_id); + struct xt_table *t; + + mutex_lock(&xt[af].mutex); + list_for_each_entry(t, &xt_net->tables[af], list) { + if (strcmp(t->name, name) == 0) { + mutex_unlock(&xt[af].mutex); + + if (t->ops) /* nat table registers with nat core, t->ops is NULL. */ + nf_unregister_net_hooks(net, t->ops, hweight32(t->valid_hooks)); + return; + } + } + mutex_unlock(&xt[af].mutex); +} +EXPORT_SYMBOL(xt_unregister_table_pre_exit); #endif #ifdef CONFIG_PROC_FS -- cgit v1.2.3 From d338693d778579b676a61346849bebd892427158 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Wed, 6 May 2026 12:07:16 +0200 Subject: netfilter: x_tables: unregister the templates first When the module is going away we need to zap the template first. Else there is a small race window where userspace could instantiate a new table after the pernet exit function has removed the current table. Fixes: fdacd57c79b7 ("netfilter: x_tables: never register tables by default") Reported-by: Tristan Madani Reviewed-by: Tristan Madani Closes: https://lore.kernel.org/netfilter-devel/20260429175613.1459342-1-tristmd@gmail.com/ Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- net/ipv4/netfilter/arptable_filter.c | 2 +- net/ipv4/netfilter/iptable_filter.c | 2 +- net/ipv4/netfilter/iptable_mangle.c | 2 +- net/ipv4/netfilter/iptable_raw.c | 2 +- net/ipv4/netfilter/iptable_security.c | 2 +- net/ipv6/netfilter/ip6table_filter.c | 2 +- net/ipv6/netfilter/ip6table_mangle.c | 2 +- net/ipv6/netfilter/ip6table_raw.c | 2 +- net/ipv6/netfilter/ip6table_security.c | 2 +- 9 files changed, 9 insertions(+), 9 deletions(-) (limited to 'net') diff --git a/net/ipv4/netfilter/arptable_filter.c b/net/ipv4/netfilter/arptable_filter.c index 393d9a8c7739..382345567a60 100644 --- a/net/ipv4/netfilter/arptable_filter.c +++ b/net/ipv4/netfilter/arptable_filter.c @@ -82,8 +82,8 @@ static int __init arptable_filter_init(void) static void __exit arptable_filter_fini(void) { - unregister_pernet_subsys(&arptable_filter_net_ops); xt_unregister_template(&packet_filter); + unregister_pernet_subsys(&arptable_filter_net_ops); kfree(arpfilter_ops); } diff --git a/net/ipv4/netfilter/iptable_filter.c b/net/ipv4/netfilter/iptable_filter.c index b2fbd9651d61..0dea754a9120 100644 --- a/net/ipv4/netfilter/iptable_filter.c +++ b/net/ipv4/netfilter/iptable_filter.c @@ -101,8 +101,8 @@ static int __init iptable_filter_init(void) static void __exit iptable_filter_fini(void) { - unregister_pernet_subsys(&iptable_filter_net_ops); xt_unregister_template(&packet_filter); + unregister_pernet_subsys(&iptable_filter_net_ops); kfree(filter_ops); } diff --git a/net/ipv4/netfilter/iptable_mangle.c b/net/ipv4/netfilter/iptable_mangle.c index a99e61996197..4d3b12492308 100644 --- a/net/ipv4/netfilter/iptable_mangle.c +++ b/net/ipv4/netfilter/iptable_mangle.c @@ -135,8 +135,8 @@ static int __init iptable_mangle_init(void) static void __exit iptable_mangle_fini(void) { - unregister_pernet_subsys(&iptable_mangle_net_ops); xt_unregister_template(&packet_mangler); + unregister_pernet_subsys(&iptable_mangle_net_ops); kfree(mangle_ops); } diff --git a/net/ipv4/netfilter/iptable_raw.c b/net/ipv4/netfilter/iptable_raw.c index 42511721e538..6f7afec7954b 100644 --- a/net/ipv4/netfilter/iptable_raw.c +++ b/net/ipv4/netfilter/iptable_raw.c @@ -100,9 +100,9 @@ static int __init iptable_raw_init(void) static void __exit iptable_raw_fini(void) { + xt_unregister_template(&packet_raw); unregister_pernet_subsys(&iptable_raw_net_ops); kfree(rawtable_ops); - xt_unregister_template(&packet_raw); } module_init(iptable_raw_init); diff --git a/net/ipv4/netfilter/iptable_security.c b/net/ipv4/netfilter/iptable_security.c index 4646bf6d7d2b..81175c20ccbe 100644 --- a/net/ipv4/netfilter/iptable_security.c +++ b/net/ipv4/netfilter/iptable_security.c @@ -89,9 +89,9 @@ static int __init iptable_security_init(void) static void __exit iptable_security_fini(void) { + xt_unregister_template(&security_table); unregister_pernet_subsys(&iptable_security_net_ops); kfree(sectbl_ops); - xt_unregister_template(&security_table); } module_init(iptable_security_init); diff --git a/net/ipv6/netfilter/ip6table_filter.c b/net/ipv6/netfilter/ip6table_filter.c index f05a9e4b2c67..cf561919bde8 100644 --- a/net/ipv6/netfilter/ip6table_filter.c +++ b/net/ipv6/netfilter/ip6table_filter.c @@ -100,8 +100,8 @@ static int __init ip6table_filter_init(void) static void __exit ip6table_filter_fini(void) { - unregister_pernet_subsys(&ip6table_filter_net_ops); xt_unregister_template(&packet_filter); + unregister_pernet_subsys(&ip6table_filter_net_ops); kfree(filter_ops); } diff --git a/net/ipv6/netfilter/ip6table_mangle.c b/net/ipv6/netfilter/ip6table_mangle.c index afa4a5703e43..1a758f2bc537 100644 --- a/net/ipv6/netfilter/ip6table_mangle.c +++ b/net/ipv6/netfilter/ip6table_mangle.c @@ -128,8 +128,8 @@ static int __init ip6table_mangle_init(void) static void __exit ip6table_mangle_fini(void) { - unregister_pernet_subsys(&ip6table_mangle_net_ops); xt_unregister_template(&packet_mangler); + unregister_pernet_subsys(&ip6table_mangle_net_ops); kfree(mangle_ops); } diff --git a/net/ipv6/netfilter/ip6table_raw.c b/net/ipv6/netfilter/ip6table_raw.c index 32d2da81c52a..923455921c1d 100644 --- a/net/ipv6/netfilter/ip6table_raw.c +++ b/net/ipv6/netfilter/ip6table_raw.c @@ -98,8 +98,8 @@ static int __init ip6table_raw_init(void) static void __exit ip6table_raw_fini(void) { - unregister_pernet_subsys(&ip6table_raw_net_ops); xt_unregister_template(&packet_raw); + unregister_pernet_subsys(&ip6table_raw_net_ops); kfree(rawtable_ops); } diff --git a/net/ipv6/netfilter/ip6table_security.c b/net/ipv6/netfilter/ip6table_security.c index 3dfd8d6ea4b9..c44834d93fc7 100644 --- a/net/ipv6/netfilter/ip6table_security.c +++ b/net/ipv6/netfilter/ip6table_security.c @@ -88,8 +88,8 @@ static int __init ip6table_security_init(void) static void __exit ip6table_security_fini(void) { - unregister_pernet_subsys(&ip6table_security_net_ops); xt_unregister_template(&security_table); + unregister_pernet_subsys(&ip6table_security_net_ops); kfree(sectbl_ops); } -- cgit v1.2.3 From b4597d5fd7d2f8cebfffd40dffb5e003cc78964c Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Wed, 6 May 2026 12:07:17 +0200 Subject: netfilter: x_tables: add and use xtables_unregister_table_exit Previous change added xtables_unregister_table_pre_exit to detach the table from the packetpath and to unlink it from the active table list. In case of rmmod, userspace that is doing set/getsockopt for this table will not be able to re-instantiate the table: 1. The larval table has been removed already 2. existing instantiated table is no longer on the xt pernet table list. This adds the second stage helper: unlink the table from the dying list, free the hook ops (if any) and do the audit notification. It replaces xt_unregister_table(). Fixes: fdacd57c79b7 ("netfilter: x_tables: never register tables by default") Reported-by: Tristan Madani Reviewed-by: Tristan Madani Closes: https://lore.kernel.org/netfilter-devel/20260429175613.1459342-1-tristmd@gmail.com/ Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- net/ipv4/netfilter/arp_tables.c | 9 ++--- net/ipv4/netfilter/ip_tables.c | 9 ++--- net/ipv4/netfilter/iptable_nat.c | 5 ++- net/ipv6/netfilter/ip6_tables.c | 9 ++--- net/ipv6/netfilter/ip6table_nat.c | 5 ++- net/netfilter/x_tables.c | 81 ++++++++++++++++++++++++++++++--------- 6 files changed, 82 insertions(+), 36 deletions(-) (limited to 'net') diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c index bd348b7bad2c..ad2259678c78 100644 --- a/net/ipv4/netfilter/arp_tables.c +++ b/net/ipv4/netfilter/arp_tables.c @@ -1501,13 +1501,11 @@ static int do_arpt_get_ctl(struct sock *sk, int cmd, void __user *user, int *len static void __arpt_unregister_table(struct net *net, struct xt_table *table) { - struct xt_table_info *private; - void *loc_cpu_entry; + struct xt_table_info *private = table->private; struct module *table_owner = table->me; + void *loc_cpu_entry; struct arpt_entry *iter; - private = xt_unregister_table(table); - /* Decrease module usage counts and free resources */ loc_cpu_entry = private->entries; xt_entry_foreach(iter, loc_cpu_entry, private->size) @@ -1515,6 +1513,7 @@ static void __arpt_unregister_table(struct net *net, struct xt_table *table) if (private->number > private->initial_entries) module_put(table_owner); xt_free_table_info(private); + kfree(table); } int arpt_register_table(struct net *net, @@ -1556,7 +1555,7 @@ int arpt_register_table(struct net *net, void arpt_unregister_table(struct net *net, const char *name) { - struct xt_table *table = xt_find_table(net, NFPROTO_ARP, name); + struct xt_table *table = xt_unregister_table_exit(net, NFPROTO_ARP, name); if (table) __arpt_unregister_table(net, table); diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c index 864489928fb5..5cbdb0815857 100644 --- a/net/ipv4/netfilter/ip_tables.c +++ b/net/ipv4/netfilter/ip_tables.c @@ -1704,12 +1704,10 @@ do_ipt_get_ctl(struct sock *sk, int cmd, void __user *user, int *len) static void __ipt_unregister_table(struct net *net, struct xt_table *table) { - struct xt_table_info *private; - void *loc_cpu_entry; + struct xt_table_info *private = table->private; struct module *table_owner = table->me; struct ipt_entry *iter; - - private = xt_unregister_table(table); + void *loc_cpu_entry; /* Decrease module usage counts and free resources */ loc_cpu_entry = private->entries; @@ -1718,6 +1716,7 @@ static void __ipt_unregister_table(struct net *net, struct xt_table *table) if (private->number > private->initial_entries) module_put(table_owner); xt_free_table_info(private); + kfree(table); } int ipt_register_table(struct net *net, const struct xt_table *table, @@ -1758,7 +1757,7 @@ int ipt_register_table(struct net *net, const struct xt_table *table, void ipt_unregister_table_exit(struct net *net, const char *name) { - struct xt_table *table = xt_find_table(net, NFPROTO_IPV4, name); + struct xt_table *table = xt_unregister_table_exit(net, NFPROTO_IPV4, name); if (table) __ipt_unregister_table(net, table); diff --git a/net/ipv4/netfilter/iptable_nat.c b/net/ipv4/netfilter/iptable_nat.c index 8fc4912e790d..a0df72554025 100644 --- a/net/ipv4/netfilter/iptable_nat.c +++ b/net/ipv4/netfilter/iptable_nat.c @@ -119,8 +119,11 @@ static int iptable_nat_table_init(struct net *net) } ret = ipt_nat_register_lookups(net); - if (ret < 0) + if (ret < 0) { + xt_unregister_table_pre_exit(net, NFPROTO_IPV4, "nat"); + synchronize_rcu(); ipt_unregister_table_exit(net, "nat"); + } kfree(repl); return ret; diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c index edf50bc7787e..9d9c3763f2f5 100644 --- a/net/ipv6/netfilter/ip6_tables.c +++ b/net/ipv6/netfilter/ip6_tables.c @@ -1713,12 +1713,10 @@ do_ip6t_get_ctl(struct sock *sk, int cmd, void __user *user, int *len) static void __ip6t_unregister_table(struct net *net, struct xt_table *table) { - struct xt_table_info *private; - void *loc_cpu_entry; + struct xt_table_info *private = table->private; struct module *table_owner = table->me; struct ip6t_entry *iter; - - private = xt_unregister_table(table); + void *loc_cpu_entry; /* Decrease module usage counts and free resources */ loc_cpu_entry = private->entries; @@ -1727,6 +1725,7 @@ static void __ip6t_unregister_table(struct net *net, struct xt_table *table) if (private->number > private->initial_entries) module_put(table_owner); xt_free_table_info(private); + kfree(table); } int ip6t_register_table(struct net *net, const struct xt_table *table, @@ -1767,7 +1766,7 @@ int ip6t_register_table(struct net *net, const struct xt_table *table, void ip6t_unregister_table_exit(struct net *net, const char *name) { - struct xt_table *table = xt_find_table(net, NFPROTO_IPV6, name); + struct xt_table *table = xt_unregister_table_exit(net, NFPROTO_IPV6, name); if (table) __ip6t_unregister_table(net, table); diff --git a/net/ipv6/netfilter/ip6table_nat.c b/net/ipv6/netfilter/ip6table_nat.c index bb8aa3fc42b4..c2394e2c94b5 100644 --- a/net/ipv6/netfilter/ip6table_nat.c +++ b/net/ipv6/netfilter/ip6table_nat.c @@ -121,8 +121,11 @@ static int ip6table_nat_table_init(struct net *net) } ret = ip6t_nat_register_lookups(net); - if (ret < 0) + if (ret < 0) { + xt_unregister_table_pre_exit(net, NFPROTO_IPV6, "nat"); + synchronize_rcu(); ip6t_unregister_table_exit(net, "nat"); + } kfree(repl); return ret; diff --git a/net/netfilter/x_tables.c b/net/netfilter/x_tables.c index 9c1e896c7b03..4e6708c23922 100644 --- a/net/netfilter/x_tables.c +++ b/net/netfilter/x_tables.c @@ -55,6 +55,9 @@ static struct list_head xt_templates[NFPROTO_NUMPROTO]; struct xt_pernet { struct list_head tables[NFPROTO_NUMPROTO]; + + /* stash area used during netns exit */ + struct list_head dead_tables[NFPROTO_NUMPROTO]; }; struct compat_delta { @@ -1634,23 +1637,6 @@ out: } EXPORT_SYMBOL_GPL(xt_register_table); -void *xt_unregister_table(struct xt_table *table) -{ - struct xt_table_info *private; - - mutex_lock(&xt[table->af].mutex); - private = table->private; - list_del(&table->list); - mutex_unlock(&xt[table->af].mutex); - audit_log_nfcfg(table->name, table->af, private->number, - AUDIT_XT_OP_UNREGISTER, GFP_KERNEL); - kfree(table->ops); - kfree(table); - - return private; -} -EXPORT_SYMBOL_GPL(xt_unregister_table); - /** * xt_unregister_table_pre_exit - pre-shutdown unregister of a table * @net: network namespace @@ -1660,6 +1646,14 @@ EXPORT_SYMBOL_GPL(xt_unregister_table); * Unregisters the specified netfilter table from the given network namespace * and also unregisters the hooks from netfilter core: no new packets will be * processed. + * + * This must be called prior to xt_unregister_table_exit() from the pernet + * .pre_exit callback. After this call, the table is no longer visible to + * the get/setsockopt path. In case of rmmod, module exit path must have + * called xt_unregister_template() prior to unregistering pernet ops to + * prevent re-instantiation of the table. + * + * See also: xt_unregister_table_exit() */ void xt_unregister_table_pre_exit(struct net *net, u8 af, const char *name) { @@ -1669,6 +1663,7 @@ void xt_unregister_table_pre_exit(struct net *net, u8 af, const char *name) mutex_lock(&xt[af].mutex); list_for_each_entry(t, &xt_net->tables[af], list) { if (strcmp(t->name, name) == 0) { + list_move(&t->list, &xt_net->dead_tables[af]); mutex_unlock(&xt[af].mutex); if (t->ops) /* nat table registers with nat core, t->ops is NULL. */ @@ -1679,6 +1674,50 @@ void xt_unregister_table_pre_exit(struct net *net, u8 af, const char *name) mutex_unlock(&xt[af].mutex); } EXPORT_SYMBOL(xt_unregister_table_pre_exit); + +/** + * xt_unregister_table_exit - remove a table during namespace teardown + * @net: the network namespace from which to unregister the table + * @af: address family (e.g., NFPROTO_IPV4, NFPROTO_IPV6) + * @name: name of the table to unregister + * + * Completes the unregister process for a table. This must be called from + * the pernet ops .exit callback. This is the second stage after + * xt_unregister_table_pre_exit(). + * + * pair with xt_unregister_table_pre_exit() during namespace shutdown. + * + * Return: the unregistered table or NULL if the table was never + * instantiated. The caller needs to kfree() the table after it + * has removed the family specific matches/targets. + */ +struct xt_table *xt_unregister_table_exit(struct net *net, u8 af, const char *name) +{ + struct xt_pernet *xt_net = net_generic(net, xt_pernet_id); + struct xt_table *table; + + mutex_lock(&xt[af].mutex); + list_for_each_entry(table, &xt_net->dead_tables[af], list) { + struct nf_hook_ops *ops = NULL; + + if (strcmp(table->name, name) != 0) + continue; + + list_del(&table->list); + + audit_log_nfcfg(table->name, table->af, table->private->number, + AUDIT_XT_OP_UNREGISTER, GFP_KERNEL); + swap(table->ops, ops); + mutex_unlock(&xt[af].mutex); + + kfree(ops); + return table; + } + mutex_unlock(&xt[af].mutex); + + return NULL; +} +EXPORT_SYMBOL_GPL(xt_unregister_table_exit); #endif #ifdef CONFIG_PROC_FS @@ -2125,8 +2164,10 @@ static int __net_init xt_net_init(struct net *net) struct xt_pernet *xt_net = net_generic(net, xt_pernet_id); int i; - for (i = 0; i < NFPROTO_NUMPROTO; i++) + for (i = 0; i < NFPROTO_NUMPROTO; i++) { INIT_LIST_HEAD(&xt_net->tables[i]); + INIT_LIST_HEAD(&xt_net->dead_tables[i]); + } return 0; } @@ -2135,8 +2176,10 @@ static void __net_exit xt_net_exit(struct net *net) struct xt_pernet *xt_net = net_generic(net, xt_pernet_id); int i; - for (i = 0; i < NFPROTO_NUMPROTO; i++) + for (i = 0; i < NFPROTO_NUMPROTO; i++) { WARN_ON_ONCE(!list_empty(&xt_net->tables[i])); + WARN_ON_ONCE(!list_empty(&xt_net->dead_tables[i])); + } } static struct pernet_operations xt_net_ops = { -- cgit v1.2.3 From b7f0544d86d439cb946515d2ef6a0a75e8626710 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Wed, 6 May 2026 12:07:18 +0200 Subject: netfilter: ebtables: move to two-stage removal scheme Like previous patches for x_tables, follow same pattern in ebtables. We can't reuse xt helpers: ebt_table struct layout is incompatible. table->ops assignment is now done while still holding the ebt mutex to make sure we never expose partially-filled table struct. Fixes: 87663c39f898 ("netfilter: ebtables: do not hook tables by default") Reviewed-by: Tristan Madani Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- net/bridge/netfilter/ebtable_broute.c | 2 +- net/bridge/netfilter/ebtable_filter.c | 2 +- net/bridge/netfilter/ebtable_nat.c | 2 +- net/bridge/netfilter/ebtables.c | 60 +++++++++++++++++++++-------------- 4 files changed, 40 insertions(+), 26 deletions(-) (limited to 'net') diff --git a/net/bridge/netfilter/ebtable_broute.c b/net/bridge/netfilter/ebtable_broute.c index 741360219552..e6f9e343b41f 100644 --- a/net/bridge/netfilter/ebtable_broute.c +++ b/net/bridge/netfilter/ebtable_broute.c @@ -128,8 +128,8 @@ static int __init ebtable_broute_init(void) static void __exit ebtable_broute_fini(void) { - unregister_pernet_subsys(&broute_net_ops); ebt_unregister_template(&broute_table); + unregister_pernet_subsys(&broute_net_ops); } module_init(ebtable_broute_init); diff --git a/net/bridge/netfilter/ebtable_filter.c b/net/bridge/netfilter/ebtable_filter.c index dacd81b12e62..02b6501c15a5 100644 --- a/net/bridge/netfilter/ebtable_filter.c +++ b/net/bridge/netfilter/ebtable_filter.c @@ -109,8 +109,8 @@ static int __init ebtable_filter_init(void) static void __exit ebtable_filter_fini(void) { - unregister_pernet_subsys(&frame_filter_net_ops); ebt_unregister_template(&frame_filter); + unregister_pernet_subsys(&frame_filter_net_ops); } module_init(ebtable_filter_init); diff --git a/net/bridge/netfilter/ebtable_nat.c b/net/bridge/netfilter/ebtable_nat.c index 0f2a8c6118d4..9985a82555c4 100644 --- a/net/bridge/netfilter/ebtable_nat.c +++ b/net/bridge/netfilter/ebtable_nat.c @@ -109,8 +109,8 @@ static int __init ebtable_nat_init(void) static void __exit ebtable_nat_fini(void) { - unregister_pernet_subsys(&frame_nat_net_ops); ebt_unregister_template(&frame_nat); + unregister_pernet_subsys(&frame_nat_net_ops); } module_init(ebtable_nat_init); diff --git a/net/bridge/netfilter/ebtables.c b/net/bridge/netfilter/ebtables.c index aea3e19875c6..3578ffbc14ae 100644 --- a/net/bridge/netfilter/ebtables.c +++ b/net/bridge/netfilter/ebtables.c @@ -42,6 +42,7 @@ struct ebt_pernet { struct list_head tables; + struct list_head dead_tables; }; struct ebt_template { @@ -1162,11 +1163,6 @@ free_newinfo: static void __ebt_unregister_table(struct net *net, struct ebt_table *table) { - mutex_lock(&ebt_mutex); - list_del(&table->list); - mutex_unlock(&ebt_mutex); - audit_log_nfcfg(table->name, AF_BRIDGE, table->private->nentries, - AUDIT_XT_OP_UNREGISTER, GFP_KERNEL); EBT_ENTRY_ITERATE(table->private->entries, table->private->entries_size, ebt_cleanup_entry, net, NULL); if (table->private->nentries) @@ -1267,13 +1263,15 @@ int ebt_register_table(struct net *net, const struct ebt_table *input_table, for (i = 0; i < num_ops; i++) ops[i].priv = table; - list_add(&table->list, &ebt_net->tables); - mutex_unlock(&ebt_mutex); - table->ops = ops; ret = nf_register_net_hooks(net, ops, num_ops); - if (ret) + if (ret) { + synchronize_rcu(); __ebt_unregister_table(net, table); + } else { + list_add(&table->list, &ebt_net->tables); + } + mutex_unlock(&ebt_mutex); audit_log_nfcfg(repl->name, AF_BRIDGE, repl->nentries, AUDIT_XT_OP_REGISTER, GFP_KERNEL); @@ -1339,7 +1337,7 @@ void ebt_unregister_template(const struct ebt_table *t) } EXPORT_SYMBOL(ebt_unregister_template); -static struct ebt_table *__ebt_find_table(struct net *net, const char *name) +void ebt_unregister_table_pre_exit(struct net *net, const char *name) { struct ebt_pernet *ebt_net = net_generic(net, ebt_pernet_id); struct ebt_table *t; @@ -1348,30 +1346,36 @@ static struct ebt_table *__ebt_find_table(struct net *net, const char *name) list_for_each_entry(t, &ebt_net->tables, list) { if (strcmp(t->name, name) == 0) { + list_move(&t->list, &ebt_net->dead_tables); mutex_unlock(&ebt_mutex); - return t; + nf_unregister_net_hooks(net, t->ops, hweight32(t->valid_hooks)); + return; } } mutex_unlock(&ebt_mutex); - return NULL; -} - -void ebt_unregister_table_pre_exit(struct net *net, const char *name) -{ - struct ebt_table *table = __ebt_find_table(net, name); - - if (table) - nf_unregister_net_hooks(net, table->ops, hweight32(table->valid_hooks)); } EXPORT_SYMBOL(ebt_unregister_table_pre_exit); void ebt_unregister_table(struct net *net, const char *name) { - struct ebt_table *table = __ebt_find_table(net, name); + struct ebt_pernet *ebt_net = net_generic(net, ebt_pernet_id); + struct ebt_table *t; - if (table) - __ebt_unregister_table(net, table); + mutex_lock(&ebt_mutex); + + list_for_each_entry(t, &ebt_net->dead_tables, list) { + if (strcmp(t->name, name) == 0) { + list_del(&t->list); + audit_log_nfcfg(t->name, AF_BRIDGE, t->private->nentries, + AUDIT_XT_OP_UNREGISTER, GFP_KERNEL); + __ebt_unregister_table(net, t); + mutex_unlock(&ebt_mutex); + return; + } + } + + mutex_unlock(&ebt_mutex); } /* userspace just supplied us with counters */ @@ -2556,11 +2560,21 @@ static int __net_init ebt_pernet_init(struct net *net) struct ebt_pernet *ebt_net = net_generic(net, ebt_pernet_id); INIT_LIST_HEAD(&ebt_net->tables); + INIT_LIST_HEAD(&ebt_net->dead_tables); return 0; } +static void __net_exit ebt_pernet_exit(struct net *net) +{ + struct ebt_pernet *ebt_net = net_generic(net, ebt_pernet_id); + + WARN_ON_ONCE(!list_empty(&ebt_net->tables)); + WARN_ON_ONCE(!list_empty(&ebt_net->dead_tables)); +} + static struct pernet_operations ebt_net_ops = { .init = ebt_pernet_init, + .exit = ebt_pernet_exit, .id = &ebt_pernet_id, .size = sizeof(struct ebt_pernet), }; -- cgit v1.2.3 From 92c603fa07bc0d6a17345de3ad7954730b8de44b Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Wed, 6 May 2026 12:07:19 +0200 Subject: netfilter: ebtables: close dangling table module init race sashiko reported for a related patch: In modules like iptable_raw.c, [..], if register_pernet_subsys() fails, the rollback might call kfree(rawtable_ops) before [..] During this window, could a concurrent userspace process find the globally visible template, trigger table_init(), [..] The table init functions must always register the template last. Otherwise, set/getsockopt can instantiate a table in a namespace while the required pernet ops (contain the destructor) isn't available. This change is also required in x_tables, handled in followup change. Fixes: 87663c39f898 ("netfilter: ebtables: do not hook tables by default") Reviewed-by: Tristan Madani Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- net/bridge/netfilter/ebtable_broute.c | 12 +++++------- net/bridge/netfilter/ebtable_filter.c | 12 +++++------- net/bridge/netfilter/ebtable_nat.c | 10 ++++------ 3 files changed, 14 insertions(+), 20 deletions(-) (limited to 'net') diff --git a/net/bridge/netfilter/ebtable_broute.c b/net/bridge/netfilter/ebtable_broute.c index e6f9e343b41f..f05c79f215ea 100644 --- a/net/bridge/netfilter/ebtable_broute.c +++ b/net/bridge/netfilter/ebtable_broute.c @@ -112,18 +112,16 @@ static struct pernet_operations broute_net_ops = { static int __init ebtable_broute_init(void) { - int ret = ebt_register_template(&broute_table, broute_table_init); + int ret = register_pernet_subsys(&broute_net_ops); if (ret) return ret; - ret = register_pernet_subsys(&broute_net_ops); - if (ret) { - ebt_unregister_template(&broute_table); - return ret; - } + ret = ebt_register_template(&broute_table, broute_table_init); + if (ret) + unregister_pernet_subsys(&broute_net_ops); - return 0; + return ret; } static void __exit ebtable_broute_fini(void) diff --git a/net/bridge/netfilter/ebtable_filter.c b/net/bridge/netfilter/ebtable_filter.c index 02b6501c15a5..0fc03b07e62a 100644 --- a/net/bridge/netfilter/ebtable_filter.c +++ b/net/bridge/netfilter/ebtable_filter.c @@ -93,18 +93,16 @@ static struct pernet_operations frame_filter_net_ops = { static int __init ebtable_filter_init(void) { - int ret = ebt_register_template(&frame_filter, frame_filter_table_init); + int ret = register_pernet_subsys(&frame_filter_net_ops); if (ret) return ret; - ret = register_pernet_subsys(&frame_filter_net_ops); - if (ret) { - ebt_unregister_template(&frame_filter); - return ret; - } + ret = ebt_register_template(&frame_filter, frame_filter_table_init); + if (ret) + unregister_pernet_subsys(&frame_filter_net_ops); - return 0; + return ret; } static void __exit ebtable_filter_fini(void) diff --git a/net/bridge/netfilter/ebtable_nat.c b/net/bridge/netfilter/ebtable_nat.c index 9985a82555c4..8a10375d8909 100644 --- a/net/bridge/netfilter/ebtable_nat.c +++ b/net/bridge/netfilter/ebtable_nat.c @@ -93,16 +93,14 @@ static struct pernet_operations frame_nat_net_ops = { static int __init ebtable_nat_init(void) { - int ret = ebt_register_template(&frame_nat, frame_nat_table_init); + int ret = register_pernet_subsys(&frame_nat_net_ops); if (ret) return ret; - ret = register_pernet_subsys(&frame_nat_net_ops); - if (ret) { - ebt_unregister_template(&frame_nat); - return ret; - } + ret = ebt_register_template(&frame_nat, frame_nat_table_init); + if (ret) + unregister_pernet_subsys(&frame_nat_net_ops); return ret; } -- cgit v1.2.3 From 16bc4b6686b2c112c10e67d6b493adc3607256d3 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Wed, 6 May 2026 12:07:20 +0200 Subject: netfilter: x_tables: close dangling table module init race Similar to the previous ebtables patch: template add exposes the table to userspace, we must do this last to rnsure the pernet ops are set up (contain the destructors). Fixes: fdacd57c79b7 ("netfilter: x_tables: never register tables by default") Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- net/ipv4/netfilter/arptable_filter.c | 23 ++++++++++++----------- net/ipv4/netfilter/iptable_filter.c | 23 ++++++++++++----------- net/ipv4/netfilter/iptable_mangle.c | 25 +++++++++++++------------ net/ipv4/netfilter/iptable_raw.c | 22 +++++++++++----------- net/ipv4/netfilter/iptable_security.c | 23 ++++++++++++----------- net/ipv6/netfilter/ip6table_filter.c | 22 +++++++++++----------- net/ipv6/netfilter/ip6table_mangle.c | 23 ++++++++++++----------- net/ipv6/netfilter/ip6table_raw.c | 20 ++++++++++---------- net/ipv6/netfilter/ip6table_security.c | 23 ++++++++++++----------- 9 files changed, 105 insertions(+), 99 deletions(-) (limited to 'net') diff --git a/net/ipv4/netfilter/arptable_filter.c b/net/ipv4/netfilter/arptable_filter.c index 382345567a60..370b635e3523 100644 --- a/net/ipv4/netfilter/arptable_filter.c +++ b/net/ipv4/netfilter/arptable_filter.c @@ -58,25 +58,26 @@ static struct pernet_operations arptable_filter_net_ops = { static int __init arptable_filter_init(void) { - int ret = xt_register_template(&packet_filter, - arptable_filter_table_init); - - if (ret < 0) - return ret; + int ret; arpfilter_ops = xt_hook_ops_alloc(&packet_filter, arpt_do_table); - if (IS_ERR(arpfilter_ops)) { - xt_unregister_template(&packet_filter); + if (IS_ERR(arpfilter_ops)) return PTR_ERR(arpfilter_ops); - } ret = register_pernet_subsys(&arptable_filter_net_ops); + if (ret < 0) + goto err_free; + + ret = xt_register_template(&packet_filter, + arptable_filter_table_init); if (ret < 0) { - xt_unregister_template(&packet_filter); - kfree(arpfilter_ops); - return ret; + unregister_pernet_subsys(&arptable_filter_net_ops); + goto err_free; } + return 0; +err_free: + kfree(arpfilter_ops); return ret; } diff --git a/net/ipv4/netfilter/iptable_filter.c b/net/ipv4/netfilter/iptable_filter.c index 0dea754a9120..672d7da1071d 100644 --- a/net/ipv4/netfilter/iptable_filter.c +++ b/net/ipv4/netfilter/iptable_filter.c @@ -77,26 +77,27 @@ static struct pernet_operations iptable_filter_net_ops = { static int __init iptable_filter_init(void) { - int ret = xt_register_template(&packet_filter, - iptable_filter_table_init); - - if (ret < 0) - return ret; + int ret; filter_ops = xt_hook_ops_alloc(&packet_filter, ipt_do_table); - if (IS_ERR(filter_ops)) { - xt_unregister_template(&packet_filter); + if (IS_ERR(filter_ops)) return PTR_ERR(filter_ops); - } ret = register_pernet_subsys(&iptable_filter_net_ops); + if (ret < 0) + goto err_free; + + ret = xt_register_template(&packet_filter, + iptable_filter_table_init); if (ret < 0) { - xt_unregister_template(&packet_filter); - kfree(filter_ops); - return ret; + unregister_pernet_subsys(&iptable_filter_net_ops); + goto err_free; } return 0; +err_free: + kfree(filter_ops); + return ret; } static void __exit iptable_filter_fini(void) diff --git a/net/ipv4/netfilter/iptable_mangle.c b/net/ipv4/netfilter/iptable_mangle.c index 4d3b12492308..13d25d9a4610 100644 --- a/net/ipv4/netfilter/iptable_mangle.c +++ b/net/ipv4/netfilter/iptable_mangle.c @@ -111,25 +111,26 @@ static struct pernet_operations iptable_mangle_net_ops = { static int __init iptable_mangle_init(void) { - int ret = xt_register_template(&packet_mangler, - iptable_mangle_table_init); - if (ret < 0) - return ret; + int ret; mangle_ops = xt_hook_ops_alloc(&packet_mangler, iptable_mangle_hook); - if (IS_ERR(mangle_ops)) { - xt_unregister_template(&packet_mangler); - ret = PTR_ERR(mangle_ops); - return ret; - } + if (IS_ERR(mangle_ops)) + return PTR_ERR(mangle_ops); ret = register_pernet_subsys(&iptable_mangle_net_ops); + if (ret < 0) + goto err_free; + + ret = xt_register_template(&packet_mangler, + iptable_mangle_table_init); if (ret < 0) { - xt_unregister_template(&packet_mangler); - kfree(mangle_ops); - return ret; + unregister_pernet_subsys(&iptable_mangle_net_ops); + goto err_free; } + return 0; +err_free: + kfree(mangle_ops); return ret; } diff --git a/net/ipv4/netfilter/iptable_raw.c b/net/ipv4/netfilter/iptable_raw.c index 6f7afec7954b..2745c22f4034 100644 --- a/net/ipv4/netfilter/iptable_raw.c +++ b/net/ipv4/netfilter/iptable_raw.c @@ -77,24 +77,24 @@ static int __init iptable_raw_init(void) pr_info("Enabling raw table before defrag\n"); } - ret = xt_register_template(table, - iptable_raw_table_init); - if (ret < 0) - return ret; - rawtable_ops = xt_hook_ops_alloc(table, ipt_do_table); - if (IS_ERR(rawtable_ops)) { - xt_unregister_template(table); + if (IS_ERR(rawtable_ops)) return PTR_ERR(rawtable_ops); - } ret = register_pernet_subsys(&iptable_raw_net_ops); + if (ret < 0) + goto err_free; + + ret = xt_register_template(table, + iptable_raw_table_init); if (ret < 0) { - xt_unregister_template(table); - kfree(rawtable_ops); - return ret; + unregister_pernet_subsys(&iptable_raw_net_ops); + goto err_free; } + return 0; +err_free: + kfree(rawtable_ops); return ret; } diff --git a/net/ipv4/netfilter/iptable_security.c b/net/ipv4/netfilter/iptable_security.c index 81175c20ccbe..491894511c54 100644 --- a/net/ipv4/netfilter/iptable_security.c +++ b/net/ipv4/netfilter/iptable_security.c @@ -65,25 +65,26 @@ static struct pernet_operations iptable_security_net_ops = { static int __init iptable_security_init(void) { - int ret = xt_register_template(&security_table, - iptable_security_table_init); - - if (ret < 0) - return ret; + int ret; sectbl_ops = xt_hook_ops_alloc(&security_table, ipt_do_table); - if (IS_ERR(sectbl_ops)) { - xt_unregister_template(&security_table); + if (IS_ERR(sectbl_ops)) return PTR_ERR(sectbl_ops); - } ret = register_pernet_subsys(&iptable_security_net_ops); + if (ret < 0) + goto err_free; + + ret = xt_register_template(&security_table, + iptable_security_table_init); if (ret < 0) { - xt_unregister_template(&security_table); - kfree(sectbl_ops); - return ret; + unregister_pernet_subsys(&iptable_security_net_ops); + goto err_free; } + return 0; +err_free: + kfree(sectbl_ops); return ret; } diff --git a/net/ipv6/netfilter/ip6table_filter.c b/net/ipv6/netfilter/ip6table_filter.c index cf561919bde8..b074fc477676 100644 --- a/net/ipv6/netfilter/ip6table_filter.c +++ b/net/ipv6/netfilter/ip6table_filter.c @@ -76,25 +76,25 @@ static struct pernet_operations ip6table_filter_net_ops = { static int __init ip6table_filter_init(void) { - int ret = xt_register_template(&packet_filter, - ip6table_filter_table_init); - - if (ret < 0) - return ret; + int ret; filter_ops = xt_hook_ops_alloc(&packet_filter, ip6t_do_table); - if (IS_ERR(filter_ops)) { - xt_unregister_template(&packet_filter); + if (IS_ERR(filter_ops)) return PTR_ERR(filter_ops); - } ret = register_pernet_subsys(&ip6table_filter_net_ops); + if (ret < 0) + goto err_free; + + ret = xt_register_template(&packet_filter, ip6table_filter_table_init); if (ret < 0) { - xt_unregister_template(&packet_filter); - kfree(filter_ops); - return ret; + unregister_pernet_subsys(&ip6table_filter_net_ops); + goto err_free; } + return 0; +err_free: + kfree(filter_ops); return ret; } diff --git a/net/ipv6/netfilter/ip6table_mangle.c b/net/ipv6/netfilter/ip6table_mangle.c index 1a758f2bc537..e6ee036a9b2c 100644 --- a/net/ipv6/netfilter/ip6table_mangle.c +++ b/net/ipv6/netfilter/ip6table_mangle.c @@ -104,25 +104,26 @@ static struct pernet_operations ip6table_mangle_net_ops = { static int __init ip6table_mangle_init(void) { - int ret = xt_register_template(&packet_mangler, - ip6table_mangle_table_init); - - if (ret < 0) - return ret; + int ret; mangle_ops = xt_hook_ops_alloc(&packet_mangler, ip6table_mangle_hook); - if (IS_ERR(mangle_ops)) { - xt_unregister_template(&packet_mangler); + if (IS_ERR(mangle_ops)) return PTR_ERR(mangle_ops); - } ret = register_pernet_subsys(&ip6table_mangle_net_ops); + if (ret < 0) + goto err_free; + + ret = xt_register_template(&packet_mangler, + ip6table_mangle_table_init); if (ret < 0) { - xt_unregister_template(&packet_mangler); - kfree(mangle_ops); - return ret; + unregister_pernet_subsys(&ip6table_mangle_net_ops); + goto err_free; } + return 0; +err_free: + kfree(mangle_ops); return ret; } diff --git a/net/ipv6/netfilter/ip6table_raw.c b/net/ipv6/netfilter/ip6table_raw.c index 923455921c1d..3b161ee875bc 100644 --- a/net/ipv6/netfilter/ip6table_raw.c +++ b/net/ipv6/netfilter/ip6table_raw.c @@ -75,24 +75,24 @@ static int __init ip6table_raw_init(void) pr_info("Enabling raw table before defrag\n"); } - ret = xt_register_template(table, ip6table_raw_table_init); - if (ret < 0) - return ret; - /* Register hooks */ rawtable_ops = xt_hook_ops_alloc(table, ip6t_do_table); - if (IS_ERR(rawtable_ops)) { - xt_unregister_template(table); + if (IS_ERR(rawtable_ops)) return PTR_ERR(rawtable_ops); - } ret = register_pernet_subsys(&ip6table_raw_net_ops); + if (ret < 0) + goto err_free; + + ret = xt_register_template(table, ip6table_raw_table_init); if (ret < 0) { - kfree(rawtable_ops); - xt_unregister_template(table); - return ret; + unregister_pernet_subsys(&ip6table_raw_net_ops); + goto err_free; } + return 0; +err_free: + kfree(rawtable_ops); return ret; } diff --git a/net/ipv6/netfilter/ip6table_security.c b/net/ipv6/netfilter/ip6table_security.c index c44834d93fc7..4bd5d97b8ab6 100644 --- a/net/ipv6/netfilter/ip6table_security.c +++ b/net/ipv6/netfilter/ip6table_security.c @@ -64,25 +64,26 @@ static struct pernet_operations ip6table_security_net_ops = { static int __init ip6table_security_init(void) { - int ret = xt_register_template(&security_table, - ip6table_security_table_init); - - if (ret < 0) - return ret; + int ret; sectbl_ops = xt_hook_ops_alloc(&security_table, ip6t_do_table); - if (IS_ERR(sectbl_ops)) { - xt_unregister_template(&security_table); + if (IS_ERR(sectbl_ops)) return PTR_ERR(sectbl_ops); - } ret = register_pernet_subsys(&ip6table_security_net_ops); + if (ret < 0) + goto err_free; + + ret = xt_register_template(&security_table, + ip6table_security_table_init); if (ret < 0) { - kfree(sectbl_ops); - xt_unregister_template(&security_table); - return ret; + unregister_pernet_subsys(&ip6table_security_net_ops); + goto err_free; } + return 0; +err_free: + kfree(sectbl_ops); return ret; } -- cgit v1.2.3 From 27414ff1b287ea9a2a11675149ec28e05539f3cc Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 7 May 2026 11:19:22 +0200 Subject: netfilter: bridge: eb_tables: close module init race sashiko reports for unrelated patch: Does the core ebtables initialization in ebtables.c suffer from a similar race? Once nf_register_sockopt() completes, the sockopts are exposed globally. sockopt has to be registered last, just like in ip/ip6/arptables. Fixes: 5b53951cfc85 ("netfilter: ebtables: use net_generic infra") Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- net/bridge/netfilter/ebtables.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/bridge/netfilter/ebtables.c b/net/bridge/netfilter/ebtables.c index 3578ffbc14ae..b9f4daac09af 100644 --- a/net/bridge/netfilter/ebtables.c +++ b/net/bridge/netfilter/ebtables.c @@ -2583,19 +2583,20 @@ static int __init ebtables_init(void) { int ret; - ret = xt_register_target(&ebt_standard_target); + ret = register_pernet_subsys(&ebt_net_ops); if (ret < 0) return ret; - ret = nf_register_sockopt(&ebt_sockopts); + + ret = xt_register_target(&ebt_standard_target); if (ret < 0) { - xt_unregister_target(&ebt_standard_target); + unregister_pernet_subsys(&ebt_net_ops); return ret; } - ret = register_pernet_subsys(&ebt_net_ops); + ret = nf_register_sockopt(&ebt_sockopts); if (ret < 0) { - nf_unregister_sockopt(&ebt_sockopts); xt_unregister_target(&ebt_standard_target); + unregister_pernet_subsys(&ebt_net_ops); return ret; } -- cgit v1.2.3 From dcb0f9aefdd604d36710fda53c25bd7cf4a3e37a Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Thu, 7 May 2026 13:00:28 +0200 Subject: netfilter: nf_conntrack_expect: restore helper propagation via expectation A recent series to fix expectations broke helper propagation via expectation, this mechanism is used by the sip and h323 helper. This also propagates the conntrack helper to expected connections. I changed semantics of exp->helper which now tells us the actual helper that created the expectation. Add an explicit assign_helper field to expectations for this purpose and update helpers to use it. Restore this feature for userspace conntrack helper via ctnetlink nfqueue integration so it is again possible to attach a helper to an expectation, where it makes sense. This is not restored via ctnetlink expectation creation as there is no client for such feature. Use the expectation layer 4 protocol number for the helper lookup for consistency. Make sure the expectation using this helper propagation mechanism also go away when the helper is unregistered. Fixes: 9c42bc9db90a ("netfilter: nf_conntrack_expect: honor expectation helper field") Fixes: 917b61fa2042 ("netfilter: ctnetlink: ignore explicit helper on new expectations") Reported-by: Ilya Maximets Tested-by: Ilya Maximets Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_conntrack_broadcast.c | 1 + net/netfilter/nf_conntrack_core.c | 7 +++++-- net/netfilter/nf_conntrack_expect.c | 1 + net/netfilter/nf_conntrack_h323_main.c | 12 ++++++------ net/netfilter/nf_conntrack_helper.c | 5 +++++ net/netfilter/nf_conntrack_netlink.c | 18 ++++++++++++++++-- net/netfilter/nf_conntrack_sip.c | 2 +- 7 files changed, 35 insertions(+), 11 deletions(-) (limited to 'net') diff --git a/net/netfilter/nf_conntrack_broadcast.c b/net/netfilter/nf_conntrack_broadcast.c index 4f39bf7c843f..75e53fde6b29 100644 --- a/net/netfilter/nf_conntrack_broadcast.c +++ b/net/netfilter/nf_conntrack_broadcast.c @@ -72,6 +72,7 @@ int nf_conntrack_broadcast_help(struct sk_buff *skb, exp->flags = NF_CT_EXPECT_PERMANENT; exp->class = NF_CT_EXPECT_CLASS_DEFAULT; rcu_assign_pointer(exp->helper, helper); + rcu_assign_pointer(exp->assign_helper, NULL); write_pnet(&exp->net, net); #ifdef CONFIG_NF_CONNTRACK_ZONES exp->zone = ct->zone; diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index b08189226320..8ba5b22a1eef 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -1811,14 +1811,17 @@ init_conntrack(struct net *net, struct nf_conn *tmpl, spin_lock_bh(&nf_conntrack_expect_lock); exp = nf_ct_find_expectation(net, zone, tuple, !tmpl || nf_ct_is_confirmed(tmpl)); if (exp) { + struct nf_conntrack_helper *assign_helper; + /* Welcome, Mr. Bond. We've been expecting you... */ __set_bit(IPS_EXPECTED_BIT, &ct->status); /* exp->master safe, refcnt bumped in nf_ct_find_expectation */ ct->master = exp->master; - if (exp->helper) { + assign_helper = rcu_dereference(exp->assign_helper); + if (assign_helper) { help = nf_ct_helper_ext_add(ct, GFP_ATOMIC); if (help) - rcu_assign_pointer(help->helper, exp->helper); + rcu_assign_pointer(help->helper, assign_helper); } #ifdef CONFIG_NF_CONNTRACK_MARK diff --git a/net/netfilter/nf_conntrack_expect.c b/net/netfilter/nf_conntrack_expect.c index 24d0576d84b7..8e943efbdf0a 100644 --- a/net/netfilter/nf_conntrack_expect.c +++ b/net/netfilter/nf_conntrack_expect.c @@ -344,6 +344,7 @@ void nf_ct_expect_init(struct nf_conntrack_expect *exp, unsigned int class, helper = rcu_dereference(help->helper); rcu_assign_pointer(exp->helper, helper); + rcu_assign_pointer(exp->assign_helper, NULL); write_pnet(&exp->net, net); #ifdef CONFIG_NF_CONNTRACK_ZONES exp->zone = ct->zone; diff --git a/net/netfilter/nf_conntrack_h323_main.c b/net/netfilter/nf_conntrack_h323_main.c index 3f5c50455b71..b2fe6554b9cf 100644 --- a/net/netfilter/nf_conntrack_h323_main.c +++ b/net/netfilter/nf_conntrack_h323_main.c @@ -643,7 +643,7 @@ static int expect_h245(struct sk_buff *skb, struct nf_conn *ct, &ct->tuplehash[!dir].tuple.src.u3, &ct->tuplehash[!dir].tuple.dst.u3, IPPROTO_TCP, NULL, &port); - rcu_assign_pointer(exp->helper, &nf_conntrack_helper_h245); + rcu_assign_pointer(exp->assign_helper, &nf_conntrack_helper_h245); nathook = rcu_dereference(nfct_h323_nat_hook); if (memcmp(&ct->tuplehash[dir].tuple.src.u3, @@ -767,7 +767,7 @@ static int expect_callforwarding(struct sk_buff *skb, nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, nf_ct_l3num(ct), &ct->tuplehash[!dir].tuple.src.u3, &addr, IPPROTO_TCP, NULL, &port); - rcu_assign_pointer(exp->helper, nf_conntrack_helper_q931); + rcu_assign_pointer(exp->assign_helper, nf_conntrack_helper_q931); nathook = rcu_dereference(nfct_h323_nat_hook); if (memcmp(&ct->tuplehash[dir].tuple.src.u3, @@ -1234,7 +1234,7 @@ static int expect_q931(struct sk_buff *skb, struct nf_conn *ct, &ct->tuplehash[!dir].tuple.src.u3 : NULL, &ct->tuplehash[!dir].tuple.dst.u3, IPPROTO_TCP, NULL, &port); - rcu_assign_pointer(exp->helper, nf_conntrack_helper_q931); + rcu_assign_pointer(exp->assign_helper, nf_conntrack_helper_q931); exp->flags = NF_CT_EXPECT_PERMANENT; /* Accept multiple calls */ nathook = rcu_dereference(nfct_h323_nat_hook); @@ -1306,7 +1306,7 @@ static int process_gcf(struct sk_buff *skb, struct nf_conn *ct, nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, nf_ct_l3num(ct), &ct->tuplehash[!dir].tuple.src.u3, &addr, IPPROTO_UDP, NULL, &port); - rcu_assign_pointer(exp->helper, nf_conntrack_helper_ras); + rcu_assign_pointer(exp->assign_helper, nf_conntrack_helper_ras); if (nf_ct_expect_related(exp, 0) == 0) { pr_debug("nf_ct_ras: expect RAS "); @@ -1523,7 +1523,7 @@ static int process_acf(struct sk_buff *skb, struct nf_conn *ct, &ct->tuplehash[!dir].tuple.src.u3, &addr, IPPROTO_TCP, NULL, &port); exp->flags = NF_CT_EXPECT_PERMANENT; - rcu_assign_pointer(exp->helper, nf_conntrack_helper_q931); + rcu_assign_pointer(exp->assign_helper, nf_conntrack_helper_q931); if (nf_ct_expect_related(exp, 0) == 0) { pr_debug("nf_ct_ras: expect Q.931 "); @@ -1577,7 +1577,7 @@ static int process_lcf(struct sk_buff *skb, struct nf_conn *ct, &ct->tuplehash[!dir].tuple.src.u3, &addr, IPPROTO_TCP, NULL, &port); exp->flags = NF_CT_EXPECT_PERMANENT; - rcu_assign_pointer(exp->helper, nf_conntrack_helper_q931); + rcu_assign_pointer(exp->assign_helper, nf_conntrack_helper_q931); if (nf_ct_expect_related(exp, 0) == 0) { pr_debug("nf_ct_ras: expect Q.931 "); diff --git a/net/netfilter/nf_conntrack_helper.c b/net/netfilter/nf_conntrack_helper.c index a715304a53d8..b594cd244fe1 100644 --- a/net/netfilter/nf_conntrack_helper.c +++ b/net/netfilter/nf_conntrack_helper.c @@ -400,6 +400,11 @@ static bool expect_iter_me(struct nf_conntrack_expect *exp, void *data) this = rcu_dereference_protected(exp->helper, lockdep_is_held(&nf_conntrack_expect_lock)); + if (this == me) + return true; + + this = rcu_dereference_protected(exp->assign_helper, + lockdep_is_held(&nf_conntrack_expect_lock)); return this == me; } diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index eda5fe4a75c8..d7209d124111 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -2634,6 +2634,7 @@ static const struct nla_policy exp_nla_policy[CTA_EXPECT_MAX+1] = { static struct nf_conntrack_expect * ctnetlink_alloc_expect(const struct nlattr *const cda[], struct nf_conn *ct, + const struct nf_conntrack_helper *assign_helper, struct nf_conntrack_tuple *tuple, struct nf_conntrack_tuple *mask); @@ -2860,6 +2861,7 @@ static int ctnetlink_glue_attach_expect(const struct nlattr *attr, struct nf_conn *ct, u32 portid, u32 report) { + struct nf_conntrack_helper *assign_helper = NULL; struct nlattr *cda[CTA_EXPECT_MAX+1]; struct nf_conntrack_tuple tuple, mask; struct nf_conntrack_expect *exp; @@ -2875,8 +2877,18 @@ ctnetlink_glue_attach_expect(const struct nlattr *attr, struct nf_conn *ct, if (err < 0) return err; + if (cda[CTA_EXPECT_HELP_NAME]) { + const char *helpname = nla_data(cda[CTA_EXPECT_HELP_NAME]); + + assign_helper = __nf_conntrack_helper_find(helpname, + nf_ct_l3num(ct), + tuple.dst.protonum); + if (!assign_helper) + return -EOPNOTSUPP; + } + exp = ctnetlink_alloc_expect((const struct nlattr * const *)cda, ct, - &tuple, &mask); + assign_helper, &tuple, &mask); if (IS_ERR(exp)) return PTR_ERR(exp); @@ -3515,6 +3527,7 @@ ctnetlink_parse_expect_nat(const struct nlattr *attr, static struct nf_conntrack_expect * ctnetlink_alloc_expect(const struct nlattr * const cda[], struct nf_conn *ct, + const struct nf_conntrack_helper *assign_helper, struct nf_conntrack_tuple *tuple, struct nf_conntrack_tuple *mask) { @@ -3568,6 +3581,7 @@ ctnetlink_alloc_expect(const struct nlattr * const cda[], struct nf_conn *ct, exp->zone = ct->zone; #endif rcu_assign_pointer(exp->helper, helper); + rcu_assign_pointer(exp->assign_helper, assign_helper); exp->tuple = *tuple; exp->mask.src.u3 = mask->src.u3; exp->mask.src.u.all = mask->src.u.all; @@ -3623,7 +3637,7 @@ ctnetlink_create_expect(struct net *net, ct = nf_ct_tuplehash_to_ctrack(h); rcu_read_lock(); - exp = ctnetlink_alloc_expect(cda, ct, &tuple, &mask); + exp = ctnetlink_alloc_expect(cda, ct, NULL, &tuple, &mask); if (IS_ERR(exp)) { err = PTR_ERR(exp); goto err_rcu; diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c index 1eb55907d470..d24bfa9e8234 100644 --- a/net/netfilter/nf_conntrack_sip.c +++ b/net/netfilter/nf_conntrack_sip.c @@ -1383,7 +1383,7 @@ static int process_register_request(struct sk_buff *skb, unsigned int protoff, nf_ct_expect_init(exp, SIP_EXPECT_SIGNALLING, nf_ct_l3num(ct), saddr, &daddr, proto, NULL, &port); exp->timeout.expires = sip_timeout * HZ; - rcu_assign_pointer(exp->helper, helper); + rcu_assign_pointer(exp->assign_helper, helper); exp->flags = NF_CT_EXPECT_PERMANENT | NF_CT_EXPECT_INACTIVE; hooks = rcu_dereference(nf_nat_sip_hooks); -- cgit v1.2.3 From d8ef54c83ad70b81735b506431affadd2f720aa1 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Thu, 7 May 2026 23:57:55 +0200 Subject: netfilter: ctnetlink: check tuple and mask in expectations created via nfqueue Ensure the expectation tuple and mask attributes are present in netlink message, otherwise null-ptr-deref is possible. Fixes: bd0779370588 ("netfilter: nfnetlink_queue: allow to attach expectations to conntracks") Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_conntrack_netlink.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'net') diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index d7209d124111..befa7e83ee49 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -2872,6 +2872,9 @@ ctnetlink_glue_attach_expect(const struct nlattr *attr, struct nf_conn *ct, if (err < 0) return err; + if (!cda[CTA_EXPECT_TUPLE] || !cda[CTA_EXPECT_MASK]) + return -EINVAL; + err = ctnetlink_glue_exp_parse((const struct nlattr * const *)cda, ct, &tuple, &mask); if (err < 0) -- cgit v1.2.3 From eb6317739b1ea3ab28791e1f91b24781905fa815 Mon Sep 17 00:00:00 2001 From: Li Xiasong Date: Thu, 7 May 2026 22:04:22 +0800 Subject: netfilter: nf_conntrack_sip: get helper before allocating expectation process_register_request() allocates an expectation and then checks whether a conntrack helper is available. If helper lookup fails, the function returns early and the allocated expectation is left behind. Reorder the code to fetch and validate helper before calling nf_ct_expect_alloc(). This keeps the logic simpler and removes the leak path while preserving existing behavior. Fixes: e14575fa7529 ("netfilter: nf_conntrack: use rcu accessors where needed") Cc: stable@vger.kernel.org Signed-off-by: Li Xiasong Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_conntrack_sip.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c index d24bfa9e8234..e69941f1a101 100644 --- a/net/netfilter/nf_conntrack_sip.c +++ b/net/netfilter/nf_conntrack_sip.c @@ -1366,6 +1366,10 @@ static int process_register_request(struct sk_buff *skb, unsigned int protoff, goto store_cseq; } + helper = rcu_dereference(nfct_help(ct)->helper); + if (!helper) + return NF_DROP; + exp = nf_ct_expect_alloc(ct); if (!exp) { nf_ct_helper_log(skb, ct, "cannot alloc expectation"); @@ -1376,10 +1380,6 @@ static int process_register_request(struct sk_buff *skb, unsigned int protoff, if (sip_direct_signalling) saddr = &ct->tuplehash[!dir].tuple.src.u3; - helper = rcu_dereference(nfct_help(ct)->helper); - if (!helper) - return NF_DROP; - nf_ct_expect_init(exp, SIP_EXPECT_SIGNALLING, nf_ct_l3num(ct), saddr, &daddr, proto, NULL, &port); exp->timeout.expires = sip_timeout * HZ; -- cgit v1.2.3 From 19f94b6fee75b3ef7fbc06f3745b9a771a8a19a4 Mon Sep 17 00:00:00 2001 From: Li Xiasong Date: Thu, 7 May 2026 22:04:23 +0800 Subject: netfilter: nft_ct: fix missing expect put in obj eval nft_ct_expect_obj_eval() allocates an expectation and may call nf_ct_expect_related(), but never drops its local reference. Add nf_ct_expect_put(exp) before return to balance allocation. Fixes: 857b46027d6f ("netfilter: nft_ct: add ct expectations support") Cc: stable@vger.kernel.org Signed-off-by: Li Xiasong Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nft_ct.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'net') diff --git a/net/netfilter/nft_ct.c b/net/netfilter/nft_ct.c index 60ee8d932fcb..fa2cc556331c 100644 --- a/net/netfilter/nft_ct.c +++ b/net/netfilter/nft_ct.c @@ -1334,6 +1334,8 @@ static void nft_ct_expect_obj_eval(struct nft_object *obj, if (nf_ct_expect_related(exp, 0) != 0) regs->verdict.code = NF_DROP; + + nf_ct_expect_put(exp); } static const struct nla_policy nft_ct_expect_policy[NFTA_CT_EXPECT_MAX + 1] = { -- cgit v1.2.3 From 7666dbb1bacc4ba522b96740cba7283d243d16e1 Mon Sep 17 00:00:00 2001 From: John Walker Date: Thu, 7 May 2026 17:07:20 -0600 Subject: wifi: cfg80211: advance loop vars in cfg80211_merge_profile() cfg80211_merge_profile() reassembles a Multi-BSSID non-transmitted BSS profile that has been split across multiple consecutive MBSSID elements. Its while-loop calls cfg80211_get_profile_continuation(ie, ielen, mbssid_elem, sub_elem) but never advances mbssid_elem or sub_elem inside the body. Each iteration therefore searches for a continuation that follows the same fixed pair; the helper returns the same next_mbssid; and the same next_sub bytes are memcpy()'d into merged_ie at a growing offset until the buffer fills. Advance both mbssid_elem and sub_elem to the just-consumed continuation so the next call to cfg80211_get_profile_continuation() searches for a further continuation beyond it (or returns NULL when none exists). A specially-crafted malicious beacon can take advantage of this bug to cause the kernel to spend an excessive amount of time in cfg80211_merge_profile (up to as much as 2ms per beacon received), which could theoretically be abused in some way. Cc: stable@vger.kernel.org Fixes: fe806e4992c9 ("cfg80211: support profile split between elements") Signed-off-by: John Walker Link: https://patch.msgid.link/20260507230720.64783-1-johnwalker0@gmail.com Signed-off-by: Johannes Berg --- net/wireless/scan.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'net') diff --git a/net/wireless/scan.c b/net/wireless/scan.c index 328af43ef832..358cbc9e43d8 100644 --- a/net/wireless/scan.c +++ b/net/wireless/scan.c @@ -2462,6 +2462,9 @@ size_t cfg80211_merge_profile(const u8 *ie, size_t ielen, memcpy(merged_ie + copied_len, next_sub->data, next_sub->datalen); copied_len += next_sub->datalen; + + mbssid_elem = next_mbssid; + sub_elem = next_sub; } return copied_len; -- cgit v1.2.3 From f03e8583532941b07761c5429de7d50766fa3110 Mon Sep 17 00:00:00 2001 From: Jiexun Wang Date: Sun, 3 May 2026 12:28:58 +0800 Subject: batman-adv: stop caching unowned originator pointers in BAT IV BAT IV keeps the last-hop neighbor address in each neigh_node, but some paths also cache an originator pointer derived from a temporary lookup. That pointer is not owned by the neigh_node and may no longer refer to a live originator entry after purge handling runs. Stop storing the auxiliary originator pointer in the BAT IV neighbor state. When BAT IV needs the neighbor originator data, resolve it from the stored neighbor address and drop the reference again after use. Fixes: c6c8fea29769 ("net: Add batman-adv meshing protocol") Cc: stable@kernel.org Reported-by: Yuan Tan Reported-by: Yifan Wu Reported-by: Juefei Pu Reported-by: Xin Liu Signed-off-by: Jiexun Wang Signed-off-by: Ren Wei [sven: avoid bonding logic for outgoing OGM] Signed-off-by: Sven Eckelmann --- net/batman-adv/bat_iv_ogm.c | 83 ++++++++++++++++++++++++++++++++------------- 1 file changed, 59 insertions(+), 24 deletions(-) (limited to 'net') diff --git a/net/batman-adv/bat_iv_ogm.c b/net/batman-adv/bat_iv_ogm.c index 618d1889c04e..74ef7dc2b2f9 100644 --- a/net/batman-adv/bat_iv_ogm.c +++ b/net/batman-adv/bat_iv_ogm.c @@ -173,19 +173,12 @@ free_orig_node_hash: static struct batadv_neigh_node * batadv_iv_ogm_neigh_new(struct batadv_hard_iface *hard_iface, const u8 *neigh_addr, - struct batadv_orig_node *orig_node, - struct batadv_orig_node *orig_neigh) + struct batadv_orig_node *orig_node) { struct batadv_neigh_node *neigh_node; neigh_node = batadv_neigh_node_get_or_create(orig_node, hard_iface, neigh_addr); - if (!neigh_node) - goto out; - - neigh_node->orig_node = orig_neigh; - -out: return neigh_node; } @@ -906,6 +899,31 @@ static u8 batadv_iv_orig_ifinfo_sum(struct batadv_orig_node *orig_node, return sum; } +/** + * batadv_iv_ogm_neigh_ifinfo_sum() - Get bcast_own sum for a last-hop neighbor + * @bat_priv: the bat priv with all the mesh interface information + * @neigh_node: last-hop neighbor of an originator + * + * Return: Number of replied (rebroadcasted) OGMs for the originator currently + * announced by the neighbor. Returns 0 if the neighbor's originator entry is + * not available anymore. + */ +static u8 batadv_iv_ogm_neigh_ifinfo_sum(struct batadv_priv *bat_priv, + const struct batadv_neigh_node *neigh_node) +{ + struct batadv_orig_node *orig_neigh; + u8 sum; + + orig_neigh = batadv_orig_hash_find(bat_priv, neigh_node->addr); + if (!orig_neigh) + return 0; + + sum = batadv_iv_orig_ifinfo_sum(orig_neigh, neigh_node->if_incoming); + batadv_orig_node_put(orig_neigh); + + return sum; +} + /** * batadv_iv_ogm_orig_update() - use OGM to update corresponding data in an * originator @@ -975,17 +993,9 @@ batadv_iv_ogm_orig_update(struct batadv_priv *bat_priv, } if (!neigh_node) { - struct batadv_orig_node *orig_tmp; - - orig_tmp = batadv_iv_ogm_orig_get(bat_priv, ethhdr->h_source); - if (!orig_tmp) - goto unlock; - neigh_node = batadv_iv_ogm_neigh_new(if_incoming, ethhdr->h_source, - orig_node, orig_tmp); - - batadv_orig_node_put(orig_tmp); + orig_node); if (!neigh_node) goto unlock; } else { @@ -1037,10 +1047,9 @@ batadv_iv_ogm_orig_update(struct batadv_priv *bat_priv, */ if (router_ifinfo && neigh_ifinfo->bat_iv.tq_avg == router_ifinfo->bat_iv.tq_avg) { - sum_orig = batadv_iv_orig_ifinfo_sum(router->orig_node, - router->if_incoming); - sum_neigh = batadv_iv_orig_ifinfo_sum(neigh_node->orig_node, - neigh_node->if_incoming); + sum_orig = batadv_iv_ogm_neigh_ifinfo_sum(bat_priv, router); + sum_neigh = batadv_iv_ogm_neigh_ifinfo_sum(bat_priv, + neigh_node); if (sum_orig >= sum_neigh) goto out; } @@ -1106,7 +1115,6 @@ static bool batadv_iv_ogm_calc_tq(struct batadv_orig_node *orig_node, if (!neigh_node) neigh_node = batadv_iv_ogm_neigh_new(if_incoming, orig_neigh_node->orig, - orig_neigh_node, orig_neigh_node); if (!neigh_node) @@ -1302,6 +1310,32 @@ out: return ret; } +/** + * batadv_orig_to_direct_router() - get direct next hop neighbor to an orig address + * @bat_priv: the bat priv with all the mesh interface information + * @orig_addr: the originator MAC address to search the best next hop router for + * @if_outgoing: the interface where the OGM should be sent to + * + * Return: A neighbor node which is the best router towards the given originator + * address. Bonding candidates are ignored. + */ +static struct batadv_neigh_node * +batadv_orig_to_direct_router(struct batadv_priv *bat_priv, u8 *orig_addr, + struct batadv_hard_iface *if_outgoing) +{ + struct batadv_neigh_node *neigh_node; + struct batadv_orig_node *orig_node; + + orig_node = batadv_orig_hash_find(bat_priv, orig_addr); + if (!orig_node) + return NULL; + + neigh_node = batadv_orig_router_get(orig_node, if_outgoing); + batadv_orig_node_put(orig_node); + + return neigh_node; +} + /** * batadv_iv_ogm_process_per_outif() - process a batman iv OGM for an outgoing * interface @@ -1372,8 +1406,9 @@ batadv_iv_ogm_process_per_outif(const struct sk_buff *skb, int ogm_offset, router = batadv_orig_router_get(orig_node, if_outgoing); if (router) { - router_router = batadv_orig_router_get(router->orig_node, - if_outgoing); + router_router = batadv_orig_to_direct_router(bat_priv, + router->addr, + if_outgoing); router_ifinfo = batadv_neigh_ifinfo_get(router, if_outgoing); } -- cgit v1.2.3 From ce425dd05d0fe7594930a0fb103634f35ac47bb6 Mon Sep 17 00:00:00 2001 From: Sven Eckelmann Date: Wed, 6 May 2026 22:20:49 +0200 Subject: batman-adv: tp_meter: fix tp_num leak on kmalloc failure When batadv_tp_start() or batadv_tp_init_recv() fail to allocate a new tp_vars object, the previously incremented bat_priv->tp_num counter is never decremented. This causes tp_num to drift upward on each allocation failure. Since only BATADV_TP_MAX_NUM sessions can be started and the count is never reduced for these failed allocations, it causes to an exhaustion of throughput meter sessions. In worst case, no new throughput meter session can be started until the mesh interface is removed. The error handling must decrement tp_num releasing the lock and aborting the creation of an throughput meter session Cc: stable@kernel.org Fixes: 33a3bb4a3345 ("batman-adv: throughput meter implementation") Signed-off-by: Sven Eckelmann --- net/batman-adv/tp_meter.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/batman-adv/tp_meter.c b/net/batman-adv/tp_meter.c index 58ca59a2799e..066c76113fc4 100644 --- a/net/batman-adv/tp_meter.c +++ b/net/batman-adv/tp_meter.c @@ -994,6 +994,7 @@ void batadv_tp_start(struct batadv_priv *bat_priv, const u8 *dst, tp_vars = kmalloc_obj(*tp_vars, GFP_ATOMIC); if (!tp_vars) { + atomic_dec(&bat_priv->tp_num); spin_unlock_bh(&bat_priv->tp_list_lock); batadv_dbg(BATADV_DBG_TP_METER, bat_priv, "Meter: %s cannot allocate list elements\n", @@ -1366,8 +1367,10 @@ batadv_tp_init_recv(struct batadv_priv *bat_priv, } tp_vars = kmalloc_obj(*tp_vars, GFP_ATOMIC); - if (!tp_vars) + if (!tp_vars) { + atomic_dec(&bat_priv->tp_num); goto out_unlock; + } ether_addr_copy(tp_vars->other_end, icmp->orig); tp_vars->role = BATADV_TP_RECEIVER; -- cgit v1.2.3 From 4ae1709a314060a196981b344610d023ea841e57 Mon Sep 17 00:00:00 2001 From: Sven Eckelmann Date: Wed, 6 May 2026 22:20:50 +0200 Subject: batman-adv: bla: prevent use-after-free when deleting claims When batadv_bla_del_backbone_claims() removes all claims for a backbone, it does this by dropping the link entry in the hash list. This list entry itself was one of the references which need to be dropped at the same time via batadv_claim_put(). But the batadv_claim_put() must not be done before the last access to the claim object in this function. Otherwise the claim might be freed already by the batadv_claim_release() function before the list entry was dropped. Cc: stable@kernel.org Fixes: 23721387c409 ("batman-adv: add basic bridge loop avoidance code") Signed-off-by: Sven Eckelmann --- net/batman-adv/bridge_loop_avoidance.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/batman-adv/bridge_loop_avoidance.c b/net/batman-adv/bridge_loop_avoidance.c index 51fe028b9088..8b77dd2ecfa4 100644 --- a/net/batman-adv/bridge_loop_avoidance.c +++ b/net/batman-adv/bridge_loop_avoidance.c @@ -318,8 +318,8 @@ batadv_bla_del_backbone_claims(struct batadv_bla_backbone_gw *backbone_gw) if (claim->backbone_gw != backbone_gw) continue; - batadv_claim_put(claim); hlist_del_rcu(&claim->hash_entry); + batadv_claim_put(claim); } spin_unlock_bh(list_lock); } -- cgit v1.2.3 From cf6b604011591865ae39ac82de8978c1120d17af Mon Sep 17 00:00:00 2001 From: Sven Eckelmann Date: Wed, 6 May 2026 22:20:51 +0200 Subject: batman-adv: bla: only purge non-released claims When batadv_bla_purge_claims() goes through the list of claims, it is only traversing the hash list with an rcu_read_lock(). Due to a potential parallel batadv_claim_put(), it can happen that it encounters a claim which was actually in the process of being released+freed by batadv_claim_release(). In this case, backbone_gw is set to NULL before the delayed RCU kfree is started. Calling batadv_bla_claim_get_backbone_gw() is then no longer allowed because it would cause a NULL-ptr derefence. To avoid this, only claims with a valid reference counter must be purged. All others are already taken care of. Cc: stable@kernel.org Fixes: 23721387c409 ("batman-adv: add basic bridge loop avoidance code") Signed-off-by: Sven Eckelmann --- net/batman-adv/bridge_loop_avoidance.c | 8 ++++++++ 1 file changed, 8 insertions(+) (limited to 'net') diff --git a/net/batman-adv/bridge_loop_avoidance.c b/net/batman-adv/bridge_loop_avoidance.c index 8b77dd2ecfa4..879ab043d57a 100644 --- a/net/batman-adv/bridge_loop_avoidance.c +++ b/net/batman-adv/bridge_loop_avoidance.c @@ -1288,6 +1288,13 @@ static void batadv_bla_purge_claims(struct batadv_priv *bat_priv, rcu_read_lock(); hlist_for_each_entry_rcu(claim, head, hash_entry) { + /* only purge claims not currently in the process of being released. + * Such claims could otherwise have a NULL-ptr backbone_gw set because + * they already went through batadv_claim_release() + */ + if (!kref_get_unless_zero(&claim->refcount)) + continue; + backbone_gw = batadv_bla_claim_get_backbone_gw(claim); if (now) goto purge_now; @@ -1313,6 +1320,7 @@ purge_now: claim->addr, claim->vid); skip: batadv_backbone_gw_put(backbone_gw); + batadv_claim_put(claim); } rcu_read_unlock(); } -- cgit v1.2.3 From ba9d20ee9076dac32c371116bacbe72480eb356c Mon Sep 17 00:00:00 2001 From: Sven Eckelmann Date: Wed, 6 May 2026 22:20:52 +0200 Subject: batman-adv: bla: put backbone reference on failed claim hash insert When batadv_bla_add_claim() fails to insert a new claim into the hash, it leaked a reference to the backbone_gw for which the claim was intended. Call batadv_backbone_gw_put() on the error path to release the reference and avoid leaking the backbone_gw object. Cc: stable@kernel.org Fixes: 3db0decf1185 ("batman-adv: Fix non-atomic bla_claim::backbone_gw access") Signed-off-by: Sven Eckelmann --- net/batman-adv/bridge_loop_avoidance.c | 1 + 1 file changed, 1 insertion(+) (limited to 'net') diff --git a/net/batman-adv/bridge_loop_avoidance.c b/net/batman-adv/bridge_loop_avoidance.c index 879ab043d57a..cec11f1251d6 100644 --- a/net/batman-adv/bridge_loop_avoidance.c +++ b/net/batman-adv/bridge_loop_avoidance.c @@ -723,6 +723,7 @@ static void batadv_bla_add_claim(struct batadv_priv *bat_priv, if (unlikely(hash_added != 0)) { /* only local changes happened. */ + batadv_backbone_gw_put(backbone_gw); kfree(claim); return; } -- cgit v1.2.3 From 481c2265286ef302327c93403a8cf7b3fe4506d0 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Mon, 4 May 2026 21:04:48 +0000 Subject: bpf: tcp: Fix type confusion in bpf_tcp_sock(). bpf_tcp_sock() only checks if sk->sk_protocol is IPPROTO_TCP, but RAW socket can bypass it: socket(AF_INET, SOCK_RAW, IPPROTO_TCP) Calling bpf_setsockopt() in SOCKOPT prog triggers out-of-bounds access to another slab object. [0] Let's use sk_is_tcp(). [0]: BUG: KASAN: slab-out-of-bounds in sol_tcp_sockopt (net/core/filter.c:5519) Read of size 8 at addr ffff88801083d760 by task test_progs/1259 CPU: 1 UID: 0 PID: 1259 Comm: test_progs Tainted: G OE 7.0.0-11175-gb5c111f4967b #1 PREEMPT(full) Tainted: [O]=OOT_MODULE, [E]=UNSIGNED_MODULE Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.17.0-debian-1.17.0-1 04/01/2014 Call Trace: dump_stack_lvl (lib/dump_stack.c:94 lib/dump_stack.c:120) print_report (mm/kasan/report.c:378 mm/kasan/report.c:482) kasan_report (mm/kasan/report.c:595) sol_tcp_sockopt (net/core/filter.c:5519) __bpf_getsockopt (net/core/filter.c:5633) bpf_sk_getsockopt (net/core/filter.c:5654) bpf_prog_629ba00a1601e9f2__setsockopt+0x86/0x22c __cgroup_bpf_run_filter_setsockopt (./include/linux/bpf.h:1402 ./include/linux/filter.h:722 ./include/linux/filter.h:729 kernel/bpf/cgroup.c:81 kernel/bpf/cgroup.c:2026) do_sock_setsockopt (net/socket.c:2363) __x64_sys_setsockopt (net/socket.c:2406) do_syscall_64 (arch/x86/entry/syscall_64.c:63) entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:121) RIP: 0033:0x7f85f82fe7de Code: 55 48 63 c9 48 63 ff 45 89 c9 48 89 e5 48 83 ec 08 6a 2c e8 34 69 f7 ff c9 c3 66 90 f3 0f 1e fa 49 89 ca b8 36 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 0a c3 66 0f 1f 84 00 00 00 00 00 48 8b 15 e1 RSP: 002b:00007ffe59dcecd8 EFLAGS: 00000202 ORIG_RAX: 0000000000000036 RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007f85f82fe7de RDX: 000000000000001c RSI: 0000000000000006 RDI: 000000000000000d RBP: 00007ffe59dcef20 R08: 000000000000003c R09: 0000000000000000 R10: 00007ffe59dcef00 R11: 0000000000000202 R12: 00007ffe59dcf268 R13: 0000000000000003 R14: 00007f85f9da5000 R15: 000055b2f3201400 The buggy address belongs to the object at ffff88801083d280 which belongs to the cache RAW of size 1792 The buggy address is located 1248 bytes inside of allocated 1792-byte region [ffff88801083d280, ffff88801083d980) Fixes: 655a51e536c0 ("bpf: Add struct bpf_tcp_sock and BPF_FUNC_tcp_sock") Reported-by: Damiano Melotti Signed-off-by: Kuniyuki Iwashima Signed-off-by: Martin KaFai Lau Acked-by: Daniel Borkmann Link: https://patch.msgid.link/20260504210610.180150-2-kuniyu@google.com --- net/core/filter.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/core/filter.c b/net/core/filter.c index bc96c18df4e0..cd88633f8dc1 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -7475,7 +7475,7 @@ u32 bpf_tcp_sock_convert_ctx_access(enum bpf_access_type type, BPF_CALL_1(bpf_tcp_sock, struct sock *, sk) { - if (sk_fullsock(sk) && sk->sk_protocol == IPPROTO_TCP) + if (sk_fullsock(sk) && sk_is_tcp(sk)) return (unsigned long)sk; return (unsigned long)NULL; -- cgit v1.2.3 From 7995b216a731db657f356f6ae37a42f445b9a0ec Mon Sep 17 00:00:00 2001 From: "Matthieu Baerts (NGI0)" Date: Mon, 4 May 2026 21:04:50 +0000 Subject: mptcp: bpf: Fix type confusion in bpf_mptcp_sock_from_subflow() bpf_mptcp_sock_from_subflow() only checks if sk->sk_protocol is IPPROTO_TCP, but RAW socket can bypass it: socket(AF_INET, SOCK_RAW, IPPROTO_TCP) In this case, it would NOT be valid to call sk_is_mptcp() which will assume sk is a pointer to a struct tcp_sock, and wrongly checks for: tcp_sk(sk)->is_mptcp. Fixes: 3bc253c2e652 ("bpf: Add bpf_skc_to_mptcp_sock_proto") Signed-off-by: Matthieu Baerts (NGI0) Signed-off-by: Kuniyuki Iwashima Signed-off-by: Martin KaFai Lau Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20260504210610.180150-4-kuniyu@google.com --- net/mptcp/bpf.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/mptcp/bpf.c b/net/mptcp/bpf.c index 8a16672b94e2..4cc16cbeb328 100644 --- a/net/mptcp/bpf.c +++ b/net/mptcp/bpf.c @@ -14,7 +14,7 @@ struct mptcp_sock *bpf_mptcp_sock_from_subflow(struct sock *sk) { - if (sk && sk_fullsock(sk) && sk->sk_protocol == IPPROTO_TCP && sk_is_mptcp(sk)) + if (sk && sk_fullsock(sk) && sk_is_tcp(sk) && sk_is_mptcp(sk)) return mptcp_sk(mptcp_subflow_ctx(sk)->conn); return NULL; -- cgit v1.2.3 From decb84b8383ab7acff94db208ef7ed19f9c55e1f Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Mon, 4 May 2026 21:04:51 +0000 Subject: bpf: tcp: Fix type confusion in bpf_skc_to_tcp_sock(). bpf_skc_to_tcp_sock() only checks if sk->sk_protocol is IPPROTO_TCP, but RAW socket can bypass it: socket(AF_INET, SOCK_RAW, IPPROTO_TCP) Let's use sk_is_tcp(). Fixes: 478cfbdf5f13 ("bpf: Add bpf_skc_to_{tcp, tcp_timewait, tcp_request}_sock() helpers") Signed-off-by: Kuniyuki Iwashima Signed-off-by: Martin KaFai Lau Link: https://patch.msgid.link/20260504210610.180150-5-kuniyu@google.com --- net/core/filter.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/core/filter.c b/net/core/filter.c index cd88633f8dc1..7d945dc2cb92 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -11963,7 +11963,7 @@ const struct bpf_func_proto bpf_skc_to_tcp6_sock_proto = { BPF_CALL_1(bpf_skc_to_tcp_sock, struct sock *, sk) { - if (sk && sk_fullsock(sk) && sk->sk_protocol == IPPROTO_TCP) + if (sk && sk_fullsock(sk) && sk_is_tcp(sk)) return (unsigned long)sk; return (unsigned long)NULL; -- cgit v1.2.3 From 843064b0a77eed3d6d63ffc53aeaa359672b4e12 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Mon, 4 May 2026 21:04:52 +0000 Subject: bpf: tcp: Fix type confusion in bpf_skc_to_tcp6_sock(). bpf_skc_to_tcp6_sock() only checks if sk->sk_protocol is IPPROTO_TCP and sk->sk_family is AF_INET6, but RAW socket can bypass it: socket(AF_INET6, SOCK_RAW, IPPROTO_TCP) Let's check sk->sk_type too. Fixes: af7ec1383361 ("bpf: Add bpf_skc_to_tcp6_sock() helper") Signed-off-by: Kuniyuki Iwashima Signed-off-by: Martin KaFai Lau Link: https://patch.msgid.link/20260504210610.180150-6-kuniyu@google.com --- net/core/filter.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/core/filter.c b/net/core/filter.c index 7d945dc2cb92..684922efd481 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -11947,7 +11947,7 @@ BPF_CALL_1(bpf_skc_to_tcp6_sock, struct sock *, sk) */ BTF_TYPE_EMIT(struct tcp6_sock); if (sk && sk_fullsock(sk) && sk->sk_protocol == IPPROTO_TCP && - sk->sk_family == AF_INET6) + sk->sk_type == SOCK_STREAM && sk->sk_family == AF_INET6) return (unsigned long)sk; return (unsigned long)NULL; -- cgit v1.2.3 From 1c2958e4ab1ed4594db16425dbcab33c56ea8330 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Mon, 4 May 2026 21:04:53 +0000 Subject: bpf: tcp: Fix type confusion in sol_tcp_sockopt(). sol_tcp_sockopt() only checks if sk->sk_protocol is IPPROTO_TCP, but RAW socket can bypass it: socket(AF_INET, SOCK_RAW, IPPROTO_TCP) Let's use sk_is_tcp(). Note that initially sol_tcp_sockopt() checked sk->sk_prot->setsockopt. Fixes: 2ab42c7b871f ("bpf: Check the protocol of a sock to agree the calls to bpf_setsockopt().") Signed-off-by: Kuniyuki Iwashima Signed-off-by: Martin KaFai Lau Link: https://patch.msgid.link/20260504210610.180150-7-kuniyu@google.com --- net/core/filter.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/core/filter.c b/net/core/filter.c index 684922efd481..ef0877eefaa7 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -5481,7 +5481,7 @@ static int sol_tcp_sockopt(struct sock *sk, int optname, char *optval, int *optlen, bool getopt) { - if (sk->sk_protocol != IPPROTO_TCP) + if (!sk_is_tcp(sk)) return -EINVAL; switch (optname) { -- cgit v1.2.3 From 18fc650ccd7fe3376eca89203668cfb8268f60df Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Sun, 26 Apr 2026 01:26:43 +0000 Subject: bpf: Free reuseport cBPF prog after RCU grace period. Eulgyu Kim reported the splat below with a repro. [0] The repro sets up a UDP reuseport group with a cBPF prog and replaces it with a new one while another thread is sending a UDP packet to the group. The reuseport prog is freed by sk_reuseport_prog_free(). bpf_prog_put() is called for "e"BPF prog to destruct through multiple stages while cBPF prog is freed immediately by bpf_release_orig_filter() and bpf_prog_free(). If a reuseport prog is detached from the setsockopt() path (reuseport_attach_prog() or reuseport_detach_prog()), sk_reuseport_prog_free() is called without waiting for RCU readers to complete, resulting in various bugs. Let's defer freeing the reuseport cBPF prog after one RCU grace period. Note "e"BPF prog is safe as is unless the fast path starts to touch fields destroyed in bpf_prog_put_deferred() and __bpf_prog_put_noref(). [0]: BUG: KASAN: vmalloc-out-of-bounds in reuseport_select_sock+0xedc/0x1220 net/core/sock_reuseport.c:596 Read of size 4 at addr ffffc9000051e004 by task slowme/10208 CPU: 6 UID: 1000 PID: 10208 Comm: slowme Not tainted 7.0.0-geb7ac95ff75e #32 PREEMPT(full) Hardware name: QEMU Ubuntu 24.04 PC v2 (i440FX + PIIX, arch_caps fix, 1996), BIOS 1.16.3-debian-1.16.3-2 04/01/2014 Call Trace: dump_stack_lvl+0xe8/0x150 lib/dump_stack.c:120 print_address_description mm/kasan/report.c:378 [inline] print_report+0xca/0x240 mm/kasan/report.c:482 kasan_report+0x118/0x150 mm/kasan/report.c:595 reuseport_select_sock+0xedc/0x1220 net/core/sock_reuseport.c:596 udp4_lib_lookup2+0x3bc/0x950 net/ipv4/udp.c:495 __udp4_lib_lookup+0x768/0xe20 net/ipv4/udp.c:723 __udp4_lib_lookup_skb+0x297/0x390 net/ipv4/udp.c:752 __udp4_lib_rcv+0x1312/0x2620 net/ipv4/udp.c:2752 ip_protocol_deliver_rcu+0x282/0x440 net/ipv4/ip_input.c:207 ip_local_deliver_finish+0x3bb/0x6f0 net/ipv4/ip_input.c:241 NF_HOOK+0x30c/0x3a0 include/linux/netfilter.h:318 NF_HOOK+0x30c/0x3a0 include/linux/netfilter.h:318 __netif_receive_skb_one_core net/core/dev.c:6181 [inline] __netif_receive_skb net/core/dev.c:6294 [inline] process_backlog+0xaa4/0x1960 net/core/dev.c:6645 __napi_poll+0xae/0x340 net/core/dev.c:7709 napi_poll net/core/dev.c:7772 [inline] net_rx_action+0x5d7/0xf50 net/core/dev.c:7929 handle_softirqs+0x22b/0x870 kernel/softirq.c:622 do_softirq+0x76/0xd0 kernel/softirq.c:523 __local_bh_enable_ip+0xf8/0x130 kernel/softirq.c:450 local_bh_enable include/linux/bottom_half.h:33 [inline] rcu_read_unlock_bh include/linux/rcupdate.h:924 [inline] __dev_queue_xmit+0x1dd7/0x3710 net/core/dev.c:4890 neigh_output include/net/neighbour.h:556 [inline] ip_finish_output2+0xca9/0x1070 net/ipv4/ip_output.c:237 NF_HOOK_COND include/linux/netfilter.h:307 [inline] ip_output+0x29f/0x450 net/ipv4/ip_output.c:438 ip_send_skb+0x45/0xc0 net/ipv4/ip_output.c:1508 udp_send_skb+0xb04/0x1510 net/ipv4/udp.c:1195 udp_sendmsg+0x1a71/0x2350 net/ipv4/udp.c:1485 sock_sendmsg_nosec net/socket.c:727 [inline] __sock_sendmsg net/socket.c:742 [inline] __sys_sendto+0x554/0x680 net/socket.c:2206 __do_sys_sendto net/socket.c:2213 [inline] __se_sys_sendto net/socket.c:2209 [inline] __x64_sys_sendto+0xde/0x100 net/socket.c:2209 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline] do_syscall_64+0x160/0xf80 arch/x86/entry/syscall_64.c:94 entry_SYSCALL_64_after_hwframe+0x77/0x7f RIP: 0033:0x415a2d Code: b3 66 2e 0f 1f 84 00 00 00 00 00 66 90 f3 0f 1e fa 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 b8 ff ff ff f7 d8 64 89 01 48 RSP: 002b:00007f6bc31e41e8 EFLAGS: 00000212 ORIG_RAX: 000000000000002c RAX: ffffffffffffffda RBX: 00007f6bc31e4cdc RCX: 0000000000415a2d RDX: 0000000000000001 RSI: 00007f6bc31e421f RDI: 0000000000000003 RBP: 00007f6bc31e4240 R08: 00007f6bc31e4220 R09: 0000000000000010 R10: 0000000000000000 R11: 0000000000000212 R12: 00007f6bc31e46c0 R13: ffffffffffffffb8 R14: 0000000000000000 R15: 00007ffc9b0d70b0 Fixes: 538950a1b752 ("soreuseport: setsockopt SO_ATTACH_REUSEPORT_[CE]BPF") Reported-by: Eulgyu Kim Reported-by: Taeyang Lee <0wn@theori.io> Signed-off-by: Kuniyuki Iwashima Signed-off-by: Daniel Borkmann Acked-by: Daniel Borkmann Link: https://lore.kernel.org/bpf/20260426012647.3233119-1-kuniyu@google.com --- net/core/filter.c | 15 ++++++++++++--- 1 file changed, 12 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/core/filter.c b/net/core/filter.c index ef0877eefaa7..70eef14edb99 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -1654,15 +1654,24 @@ err_prog_put: return err; } +static void sk_reuseport_prog_free_rcu(struct rcu_head *rcu) +{ + struct bpf_prog_aux *aux = container_of(rcu, struct bpf_prog_aux, rcu); + struct bpf_prog *prog = aux->prog; + + bpf_release_orig_filter(prog); + bpf_prog_free(prog); +} + void sk_reuseport_prog_free(struct bpf_prog *prog) { if (!prog) return; - if (prog->type == BPF_PROG_TYPE_SK_REUSEPORT) - bpf_prog_put(prog); + if (bpf_prog_was_classic(prog)) + call_rcu(&prog->aux->rcu, sk_reuseport_prog_free_rcu); else - bpf_prog_destroy(prog); + bpf_prog_put(prog); } static inline int __bpf_try_make_writable(struct sk_buff *skb, -- cgit v1.2.3 From 97c8a3c1f73d828de43a5a88e8a9a143efb2b661 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Wed, 6 May 2026 03:59:18 +0000 Subject: tcp: Fix potential UAF in reqsk_timer_handler(). When TCP socket migration fails at inet_ehash_insert() in reqsk_timer_handler(), we jump to the no_ownership: label and free the new reqsk immediately with __reqsk_free(). Thus, we must stop the new reqsk's timer before jumping to the label, but the timer might be missed since the cited commit, resulting in UAF. As we are in the original reqsk's timer context, we can safely call timer_delete_sync() for the new reqsk. Let's pass false to __inet_csk_reqsk_queue_drop() to stop the new reqsk's timer. Fixes: 83fccfc3940c ("inet: fix potential deadlock in reqsk_queue_unlink()") Reported-by: Damiano Melotti Signed-off-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20260506035954.1563147-2-kuniyu@google.com Signed-off-by: Jakub Kicinski --- net/ipv4/inet_connection_sock.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index 928654c34156..971f9db2c586 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -1108,7 +1108,7 @@ static void reqsk_timer_handler(struct timer_list *t) if (!inet_ehash_insert(req_to_sk(nreq), req_to_sk(oreq), NULL)) { /* delete timer */ - __inet_csk_reqsk_queue_drop(sk_listener, nreq, true); + __inet_csk_reqsk_queue_drop(sk_listener, nreq, false); goto no_ownership; } -- cgit v1.2.3 From 7eca3292cac7c26dad4c236f51ba225c39a0523f Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Wed, 6 May 2026 03:59:19 +0000 Subject: tcp: Fix imbalanced icsk_accept_queue count. When TCP socket migration happens in reqsk_timer_handler(), @sk_listener will be updated with the new listener. When we call __inet_csk_reqsk_queue_drop(), the listener must be the one stored in req->rsk_listener. The cited commit accidentally replaced oreq->rsk_listener with sk_listener, leading to imbalanced icsk_accept_queue count. Let's pass the correct listener to __inet_csk_reqsk_queue_drop(). Fixes: e8c526f2bdf1 ("tcp/dccp: Don't use timer_pending() in reqsk_queue_unlink().") Reported-by: Damiano Melotti Signed-off-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20260506035954.1563147-3-kuniyu@google.com Signed-off-by: Jakub Kicinski --- net/ipv4/inet_connection_sock.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index 971f9db2c586..dbcd37dfdc15 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -1134,7 +1134,7 @@ no_ownership: } drop: - __inet_csk_reqsk_queue_drop(sk_listener, oreq, true); + __inet_csk_reqsk_queue_drop(oreq->rsk_listener, oreq, true); reqsk_put(oreq); } -- cgit v1.2.3 From 7ce5556f255a680d80daa31b1cedecf7f89e2c22 Mon Sep 17 00:00:00 2001 From: Maoyi Xie Date: Wed, 6 May 2026 16:24:15 +0800 Subject: ipv6: flowlabel: take ip6_fl_lock across mem_check and fl_intern mem_check() in net/ipv6/ip6_flowlabel.c reads fl_size without holding ip6_fl_lock. fl_intern() takes the lock immediately afterwards. The two checks therefore race against concurrent fl_intern, ip6_fl_gc and ip6_fl_purge writers, which makes the mem_check budget check approximate. Move spin_lock_bh(&ip6_fl_lock) and the matching unlock from fl_intern() into its only caller ipv6_flowlabel_get(). The mem_check() call now runs under the same critical section as the fl_intern() insert, so the budget check is exact. With all writers and the read of fl_size under ip6_fl_lock, convert fl_size from atomic_t to plain int. The four sites that update or read fl_size are fl_intern (insert path), ip6_fl_gc (garbage collector, the !sched check and the per-entry decrement), ip6_fl_purge (per-netns purge), and mem_check (budget check), and all four now run under ip6_fl_lock. This is a prerequisite for adding a per-netns budget alongside fl_size. The follow-up patch adds netns_ipv6::flowlabel_count and folds it into mem_check(). Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Suggested-by: Willem de Bruijn Reviewed-by: Willem de Bruijn Signed-off-by: Maoyi Xie Link: https://patch.msgid.link/20260506082416.2259567-2-maoyixie.tju@gmail.com Signed-off-by: Jakub Kicinski --- net/ipv6/ip6_flowlabel.c | 34 +++++++++++++++++++++------------- 1 file changed, 21 insertions(+), 13 deletions(-) (limited to 'net') diff --git a/net/ipv6/ip6_flowlabel.c b/net/ipv6/ip6_flowlabel.c index c92f98c6f6ec..a8974643195a 100644 --- a/net/ipv6/ip6_flowlabel.c +++ b/net/ipv6/ip6_flowlabel.c @@ -40,7 +40,7 @@ #define FL_HASH_MASK 255 #define FL_HASH(l) (ntohl(l)&FL_HASH_MASK) -static atomic_t fl_size = ATOMIC_INIT(0); +static int fl_size; static struct ip6_flowlabel __rcu *fl_ht[FL_HASH_MASK+1]; static void ip6_fl_gc(struct timer_list *unused); @@ -163,7 +163,7 @@ static void ip6_fl_gc(struct timer_list *unused) if (time_after_eq(now, ttd)) { *flp = fl->next; fl_free(fl); - atomic_dec(&fl_size); + fl_size--; continue; } if (!sched || time_before(ttd, sched)) @@ -172,7 +172,7 @@ static void ip6_fl_gc(struct timer_list *unused) flp = &fl->next; } } - if (!sched && atomic_read(&fl_size)) + if (!sched && fl_size) sched = now + FL_MAX_LINGER; if (sched) { mod_timer(&ip6_fl_gc_timer, sched); @@ -196,7 +196,7 @@ static void __net_exit ip6_fl_purge(struct net *net) atomic_read(&fl->users) == 0) { *flp = fl->next; fl_free(fl); - atomic_dec(&fl_size); + fl_size--; continue; } flp = &fl->next; @@ -210,10 +210,10 @@ static struct ip6_flowlabel *fl_intern(struct net *net, { struct ip6_flowlabel *lfl; + lockdep_assert_held(&ip6_fl_lock); + fl->label = label & IPV6_FLOWLABEL_MASK; - rcu_read_lock(); - spin_lock_bh(&ip6_fl_lock); if (label == 0) { for (;;) { fl->label = htonl(get_random_u32())&IPV6_FLOWLABEL_MASK; @@ -235,8 +235,6 @@ static struct ip6_flowlabel *fl_intern(struct net *net, lfl = __fl_lookup(net, fl->label); if (lfl) { atomic_inc(&lfl->users); - spin_unlock_bh(&ip6_fl_lock); - rcu_read_unlock(); return lfl; } } @@ -244,9 +242,7 @@ static struct ip6_flowlabel *fl_intern(struct net *net, fl->lastuse = jiffies; fl->next = fl_ht[FL_HASH(fl->label)]; rcu_assign_pointer(fl_ht[FL_HASH(fl->label)], fl); - atomic_inc(&fl_size); - spin_unlock_bh(&ip6_fl_lock); - rcu_read_unlock(); + fl_size++; return NULL; } @@ -464,10 +460,14 @@ done: static int mem_check(struct sock *sk) { - int room = FL_MAX_SIZE - atomic_read(&fl_size); + int room; struct ipv6_fl_socklist *sfl; int count = 0; + lockdep_assert_held(&ip6_fl_lock); + + room = FL_MAX_SIZE - fl_size; + if (room > FL_MAX_SIZE - FL_MAX_PER_SOCK) return 0; @@ -692,11 +692,19 @@ release: if (!sfl1) goto done; + rcu_read_lock(); + spin_lock_bh(&ip6_fl_lock); err = mem_check(sk); + if (err == 0) + fl1 = fl_intern(net, fl, freq->flr_label); + else + fl1 = NULL; + spin_unlock_bh(&ip6_fl_lock); + rcu_read_unlock(); + if (err != 0) goto done; - fl1 = fl_intern(net, fl, freq->flr_label); if (fl1) goto recheck; -- cgit v1.2.3 From e68eadffb724b36ffd3d5619e0efcaf29ec2a175 Mon Sep 17 00:00:00 2001 From: Maoyi Xie Date: Wed, 6 May 2026 16:24:16 +0800 Subject: ipv6: flowlabel: enforce per-netns limit for unprivileged callers fl_size, fl_ht and ip6_fl_lock in net/ipv6/ip6_flowlabel.c are file scope and shared across netns. mem_check() reads fl_size to decide whether to deny non-CAP_NET_ADMIN callers. capable() runs against init_user_ns, so an unprivileged user in any non-init userns can push fl_size past FL_MAX_SIZE - FL_MAX_SIZE / 4 and starve every other unprivileged userns on the host. Add struct netns_ipv6::flowlabel_count, bumped and decremented next to fl_size in fl_intern, ip6_fl_gc and ip6_fl_purge. The new field fills the existing 4-byte hole after ipmr_seq, so struct netns_ipv6 stays the same size on 64-bit builds. Bump FL_MAX_SIZE from 4096 to 8192. It has been 4096 since the file was added. Machines and connection counts have grown. mem_check() folds an extra per-netns ceiling into the existing non-CAP_NET_ADMIN conditional. The ceiling is half of the total budget that unprivileged callers have ever been able to use, i.e. (FL_MAX_SIZE - FL_MAX_SIZE / 4) / 2 = 3072 entries. With FL_MAX_SIZE doubled, this preserves the original per-user reach of 3K (what an unprivileged caller could already obtain before this change), while forcing an attacker to spread allocations across at least two netns to exhaust the global non-CAP_NET_ADMIN budget. CAP_NET_ADMIN against init_user_ns still bypasses both caps. The previous patch took ip6_fl_lock across mem_check and fl_intern, so the new flowlabel_count read in mem_check and the new flowlabel_count++ in fl_intern run under the same critical section. flowlabel_count is therefore plain int, like fl_size. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Suggested-by: Willem de Bruijn Reviewed-by: Willem de Bruijn Cc: stable@vger.kernel.org # v5.15+ Signed-off-by: Maoyi Xie Link: https://patch.msgid.link/20260506082416.2259567-3-maoyixie.tju@gmail.com Signed-off-by: Jakub Kicinski --- net/ipv6/ip6_flowlabel.c | 14 +++++++++++--- 1 file changed, 11 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/ipv6/ip6_flowlabel.c b/net/ipv6/ip6_flowlabel.c index a8974643195a..b1ccdf0dc646 100644 --- a/net/ipv6/ip6_flowlabel.c +++ b/net/ipv6/ip6_flowlabel.c @@ -36,7 +36,7 @@ /* FL hash table */ #define FL_MAX_PER_SOCK 32 -#define FL_MAX_SIZE 4096 +#define FL_MAX_SIZE 8192 #define FL_HASH_MASK 255 #define FL_HASH(l) (ntohl(l)&FL_HASH_MASK) @@ -162,8 +162,9 @@ static void ip6_fl_gc(struct timer_list *unused) ttd = fl->expires; if (time_after_eq(now, ttd)) { *flp = fl->next; - fl_free(fl); fl_size--; + fl->fl_net->ipv6.flowlabel_count--; + fl_free(fl); continue; } if (!sched || time_before(ttd, sched)) @@ -197,6 +198,7 @@ static void __net_exit ip6_fl_purge(struct net *net) *flp = fl->next; fl_free(fl); fl_size--; + net->ipv6.flowlabel_count--; continue; } flp = &fl->next; @@ -243,6 +245,7 @@ static struct ip6_flowlabel *fl_intern(struct net *net, fl->next = fl_ht[FL_HASH(fl->label)]; rcu_assign_pointer(fl_ht[FL_HASH(fl->label)], fl); fl_size++; + net->ipv6.flowlabel_count++; return NULL; } @@ -460,6 +463,9 @@ done: static int mem_check(struct sock *sk) { + const int unpriv_total_limit = FL_MAX_SIZE - (FL_MAX_SIZE / 4); + const int unpriv_user_limit = unpriv_total_limit / 2; + struct net *net = sock_net(sk); int room; struct ipv6_fl_socklist *sfl; int count = 0; @@ -478,7 +484,9 @@ static int mem_check(struct sock *sk) if (room <= 0 || ((count >= FL_MAX_PER_SOCK || - (count > 0 && room < FL_MAX_SIZE/2) || room < FL_MAX_SIZE/4) && + (count > 0 && room < FL_MAX_SIZE / 2) || + room < FL_MAX_SIZE / 4 || + net->ipv6.flowlabel_count >= unpriv_user_limit) && !capable(CAP_NET_ADMIN))) return -ENOBUFS; -- cgit v1.2.3 From 58e2330bd45572a6e3d46ea94cf7a9641f43591a Mon Sep 17 00:00:00 2001 From: Dragos Tatulea Date: Wed, 6 May 2026 09:08:08 +0000 Subject: net: napi: Avoid gro timer misfiring at end of busypoll When in irq deferral mode (defer-hard-irqs > 0), a short enough gro-flush timeout can trigger before NAPI_STATE_SCHED is cleared if the last poll in busy_poll_stop() takes too long. This can have the effect of leaving the queue stuck with interrupts disabled and no timer armed which results in a tx timeout if there is no subsequent busypoll cycle. To prevent this, defer the gro-flush timer arm after the last poll. Fixes: 7fd3253a7de6 ("net: Introduce preferred busy-polling") Co-developed-by: Martin Karsten Signed-off-by: Martin Karsten Signed-off-by: Dragos Tatulea Reviewed-by: Tariq Toukan Reviewed-by: Cosmin Ratiu Reviewed-by: Joe Damato Link: https://patch.msgid.link/20260506090808.820559-2-dtatulea@nvidia.com Signed-off-by: Jakub Kicinski --- net/core/dev.c | 21 ++++++++++++--------- 1 file changed, 12 insertions(+), 9 deletions(-) (limited to 'net') diff --git a/net/core/dev.c b/net/core/dev.c index 8bfa8313ef62..0c6c270d9f7d 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -6862,9 +6862,9 @@ static void skb_defer_free_flush(void) #if defined(CONFIG_NET_RX_BUSY_POLL) -static void __busy_poll_stop(struct napi_struct *napi, bool skip_schedule) +static void __busy_poll_stop(struct napi_struct *napi, unsigned long timeout) { - if (!skip_schedule) { + if (!timeout) { gro_normal_list(&napi->gro); __napi_schedule(napi); return; @@ -6874,6 +6874,8 @@ static void __busy_poll_stop(struct napi_struct *napi, bool skip_schedule) gro_flush_normal(&napi->gro, HZ >= 1000); clear_bit(NAPI_STATE_SCHED, &napi->state); + hrtimer_start(&napi->timer, ns_to_ktime(timeout), + HRTIMER_MODE_REL_PINNED); } enum { @@ -6885,8 +6887,7 @@ static void busy_poll_stop(struct napi_struct *napi, void *have_poll_lock, unsigned flags, u16 budget) { struct bpf_net_context __bpf_net_ctx, *bpf_net_ctx; - bool skip_schedule = false; - unsigned long timeout; + unsigned long timeout = 0; int rc; /* Busy polling means there is a high chance device driver hard irq @@ -6906,10 +6907,12 @@ static void busy_poll_stop(struct napi_struct *napi, void *have_poll_lock, if (flags & NAPI_F_PREFER_BUSY_POLL) { napi->defer_hard_irqs_count = napi_get_defer_hard_irqs(napi); - timeout = napi_get_gro_flush_timeout(napi); - if (napi->defer_hard_irqs_count && timeout) { - hrtimer_start(&napi->timer, ns_to_ktime(timeout), HRTIMER_MODE_REL_PINNED); - skip_schedule = true; + if (napi->defer_hard_irqs_count) { + /* A short enough gro flush timeout and long enough + * poll can result in timer firing too early. + * Timer will be armed later if necessary. + */ + timeout = napi_get_gro_flush_timeout(napi); } } @@ -6924,7 +6927,7 @@ static void busy_poll_stop(struct napi_struct *napi, void *have_poll_lock, trace_napi_poll(napi, rc, budget); netpoll_poll_unlock(have_poll_lock); if (rc == budget) - __busy_poll_stop(napi, skip_schedule); + __busy_poll_stop(napi, timeout); bpf_net_ctx_clear(bpf_net_ctx); local_bh_enable(); } -- cgit v1.2.3 From 4908f1395fb1b832ceec11584af649874a2732ea Mon Sep 17 00:00:00 2001 From: Quan Sun <2022090917019@std.uestc.edu.cn> Date: Thu, 7 May 2026 21:17:38 +0800 Subject: net: ethtool: fix NULL pointer dereference in phy_reply_size In phy_prepare_data(), several strings such as 'name', 'drvname', 'upstream_sfp_name', and 'downstream_sfp_name' are allocated using kstrdup(). However, these allocations were not checked for failure. If kstrdup() fails for 'name', it returns NULL while the function continues. This leads to a kernel NULL pointer dereference and panic later in phy_reply_size() when it unconditionally calls strlen() on the NULL pointer. While other strings like 'upstream_sfp_name' might be checked before access in certain code paths, failing to handle these allocations consistently can lead to incomplete data reporting or hidden bugs. Fix this by adding proper NULL checks for all kstrdup() calls in phy_prepare_data() and implement a centralized error handling path using goto labels to ensure all previously allocated resources are freed on failure. Fixes: 9dd2ad5e92b9 ("net: ethtool: phy: Convert the PHY_GET command to generic phy dump") Signed-off-by: Quan Sun <2022090917019@std.uestc.edu.cn> Reviewed-by: Maxime Chevallier Link: https://patch.msgid.link/20260507131738.1173835-1-2022090917019@std.uestc.edu.cn Signed-off-by: Jakub Kicinski --- net/ethtool/phy.c | 32 ++++++++++++++++++++++++++++++-- 1 file changed, 30 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/ethtool/phy.c b/net/ethtool/phy.c index d4e6887055ab..f76d94d848d6 100644 --- a/net/ethtool/phy.c +++ b/net/ethtool/phy.c @@ -76,6 +76,7 @@ static int phy_prepare_data(const struct ethnl_req_info *req_info, struct nlattr **tb = info->attrs; struct phy_device_node *pdn; struct phy_device *phydev; + int ret; /* RTNL is held by the caller */ phydev = ethnl_req_get_phydev(req_info, tb, ETHTOOL_A_PHY_HEADER, @@ -88,8 +89,17 @@ static int phy_prepare_data(const struct ethnl_req_info *req_info, return -EOPNOTSUPP; rep_data->phyindex = phydev->phyindex; + rep_data->name = kstrdup(dev_name(&phydev->mdio.dev), GFP_KERNEL); + if (!rep_data->name) + return -ENOMEM; + rep_data->drvname = kstrdup(phydev->drv->name, GFP_KERNEL); + if (!rep_data->drvname) { + ret = -ENOMEM; + goto err_free_name; + } + rep_data->upstream_type = pdn->upstream_type; if (pdn->upstream_type == PHY_UPSTREAM_PHY) { @@ -97,15 +107,33 @@ static int phy_prepare_data(const struct ethnl_req_info *req_info, rep_data->upstream_index = upstream->phyindex; } - if (pdn->parent_sfp_bus) + if (pdn->parent_sfp_bus) { rep_data->upstream_sfp_name = kstrdup(sfp_get_name(pdn->parent_sfp_bus), GFP_KERNEL); + if (!rep_data->upstream_sfp_name) { + ret = -ENOMEM; + goto err_free_drvname; + } + } - if (phydev->sfp_bus) + if (phydev->sfp_bus) { rep_data->downstream_sfp_name = kstrdup(sfp_get_name(phydev->sfp_bus), GFP_KERNEL); + if (!rep_data->downstream_sfp_name) { + ret = -ENOMEM; + goto err_free_upstream_sfp; + } + } return 0; + +err_free_upstream_sfp: + kfree(rep_data->upstream_sfp_name); +err_free_drvname: + kfree(rep_data->drvname); +err_free_name: + kfree(rep_data->name); + return ret; } static int phy_fill_reply(struct sk_buff *skb, -- cgit v1.2.3 From efda25ee84325385f859d10872590e90ce837243 Mon Sep 17 00:00:00 2001 From: Alice Ryhl Date: Wed, 6 May 2026 20:07:13 +0000 Subject: genetlink: free the skb on 'group >= family->n_mcgrps' These methods generally consume ownership of the provided skb, so even if an error path is encountered, the skb is freed. This is because the very first thing they do after some initial setup is to unconditionally consume the skb via consume_skb(skb). Any subsequent errors lead to the core netlink layer freeing the skb. However, there is one check that occurs before ownership is passed, which is the check for the group index. So if this error condition is encountered, then the skb is leaked. This error condition is generally considered a violation of the netlink API, so it's not expected to occur under normal circumstances. For the same reason, no callers check for this error condition, and no callers need to be adjusted. However, we should still follow the same ownership semantics of the rest of the function. Thus, free the skb in this codepath. Suggested-by: Andrew Lunn Suggested-by: Matthew Maurer Fixes: 2a94fe48f32c ("genetlink: make multicast groups const, prevent abuse") Link: https://lore.kernel.org/r/845b36ba-7b3a-41f2-acb2-b284f253e2ca@lunn.ch Signed-off-by: Alice Ryhl Link: https://patch.msgid.link/20260506-genlmsg-return-v2-1-a63ee2a055d6@google.com Signed-off-by: Jakub Kicinski --- net/netlink/genetlink.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/netlink/genetlink.c b/net/netlink/genetlink.c index d251d894afd4..0da39eaed255 100644 --- a/net/netlink/genetlink.c +++ b/net/netlink/genetlink.c @@ -1972,8 +1972,10 @@ int genlmsg_multicast_allns(const struct genl_family *family, struct sk_buff *skb, u32 portid, unsigned int group) { - if (WARN_ON_ONCE(group >= family->n_mcgrps)) + if (WARN_ON_ONCE(group >= family->n_mcgrps)) { + kfree_skb(skb); return -EINVAL; + } group = family->mcgrp_offset + group; return genlmsg_mcast(skb, portid, group); @@ -1986,8 +1988,10 @@ void genl_notify(const struct genl_family *family, struct sk_buff *skb, struct net *net = genl_info_net(info); struct sock *sk = net->genl_sock; - if (WARN_ON_ONCE(group >= family->n_mcgrps)) + if (WARN_ON_ONCE(group >= family->n_mcgrps)) { + kfree_skb(skb); return; + } group = family->mcgrp_offset + group; nlmsg_notify(sk, skb, info->snd_portid, group, -- cgit v1.2.3 From a77d5a069d959dc45f5f472d48cba37d8cba0f1c Mon Sep 17 00:00:00 2001 From: Mohsin Bashir Date: Wed, 6 May 2026 16:37:45 -0700 Subject: net: shaper: Reject reparenting of existing nodes When an existing node-scope shaper is moved to a different parent via the group operation, the framework fails to update the leaves count on both the old and new parent shapers. Only newly created nodes (handle.id == NET_SHAPER_ID_UNSPEC) trigger the parent leaves increment at line 1039. This causes the parent's leaves counter to diverge from the actual number of children in the xarray. When the node is later deleted, pre_del_node() allocates an array sized by the stale leaves count, but the xarray iteration finds more children than expected, hitting the WARN_ON_ONCE guard and returning -EINVAL. Rather than adding reparenting support with complex leaves count bookkeeping, reject group calls that attempt to change an existing node's parent. Updates to an existing node's rate or leaves under the same parent remain permitted. We expect that for any modification of the topology user should always create new groups and let the kernel garbage collect the leaf-less nodes. Fixes: 5d5d4700e75d ("net-shapers: implement NL group operation") Signed-off-by: Mohsin Bashir Link: https://patch.msgid.link/20260506233745.111895-1-mohsin.bashr@gmail.com Signed-off-by: Jakub Kicinski --- net/shaper/shaper.c | 30 +++++++++++++++++++++++------- 1 file changed, 23 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/net/shaper/shaper.c b/net/shaper/shaper.c index 94bc9c7382ea..1069fa4eb9f6 100644 --- a/net/shaper/shaper.c +++ b/net/shaper/shaper.c @@ -964,15 +964,22 @@ static int __net_shaper_group(struct net_shaper_binding *binding, int i, ret; if (node->handle.scope == NET_SHAPER_SCOPE_NODE) { + struct net_shaper *cur = NULL; + new_node = node->handle.id == NET_SHAPER_ID_UNSPEC; - if (!new_node && !net_shaper_lookup(binding, &node->handle)) { - /* The related attribute is not available when - * reaching here from the delete() op. - */ - NL_SET_ERR_MSG_FMT(extack, "Node shaper %d:%d does not exists", - node->handle.scope, node->handle.id); - return -ENOENT; + if (!new_node) { + cur = net_shaper_lookup(binding, &node->handle); + if (!cur) { + /* The related attribute is not available + * when reaching here from the delete() op. + */ + NL_SET_ERR_MSG_FMT(extack, + "Node shaper %d:%d does not exist", + node->handle.scope, + node->handle.id); + return -ENOENT; + } } /* When unspecified, the node parent scope is inherited from @@ -986,6 +993,15 @@ static int __net_shaper_group(struct net_shaper_binding *binding, return ret; } + if (cur && net_shaper_handle_cmp(&cur->parent, + &node->parent)) { + NL_SET_ERR_MSG_FMT(extack, + "Cannot reparent node shaper %d:%d", + node->handle.scope, + node->handle.id); + return -EOPNOTSUPP; + } + } else { net_shaper_default_parent(&node->handle, &node->parent); } -- cgit v1.2.3 From abb5f36771cc4c05899b34000829a787572a8817 Mon Sep 17 00:00:00 2001 From: Ben Morris Date: Thu, 7 May 2026 17:14:55 -0700 Subject: sctp: revalidate list cursor after sctp_sendmsg_to_asoc() in SCTP_SENDALL The SCTP_SENDALL path in sctp_sendmsg() iterates ep->asocs with list_for_each_entry_safe(), which caches the next entry in @tmp before the loop body runs. The body calls sctp_sendmsg_to_asoc(), which may drop the socket lock inside sctp_wait_for_sndbuf(). While the lock is dropped, another thread can SCTP_SOCKOPT_PEELOFF the association cached in @tmp, migrating it to a new endpoint via sctp_sock_migrate() (list_del_init() + list_add_tail() to newep->asocs), and optionally close the new socket which frees the association via kfree_rcu(). The cached @tmp can also be freed by a network ABORT for that association, processed in softirq while the lock is dropped. sctp_wait_for_sndbuf() revalidates @asoc (the current entry) on re-lock via the "sk != asoc->base.sk" and "asoc->base.dead" checks, but nothing revalidates @tmp. After a successful return, the iterator advances to the stale @tmp, yielding either a use-after-free (if the peeled socket was closed) or a list-walk onto the new endpoint's list head (type confusion of &newep->asocs as a struct sctp_association *). Both are reachable from CapEff=0; the type-confusion path gives controlled indirect call via the outqueue.sched->init_sid pointer. Fix by re-deriving @tmp from @asoc after sctp_sendmsg_to_asoc() returns. @asoc is known to still be on ep->asocs at that point: the only callers that list_del an association from ep->asocs are sctp_association_free() (which sets asoc->base.dead) and sctp_assoc_migrate() (which changes asoc->base.sk), and sctp_wait_for_sndbuf() checks both under the lock before any successful return; a tripped check propagates as err < 0 and the loop bails before the re-derive. The SCTP_ABORT path in sctp_sendmsg_check_sflags() returns 0 and the loop hits 'continue' before sctp_sendmsg_to_asoc() is ever called, so the @tmp cached by list_for_each_entry_safe() still covers the lock-held free that ba59fb027307 ("sctp: walk the list of asoc safely") was added for. Fixes: 4910280503f3 ("sctp: add support for snd flag SCTP_SENDALL process in sendmsg") Cc: stable@vger.kernel.org Signed-off-by: Ben Morris Acked-by: Xin Long Link: https://patch.msgid.link/20260508001455.3137-1-joycathacker@gmail.com Signed-off-by: Jakub Kicinski --- net/sctp/socket.c | 9 +++++++++ 1 file changed, 9 insertions(+) (limited to 'net') diff --git a/net/sctp/socket.c b/net/sctp/socket.c index 58d0d9747f0b..1d2568bb6bc2 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -1986,6 +1986,15 @@ static int sctp_sendmsg(struct sock *sk, struct msghdr *msg, size_t msg_len) goto out_unlock; iov_iter_revert(&msg->msg_iter, err); + + /* sctp_sendmsg_to_asoc() may have released the socket + * lock (sctp_wait_for_sndbuf), during which other + * associations on ep->asocs could have been peeled + * off or freed. @asoc itself is revalidated by the + * base.dead and base.sk checks in sctp_wait_for_sndbuf, + * so re-derive the cached cursor from it. + */ + tmp = list_next_entry(asoc, asocs); } goto out_unlock; -- cgit v1.2.3 From bf6d507f7e3c65751d52fd8caf1ea4e003922624 Mon Sep 17 00:00:00 2001 From: Linpu Yu Date: Fri, 8 May 2026 22:43:43 +0800 Subject: xskmap: reject TX-only AF_XDP sockets XSKMAP entries are used as redirect targets for incoming XDP frames. A TX-only AF_XDP socket lacks an Rx ring and cannot handle redirected traffic, but xsk_map_update_elem() currently allows such sockets to be inserted into the map. Redirecting packets to such a socket on the veth generic-XDP path causes a kernel crash in xsk_generic_rcv(). This became possible after xsk_is_setup_for_bpf_map() was removed from the XSKMAP update path, which allowed bound TX-only sockets to be inserted into the map. Reject TX-only sockets during XSKMAP updates to avoid the crash. They remain fully operational for pure Tx purposes outside XSKMAP. Fixes: 968be23ceaca ("xsk: Fix possible segfault at xskmap entry insertion") Reported-by: Juefei Pu Reported-by: Yuan Tan Reported-by: Xin Liu Signed-off-by: Yifan Wu Signed-off-by: Linpu Yu Reviewed-by: Jason Xing Link: https://lore.kernel.org/r/20260508144344.694-1-linpu5433@gmail.com Signed-off-by: Alexei Starovoitov --- net/xdp/xskmap.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'net') diff --git a/net/xdp/xskmap.c b/net/xdp/xskmap.c index afa457506274..3bff346308d0 100644 --- a/net/xdp/xskmap.c +++ b/net/xdp/xskmap.c @@ -184,6 +184,10 @@ static long xsk_map_update_elem(struct bpf_map *map, void *key, void *value, } xs = (struct xdp_sock *)sock->sk; + if (!READ_ONCE(xs->rx)) { + sockfd_put(sock); + return -ENOBUFS; + } map_entry = &m->xsk_map[i]; node = xsk_map_node_alloc(m, map_entry); -- cgit v1.2.3 From aa54b1d27fe0c2b78e664a34fd0fdf7cd1960d71 Mon Sep 17 00:00:00 2001 From: Hyunwoo Kim Date: Fri, 8 May 2026 17:53:09 +0900 Subject: rxrpc: Also unshare DATA/RESPONSE packets when paged frags are present The DATA-packet handler in rxrpc_input_call_event() and the RESPONSE handler in rxrpc_verify_response() copy the skb to a linear one before calling into the security ops only when skb_cloned() is true. An skb that is not cloned but still carries externally-owned paged fragments (e.g. SKBFL_SHARED_FRAG set by splice() into a UDP socket via __ip_append_data, or a chained skb_has_frag_list()) falls through to the in-place decryption path, which binds the frag pages directly into the AEAD/skcipher SGL via skb_to_sgvec(). Extend the gate to also unshare when skb_has_frag_list() or skb_has_shared_frag() is true. This catches the splice-loopback vector and other externally-shared frag sources while preserving the zero-copy fast path for skbs whose frags are kernel-private (e.g. NIC page_pool RX, GRO). The OOM/trace handling already in place is reused. Fixes: d0d5c0cd1e71 ("rxrpc: Use skb_unshare() rather than skb_cow_data()") Cc: stable@vger.kernel.org Signed-off-by: Hyunwoo Kim Reviewed-by: Jiayuan Chen Acked-by: David Howells Signed-off-by: Linus Torvalds --- net/rxrpc/call_event.c | 4 +++- net/rxrpc/conn_event.c | 3 ++- 2 files changed, 5 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/rxrpc/call_event.c b/net/rxrpc/call_event.c index fdd683261226..2b19b252225e 100644 --- a/net/rxrpc/call_event.c +++ b/net/rxrpc/call_event.c @@ -334,7 +334,9 @@ bool rxrpc_input_call_event(struct rxrpc_call *call) if (sp->hdr.type == RXRPC_PACKET_TYPE_DATA && sp->hdr.securityIndex != 0 && - skb_cloned(skb)) { + (skb_cloned(skb) || + skb_has_frag_list(skb) || + skb_has_shared_frag(skb))) { /* Unshare the packet so that it can be * modified by in-place decryption. */ diff --git a/net/rxrpc/conn_event.c b/net/rxrpc/conn_event.c index a2130d25aaa9..442414d90ba1 100644 --- a/net/rxrpc/conn_event.c +++ b/net/rxrpc/conn_event.c @@ -245,7 +245,8 @@ static int rxrpc_verify_response(struct rxrpc_connection *conn, { int ret; - if (skb_cloned(skb)) { + if (skb_cloned(skb) || skb_has_frag_list(skb) || + skb_has_shared_frag(skb)) { /* Copy the packet if shared so that we can do in-place * decryption. */ -- cgit v1.2.3 From c00b472a1322d4f5424cd7b6c7d00270eae673bd Mon Sep 17 00:00:00 2001 From: Jeff Layton Date: Sat, 11 Apr 2026 17:12:16 -0400 Subject: sunrpc: start cache request seqno at 1 to fix netlink GET_REQS sunrpc_cache_requests_snapshot() filters requests with crq->seqno <= min_seqno. The min_seqno for the first netlink dump call is cb->args[0] which is 0. Since next_seqno was initialized to 0, the very first cache request got seqno=0 and was silently skipped by the snapshot (0 <= 0 is true). This caused netlink-based GET_REQS to return 0 pending requests even when a request was queued, preventing mountd from resolving cache entries (particularly expkey/nfsd.fh). The unresolved CACHE_PENDING state blocked all further notifications for the entry, leading to permanent NFS4ERR_DELAY hangs. Start next_seqno at 1 so all requests have seqno >= 1 and pass the snapshot filter when min_seqno is 0. Fixes: facc4e3c8042 ("sunrpc: split cache_detail queue into request and reader lists") Signed-off-by: Jeff Layton Signed-off-by: Chuck Lever --- net/sunrpc/cache.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/sunrpc/cache.c b/net/sunrpc/cache.c index 7081c1214e6c..b5474ce534fb 100644 --- a/net/sunrpc/cache.c +++ b/net/sunrpc/cache.c @@ -403,7 +403,7 @@ void sunrpc_init_cache_detail(struct cache_detail *cd) INIT_LIST_HEAD(&cd->readers); spin_lock_init(&cd->queue_lock); init_waitqueue_head(&cd->queue_wait); - cd->next_seqno = 0; + cd->next_seqno = 1; spin_lock(&cache_list_lock); cd->nextcheck = 0; cd->entries = 0; -- cgit v1.2.3 From 7dbac7680eb629b3b4dc7e98c34f943b8814c0c8 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Wed, 6 May 2026 21:23:28 +0800 Subject: xfrm: ipcomp: Free destination pages on acomp errors Move the out_free_req label up by a couple of lines so that the allocated dst SG list gets freed on error as well as success. Fixes: eb2953d26971 ("xfrm: ipcomp: Use crypto_acomp interface") Cc: stable@kernel.org Reported-by: Yuan Tan Reported-by: Yifan Wu Reported-by: Juefei Pu Reported-by: Xin Liu Reported-by: Yilin Zhu Signed-off-by: Herbert Xu Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_ipcomp.c | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/xfrm/xfrm_ipcomp.c b/net/xfrm/xfrm_ipcomp.c index 5f38dff16177..671d48f8c937 100644 --- a/net/xfrm/xfrm_ipcomp.c +++ b/net/xfrm/xfrm_ipcomp.c @@ -51,11 +51,15 @@ static int ipcomp_post_acomp(struct sk_buff *skb, int err, int hlen) struct scatterlist *dsg; int len, dlen; - if (unlikely(err)) - goto out_free_req; + if (unlikely(!req)) + return err; extra = acomp_request_extra(req); dsg = extra->sg; + + if (unlikely(err)) + goto out_free_req; + dlen = req->dlen; pskb_trim_unique(skb, 0); @@ -84,10 +88,10 @@ static int ipcomp_post_acomp(struct sk_buff *skb, int err, int hlen) skb_shinfo(skb)->nr_frags++; } while ((dlen -= len)); - for (; dsg; dsg = sg_next(dsg)) +out_free_req: + for (; dsg && sg_page(dsg); dsg = sg_next(dsg)) __free_page(sg_page(dsg)); -out_free_req: acomp_request_free(req); return err; } -- cgit v1.2.3 From 4c79fc2d598694bda845b46229c9d48b65042970 Mon Sep 17 00:00:00 2001 From: Raphael Zimmer Date: Wed, 22 Apr 2026 10:47:13 +0200 Subject: libceph: Fix potential out-of-bounds access in crush_decode() A message of type CEPH_MSG_OSD_MAP containing a crush map with at least one bucket has two fields holding the bucket algorithm. If the values in these two fields differ, an out-of-bounds access can occur. This is the case because the first algorithm field (alg) is used to allocate the correct amount of memory for a bucket of this type, while the second algorithm field inside the bucket (b->alg) is used in the subsequent processing. This patch fixes the issue by adding a check that compares alg and b->alg and aborts the processing in case they differ. Furthermore, b->alg is set to 0 in this case, because the destruction of the crush map also uses this field to determine the bucket type, which can again result in an out-of-bounds access when trying to free the memory pointed to by the fields of the bucket. To correctly free the memory allocated for the bucket in such a case, the corresponding call to kfree is moved from the algorithm-specific crush_destroy_bucket functions to the generic crush_destroy_bucket(). Cc: stable@vger.kernel.org Signed-off-by: Raphael Zimmer Reviewed-by: Ilya Dryomov Signed-off-by: Ilya Dryomov --- net/ceph/crush/crush.c | 6 +----- net/ceph/osdmap.c | 4 ++++ 2 files changed, 5 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/ceph/crush/crush.c b/net/ceph/crush/crush.c index 254ded0b05f6..521aec1d5fc0 100644 --- a/net/ceph/crush/crush.c +++ b/net/ceph/crush/crush.c @@ -47,7 +47,6 @@ int crush_get_bucket_item_weight(const struct crush_bucket *b, int p) void crush_destroy_bucket_uniform(struct crush_bucket_uniform *b) { kfree(b->h.items); - kfree(b); } void crush_destroy_bucket_list(struct crush_bucket_list *b) @@ -55,14 +54,12 @@ void crush_destroy_bucket_list(struct crush_bucket_list *b) kfree(b->item_weights); kfree(b->sum_weights); kfree(b->h.items); - kfree(b); } void crush_destroy_bucket_tree(struct crush_bucket_tree *b) { kfree(b->h.items); kfree(b->node_weights); - kfree(b); } void crush_destroy_bucket_straw(struct crush_bucket_straw *b) @@ -70,14 +67,12 @@ void crush_destroy_bucket_straw(struct crush_bucket_straw *b) kfree(b->straws); kfree(b->item_weights); kfree(b->h.items); - kfree(b); } void crush_destroy_bucket_straw2(struct crush_bucket_straw2 *b) { kfree(b->item_weights); kfree(b->h.items); - kfree(b); } void crush_destroy_bucket(struct crush_bucket *b) @@ -99,6 +94,7 @@ void crush_destroy_bucket(struct crush_bucket *b) crush_destroy_bucket_straw2((struct crush_bucket_straw2 *)b); break; } + kfree(b); } /** diff --git a/net/ceph/osdmap.c b/net/ceph/osdmap.c index c89e66d4fcb7..a87268058e61 100644 --- a/net/ceph/osdmap.c +++ b/net/ceph/osdmap.c @@ -516,6 +516,10 @@ static struct crush_map *crush_decode(void *pbyval, void *end) b->id = ceph_decode_32(p); b->type = ceph_decode_16(p); b->alg = ceph_decode_8(p); + if (b->alg != alg) { + b->alg = 0; + goto bad; + } b->hash = ceph_decode_8(p); b->weight = ceph_decode_32(p); b->size = ceph_decode_32(p); -- cgit v1.2.3 From 596f91294b351866956808b1ecb8dfae15382a6d Mon Sep 17 00:00:00 2001 From: Raphael Zimmer Date: Fri, 24 Apr 2026 15:37:37 +0200 Subject: libceph: Fix unnecessarily high ceph_decode_need() for uniform bucket In crush_decode_uniform_bucket(), the item_weight field of the bucket is set. This is a single field of type u32 since the uniform bucket uses the same weight for all items. The value in ceph_decode_need() is set to (1+b->h.size) * sizeof(u32), which is higher than actually needed. This patch removes the call to ceph_decode_need() with the unnecessarily high value and switches the subsequent operation from ceph_decode_32() to ceph_decode_32_safe(), which already includes the correct bounds check. Signed-off-by: Raphael Zimmer Reviewed-by: Ilya Dryomov Signed-off-by: Ilya Dryomov --- net/ceph/osdmap.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'net') diff --git a/net/ceph/osdmap.c b/net/ceph/osdmap.c index a87268058e61..669348d883f0 100644 --- a/net/ceph/osdmap.c +++ b/net/ceph/osdmap.c @@ -72,8 +72,7 @@ static int crush_decode_uniform_bucket(void **p, void *end, struct crush_bucket_uniform *b) { dout("crush_decode_uniform_bucket %p to %p\n", *p, end); - ceph_decode_need(p, end, (1+b->h.size) * sizeof(u32), bad); - b->item_weight = ceph_decode_32(p); + ceph_decode_32_safe(p, end, b->item_weight, bad); return 0; bad: return -EINVAL; -- cgit v1.2.3 From 821365487aa58d06bda65c676ba215d506ba9768 Mon Sep 17 00:00:00 2001 From: Raphael Zimmer Date: Tue, 28 Apr 2026 14:15:46 +0200 Subject: libceph: Fix potential out-of-bounds access in __ceph_x_decrypt() In __ceph_x_decrypt(), a part of the buffer p is interpreted as a ceph_x_encrypt_header, and the magic field of this struct is accessed. This happens without any guarantee that the buffer is large enough to hold this struct. The function parameter ciphertext_len represents the length of the ciphertext to decrypt and is guaranteed to be at most the remaining size of the allocated buffer p. However, this value is not necessarily greater than sizeof(ceph_x_encrypt_header). E.g., a message frame of type FRAME_TAG_AUTH_REPLY_MORE, that is just as long to hold the ciphertext at its end with a ciphertext_len of 8 or less, can trigger an out-of-bounds memory access when accessing hdr->magic. This patch fixes the issue by adding a check to ensure that the decrypted plaintext in the buffer is large enough to represent at least the ceph_x_encrypt_header. Cc: stable@vger.kernel.org Signed-off-by: Raphael Zimmer Reviewed-by: Ilya Dryomov Signed-off-by: Ilya Dryomov --- net/ceph/auth_x.c | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'net') diff --git a/net/ceph/auth_x.c b/net/ceph/auth_x.c index 692e0b868822..9e64e82d0b63 100644 --- a/net/ceph/auth_x.c +++ b/net/ceph/auth_x.c @@ -115,6 +115,11 @@ static int __ceph_x_decrypt(const struct ceph_crypto_key *key, int usage_slot, if (ret) return ret; + if (plaintext_len < sizeof(*hdr)) { + pr_err("%s plaintext too small %d\n", __func__, plaintext_len); + return -EINVAL; + } + hdr = p + ceph_crypt_data_offset(key); if (le64_to_cpu(hdr->magic) != CEPHX_ENC_MAGIC) { pr_err("%s bad magic\n", __func__); -- cgit v1.2.3 From 94f3b133168d1c49895e7cc6afbcf1cc0b354602 Mon Sep 17 00:00:00 2001 From: Luxiao Xu Date: Mon, 11 May 2026 18:52:09 +0200 Subject: batman-adv: fix tp_meter counter underflow during shutdown batadv_tp_sender_shutdown() unconditionally decrements the "sending" atomic counter. If multiple paths (e.g. timeout, user cancel, and normal finish) call this function, the counter can underflow to -1. Since the sender logic treats any non-zero value as "still sending", a negative value causes the sender kthread to loop indefinitely. This leads to a use-after-free when the interface is removed while the zombie thread is still active. Fix this by using atomic_xchg() to ensure the counter only transitions from 1 to 0 once. Fixes: 33a3bb4a3345 ("batman-adv: throughput meter implementation") Cc: stable@kernel.org Reported-by: Yuan Tan Reported-by: Yifan Wu Reported-by: Juefei Pu Reported-by: Xin Liu Signed-off-by: Luxiao Xu Signed-off-by: Ren Wei [sven: added missing change in batadv_tp_send] Signed-off-by: Sven Eckelmann --- net/batman-adv/tp_meter.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/batman-adv/tp_meter.c b/net/batman-adv/tp_meter.c index 066c76113fc4..a4397aa881dd 100644 --- a/net/batman-adv/tp_meter.c +++ b/net/batman-adv/tp_meter.c @@ -451,7 +451,7 @@ static void batadv_tp_sender_end(struct batadv_priv *bat_priv, static void batadv_tp_sender_shutdown(struct batadv_tp_vars *tp_vars, enum batadv_tp_meter_reason reason) { - if (!atomic_dec_and_test(&tp_vars->sending)) + if (atomic_xchg(&tp_vars->sending, 0) != 1) return; tp_vars->reason = reason; @@ -885,7 +885,7 @@ static int batadv_tp_send(void *arg) "Meter: %s() cannot send packets (%d)\n", __func__, err); /* ensure nobody else tries to stop the thread now */ - if (atomic_dec_and_test(&tp_vars->sending)) + if (atomic_xchg(&tp_vars->sending, 0) == 1) tp_vars->reason = err; break; } -- cgit v1.2.3 From 77098e4bea37af51d3962efa88a5af2ea5e1ac57 Mon Sep 17 00:00:00 2001 From: Sven Eckelmann Date: Sun, 10 May 2026 11:31:03 +0200 Subject: batman-adv: tp_meter: fix tp_vars reference leak in receiver shutdown The receiver shutdown timer handler, batadv_tp_receiver_shutdown(), is responsible for releasing the tp_vars reference it holds. However, the existing logic for coordinating this release with batadv_tp_stop_all() was flawed. timer_shutdown_sync() guarantees the timer will not fire again after it returns, but it returns non-zero only when the timer was pending at the time of the call. If the timer had already expired (and batadv_tp_stop_all() would unsucessfully try to rearm itself), batadv_tp_stop_all() skips its batadv_tp_vars_put(), and batadv_tp_receiver_shutdown() fails to put its own reference as well. Fix this by introducing a new atomic variable receiving that is set to 1 when the receiver is initialized and cleared atomically with atomic_xchg() by whichever side claims it first. Only the side that observes the transition from 1 to 0 is responsible for releasing the tp_vars timer reference, eliminating the uncertainty. Cc: stable@kernel.org Fixes: 3d3cf6a7314a ("batman-adv: stop tp_meter sessions during mesh teardown") Signed-off-by: Sven Eckelmann --- net/batman-adv/tp_meter.c | 13 +++++++++++-- net/batman-adv/types.h | 3 +++ 2 files changed, 14 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/batman-adv/tp_meter.c b/net/batman-adv/tp_meter.c index a4397aa881dd..ca6c3f6374bc 100644 --- a/net/batman-adv/tp_meter.c +++ b/net/batman-adv/tp_meter.c @@ -8,6 +8,7 @@ #include "main.h" #include +#include #include #include #include @@ -1156,6 +1157,9 @@ static void batadv_tp_receiver_shutdown(struct timer_list *t) spin_unlock_bh(&tp_vars->unacked_lock); /* drop reference of timer */ + if (WARN_ON(atomic_xchg(&tp_vars->receiving, 0) != 1)) + return; + batadv_tp_vars_put(tp_vars); } @@ -1374,6 +1378,7 @@ batadv_tp_init_recv(struct batadv_priv *bat_priv, ether_addr_copy(tp_vars->other_end, icmp->orig); tp_vars->role = BATADV_TP_RECEIVER; + atomic_set(&tp_vars->receiving, 1); memcpy(tp_vars->session, icmp->session, sizeof(tp_vars->session)); tp_vars->last_recv = BATADV_TP_FIRST_SEQ; tp_vars->bat_priv = bat_priv; @@ -1546,8 +1551,12 @@ void batadv_tp_stop_all(struct batadv_priv *bat_priv) break; case BATADV_TP_RECEIVER: batadv_tp_list_detach(tp_var); - if (timer_shutdown_sync(&tp_var->timer)) - batadv_tp_vars_put(tp_var); + timer_shutdown_sync(&tp_var->timer); + + if (atomic_xchg(&tp_var->receiving, 0) != 1) + break; + + batadv_tp_vars_put(tp_var); break; } diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h index daa06f421154..b9c0b7779122 100644 --- a/net/batman-adv/types.h +++ b/net/batman-adv/types.h @@ -1323,6 +1323,9 @@ struct batadv_tp_vars { /** @sending: sending binary semaphore: 1 if sending, 0 is not */ atomic_t sending; + /** @receiving: receiving binary semaphore: 1 if receiving, 0 is not */ + atomic_t receiving; + /** @reason: reason for a stopped session */ enum batadv_tp_meter_reason reason; -- cgit v1.2.3 From 35d0ed82d03e5ee77ea4f31f20e29562a7721649 Mon Sep 17 00:00:00 2001 From: Raphael Zimmer Date: Tue, 5 May 2026 11:08:12 +0200 Subject: libceph: Fix potential out-of-bounds access in osdmap_decode() When decoding osd_state and osd_weight from an incoming osdmap in osdmap_decode(), both are decoded for each osd, i.e., map->max_osd times. The ceph_decode_need() check only accounts for sizeof(*map->osd_weight) once. This can potentially result in an out-of-bounds memory access if the incoming message is corrupted such that the max_osd value exceeds the actual content of the osdmap message. This patch fixes the issue by changing the corresponding part in the ceph_decode_need() check to account for map->max_osd*sizeof(*map->osd_weight). Cc: stable@vger.kernel.org Fixes: dcbc919a5dc8 ("libceph: switch osdmap decoding to use ceph_decode_entity_addr") Signed-off-by: Raphael Zimmer Reviewed-by: Ilya Dryomov Signed-off-by: Ilya Dryomov --- net/ceph/osdmap.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/ceph/osdmap.c b/net/ceph/osdmap.c index 669348d883f0..2095e73ccf6c 100644 --- a/net/ceph/osdmap.c +++ b/net/ceph/osdmap.c @@ -1705,7 +1705,7 @@ static int osdmap_decode(void **p, void *end, bool msgr2, ceph_decode_need(p, end, 3*sizeof(u32) + map->max_osd*(struct_v >= 5 ? sizeof(u32) : sizeof(u8)) + - sizeof(*map->osd_weight), e_inval); + map->max_osd*sizeof(*map->osd_weight), e_inval); if (ceph_decode_32(p) != map->max_osd) goto e_inval; -- cgit v1.2.3 From e174929793195e0cd6a4adb0cad731b39f9019b4 Mon Sep 17 00:00:00 2001 From: Allison Henderson Date: Tue, 5 May 2026 16:43:36 -0700 Subject: net/rds: reset op_nents when zerocopy page pin fails When iov_iter_get_pages2() fails in rds_message_zcopy_from_user(), the pinned pages are released with put_page(), and rm->data.op_mmp_znotifier is cleared. But we fail to properly clear rm->data.op_nents. Later when rds_message_purge() is called from rds_sendmsg() the cleanup loop iterates over the incorrectly non zero number of op_nents and frees them again. Fix this by properly resetting op_nents when it should be in rds_message_zcopy_from_user(). Fixes: 0cebaccef3ac ("rds: zerocopy Tx support.") Signed-off-by: Allison Henderson Reviewed-by: Simon Horman Link: https://patch.msgid.link/20260505234336.2132721-1-achender@kernel.org Signed-off-by: Jakub Kicinski --- net/rds/message.c | 1 + 1 file changed, 1 insertion(+) (limited to 'net') diff --git a/net/rds/message.c b/net/rds/message.c index 25fedcb3cd00..7feb0eb6537d 100644 --- a/net/rds/message.c +++ b/net/rds/message.c @@ -448,6 +448,7 @@ static int rds_message_zcopy_from_user(struct rds_message *rm, struct iov_iter * for (i = 0; i < rm->data.op_nents; i++) put_page(sg_page(&rm->data.op_sg[i])); + rm->data.op_nents = 0; mmp = &rm->data.op_mmp_znotifier->z_mmp; mm_unaccount_pinned_pages(mmp); ret = -EFAULT; -- cgit v1.2.3 From 03cb001ef87b3f8d859cf7f96329acf3d6235d29 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Fri, 8 May 2026 12:08:46 +0000 Subject: tcp: Fix out-of-bounds access for twsk in tcp_ao_established_key(). lockdep_sock_is_held() was added in tcp_ao_established_key() by the cited commit. It can be called from tcp_v[46]_timewait_ack() with twsk. Since it does not have sk->sk_lock, the lockdep annotation results in out-of-bound access. $ pahole -C tcp_timewait_sock vmlinux | grep size /* size: 288, cachelines: 5, members: 8 */ $ pahole -C sock vmlinux | grep sk_lock socket_lock_t sk_lock; /* 440 192 */ Let's not use lockdep_sock_is_held() for TCP_TIME_WAIT. Fixes: 6b2d11e2d8fc ("net/tcp: Add missing lockdep annotations for TCP-AO hlist traversals") Reported-by: Damiano Melotti Signed-off-by: Kuniyuki Iwashima Reviewed-by: Eric Dumazet Link: https://patch.msgid.link/20260508120853.4098365-1-kuniyu@google.com Signed-off-by: Jakub Kicinski --- net/ipv4/tcp_ao.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv4/tcp_ao.c b/net/ipv4/tcp_ao.c index a97cdf3e6af4..0a4b38b315fe 100644 --- a/net/ipv4/tcp_ao.c +++ b/net/ipv4/tcp_ao.c @@ -116,7 +116,8 @@ struct tcp_ao_key *tcp_ao_established_key(const struct sock *sk, { struct tcp_ao_key *key; - hlist_for_each_entry_rcu(key, &ao->head, node, lockdep_sock_is_held(sk)) { + hlist_for_each_entry_rcu(key, &ao->head, node, + sk_fullsock(sk) && lockdep_sock_is_held(sk)) { if ((sndid >= 0 && key->sndid != sndid) || (rcvid >= 0 && key->rcvid != rcvid)) continue; -- cgit v1.2.3 From f8e64e956a635b054227f56e9586a1997add0646 Mon Sep 17 00:00:00 2001 From: Davide Caratti Date: Fri, 8 May 2026 19:05:10 +0200 Subject: net/sched: dualpi2: initialize timer earlier in dualpi2_init() 'pi2_timer' needs to be initialized in all error paths of dualpi2_init(): otherwise, a failure in qdisc_create_dflt() causes the following crash in dualpi2_destroy(): # tc qdisc add dev crash0 handle 1: root dualpi2 BUG: kernel NULL pointer dereference, address: 0000000000000010 #PF: supervisor read access in kernel mode #PF: error_code(0x0000) - not-present page PGD 0 P4D 0 Oops: Oops: 0000 [#1] SMP PTI CPU: 4 UID: 0 PID: 471 Comm: tc Tainted: G E 7.1.0-rc1-virtme #2 PREEMPT(full) Tainted: [E]=UNSIGNED_MODULE Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011 RIP: 0010:hrtimer_active+0x39/0x60 Code: f9 eb 23 0f b6 41 38 3c 01 0f 87 87 64 c0 ff 83 e0 01 75 33 48 39 4a 50 74 28 44 3b 42 10 75 06 48 3b 51 30 74 21 48 8b 51 30 <44> 8b 42 10 41 f6 c0 01 74 cf f3 90 44 8b 42 10 41 f6 c0 01 74 c3 RSP: 0018:ffffd0db80b93620 EFLAGS: 00010282 RAX: ffffffffc0400320 RBX: ffff8cf24a4c86b8 RCX: ffff8cf24a4c86b8 RDX: 0000000000000000 RSI: ffff8cf2429c2ab0 RDI: ffff8cf24a4c86b8 RBP: 00000000fffffff4 R08: 0000000000000003 R09: 0000000000000000 R10: 0000000000000001 R11: ffff8cf24a39c500 R12: ffff8cf24822c000 R13: ffffd0db80b936c0 R14: ffffffffc02cf360 R15: 00000000ffffffff FS: 00007fbc01706580(0000) GS:ffff8cf2dc759000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000000000010 CR3: 0000000008e02003 CR4: 0000000000172ef0 Call Trace: hrtimer_cancel+0x15/0x40 dualpi2_destroy+0x20/0x40 [sch_dualpi2] qdisc_create+0x230/0x570 tc_modify_qdisc+0x716/0xc10 rtnetlink_rcv_msg+0x188/0x780 netlink_rcv_skb+0xcd/0x150 netlink_unicast+0x1ba/0x290 netlink_sendmsg+0x242/0x4d0 ____sys_sendmsg+0x39e/0x3e0 ___sys_sendmsg+0xe1/0x130 __sys_sendmsg+0xad/0x110 do_syscall_64+0x14f/0xf80 entry_SYSCALL_64_after_hwframe+0x77/0x7f RIP: 0033:0x7fbc0188b08e Code: 4d 89 d8 e8 94 bd 00 00 4c 8b 5d f8 41 8b 93 08 03 00 00 59 5e 48 83 f8 fc 74 11 c9 c3 0f 1f 80 00 00 00 00 48 8b 45 10 0f 05 c3 83 e2 39 83 fa 08 75 e7 e8 03 ff ff ff 0f 1f 00 f3 0f 1e fa RSP: 002b:00007fff593260e0 EFLAGS: 00000202 ORIG_RAX: 000000000000002e RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007fbc0188b08e RDX: 0000000000000000 RSI: 00007fff59326190 RDI: 0000000000000003 RBP: 00007fff593260f0 R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000000 R11: 0000000000000202 R12: 000055f06124f260 R13: 0000000069fca043 R14: 000055f061255640 R15: 000055f06124d3f8 Modules linked in: sch_dualpi2(E) CR2: 0000000000000010 [1] https://lore.kernel.org/netdev/2e78e01c504c633ebdff18d041833cf2e079a3a4.1607020450.git.dcaratti@redhat.com/ [2] https://lore.kernel.org/netdev/20200725201707.16909-1-xiyou.wangcong@gmail.com/ v2: - rebased on top of latest net.git Fixes: 320d031ad6e4 ("sched: Struct definition and parsing of dualpi2 qdisc") Signed-off-by: Davide Caratti Reviewed-by: Eric Dumazet Link: https://patch.msgid.link/1faca91179702b31da5d87653e1e036543e32722.1778259798.git.dcaratti@redhat.com Signed-off-by: Jakub Kicinski --- net/sched/sch_dualpi2.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/sched/sch_dualpi2.c b/net/sched/sch_dualpi2.c index 241e6a46bd00..a22489c14458 100644 --- a/net/sched/sch_dualpi2.c +++ b/net/sched/sch_dualpi2.c @@ -938,6 +938,8 @@ static int dualpi2_init(struct Qdisc *sch, struct nlattr *opt, int err; sch->flags |= TCQ_F_DEQUEUE_DROPS; + hrtimer_setup(&q->pi2_timer, dualpi2_timer, CLOCK_MONOTONIC, + HRTIMER_MODE_ABS_PINNED_SOFT); q->l_queue = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops, TC_H_MAKE(sch->handle, 1), extack); @@ -950,8 +952,6 @@ static int dualpi2_init(struct Qdisc *sch, struct nlattr *opt, q->sch = sch; dualpi2_reset_default(sch); - hrtimer_setup(&q->pi2_timer, dualpi2_timer, CLOCK_MONOTONIC, - HRTIMER_MODE_ABS_PINNED_SOFT); if (opt && nla_len(opt)) { err = dualpi2_change(sch, opt, extack); -- cgit v1.2.3 From 1e9fab756f8395096d5bba7be0c373c4c8f5d165 Mon Sep 17 00:00:00 2001 From: Sven Eckelmann Date: Sat, 2 May 2026 19:08:37 +0200 Subject: batman-adv: tt: reject oversized local TVLV buffers The commit 3a359bf5c61d ("batman-adv: reject oversized global TT response buffers") added a check to ensure that a global return buffer size can be stored in an u16. The same buffer handling also exists for the local data buffer but was not touched. A similar check should be also be in place for the local TVLV buffer. It doesn't have the similar attack surface because it is only generated from locally discovered MAC addresses but the dynamic nature could still cause temporarily to large buffers. Cc: stable@kernel.org Fixes: 7ea7b4a14275 ("batman-adv: make the TT CRC logic VLAN specific") Signed-off-by: Sven Eckelmann --- net/batman-adv/translation-table.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/batman-adv/translation-table.c b/net/batman-adv/translation-table.c index 05cddcf994f6..06548dae1039 100644 --- a/net/batman-adv/translation-table.c +++ b/net/batman-adv/translation-table.c @@ -877,12 +877,12 @@ batadv_tt_prepare_tvlv_local_data(struct batadv_priv *bat_priv, { struct batadv_tvlv_tt_vlan_data *tt_vlan; struct batadv_meshif_vlan *vlan; + size_t change_offset; u16 num_vlan = 0; u16 vlan_entries = 0; u16 total_entries = 0; u16 tvlv_len; u8 *tt_change_ptr; - int change_offset; spin_lock_bh(&bat_priv->meshif_vlan_list_lock); hlist_for_each_entry(vlan, &bat_priv->meshif_vlan_list, list) { @@ -900,8 +900,10 @@ batadv_tt_prepare_tvlv_local_data(struct batadv_priv *bat_priv, if (*tt_len < 0) *tt_len = batadv_tt_len(total_entries); - tvlv_len = *tt_len; - tvlv_len += change_offset; + if (check_add_overflow(*tt_len, change_offset, &tvlv_len)) { + tvlv_len = 0; + goto out; + } *tt_data = kmalloc(tvlv_len, GFP_ATOMIC); if (!*tt_data) { -- cgit v1.2.3 From b64963a2ceeb7529310b6cf253a1e540784422f4 Mon Sep 17 00:00:00 2001 From: Sven Eckelmann Date: Sat, 2 May 2026 19:53:21 +0200 Subject: batman-adv: tt: fix negative tt_buff_len batadv_orig_node::tt_buff_len was declared as s16, but the field is never intended to hold a negative value. When a value greater than 32767 is assigned, it wraps to a negative signed integer. In batadv_send_other_tt_response(), tt_buff_len is temporarily widened to s32. The incorrectly negative s16 value propagates into the s32, causing batadv_tt_prepare_tvlv_global_data() to allocate a full sized buffer but populates only a small portion of it with the collected changeset. All remaining bits are kept uninitialized. Using an u16 avoids this type confusion and ensures that no (negative) sign extension is performed in batadv_send_other_tt_response(). Cc: stable@kernel.org Fixes: a73105b8d4c7 ("batman-adv: improved client announcement mechanism") Signed-off-by: Sven Eckelmann --- net/batman-adv/types.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h index b9c0b7779122..888f337a194b 100644 --- a/net/batman-adv/types.h +++ b/net/batman-adv/types.h @@ -452,7 +452,7 @@ struct batadv_orig_node { * @tt_buff_len: length of the last tt changeset this node received * from the orig node */ - s16 tt_buff_len; + u16 tt_buff_len; /** @tt_buff_lock: lock that protects tt_buff and tt_buff_len */ spinlock_t tt_buff_lock; -- cgit v1.2.3 From fc92cdfcb295cefa4344d71a527d61b638b7bfc4 Mon Sep 17 00:00:00 2001 From: Sven Eckelmann Date: Sat, 2 May 2026 19:53:21 +0200 Subject: batman-adv: tt: fix negative last_changeset_len batadv_piv_tt::last_changeset_len len was declared as s16, but the field is never intended to hold a negative value. When a value greater than 32767 is assigned, it wraps to a negative signed integer. In batadv_send_my_tt_response(), last_changeset_len is temporarily widened to s32. The incorrectly negative s16 value propagates into the s32, causing batadv_tt_prepare_tvlv_local_data() to allocate a full sized buffer but populates only a small portion of it with the collected changeset. All remaining bits are kept uninitialized. Using an u16 avoids this type confusion and ensures that no (negative) sign extension is performed in batadv_send_my_tt_response(). Cc: stable@kernel.org Fixes: a73105b8d4c7 ("batman-adv: improved client announcement mechanism") Signed-off-by: Sven Eckelmann --- net/batman-adv/types.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h index 888f337a194b..739439e2b235 100644 --- a/net/batman-adv/types.h +++ b/net/batman-adv/types.h @@ -993,7 +993,7 @@ struct batadv_priv_tt { * @last_changeset_len: length of last tt changeset this host has * generated */ - s16 last_changeset_len; + u16 last_changeset_len; /** * @last_changeset_lock: lock protecting last_changeset & -- cgit v1.2.3 From 94d27005016be15ffc638b2ecbc4d58805ad7b48 Mon Sep 17 00:00:00 2001 From: Sven Eckelmann Date: Sat, 2 May 2026 19:47:11 +0200 Subject: batman-adv: tt: fix TOCTOU race for reported vlans The local TT based TVLV is generated by first checking the number of VLANs which have at least one TT entry. A new buffer with the correct size for the VLANs is then allocated. Only then, the list of VLANs s used to fill the VLAN entries in the buffer. During this time, the meshif_vlan_list_lock is held. But the actual number of TT entries of each VLAN can still increase during this time - just not the number of VLANs in the list. But the prefilter used in the buffer size calculation might still cause an increase of the number of VLANs which need to be stored. Simply because a VLAN might now suddenly have at least one entry when it had none in the pre-alloc check - and then needs to occupy space which was not allocated. It is better to overestimate the buffer size at the beginning and then fill the buffer only with the VLANs which are not empty. Cc: stable@kernel.org Fixes: 16116dac2339 ("batman-adv: prevent TT request storms by not sending inconsistent TT TLVLs") Signed-off-by: Sven Eckelmann --- net/batman-adv/translation-table.c | 13 +++++++++---- 1 file changed, 9 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/batman-adv/translation-table.c b/net/batman-adv/translation-table.c index 06548dae1039..f009cbf8a276 100644 --- a/net/batman-adv/translation-table.c +++ b/net/batman-adv/translation-table.c @@ -887,11 +887,8 @@ batadv_tt_prepare_tvlv_local_data(struct batadv_priv *bat_priv, spin_lock_bh(&bat_priv->meshif_vlan_list_lock); hlist_for_each_entry(vlan, &bat_priv->meshif_vlan_list, list) { vlan_entries = atomic_read(&vlan->tt.num_entries); - if (vlan_entries < 1) - continue; - - num_vlan++; total_entries += vlan_entries; + num_vlan++; } change_offset = struct_size(*tt_data, vlan_data, num_vlan); @@ -916,6 +913,7 @@ batadv_tt_prepare_tvlv_local_data(struct batadv_priv *bat_priv, (*tt_data)->num_vlan = htons(num_vlan); tt_vlan = (*tt_data)->vlan_data; + num_vlan = 0; hlist_for_each_entry(vlan, &bat_priv->meshif_vlan_list, list) { vlan_entries = atomic_read(&vlan->tt.num_entries); if (vlan_entries < 1) @@ -926,8 +924,15 @@ batadv_tt_prepare_tvlv_local_data(struct batadv_priv *bat_priv, tt_vlan->reserved = 0; tt_vlan++; + num_vlan++; } + /* recalculate in case number of VLANs reduced */ + change_offset = struct_size(*tt_data, vlan_data, num_vlan); + tvlv_len = *tt_len + change_offset; + + (*tt_data)->num_vlan = htons(num_vlan); + tt_change_ptr = (u8 *)*tt_data + change_offset; *tt_change = (struct batadv_tvlv_tt_change *)tt_change_ptr; -- cgit v1.2.3 From fa1bd704940b5bcbc32c0b28db9167405c8ee5e0 Mon Sep 17 00:00:00 2001 From: Sven Eckelmann Date: Sat, 2 May 2026 20:47:34 +0200 Subject: batman-adv: tt: avoid empty VLAN responses The commit 16116dac2339 ("batman-adv: prevent TT request storms by not sending inconsistent TT TLVLs") added checks to the local (direct) TT response code. But the response can also be done indirectly by another node using the global TT state. To avoid such inconsistency states reported in the original fix, also avoid sending empty VLANs for replies from the global TT state. Cc: stable@kernel.org Fixes: 7ea7b4a14275 ("batman-adv: make the TT CRC logic VLAN specific") Signed-off-by: Sven Eckelmann --- net/batman-adv/translation-table.c | 20 +++++++++++++++++--- 1 file changed, 17 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/batman-adv/translation-table.c b/net/batman-adv/translation-table.c index f009cbf8a276..2259b241e0b5 100644 --- a/net/batman-adv/translation-table.c +++ b/net/batman-adv/translation-table.c @@ -797,24 +797,26 @@ batadv_tt_prepare_tvlv_global_data(struct batadv_orig_node *orig_node, s32 *tt_len) { u16 num_vlan = 0; - u16 num_entries = 0; u16 tvlv_len = 0; unsigned int change_offset; struct batadv_tvlv_tt_vlan_data *tt_vlan; struct batadv_orig_node_vlan *vlan; + u16 total_entries = 0; u8 *tt_change_ptr; + int vlan_entries; spin_lock_bh(&orig_node->vlan_list_lock); hlist_for_each_entry(vlan, &orig_node->vlan_list, list) { + vlan_entries = atomic_read(&vlan->tt.num_entries); + total_entries += vlan_entries; num_vlan++; - num_entries += atomic_read(&vlan->tt.num_entries); } change_offset = struct_size(*tt_data, vlan_data, num_vlan); /* if tt_len is negative, allocate the space needed by the full table */ if (*tt_len < 0) - *tt_len = batadv_tt_len(num_entries); + *tt_len = batadv_tt_len(total_entries); if (change_offset > U16_MAX || *tt_len > U16_MAX - change_offset) { *tt_len = 0; @@ -835,14 +837,26 @@ batadv_tt_prepare_tvlv_global_data(struct batadv_orig_node *orig_node, (*tt_data)->num_vlan = htons(num_vlan); tt_vlan = (*tt_data)->vlan_data; + num_vlan = 0; hlist_for_each_entry(vlan, &orig_node->vlan_list, list) { + vlan_entries = atomic_read(&vlan->tt.num_entries); + if (vlan_entries < 1) + continue; + tt_vlan->vid = htons(vlan->vid); tt_vlan->crc = htonl(vlan->tt.crc); tt_vlan->reserved = 0; tt_vlan++; + num_vlan++; } + /* recalculate in case number of VLANs reduced */ + change_offset = struct_size(*tt_data, vlan_data, num_vlan); + tvlv_len = *tt_len + change_offset; + + (*tt_data)->num_vlan = htons(num_vlan); + tt_change_ptr = (u8 *)*tt_data + change_offset; *tt_change = (struct batadv_tvlv_tt_change *)tt_change_ptr; -- cgit v1.2.3 From 99d9958fa10fb684b2a8e2c48a8d704122721420 Mon Sep 17 00:00:00 2001 From: Sven Eckelmann Date: Sat, 2 May 2026 21:25:19 +0200 Subject: batman-adv: tt: prevent TVLV entry number overflow The helpers to prepare the buffers for the local and global TT based replies are trying to sum up all TT entries which can be found for each VLAN. In theory, this sum can be too big for an u16 and therefore overflow. A too small buffer would then be allocated for the TVLV. The too small buffer will be handled gracefully by batadv_tt_tvlv_generate() and is not causing a buffer overflow - just a truncated reply. But this overflow shouldn't have happened in the first and the too small buffer should never have been allocated when an overflow was detected. Cc: stable@kernel.org Fixes: 7ea7b4a14275 ("batman-adv: make the TT CRC logic VLAN specific") Signed-off-by: Sven Eckelmann --- net/batman-adv/translation-table.c | 20 +++++++++++++++++--- 1 file changed, 17 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/batman-adv/translation-table.c b/net/batman-adv/translation-table.c index 2259b241e0b5..9f6e67771ffa 100644 --- a/net/batman-adv/translation-table.c +++ b/net/batman-adv/translation-table.c @@ -804,11 +804,18 @@ batadv_tt_prepare_tvlv_global_data(struct batadv_orig_node *orig_node, u16 total_entries = 0; u8 *tt_change_ptr; int vlan_entries; + u16 sum_entries; spin_lock_bh(&orig_node->vlan_list_lock); hlist_for_each_entry(vlan, &orig_node->vlan_list, list) { vlan_entries = atomic_read(&vlan->tt.num_entries); - total_entries += vlan_entries; + + if (check_add_overflow(vlan_entries, total_entries, &sum_entries)) { + *tt_len = 0; + goto out; + } + + total_entries = sum_entries; num_vlan++; } @@ -893,15 +900,22 @@ batadv_tt_prepare_tvlv_local_data(struct batadv_priv *bat_priv, struct batadv_meshif_vlan *vlan; size_t change_offset; u16 num_vlan = 0; - u16 vlan_entries = 0; u16 total_entries = 0; u16 tvlv_len; u8 *tt_change_ptr; + int vlan_entries; + u16 sum_entries; spin_lock_bh(&bat_priv->meshif_vlan_list_lock); hlist_for_each_entry(vlan, &bat_priv->meshif_vlan_list, list) { vlan_entries = atomic_read(&vlan->tt.num_entries); - total_entries += vlan_entries; + + if (check_add_overflow(vlan_entries, total_entries, &sum_entries)) { + tvlv_len = 0; + goto out; + } + + total_entries = sum_entries; num_vlan++; } -- cgit v1.2.3 From 911f54771ca97947cfdca360e9e9b4147a330740 Mon Sep 17 00:00:00 2001 From: Quan Sun <2022090917019@std.uestc.edu.cn> Date: Fri, 8 May 2026 20:46:36 +0800 Subject: net: hsr: fix NULL pointer dereference in hsr_get_node_data() In the HSR (High-availability Seamless Redundancy) protocol, node information is maintained in the node_db. When a supervision frame is received, node->addr_B_port is updated to track the receiving port type (e.g., HSR_PT_SLAVE_B). If the underlying physical interface associated with this slave port is removed (e.g., via `ip link del`), hsr_del_port() frees the hsr_port object. However, the stale node->addr_B_port reference is kept in the node_db until the node ages out. Subsequently, if userspace queries the node status via the Netlink command HSR_C_GET_NODE_STATUS, the kernel calls hsr_get_node_data(). This function unconditionally dereferences the pointer returned by hsr_port_get_hsr(): if (node->addr_B_port != HSR_PT_NONE) { port = hsr_port_get_hsr(hsr, node->addr_B_port); *addr_b_ifindex = port->dev->ifindex; // <-- NULL deref } If the slave port has been deleted, hsr_port_get_hsr() returns NULL, resulting in a kernel panic. Oops: general protection fault, probably for non-canonical address KASAN: null-ptr-deref in range [0x0000000000000010-0x0000000000000017] RIP: 0010:hsr_get_node_data+0x7b6/0x9e0 Call Trace: hsr_get_node_status+0x445/0xa40 Fix this by adding a proper NULL pointer check. If the port lookup fails due to a stale port type, gracefully treat it as if no valid port exists and assign -1 to the interface index. Steps to reproduce: 1. Create an HSR interface with two slave devices. 2. Receive a supervision frame to populate node_db with addr_B_port assigned to SLAVE_B. 3. Delete the underlying slave device B. 4. Send an HSR_C_GET_NODE_STATUS Netlink message. Fixes: c5a759117210 ("net/hsr: Use list_head (and rcu) instead of array for slave devices.") Signed-off-by: Quan Sun <2022090917019@std.uestc.edu.cn> Link: https://patch.msgid.link/20260508124636.1462346-1-2022090917019@std.uestc.edu.cn Signed-off-by: Paolo Abeni --- net/hsr/hsr_framereg.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/hsr/hsr_framereg.c b/net/hsr/hsr_framereg.c index d09875b33588..124619920d38 100644 --- a/net/hsr/hsr_framereg.c +++ b/net/hsr/hsr_framereg.c @@ -889,7 +889,10 @@ int hsr_get_node_data(struct hsr_priv *hsr, if (node->addr_B_port != HSR_PT_NONE) { port = hsr_port_get_hsr(hsr, node->addr_B_port); - *addr_b_ifindex = port->dev->ifindex; + if (port) + *addr_b_ifindex = port->dev->ifindex; + else + *addr_b_ifindex = -1; } else { *addr_b_ifindex = -1; } -- cgit v1.2.3 From 5f344d809e015fba3709e5219428c00b8ac5d7df Mon Sep 17 00:00:00 2001 From: Stefano Garzarella Date: Fri, 8 May 2026 18:44:10 +0200 Subject: vsock/virtio: fix length and offset in tap skb for split packets virtio_transport_build_skb() builds a new skb to be delivered to the vsockmon tap device. To build the new skb, it uses the original skb data length as payload length, but as the comment notes, the original packet stored in the skb may have been split in multiple packets, so we need to use the length in the header, which is correctly updated before the packet is delivered to the tap, and the offset for the data. This was also similar to what we did before commit 71dc9ec9ac7d ("virtio/vsock: replace virtio_vsock_pkt with sk_buff") where we probably missed something during the skb conversion. Also update the comment above, which was left stale by the skb conversion and still mentioned a buffer pointer that no longer exists. Fixes: 71dc9ec9ac7d ("virtio/vsock: replace virtio_vsock_pkt with sk_buff") Signed-off-by: Stefano Garzarella Reviewed-by: Bobby Eshleman Reviewed-by: Arseniy Krasnov Link: https://patch.msgid.link/20260508164411.261440-2-sgarzare@redhat.com Acked-by: Michael S. Tsirkin Signed-off-by: Paolo Abeni --- net/vmw_vsock/virtio_transport_common.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/vmw_vsock/virtio_transport_common.c b/net/vmw_vsock/virtio_transport_common.c index 9b8014516f4f..a678d5d75704 100644 --- a/net/vmw_vsock/virtio_transport_common.c +++ b/net/vmw_vsock/virtio_transport_common.c @@ -166,12 +166,12 @@ static struct sk_buff *virtio_transport_build_skb(void *opaque) struct sk_buff *skb; size_t payload_len; - /* A packet could be split to fit the RX buffer, so we can retrieve - * the payload length from the header and the buffer pointer taking - * care of the offset in the original packet. + /* A packet could be split to fit the RX buffer, so we use + * the payload length from the header, which has been updated + * by the sender to reflect the fragment size. */ pkt_hdr = virtio_vsock_hdr(pkt); - payload_len = pkt->len; + payload_len = le32_to_cpu(pkt_hdr->len); skb = alloc_skb(sizeof(*hdr) + sizeof(*pkt_hdr) + payload_len, GFP_ATOMIC); @@ -219,7 +219,8 @@ static struct sk_buff *virtio_transport_build_skb(void *opaque) virtio_transport_copy_nonlinear_skb(pkt, data, payload_len); } else { - skb_put_data(skb, pkt->data, payload_len); + skb_put_data(skb, pkt->data + VIRTIO_VSOCK_SKB_CB(pkt)->offset, + payload_len); } } -- cgit v1.2.3 From 3a3e3d90cbc79600544536723911657730759af3 Mon Sep 17 00:00:00 2001 From: Stefano Garzarella Date: Fri, 8 May 2026 18:44:11 +0200 Subject: vsock/virtio: fix empty payload in tap skb for non-linear buffers For non-linear skbs, virtio_transport_build_skb() goes through virtio_transport_copy_nonlinear_skb() to copy the original payload in the new skb to be delivered to the vsockmon tap device. This manually initializes an iov_iter but does not set iov_iter.count. Since the iov_iter is zero-initialized, the copy length is zero and no payload is actually copied to the monitor interface, leaving data un-initialized. Fix this by removing the linear vs non-linear split and using skb_copy_datagram_iter() with iov_iter_kvec() for all cases, as vhost-vsock already does. This handles both linear and non-linear skbs, properly initializes the iov_iter, and removes the now unused virtio_transport_copy_nonlinear_skb(). While touching this code, let's also check the return value of skb_copy_datagram_iter(), even though it's unlikely to fail. Fixes: 4b0bf10eb077 ("vsock/virtio: non-linear skb handling for tap") Reported-by: Yiqi Sun Signed-off-by: Stefano Garzarella Reviewed-by: Bobby Eshleman Reviewed-by: Arseniy Krasnov Link: https://patch.msgid.link/20260508164411.261440-3-sgarzare@redhat.com Acked-by: Michael S. Tsirkin Signed-off-by: Paolo Abeni --- net/vmw_vsock/virtio_transport_common.c | 40 ++++++++++----------------------- 1 file changed, 12 insertions(+), 28 deletions(-) (limited to 'net') diff --git a/net/vmw_vsock/virtio_transport_common.c b/net/vmw_vsock/virtio_transport_common.c index a678d5d75704..989cc252d3d3 100644 --- a/net/vmw_vsock/virtio_transport_common.c +++ b/net/vmw_vsock/virtio_transport_common.c @@ -136,27 +136,6 @@ static void virtio_transport_init_hdr(struct sk_buff *skb, hdr->fwd_cnt = cpu_to_le32(0); } -static void virtio_transport_copy_nonlinear_skb(const struct sk_buff *skb, - void *dst, - size_t len) -{ - struct iov_iter iov_iter = { 0 }; - struct kvec kvec; - size_t to_copy; - - kvec.iov_base = dst; - kvec.iov_len = len; - - iov_iter.iter_type = ITER_KVEC; - iov_iter.kvec = &kvec; - iov_iter.nr_segs = 1; - - to_copy = min_t(size_t, len, skb->len); - - skb_copy_datagram_iter(skb, VIRTIO_VSOCK_SKB_CB(skb)->offset, - &iov_iter, to_copy); -} - /* Packet capture */ static struct sk_buff *virtio_transport_build_skb(void *opaque) { @@ -214,13 +193,18 @@ static struct sk_buff *virtio_transport_build_skb(void *opaque) skb_put_data(skb, pkt_hdr, sizeof(*pkt_hdr)); if (payload_len) { - if (skb_is_nonlinear(pkt)) { - void *data = skb_put(skb, payload_len); - - virtio_transport_copy_nonlinear_skb(pkt, data, payload_len); - } else { - skb_put_data(skb, pkt->data + VIRTIO_VSOCK_SKB_CB(pkt)->offset, - payload_len); + struct iov_iter iov_iter; + struct kvec kvec; + void *data = skb_put(skb, payload_len); + + kvec.iov_base = data; + kvec.iov_len = payload_len; + iov_iter_kvec(&iov_iter, ITER_DEST, &kvec, 1, payload_len); + + if (skb_copy_datagram_iter(pkt, VIRTIO_VSOCK_SKB_CB(pkt)->offset, + &iov_iter, payload_len)) { + kfree_skb(skb); + return NULL; } } -- cgit v1.2.3 From 7cee43fcb0c3f71441d2faaa8c2202b6a88b6bef Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Sun, 10 May 2026 12:28:55 -0700 Subject: net: shaper: flip the polarity of the valid flag The usual way of inserting entries which are not yet fully ready into XArray is to have a VALID flag. The shaper code has a NOT_VALID flag. Since XArray code does not let us create entries with marks already set - the creation of entries is currently not atomic. Flip the polarity of the VALID flag. This closes the tiny race in net_shaper_pre_insert() of entries being created without the NOT_VALID flag. Fixes: 93954b40f6a4 ("net-shapers: implement NL set and delete operations") Signed-off-by: Jakub Kicinski Link: https://patch.msgid.link/20260510192904.3987113-2-kuba@kernel.org Signed-off-by: Paolo Abeni --- net/shaper/shaper.c | 34 +++++++++++++++++----------------- 1 file changed, 17 insertions(+), 17 deletions(-) (limited to 'net') diff --git a/net/shaper/shaper.c b/net/shaper/shaper.c index 1069fa4eb9f6..d2b8f1f951b1 100644 --- a/net/shaper/shaper.c +++ b/net/shaper/shaper.c @@ -275,11 +275,13 @@ static void net_shaper_default_parent(const struct net_shaper_handle *handle, parent->id = 0; } -/* - * MARK_0 is already in use due to XA_FLAGS_ALLOC, can't reuse such flag as - * it's cleared by xa_store(). +/* MARK_0 is already in use due to XA_FLAGS_ALLOC. The VALID mark is set on + * an entry only after the device-side configuration has completed + * successfully (see net_shaper_commit()). Lookups and dumps must filter on + * this mark to avoid exposing tentative entries inserted by + * net_shaper_pre_insert() while the driver call is still in flight. */ -#define NET_SHAPER_NOT_VALID XA_MARK_1 +#define NET_SHAPER_VALID XA_MARK_1 static struct net_shaper * net_shaper_lookup(struct net_shaper_binding *binding, @@ -289,8 +291,8 @@ net_shaper_lookup(struct net_shaper_binding *binding, struct net_shaper_hierarchy *hierarchy; hierarchy = net_shaper_hierarchy_rcu(binding); - if (!hierarchy || xa_get_mark(&hierarchy->shapers, index, - NET_SHAPER_NOT_VALID)) + if (!hierarchy || !xa_get_mark(&hierarchy->shapers, index, + NET_SHAPER_VALID)) return NULL; return xa_load(&hierarchy->shapers, index); @@ -370,13 +372,10 @@ static int net_shaper_pre_insert(struct net_shaper_binding *binding, goto free_id; } - /* Mark 'tentative' shaper inside the hierarchy container. - * xa_set_mark is a no-op if the previous store fails. + /* Insert as 'tentative' (no VALID mark). The mark will be set by + * net_shaper_commit() once the driver-side configuration succeeds. */ - xa_lock(&hierarchy->shapers); - prev = __xa_store(&hierarchy->shapers, index, cur, GFP_KERNEL); - __xa_set_mark(&hierarchy->shapers, index, NET_SHAPER_NOT_VALID); - xa_unlock(&hierarchy->shapers); + prev = xa_store(&hierarchy->shapers, index, cur, GFP_KERNEL); if (xa_err(prev)) { NL_SET_ERR_MSG(extack, "Can't insert shaper into device store"); kfree_rcu(cur, rcu); @@ -413,8 +412,7 @@ static void net_shaper_commit(struct net_shaper_binding *binding, /* Successful update: drop the tentative mark * and update the hierarchy container. */ - __xa_clear_mark(&hierarchy->shapers, index, - NET_SHAPER_NOT_VALID); + __xa_set_mark(&hierarchy->shapers, index, NET_SHAPER_VALID); *cur = shapers[i]; } xa_unlock(&hierarchy->shapers); @@ -431,8 +429,9 @@ static void net_shaper_rollback(struct net_shaper_binding *binding) return; xa_lock(&hierarchy->shapers); - xa_for_each_marked(&hierarchy->shapers, index, cur, - NET_SHAPER_NOT_VALID) { + xa_for_each(&hierarchy->shapers, index, cur) { + if (xa_get_mark(&hierarchy->shapers, index, NET_SHAPER_VALID)) + continue; __xa_erase(&hierarchy->shapers, index); kfree(cur); } @@ -836,7 +835,8 @@ int net_shaper_nl_get_dumpit(struct sk_buff *skb, goto out_unlock; for (; (shaper = xa_find(&hierarchy->shapers, &ctx->start_index, - U32_MAX, XA_PRESENT)); ctx->start_index++) { + U32_MAX, NET_SHAPER_VALID)); + ctx->start_index++) { ret = net_shaper_fill_one(skb, binding, shaper, info); if (ret) break; -- cgit v1.2.3 From 235fb5376139c3419f2218349f1fa2f06f24f7ad Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Sun, 10 May 2026 12:28:56 -0700 Subject: net: shaper: fix trivial ordering issue in net_shaper_commit() We should update the entry before we mark it as valid. Fixes: 93954b40f6a4 ("net-shapers: implement NL set and delete operations") Signed-off-by: Jakub Kicinski Link: https://patch.msgid.link/20260510192904.3987113-3-kuba@kernel.org Signed-off-by: Paolo Abeni --- net/shaper/shaper.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/shaper/shaper.c b/net/shaper/shaper.c index d2b8f1f951b1..86319ddbf290 100644 --- a/net/shaper/shaper.c +++ b/net/shaper/shaper.c @@ -295,6 +295,10 @@ net_shaper_lookup(struct net_shaper_binding *binding, NET_SHAPER_VALID)) return NULL; + /* Pairs with smp_wmb() in net_shaper_commit(): if the entry is + * valid, its contents must be visible too. + */ + smp_rmb(); return xa_load(&hierarchy->shapers, index); } @@ -412,8 +416,9 @@ static void net_shaper_commit(struct net_shaper_binding *binding, /* Successful update: drop the tentative mark * and update the hierarchy container. */ - __xa_set_mark(&hierarchy->shapers, index, NET_SHAPER_VALID); *cur = shapers[i]; + smp_wmb(); + __xa_set_mark(&hierarchy->shapers, index, NET_SHAPER_VALID); } xa_unlock(&hierarchy->shapers); } @@ -837,6 +842,10 @@ int net_shaper_nl_get_dumpit(struct sk_buff *skb, for (; (shaper = xa_find(&hierarchy->shapers, &ctx->start_index, U32_MAX, NET_SHAPER_VALID)); ctx->start_index++) { + /* Pairs with smp_wmb() in net_shaper_commit(): the entry + * is marked VALID, so its contents must be visible too. + */ + smp_rmb(); ret = net_shaper_fill_one(skb, binding, shaper, info); if (ret) break; -- cgit v1.2.3 From a9a2fa1da619f276580b0d4c5d12efac89e8642b Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Sun, 10 May 2026 12:28:57 -0700 Subject: net: shaper: reject duplicate leaves in GROUP request net_shaper_nl_group_doit() does not deduplicate NET_SHAPER_A_LEAVES entries. When userspace supplies the same leaf handle twice, the same old-parent pointer lands twice in old_nodes[]. The cleanup loop double frees the parent. Of course the same parent may still be in old_nodes[] twice if we are moving multiple of its leaves. Note that this patch also implicitly fixes the fact that the i >= leaves_count path forgets to set ret. Fixes: 5d5d4700e75d ("net-shapers: implement NL group operation") Signed-off-by: Jakub Kicinski Link: https://patch.msgid.link/20260510192904.3987113-4-kuba@kernel.org Signed-off-by: Paolo Abeni --- net/shaper/shaper.c | 60 +++++++++++++++++++++++++++++++++++++++-------------- 1 file changed, 45 insertions(+), 15 deletions(-) (limited to 'net') diff --git a/net/shaper/shaper.c b/net/shaper/shaper.c index 86319ddbf290..c8960821cf23 100644 --- a/net/shaper/shaper.c +++ b/net/shaper/shaper.c @@ -941,6 +941,46 @@ static int net_shaper_handle_cmp(const struct net_shaper_handle *a, return memcmp(a, b, sizeof(*a)); } +static int net_shaper_parse_leaves(struct net_shaper_binding *binding, + struct genl_info *info, + const struct net_shaper *node, + struct net_shaper *leaves, + int leaves_count) +{ + struct nlattr *attr; + int i, j, ret, rem; + + i = 0; + nla_for_each_attr_type(attr, NET_SHAPER_A_LEAVES, + genlmsg_data(info->genlhdr), + genlmsg_len(info->genlhdr), rem) { + if (WARN_ON_ONCE(i >= leaves_count)) + return -EINVAL; + + ret = net_shaper_parse_leaf(binding, attr, info, + node, &leaves[i]); + if (ret) + return ret; + + /* Reject duplicates */ + for (j = 0; j < i; j++) { + if (net_shaper_handle_cmp(&leaves[i].handle, + &leaves[j].handle)) + continue; + + NL_SET_ERR_MSG_ATTR_FMT(info->extack, attr, + "Duplicate leaf shaper %d:%d", + leaves[i].handle.scope, + leaves[i].handle.id); + return -EINVAL; + } + + i++; + } + + return 0; +} + static int net_shaper_parent_from_leaves(int leaves_count, const struct net_shaper *leaves, struct net_shaper *node, @@ -1197,10 +1237,9 @@ int net_shaper_nl_group_doit(struct sk_buff *skb, struct genl_info *info) struct net_shaper **old_nodes, *leaves, node = {}; struct net_shaper_hierarchy *hierarchy; struct net_shaper_binding *binding; - int i, ret, rem, leaves_count; + int i, ret, leaves_count; int old_nodes_count = 0; struct sk_buff *msg; - struct nlattr *attr; if (GENL_REQ_ATTR_CHECK(info, NET_SHAPER_A_LEAVES)) return -EINVAL; @@ -1228,19 +1267,10 @@ int net_shaper_nl_group_doit(struct sk_buff *skb, struct genl_info *info) if (ret) goto free_leaves; - i = 0; - nla_for_each_attr_type(attr, NET_SHAPER_A_LEAVES, - genlmsg_data(info->genlhdr), - genlmsg_len(info->genlhdr), rem) { - if (WARN_ON_ONCE(i >= leaves_count)) - goto free_leaves; - - ret = net_shaper_parse_leaf(binding, attr, info, - &node, &leaves[i]); - if (ret) - goto free_leaves; - i++; - } + ret = net_shaper_parse_leaves(binding, info, &node, + leaves, leaves_count); + if (ret) + goto free_leaves; /* Prepare the msg reply in advance, to avoid device operation * rollback on allocation failure. -- cgit v1.2.3 From 8054f85b83f42a37d482fc77ea7c9ff06a9407d9 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Sun, 10 May 2026 12:28:59 -0700 Subject: net: shaper: set ret to -ENOMEM when genlmsg_new() fails in group_doit genlmsg_new() alloc failure path in net_shaper_nl_group_doit() forgets to set ret before jumping to error handling. Fixes: 5d5d4700e75d ("net-shapers: implement NL group operation") Signed-off-by: Jakub Kicinski Link: https://patch.msgid.link/20260510192904.3987113-6-kuba@kernel.org Signed-off-by: Paolo Abeni --- net/shaper/shaper.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/shaper/shaper.c b/net/shaper/shaper.c index c8960821cf23..12e5e0c18643 100644 --- a/net/shaper/shaper.c +++ b/net/shaper/shaper.c @@ -1276,8 +1276,10 @@ int net_shaper_nl_group_doit(struct sk_buff *skb, struct genl_info *info) * rollback on allocation failure. */ msg = genlmsg_new(net_shaper_handle_size(), GFP_KERNEL); - if (!msg) + if (!msg) { + ret = -ENOMEM; goto free_leaves; + } hierarchy = net_shaper_hierarchy_setup(binding); if (!hierarchy) { -- cgit v1.2.3 From 0f9a857e34d0f8c018a3e4435c6f0e92e8d2f38c Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Sun, 10 May 2026 12:29:00 -0700 Subject: net: shaper: fix undersized reply skb allocation in GROUP command net_shaper_group_send_reply() writes both the NET_SHAPER_A_IFINDEX attribute (via net_shaper_fill_binding()) and the nested NET_SHAPER_A_HANDLE attribute (via net_shaper_fill_handle()), but the reply skb at the call site in net_shaper_nl_group_doit() is allocated using net_shaper_handle_size(), which only accounts for the nested handle. The allocation is therefore short by nla_total_size(sizeof(u32)) (8 bytes) for the IFINDEX attribute. In practice the slab allocator rounds up the small allocation so the bug is latent, but the size accounting is wrong and could bite if the reply grew further. Introduce net_shaper_group_reply_size() that accounts for the full reply payload and use it both at the genlmsg_new() call site and in the defensive WARN_ONCE message. Fixes: 5d5d4700e75d ("net-shapers: implement NL group operation") Signed-off-by: Jakub Kicinski Link: https://patch.msgid.link/20260510192904.3987113-7-kuba@kernel.org Signed-off-by: Paolo Abeni --- net/shaper/shaper.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/shaper/shaper.c b/net/shaper/shaper.c index 12e5e0c18643..08fde2d9e8aa 100644 --- a/net/shaper/shaper.c +++ b/net/shaper/shaper.c @@ -90,6 +90,12 @@ static int net_shaper_handle_size(void) nla_total_size(sizeof(u32))); } +static int net_shaper_group_reply_size(void) +{ + return nla_total_size(sizeof(u32)) + /* NET_SHAPER_A_IFINDEX */ + net_shaper_handle_size(); /* NET_SHAPER_A_HANDLE */ +} + static int net_shaper_fill_binding(struct sk_buff *msg, const struct net_shaper_binding *binding, u32 type) @@ -1227,7 +1233,7 @@ static int net_shaper_group_send_reply(struct net_shaper_binding *binding, free_msg: /* Should never happen as msg is pre-allocated with enough space. */ WARN_ONCE(true, "calculated message payload length (%d)", - net_shaper_handle_size()); + net_shaper_group_reply_size()); nlmsg_free(msg); return -EMSGSIZE; } @@ -1275,7 +1281,7 @@ int net_shaper_nl_group_doit(struct sk_buff *skb, struct genl_info *info) /* Prepare the msg reply in advance, to avoid device operation * rollback on allocation failure. */ - msg = genlmsg_new(net_shaper_handle_size(), GFP_KERNEL); + msg = genlmsg_new(net_shaper_group_reply_size(), GFP_KERNEL); if (!msg) { ret = -ENOMEM; goto free_leaves; -- cgit v1.2.3 From 8d5806c600fddb907ebe378f9c366d4b52ac3a39 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Sun, 10 May 2026 12:29:02 -0700 Subject: net: shaper: reject handle IDs exceeding internal bit-width net_shaper_parse_handle() reads the user-supplied handle ID via nla_get_u32(), accepting the full u32 range. However, the xarray key is built by net_shaper_handle_to_index() using FIELD_PREP(NET_SHAPER_ID_MASK, handle->id), where NET_SHAPER_ID_MASK is GENMASK(25, 0) - only 26 bits wide. FIELD_PREP silently masks off the upper bits at runtime. A user-supplied NODE id like 0x04000123 becomes id 0x123. Additionally, a user-supplied id equal to NET_SHAPER_ID_UNSPEC (0x03FFFFFF, which is NET_SHAPER_ID_MASK itself) would collide with the sentinel used internally by the group operation to signal "allocate a new NODE id". Reject user-supplied IDs >= NET_SHAPER_ID_MASK (i.e., >= 0x03FFFFFF) in the policy. Fixes: 4b623f9f0f59 ("net-shapers: implement NL get operation") Signed-off-by: Jakub Kicinski Link: https://patch.msgid.link/20260510192904.3987113-9-kuba@kernel.org Signed-off-by: Paolo Abeni --- net/shaper/shaper.c | 4 +++- net/shaper/shaper_nl_gen.c | 7 ++++++- net/shaper/shaper_nl_gen.h | 2 ++ 3 files changed, 11 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/shaper/shaper.c b/net/shaper/shaper.c index 08fde2d9e8aa..eb049847fed6 100644 --- a/net/shaper/shaper.c +++ b/net/shaper/shaper.c @@ -21,6 +21,8 @@ #define NET_SHAPER_ID_UNSPEC NET_SHAPER_ID_MASK +static_assert(NET_SHAPER_ID_UNSPEC == NET_SHAPER_MAX_HANDLE_ID + 1); + struct net_shaper_hierarchy { struct xarray shapers; }; @@ -360,7 +362,7 @@ static int net_shaper_pre_insert(struct net_shaper_binding *binding, handle->id == NET_SHAPER_ID_UNSPEC) { u32 min, max; - handle->id = NET_SHAPER_ID_MASK - 1; + handle->id = NET_SHAPER_MAX_HANDLE_ID; max = net_shaper_handle_to_index(handle); handle->id = 0; min = net_shaper_handle_to_index(handle); diff --git a/net/shaper/shaper_nl_gen.c b/net/shaper/shaper_nl_gen.c index 9b29be3ef19a..76eff85ec66d 100644 --- a/net/shaper/shaper_nl_gen.c +++ b/net/shaper/shaper_nl_gen.c @@ -11,10 +11,15 @@ #include +/* Integer value ranges */ +static const struct netlink_range_validation net_shaper_a_handle_id_range = { + .max = NET_SHAPER_MAX_HANDLE_ID, +}; + /* Common nested types */ const struct nla_policy net_shaper_handle_nl_policy[NET_SHAPER_A_HANDLE_ID + 1] = { [NET_SHAPER_A_HANDLE_SCOPE] = NLA_POLICY_MAX(NLA_U32, 3), - [NET_SHAPER_A_HANDLE_ID] = { .type = NLA_U32, }, + [NET_SHAPER_A_HANDLE_ID] = NLA_POLICY_FULL_RANGE(NLA_U32, &net_shaper_a_handle_id_range), }; const struct nla_policy net_shaper_leaf_info_nl_policy[NET_SHAPER_A_WEIGHT + 1] = { diff --git a/net/shaper/shaper_nl_gen.h b/net/shaper/shaper_nl_gen.h index 42c46c52c775..2406652a9014 100644 --- a/net/shaper/shaper_nl_gen.h +++ b/net/shaper/shaper_nl_gen.h @@ -12,6 +12,8 @@ #include +#define NET_SHAPER_MAX_HANDLE_ID 67108862 + /* Common nested types */ extern const struct nla_policy net_shaper_handle_nl_policy[NET_SHAPER_A_HANDLE_ID + 1]; extern const struct nla_policy net_shaper_leaf_info_nl_policy[NET_SHAPER_A_WEIGHT + 1]; -- cgit v1.2.3 From b62b29e6de6711f5918940aa6ff2bbab6d6af502 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Sun, 10 May 2026 12:29:03 -0700 Subject: net: shaper: enforce singleton NETDEV scope with id 0 The NETDEV scope represents a singleton root shaper in the per-device hierarchy. All code assumes NETDEV shapers have id 0: net_shaper_default_parent() hardcodes parent->id = 0 when returning the NETDEV parent for QUEUE/NODE children, and the UAPI documentation describes NETDEV scope as "the main shaper" (singular, not plural). Make sure we reject non-0 IDs. Fixes: 4b623f9f0f59 ("net-shapers: implement NL get operation") Signed-off-by: Jakub Kicinski Link: https://patch.msgid.link/20260510192904.3987113-10-kuba@kernel.org Signed-off-by: Paolo Abeni --- net/shaper/shaper.c | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'net') diff --git a/net/shaper/shaper.c b/net/shaper/shaper.c index eb049847fed6..4ae3ee6764a0 100644 --- a/net/shaper/shaper.c +++ b/net/shaper/shaper.c @@ -482,6 +482,12 @@ static int net_shaper_parse_handle(const struct nlattr *attr, else if (handle->scope == NET_SHAPER_SCOPE_NODE) id = NET_SHAPER_ID_UNSPEC; + if (id && handle->scope == NET_SHAPER_SCOPE_NETDEV) { + NL_SET_ERR_MSG_ATTR(info->extack, id_attr, + "Netdev scope is a singleton, must use ID 0"); + return -EINVAL; + } + handle->id = id; return 0; } -- cgit v1.2.3 From ce372e869f9f492f3d5aa9a0ae75ed52c61d2d6f Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Sun, 10 May 2026 12:29:04 -0700 Subject: net: shaper: reject QUEUE scope handle with missing id net_shaper_parse_handle() does not enforce that the user provides the handle ID. For NODE the ID defaults to UNSPEC for all other cases it defaults to 0. For NETDEV 0 is the only option. For QUEUE defaulting to 0 makes less intuitive sense. Specifically because the behavior should (IMHO) be the same for all cases where there may be more than one ID (QUEUE and NODE). We should either document this as intentional or reject. I picked the latter with no strong conviction. Fixes: 4b623f9f0f59 ("net-shapers: implement NL get operation") Signed-off-by: Jakub Kicinski Link: https://patch.msgid.link/20260510192904.3987113-11-kuba@kernel.org Signed-off-by: Paolo Abeni --- net/shaper/shaper.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/shaper/shaper.c b/net/shaper/shaper.c index 4ae3ee6764a0..b1c65110f04d 100644 --- a/net/shaper/shaper.c +++ b/net/shaper/shaper.c @@ -477,10 +477,15 @@ static int net_shaper_parse_handle(const struct nlattr *attr, * shaper (any other value). */ id_attr = tb[NET_SHAPER_A_HANDLE_ID]; - if (id_attr) + if (id_attr) { id = nla_get_u32(id_attr); - else if (handle->scope == NET_SHAPER_SCOPE_NODE) + } else if (handle->scope == NET_SHAPER_SCOPE_NODE) { id = NET_SHAPER_ID_UNSPEC; + } else if (handle->scope == NET_SHAPER_SCOPE_QUEUE) { + NL_SET_ERR_ATTR_MISS(info->extack, attr, + NET_SHAPER_A_HANDLE_ID); + return -EINVAL; + } if (id && handle->scope == NET_SHAPER_SCOPE_NETDEV) { NL_SET_ERR_MSG_ATTR(info->extack, id_attr, -- cgit v1.2.3 From d289478cfc0bcf81c7914200d6abdcb78bd04ded Mon Sep 17 00:00:00 2001 From: Raphael Zimmer Date: Tue, 12 May 2026 09:29:30 +0200 Subject: libceph: handle rbtree insertion error in decode_choose_args() A message of type CEPH_MSG_OSD_MAP contains an OSD map that itself contains a CRUSH map. The received CRUSH map may optionally contain choose_args that get decoded in decode_choose_args(). In this function, num_choose_arg_maps is read from the message, and a corresponding number of crush_choose_arg_maps gets decoded afterwards. Each crush_choose_arg_map has a choose_args_index, which serves as the key when inserting it into the choose_args rbtree of the decoded crush_map. If a (potentially corrupted) message contains two crush_choose_arg_maps with the same index, the assertion in insert_choose_arg_map() triggers a kernel BUG when trying to insert the second crush_choose_arg_map. This patch fixes the issue by switching to the non-asserting rbtree insertion function and rejecting the message if the insertion fails. [ idryomov: changelog ] Cc: stable@vger.kernel.org Signed-off-by: Raphael Zimmer Reviewed-by: Ilya Dryomov Signed-off-by: Ilya Dryomov --- net/ceph/osdmap.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/ceph/osdmap.c b/net/ceph/osdmap.c index 2095e73ccf6c..0752a3808b91 100644 --- a/net/ceph/osdmap.c +++ b/net/ceph/osdmap.c @@ -392,7 +392,10 @@ static int decode_choose_args(void **p, void *end, struct crush_map *c) goto e_inval; } - insert_choose_arg_map(&c->choose_args, arg_map); + if (!__insert_choose_arg_map(&c->choose_args, arg_map)) { + ret = -EEXIST; + goto fail; + } } return 0; -- cgit v1.2.3 From 28b0a2ab8c82d0bbdeb8013029c67c978ce6e4bf Mon Sep 17 00:00:00 2001 From: Raphael Zimmer Date: Tue, 12 May 2026 18:16:40 +0200 Subject: libceph: Fix potential null-ptr-deref in decode_choose_args() A message of type CEPH_MSG_OSD_MAP contains an OSD map that itself contains a CRUSH map. When decoding this CRUSH map in crush_decode(), an array of max_buckets CRUSH buckets is decoded, where some indices may not refer to actual buckets and are therefore set to NULL. The received CRUSH map may optionally contain choose_args that get decoded in decode_choose_args(). When decoding a crush_choose_arg_map, a series of choose_args for different buckets is decoded, with the bucket_index being read from the incoming message. It is only checked that the bucket index does not exceed max_buckets, but not that it doesn't point to an index with a NULL bucket. If a (potentially corrupted) message contains a crush_choose_arg_map including such a bucket_index, a null pointer dereference may occur in the subsequent processing when attempting to access the bucket with the given index. This patch fixes the issue by extending the affected check. Now, it is only attempted to access the bucket if it is not NULL. Cc: stable@vger.kernel.org Signed-off-by: Raphael Zimmer Reviewed-by: Ilya Dryomov Signed-off-by: Ilya Dryomov --- net/ceph/osdmap.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/ceph/osdmap.c b/net/ceph/osdmap.c index 0752a3808b91..8b5b0587a0cf 100644 --- a/net/ceph/osdmap.c +++ b/net/ceph/osdmap.c @@ -388,7 +388,8 @@ static int decode_choose_args(void **p, void *end, struct crush_map *c) goto fail; if (arg->ids_size && - arg->ids_size != c->buckets[bucket_index]->size) + (!c->buckets[bucket_index] || + arg->ids_size != c->buckets[bucket_index]->size)) goto e_inval; } -- cgit v1.2.3 From e3adf69f8eb121a9128c2b0029efd050d3649153 Mon Sep 17 00:00:00 2001 From: David Carlier Date: Sat, 9 May 2026 22:50:46 +0100 Subject: net: ethtool: phy: avoid NULL deref when PHY driver is unbound phydev->drv can become NULL while the phy_device is still attached to its net_device, namely after the PHY driver is unbound via sysfs: echo > /sys/bus/mdio_bus/drivers//unbind phy_remove() clears phydev->drv but doesn't call phy_detach(), so the phy_device stays in the link topology xarray and ethnl_req_get_phydev() still hands it back. ETHTOOL_MSG_PHY_GET then oopses on: rep_data->drvname = kstrdup(phydev->drv->name, GFP_KERNEL); drvname is already treated as optional by phy_reply_size(), phy_fill_reply() and phy_cleanup_data(), so just skip the allocation when there is no driver bound. Fixes: 9dd2ad5e92b9 ("net: ethtool: phy: Convert the PHY_GET command to generic phy dump") Cc: stable@vger.kernel.org # 6.13.x Signed-off-by: David Carlier Reviewed-by: Maxime Chevallier Tested-by: Maxime Chevallier Link: https://patch.msgid.link/20260509215046.107157-1-devnexen@gmail.com Signed-off-by: Jakub Kicinski --- net/ethtool/phy.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/ethtool/phy.c b/net/ethtool/phy.c index f76d94d848d6..ddc6eab701ed 100644 --- a/net/ethtool/phy.c +++ b/net/ethtool/phy.c @@ -94,10 +94,12 @@ static int phy_prepare_data(const struct ethnl_req_info *req_info, if (!rep_data->name) return -ENOMEM; - rep_data->drvname = kstrdup(phydev->drv->name, GFP_KERNEL); - if (!rep_data->drvname) { - ret = -ENOMEM; - goto err_free_name; + if (phydev->drv) { + rep_data->drvname = kstrdup(phydev->drv->name, GFP_KERNEL); + if (!rep_data->drvname) { + ret = -ENOMEM; + goto err_free_name; + } } rep_data->upstream_type = pdn->upstream_type; -- cgit v1.2.3 From f9e2342046ef1560d35bcd4a4b1197648ffd151d Mon Sep 17 00:00:00 2001 From: Wei Yang Date: Sat, 9 May 2026 20:23:58 +0800 Subject: net: atm: fix skb leak in sigd_send() default branch The default branch in sigd_send() calls sock_put() and returns -EINVAL without freeing the skb, while all other exit paths do so. Add the missing dev_kfree_skb() before sock_put() to fix the leak. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: Wei Yang Link: https://patch.msgid.link/20260509122358.1102997-1-albin_yang@163.com Signed-off-by: Jakub Kicinski --- net/atm/signaling.c | 1 + 1 file changed, 1 insertion(+) (limited to 'net') diff --git a/net/atm/signaling.c b/net/atm/signaling.c index 358fbe5e4d1d..b991d937205a 100644 --- a/net/atm/signaling.c +++ b/net/atm/signaling.c @@ -179,6 +179,7 @@ as_indicate_complete: break; default: pr_alert("bad message type %d\n", (int)msg->type); + dev_kfree_skb(skb); /* Paired with find_get_vcc(msg->vcc) above */ sock_put(sk); return -EINVAL; -- cgit v1.2.3 From a3fdd924d88c30b9f488636ce0e4696012cf5511 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Nicol=C3=B2=20Coccia?= Date: Sun, 10 May 2026 12:34:13 -0400 Subject: net/smc: fix sleep-inside-lock in __smc_setsockopt() causing local DoS MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit A logic flaw in __smc_setsockopt() allows a local unprivileged user to cause a Denial of Service (DoS) by holding the socket lock indefinitely. The function __smc_setsockopt() calls copy_from_sockptr() while holding lock_sock(sk). By passing a userfaultfd-monitored memory page (or FUSE-backed memory on systems where unprivileged userfaultfd is disabled) as the optval, an attacker can halt execution during the copy operation, keeping the lock held. Combined with asynchronous tear-down operations like shutdown(), this exhausts the kernel wq (kworkers) and triggers the hung task watchdog. [ 240.123456] INFO: task kworker/u8:2 blocked for more than 120 seconds. [ 240.123489] Call Trace: [ 240.123501] smc_shutdown+... [ 240.123512] lock_sock_nested+... This patch moves the user-space copy outside the lock_sock() critical section to prevent the issue. Fixes: a6a6fe27bab4 ("net/smc: Dynamic control handshake limitation by socket options") Signed-off-by: Nicolò Coccia Reviewed-by: Dust Li Tested-by: Dust Li Signed-off-by: Jakub Kicinski --- net/smc/af_smc.c | 17 ++++++++--------- 1 file changed, 8 insertions(+), 9 deletions(-) (limited to 'net') diff --git a/net/smc/af_smc.c b/net/smc/af_smc.c index 185dbed7de5d..da28652f6810 100644 --- a/net/smc/af_smc.c +++ b/net/smc/af_smc.c @@ -3054,18 +3054,17 @@ static int __smc_setsockopt(struct socket *sock, int level, int optname, smc = smc_sk(sk); + /* pre-fetch user data outside the lock */ + if (optname == SMC_LIMIT_HS) { + if (optlen < sizeof(int)) + return -EINVAL; + if (copy_from_sockptr(&val, optval, sizeof(int))) + return -EFAULT; + } + lock_sock(sk); switch (optname) { case SMC_LIMIT_HS: - if (optlen < sizeof(int)) { - rc = -EINVAL; - break; - } - if (copy_from_sockptr(&val, optval, sizeof(int))) { - rc = -EFAULT; - break; - } - smc->limit_smc_hs = !!val; rc = 0; break; -- cgit v1.2.3 From 7bf563badd37cb796df5477d2b78bb64148a1268 Mon Sep 17 00:00:00 2001 From: Xiang Mei Date: Sun, 10 May 2026 15:26:40 -0700 Subject: net/smc: avoid NULL deref of conn->lnk in smc_msg_event tracepoint The smc_msg_event tracepoint class, shared by smc_tx_sendmsg and smc_rx_recvmsg, unconditionally dereferences smc->conn.lnk: __string(name, smc->conn.lnk->ibname) conn->lnk is only set for SMC-R; for SMC-D it is NULL. Other code on these paths already handles this (e.g. !conn->lnk in SMC_STAT_RMB_TX_SIZE_SMALL()). With the tracepoint enabled, the first sendmsg()/recvmsg() on an SMC-D socket crashes: Oops: general protection fault, probably for non-canonical address KASAN: null-ptr-deref in range [...] RIP: 0010:strlen+0x1e/0xa0 Call Trace: trace_event_raw_event_smc_msg_event (net/smc/smc_tracepoint.h:44) smc_rx_recvmsg (net/smc/smc_rx.c:515) smc_recvmsg (net/smc/af_smc.c:2859) __sys_recvfrom (net/socket.c:2315) __x64_sys_recvfrom (net/socket.c:2326) do_syscall_64 The faulting address 0x3e0 is offsetof(struct smc_link, ibname), confirming the NULL ->lnk deref. Enabling the tracepoint requires root, but the trigger itself is unprivileged: socket(AF_SMC, ...) has no capability check, and SMC-D negotiation needs no admin step on s390 or on x86 with the loopback ISM device loaded. Log an empty device name for SMC-D instead of dereferencing NULL. Fixes: aff3083f10bf ("net/smc: Introduce tracepoints for tx and rx msg") Reported-by: Weiming Shi Signed-off-by: Xiang Mei Reviewed-by: Dust Li Reviewed-by: Sidraya Jayagond Signed-off-by: Jakub Kicinski --- net/smc/smc_tracepoint.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/smc/smc_tracepoint.h b/net/smc/smc_tracepoint.h index a9a6e3c1113a..53da84f57fd6 100644 --- a/net/smc/smc_tracepoint.h +++ b/net/smc/smc_tracepoint.h @@ -51,7 +51,7 @@ DECLARE_EVENT_CLASS(smc_msg_event, __field(const void *, smc) __field(u64, net_cookie) __field(size_t, len) - __string(name, smc->conn.lnk->ibname) + __string(name, smc->conn.lnk ? smc->conn.lnk->ibname : "") ), TP_fast_assign( -- cgit v1.2.3 From 3d042592ebd4c7e44974d556de0b727cb7db4dab Mon Sep 17 00:00:00 2001 From: Chenguang Zhao Date: Mon, 11 May 2026 09:43:43 +0800 Subject: ethtool: fix ethnl_bitmap32_not_zero() bit interval semantics ethnl_bitmap32_not_zero() should return true if some bit in [start, end) is set: - Fix inverted memchr_inv() sense: return true when the scan finds a non-zero byte, not when the middle words are all zero. - Return false for an empty interval (end <= start). - When end is 32-bit aligned, indices in [start, end) do not include any bits from map[end_word]; return false after earlier checks found no non-zero data. Fixes: 10b518d4e6dd ("ethtool: netlink bitset handling") Signed-off-by: Chenguang Zhao Signed-off-by: Jakub Kicinski --- net/ethtool/bitset.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/ethtool/bitset.c b/net/ethtool/bitset.c index 8bb98d3ea3db..a3a2cc6480a0 100644 --- a/net/ethtool/bitset.c +++ b/net/ethtool/bitset.c @@ -92,7 +92,7 @@ static bool ethnl_bitmap32_not_zero(const u32 *map, unsigned int start, u32 mask; if (end <= start) - return true; + return false; if (start % 32) { mask = ethnl_upper_bits(start); @@ -105,11 +105,11 @@ static bool ethnl_bitmap32_not_zero(const u32 *map, unsigned int start, start_word++; } - if (!memchr_inv(map + start_word, '\0', - (end_word - start_word) * sizeof(u32))) + if (memchr_inv(map + start_word, '\0', + (end_word - start_word) * sizeof(u32))) return true; if (end % 32 == 0) - return true; + return false; return map[end_word] & ethnl_lower_bits(end); } -- cgit v1.2.3 From 320fb29ea23cfa1aeef32563da8748247db896ea Mon Sep 17 00:00:00 2001 From: Jamal Hadi Salim Date: Mon, 11 May 2026 14:30:57 -0400 Subject: net/sched: sch_cbs: Call qdisc_reset for child qdisc During a reset, CBS is not calling reset on its child qdisc, which might cause qlen/backlog accounting issues. For example, if we have CBS with a QFQ parent and a netem child with delay, we can create a scenario where the parent's qlen underflows. QFQ, specifically, uses qlen to check whether it should deference a pointer, so this scenario may cause a null-ptr deref in QFQ: [ 43.875639][ T319] Oops: general protection fault, probably for non-canonical address 0xdffffc0000000009: 0000 [#1] SMP KASAN NOPTI [ 43.876124][ T319] KASAN: null-ptr-deref in range [0x0000000000000048-0x000000000000004f] [ 43.876417][ T319] CPU: 10 UID: 0 PID: 319 Comm: ping Not tainted 7.0.0-13039-ge728258debd5 #773 PREEMPT(full) [ 43.876751][ T319] Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011 [ 43.876949][ T319] RIP: 0010:qfq_dequeue+0x35c/0x1650 [ 43.877123][ T319] Code: 00 fc ff df 80 3c 02 00 0f 85 17 0e 00 00 4c 8d 73 48 48 89 9d b8 02 00 00 48 b8 00 00 00 00 00 fc ff df 4c 89 f2 48 c1 ea 03 <80> 3c 02 00 0f 85 76 0c 00 00 48 b8 00 00 00 00 00 fc ff df 4c 8b [ 43.877648][ T319] RSP: 0018:ffff8881017ef4f0 EFLAGS: 00010216 [ 43.877845][ T319] RAX: dffffc0000000000 RBX: 0000000000000000 RCX: dffffc0000000000 [ 43.878073][ T319] RDX: 0000000000000009 RSI: 0000000c40000000 RDI: ffff88810eef02b0 [ 43.878306][ T319] RBP: ffff88810eef0000 R08: ffff88810eef0280 R09: 1ffff1102120fd63 [ 43.878523][ T319] R10: 1ffff1102120fd66 R11: 1ffff1102120fd67 R12: 0000000c40000000 [ 43.878742][ T319] R13: ffff88810eef02b8 R14: 0000000000000048 R15: 0000000020000000 [ 43.878959][ T319] FS: 00007f9c51c47c40(0000) GS:ffff88817a0be000(0000) knlGS:0000000000000000 [ 43.879214][ T319] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 43.879403][ T319] CR2: 000055e69a2230a8 CR3: 000000010c07a000 CR4: 0000000000750ef0 [ 43.879621][ T319] PKRU: 55555554 [ 43.879735][ T319] Call Trace: [ 43.879844][ T319] [ 43.879924][ T319] __qdisc_run+0x169/0x1900 [ 43.880075][ T319] ? dev_qdisc_enqueue+0x8b/0x210 [ 43.880222][ T319] __dev_queue_xmit+0x2346/0x37a0 [ 43.880376][ T319] ? register_lock_class+0x3f/0x800 [ 43.880531][ T319] ? srso_alias_return_thunk+0x5/0xfbef5 [ 43.880684][ T319] ? __pfx___dev_queue_xmit+0x10/0x10 [ 43.880834][ T319] ? srso_alias_return_thunk+0x5/0xfbef5 [ 43.880977][ T319] ? __lock_acquire+0x819/0x1df0 [ 43.881124][ T319] ? srso_alias_return_thunk+0x5/0xfbef5 [ 43.881275][ T319] ? srso_alias_return_thunk+0x5/0xfbef5 [ 43.881418][ T319] ? __asan_memcpy+0x3c/0x60 [ 43.881563][ T319] ? srso_alias_return_thunk+0x5/0xfbef5 [ 43.881708][ T319] ? eth_header+0x165/0x1a0 [ 43.881853][ T319] ? lockdep_hardirqs_on_prepare+0xdb/0x1a0 [ 43.882031][ T319] ? srso_alias_return_thunk+0x5/0xfbef5 [ 43.882174][ T319] ? neigh_resolve_output+0x3cc/0x7e0 [ 43.882325][ T319] ? srso_alias_return_thunk+0x5/0xfbef5 [ 43.882471][ T319] ip_finish_output2+0x6b6/0x1e10 Fix this by calling qdisc_reset for CBS' child qdisc. Sashiko caught an issue which could result in a null ptr deref if qdisc_create_dflt() is invoked on an unitialised cbs qdisc which is exposed by this patch. We add an early return if the qdisc is null to address this. This is a similar approach used by two other fixes[1][2]. The proper fix for this specific issue elucidated by sashiko is to remove the call to qdisc_reset when qdisc_create_dflt fails. Since the dflt qdisc isn't attached anywhere yet at that point, calling the reset callback doesn't make much sense (and as stated has been a source of two other bugs). We plan on submitting this fix in a later patch. [1] https://lore.kernel.org/netdev/20221018063201.306474-2-shaozhengchao@huawei.com/ [2] https://lore.kernel.org/netdev/20221018063201.306474-4-shaozhengchao@huawei.com/ Fixes: 585d763af09c ("net/sched: Introduce Credit Based Shaper (CBS) qdisc") Reported-by: Junyoung Jang Tested-by: Junyoung Jang Tested-by: Victor Nogueira Acked-by: Vinicius Costa Gomes Signed-off-by: Jamal Hadi Salim Signed-off-by: Jakub Kicinski --- net/sched/sch_cbs.c | 16 +++++++++++++++- 1 file changed, 15 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/sched/sch_cbs.c b/net/sched/sch_cbs.c index 8c9a0400c862..0f953bd46b58 100644 --- a/net/sched/sch_cbs.c +++ b/net/sched/sch_cbs.c @@ -243,6 +243,20 @@ static struct sk_buff *cbs_dequeue(struct Qdisc *sch) return q->dequeue(sch); } +static void cbs_reset(struct Qdisc *sch) +{ + struct cbs_sched_data *q = qdisc_priv(sch); + + /* Nothing to do if we couldn't create the underlying qdisc */ + if (!q->qdisc) + return; + + qdisc_reset(q->qdisc); + qdisc_watchdog_cancel(&q->watchdog); + q->credits = 0; + q->last = 0; +} + static const struct nla_policy cbs_policy[TCA_CBS_MAX + 1] = { [TCA_CBS_PARMS] = { .len = sizeof(struct tc_cbs_qopt) }, }; @@ -540,7 +554,7 @@ static struct Qdisc_ops cbs_qdisc_ops __read_mostly = { .dequeue = cbs_dequeue, .peek = qdisc_peek_dequeued, .init = cbs_init, - .reset = qdisc_reset_queue, + .reset = cbs_reset, .destroy = cbs_destroy, .change = cbs_change, .dump = cbs_dump, -- cgit v1.2.3 From b84c5632c7b31f8910167075a8128cfb9e50fcfe Mon Sep 17 00:00:00 2001 From: Faicker Mo Date: Mon, 11 May 2026 22:05:51 +0800 Subject: net: net_failover: Fix the deadlock in slave register There is netdev_lock_ops() before the NETDEV_REGISTER notifier in register_netdevice(), so use the non-locking functions in net_failover_slave_register(). failover_slave_register() in failover_existing_slave_register() adds lock and unlock ops too. Call Trace: __schedule+0x30d/0x7a0 schedule+0x27/0x90 schedule_preempt_disabled+0x15/0x30 __mutex_lock.constprop.0+0x538/0x9e0 __mutex_lock_slowpath+0x13/0x20 mutex_lock+0x3b/0x50 dev_set_mtu+0x40/0xe0 net_failover_slave_register+0x24/0x280 failover_slave_register+0x103/0x1b0 failover_event+0x15e/0x210 ? dropmon_net_event+0xac/0xe0 notifier_call_chain+0x5e/0xe0 raw_notifier_call_chain+0x16/0x30 call_netdevice_notifiers_info+0x52/0xa0 register_netdevice+0x5f4/0x7c0 register_netdev+0x1e/0x40 _mlx5e_probe+0xe2/0x370 [mlx5_core] mlx5e_probe+0x59/0x70 [mlx5_core] ? __pfx_mlx5e_probe+0x10/0x10 [mlx5_core] Fixes: 4c975fd70002 ("net: hold instance lock during NETDEV_REGISTER/UP") Signed-off-by: Faicker Mo Signed-off-by: Jakub Kicinski --- net/core/failover.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/core/failover.c b/net/core/failover.c index 11bb183c7a1b..e43c59cd6868 100644 --- a/net/core/failover.c +++ b/net/core/failover.c @@ -12,6 +12,7 @@ #include #include #include +#include #include static LIST_HEAD(failover_list); @@ -221,8 +222,11 @@ failover_existing_slave_register(struct net_device *failover_dev) for_each_netdev(net, dev) { if (netif_is_failover(dev)) continue; - if (ether_addr_equal(failover_dev->perm_addr, dev->perm_addr)) + if (ether_addr_equal(failover_dev->perm_addr, dev->perm_addr)) { + netdev_lock_ops(dev); failover_slave_register(dev); + netdev_unlock_ops(dev); + } } rtnl_unlock(); } -- cgit v1.2.3 From 742b04d0550b0ec89dcbc99537ec88653bd1ad90 Mon Sep 17 00:00:00 2001 From: David Ahern Date: Wed, 13 May 2026 10:49:14 -0600 Subject: xfrm: Check for underflow in xfrm_state_mtu Leo Lin reported OOB write issue in esp component: xfrm_state_mtu() returns u32 but performs its arithmetic in unsigned modulo-2^32 space using an attacker-influenced "header_len + authsize + net_adj" subtracted from a small "mtu" argument. A nobody user can install an IPv4 ESP tunnel SA with a large authentication key (XFRMA_ALG_AUTH_TRUNC, e.g. hmac(sha512), 64-byte key, 64-byte trunc), configure a small interface MTU (68 bytes), and set XFRMA_TFCPAD to a large value. When a single UDP datagram is then sent through the tunnel, xfrm_state_mtu() underflows to a near-2^32 value, and esp_output() consumes it as a signed int via: padto = min(x->tfcpad, xfrm_state_mtu(x, mtu_cached)) esp.tfclen = padto - skb->len (assigned to int) esp.tfclen ends up negative (e.g. -207). It is sign-extended to size_t when passed to memset() inside esp_output_fill_trailer(), producing a ~16 EB write of zeroes at skb_tail_pointer(skb). KASAN logs it as "Write of size 18446744073709551537 at addr ffff888...". Check for underflow and return 1. This causes the sendmsg attempt to fail with ENETUNREACH. Fixes: c5c252389374 ("[XFRM]: Optimize MTU calculation") Reported-by: Leo Lin Assisted-by: Codex:26.506.31004 Signed-off-by: David Ahern Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_state.c | 19 ++++++++++++++++--- 1 file changed, 16 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c index 395d82411a87..589c3b6e4679 100644 --- a/net/xfrm/xfrm_state.c +++ b/net/xfrm/xfrm_state.c @@ -3114,10 +3114,14 @@ u32 xfrm_state_mtu(struct xfrm_state *x, int mtu) const struct xfrm_type *type = READ_ONCE(x->type); struct crypto_aead *aead; u32 blksize, net_adj = 0; + u32 overhead, payload_mtu; if (x->km.state != XFRM_STATE_VALID || - !type || type->proto != IPPROTO_ESP) + !type || type->proto != IPPROTO_ESP) { + if (mtu <= x->props.header_len) + return 1; return mtu - x->props.header_len; + } aead = x->data; blksize = ALIGN(crypto_aead_blocksize(aead), 4); @@ -3140,8 +3144,17 @@ u32 xfrm_state_mtu(struct xfrm_state *x, int mtu) break; } - return ((mtu - x->props.header_len - crypto_aead_authsize(aead) - - net_adj) & ~(blksize - 1)) + net_adj - 2; + overhead = x->props.header_len + crypto_aead_authsize(aead) + net_adj; + if (mtu <= overhead) + return 1; + + payload_mtu = mtu - overhead; + payload_mtu &= ~(blksize - 1); + if (payload_mtu <= 2) + return 1; + + return payload_mtu + net_adj - 2; + } EXPORT_SYMBOL_GPL(xfrm_state_mtu); -- cgit v1.2.3 From 277740023def559a4a2ddc3e8e784ee37a0f16a9 Mon Sep 17 00:00:00 2001 From: Xiang Mei Date: Sun, 10 May 2026 23:21:38 -0700 Subject: net/smc: reject CHID-0 ACCEPT that matches an empty ism_dev slot On the SMC-D client, slot 0 of ini->ism_dev[]/ini->ism_chid[] is reserved for an SMC-Dv1 device. smc_find_ism_v2_device_clnt() populates V2 entries starting at index 1, so when no V1 device is selected slot 0 is left in its kzalloc()'ed state with ism_dev[0] == NULL and ism_chid[0] == 0. smc_v2_determine_accepted_chid() then matches the peer's CHID against the array starting from index 0 using the CHID alone. A malicious peer replying to a SMC-Dv2-only proposal with d1.chid == 0 matches the empty slot, ini->ism_selected becomes 0, and the subsequent ism_dev[0]->lgr_lock dereference in smc_conn_create() faults at offsetof(struct smcd_dev, lgr_lock) == 0x68: BUG: KASAN: null-ptr-deref in _raw_spin_lock_bh+0x79/0xe0 Write of size 4 at addr 0000000000000068 by task exploit/144 Call Trace: _raw_spin_lock_bh smc_conn_create (net/smc/smc_core.c:1997) __smc_connect (net/smc/af_smc.c:1447) smc_connect (net/smc/af_smc.c:1720) __sys_connect __x64_sys_connect do_syscall_64 Require ism_dev[i] to be non-NULL before accepting a CHID match. Fixes: a7c9c5f4af7f ("net/smc: CLC accept / confirm V2") Reported-by: Weiming Shi Assisted-by: Claude:claude-opus-4-7 Signed-off-by: Xiang Mei Link: https://patch.msgid.link/20260511062138.2839584-1-xmei5@asu.edu Signed-off-by: Paolo Abeni --- net/smc/af_smc.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/smc/af_smc.c b/net/smc/af_smc.c index da28652f6810..dffbd529762d 100644 --- a/net/smc/af_smc.c +++ b/net/smc/af_smc.c @@ -1400,7 +1400,8 @@ smc_v2_determine_accepted_chid(struct smc_clc_msg_accept_confirm *aclc, int i; for (i = 0; i < ini->ism_offered_cnt + 1; i++) { - if (ini->ism_chid[i] == ntohs(aclc->d1.chid)) { + if (ini->ism_dev[i] && + ini->ism_chid[i] == ntohs(aclc->d1.chid)) { ini->ism_selected = i; return 0; } -- cgit v1.2.3 From 285943c6e7ca309bbea84b253745154241d9788a Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Mon, 11 May 2026 10:49:17 -0700 Subject: net: tls: fix off-by-one in sg_chain entry count for wrapped sk_msg ring MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When an sk_msg scatterlist ring wraps (sg.end < sg.start), tls_push_record() chains the tail portion of the ring to the head using sg_chain(). An extra entry in the sg array is reserved for this: struct sk_msg_sg { [...] /* The extra two elements: * 1) used for chaining the front and sections when the list becomes * partitioned (e.g. end < start). The crypto APIs require the * chaining; * 2) to chain tailer SG entries after the message. */ struct scatterlist data[MAX_MSG_FRAGS + 2]; The current code uses MAX_SKB_FRAGS + 1 as the ring size: sg_chain(&msg_pl->sg.data[msg_pl->sg.start], MAX_SKB_FRAGS - msg_pl->sg.start + 1, msg_pl->sg.data); This places the chain pointer at sg_chain(data[start], (MAX_SKB_FRAGS - msg_start + 1) .. = &data[start] + (MAX_SKB_FRAGS - msg_start + 1) - 1 = data[start + (MAX_SKB_FRAGS - start + 1) - 1] = data[MAX_SKB_FRAGS] instead of the true last entry. This is likely due to a "race" of the commit under Fixes landing close to commit 031097d9e079 ("bpf: sk_msg, zap ingress queue on psock down") Convert to ARRAY_SIZE and drop the data[start] / - start (as suggested by Sabrina). Reported-by: 钱一铭 Fixes: 9aaaa56845a0 ("bpf: Sockmap/tls, skmsg can have wrapped skmsg that needs extra chaining") Signed-off-by: Jakub Kicinski Reviewed-by: Sabrina Dubroca Link: https://patch.msgid.link/20260511174920.433155-2-kuba@kernel.org Signed-off-by: Paolo Abeni --- net/tls/tls_sw.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/tls/tls_sw.c b/net/tls/tls_sw.c index 2590e855f6a5..2608b0c01849 100644 --- a/net/tls/tls_sw.c +++ b/net/tls/tls_sw.c @@ -800,11 +800,9 @@ static int tls_push_record(struct sock *sk, int flags, sg_mark_end(sk_msg_elem(msg_pl, i)); } - if (msg_pl->sg.end < msg_pl->sg.start) { - sg_chain(&msg_pl->sg.data[msg_pl->sg.start], - MAX_SKB_FRAGS - msg_pl->sg.start + 1, + if (msg_pl->sg.end < msg_pl->sg.start) + sg_chain(msg_pl->sg.data, ARRAY_SIZE(msg_pl->sg.data), msg_pl->sg.data); - } i = msg_pl->sg.start; sg_chain(rec->sg_aead_in, 2, &msg_pl->sg.data[i]); -- cgit v1.2.3 From ff26a0e8377dec07e4a7230db7675bed1b9a6d03 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Mon, 11 May 2026 10:49:18 -0700 Subject: net: tls: prevent chain-after-chain in plain text SG Sashiko points out that if end = 0 (start != 0) the current code will create a chain link to content type right after the wrap link: This would create a chain where the wrap link points directly to another chain link. The scatterlist API sg_next iterator does not recursively resolve consecutive chain links. meaning this is illegal input to crypto. The wrapping link is unnecessary if end = 0. end is the entry after the last one used so end = 0 means there's nothing pushed after the wrap: end start i v v v [ ]...[ ][ d ][ d ][ d ][ d ][rsv for wrap] Skip the wrapping in this case. TLS 1.3 can use the "wrapping slot" for it's chaining if end = 0. This avoids the chain-after-chain. Move the wrap chaining before marking END and chaining off content type, that feels like more logical ordering to me, but should not matter from functional perspective. Reported-by: Sashiko Fixes: 9aaaa56845a0 ("bpf: Sockmap/tls, skmsg can have wrapped skmsg that needs extra chaining") Signed-off-by: Jakub Kicinski Link: https://patch.msgid.link/20260511174920.433155-3-kuba@kernel.org Signed-off-by: Paolo Abeni --- net/tls/tls_sw.c | 24 ++++++++++++++++++------ 1 file changed, 18 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/net/tls/tls_sw.c b/net/tls/tls_sw.c index 2608b0c01849..3bfdaf5e64f5 100644 --- a/net/tls/tls_sw.c +++ b/net/tls/tls_sw.c @@ -789,21 +789,33 @@ static int tls_push_record(struct sock *sk, int flags, i = msg_pl->sg.end; sk_msg_iter_var_prev(i); + /* msg_pl->sg.data is a ring; data[MAX+1] is reserved for the wrap + * link (frags won't use it). 'i' is now the last filled entry: + * + * i end start + * v v v [ rsv ] + * [ d ][ d ][ ][ ]...[ ][ d ][ d ][ d ][chain] + * ^ END v + * `-----------------------------------------' + * + * Note that SGL does not allow chain-after-chain, so for TLS 1.3, + * we must make sure we don't create the wrap entry and then chain + * link to content_type immediately at index 0. + */ + if (i < msg_pl->sg.start) + sg_chain(msg_pl->sg.data, ARRAY_SIZE(msg_pl->sg.data), + msg_pl->sg.data); + rec->content_type = record_type; if (prot->version == TLS_1_3_VERSION) { /* Add content type to end of message. No padding added */ sg_set_buf(&rec->sg_content_type, &rec->content_type, 1); sg_mark_end(&rec->sg_content_type); - sg_chain(msg_pl->sg.data, msg_pl->sg.end + 1, - &rec->sg_content_type); + sg_chain(msg_pl->sg.data, i + 2, &rec->sg_content_type); } else { sg_mark_end(sk_msg_elem(msg_pl, i)); } - if (msg_pl->sg.end < msg_pl->sg.start) - sg_chain(msg_pl->sg.data, ARRAY_SIZE(msg_pl->sg.data), - msg_pl->sg.data); - i = msg_pl->sg.start; sg_chain(rec->sg_aead_in, 2, &msg_pl->sg.data[i]); -- cgit v1.2.3 From e83f5e24da741fa9405aeeff00b08c5ee7c37b88 Mon Sep 17 00:00:00 2001 From: Jiexun Wang Date: Wed, 6 May 2026 19:43:30 +0800 Subject: Bluetooth: serialize accept_q access bt_sock_poll() walks the accept queue without synchronization, while child teardown can unlink the same socket and drop its last reference. The unsynchronized accept queue walk has existed since the initial Bluetooth import. Protect accept_q with a dedicated lock for queue updates and polling. Also rework bt_accept_dequeue() to take temporary child references under the queue lock before dropping it and locking the child socket. Fixes: 1da177e4c3f41524e886b7f1b8a0c1fc7321cac2 ("Linux-2.6.12-rc2") Cc: stable@vger.kernel.org Reported-by: Jann Horn Reported-by: Yuan Tan Reported-by: Yifan Wu Reported-by: Juefei Pu Reported-by: Xin Liu Signed-off-by: Jiexun Wang Signed-off-by: Ren Wei Signed-off-by: Jiexun Wang Reviewed-by: Jann Horn Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/af_bluetooth.c | 87 +++++++++++++++++++++++++++++++++----------- 1 file changed, 65 insertions(+), 22 deletions(-) (limited to 'net') diff --git a/net/bluetooth/af_bluetooth.c b/net/bluetooth/af_bluetooth.c index 33d053d63407..9d68dd86023c 100644 --- a/net/bluetooth/af_bluetooth.c +++ b/net/bluetooth/af_bluetooth.c @@ -154,6 +154,7 @@ struct sock *bt_sock_alloc(struct net *net, struct socket *sock, sock_init_data(sock, sk); INIT_LIST_HEAD(&bt_sk(sk)->accept_q); + spin_lock_init(&bt_sk(sk)->accept_q_lock); sock_reset_flag(sk, SOCK_ZAPPED); @@ -214,6 +215,7 @@ void bt_accept_enqueue(struct sock *parent, struct sock *sk, bool bh) { const struct cred *old_cred; struct pid *old_pid; + struct bt_sock *par = bt_sk(parent); BT_DBG("parent %p, sk %p", parent, sk); @@ -224,9 +226,13 @@ void bt_accept_enqueue(struct sock *parent, struct sock *sk, bool bh) else lock_sock_nested(sk, SINGLE_DEPTH_NESTING); - list_add_tail(&bt_sk(sk)->accept_q, &bt_sk(parent)->accept_q); bt_sk(sk)->parent = parent; + spin_lock_bh(&par->accept_q_lock); + list_add_tail(&bt_sk(sk)->accept_q, &par->accept_q); + sk_acceptq_added(parent); + spin_unlock_bh(&par->accept_q_lock); + /* Copy credentials from parent since for incoming connections the * socket is allocated by the kernel. */ @@ -244,8 +250,6 @@ void bt_accept_enqueue(struct sock *parent, struct sock *sk, bool bh) bh_unlock_sock(sk); else release_sock(sk); - - sk_acceptq_added(parent); } EXPORT_SYMBOL(bt_accept_enqueue); @@ -254,45 +258,72 @@ EXPORT_SYMBOL(bt_accept_enqueue); */ void bt_accept_unlink(struct sock *sk) { + struct sock *parent = bt_sk(sk)->parent; + BT_DBG("sk %p state %d", sk, sk->sk_state); + spin_lock_bh(&bt_sk(parent)->accept_q_lock); list_del_init(&bt_sk(sk)->accept_q); - sk_acceptq_removed(bt_sk(sk)->parent); + sk_acceptq_removed(parent); + spin_unlock_bh(&bt_sk(parent)->accept_q_lock); bt_sk(sk)->parent = NULL; sock_put(sk); } EXPORT_SYMBOL(bt_accept_unlink); +static struct sock *bt_accept_get(struct sock *parent, struct sock *sk) +{ + struct bt_sock *bt = bt_sk(parent); + struct sock *next = NULL; + + /* accept_q is modified from child teardown paths too, so take a + * temporary reference before dropping the queue lock. + */ + spin_lock_bh(&bt->accept_q_lock); + + if (sk) { + if (bt_sk(sk)->parent != parent) + goto out; + + if (!list_is_last(&bt_sk(sk)->accept_q, &bt->accept_q)) { + next = &list_next_entry(bt_sk(sk), accept_q)->sk; + sock_hold(next); + } + } else if (!list_empty(&bt->accept_q)) { + next = &list_first_entry(&bt->accept_q, + struct bt_sock, accept_q)->sk; + sock_hold(next); + } + +out: + spin_unlock_bh(&bt->accept_q_lock); + return next; +} + struct sock *bt_accept_dequeue(struct sock *parent, struct socket *newsock) { - struct bt_sock *s, *n; - struct sock *sk; + struct sock *sk, *next; BT_DBG("parent %p", parent); restart: - list_for_each_entry_safe(s, n, &bt_sk(parent)->accept_q, accept_q) { - sk = (struct sock *)s; - + for (sk = bt_accept_get(parent, NULL); sk; sk = next) { /* Prevent early freeing of sk due to unlink and sock_kill */ - sock_hold(sk); lock_sock(sk); /* Check sk has not already been unlinked via * bt_accept_unlink() due to serialisation caused by sk locking */ - if (!bt_sk(sk)->parent) { + if (bt_sk(sk)->parent != parent) { BT_DBG("sk %p, already unlinked", sk); release_sock(sk); sock_put(sk); - /* Restart the loop as sk is no longer in the list - * and also avoid a potential infinite loop because - * list_for_each_entry_safe() is not thread safe. - */ goto restart; } + next = bt_accept_get(parent, sk); + /* sk is safely in the parent list so reduce reference count */ sock_put(sk); @@ -310,6 +341,8 @@ restart: sock_graft(sk, newsock); release_sock(sk); + if (next) + sock_put(next); return sk; } @@ -518,18 +551,28 @@ EXPORT_SYMBOL(bt_sock_stream_recvmsg); static inline __poll_t bt_accept_poll(struct sock *parent) { - struct bt_sock *s, *n; + struct bt_sock *bt = bt_sk(parent); + struct bt_sock *s; struct sock *sk; + __poll_t mask = 0; + + spin_lock_bh(&bt->accept_q_lock); + list_for_each_entry(s, &bt->accept_q, accept_q) { + int state; - list_for_each_entry_safe(s, n, &bt_sk(parent)->accept_q, accept_q) { sk = (struct sock *)s; - if (sk->sk_state == BT_CONNECTED || - (test_bit(BT_SK_DEFER_SETUP, &bt_sk(parent)->flags) && - sk->sk_state == BT_CONNECT2)) - return EPOLLIN | EPOLLRDNORM; + state = READ_ONCE(sk->sk_state); + + if (state == BT_CONNECTED || + (test_bit(BT_SK_DEFER_SETUP, &bt->flags) && + state == BT_CONNECT2)) { + mask = EPOLLIN | EPOLLRDNORM; + break; + } } + spin_unlock_bh(&bt->accept_q_lock); - return 0; + return mask; } __poll_t bt_sock_poll(struct file *file, struct socket *sock, -- cgit v1.2.3 From 3374ef8cf99368a40f7efd51a2a375a4c5dc6f0d Mon Sep 17 00:00:00 2001 From: Michael Bommarito Date: Mon, 11 May 2026 08:26:41 -0400 Subject: Bluetooth: L2CAP: ecred_reconfigure: send packed pdu, not stack pointer Commit 1c08108f3014 ("Bluetooth: L2CAP: Avoid -Wflex-array-member-not-at-end warnings") converted the on-stack request PDU in l2cap_ecred_reconfigure() from an explicit packed struct to DEFINE_RAW_FLEX(), but did not adjust the size and source-pointer arguments to l2cap_send_cmd(): - struct { - struct l2cap_ecred_reconf_req req; - __le16 scid; - } pdu; + DEFINE_RAW_FLEX(struct l2cap_ecred_reconf_req, pdu, scid, 1); ... l2cap_send_cmd(conn, chan->ident, L2CAP_ECRED_RECONF_REQ, sizeof(pdu), &pdu); After the conversion, DEFINE_RAW_FLEX() expands to declare an anonymous union pdu_u plus a local pointer "pdu" pointing at it. Therefore: - sizeof(pdu) is now sizeof(struct l2cap_ecred_reconf_req *) = 8 on 64-bit (4 on 32-bit), not the 6 bytes of (mtu, mps, scid[1]). - &pdu is the address of the local pointer's stack storage, not the address of the request payload. l2cap_send_cmd() forwards (data, count) to l2cap_build_cmd(), which calls skb_put_data(skb, data, count). The L2CAP_ECRED_RECONFIGURE_REQ packet body therefore contains 8 bytes copied from the kernel stack starting at &pdu -- the 8 bytes overlap the pdu pointer's value, leaking a kernel stack address to the paired Bluetooth peer. The intended (mtu, mps, scid) fields are not transmitted at all, so the peer rejects the request as malformed and the L2CAP_ECRED_RECONFIGURE feature itself has been broken for the local-side initiator since the introducing commit landed. The sibling site l2cap_ecred_conn_req() in the same commit was converted correctly (sizeof(*pdu) + len, pdu); only this site was missed. Restore the original semantics: pass the full flex-struct size via struct_size(pdu, scid, 1) and the pdu pointer (the struct address) as the source. Validated on a stock 7.0-based host kernel via the real call path: setsockopt(SOL_BLUETOOTH, BT_RCVMTU, ...) on a BT_CONNECTED L2CAP_MODE_EXT_FLOWCTL socket emits an L2CAP_ECRED_RECONFIGURE_REQ whose body is 8 bytes (the on-stack pdu local's value) rather than the expected 6. Three captures from fresh socket / fresh hciemu peer on the same host -- low bytes vary per call, high 0xffff confirms a kernel virtual address (KASLR-randomised stack slot, not a fixed string): RECONF_REQ body (ident=0x02 len=8): 42 fb 54 af 0e ca ff ff RECONF_REQ body (ident=0x02 len=8): 52 3d 2e af 0e ca ff ff RECONF_REQ body (ident=0x02 len=8): b2 fc 5b af 0e ca ff ff After this patch the body is 6 bytes carrying the expected little-endian (mtu, mps, scid). Cc: stable@vger.kernel.org Fixes: 1c08108f3014 ("Bluetooth: L2CAP: Avoid -Wflex-array-member-not-at-end warnings") Assisted-by: Claude:claude-opus-4-7 Signed-off-by: Michael Bommarito Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/l2cap_core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c index 7701528f1167..fdccd62ccca8 100644 --- a/net/bluetooth/l2cap_core.c +++ b/net/bluetooth/l2cap_core.c @@ -7274,7 +7274,7 @@ static void l2cap_ecred_reconfigure(struct l2cap_chan *chan) chan->ident = l2cap_get_ident(conn); l2cap_send_cmd(conn, chan->ident, L2CAP_ECRED_RECONF_REQ, - sizeof(pdu), &pdu); + struct_size(pdu, scid, 1), pdu); } int l2cap_chan_reconfigure(struct l2cap_chan *chan, __u16 mtu) -- cgit v1.2.3 From 9cd3f16c320bfdadd4509358122368deb56a5741 Mon Sep 17 00:00:00 2001 From: Ruide Cao Date: Wed, 13 May 2026 11:58:15 +0800 Subject: batman-adv: fix fragment reassembly length accounting batman-adv keeps a running payload length for queued fragments and uses it to validate a fragment chain before reassembly. That accounting currently allows the accumulated fragment length to be truncated during updates. As a result, malformed fragment chains can bypass the intended validation and drive reassembly with inconsistent length state, leading to a local denial of service. Fix the accounting by storing the accumulated length in a length-typed field and rejecting update overflows before the existing validation logic runs. The fix was verified against the original reproducer and against valid fragment reassembly paths. Fixes: 610bfc6bc99b ("batman-adv: Receive fragmented packets and merge") Cc: stable@kernel.org Reported-by: Yuan Tan Reported-by: Yifan Wu Reported-by: Juefei Pu Reported-by: Xin Liu Signed-off-by: Ruide Cao Tested-by: Ren Wei Signed-off-by: Ren Wei Signed-off-by: Sven Eckelmann --- net/batman-adv/fragmentation.c | 23 +++++++++++++++++------ net/batman-adv/types.h | 2 +- 2 files changed, 18 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/net/batman-adv/fragmentation.c b/net/batman-adv/fragmentation.c index f4e45cc25816..1152c2ce0c1e 100644 --- a/net/batman-adv/fragmentation.c +++ b/net/batman-adv/fragmentation.c @@ -17,6 +17,7 @@ #include #include #include +#include #include #include #include @@ -80,9 +81,9 @@ void batadv_frag_purge_orig(struct batadv_orig_node *orig_node, * * Return: the maximum size of payload that can be fragmented. */ -static int batadv_frag_size_limit(void) +static size_t batadv_frag_size_limit(void) { - int limit = BATADV_FRAG_MAX_FRAG_SIZE; + size_t limit = BATADV_FRAG_MAX_FRAG_SIZE; limit -= sizeof(struct batadv_frag_packet); limit *= BATADV_FRAG_MAX_FRAGMENTS; @@ -143,7 +144,9 @@ static bool batadv_frag_insert_packet(struct batadv_orig_node *orig_node, struct batadv_frag_packet *frag_packet; u8 bucket; u16 seqno, hdr_size = sizeof(struct batadv_frag_packet); + bool overflow = false; bool ret = false; + size_t data_len; /* Linearize packet to avoid linearizing 16 packets in a row when doing * the later merge. Non-linear merge should be added to remove this @@ -153,6 +156,7 @@ static bool batadv_frag_insert_packet(struct batadv_orig_node *orig_node, goto err; frag_packet = (struct batadv_frag_packet *)skb->data; + data_len = skb->len - hdr_size; seqno = ntohs(frag_packet->seqno); bucket = seqno % BATADV_FRAG_BUFFER_COUNT; @@ -171,7 +175,7 @@ static bool batadv_frag_insert_packet(struct batadv_orig_node *orig_node, spin_lock_bh(&chain->lock); if (batadv_frag_init_chain(chain, seqno)) { hlist_add_head(&frag_entry_new->list, &chain->fragment_list); - chain->size = skb->len - hdr_size; + chain->size = data_len; chain->timestamp = jiffies; chain->total_size = ntohs(frag_packet->total_size); ret = true; @@ -188,7 +192,11 @@ static bool batadv_frag_insert_packet(struct batadv_orig_node *orig_node, if (frag_entry_curr->no < frag_entry_new->no) { hlist_add_before(&frag_entry_new->list, &frag_entry_curr->list); - chain->size += skb->len - hdr_size; + + if (check_add_overflow(chain->size, data_len, + &chain->size)) + overflow = true; + chain->timestamp = jiffies; ret = true; goto out; @@ -201,13 +209,16 @@ static bool batadv_frag_insert_packet(struct batadv_orig_node *orig_node, /* Reached the end of the list, so insert after 'frag_entry_last'. */ if (likely(frag_entry_last)) { hlist_add_behind(&frag_entry_new->list, &frag_entry_last->list); - chain->size += skb->len - hdr_size; + + if (check_add_overflow(chain->size, data_len, &chain->size)) + overflow = true; + chain->timestamp = jiffies; ret = true; } out: - if (chain->size > batadv_frag_size_limit() || + if (overflow || chain->size > batadv_frag_size_limit() || chain->total_size != ntohs(frag_packet->total_size) || chain->total_size > batadv_frag_size_limit()) { /* Clear chain if total size of either the list or the packet diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h index 739439e2b235..c8c3e8064f00 100644 --- a/net/batman-adv/types.h +++ b/net/batman-adv/types.h @@ -301,7 +301,7 @@ struct batadv_frag_table_entry { u16 seqno; /** @size: accumulated size of packets in list */ - u16 size; + size_t size; /** @total_size: expected size of the assembled packet */ u16 total_size; -- cgit v1.2.3 From a340a51ed801eab7bb454150c226323b865263cc Mon Sep 17 00:00:00 2001 From: Ruijie Li Date: Thu, 14 May 2026 16:13:25 +0800 Subject: batman-adv: clear current gateway during teardown batadv_gw_node_free() removes the gateway list entries during mesh teardown, but it does not clear the currently selected gateway. This leaves stale gateway state behind across cleanup and can break a later mesh recreation. Clear bat_priv->gw.curr_gw before walking the gateway list so the selected gateway reference is dropped as part of teardown. Fixes: 2265c1410864 ("batman-adv: gateway election code refactoring") Cc: stable@kernel.org Reported-by: Yuan Tan Reported-by: Yifan Wu Reported-by: Juefei Pu Reported-by: Xin Liu Signed-off-by: Ruijie Li Signed-off-by: Zhanpeng Li Signed-off-by: Ren Wei Signed-off-by: Sven Eckelmann --- net/batman-adv/gateway_client.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'net') diff --git a/net/batman-adv/gateway_client.c b/net/batman-adv/gateway_client.c index 51e9c081a2a4..a9d0346e8332 100644 --- a/net/batman-adv/gateway_client.c +++ b/net/batman-adv/gateway_client.c @@ -478,10 +478,14 @@ void batadv_gw_node_delete(struct batadv_priv *bat_priv, */ void batadv_gw_node_free(struct batadv_priv *bat_priv) { + struct batadv_gw_node *curr_gw; struct batadv_gw_node *gw_node; struct hlist_node *node_tmp; spin_lock_bh(&bat_priv->gw.list_lock); + curr_gw = rcu_replace_pointer(bat_priv->gw.curr_gw, NULL, true); + batadv_gw_node_put(curr_gw); + hlist_for_each_entry_safe(gw_node, node_tmp, &bat_priv->gw.gateway_list, list) { hlist_del_init_rcu(&gw_node->list); -- cgit v1.2.3 From 2d8826a2d3657cea66fb0370f9e521575a673871 Mon Sep 17 00:00:00 2001 From: Sven Eckelmann Date: Wed, 13 May 2026 09:01:34 +0200 Subject: batman-adv: dat: handle forward allocation error batadv_dat_forward_data() calls pskb_copy_for_clone() to duplicate an skb for each DHT candidate, but does not check the return value before passing it to batadv_send_skb_prepare_unicast_4addr(). That function dereferences the skb unconditionally, so a failed allocation triggers a NULL pointer dereference. Skip forwarding to the current DHT candidate on allocation failure. Cc: stable@kernel.org Fixes: 785ea1144182 ("batman-adv: Distributed ARP Table - create DHT helper functions") Reported-by: Yuan Tan Reported-by: Yifan Wu Reported-by: Juefei Pu Reported-by: Xin Liu Reviewed-by: Yuan Tan Signed-off-by: Sven Eckelmann --- net/batman-adv/distributed-arp-table.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'net') diff --git a/net/batman-adv/distributed-arp-table.c b/net/batman-adv/distributed-arp-table.c index 3efc4cf50b46..0a8bd95e2f99 100644 --- a/net/batman-adv/distributed-arp-table.c +++ b/net/batman-adv/distributed-arp-table.c @@ -696,6 +696,9 @@ static bool batadv_dat_forward_data(struct batadv_priv *bat_priv, goto free_orig; tmp_skb = pskb_copy_for_clone(skb, GFP_ATOMIC); + if (!tmp_skb) + goto free_neigh; + if (!batadv_send_skb_prepare_unicast_4addr(bat_priv, tmp_skb, cand[i].orig_node, packet_subtype)) { -- cgit v1.2.3 From 6c65cf23d4c6170fcf5714c32aa64689718cb142 Mon Sep 17 00:00:00 2001 From: Sven Eckelmann Date: Wed, 13 May 2026 09:01:35 +0200 Subject: batman-adv: tp_meter: avoid use of uninit sender vars batadv_tp_recv_ack() and batadv_tp_stop() are only valid for tp_vars in the BATADV_TP_SENDER role. When called with a BATADV_TP_RECEIVER role, it proceeds to read sender-only members that were never initialized, leading to undefined behavior. This can be triggered when a node that is currently acting as a receiver in an ongoing tp_meter session receives a malicious ACK packet. Guard against this by checking tp_vars->role immediately after the lookup and bailing out if it is not BATADV_TP_SENDER, before any of those members are accessed. Cc: stable@kernel.org Fixes: 33a3bb4a3345 ("batman-adv: throughput meter implementation") Reported-by: Yuan Tan Reported-by: Yifan Wu Reported-by: Juefei Pu Reported-by: Xin Liu Reviewed-by: Yuan Tan Signed-off-by: Sven Eckelmann --- net/batman-adv/tp_meter.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/batman-adv/tp_meter.c b/net/batman-adv/tp_meter.c index ca6c3f6374bc..a3593d104caa 100644 --- a/net/batman-adv/tp_meter.c +++ b/net/batman-adv/tp_meter.c @@ -664,6 +664,9 @@ static void batadv_tp_recv_ack(struct batadv_priv *bat_priv, if (unlikely(!tp_vars)) return; + if (unlikely(tp_vars->role != BATADV_TP_SENDER)) + goto out; + if (unlikely(atomic_read(&tp_vars->sending) == 0)) goto out; @@ -1101,12 +1104,16 @@ void batadv_tp_stop(struct batadv_priv *bat_priv, const u8 *dst, if (!tp_vars) { batadv_dbg(BATADV_DBG_TP_METER, bat_priv, "Meter: trying to interrupt an already over connection\n"); - goto out; + goto out_put_orig_node; } + if (unlikely(tp_vars->role != BATADV_TP_SENDER)) + goto out_put_tp_vars; + batadv_tp_sender_shutdown(tp_vars, return_value); +out_put_tp_vars: batadv_tp_vars_put(tp_vars); -out: +out_put_orig_node: batadv_orig_node_put(orig_node); } -- cgit v1.2.3 From d2bfdbb69cf87676981b1043010b6224d84c6d3a Mon Sep 17 00:00:00 2001 From: Maoyi Xie Date: Tue, 12 May 2026 22:28:07 +0800 Subject: rds_tcp: close NULL deref window in rds_tcp_set_callbacks rds_tcp_set_callbacks() links a new rds_tcp_connection onto rds_tcp_tc_list under rds_tcp_tc_list_lock. It releases the lock, then assigns tc->t_sock = sock outside the lock. rds_tcp_tc_info() and rds6_tcp_tc_info() walk rds_tcp_tc_list under the same lock. Both dereference tc->t_sock->sk without a NULL check. A reader can acquire rds_tcp_tc_list_lock between the writer's spin_unlock and the t_sock store. It then sees a list entry whose t_sock is NULL. The dereference of tc->t_sock->sk is a NULL access. Move tc->t_sock = sock inside rds_tcp_tc_list_lock, before list_add_tail. A reader holding the lock then observes the linkage and the t_sock store together. The restore path is safe. rds_tcp_restore_callbacks() does list_del_init inside the lock. The matching tc->t_sock = NULL after unlink is harmless to readers holding the lock. Fixes: 70041088e3b9 ("RDS: Add TCP transport to RDS") Suggested-by: Simon Horman Signed-off-by: Maoyi Xie Reviewed-by: Allison Henderson Link: https://patch.msgid.link/20260512142807.1855619-1-maoyi.xie@ntu.edu.sg Signed-off-by: Jakub Kicinski --- net/rds/tcp.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/rds/tcp.c b/net/rds/tcp.c index 654e23d13e3d..5830b31a1f37 100644 --- a/net/rds/tcp.c +++ b/net/rds/tcp.c @@ -198,8 +198,13 @@ void rds_tcp_set_callbacks(struct socket *sock, struct rds_conn_path *cp) rdsdebug("setting sock %p callbacks to tc %p\n", sock, tc); write_lock_bh(&sock->sk->sk_callback_lock); - /* done under the callback_lock to serialize with write_space */ + /* done under the callback_lock to serialize with write_space. + * Set t_sock inside rds_tcp_tc_list_lock so readers walking + * rds_tcp_tc_list under the same lock cannot observe an + * entry whose t_sock is NULL. + */ spin_lock(&rds_tcp_tc_list_lock); + tc->t_sock = sock; list_add_tail(&tc->t_list_item, &rds_tcp_tc_list); #if IS_ENABLED(CONFIG_IPV6) rds6_tcp_tc_count++; @@ -211,8 +216,6 @@ void rds_tcp_set_callbacks(struct socket *sock, struct rds_conn_path *cp) /* accepted sockets need our listen data ready undone */ if (sock->sk->sk_data_ready == rds_tcp_listen_data_ready) sock->sk->sk_data_ready = sock->sk->sk_user_data; - - tc->t_sock = sock; if (!tc->t_rtn) tc->t_rtn = net_generic(sock_net(sock->sk), rds_tcp_netid); tc->t_cpath = cp; -- cgit v1.2.3 From f84eca5817390257cef78013d0112481c503b4a3 Mon Sep 17 00:00:00 2001 From: William Bowling Date: Wed, 13 May 2026 04:16:35 +0000 Subject: net: skbuff: preserve shared-frag marker during coalescing skb_try_coalesce() can attach paged frags from @from to @to. If @from has SKBFL_SHARED_FRAG set, the resulting @to skb can contain the same externally-owned or page-cache-backed frags, but the shared-frag marker is currently lost. That breaks the invariant relied on by later in-place writers. In particular, ESP input checks skb_has_shared_frag() before deciding whether an uncloned nonlinear skb can skip skb_cow_data(). If TCP receive coalescing has moved shared frags into an unmarked skb, ESP can see skb_has_shared_frag() as false and decrypt in place over page-cache backed frags. Propagate SKBFL_SHARED_FRAG when skb_try_coalesce() transfers paged frags. The tailroom copy path does not need the marker because it copies bytes into @to's linear data rather than transferring frag descriptors. Fixes: cef401de7be8 ("net: fix possible wrong checksum generation") Fixes: f4c50a4034e6 ("xfrm: esp: avoid in-place decrypt on shared skb frags") Signed-off-by: William Bowling Reviewed-by: Eric Dumazet Tested-by: Jiayuan Chen Link: https://patch.msgid.link/20260513041635.1289541-1-vakzz@zellic.io Signed-off-by: Jakub Kicinski --- net/core/skbuff.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'net') diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 7dad68e3b518..9c4e8d331d6d 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -6200,6 +6200,8 @@ bool skb_try_coalesce(struct sk_buff *to, struct sk_buff *from, from_shinfo->frags, from_shinfo->nr_frags * sizeof(skb_frag_t)); to_shinfo->nr_frags += from_shinfo->nr_frags; + if (from_shinfo->nr_frags) + to_shinfo->flags |= from_shinfo->flags & SKBFL_SHARED_FRAG; if (!skb_cloned(from)) from_shinfo->nr_frags = 0; -- cgit v1.2.3 From f508262ae9f21fe0e6c0749948b9dc7dd5a62a70 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Wed, 13 May 2026 08:58:25 -0400 Subject: tls: Preserve sk_err across recvmsg() when data has been copied The sk_err check in tls_rx_rec_wait() consumes the error via sock_error(), which clears sk_err atomically. When the caller (tls_sw_recvmsg, tls_sw_splice_read, or tls_sw_read_sock) already has bytes copied to userspace, it returns those bytes and discards the error from this call. sk_err is now zero on the socket, so the next read syscall observes only RCV_SHUTDOWN and reports a clean EOF instead of the actual error (typically -ECONNRESET). The race is reachable when tls_read_flush_backlog()'s periodic sk_flush_backlog() triggers tcp_reset() in the middle of a multi-record read. Pass a has_copied flag to tls_rx_rec_wait(). When has_copied is false, consume sk_err via sock_error() as before. When has_copied is true, report the error from READ_ONCE() but leave sk_err set: the caller returns the byte count and discards the err from this call, and the next read syscall surfaces the preserved sk_err. This mirrors the tcp_recvmsg() preserve-and-surface pattern. The decrypt-abort path is unaffected: tls_err_abort() raises sk_err to EBADMSG after tls_rx_rec_wait() returns, and nothing on the caller's return path consumes it, so the EBADMSG surfaces on the next read. tls_sw_splice_read() passes has_copied=false: it processes one record per call, so no bytes have been copied within the function when tls_rx_rec_wait() runs. A reset that arrives between iterations of splice_direct_to_actor() (the sendfile() path) is still consumed by sock_error() in the later call, and the outer loop returns the prior iterations' byte count and drops the error. tcp_splice_read() exhibits the same pattern at the iteration boundary; addressing it belongs at the splice_direct_to_actor() layer and is out of scope here. Fixes: c46b01839f7a ("tls: rx: periodically flush socket backlog") Suggested-by: Jakub Kicinski Signed-off-by: Chuck Lever Link: https://patch.msgid.link/20260513125825.205189-1-cel@kernel.org Signed-off-by: Jakub Kicinski --- net/tls/tls_sw.c | 20 ++++++++++++++------ 1 file changed, 14 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/net/tls/tls_sw.c b/net/tls/tls_sw.c index 3bfdaf5e64f5..964ebc268ee4 100644 --- a/net/tls/tls_sw.c +++ b/net/tls/tls_sw.c @@ -1366,9 +1366,14 @@ unlock: mutex_unlock(&tls_ctx->tx_lock); } +/* When has_copied is true the caller has already moved bytes to + * userspace. Report sk_err but leave it set so the next read + * surfaces it instead of a spurious EOF, otherwise sk_err is + * consumed via sock_error(). + */ static int tls_rx_rec_wait(struct sock *sk, struct sk_psock *psock, bool nonblock, - bool released) + bool released, bool has_copied) { struct tls_context *tls_ctx = tls_get_ctx(sk); struct tls_sw_context_rx *ctx = tls_sw_ctx_rx(tls_ctx); @@ -1386,8 +1391,11 @@ tls_rx_rec_wait(struct sock *sk, struct sk_psock *psock, bool nonblock, if (!sk_psock_queue_empty(psock)) return 0; - if (sk->sk_err) + if (sk->sk_err) { + if (has_copied) + return -READ_ONCE(sk->sk_err); return sock_error(sk); + } if (ret < 0) return ret; @@ -1423,7 +1431,7 @@ tls_rx_rec_wait(struct sock *sk, struct sk_psock *psock, bool nonblock, } if (unlikely(!tls_strp_msg_load(&ctx->strp, released))) - return tls_rx_rec_wait(sk, psock, nonblock, false); + return tls_rx_rec_wait(sk, psock, nonblock, false, has_copied); return 1; } @@ -2110,7 +2118,7 @@ int tls_sw_recvmsg(struct sock *sk, int to_decrypt, chunk; err = tls_rx_rec_wait(sk, psock, flags & MSG_DONTWAIT, - released); + released, !!(decrypted + copied)); if (err <= 0) { if (psock) { chunk = sk_msg_recvmsg(sk, psock, msg, len, @@ -2297,7 +2305,7 @@ ssize_t tls_sw_splice_read(struct socket *sock, loff_t *ppos, struct tls_decrypt_arg darg; err = tls_rx_rec_wait(sk, NULL, flags & SPLICE_F_NONBLOCK, - true); + true, false); if (err <= 0) goto splice_read_end; @@ -2383,7 +2391,7 @@ int tls_sw_read_sock(struct sock *sk, read_descriptor_t *desc, } else { struct tls_decrypt_arg darg; - err = tls_rx_rec_wait(sk, NULL, true, released); + err = tls_rx_rec_wait(sk, NULL, true, released, !!copied); if (err <= 0) goto read_sock_end; -- cgit v1.2.3 From bc62216dc8e221e3781afa14430f45208bfa9af9 Mon Sep 17 00:00:00 2001 From: Sven Eckelmann Date: Wed, 13 May 2026 09:01:36 +0200 Subject: batman-adv: frag: disallow unicast fragment in fragment batadv_frag_skb_buffer() is called by batadv_batman_skb_recv() when a BATADV_UNICAST_FRAG packet is received. Once all fragments are collected and the packet is reassembled, batadv_recv_frag_packet() calls batadv_batman_skb_recv() again to process the defragmented payload. A malicious sender can craft a BATADV_UNICAST_FRAG packet whose reassembled payload is itself a BATADV_UNICAST_FRAG packet (matryoshka-style nesting). Each nesting level recurses through batadv_batman_skb_recv() without bound, growing the kernel stack until it is exhausted. Since refragmentation or fragments in fragments are not actually allowed, discard all packets which are still BATADV_UNICAST_FRAG packets after the defragmentation process. Cc: stable@kernel.org Fixes: 610bfc6bc99b ("batman-adv: Receive fragmented packets and merge") Reported-by: Yuan Tan Reported-by: Yifan Wu Reported-by: Juefei Pu Reported-by: Xin Liu Reviewed-by: Yuan Tan Signed-off-by: Sven Eckelmann --- net/batman-adv/fragmentation.c | 35 +++++++++++++++++++++++++++++++++++ 1 file changed, 35 insertions(+) (limited to 'net') diff --git a/net/batman-adv/fragmentation.c b/net/batman-adv/fragmentation.c index 1152c2ce0c1e..4a594aa2ebf6 100644 --- a/net/batman-adv/fragmentation.c +++ b/net/batman-adv/fragmentation.c @@ -304,6 +304,31 @@ free: return skb_out; } +/** + * batadv_skb_is_frag() - check if newly merged skb is gain a unicast packet + * @skb: newly merged skb + * + * Return: if newly skb is of type BATADV_UNICAST_FRAG + */ +static bool batadv_skb_is_frag(struct sk_buff *skb) +{ + struct batadv_ogm_packet *batadv_ogm_packet; + + /* packet should hold at least type and version */ + if (unlikely(!pskb_may_pull(skb, 2))) + return false; + + batadv_ogm_packet = (struct batadv_ogm_packet *)skb->data; + + if (batadv_ogm_packet->version != BATADV_COMPAT_VERSION) + return false; + + if (batadv_ogm_packet->packet_type != BATADV_UNICAST_FRAG) + return false; + + return true; +} + /** * batadv_frag_skb_buffer() - buffer fragment for later merge * @skb: skb to buffer @@ -337,6 +362,16 @@ bool batadv_frag_skb_buffer(struct sk_buff **skb, if (!skb_out) goto out_err; + /* fragment in fragment is not allowed. otherwise it is possible + * to exhaust the stack when receiving a matryoshka-style + * "fragments in a fragment packet" + */ + if (batadv_skb_is_frag(skb_out)) { + kfree_skb(skb_out); + skb_out = NULL; + goto out_err; + } + out: ret = true; out_err: -- cgit v1.2.3 From d5487249a81ea658717614009c8f46acc5b7101a Mon Sep 17 00:00:00 2001 From: Sven Eckelmann Date: Wed, 13 May 2026 10:43:54 +0200 Subject: batman-adv: tp_meter: directly shut down timer on cleanup batadv_tp_sender_cleanup() was calling timer_delete_sync() followed by timer_delete() to guard against the timer handler re-arming itself between the two calls. This double-deletion hack relied on the sending status being set to 0 to suppress re-arming. Replace both calls with a single timer_shutdown_sync(). This function both waits for any running timer callback to complete (like timer_delete_sync()) and permanently disarms the timer so it cannot be re-armed afterwards, making re-arming prevention unconditional and self-documenting. The re-arming property is also required because otherwise: 1. context 0 (batadv_tp_recv_ack()) checks in batadv_tp_reset_sender_timer() if sending is still 1 -> it is 2. context 1 changes in batadv_tp_sender_shutdown() sending to 0 and in this process forces the kthread to stop timer in batadv_tp_sender_cleanup() 3. context 0 continues in batadv_tp_reset_sender_timer() and rearms the timer -> but the reference for it is already gone Cc: stable@kernel.org Fixes: 33a3bb4a3345 ("batman-adv: throughput meter implementation") Signed-off-by: Sven Eckelmann --- net/batman-adv/tp_meter.c | 8 +------- 1 file changed, 1 insertion(+), 7 deletions(-) (limited to 'net') diff --git a/net/batman-adv/tp_meter.c b/net/batman-adv/tp_meter.c index a3593d104caa..1fd1526059d8 100644 --- a/net/batman-adv/tp_meter.c +++ b/net/batman-adv/tp_meter.c @@ -401,13 +401,7 @@ static void batadv_tp_sender_cleanup(struct batadv_tp_vars *tp_vars) batadv_tp_list_detach(tp_vars); /* kill the timer and remove its reference */ - timer_delete_sync(&tp_vars->timer); - /* the worker might have rearmed itself therefore we kill it again. Note - * that if the worker should run again before invoking the following - * timer_delete(), it would not re-arm itself once again because the status - * is OFF now - */ - timer_delete(&tp_vars->timer); + timer_shutdown_sync(&tp_vars->timer); batadv_tp_vars_put(tp_vars); } -- cgit v1.2.3 From 915fab69823a14c170dbaa3b41978768e0fe62fc Mon Sep 17 00:00:00 2001 From: Michael Bommarito Date: Tue, 12 May 2026 16:51:14 -0400 Subject: ipv4: raw: reject IP_HDRINCL packets with ihl < 5 raw_send_hdrinc() validates that the caller-supplied IPv4 header fits within the message length: iphlen = iph->ihl * 4; err = -EINVAL; if (iphlen > length) goto error_free; if (iphlen >= sizeof(*iph)) { /* fix up saddr, tot_len, id, csum, transport_header */ } It does not, however, reject ihl < 5. For such a packet the "if (iphlen >= sizeof(*iph))" branch is skipped, leaving the crafted iphdr untouched, but the packet is still handed to __ip_local_out() and onward. Downstream consumers that read iph->ihl assume a sane value: net/ipv4/ah4.c:ah_output() in particular subtracts sizeof(struct iphdr) from top_iph->ihl * 4 and passes the (signed-int-negative, then cast to size_t) result to memcpy(), producing an OOB access of length close to SIZE_MAX and a host kernel panic. An IPv4 header with ihl < 5 is malformed by definition (RFC 791: "Internet Header Length is the length of the internet header in 32 bit words ... Note that the minimum value for a correct header is 5."). The kernel should not be willing to inject such a packet into its own output path. Reject "iphlen < sizeof(*iph)" alongside the existing "iphlen > length" check. This matches the principle that locally constructed packets that re-enter the IP stack must pass the same basic sanity tests that a foreign packet would be subjected to. Once this lands, the "if (iphlen >= sizeof(*iph))" wrapper around the fixup branch becomes redundant; left in place to keep the patch minimal and backport-friendly. A follow-up can unwrap it. Note that commit 86f4c90a1c5c ("ipv4, ipv6: ensure raw socket message is big enough to hold an IP header") ensures the message buffer is large enough to hold an iphdr, but does not constrain the self-reported iph->ihl. Reachability: the malformed packet source is any caller with CAP_NET_RAW, including an unprivileged process in a user+net namespace on a kernel with CONFIG_USER_NS=y. The reproduced AH crash also requires a matching xfrm AH policy on the outgoing route; a container granted CAP_NET_ADMIN can install that state and policy in its netns. Loopback bypasses xfrm_output, so the trigger uses a real netdev. Reproduced on UML + KASAN: kernel-mode fault at addr 0x0 with memcpy_orig at the crash site. Same shape reproduces inside a rootless Docker container with --cap-add NET_ADMIN on a stock distro kernel. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Cc: stable@vger.kernel.org Suggested-by: Herbert Xu Signed-off-by: Michael Bommarito Link: https://patch.msgid.link/77ec2b5e8111961c2c39883c92e8aa2709039c17.1778614451.git.michael.bommarito@gmail.com Signed-off-by: Jakub Kicinski --- net/ipv4/raw.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c index 5aaf9c62c8e1..68e88cb3e55c 100644 --- a/net/ipv4/raw.c +++ b/net/ipv4/raw.c @@ -391,7 +391,7 @@ static int raw_send_hdrinc(struct sock *sk, struct flowi4 *fl4, * in, reject the frame as invalid */ err = -EINVAL; - if (iphlen > length) + if (iphlen > length || iphlen < sizeof(*iph)) goto error_free; if (iphlen >= sizeof(*iph)) { -- cgit v1.2.3 From ae38d9179190a956e2a87a69ef1dd6f451b51c4d Mon Sep 17 00:00:00 2001 From: Stefano Garzarella Date: Thu, 14 May 2026 11:29:48 +0200 Subject: vsock/virtio: fix zerocopy completion for multi-skb sends When a large message is fragmented into multiple skbs, the zerocopy uarg is only allocated and attached to the last skb in the loop. Non-final skbs carry pinned user pages with no completion tracking, so the kernel has no way to notify userspace when those pages are safe to reuse. If the loop breaks early the uarg is never allocated at all, leaking pinned pages with no completion notification. Fix this by following the approach used by TCP: allocate the zerocopy uarg (if not provided by the caller) before the send loop and attach it to every skb via skb_zcopy_set(), which takes a reference per skb. Each skb's completion properly decrements the refcount, and the notification only fires after the last skb is freed. On failure, if no data was sent, the uarg is cleanly aborted via net_zcopy_put_abort(). This issue was initially discovered by sashiko while reviewing commit 1cb36e252211 ("vsock/virtio: fix MSG_ZEROCOPY pinned-pages accounting") but was pre-existing. Fixes: 581512a6dc93 ("vsock/virtio: MSG_ZEROCOPY flag support") Closes: https://sashiko.dev/#/patchset/20260420132051.217589-1-sgarzare%40redhat.com Reported-by: Maher Azzouzi Signed-off-by: Stefano Garzarella Acked-by: Michael S. Tsirkin Acked-by: Arseniy Krasnov Link: https://patch.msgid.link/20260514092948.268720-1-sgarzare@redhat.com Signed-off-by: Jakub Kicinski --- net/vmw_vsock/virtio_transport_common.c | 83 ++++++++++++++------------------- 1 file changed, 34 insertions(+), 49 deletions(-) (limited to 'net') diff --git a/net/vmw_vsock/virtio_transport_common.c b/net/vmw_vsock/virtio_transport_common.c index 989cc252d3d3..1e3409d28164 100644 --- a/net/vmw_vsock/virtio_transport_common.c +++ b/net/vmw_vsock/virtio_transport_common.c @@ -70,34 +70,6 @@ static bool virtio_transport_can_zcopy(const struct virtio_transport *t_ops, return true; } -static int virtio_transport_init_zcopy_skb(struct vsock_sock *vsk, - struct sk_buff *skb, - struct msghdr *msg, - size_t pkt_len, - bool zerocopy) -{ - struct ubuf_info *uarg; - - if (msg->msg_ubuf) { - uarg = msg->msg_ubuf; - net_zcopy_get(uarg); - } else { - struct ubuf_info_msgzc *uarg_zc; - - uarg = msg_zerocopy_realloc(sk_vsock(vsk), - pkt_len, NULL, false); - if (!uarg) - return -1; - - uarg_zc = uarg_to_msgzc(uarg); - uarg_zc->zerocopy = zerocopy ? 1 : 0; - } - - skb_zcopy_init(skb, uarg); - - return 0; -} - static int virtio_transport_fill_skb(struct sk_buff *skb, struct virtio_vsock_pkt_info *info, size_t len, @@ -317,8 +289,10 @@ static int virtio_transport_send_pkt_info(struct vsock_sock *vsk, u32 src_cid, src_port, dst_cid, dst_port; const struct virtio_transport *t_ops; struct virtio_vsock_sock *vvs; + struct ubuf_info *uarg = NULL; u32 pkt_len = info->pkt_len; bool can_zcopy = false; + bool have_uref = false; u32 rest_len; int ret; @@ -360,6 +334,25 @@ static int virtio_transport_send_pkt_info(struct vsock_sock *vsk, if (can_zcopy) max_skb_len = min_t(u32, VIRTIO_VSOCK_MAX_PKT_BUF_SIZE, (MAX_SKB_FRAGS * PAGE_SIZE)); + + if (info->msg->msg_flags & MSG_ZEROCOPY && + info->op == VIRTIO_VSOCK_OP_RW) { + uarg = info->msg->msg_ubuf; + + if (!uarg) { + uarg = msg_zerocopy_realloc(sk_vsock(vsk), + pkt_len, NULL, false); + if (!uarg) { + virtio_transport_put_credit(vvs, pkt_len); + return -ENOMEM; + } + + if (!can_zcopy) + uarg_to_msgzc(uarg)->zerocopy = 0; + + have_uref = true; + } + } } rest_len = pkt_len; @@ -378,27 +371,7 @@ static int virtio_transport_send_pkt_info(struct vsock_sock *vsk, break; } - /* We process buffer part by part, allocating skb on - * each iteration. If this is last skb for this buffer - * and MSG_ZEROCOPY mode is in use - we must allocate - * completion for the current syscall. - * - * Pass pkt_len because msg iter is already consumed - * by virtio_transport_fill_skb(), so iter->count - * can not be used for RLIMIT_MEMLOCK pinned-pages - * accounting done by msg_zerocopy_realloc(). - */ - if (info->msg && info->msg->msg_flags & MSG_ZEROCOPY && - skb_len == rest_len && info->op == VIRTIO_VSOCK_OP_RW) { - if (virtio_transport_init_zcopy_skb(vsk, skb, - info->msg, - pkt_len, - can_zcopy)) { - kfree_skb(skb); - ret = -ENOMEM; - break; - } - } + skb_zcopy_set(skb, uarg, NULL); virtio_transport_inc_tx_pkt(vvs, skb); @@ -422,6 +395,18 @@ static int virtio_transport_send_pkt_info(struct vsock_sock *vsk, virtio_transport_put_credit(vvs, rest_len); + /* msg_zerocopy_realloc() initializes the ubuf_info refcnt to 1. + * skb_zcopy_set() increases it for each skb, so we can drop that + * initial reference to keep it balanced. + */ + if (have_uref) { + if (rest_len == pkt_len) + /* No data sent, abort the notification. */ + net_zcopy_put_abort(uarg, true); + else + net_zcopy_put(uarg); + } + /* Return number of bytes, if any data has been sent. */ if (rest_len != pkt_len) ret = pkt_len - rest_len; -- cgit v1.2.3 From aaec7096f9961eb223b5b149abe9495525c205d9 Mon Sep 17 00:00:00 2001 From: Michael Bommarito Date: Wed, 13 May 2026 19:38:38 -0400 Subject: net: hsr: defer node table free until after RCU readers HSR node-list and node-status generic-netlink operations run under rcu_read_lock(). They walk hsr->node_db through hsr_get_next_node() and hsr_get_node_data(), but RTM_DELLINK teardown removes the same node table with plain list_del() and frees each node immediately. That lets a generic-netlink reader hold a struct hsr_node pointer across hsr_dellink(). In a KASAN build, widening the reader window after hsr_get_next_node() obtains the node reproduces a slab-use-after-free when the reader copies node->macaddress_A; the freeing stack is hsr_del_nodes() from hsr_dellink(). Use list_del_rcu() and defer the free through the existing hsr_free_node_rcu() callback. This matches the lifetime rule used by the HSR prune paths, which already delete nodes with list_del_rcu() and call_rcu(). Fixes: b9a1e627405d ("hsr: implement dellink to clean up resources") Cc: stable@vger.kernel.org # v5.3+ Signed-off-by: Michael Bommarito Link: https://patch.msgid.link/20260513233838.3064715-2-michael.bommarito@gmail.com Signed-off-by: Jakub Kicinski --- net/hsr/hsr_framereg.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/hsr/hsr_framereg.c b/net/hsr/hsr_framereg.c index 124619920d38..b514e43766ef 100644 --- a/net/hsr/hsr_framereg.c +++ b/net/hsr/hsr_framereg.c @@ -163,8 +163,8 @@ void hsr_del_nodes(struct list_head *node_db) struct hsr_node *tmp; list_for_each_entry_safe(node, tmp, node_db, mac_list) { - list_del(&node->mac_list); - hsr_free_node(node); + list_del_rcu(&node->mac_list); + call_rcu(&node->rcu_head, hsr_free_node_rcu); } } -- cgit v1.2.3 From 79d8be262377f7112cfa3088dfc4142d5a2533f3 Mon Sep 17 00:00:00 2001 From: Michael Bommarito Date: Fri, 15 May 2026 11:45:31 -0400 Subject: xfrm: ah: use skb_to_full_sk in async output callbacks When AH output is offloaded to an asynchronous crypto provider (hardware accelerators such as AMD CCP, or a forced-async software shim used for testing), the digest completion fires ah_output_done() / ah6_output_done() on a workqueue. The egress skb at that point may have been originated by a TCP listener sending a SYN-ACK, which sets skb->sk to a request_sock via skb_set_owner_edemux(); it may also have been originated by an inet_timewait_sock retransmit. Neither is a full struct sock, and passing the raw skb->sk to xfrm_output_resume() then forwards a non-full socket through the rest of the xfrm output chain. xfrm_output_resume() and its downstream consumers expect a full sk where they dereference at all. The natural egress path through ah_output_done() does not crash today because the consumers that read past sock_common are either gated by sk_fullsock() or short-circuit on flags that are clear on a fresh request_sock; an exhaustive walk of the 50 most plausible consumers under sch_fq, dev_queue_xmit, netfilter, tc-egress and cgroup-egress BPF found no current unguarded deref. The bug is still a real type confusion that future consumer changes could turn into a memory-corruption primitive. This is the same bug class fixed for ESP in commit 1620c88887b1 ("xfrm: Fix the usage of skb->sk"). Apply the analogous fix to AH: convert skb->sk to a full socket pointer (or NULL) via skb_to_full_sk() before handing it to xfrm_output_resume(). The same async AH callbacks were touched recently for an independent ESN-related ICV layout bug in commit ec54093e6a8f ("xfrm: ah: account for ESN high bits in async callbacks"); the sk type-confusion addressed here is orthogonal. This patch is part of an ongoing audit of the AH callback paths; an ah_output ihl-validation hardening series is also currently under review on netdev. Reproduced under UML + KASAN + lockdep with a forced-async hmac(sha1) shim that registers at priority 9999 and wraps the sync in-tree hmac-sha1-lib. With the shim loaded, ah_output_done runs on every SYN-ACK egress through a transport-mode AH SA and skb->sk arrives as a request_sock (TCP_NEW_SYN_RECV); after this patch, xfrm_output_resume() receives the listener (the result of sk_to_full_sk()) and consumer derefs land on full-sock fields as intended. Fixes: 9ab1265d5231 ("xfrm: Use actual socket sk instead of skb socket for xfrm_output_resume") Cc: stable@vger.kernel.org Assisted-by: Claude:claude-opus-4-7 Signed-off-by: Michael Bommarito Signed-off-by: Steffen Klassert --- net/ipv4/ah4.c | 2 +- net/ipv6/ah6.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/ipv4/ah4.c b/net/ipv4/ah4.c index 4366cbac3f06..6fd642d2278d 100644 --- a/net/ipv4/ah4.c +++ b/net/ipv4/ah4.c @@ -143,7 +143,7 @@ static void ah_output_done(void *data, int err) } kfree(AH_SKB_CB(skb)->tmp); - xfrm_output_resume(skb->sk, skb, err); + xfrm_output_resume(skb_to_full_sk(skb), skb, err); } static int ah_output(struct xfrm_state *x, struct sk_buff *skb) diff --git a/net/ipv6/ah6.c b/net/ipv6/ah6.c index de1e68199a01..76f7a2de9108 100644 --- a/net/ipv6/ah6.c +++ b/net/ipv6/ah6.c @@ -337,7 +337,7 @@ static void ah6_output_done(void *data, int err) ah6_restore_hdrs(top_iph, iph_ext, extlen); kfree(AH_SKB_CB(skb)->tmp); - xfrm_output_resume(skb->sk, skb, err); + xfrm_output_resume(skb_to_full_sk(skb), skb, err); } static int ah6_output(struct xfrm_state *x, struct sk_buff *skb) -- cgit v1.2.3 From 1afc25ae75288b3ce59e9e5a4b448bd354c9e565 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Sat, 9 May 2026 10:27:06 +0200 Subject: netfilter: nf_conntrack_helper: fix possible null deref during error log Reported by sashiko: there is a small race window. If a helper module is unloaded or a userspace-defined helper is removed, nf_conntrack_helper_unregister() sets ->helper to NULL. Handle this safely. This needs a second patch to close related race during nf_conntrack_helper_unregister(). Fixes: b20ab9cc63ca ("netfilter: nf_ct_helper: better logging for dropped packets") Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_conntrack_helper.c | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/netfilter/nf_conntrack_helper.c b/net/netfilter/nf_conntrack_helper.c index b594cd244fe1..17e971bd4c74 100644 --- a/net/netfilter/nf_conntrack_helper.c +++ b/net/netfilter/nf_conntrack_helper.c @@ -321,8 +321,8 @@ __printf(3, 4) void nf_ct_helper_log(struct sk_buff *skb, const struct nf_conn *ct, const char *fmt, ...) { + const char *helper_name = "(null)"; const struct nf_conn_help *help; - const struct nf_conntrack_helper *helper; struct va_format vaf; va_list args; @@ -331,14 +331,17 @@ void nf_ct_helper_log(struct sk_buff *skb, const struct nf_conn *ct, vaf.fmt = fmt; vaf.va = &args; - /* Called from the helper function, this call never fails */ help = nfct_help(ct); + if (help) { + const struct nf_conntrack_helper *helper; - /* rcu_read_lock()ed by nf_hook_thresh */ - helper = rcu_dereference(help->helper); + helper = rcu_dereference(help->helper); + if (helper) + helper_name = helper->name; + } nf_log_packet(nf_ct_net(ct), nf_ct_l3num(ct), 0, skb, NULL, NULL, NULL, - "nf_ct_%s: dropping packet: %pV ", helper->name, &vaf); + "helper %s dropping packet: %pV ", helper_name, &vaf); va_end(args); } -- cgit v1.2.3 From 5522d65d81a711c60a9969d37a485d48d0ad1496 Mon Sep 17 00:00:00 2001 From: Julian Anastasov Date: Sun, 10 May 2026 13:46:05 +0300 Subject: ipvs: avoid possible loop in ip_vs_dst_event on resizing Sashiko points out that unprivileged user can frequently call ip_vs_flush() or ip_vs_del_service() to trigger svc_table_changes updates that can lead to infinite loop in ip_vs_dst_event(). This can also happen if the user triggers frequent table resizing without deleting all services. We should also consider the possible effects if the user triggers many NETDEV_DOWN events. One way to solve it is to hold svc_resize_sem in ip_vs_dst_event() but this can block the dev notifier during the whole resizing process. Instead, use new rw_semaphore svc_replace_sem to protect just the svc_table replacement which is a short code section. Then hold svc_replace_sem in ip_vs_dst_event() to serialize with replacing the svc_table. As result, loop is avoided as there is no need to repeat the table walking from the start. By this way changes in svc_table_changes can happen only when all services are removed and all dev references dropped which allows us to abort the table walking. As IP_VS_WORK_SVC_NORESIZE is the flag used to stop the svc_resize_work under service_mutex, we should check only this flag often but not while under service_mutex. To remove the mutex_trylock() for service_mutex in the second phase where the resizer installs the new table after rehashing, we will avoid holding the service_mutex there. As result, the code in configuration context which is under service_mutex should access ipvs->svc_table under RCU because it can be replaced at anytime and released after a RCU grace period. As for ip_vs_zero_all(), it needs different solution as a table walker which can escape single RCU read-side critical section: to hold the svc_replace_sem to prevent table to be replaced. In ip_vs_status_show() prefer to hold svc_replace_sem to avoid many loops, just detect if the svc_table is removed. Prefer the newly attached table for the u_thresh/l_thresh checks to know when to grow/shrink while adding or deleting services because the new table size is based on the latest parameters. Link: https://sashiko.dev/#/patchset/20260505001648.360569-1-pablo%40netfilter.org Fixes: 840aac3d900d ("ipvs: use resizable hash table for services") Signed-off-by: Julian Anastasov Signed-off-by: Pablo Neira Ayuso --- net/netfilter/ipvs/ip_vs_ctl.c | 187 +++++++++++++++++++++++++++-------------- 1 file changed, 122 insertions(+), 65 deletions(-) (limited to 'net') diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c index c7c7f6a7a9f6..bd9cae44d214 100644 --- a/net/netfilter/ipvs/ip_vs_ctl.c +++ b/net/netfilter/ipvs/ip_vs_ctl.c @@ -327,18 +327,22 @@ ip_vs_use_count_dec(void) /* Service hashing: * Operation Locking order * --------------------------------------------------------------------------- - * add table service_mutex, svc_resize_sem(W) - * del table service_mutex - * move between tables svc_resize_sem(W), seqcount_t(W), bit lock - * add/del service service_mutex, bit lock + * add first table service_mutex + * attach new table service_mutex + * add/del service service_mutex, RCU, bit lock + * move between tables (rehash) svc_resize_sem(W), seqcount_t(W), bit lock + * replace old with attached svc_resize_sem(W), svc_replace_sem(W) * find service RCU, seqcount_t(R) * walk services(blocking) service_mutex, svc_resize_sem(R) * walk services(non-blocking) RCU, seqcount_t(R) + * walk services(non-blocking) svc_resize_sem(R), RCU, seqcount_t(R) + * walk services(non-blocking) svc_replace_sem(R), RCU, seqcount_t(R) + * del table service_mutex after stopped work * - * - new tables are linked/unlinked under service_mutex and svc_resize_sem - * - new table is linked on resizing and all operations can run in parallel - * in 2 tables until the new table is registered as current one - * - two contexts can modify buckets: config and table resize, both in + * - new table is attached on resizing under service_mutex and all operations + * can run in parallel in 2 tables until the new table is registered as current + * one + * - two contexts can modify buckets: config and table resize (work), both in * process context * - only table resizer can move entries, so we do not protect t->seqc[] * items with t->lock[] @@ -346,9 +350,13 @@ ip_vs_use_count_dec(void) * services are moved to new table * - move operations may disturb readers: find operation will not miss entries * but walkers may see same entry twice if they are forced to retry chains - * - walkers using cond_resched_rcu() on !PREEMPT_RCU may need to hold - * service_mutex to disallow new tables to be installed or to check + * or to walk the newly attached second table + * - walkers using cond_resched_rcu() on !PREEMPT_RCU may need to check * svc_table_changes and repeat the RCU read section if new table is installed + * - walkers may serialize with the whole resizing process (svc_resize_sem) + * to prevent seeing same service twice or just with the svc_table + * replace (svc_replace_sem) when we can see entries twice but we + * prefer to run concurrently with the rehashing. */ /* @@ -387,9 +395,16 @@ static int ip_vs_svc_hash(struct ip_vs_service *svc) /* increase its refcnt because it is referenced by the svc table */ atomic_inc(&svc->refcnt); + /* We know if new table is attached under service_mutex but rely on + * RCU to hold the old table to be freed in resizer + */ + rcu_read_lock(); + + /* This can be the old or the new table */ + t = rcu_dereference(ipvs->svc_table); + /* New entries go into recent table */ - t = rcu_dereference_protected(ipvs->svc_table, 1); - t = rcu_dereference_protected(t->new_tbl, 1); + t = rcu_dereference(t->new_tbl); if (svc->fwmark == 0) { /* @@ -410,6 +425,8 @@ static int ip_vs_svc_hash(struct ip_vs_service *svc) hlist_bl_add_head_rcu(&svc->s_list, head); hlist_bl_unlock(head); + rcu_read_unlock(); + return 1; } @@ -432,7 +449,13 @@ static int ip_vs_svc_unhash(struct ip_vs_service *svc) return 0; } - t = rcu_dereference_protected(ipvs->svc_table, 1); + /* We know if new table is attached under service_mutex but rely on + * RCU to hold the old table to be freed in resizer + */ + rcu_read_lock(); + + /* This can be the old or the new table */ + t = rcu_dereference(ipvs->svc_table); hash_key = READ_ONCE(svc->hash_key); /* We need to lock the bucket in the right table */ if (ip_vs_rht_same_table(t, hash_key)) { @@ -443,13 +466,13 @@ static int ip_vs_svc_unhash(struct ip_vs_service *svc) /* Moved to new table ? */ if (hash_key != hash_key2) { hlist_bl_unlock(head); - t = rcu_dereference_protected(t->new_tbl, 1); + t = rcu_dereference(t->new_tbl); head = t->buckets + (hash_key2 & t->mask); hlist_bl_lock(head); } } else { /* It is already moved to new table */ - t = rcu_dereference_protected(t->new_tbl, 1); + t = rcu_dereference(t->new_tbl); head = t->buckets + (hash_key & t->mask); hlist_bl_lock(head); } @@ -459,6 +482,8 @@ static int ip_vs_svc_unhash(struct ip_vs_service *svc) svc->flags &= ~IP_VS_SVC_F_HASHED; atomic_dec(&svc->refcnt); hlist_bl_unlock(head); + + rcu_read_unlock(); return 1; } @@ -666,15 +691,14 @@ static void svc_resize_work_handler(struct work_struct *work) goto unlock_sem; more_work = false; clear_bit(IP_VS_WORK_SVC_RESIZE, &ipvs->work_flags); - if (!READ_ONCE(ipvs->enable) || - test_bit(IP_VS_WORK_SVC_NORESIZE, &ipvs->work_flags)) + if (!READ_ONCE(ipvs->enable)) goto unlock_m; t = rcu_dereference_protected(ipvs->svc_table, 1); /* Do nothing if table is removed */ if (!t) goto unlock_m; - /* New table needs to be registered? BUG! */ - if (t != rcu_dereference_protected(t->new_tbl, 1)) + /* New table already attached? BUG! */ + if (t != rcu_access_pointer(t->new_tbl)) goto unlock_m; lfactor = sysctl_svc_lfactor(ipvs); @@ -691,6 +715,7 @@ static void svc_resize_work_handler(struct work_struct *work) /* Flip the table_id */ t_new->table_id = t->table_id ^ IP_VS_RHT_TABLE_ID_MASK; + /* Attach new table */ rcu_assign_pointer(t->new_tbl, t_new); /* Allow add/del to new_tbl while moving from old table */ mutex_unlock(&ipvs->service_mutex); @@ -698,8 +723,8 @@ static void svc_resize_work_handler(struct work_struct *work) ip_vs_rht_for_each_bucket(t, bucket, head) { same_bucket: if (++limit >= 16) { - if (!READ_ONCE(ipvs->enable) || - test_bit(IP_VS_WORK_SVC_NORESIZE, + /* Check if work is stopped */ + if (test_bit(IP_VS_WORK_SVC_NORESIZE, &ipvs->work_flags)) goto unlock_sem; if (resched_score >= 100) { @@ -764,16 +789,12 @@ same_bucket: goto same_bucket; } - /* Tables can be switched only under service_mutex */ - while (!mutex_trylock(&ipvs->service_mutex)) { - cond_resched(); - if (!READ_ONCE(ipvs->enable) || - test_bit(IP_VS_WORK_SVC_NORESIZE, &ipvs->work_flags)) - goto unlock_sem; - } - if (!READ_ONCE(ipvs->enable) || - test_bit(IP_VS_WORK_SVC_NORESIZE, &ipvs->work_flags)) - goto unlock_m; + /* Serialize with readers that don't like svc_table changes */ + down_write(&ipvs->svc_replace_sem); + + /* Check if work is stopped to avoid synchronize_rcu() */ + if (test_bit(IP_VS_WORK_SVC_NORESIZE, &ipvs->work_flags)) + goto unlock_repl; rcu_assign_pointer(ipvs->svc_table, t_new); /* Inform readers that new table is installed */ @@ -781,8 +802,8 @@ same_bucket: atomic_inc(&ipvs->svc_table_changes); t_free = t; -unlock_m: - mutex_unlock(&ipvs->service_mutex); +unlock_repl: + up_write(&ipvs->svc_replace_sem); unlock_sem: up_write(&ipvs->svc_resize_sem); @@ -801,6 +822,11 @@ out: test_bit(IP_VS_WORK_SVC_NORESIZE, &ipvs->work_flags)) return; queue_delayed_work(system_unbound_wq, &ipvs->svc_resize_work, 1); + return; + +unlock_m: + mutex_unlock(&ipvs->service_mutex); + goto unlock_sem; } static inline void @@ -1691,6 +1717,7 @@ ip_vs_add_service(struct netns_ipvs *ipvs, struct ip_vs_service_user_kern *u, struct ip_vs_pe *pe = NULL; int ret_hooks = -1; int ret = 0; + bool grow; /* increase the module use count */ if (!ip_vs_use_count_inc()) @@ -1732,16 +1759,25 @@ ip_vs_add_service(struct netns_ipvs *ipvs, struct ip_vs_service_user_kern *u, } #endif - t = rcu_dereference_protected(ipvs->svc_table, 1); + /* The old table can be freed, protect it with RCU */ + rcu_read_lock(); + t = rcu_dereference(ipvs->svc_table); if (!t) { int lfactor = sysctl_svc_lfactor(ipvs); int new_size = ip_vs_svc_desired_size(ipvs, NULL, lfactor); + rcu_read_unlock(); t_new = ip_vs_svc_table_alloc(ipvs, new_size, lfactor); if (!t_new) { ret = -ENOMEM; goto out_err; } + grow = false; + } else { + /* Even the currently attached new table may need to grow */ + t = rcu_dereference(t->new_tbl); + grow = ip_vs_get_num_services(ipvs) + 1 > t->u_thresh; + rcu_read_unlock(); } if (!rcu_dereference_protected(ipvs->conn_tab, 1)) { @@ -1800,6 +1836,7 @@ ip_vs_add_service(struct netns_ipvs *ipvs, struct ip_vs_service_user_kern *u, goto out_err; if (t_new) { + /* Add table for first time */ clear_bit(IP_VS_WORK_SVC_NORESIZE, &ipvs->work_flags); rcu_assign_pointer(ipvs->svc_table, t_new); t_new = NULL; @@ -1831,8 +1868,7 @@ ip_vs_add_service(struct netns_ipvs *ipvs, struct ip_vs_service_user_kern *u, ip_vs_svc_hash(svc); /* Schedule resize work */ - if (t && ip_vs_get_num_services(ipvs) > t->u_thresh && - !test_and_set_bit(IP_VS_WORK_SVC_RESIZE, &ipvs->work_flags)) + if (grow && !test_and_set_bit(IP_VS_WORK_SVC_RESIZE, &ipvs->work_flags)) queue_delayed_work(system_unbound_wq, &ipvs->svc_resize_work, 1); @@ -2054,7 +2090,6 @@ static int ip_vs_del_service(struct ip_vs_service *svc) return -EEXIST; ipvs = svc->ipvs; ip_vs_unlink_service(svc, false); - t = rcu_dereference_protected(ipvs->svc_table, 1); /* Drop the table if no more services */ ns = ip_vs_get_num_services(ipvs); @@ -2062,6 +2097,7 @@ static int ip_vs_del_service(struct ip_vs_service *svc) /* Stop the resizer and drop the tables */ set_bit(IP_VS_WORK_SVC_NORESIZE, &ipvs->work_flags); cancel_delayed_work_sync(&ipvs->svc_resize_work); + t = rcu_dereference_protected(ipvs->svc_table, 1); if (t) { rcu_assign_pointer(ipvs->svc_table, NULL); /* Inform readers that table is removed */ @@ -2075,11 +2111,19 @@ static int ip_vs_del_service(struct ip_vs_service *svc) t = p; } } - } else if (ns <= t->l_thresh && - !test_and_set_bit(IP_VS_WORK_SVC_RESIZE, - &ipvs->work_flags)) { - queue_delayed_work(system_unbound_wq, &ipvs->svc_resize_work, - 1); + } else { + bool shrink; + + rcu_read_lock(); + t = rcu_dereference(ipvs->svc_table); + /* Even the currently attached new table may need to shrink */ + t = rcu_dereference(t->new_tbl); + shrink = ns <= t->l_thresh; + rcu_read_unlock(); + if (shrink && !test_and_set_bit(IP_VS_WORK_SVC_RESIZE, + &ipvs->work_flags)) + queue_delayed_work(system_unbound_wq, + &ipvs->svc_resize_work, 1); } return 0; } @@ -2184,17 +2228,21 @@ static int ip_vs_dst_event(struct notifier_block *this, unsigned long event, struct ip_vs_service *svc; struct hlist_bl_node *e; struct ip_vs_dest *dest; - int old_gen, new_gen; + int old_gen; if (event != NETDEV_DOWN || !ipvs) return NOTIFY_DONE; IP_VS_DBG(3, "%s() dev=%s\n", __func__, dev->name); + /* Allow concurrent rehashing on resize but to avoid loop + * serialize with installing the new table. + */ + down_read(&ipvs->svc_replace_sem); + old_gen = atomic_read(&ipvs->svc_table_changes); rcu_read_lock(); -repeat: smp_rmb(); /* ipvs->svc_table and svc_table_changes */ ip_vs_rht_walk_buckets_rcu(ipvs->svc_table, head) { hlist_bl_for_each_entry_rcu(svc, e, head, s_list) { @@ -2207,17 +2255,17 @@ repeat: } resched_score++; if (resched_score >= 100) { - resched_score = 0; cond_resched_rcu(); - new_gen = atomic_read(&ipvs->svc_table_changes); - /* New table installed ? */ - if (old_gen != new_gen) { - old_gen = new_gen; - goto repeat; - } + /* Flushed? So no more dev refs */ + if (atomic_read(&ipvs->svc_table_changes) != old_gen) + goto done; + resched_score = 0; } } + +done: rcu_read_unlock(); + up_read(&ipvs->svc_replace_sem); return NOTIFY_DONE; } @@ -2244,6 +2292,10 @@ static int ip_vs_zero_all(struct netns_ipvs *ipvs) struct ip_vs_service *svc; struct hlist_bl_node *e; + /* svc_table can not be replaced (svc_replace_sem) or + * removed (service_mutex) + */ + down_read(&ipvs->svc_replace_sem); rcu_read_lock(); ip_vs_rht_walk_buckets_rcu(ipvs->svc_table, head) { @@ -2259,6 +2311,7 @@ static int ip_vs_zero_all(struct netns_ipvs *ipvs) } rcu_read_unlock(); + up_read(&ipvs->svc_replace_sem); ip_vs_zero_stats(&ipvs->tot_stats->s); return 0; @@ -3062,6 +3115,7 @@ static int ip_vs_status_show(struct seq_file *seq, void *v) u32 sum; int i; + /* Info for conns */ rcu_read_lock(); t = rcu_dereference(ipvs->conn_tab); @@ -3123,6 +3177,12 @@ repeat_conn: } after_conns: + rcu_read_unlock(); + + /* Info for services */ + down_read(&ipvs->svc_replace_sem); + rcu_read_lock(); + t = rcu_dereference(ipvs->svc_table); count = ip_vs_get_num_services(ipvs); @@ -3133,9 +3193,7 @@ after_conns: if (!count) goto after_svc; old_gen = atomic_read(&ipvs->svc_table_changes); - loops = 0; -repeat_svc: smp_rmb(); /* ipvs->svc_table and svc_table_changes */ memset(counts, 0, sizeof(counts)); ip_vs_rht_for_each_table_rcu(ipvs->svc_table, t, pt) { @@ -3157,15 +3215,10 @@ repeat_svc: if (resched_score >= 100) { resched_score = 0; cond_resched_rcu(); - new_gen = atomic_read(&ipvs->svc_table_changes); - /* New table installed ? */ - if (old_gen != new_gen) { - /* Too many changes? */ - if (++loops >= 5) - goto after_svc; - old_gen = new_gen; - goto repeat_svc; - } + /* Flushed? */ + if (atomic_read(&ipvs->svc_table_changes) != + old_gen) + goto after_svc; } counts[count]++; } @@ -3184,6 +3237,9 @@ repeat_svc: } after_svc: + rcu_read_unlock(); + up_read(&ipvs->svc_replace_sem); + seq_printf(seq, "Stats thread slots:\t%d (max %lu)\n", ipvs->est_kt_count, ipvs->est_max_threads); seq_printf(seq, "Stats chain max len:\t%d\n", ipvs->est_chain_max); @@ -3191,7 +3247,6 @@ after_svc: ipvs->est_chain_max * IPVS_EST_CHAIN_FACTOR * IPVS_EST_NTICKS); - rcu_read_unlock(); return 0; } @@ -3503,7 +3558,7 @@ __ip_vs_get_service_entries(struct netns_ipvs *ipvs, int ret = 0; lockdep_assert_held(&ipvs->svc_resize_sem); - /* All service modifications are disabled, go ahead */ + /* All svc_table modifications are disabled, go ahead */ ip_vs_rht_walk_buckets(ipvs->svc_table, head) { hlist_bl_for_each_entry(svc, e, head, s_list) { /* Only expose IPv4 entries to old interface */ @@ -3687,7 +3742,7 @@ do_ip_vs_get_ctl(struct sock *sk, int cmd, void __user *user, int *len) pr_err("length: %u != %zu\n", *len, size); return -EINVAL; } - /* Protect against table resizer moving the entries. + /* Prevent modifications to the list with services. * Try reverse locking, so that we do not hold the mutex * while waiting for semaphore. */ @@ -4029,6 +4084,7 @@ static int ip_vs_genl_dump_services(struct sk_buff *skb, int start = cb->args[0]; int idx = 0; + /* Make sure we do not see same service twice during resize */ down_read(&ipvs->svc_resize_sem); rcu_read_lock(); ip_vs_rht_walk_buckets_safe_rcu(ipvs->svc_table, head) { @@ -5072,6 +5128,7 @@ int __net_init ip_vs_control_net_init(struct netns_ipvs *ipvs) /* Initialize service_mutex, svc_table per netns */ __mutex_init(&ipvs->service_mutex, "ipvs->service_mutex", &__ipvs_service_key); init_rwsem(&ipvs->svc_resize_sem); + init_rwsem(&ipvs->svc_replace_sem); INIT_DELAYED_WORK(&ipvs->svc_resize_work, svc_resize_work_handler); atomic_set(&ipvs->svc_table_changes, 0); RCU_INIT_POINTER(ipvs->svc_table, NULL); -- cgit v1.2.3 From 53d7fd878c28b28e03769071d1f28ef031a060ad Mon Sep 17 00:00:00 2001 From: Jozsef Kadlecsik Date: Thu, 14 May 2026 10:55:10 +0200 Subject: netfilter: ipset: fix a potential dump-destroy race When dumping sets in order to create the proper order for restore, the list type of sets dumped last. Therefore internally we run the dumping loop twice: first with all non-list type of sets and skipping the list type ones and then secondly for the list type of sets. Sashiko noticed that there's a potential race between dump and destroy if in the first loop the last set was a list type of set: its pointer remains unreferenced and a concurrent destroy can free it. Fix the issue by resetting the variable holding the pointer. Signed-off-by: Jozsef Kadlecsik Signed-off-by: Pablo Neira Ayuso --- net/netfilter/ipset/ip_set_core.c | 1 + 1 file changed, 1 insertion(+) (limited to 'net') diff --git a/net/netfilter/ipset/ip_set_core.c b/net/netfilter/ipset/ip_set_core.c index c5a26236a0bb..0874029cb0f2 100644 --- a/net/netfilter/ipset/ip_set_core.c +++ b/net/netfilter/ipset/ip_set_core.c @@ -1613,6 +1613,7 @@ dump_last: ((dump_type == DUMP_ALL) == !!(set->type->features & IPSET_DUMP_LAST))) { write_unlock_bh(&ip_set_ref_lock); + set = NULL; continue; } pr_debug("List set: %s\n", set->name); -- cgit v1.2.3 From b6a91f68ebfed9c38e0e9150f58a9b85da07181c Mon Sep 17 00:00:00 2001 From: Yizhou Zhao Date: Tue, 12 May 2026 01:30:41 +0800 Subject: netfilter: nft_inner: Fix IPv6 inner_thoff desync MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit In nft_inner_parse_l2l3(), when processing inner IPv6 packets, ipv6_find_hdr() correctly computes the transport header offset traversing all extension headers, but the result is immediately overwritten with nhoff + sizeof(_ip6h) (40 bytes), which only accounts for the IPv6 base header. This creates a desync between inner_thoff (wrong — points to extension header start) and l4proto (correct — e.g., IPPROTO_TCP), enabling transport header forgery and potential firewall bypass. This issue affects stable versions from Linux 6.2. For comparison, the normal (non-inner) IPv6 path correctly preserves ipv6_find_hdr()'s result. Removing the incorrect overwrite ensures that ipv6_find_hdr()'s calculated transport header offset is preserved, thereby fixing the desynchronization. Fixes: 3a07327d10a0 ("netfilter: nft_inner: support for inner tunnel header matching") Cc: stable@vger.kernel.org Reported-by: Yizhou Zhao Reported-by: Yuxiang Yang Reported-by: Xuewei Feng Reported-by: Qi Li Reported-by: Ke Xu Assisted-by: GLM:5.1 Z.ai Signed-off-by: Yizhou Zhao Reviewed-by: Fernando Fernandez Mancera Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nft_inner.c | 1 - 1 file changed, 1 deletion(-) (limited to 'net') diff --git a/net/netfilter/nft_inner.c b/net/netfilter/nft_inner.c index 03ffb1159fc1..859aa38e333b 100644 --- a/net/netfilter/nft_inner.c +++ b/net/netfilter/nft_inner.c @@ -163,7 +163,6 @@ static int nft_inner_parse_l2l3(const struct nft_inner *priv, return -1; if (fragoff == 0) { - thoff = nhoff + sizeof(_ip6h); ctx->flags |= NFT_PAYLOAD_CTX_INNER_TH; ctx->inner_thoff = thoff; ctx->l4proto = l4proto; -- cgit v1.2.3 From 0d3a282ab5f165fc207ff49ea5b6ad8f54616bd6 Mon Sep 17 00:00:00 2001 From: Nan Li Date: Tue, 12 May 2026 16:50:01 +0800 Subject: netfilter: ipset: stop hash:* range iteration at end The following hash set variants: hash:ip,mark hash:ip,port hash:ip,port,ip hash:ip,port,net iterate IPv4 ranges with a 32-bit iterator. The iterator must stop once the last address in the requested range has been processed. Advancing it once more can move the traversal state past the end of the request, so a later retry may continue from an unintended position. Handle the iterator increment explicitly at the end of the loop and stop once the upper bound has been processed. This keeps the existing retry behaviour intact for valid ranges while preventing traversal from continuing past the original boundary. Fixes: 48596a8ddc46 ("netfilter: ipset: Fix adding an IPv4 range containing more than 2^31 addresses") Cc: stable@kernel.org Reported-by: Yuan Tan Reported-by: Yifan Wu Reported-by: Juefei Pu Reported-by: Xin Liu Signed-off-by: Nan Li Signed-off-by: Ren Wei Signed-off-by: Pablo Neira Ayuso --- net/netfilter/ipset/ip_set_hash_ipmark.c | 6 +++++- net/netfilter/ipset/ip_set_hash_ipport.c | 5 ++++- net/netfilter/ipset/ip_set_hash_ipportip.c | 5 ++++- net/netfilter/ipset/ip_set_hash_ipportnet.c | 5 ++++- 4 files changed, 17 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/netfilter/ipset/ip_set_hash_ipmark.c b/net/netfilter/ipset/ip_set_hash_ipmark.c index a22ec1a6f6ec..e26ca2a370e3 100644 --- a/net/netfilter/ipset/ip_set_hash_ipmark.c +++ b/net/netfilter/ipset/ip_set_hash_ipmark.c @@ -150,7 +150,7 @@ hash_ipmark4_uadt(struct ip_set *set, struct nlattr *tb[], if (retried) ip = ntohl(h->next.ip); - for (; ip <= ip_to; ip++, i++) { + for (; ip <= ip_to; i++) { e.ip = htonl(ip); if (i > IPSET_MAX_RANGE) { hash_ipmark4_data_next(&h->next, &e); @@ -162,6 +162,10 @@ hash_ipmark4_uadt(struct ip_set *set, struct nlattr *tb[], return ret; ret = 0; + + if (ip == ip_to) + break; + ip++; } return ret; } diff --git a/net/netfilter/ipset/ip_set_hash_ipport.c b/net/netfilter/ipset/ip_set_hash_ipport.c index e977b5a9c48d..41ca24a22a02 100644 --- a/net/netfilter/ipset/ip_set_hash_ipport.c +++ b/net/netfilter/ipset/ip_set_hash_ipport.c @@ -186,7 +186,7 @@ hash_ipport4_uadt(struct ip_set *set, struct nlattr *tb[], if (retried) ip = ntohl(h->next.ip); - for (; ip <= ip_to; ip++) { + for (; ip <= ip_to;) { p = retried && ip == ntohl(h->next.ip) ? ntohs(h->next.port) : port; for (; p <= port_to; p++, i++) { @@ -203,6 +203,9 @@ hash_ipport4_uadt(struct ip_set *set, struct nlattr *tb[], ret = 0; } + if (ip == ip_to) + break; + ip++; } return ret; } diff --git a/net/netfilter/ipset/ip_set_hash_ipportip.c b/net/netfilter/ipset/ip_set_hash_ipportip.c index 39a01934b153..b9ac2efaa15c 100644 --- a/net/netfilter/ipset/ip_set_hash_ipportip.c +++ b/net/netfilter/ipset/ip_set_hash_ipportip.c @@ -182,7 +182,7 @@ hash_ipportip4_uadt(struct ip_set *set, struct nlattr *tb[], if (retried) ip = ntohl(h->next.ip); - for (; ip <= ip_to; ip++) { + for (; ip <= ip_to;) { p = retried && ip == ntohl(h->next.ip) ? ntohs(h->next.port) : port; for (; p <= port_to; p++, i++) { @@ -199,6 +199,9 @@ hash_ipportip4_uadt(struct ip_set *set, struct nlattr *tb[], ret = 0; } + if (ip == ip_to) + break; + ip++; } return ret; } diff --git a/net/netfilter/ipset/ip_set_hash_ipportnet.c b/net/netfilter/ipset/ip_set_hash_ipportnet.c index 5c6de605a9fb..2d6652d43199 100644 --- a/net/netfilter/ipset/ip_set_hash_ipportnet.c +++ b/net/netfilter/ipset/ip_set_hash_ipportnet.c @@ -274,7 +274,7 @@ hash_ipportnet4_uadt(struct ip_set *set, struct nlattr *tb[], p = port; ip2 = ip2_from; } - for (; ip <= ip_to; ip++) { + for (; ip <= ip_to;) { e.ip = htonl(ip); for (; p <= port_to; p++) { e.port = htons(p); @@ -298,6 +298,9 @@ hash_ipportnet4_uadt(struct ip_set *set, struct nlattr *tb[], ip2 = ip2_from; } p = port; + if (ip == ip_to) + break; + ip++; } return ret; } -- cgit v1.2.3 From a6cb3ff979855f7f0ee9450a947fe8f96c2ba37a Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Tue, 12 May 2026 11:30:49 +0200 Subject: netfilter: nft_inner: release local_lock before re-enabling softirqs Quoting sashiko: In the error path, local_bh_enable() is called before local_unlock_nested_bh(). Fixes: ba36fada9ab4 ("netfilter: nft_inner: Use nested-BH locking for nft_pcpu_tun_ctx") Signed-off-by: Florian Westphal Reviewed-by: Fernando Fernandez Mancera Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nft_inner.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/netfilter/nft_inner.c b/net/netfilter/nft_inner.c index 859aa38e333b..d14ca157910b 100644 --- a/net/netfilter/nft_inner.c +++ b/net/netfilter/nft_inner.c @@ -246,8 +246,8 @@ static bool nft_inner_restore_tun_ctx(const struct nft_pktinfo *pkt, local_lock_nested_bh(&nft_pcpu_tun_ctx.bh_lock); this_cpu_tun_ctx = this_cpu_ptr(&nft_pcpu_tun_ctx.ctx); if (this_cpu_tun_ctx->cookie != (unsigned long)pkt->skb) { - local_bh_enable(); local_unlock_nested_bh(&nft_pcpu_tun_ctx.bh_lock); + local_bh_enable(); return false; } *tun_ctx = *this_cpu_tun_ctx; -- cgit v1.2.3 From 4322dcde6b4173c2d8e8e6118ed290794263bcc8 Mon Sep 17 00:00:00 2001 From: Zhengchuan Liang Date: Wed, 13 May 2026 15:57:17 +0800 Subject: netfilter: ip6t_hbh: reject oversized option lists struct ip6t_opts stores at most IP6T_OPTS_OPTSNR option descriptors, but hbh_mt6_check() does not reject larger optsnr values supplied from userspace. Validate optsnr in the rule setup path so only match data that fits the fixed-size opts array can be installed. This follows the existing xtables pattern of rejecting invalid user-provided counts in checkentry() and keeps the packet matching path unchanged. `struct ip6t_opts` has a fixed `opts[IP6T_OPTS_OPTSNR]` array, where `IP6T_OPTS_OPTSNR` is 16, then off-by-one array access is possible: [ 137.924693][ T8692] UBSAN: array-index-out-of-bounds in ../net/ipv6/netfilter/ip6t_hbh.c:110:29 [ 137.926167][ T8692] index 16 is out of range for type '__u16 [16]' Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Cc: stable@kernel.org Reported-by: Yuan Tan Reported-by: Yifan Wu Reported-by: Juefei Pu Reported-by: Xin Liu Signed-off-by: Zhengchuan Liang Signed-off-by: Ren Wei Signed-off-by: Pablo Neira Ayuso --- net/ipv6/netfilter/ip6t_hbh.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'net') diff --git a/net/ipv6/netfilter/ip6t_hbh.c b/net/ipv6/netfilter/ip6t_hbh.c index e7a3fb9355ee..450dd53846a2 100644 --- a/net/ipv6/netfilter/ip6t_hbh.c +++ b/net/ipv6/netfilter/ip6t_hbh.c @@ -168,6 +168,10 @@ static int hbh_mt6_check(const struct xt_mtchk_param *par) pr_debug("unknown flags %X\n", optsinfo->invflags); return -EINVAL; } + if (optsinfo->optsnr > IP6T_OPTS_OPTSNR) { + pr_debug("too many supported opts specified\n"); + return -EINVAL; + } if (optsinfo->flags & IP6T_OPTS_NSTRICT) { pr_debug("Not strict - not implemented"); -- cgit v1.2.3 From c0c42a0fb27144c1cd7509f94bec0d3bcca98c72 Mon Sep 17 00:00:00 2001 From: Jozsef Kadlecsik Date: Thu, 14 May 2026 10:55:11 +0200 Subject: netfilter: ipset: Fix data race between add and list header in all hash types The "ipset list -terse" command is actually a dump operation which may run parallel with "ipset add" commands, which can trigger an internal resizing of the hash type of sets just being dumped. However, dumping just the header part of the set was not protected against underlying resizing. Fix it by protecting the header dumping part as well. Fixes: c4c997839cf9 ("netfilter: ipset: Fix parallel resizing and listing of the same set") Signed-off-by: Jozsef Kadlecsik Signed-off-by: Pablo Neira Ayuso --- net/netfilter/ipset/ip_set_core.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/netfilter/ipset/ip_set_core.c b/net/netfilter/ipset/ip_set_core.c index 0874029cb0f2..3706b4a85a0f 100644 --- a/net/netfilter/ipset/ip_set_core.c +++ b/net/netfilter/ipset/ip_set_core.c @@ -1649,13 +1649,13 @@ dump_last: if (cb->args[IPSET_CB_PROTO] > IPSET_PROTOCOL_MIN && nla_put_net16(skb, IPSET_ATTR_INDEX, htons(index))) goto nla_put_failure; + if (set->variant->uref) + set->variant->uref(set, cb, true); ret = set->variant->head(set, skb); if (ret < 0) goto release_refcount; if (dump_flags & IPSET_FLAG_LIST_HEADER) goto next_set; - if (set->variant->uref) - set->variant->uref(set, cb, true); fallthrough; default: ret = set->variant->list(set, skb, cb); -- cgit v1.2.3 From 2358f7427ccd6ec8867a48205d8fcec973683a3f Mon Sep 17 00:00:00 2001 From: Jozsef Kadlecsik Date: Fri, 8 May 2026 22:58:58 +0200 Subject: netfilter: ipset: Fix data race between add and dump in all hash types When adding a new entry to the next position in the existing hash bucket, the position index was incremented too early and parallel dump could read it before the entry was populated with the value. Move the setting of the position index after populating the entry. v2: Position counting fixed, noticed by Florian Westphal. Fixes: 18f84d41d34f ("netfilter: ipset: Introduce RCU locking in hash:* types") Reported-by: syzbot+786c889f046e8b003ca6@syzkaller.appspotmail.com Reported-by: syzbot+1da17e4b41d795df059e@syzkaller.appspotmail.com Reported-by: syzbot+421c5f3ff8e9493084d9@syzkaller.appspotmail.com Signed-off-by: Jozsef Kadlecsik Signed-off-by: Pablo Neira Ayuso --- net/netfilter/ipset/ip_set_hash_gen.h | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/netfilter/ipset/ip_set_hash_gen.h b/net/netfilter/ipset/ip_set_hash_gen.h index b79e5dd2af03..133ce4611eed 100644 --- a/net/netfilter/ipset/ip_set_hash_gen.h +++ b/net/netfilter/ipset/ip_set_hash_gen.h @@ -844,7 +844,7 @@ mtype_add(struct ip_set *set, void *value, const struct ip_set_ext *ext, const struct mtype_elem *d = value; struct mtype_elem *data; struct hbucket *n, *old = ERR_PTR(-ENOENT); - int i, j = -1, ret; + int i, j = -1, npos = 0, ret; bool flag_exist = flags & IPSET_FLAG_EXIST; bool deleted = false, forceadd = false, reuse = false; u32 r, key, multi = 0, elements, maxelem; @@ -889,6 +889,7 @@ mtype_add(struct ip_set *set, void *value, const struct ip_set_ext *ext, ext_size(AHASH_INIT_SIZE, set->dsize); goto copy_elem; } + npos = n->pos; for (i = 0; i < n->pos; i++) { if (!test_bit(i, n->used)) { /* Reuse first deleted entry */ @@ -962,7 +963,8 @@ mtype_add(struct ip_set *set, void *value, const struct ip_set_ext *ext, } copy_elem: - j = n->pos++; + j = npos; + npos = n->pos + 1; data = ahash_data(n, j, set->dsize); copy_data: t->hregion[r].elements++; @@ -985,6 +987,7 @@ overwrite_extensions: if (SET_WITH_TIMEOUT(set)) ip_set_timeout_set(ext_timeout(data, set), ext->timeout); smp_mb__before_atomic(); + n->pos = npos; set_bit(j, n->used); if (old != ERR_PTR(-ENOENT)) { rcu_assign_pointer(hbucket(t, key), n); -- cgit v1.2.3 From 7f7445840b7771338618930e45ee641104b38ed8 Mon Sep 17 00:00:00 2001 From: Jozsef Kadlecsik Date: Thu, 14 May 2026 10:55:13 +0200 Subject: netfilter: ipset: annotate "pos" for concurrent readers/writers The "pos" structure member of struct hbucket stores the first free slot in the hash bucket of a hash type of set and there are concurrent readers/writers. Annotate accesses properly. Fixes: 18f84d41d34f ("netfilter: ipset: Introduce RCU locking in hash:* types") Signed-off-by: Jozsef Kadlecsik Signed-off-by: Pablo Neira Ayuso --- net/netfilter/ipset/ip_set_hash_gen.h | 62 +++++++++++++++++++++-------------- 1 file changed, 38 insertions(+), 24 deletions(-) (limited to 'net') diff --git a/net/netfilter/ipset/ip_set_hash_gen.h b/net/netfilter/ipset/ip_set_hash_gen.h index 133ce4611eed..04e4627ddfc1 100644 --- a/net/netfilter/ipset/ip_set_hash_gen.h +++ b/net/netfilter/ipset/ip_set_hash_gen.h @@ -386,8 +386,9 @@ static void mtype_ext_cleanup(struct ip_set *set, struct hbucket *n) { int i; + u8 pos = smp_load_acquire(&n->pos); - for (i = 0; i < n->pos; i++) + for (i = 0; i < pos; i++) if (test_bit(i, n->used)) ip_set_ext_destroy(set, ahash_data(n, i, set->dsize)); } @@ -490,7 +491,7 @@ mtype_gc_do(struct ip_set *set, struct htype *h, struct htable *t, u32 r) #ifdef IP_SET_HASH_WITH_NETS u8 k; #endif - u8 htable_bits = t->htable_bits; + u8 pos, htable_bits = t->htable_bits; spin_lock_bh(&t->hregion[r].lock); for (i = ahash_bucket_start(r, htable_bits); @@ -498,7 +499,8 @@ mtype_gc_do(struct ip_set *set, struct htype *h, struct htable *t, u32 r) n = __ipset_dereference(hbucket(t, i)); if (!n) continue; - for (j = 0, d = 0; j < n->pos; j++) { + pos = smp_load_acquire(&n->pos); + for (j = 0, d = 0; j < pos; j++) { if (!test_bit(j, n->used)) { d++; continue; @@ -534,7 +536,7 @@ mtype_gc_do(struct ip_set *set, struct htype *h, struct htable *t, u32 r) /* Still try to delete expired elements. */ continue; tmp->size = n->size - AHASH_INIT_SIZE; - for (j = 0, d = 0; j < n->pos; j++) { + for (j = 0, d = 0; j < pos; j++) { if (!test_bit(j, n->used)) continue; data = ahash_data(n, j, dsize); @@ -623,7 +625,7 @@ mtype_resize(struct ip_set *set, bool retried) { struct htype *h = set->data; struct htable *t, *orig; - u8 htable_bits; + u8 pos, htable_bits; size_t hsize, dsize = set->dsize; #ifdef IP_SET_HASH_WITH_NETS u8 flags; @@ -685,7 +687,8 @@ retry: n = __ipset_dereference(hbucket(orig, i)); if (!n) continue; - for (j = 0; j < n->pos; j++) { + pos = smp_load_acquire(&n->pos); + for (j = 0; j < pos; j++) { if (!test_bit(j, n->used)) continue; data = ahash_data(n, j, dsize); @@ -809,9 +812,10 @@ mtype_ext_size(struct ip_set *set, u32 *elements, size_t *ext_size) { struct htype *h = set->data; const struct htable *t; - u32 i, j, r; struct hbucket *n; struct mtype_elem *data; + u32 i, j, r; + u8 pos; t = rcu_dereference_bh(h->table); for (r = 0; r < ahash_numof_locks(t->htable_bits); r++) { @@ -820,7 +824,8 @@ mtype_ext_size(struct ip_set *set, u32 *elements, size_t *ext_size) n = rcu_dereference_bh(hbucket(t, i)); if (!n) continue; - for (j = 0; j < n->pos; j++) { + pos = smp_load_acquire(&n->pos); + for (j = 0; j < pos; j++) { if (!test_bit(j, n->used)) continue; data = ahash_data(n, j, set->dsize); @@ -844,10 +849,11 @@ mtype_add(struct ip_set *set, void *value, const struct ip_set_ext *ext, const struct mtype_elem *d = value; struct mtype_elem *data; struct hbucket *n, *old = ERR_PTR(-ENOENT); - int i, j = -1, npos = 0, ret; + int i, j = -1, ret; bool flag_exist = flags & IPSET_FLAG_EXIST; bool deleted = false, forceadd = false, reuse = false; u32 r, key, multi = 0, elements, maxelem; + u8 npos = 0; rcu_read_lock_bh(); t = rcu_dereference_bh(h->table); @@ -889,8 +895,8 @@ mtype_add(struct ip_set *set, void *value, const struct ip_set_ext *ext, ext_size(AHASH_INIT_SIZE, set->dsize); goto copy_elem; } - npos = n->pos; - for (i = 0; i < n->pos; i++) { + npos = smp_load_acquire(&n->pos); + for (i = 0; i < npos; i++) { if (!test_bit(i, n->used)) { /* Reuse first deleted entry */ if (j == -1) { @@ -934,7 +940,7 @@ mtype_add(struct ip_set *set, void *value, const struct ip_set_ext *ext, if (elements >= maxelem) goto set_full; /* Create a new slot */ - if (n->pos >= n->size) { + if (npos >= n->size) { #ifdef IP_SET_HASH_WITH_MULTI if (h->bucketsize >= AHASH_MAX_TUNED) goto set_full; @@ -963,8 +969,7 @@ mtype_add(struct ip_set *set, void *value, const struct ip_set_ext *ext, } copy_elem: - j = npos; - npos = n->pos + 1; + j = npos++; data = ahash_data(n, j, set->dsize); copy_data: t->hregion[r].elements++; @@ -987,7 +992,8 @@ overwrite_extensions: if (SET_WITH_TIMEOUT(set)) ip_set_timeout_set(ext_timeout(data, set), ext->timeout); smp_mb__before_atomic(); - n->pos = npos; + /* Ensure all data writes are visible before updating position */ + smp_store_release(&n->pos, npos); set_bit(j, n->used); if (old != ERR_PTR(-ENOENT)) { rcu_assign_pointer(hbucket(t, key), n); @@ -1046,6 +1052,7 @@ mtype_del(struct ip_set *set, void *value, const struct ip_set_ext *ext, int i, j, k, r, ret = -IPSET_ERR_EXIST; u32 key, multi = 0; size_t dsize = set->dsize; + u8 pos; /* Userspace add and resize is excluded by the mutex. * Kernespace add does not trigger resize. @@ -1061,7 +1068,8 @@ mtype_del(struct ip_set *set, void *value, const struct ip_set_ext *ext, n = rcu_dereference_bh(hbucket(t, key)); if (!n) goto out; - for (i = 0, k = 0; i < n->pos; i++) { + pos = smp_load_acquire(&n->pos); + for (i = 0, k = 0; i < pos; i++) { if (!test_bit(i, n->used)) { k++; continue; @@ -1075,8 +1083,8 @@ mtype_del(struct ip_set *set, void *value, const struct ip_set_ext *ext, ret = 0; clear_bit(i, n->used); smp_mb__after_atomic(); - if (i + 1 == n->pos) - n->pos--; + if (i + 1 == pos) + smp_store_release(&n->pos, --pos); t->hregion[r].elements--; #ifdef IP_SET_HASH_WITH_NETS for (j = 0; j < IPSET_NET_COUNT; j++) @@ -1097,11 +1105,11 @@ mtype_del(struct ip_set *set, void *value, const struct ip_set_ext *ext, x->flags = flags; } } - for (; i < n->pos; i++) { + for (; i < pos; i++) { if (!test_bit(i, n->used)) k++; } - if (k == n->pos) { + if (k == pos) { t->hregion[r].ext_size -= ext_size(n->size, dsize); rcu_assign_pointer(hbucket(t, key), NULL); kfree_rcu(n, rcu); @@ -1112,7 +1120,7 @@ mtype_del(struct ip_set *set, void *value, const struct ip_set_ext *ext, if (!tmp) goto out; tmp->size = n->size - AHASH_INIT_SIZE; - for (j = 0, k = 0; j < n->pos; j++) { + for (j = 0, k = 0; j < pos; j++) { if (!test_bit(j, n->used)) continue; data = ahash_data(n, j, dsize); @@ -1173,6 +1181,7 @@ mtype_test_cidrs(struct ip_set *set, struct mtype_elem *d, int ret, i, j = 0; #endif u32 key, multi = 0; + u8 pos; pr_debug("test by nets\n"); for (; j < NLEN && h->nets[j].cidr[0] && !multi; j++) { @@ -1190,7 +1199,8 @@ mtype_test_cidrs(struct ip_set *set, struct mtype_elem *d, n = rcu_dereference_bh(hbucket(t, key)); if (!n) continue; - for (i = 0; i < n->pos; i++) { + pos = smp_load_acquire(&n->pos); + for (i = 0; i < pos; i++) { if (!test_bit(i, n->used)) continue; data = ahash_data(n, i, set->dsize); @@ -1224,6 +1234,7 @@ mtype_test(struct ip_set *set, void *value, const struct ip_set_ext *ext, struct mtype_elem *data; int i, ret = 0; u32 key, multi = 0; + u8 pos; rcu_read_lock_bh(); t = rcu_dereference_bh(h->table); @@ -1246,7 +1257,8 @@ mtype_test(struct ip_set *set, void *value, const struct ip_set_ext *ext, ret = 0; goto out; } - for (i = 0; i < n->pos; i++) { + pos = smp_load_acquire(&n->pos); + for (i = 0; i < pos; i++) { if (!test_bit(i, n->used)) continue; data = ahash_data(n, i, set->dsize); @@ -1363,6 +1375,7 @@ mtype_list(const struct ip_set *set, /* We assume that one hash bucket fills into one page */ void *incomplete; int i, ret = 0; + u8 pos; atd = nla_nest_start(skb, IPSET_ATTR_ADT); if (!atd) @@ -1381,7 +1394,8 @@ mtype_list(const struct ip_set *set, cb->args[IPSET_CB_ARG0], t, n); if (!n) continue; - for (i = 0; i < n->pos; i++) { + pos = smp_load_acquire(&n->pos); + for (i = 0; i < pos; i++) { if (!test_bit(i, n->used)) continue; e = ahash_data(n, i, set->dsize); -- cgit v1.2.3 From b2870fc21601db9133bc70c48c603b487614fa3b Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Thu, 14 May 2026 16:46:38 +0200 Subject: netfilter: br_netfilter: Reallocate headroom if necessary in neigh_hh_bridge() neigh_hh_bridge() assumes the skb always has sufficient headroom to copy the aligned L2 header. This assumption can trigger the crash reported below using the following netfilter setup: $modprobe br_netfilter $sysctl -w net.bridge.bridge-nf-call-iptables=1 $root@OpenWrt:~# nft list ruleset table ip nat { chain prerouting { type nat hook prerouting priority dstnat; policy accept; ip daddr 192.168.83.123 dnat to 192.168.83.120 } } - iperf3 client (192.168.83.119) --> bridge (192.168.83.118) --> iperf3 server (192.168.83.120) the iperf3 client is sending packet for 192.168.83.123 to the bridge device. [ 1579.036575] Unable to handle kernel write to read-only memory at virtual address ffffff8004d76ffe [ 1579.045482] Mem abort info: [ 1579.048273] ESR = 0x000000009600004f [ 1579.052024] EC = 0x25: DABT (current EL), IL = 32 bits [ 1579.057363] SET = 0, FnV = 0 [ 1579.060417] EA = 0, S1PTW = 0 [ 1579.063550] FSC = 0x0f: level 3 permission fault [ 1579.068345] Data abort info: [ 1579.071224] ISV = 0, ISS = 0x0000004f, ISS2 = 0x00000000 [ 1579.076720] CM = 0, WnR = 1, TnD = 0, TagAccess = 0 [ 1579.081770] GCS = 0, Overlay = 0, DirtyBit = 0, Xs = 0 [ 1579.087092] swapper pgtable: 4k pages, 39-bit VAs, pgdp=0000000080dc4000 [ 1579.093794] [ffffff8004d76ffe] pgd=180000009ffff003, p4d=180000009ffff003, pud=180000009ffff003, pmd=180000009ffe3003, pte=0060000084d76787 [ 1579.106343] Internal error: Oops: 000000009600004f [#1] SMP [ 1579.193824] CPU: 0 UID: 0 PID: 235 Comm: napi/qdma_eth-3 Tainted: G O 6.12.57 #0 [ 1579.202614] Tainted: [O]=OOT_MODULE [ 1579.206102] Hardware name: Airoha AN7581 Evaluation Board (DT) [ 1579.211929] pstate: 60400005 (nZCv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--) [ 1579.218889] pc : br_nf_pre_routing_finish_bridge+0x1ac/0xcc8 [br_netfilter] [ 1579.225859] lr : br_nf_pre_routing_finish_bridge+0x18c/0xcc8 [br_netfilter] [ 1579.232822] sp : ffffffc0817cba20 [ 1579.236128] x29: ffffffc0817cba20 x28: 0000000000000000 x27: ffffff8002b89000 [ 1579.243273] x26: ffffff8004d7700e x25: 0000000000000008 x24: 0000000000000000 [ 1579.250416] x23: ffffffc08179d4c0 x22: 0000000000000000 x21: ffffffc08179d4c0 [ 1579.257561] x20: ffffff8004d9b800 x19: ffffff8015010000 x18: 0000000000000014 [ 1579.264704] x17: ffffffbf9e930000 x16: ffffffc0817c8000 x15: 0000000000000070 [ 1579.271848] x14: 0000000000000080 x13: 0000000000000001 x12: 0000000000000000 [ 1579.278993] x11: ffffffc0798caae0 x10: ffffff8014db6fd8 x9 : 0000000000000000 [ 1579.286136] x8 : 0000000000000003 x7 : ffffffc08171f628 x6 : 000000001a3b83d3 [ 1579.293281] x5 : 0000000000000000 x4 : 1beb76f22fee0000 x3 : ffffff8004d7700e [ 1579.300425] x2 : 0000000000000000 x1 : ffffff8004d9b8bc x0 : ffffff80026ed000 [ 1579.307570] Call trace: [ 1579.310018] br_nf_pre_routing_finish_bridge+0x1ac/0xcc8 [br_netfilter] [ 1579.316632] br_nf_hook_thresh+0xd4/0x14bc [br_netfilter] [ 1579.322032] br_nf_hook_thresh+0x250/0x14bc [br_netfilter] [ 1579.327517] br_nf_hook_thresh+0x76c/0x14bc [br_netfilter] [ 1579.333003] br_handle_frame+0x180/0x480 [ 1579.336935] __netif_receive_skb_core.constprop.0+0x540/0xf40 [ 1579.342682] __netif_receive_skb_one_core+0x28/0x50 [ 1579.347561] process_backlog+0x98/0x1e0 [ 1579.351398] __napi_poll+0x34/0x1c4 [ 1579.354887] net_rx_action+0x178/0x330 [ 1579.358638] handle_softirqs+0x108/0x2d4 [ 1579.362560] __do_softirq+0x10/0x18 [ 1579.366051] ____do_softirq+0xc/0x20 [ 1579.369627] call_on_irq_stack+0x30/0x4c [ 1579.373550] do_softirq_own_stack+0x18/0x20 [ 1579.377734] do_softirq+0x4c/0x60 [ 1579.381050] __local_bh_enable_ip+0x88/0x98 [ 1579.385234] napi_threaded_poll_loop+0x188/0x21c [ 1579.389853] napi_threaded_poll+0x70/0x80 [ 1579.393863] kthread+0xd8/0xdc [ 1579.396918] ret_from_fork+0x10/0x20 [ 1579.400499] Code: 88dffc22 3707ffc2 f9406663 f9406684 (f81f0064) [ 1579.406589] ---[ end trace 0000000000000000 ]--- [ 1579.411209] Kernel panic - not syncing: Oops: Fatal exception in interrupt [ 1579.418083] SMP: stopping secondary CPUs [ 1579.422012] Kernel Offset: disabled Fix the issue reallocating the skb headroom if necessary in neigh_hh_bridge routine. Fixes: e179e6322ac33 ("netfilter: bridge-netfilter: Fix MAC header handling with IP DNAT") Reviewed-by: Ido Schimmel Signed-off-by: Lorenzo Bianconi Signed-off-by: Pablo Neira Ayuso --- net/bridge/br_netfilter_hooks.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/bridge/br_netfilter_hooks.c b/net/bridge/br_netfilter_hooks.c index 0ab1c94db4b9..0a394e5f4391 100644 --- a/net/bridge/br_netfilter_hooks.c +++ b/net/bridge/br_netfilter_hooks.c @@ -297,7 +297,11 @@ int br_nf_pre_routing_finish_bridge(struct net *net, struct sock *sk, struct sk_ goto free_skb; } - neigh_hh_bridge(&neigh->hh, skb); + if (neigh_hh_bridge(&neigh->hh, skb)) { + neigh_release(neigh); + goto free_skb; + } + skb->dev = br_indev; ret = br_handle_frame_finish(net, sk, skb); -- cgit v1.2.3 From e196115ec330a18de415bdb9f5071aa9f08e53ce Mon Sep 17 00:00:00 2001 From: Haoze Xie Date: Fri, 15 May 2026 11:19:02 +0800 Subject: netfilter: nf_queue: hold bridge skb->dev while queued br_pass_frame_up() rewrites skb->dev from the ingress port to the bridge master before queueing bridge LOCAL_IN packets. NFQUEUE only holds references on state.in/out and bridge physdevs, so a queued bridge packet can retain a freed bridge master in skb->dev until reinjection. When the verdict is reinjected later, br_netif_receive_skb() re-enters the receive path with skb->dev still pointing at the freed bridge master, triggering a use-after-free. Store skb->dev in the queue entry, hold a reference on it for the queue lifetime, and use the saved device when dropping queued packets during NETDEV_DOWN handling. Fixes: ac2863445686 ("netfilter: bridge: add nf_afinfo to enable queuing to userspace") Cc: stable@kernel.org Reported-by: Yuan Tan Reported-by: Yifan Wu Reported-by: Juefei Pu Reported-by: Xin Liu Signed-off-by: Haoze Xie Signed-off-by: Ren Wei Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_queue.c | 4 +++- net/netfilter/nfnetlink_queue.c | 2 ++ 2 files changed, 5 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/netfilter/nf_queue.c b/net/netfilter/nf_queue.c index a6c81c04b3a5..57b450024a99 100644 --- a/net/netfilter/nf_queue.c +++ b/net/netfilter/nf_queue.c @@ -61,6 +61,7 @@ static void nf_queue_entry_release_refs(struct nf_queue_entry *entry) struct nf_hook_state *state = &entry->state; /* Release those devices we held, or Alexey will kill me. */ + dev_put(entry->skb_dev); dev_put(state->in); dev_put(state->out); if (state->sk) @@ -102,6 +103,7 @@ bool nf_queue_entry_get_refs(struct nf_queue_entry *entry) if (state->sk && !refcount_inc_not_zero(&state->sk->sk_refcnt)) return false; + dev_hold(entry->skb_dev); dev_hold(state->in); dev_hold(state->out); @@ -202,11 +204,11 @@ static int __nf_queue(struct sk_buff *skb, const struct nf_hook_state *state, *entry = (struct nf_queue_entry) { .skb = skb, + .skb_dev = skb->dev, .state = *state, .hook_index = index, .size = sizeof(*entry) + route_key_size, }; - __nf_queue_entry_init_physdevs(entry); if (!nf_queue_entry_get_refs(entry)) { diff --git a/net/netfilter/nfnetlink_queue.c b/net/netfilter/nfnetlink_queue.c index 58304fd1f70f..984a0eb9e149 100644 --- a/net/netfilter/nfnetlink_queue.c +++ b/net/netfilter/nfnetlink_queue.c @@ -1212,6 +1212,8 @@ dev_cmp(struct nf_queue_entry *entry, unsigned long ifindex) if (physinif == ifindex || physoutif == ifindex) return 1; #endif + if (entry->skb_dev && entry->skb_dev->ifindex == ifindex) + return 1; if (entry->state.in) if (entry->state.in->ifindex == ifindex) return 1; -- cgit v1.2.3 From 0039ac8305064e455f04d412ec3896c4fe41d04f Mon Sep 17 00:00:00 2001 From: Sven Eckelmann Date: Sat, 16 May 2026 22:10:08 +0200 Subject: batman-adv: fix batadv_skb_is_frag() kernel-doc The kernel-doc comment for batadv_skb_is_frag() contained two errors: * the function description referred to "gain a unicast packet" instead of "contains unicast fragment". * the Return section omitted "merged" from "newly skb", leaving the description grammatically incorrect and inconsistent with the function description. Fixes: bc62216dc8e2 ("batman-adv: frag: disallow unicast fragment in fragment") Signed-off-by: Sven Eckelmann --- net/batman-adv/fragmentation.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/batman-adv/fragmentation.c b/net/batman-adv/fragmentation.c index 4a594aa2ebf6..e9553db42349 100644 --- a/net/batman-adv/fragmentation.c +++ b/net/batman-adv/fragmentation.c @@ -305,10 +305,10 @@ free: } /** - * batadv_skb_is_frag() - check if newly merged skb is gain a unicast packet + * batadv_skb_is_frag() - check if newly merged skb contains unicast fragment * @skb: newly merged skb * - * Return: if newly skb is of type BATADV_UNICAST_FRAG + * Return: if newly merged skb is of type BATADV_UNICAST_FRAG */ static bool batadv_skb_is_frag(struct sk_buff *skb) { -- cgit v1.2.3 From 9e7f36ab5b7bf68463faa5f7b926fea8f35597bb Mon Sep 17 00:00:00 2001 From: Weiming Shi Date: Thu, 14 May 2026 05:38:08 -0700 Subject: net: appletalk: fix NULL pointer dereference in aarp_send_ddp() aarp_send_ddp() calls atalk_find_dev_addr(dev) in the LocalTalk fast path without checking for NULL. When the device has no AppleTalk interface configured (dev->atalk_ptr == NULL), this leads to a NULL pointer dereference at the at->s_net access. KASAN: null-ptr-deref in range [0x0000000000000000-0x0000000000000007] RIP: 0010:aarp_send_ddp (net/appletalk/aarp.c:552 (discriminator 2)) Call Trace: atalk_sendmsg (net/appletalk/ddp.c:1715) __sys_sendto (net/socket.c:2265 (discriminator 1)) __x64_sys_sendto (net/socket.c:2272) do_syscall_64 (arch/x86/entry/syscall_64.c:94) entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:121) Add a NULL check consistent with the other callers of atalk_find_dev_addr(). Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Reported-by: Xiang Mei Signed-off-by: Weiming Shi Link: https://patch.msgid.link/20260514123806.3085961-3-bestswngs@gmail.com Signed-off-by: Jakub Kicinski --- net/appletalk/aarp.c | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'net') diff --git a/net/appletalk/aarp.c b/net/appletalk/aarp.c index e7315c01a299..30493ea3c010 100644 --- a/net/appletalk/aarp.c +++ b/net/appletalk/aarp.c @@ -542,6 +542,11 @@ int aarp_send_ddp(struct net_device *dev, struct sk_buff *skb, struct ddpehdr *ddp = (struct ddpehdr *)skb->data; int ft = 2; + if (!at) { + kfree_skb(skb); + return NET_XMIT_DROP; + } + /* * Compressible ? * -- cgit v1.2.3 From bae3ee802c21e83ad1eb805519e6f32ea528b4d2 Mon Sep 17 00:00:00 2001 From: Ilya Maximets Date: Thu, 14 May 2026 20:46:31 +0200 Subject: openvswitch: vport: fix race between linking and the device notifier Sashiko reports that it is technically possible that we got the device reference, but by the time we're linking it to the OVS datapath, it may be already in the process of being deleted. In this case if the notifier wins the race for RTNL, it will see that the device is not yet in the OVS datapath (ovs_netdev_get_vport() will fail in the dp_device_event()) and will do nothing. Then the ovs_netdev_link() will take the RTNL and link the unregistering device to OVS datapath. Eventually, netdev_wait_allrefs_any() will re-broadcast the event and the device will be properly detached, but it will take at least a second before that happens, so it's not something we should rely on. Let's avoid linking the non-registered device in the first place. Note: As per documentation, RTNL doesn't protect the reg_state, but it actually does for all the state transitions we care about here, so it should not be necessary to use READ_ONCE or taking the instance lock. We can still do that, but we have a few more places even in this file where the reg_state is accessed without those while under RTNL, and many more places like this across the kernel code, so it might make more sense to change all of them in a more centralized fashion in the future, if necessary. Fixes: ccb1352e76cf ("net: Add Open vSwitch kernel components.") Signed-off-by: Ilya Maximets Reviewed-by: Aaron Conole Acked-by: Eelco Chaudron Link: https://patch.msgid.link/20260514184702.2461435-1-i.maximets@ovn.org Signed-off-by: Jakub Kicinski --- net/openvswitch/vport-netdev.c | 8 ++++++++ 1 file changed, 8 insertions(+) (limited to 'net') diff --git a/net/openvswitch/vport-netdev.c b/net/openvswitch/vport-netdev.c index c42642075685..e7e8490a53d8 100644 --- a/net/openvswitch/vport-netdev.c +++ b/net/openvswitch/vport-netdev.c @@ -83,6 +83,14 @@ struct vport *ovs_netdev_link(struct vport *vport, bool tunnel) } rtnl_lock(); + /* Do not link devices that are not registered to avoid a potential + * race with the NETDEV_UNREGISTER notification in dp_device_event(). + */ + if (vport->dev->reg_state != NETREG_REGISTERED) { + err = -ENODEV; + goto error_put_unlock; + } + err = netdev_master_upper_dev_link(vport->dev, get_dpdev(vport->dp), NULL, NULL, NULL); -- cgit v1.2.3 From f8ce8b8331a1bc44ad4905886a482214d428b253 Mon Sep 17 00:00:00 2001 From: Sven Eckelmann Date: Sat, 9 May 2026 22:44:12 +0200 Subject: batman-adv: v: stop OGMv2 on disabled interface When a batadv_hard_iface is disabled, its mesh_iface pointer is set to NULL. However, batadv_v_ogm_send_meshif() may still dispatch OGMs via batadv_v_ogm_queue_on_if() for interfaces that have since lost their mesh_iface association. This results in a NULL pointer dereference when batadv_v_ogm_queue_on_if() unconditionally calls netdev_priv() on the now NULL hard_iface->mesh_iface to retrieve the batadv_priv. It is necessary to ensure that the batadv_v_ogm_queue_on_if() checks that it is using the same mesh_iface for which batadv_v_ogm_send_meshif() was called. Cc: stable@kernel.org Fixes: 0da0035942d4 ("batman-adv: OGMv2 - add basic infrastructure") Reported-by: Yuan Tan Reported-by: Yifan Wu Reported-by: Juefei Pu Reported-by: Xin Liu Reviewed-by: Yuan Tan Signed-off-by: Sven Eckelmann --- net/batman-adv/bat_v_ogm.c | 33 +++++++++++++++++++++------------ 1 file changed, 21 insertions(+), 12 deletions(-) (limited to 'net') diff --git a/net/batman-adv/bat_v_ogm.c b/net/batman-adv/bat_v_ogm.c index e3870492dab7..e955b4940c72 100644 --- a/net/batman-adv/bat_v_ogm.c +++ b/net/batman-adv/bat_v_ogm.c @@ -113,14 +113,14 @@ static void batadv_v_ogm_start_timer(struct batadv_priv *bat_priv) /** * batadv_v_ogm_send_to_if() - send a batman ogm using a given interface + * @bat_priv: the bat priv with all the mesh interface information * @skb: the OGM to send * @hard_iface: the interface to use to send the OGM */ -static void batadv_v_ogm_send_to_if(struct sk_buff *skb, +static void batadv_v_ogm_send_to_if(struct batadv_priv *bat_priv, + struct sk_buff *skb, struct batadv_hard_iface *hard_iface) { - struct batadv_priv *bat_priv = netdev_priv(hard_iface->mesh_iface); - if (hard_iface->if_status != BATADV_IF_ACTIVE) { kfree_skb(skb); return; @@ -187,6 +187,7 @@ static void batadv_v_ogm_aggr_list_free(struct batadv_hard_iface *hard_iface) /** * batadv_v_ogm_aggr_send() - flush & send aggregation queue + * @bat_priv: the bat priv with all the mesh interface information * @hard_iface: the interface with the aggregation queue to flush * * Aggregates all OGMv2 packets currently in the aggregation queue into a @@ -196,7 +197,8 @@ static void batadv_v_ogm_aggr_list_free(struct batadv_hard_iface *hard_iface) * * Caller needs to hold the hard_iface->bat_v.aggr_list.lock. */ -static void batadv_v_ogm_aggr_send(struct batadv_hard_iface *hard_iface) +static void batadv_v_ogm_aggr_send(struct batadv_priv *bat_priv, + struct batadv_hard_iface *hard_iface) { unsigned int aggr_len = hard_iface->bat_v.aggr_len; struct sk_buff *skb_aggr; @@ -226,27 +228,32 @@ static void batadv_v_ogm_aggr_send(struct batadv_hard_iface *hard_iface) consume_skb(skb); } - batadv_v_ogm_send_to_if(skb_aggr, hard_iface); + batadv_v_ogm_send_to_if(bat_priv, skb_aggr, hard_iface); } /** * batadv_v_ogm_queue_on_if() - queue a batman ogm on a given interface + * @bat_priv: the bat priv with all the mesh interface information * @skb: the OGM to queue * @hard_iface: the interface to queue the OGM on */ -static void batadv_v_ogm_queue_on_if(struct sk_buff *skb, +static void batadv_v_ogm_queue_on_if(struct batadv_priv *bat_priv, + struct sk_buff *skb, struct batadv_hard_iface *hard_iface) { - struct batadv_priv *bat_priv = netdev_priv(hard_iface->mesh_iface); + if (hard_iface->mesh_iface != bat_priv->mesh_iface) { + kfree_skb(skb); + return; + } if (!atomic_read(&bat_priv->aggregated_ogms)) { - batadv_v_ogm_send_to_if(skb, hard_iface); + batadv_v_ogm_send_to_if(bat_priv, skb, hard_iface); return; } spin_lock_bh(&hard_iface->bat_v.aggr_list.lock); if (!batadv_v_ogm_queue_left(skb, hard_iface)) - batadv_v_ogm_aggr_send(hard_iface); + batadv_v_ogm_aggr_send(bat_priv, hard_iface); hard_iface->bat_v.aggr_len += batadv_v_ogm_len(skb); __skb_queue_tail(&hard_iface->bat_v.aggr_list, skb); @@ -343,7 +350,7 @@ static void batadv_v_ogm_send_meshif(struct batadv_priv *bat_priv) break; } - batadv_v_ogm_queue_on_if(skb_tmp, hard_iface); + batadv_v_ogm_queue_on_if(bat_priv, skb_tmp, hard_iface); batadv_hardif_put(hard_iface); } rcu_read_unlock(); @@ -383,12 +390,14 @@ void batadv_v_ogm_aggr_work(struct work_struct *work) { struct batadv_hard_iface_bat_v *batv; struct batadv_hard_iface *hard_iface; + struct batadv_priv *bat_priv; batv = container_of(work, struct batadv_hard_iface_bat_v, aggr_wq.work); hard_iface = container_of(batv, struct batadv_hard_iface, bat_v); + bat_priv = netdev_priv(hard_iface->mesh_iface); spin_lock_bh(&hard_iface->bat_v.aggr_list.lock); - batadv_v_ogm_aggr_send(hard_iface); + batadv_v_ogm_aggr_send(bat_priv, hard_iface); spin_unlock_bh(&hard_iface->bat_v.aggr_list.lock); batadv_v_ogm_start_queue_timer(hard_iface); @@ -578,7 +587,7 @@ static void batadv_v_ogm_forward(struct batadv_priv *bat_priv, if_outgoing->net_dev->name, ntohl(ogm_forward->throughput), ogm_forward->ttl, if_incoming->net_dev->name); - batadv_v_ogm_queue_on_if(skb, if_outgoing); + batadv_v_ogm_queue_on_if(bat_priv, skb, if_outgoing); out: batadv_orig_ifinfo_put(orig_ifinfo); -- cgit v1.2.3 From 501368506563e151b322c8c3f228b796e615b90d Mon Sep 17 00:00:00 2001 From: Sven Eckelmann Date: Thu, 14 May 2026 16:33:12 +0200 Subject: batman-adv: tvlv: abort OGM send on tvlv append failure batadv_tvlv_container_ogm_append() could fail in two ways: a memory allocation failure when resizing the packet buffer, or the tvlv data exceeding U16_MAX bytes. In both cases the function previously returned the old (now stale) tvlv_value_len rather than signalling an error, causing the OGM/OGM2 send path to transmit a packet whose TVLV length field no longer matched the actual buffer contents. And because it also didn't fill in the new TVLV data, sending either uninitialized or corrupted data on the wire. All errors in batadv_tvlv_container_ogm_append() must be forwarded to the caller. And the caller must abort the send of the OGM2. For B.A.T.M.A.N. IV, it is currently not allowed to abort the send. The non-TVLV part of the OGM must be queued up instead. Cc: stable@kernel.org Fixes: ef26157747d4 ("batman-adv: tvlv - basic infrastructure") Signed-off-by: Sven Eckelmann --- net/batman-adv/bat_iv_ogm.c | 16 +++++++++++++--- net/batman-adv/bat_v_ogm.c | 26 ++++++++++++++------------ net/batman-adv/tvlv.c | 17 ++++++++++++----- net/batman-adv/tvlv.h | 2 +- 4 files changed, 40 insertions(+), 21 deletions(-) (limited to 'net') diff --git a/net/batman-adv/bat_iv_ogm.c b/net/batman-adv/bat_iv_ogm.c index 74ef7dc2b2f9..7ad26128b5f7 100644 --- a/net/batman-adv/bat_iv_ogm.c +++ b/net/batman-adv/bat_iv_ogm.c @@ -790,6 +790,7 @@ static void batadv_iv_ogm_schedule_buff(struct batadv_hard_iface *hard_iface) u32 seqno; u16 tvlv_len = 0; unsigned long send_time; + int ret; lockdep_assert_held(&hard_iface->bat_iv.ogm_buff_mutex); @@ -813,9 +814,18 @@ static void batadv_iv_ogm_schedule_buff(struct batadv_hard_iface *hard_iface) * appended as it may alter the tt tvlv container */ batadv_tt_local_commit_changes(bat_priv); - tvlv_len = batadv_tvlv_container_ogm_append(bat_priv, ogm_buff, - ogm_buff_len, - BATADV_OGM_HLEN); + ret = batadv_tvlv_container_ogm_append(bat_priv, ogm_buff, + ogm_buff_len, + BATADV_OGM_HLEN); + if (ret < 0) { + /* OGMs must be queued even when the buffer allocation for + * TVLVs failed. just fall back to the non-TVLV version + */ + ret = 0; + *ogm_buff_len = BATADV_OGM_HLEN; + } + + tvlv_len = ret; } batadv_ogm_packet = (struct batadv_ogm_packet *)(*ogm_buff); diff --git a/net/batman-adv/bat_v_ogm.c b/net/batman-adv/bat_v_ogm.c index e955b4940c72..d66ca77b1aaa 100644 --- a/net/batman-adv/bat_v_ogm.c +++ b/net/batman-adv/bat_v_ogm.c @@ -269,10 +269,10 @@ static void batadv_v_ogm_send_meshif(struct batadv_priv *bat_priv) struct batadv_hard_iface *hard_iface; struct batadv_ogm2_packet *ogm_packet; struct sk_buff *skb, *skb_tmp; - unsigned char *ogm_buff; + unsigned char **ogm_buff; struct list_head *iter; - int ogm_buff_len; - u16 tvlv_len = 0; + int *ogm_buff_len; + u16 tvlv_len; int ret; lockdep_assert_held(&bat_priv->bat_v.ogm_buff_mutex); @@ -280,25 +280,27 @@ static void batadv_v_ogm_send_meshif(struct batadv_priv *bat_priv) if (atomic_read(&bat_priv->mesh_state) == BATADV_MESH_DEACTIVATING) goto out; - ogm_buff = bat_priv->bat_v.ogm_buff; - ogm_buff_len = bat_priv->bat_v.ogm_buff_len; + ogm_buff = &bat_priv->bat_v.ogm_buff; + ogm_buff_len = &bat_priv->bat_v.ogm_buff_len; + /* tt changes have to be committed before the tvlv data is * appended as it may alter the tt tvlv container */ batadv_tt_local_commit_changes(bat_priv); - tvlv_len = batadv_tvlv_container_ogm_append(bat_priv, &ogm_buff, - &ogm_buff_len, - BATADV_OGM2_HLEN); + ret = batadv_tvlv_container_ogm_append(bat_priv, ogm_buff, + ogm_buff_len, + BATADV_OGM2_HLEN); + if (ret < 0) + goto reschedule; - bat_priv->bat_v.ogm_buff = ogm_buff; - bat_priv->bat_v.ogm_buff_len = ogm_buff_len; + tvlv_len = ret; - skb = netdev_alloc_skb_ip_align(NULL, ETH_HLEN + ogm_buff_len); + skb = netdev_alloc_skb_ip_align(NULL, ETH_HLEN + *ogm_buff_len); if (!skb) goto reschedule; skb_reserve(skb, ETH_HLEN); - skb_put_data(skb, ogm_buff, ogm_buff_len); + skb_put_data(skb, *ogm_buff, *ogm_buff_len); ogm_packet = (struct batadv_ogm2_packet *)skb->data; ogm_packet->seqno = htonl(atomic_read(&bat_priv->bat_v.ogm_seqno)); diff --git a/net/batman-adv/tvlv.c b/net/batman-adv/tvlv.c index 8129a3f9c44d..46ed61dbf087 100644 --- a/net/batman-adv/tvlv.c +++ b/net/batman-adv/tvlv.c @@ -8,6 +8,7 @@ #include #include +#include #include #include #include @@ -306,9 +307,10 @@ static bool batadv_tvlv_realloc_packet_buff(unsigned char **packet_buff, * The ogm packet might be enlarged or shrunk depending on the current size * and the size of the to-be-appended tvlv containers. * - * Return: size of all appended tvlv containers in bytes. + * Return: size of all appended tvlv containers in bytes (max U16_MAX), negative + * if operation failed */ -u16 batadv_tvlv_container_ogm_append(struct batadv_priv *bat_priv, +int batadv_tvlv_container_ogm_append(struct batadv_priv *bat_priv, unsigned char **packet_buff, int *packet_buff_len, int packet_min_len) { @@ -316,6 +318,7 @@ u16 batadv_tvlv_container_ogm_append(struct batadv_priv *bat_priv, struct batadv_tvlv_hdr *tvlv_hdr; u16 tvlv_value_len; void *tvlv_value; + int tvlv_len_ret; bool ret; spin_lock_bh(&bat_priv->tvlv.container_list_lock); @@ -323,9 +326,12 @@ u16 batadv_tvlv_container_ogm_append(struct batadv_priv *bat_priv, ret = batadv_tvlv_realloc_packet_buff(packet_buff, packet_buff_len, packet_min_len, tvlv_value_len); - - if (!ret) + if (!ret) { + tvlv_len_ret = -ENOMEM; goto end; + } + + tvlv_len_ret = tvlv_value_len; if (!tvlv_value_len) goto end; @@ -344,7 +350,8 @@ u16 batadv_tvlv_container_ogm_append(struct batadv_priv *bat_priv, end: spin_unlock_bh(&bat_priv->tvlv.container_list_lock); - return tvlv_value_len; + + return tvlv_len_ret; } /** diff --git a/net/batman-adv/tvlv.h b/net/batman-adv/tvlv.h index e5697230d991..f96f6b3f44a0 100644 --- a/net/batman-adv/tvlv.h +++ b/net/batman-adv/tvlv.h @@ -16,7 +16,7 @@ void batadv_tvlv_container_register(struct batadv_priv *bat_priv, u8 type, u8 version, void *tvlv_value, u16 tvlv_value_len); -u16 batadv_tvlv_container_ogm_append(struct batadv_priv *bat_priv, +int batadv_tvlv_container_ogm_append(struct batadv_priv *bat_priv, unsigned char **packet_buff, int *packet_buff_len, int packet_min_len); void batadv_tvlv_ogm_receive(struct batadv_priv *bat_priv, -- cgit v1.2.3 From f50487e3566358b2b982b7801945e858c78ad9ab Mon Sep 17 00:00:00 2001 From: Sven Eckelmann Date: Sat, 9 May 2026 21:55:29 +0200 Subject: batman-adv: tvlv: reject oversized TVLV packets batadv_tvlv_container_ogm_append() builds a TVLV packet section from the tvlv.container_list. The total size of this section is computed by batadv_tvlv_container_list_size(), which sums the sizes of all registered containers. The return type and accumulator in batadv_tvlv_container_list_size() were u16. If the accumulated size exceeds U16_MAX, the value wraps around, causing the subsequent allocation in batadv_tvlv_container_ogm_append() to be undersized. The memcpy-style copy that follows would then write beyond the end of the allocated buffer, corrupting kernel memory. Fix this by widening the return type of batadv_tvlv_container_list_size() to size_t. In batadv_tvlv_container_ogm_append(), check the computed length against U16_MAX before proceeding, and bail out as if the allocation had failed when the limit is exceeded. Cc: stable@kernel.org Fixes: ef26157747d4 ("batman-adv: tvlv - basic infrastructure") Reported-by: Yuan Tan Reported-by: Yifan Wu Reported-by: Juefei Pu Reported-by: Xin Liu Reviewed-by: Yuan Tan Signed-off-by: Sven Eckelmann --- net/batman-adv/tvlv.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/batman-adv/tvlv.c b/net/batman-adv/tvlv.c index 46ed61dbf087..cc6ac580c620 100644 --- a/net/batman-adv/tvlv.c +++ b/net/batman-adv/tvlv.c @@ -13,6 +13,7 @@ #include #include #include +#include #include #include #include @@ -160,10 +161,10 @@ batadv_tvlv_container_get(struct batadv_priv *bat_priv, u8 type, u8 version) * * Return: size of all currently registered tvlv containers in bytes. */ -static u16 batadv_tvlv_container_list_size(struct batadv_priv *bat_priv) +static size_t batadv_tvlv_container_list_size(struct batadv_priv *bat_priv) { struct batadv_tvlv_container *tvlv; - u16 tvlv_len = 0; + size_t tvlv_len = 0; lockdep_assert_held(&bat_priv->tvlv.container_list_lock); @@ -316,13 +317,17 @@ int batadv_tvlv_container_ogm_append(struct batadv_priv *bat_priv, { struct batadv_tvlv_container *tvlv; struct batadv_tvlv_hdr *tvlv_hdr; - u16 tvlv_value_len; + size_t tvlv_value_len; void *tvlv_value; int tvlv_len_ret; bool ret; spin_lock_bh(&bat_priv->tvlv.container_list_lock); tvlv_value_len = batadv_tvlv_container_list_size(bat_priv); + if (tvlv_value_len > U16_MAX) { + tvlv_len_ret = -E2BIG; + goto end; + } ret = batadv_tvlv_realloc_packet_buff(packet_buff, packet_buff_len, packet_min_len, tvlv_value_len); -- cgit v1.2.3 From 71dce47f0758537fff78fddb5fb0d4632d29b29f Mon Sep 17 00:00:00 2001 From: Sven Eckelmann Date: Wed, 13 May 2026 23:38:54 +0200 Subject: batman-adv: tp_meter: fix race condition in send error reporting batadv_tp_sender_shutdown() previously used two separate variables to track session state: sending (an atomic flag indicating whether the session was active) and reason (a plain enum storing the stop reason). This introduced a race window between the two writes: after sending was cleared to 0, batadv_tp_send() could observe the stopped state and call batadv_tp_sender_end() before reason was written, causing the wrong stop reason to be reported to the caller. Fix this by consolidating both variables into a single atomic send_result, which holds 0 while the session is running and the stop reason once it ends. Cc: stable@kernel.org Fixes: 33a3bb4a3345 ("batman-adv: throughput meter implementation") Signed-off-by: Sven Eckelmann --- net/batman-adv/tp_meter.c | 40 +++++++++++++++++++++++++--------------- net/batman-adv/types.h | 10 +++++----- 2 files changed, 30 insertions(+), 20 deletions(-) (limited to 'net') diff --git a/net/batman-adv/tp_meter.c b/net/batman-adv/tp_meter.c index 1fd1526059d8..3ce6d9b2c9f3 100644 --- a/net/batman-adv/tp_meter.c +++ b/net/batman-adv/tp_meter.c @@ -413,11 +413,14 @@ static void batadv_tp_sender_cleanup(struct batadv_tp_vars *tp_vars) static void batadv_tp_sender_end(struct batadv_priv *bat_priv, struct batadv_tp_vars *tp_vars) { + enum batadv_tp_meter_reason reason; u32 session_cookie; + reason = atomic_read(&tp_vars->send_result); + batadv_dbg(BATADV_DBG_TP_METER, bat_priv, "Test towards %pM finished..shutting down (reason=%d)\n", - tp_vars->other_end, tp_vars->reason); + tp_vars->other_end, reason); batadv_dbg(BATADV_DBG_TP_METER, bat_priv, "Last timing stats: SRTT=%ums RTTVAR=%ums RTO=%ums\n", @@ -430,7 +433,7 @@ static void batadv_tp_sender_end(struct batadv_priv *bat_priv, session_cookie = batadv_tp_session_cookie(tp_vars->session, tp_vars->icmp_uid); - batadv_tp_batctl_notify(tp_vars->reason, + batadv_tp_batctl_notify(reason, tp_vars->other_end, bat_priv, tp_vars->start_time, @@ -446,10 +449,18 @@ static void batadv_tp_sender_end(struct batadv_priv *bat_priv, static void batadv_tp_sender_shutdown(struct batadv_tp_vars *tp_vars, enum batadv_tp_meter_reason reason) { - if (atomic_xchg(&tp_vars->sending, 0) != 1) - return; + atomic_cmpxchg(&tp_vars->send_result, 0, reason); +} - tp_vars->reason = reason; +/** + * batadv_tp_sender_stopped() - check if tp session was stopped with reason + * @tp_vars: the private data of the current TP meter session + * + * Return: whether stop reason was found + */ +static bool batadv_tp_sender_stopped(struct batadv_tp_vars *tp_vars) +{ + return atomic_read(&tp_vars->send_result) != 0; } /** @@ -479,7 +490,7 @@ static void batadv_tp_reset_sender_timer(struct batadv_tp_vars *tp_vars) /* most of the time this function is invoked while normal packet * reception... */ - if (unlikely(atomic_read(&tp_vars->sending) == 0)) + if (unlikely(batadv_tp_sender_stopped(tp_vars))) /* timer ref will be dropped in batadv_tp_sender_cleanup */ return; @@ -499,7 +510,7 @@ static void batadv_tp_sender_timeout(struct timer_list *t) struct batadv_tp_vars *tp_vars = timer_container_of(tp_vars, t, timer); struct batadv_priv *bat_priv = tp_vars->bat_priv; - if (atomic_read(&tp_vars->sending) == 0) + if (batadv_tp_sender_stopped(tp_vars)) return; /* if the user waited long enough...shutdown the test */ @@ -661,7 +672,7 @@ static void batadv_tp_recv_ack(struct batadv_priv *bat_priv, if (unlikely(tp_vars->role != BATADV_TP_SENDER)) goto out; - if (unlikely(atomic_read(&tp_vars->sending) == 0)) + if (unlikely(batadv_tp_sender_stopped(tp_vars))) goto out; /* old ACK? silently drop it.. */ @@ -827,21 +838,21 @@ static int batadv_tp_send(void *arg) if (unlikely(tp_vars->role != BATADV_TP_SENDER)) { err = BATADV_TP_REASON_DST_UNREACHABLE; - tp_vars->reason = err; + batadv_tp_sender_shutdown(tp_vars, err); goto out; } orig_node = batadv_orig_hash_find(bat_priv, tp_vars->other_end); if (unlikely(!orig_node)) { err = BATADV_TP_REASON_DST_UNREACHABLE; - tp_vars->reason = err; + batadv_tp_sender_shutdown(tp_vars, err); goto out; } primary_if = batadv_primary_if_get_selected(bat_priv); if (unlikely(!primary_if)) { err = BATADV_TP_REASON_DST_UNREACHABLE; - tp_vars->reason = err; + batadv_tp_sender_shutdown(tp_vars, err); goto out; } @@ -860,7 +871,7 @@ static int batadv_tp_send(void *arg) queue_delayed_work(batadv_event_workqueue, &tp_vars->finish_work, msecs_to_jiffies(tp_vars->test_length)); - while (atomic_read(&tp_vars->sending) != 0) { + while (!batadv_tp_sender_stopped(tp_vars)) { if (unlikely(!batadv_tp_avail(tp_vars, payload_len))) { batadv_tp_wait_available(tp_vars, payload_len); continue; @@ -883,8 +894,7 @@ static int batadv_tp_send(void *arg) "Meter: %s() cannot send packets (%d)\n", __func__, err); /* ensure nobody else tries to stop the thread now */ - if (atomic_xchg(&tp_vars->sending, 0) == 1) - tp_vars->reason = err; + batadv_tp_sender_shutdown(tp_vars, err); break; } @@ -1006,7 +1016,7 @@ void batadv_tp_start(struct batadv_priv *bat_priv, const u8 *dst, ether_addr_copy(tp_vars->other_end, dst); kref_init(&tp_vars->refcount); tp_vars->role = BATADV_TP_SENDER; - atomic_set(&tp_vars->sending, 1); + atomic_set(&tp_vars->send_result, 0); memcpy(tp_vars->session, session_id, sizeof(session_id)); tp_vars->icmp_uid = icmp_uid; diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h index c8c3e8064f00..fb0e4cb89d79 100644 --- a/net/batman-adv/types.h +++ b/net/batman-adv/types.h @@ -1320,15 +1320,15 @@ struct batadv_tp_vars { /** @role: receiver/sender modi */ enum batadv_tp_meter_role role; - /** @sending: sending binary semaphore: 1 if sending, 0 is not */ - atomic_t sending; + /** + * @send_result: 0 when sending is ongoing and otherwise + * enum batadv_tp_meter_reason + */ + atomic_t send_result; /** @receiving: receiving binary semaphore: 1 if receiving, 0 is not */ atomic_t receiving; - /** @reason: reason for a stopped session */ - enum batadv_tp_meter_reason reason; - /** @finish_work: work item for the finishing procedure */ struct delayed_work finish_work; -- cgit v1.2.3 From ff24f2ecfd94c07a2b89bac497433e3b23271cac Mon Sep 17 00:00:00 2001 From: Sven Eckelmann Date: Sat, 16 May 2026 12:33:41 +0200 Subject: batman-adv: tp_meter: avoid role confusion in tp_list Session lookups in tp_list matched only on destination address (and optionally session ID), leaving role validation to the caller. If two sessions with the same other_end coexisted (one as sender, one as receiver) a lookup could silently return the wrong one, causing the caller's role to bail out early, potentially skipping necessary cleanup. Move the role check into the lookup functions themselves so the correct entry is always returned, or none at all. Since batadv_tp_start() legitimately needs to detect any active session to a destination regardless of role, introduce a dedicated helper for that case rather than bending the existing lookup semantics. Cc: stable@kernel.org Fixes: 33a3bb4a3345 ("batman-adv: throughput meter implementation") Signed-off-by: Sven Eckelmann --- net/batman-adv/tp_meter.c | 59 +++++++++++++++++++++++++++++------------------ 1 file changed, 36 insertions(+), 23 deletions(-) (limited to 'net') diff --git a/net/batman-adv/tp_meter.c b/net/batman-adv/tp_meter.c index 3ce6d9b2c9f3..0fc4ca78e84e 100644 --- a/net/batman-adv/tp_meter.c +++ b/net/batman-adv/tp_meter.c @@ -255,6 +255,7 @@ static void batadv_tp_batctl_error_notify(enum batadv_tp_meter_reason reason, * batadv_tp_list_find() - find a tp_vars object in the global list * @bat_priv: the bat priv with all the mesh interface information * @dst: the other endpoint MAC address to look for + * @role: role of the session * * Look for a tp_vars object matching dst as end_point and return it after * having increment the refcounter. Return NULL is not found @@ -262,7 +263,8 @@ static void batadv_tp_batctl_error_notify(enum batadv_tp_meter_reason reason, * Return: matching tp_vars or NULL when no tp_vars with @dst was found */ static struct batadv_tp_vars *batadv_tp_list_find(struct batadv_priv *bat_priv, - const u8 *dst) + const u8 *dst, + enum batadv_tp_meter_role role) { struct batadv_tp_vars *pos, *tp_vars = NULL; @@ -271,6 +273,9 @@ static struct batadv_tp_vars *batadv_tp_list_find(struct batadv_priv *bat_priv, if (!batadv_compare_eth(pos->other_end, dst)) continue; + if (pos->role != role) + continue; + /* most of the time this function is invoked during the normal * process..it makes sens to pay more when the session is * finished and to speed the process up during the measurement @@ -286,12 +291,33 @@ static struct batadv_tp_vars *batadv_tp_list_find(struct batadv_priv *bat_priv, return tp_vars; } +/** + * batadv_tp_list_active() - check if session from/to destination is ongoing + * @bat_priv: the bat priv with all the mesh interface information + * @dst: the other endpoint MAC address to look for + * + * Return: if matching session with @dst was found + */ +static bool batadv_tp_list_active(struct batadv_priv *bat_priv, const u8 *dst) + __must_hold(&bat_priv->tp_list_lock) +{ + struct batadv_tp_vars *tp_vars; + + hlist_for_each_entry_rcu(tp_vars, &bat_priv->tp_list, list) { + if (batadv_compare_eth(tp_vars->other_end, dst)) + return true; + } + + return false; +} + /** * batadv_tp_list_find_session() - find tp_vars session object in the global * list * @bat_priv: the bat priv with all the mesh interface information * @dst: the other endpoint MAC address to look for * @session: session identifier + * @role: role of the session * * Look for a tp_vars object matching dst as end_point, session as tp meter * session and return it after having increment the refcounter. Return NULL @@ -301,7 +327,7 @@ static struct batadv_tp_vars *batadv_tp_list_find(struct batadv_priv *bat_priv, */ static struct batadv_tp_vars * batadv_tp_list_find_session(struct batadv_priv *bat_priv, const u8 *dst, - const u8 *session) + const u8 *session, enum batadv_tp_meter_role role) { struct batadv_tp_vars *pos, *tp_vars = NULL; @@ -313,6 +339,9 @@ batadv_tp_list_find_session(struct batadv_priv *bat_priv, const u8 *dst, if (memcmp(pos->session, session, sizeof(pos->session)) != 0) continue; + if (pos->role != role) + continue; + /* most of the time this function is invoked during the normal * process..it makes sense to pay more when the session is * finished and to speed the process up during the measurement @@ -665,13 +694,10 @@ static void batadv_tp_recv_ack(struct batadv_priv *bat_priv, /* find the tp_vars */ tp_vars = batadv_tp_list_find_session(bat_priv, icmp->orig, - icmp->session); + icmp->session, BATADV_TP_SENDER); if (unlikely(!tp_vars)) return; - if (unlikely(tp_vars->role != BATADV_TP_SENDER)) - goto out; - if (unlikely(batadv_tp_sender_stopped(tp_vars))) goto out; @@ -980,10 +1006,8 @@ void batadv_tp_start(struct batadv_priv *bat_priv, const u8 *dst, return; } - tp_vars = batadv_tp_list_find(bat_priv, dst); - if (tp_vars) { + if (batadv_tp_list_active(bat_priv, dst)) { spin_unlock_bh(&bat_priv->tp_list_lock); - batadv_tp_vars_put(tp_vars); batadv_dbg(BATADV_DBG_TP_METER, bat_priv, "Meter: test to or from the same node already ongoing, aborting\n"); batadv_tp_batctl_error_notify(BATADV_TP_REASON_ALREADY_ONGOING, @@ -1104,18 +1128,14 @@ void batadv_tp_stop(struct batadv_priv *bat_priv, const u8 *dst, if (!orig_node) return; - tp_vars = batadv_tp_list_find(bat_priv, orig_node->orig); + tp_vars = batadv_tp_list_find(bat_priv, orig_node->orig, BATADV_TP_SENDER); if (!tp_vars) { batadv_dbg(BATADV_DBG_TP_METER, bat_priv, "Meter: trying to interrupt an already over connection\n"); goto out_put_orig_node; } - if (unlikely(tp_vars->role != BATADV_TP_SENDER)) - goto out_put_tp_vars; - batadv_tp_sender_shutdown(tp_vars, return_value); -out_put_tp_vars: batadv_tp_vars_put(tp_vars); out_put_orig_node: batadv_orig_node_put(orig_node); @@ -1371,7 +1391,7 @@ batadv_tp_init_recv(struct batadv_priv *bat_priv, goto out_unlock; tp_vars = batadv_tp_list_find_session(bat_priv, icmp->orig, - icmp->session); + icmp->session, BATADV_TP_RECEIVER); if (tp_vars) goto out_unlock; @@ -1442,7 +1462,7 @@ static void batadv_tp_recv_msg(struct batadv_priv *bat_priv, } } else { tp_vars = batadv_tp_list_find_session(bat_priv, icmp->orig, - icmp->session); + icmp->session, BATADV_TP_RECEIVER); if (!tp_vars) { batadv_dbg(BATADV_DBG_TP_METER, bat_priv, "Unexpected packet from %pM!\n", @@ -1451,13 +1471,6 @@ static void batadv_tp_recv_msg(struct batadv_priv *bat_priv, } } - if (unlikely(tp_vars->role != BATADV_TP_RECEIVER)) { - batadv_dbg(BATADV_DBG_TP_METER, bat_priv, - "Meter: dropping packet: not expected (role=%u)\n", - tp_vars->role); - goto out; - } - tp_vars->last_recv_time = jiffies; /* if the packet is a duplicate, it may be the case that an ACK has been -- cgit v1.2.3 From 20c2d6a20ca936f5aaa6dd40f73f262ac45c87cc Mon Sep 17 00:00:00 2001 From: Sven Eckelmann Date: Thu, 14 May 2026 19:22:02 +0200 Subject: batman-adv: mcast: fix use-after-free in orig_node RCU release MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit batadv_mcast_purge_orig() removes entries from RCU-protected hlists but does not wait for an RCU grace period before returning. Concurrent RCU readers may still accesses references to those entries at the point of removal. RCU-protected readers trying to operate on entries like orig->mcast_want_all_ipv6_node will then access already freed memory. Fix this by moving batadv_mcast_purge_orig() to batadv_orig_node_release(), just before the call_rcu() invocation. This ensures RCU readers that were active at purge time have drained before the orig_node memory is reclaimed. Cc: stable@kernel.org Fixes: ab49886e3da7 ("batman-adv: Add IPv4 link-local/IPv6-ll-all-nodes multicast support") Acked-by: Linus Lüssing Signed-off-by: Sven Eckelmann --- net/batman-adv/originator.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/batman-adv/originator.c b/net/batman-adv/originator.c index b3468ccab535..ad4921b659d9 100644 --- a/net/batman-adv/originator.c +++ b/net/batman-adv/originator.c @@ -835,8 +835,6 @@ static void batadv_orig_node_free_rcu(struct rcu_head *rcu) orig_node = container_of(rcu, struct batadv_orig_node, rcu); - batadv_mcast_purge_orig(orig_node); - batadv_frag_purge_orig(orig_node, NULL); kfree(orig_node->tt_buff); @@ -887,6 +885,8 @@ void batadv_orig_node_release(struct kref *ref) } spin_unlock_bh(&orig_node->vlan_list_lock); + batadv_mcast_purge_orig(orig_node); + call_rcu(&orig_node->rcu, batadv_orig_node_free_rcu); } -- cgit v1.2.3 From aa3153bd139a6c48667dcd02608d3b2c80bff02c Mon Sep 17 00:00:00 2001 From: Sven Eckelmann Date: Fri, 15 May 2026 22:00:40 +0200 Subject: batman-adv: iv: recover OGM scheduling after forward packet error MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When batadv_iv_ogm_schedule_buff() fails to allocate and queue a forward packet for OGM transmission, the work item that drives periodic OGM scheduling is never re-armed. This silently halts transmission of the node's own OGMs on the affected interface — only OGMs from other peers continue to be aggregated and forwarded. Fix this by tracking whether batadv_iv_ogm_queue_add() (and transitively batadv_iv_ogm_aggregate_new()) successfully scheduled a forward packet. When scheduling fails, batadv_iv_ogm_schedule_buff() falls back to queuing a dedicated recovery work item (reschedule_work) that fires after one originator interval and calls batadv_iv_ogm_schedule() again. Cc: stable@kernel.org Fixes: c6c8fea29769 ("net: Add batman-adv meshing protocol") Signed-off-by: Sven Eckelmann --- net/batman-adv/bat_iv_ogm.c | 76 +++++++++++++++++++++++++++++++++------------ net/batman-adv/types.h | 3 ++ 2 files changed, 60 insertions(+), 19 deletions(-) (limited to 'net') diff --git a/net/batman-adv/bat_iv_ogm.c b/net/batman-adv/bat_iv_ogm.c index 7ad26128b5f7..b8b1b997960a 100644 --- a/net/batman-adv/bat_iv_ogm.c +++ b/net/batman-adv/bat_iv_ogm.c @@ -224,6 +224,8 @@ static void batadv_iv_ogm_iface_disable(struct batadv_hard_iface *hard_iface) hard_iface->bat_iv.ogm_buff = NULL; mutex_unlock(&hard_iface->bat_iv.ogm_buff_mutex); + + cancel_delayed_work_sync(&hard_iface->bat_iv.reschedule_work); } static void batadv_iv_ogm_iface_update_mac(struct batadv_hard_iface *hard_iface) @@ -536,8 +538,10 @@ out: * @if_incoming: interface where the packet was received * @if_outgoing: interface for which the retransmission should be considered * @own_packet: true if it is a self-generated ogm + * + * Return: whether forward packet was scheduled */ -static void batadv_iv_ogm_aggregate_new(const unsigned char *packet_buff, +static bool batadv_iv_ogm_aggregate_new(const unsigned char *packet_buff, int packet_len, unsigned long send_time, bool direct_link, struct batadv_hard_iface *if_incoming, @@ -561,13 +565,13 @@ static void batadv_iv_ogm_aggregate_new(const unsigned char *packet_buff, skb = netdev_alloc_skb_ip_align(NULL, skb_size); if (!skb) - return; + return false; forw_packet_aggr = batadv_forw_packet_alloc(if_incoming, if_outgoing, queue_left, bat_priv, skb); if (!forw_packet_aggr) { kfree_skb(skb); - return; + return false; } forw_packet_aggr->skb->priority = TC_PRIO_CONTROL; @@ -590,6 +594,8 @@ static void batadv_iv_ogm_aggregate_new(const unsigned char *packet_buff, batadv_iv_send_outstanding_bat_ogm_packet); batadv_forw_packet_ogmv1_queue(bat_priv, forw_packet_aggr, send_time); + + return true; } /* aggregate a new packet into the existing ogm packet */ @@ -617,8 +623,10 @@ static void batadv_iv_ogm_aggregate(struct batadv_forw_packet *forw_packet_aggr, * @if_outgoing: interface for which the retransmission should be considered * @own_packet: true if it is a self-generated ogm * @send_time: timestamp (jiffies) when the packet is to be sent + * + * Return: whether forward packet was scheduled */ -static void batadv_iv_ogm_queue_add(struct batadv_priv *bat_priv, +static bool batadv_iv_ogm_queue_add(struct batadv_priv *bat_priv, unsigned char *packet_buff, int packet_len, struct batadv_hard_iface *if_incoming, @@ -670,14 +678,16 @@ static void batadv_iv_ogm_queue_add(struct batadv_priv *bat_priv, if (!own_packet && atomic_read(&bat_priv->aggregated_ogms)) send_time += max_aggregation_jiffies; - batadv_iv_ogm_aggregate_new(packet_buff, packet_len, - send_time, direct_link, - if_incoming, if_outgoing, - own_packet); + return batadv_iv_ogm_aggregate_new(packet_buff, packet_len, + send_time, direct_link, + if_incoming, if_outgoing, + own_packet); } else { batadv_iv_ogm_aggregate(forw_packet_aggr, packet_buff, packet_len, direct_link); spin_unlock_bh(&bat_priv->forw_bat_list_lock); + + return true; } } @@ -790,6 +800,8 @@ static void batadv_iv_ogm_schedule_buff(struct batadv_hard_iface *hard_iface) u32 seqno; u16 tvlv_len = 0; unsigned long send_time; + bool reschedule = false; + bool scheduled; int ret; lockdep_assert_held(&hard_iface->bat_iv.ogm_buff_mutex); @@ -818,11 +830,8 @@ static void batadv_iv_ogm_schedule_buff(struct batadv_hard_iface *hard_iface) ogm_buff_len, BATADV_OGM_HLEN); if (ret < 0) { - /* OGMs must be queued even when the buffer allocation for - * TVLVs failed. just fall back to the non-TVLV version - */ - ret = 0; - *ogm_buff_len = BATADV_OGM_HLEN; + reschedule = true; + goto out; } tvlv_len = ret; @@ -844,8 +853,11 @@ static void batadv_iv_ogm_schedule_buff(struct batadv_hard_iface *hard_iface) /* OGMs from secondary interfaces are only scheduled on their * respective interfaces. */ - batadv_iv_ogm_queue_add(bat_priv, *ogm_buff, *ogm_buff_len, - hard_iface, hard_iface, 1, send_time); + scheduled = batadv_iv_ogm_queue_add(bat_priv, *ogm_buff, *ogm_buff_len, + hard_iface, hard_iface, 1, send_time); + if (!scheduled) + reschedule = true; + goto out; } @@ -857,15 +869,28 @@ static void batadv_iv_ogm_schedule_buff(struct batadv_hard_iface *hard_iface) if (!kref_get_unless_zero(&tmp_hard_iface->refcount)) continue; - batadv_iv_ogm_queue_add(bat_priv, *ogm_buff, - *ogm_buff_len, hard_iface, - tmp_hard_iface, 1, send_time); - + scheduled = batadv_iv_ogm_queue_add(bat_priv, *ogm_buff, + *ogm_buff_len, hard_iface, + tmp_hard_iface, 1, send_time); batadv_hardif_put(tmp_hard_iface); + + if (!scheduled && tmp_hard_iface == hard_iface) + reschedule = true; } rcu_read_unlock(); out: + if (reschedule) { + /* there was a failure scheduling the own forward packet. + * as result, the batadv_iv_send_outstanding_bat_ogm_packet() + * work item is no longer scheduled. it is therefore necessary + * to reschedule it manually + */ + queue_delayed_work(batadv_event_workqueue, + &hard_iface->bat_iv.reschedule_work, + msecs_to_jiffies(atomic_read(&bat_priv->orig_interval))); + } + batadv_hardif_put(primary_if); } @@ -880,6 +905,17 @@ static void batadv_iv_ogm_schedule(struct batadv_hard_iface *hard_iface) mutex_unlock(&hard_iface->bat_iv.ogm_buff_mutex); } +static void batadv_iv_ogm_reschedule(struct work_struct *work) +{ + struct delayed_work *delayed_work = to_delayed_work(work); + struct batadv_hard_iface *hard_iface; + + hard_iface = container_of(delayed_work, + struct batadv_hard_iface, + bat_iv.reschedule_work); + batadv_iv_ogm_schedule(hard_iface); +} + /** * batadv_iv_orig_ifinfo_sum() - Get bcast_own sum for originator over interface * @orig_node: originator which reproadcasted the OGMs directly @@ -2272,6 +2308,8 @@ batadv_iv_ogm_neigh_is_sob(struct batadv_neigh_node *neigh1, static void batadv_iv_iface_enabled(struct batadv_hard_iface *hard_iface) { + INIT_DELAYED_WORK(&hard_iface->bat_iv.reschedule_work, batadv_iv_ogm_reschedule); + /* begin scheduling originator messages on that interface */ batadv_iv_ogm_schedule(hard_iface); } diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h index fb0e4cb89d79..821ada05d86a 100644 --- a/net/batman-adv/types.h +++ b/net/batman-adv/types.h @@ -83,6 +83,9 @@ struct batadv_hard_iface_bat_iv { /** @ogm_seqno: OGM sequence number - used to identify each OGM */ atomic_t ogm_seqno; + /** @reschedule_work: recover OGM schedule after schedule error */ + struct delayed_work reschedule_work; + /** @ogm_buff_mutex: lock protecting ogm_buff and ogm_buff_len */ struct mutex ogm_buff_mutex; }; -- cgit v1.2.3 From 0459430add32ea41f3e2ef9351610e6d33627a6b Mon Sep 17 00:00:00 2001 From: Sven Eckelmann Date: Sun, 10 May 2026 11:43:20 +0200 Subject: batman-adv: bla: fix report_work leak on backbone_gw purge batadv_bla_purge_backbone_gw() removes stale backbone gateway entries, but fails to properly handle their associated report_work: - If report_work is running, the purge must wait for it to finish before freeing the backbone_gw, otherwise the worker may access freed memory (e.g. bat_priv). - If report_work is pending, the purge must cancel it and release the reference held for that pending work item. The previous implementation called hlist_for_each_entry_safe() inside a spin_lock_bh() section, but cancel_work_sync() may sleep and therefore cannot be called from within a spinlock-protected region. Restructure the loop to handle one entry per spinlock critical section: acquire the lock, find the next entry to purge, remove it from the hash list, then release the lock before calling cancel_work_sync() and dropping the hash_entry reference. Repeat until no more entries require purging. Cc: stable@kernel.org Fixes: 23721387c409 ("batman-adv: add basic bridge loop avoidance code") Reviewed-by: Simon Wunderlich Signed-off-by: Sven Eckelmann --- net/batman-adv/bridge_loop_avoidance.c | 54 ++++++++++++++++++++++------------ 1 file changed, 35 insertions(+), 19 deletions(-) (limited to 'net') diff --git a/net/batman-adv/bridge_loop_avoidance.c b/net/batman-adv/bridge_loop_avoidance.c index cec11f1251d6..df1dfdf4a1a1 100644 --- a/net/batman-adv/bridge_loop_avoidance.c +++ b/net/batman-adv/bridge_loop_avoidance.c @@ -1224,6 +1224,7 @@ static void batadv_bla_purge_backbone_gw(struct batadv_priv *bat_priv, int now) struct hlist_head *head; struct batadv_hashtable *hash; spinlock_t *list_lock; /* protects write access to the hash lists */ + bool purged; int i; hash = bat_priv->bla.backbone_hash; @@ -1234,30 +1235,45 @@ static void batadv_bla_purge_backbone_gw(struct batadv_priv *bat_priv, int now) head = &hash->table[i]; list_lock = &hash->list_locks[i]; - spin_lock_bh(list_lock); - hlist_for_each_entry_safe(backbone_gw, node_tmp, - head, hash_entry) { - if (now) - goto purge_now; - if (!batadv_has_timed_out(backbone_gw->lasttime, - BATADV_BLA_BACKBONE_TIMEOUT)) - continue; + do { + purged = false; - batadv_dbg(BATADV_DBG_BLA, backbone_gw->bat_priv, - "%s(): backbone gw %pM timed out\n", - __func__, backbone_gw->orig); + spin_lock_bh(list_lock); + hlist_for_each_entry_safe(backbone_gw, node_tmp, + head, hash_entry) { + if (now) + goto purge_now; + if (!batadv_has_timed_out(backbone_gw->lasttime, + BATADV_BLA_BACKBONE_TIMEOUT)) + continue; + + batadv_dbg(BATADV_DBG_BLA, backbone_gw->bat_priv, + "%s(): backbone gw %pM timed out\n", + __func__, backbone_gw->orig); purge_now: - /* don't wait for the pending request anymore */ - if (atomic_read(&backbone_gw->request_sent)) - atomic_dec(&bat_priv->bla.num_requests); + purged = true; - batadv_bla_del_backbone_claims(backbone_gw); + /* don't wait for the pending request anymore */ + if (atomic_read(&backbone_gw->request_sent)) + atomic_dec(&bat_priv->bla.num_requests); - hlist_del_rcu(&backbone_gw->hash_entry); - batadv_backbone_gw_put(backbone_gw); - } - spin_unlock_bh(list_lock); + batadv_bla_del_backbone_claims(backbone_gw); + + hlist_del_rcu(&backbone_gw->hash_entry); + break; + } + spin_unlock_bh(list_lock); + + if (purged) { + /* reference for pending report_work */ + if (cancel_work_sync(&backbone_gw->report_work)) + batadv_backbone_gw_put(backbone_gw); + + /* reference for hash_entry */ + batadv_backbone_gw_put(backbone_gw); + } + } while (purged); } } -- cgit v1.2.3 From 83ab69bd12b80f6ea169c8bea6977701b53a043d Mon Sep 17 00:00:00 2001 From: Sven Eckelmann Date: Tue, 12 May 2026 09:13:31 +0200 Subject: batman-adv: bla: avoid double decrement of bla.num_requests The bla.num_requests is increased when no request_sent was in progress. And it is decremented in various places (announcement was received, backbone is purged, periodic work). But the check if the request_sent is actually set to a specific state and the atomic_dec/_inc are not safe because they are not atomic (TOCTOU) and multiple such code portions can run concurrently. At the same time, it is necessary to modify request_sent (state) and bla.num_requests atomically. Otherwise batadv_bla_send_request() might set request_sent to 1 and is interrupted. batadv_handle_announce() can then set request_sent back to 0 and decrement num_requests before batadv_bla_send_request() incremented it. The two operations must therefore be locked. And since state (request_sent) and wait_periods are only accessed inside this lock, they can be converted to simpler datatypes. And to avoid that the bla.num_requests is touched by a parallel running context with a valid backbone_gw reference after batadv_bla_purge_backbone_gw() ran, a third state "stopped" is required to correctly signal that a backbone_gw is in the state of being cleaned up. Cc: stable@kernel.org Fixes: 23721387c409 ("batman-adv: add basic bridge loop avoidance code") Signed-off-by: Sven Eckelmann --- net/batman-adv/bridge_loop_avoidance.c | 51 +++++++++++++++++++++++----------- net/batman-adv/mesh-interface.c | 1 + net/batman-adv/types.h | 39 ++++++++++++++++++++------ 3 files changed, 67 insertions(+), 24 deletions(-) (limited to 'net') diff --git a/net/batman-adv/bridge_loop_avoidance.c b/net/batman-adv/bridge_loop_avoidance.c index df1dfdf4a1a1..1bef12e659cb 100644 --- a/net/batman-adv/bridge_loop_avoidance.c +++ b/net/batman-adv/bridge_loop_avoidance.c @@ -514,8 +514,8 @@ batadv_bla_get_backbone_gw(struct batadv_priv *bat_priv, const u8 *orig, entry->crc = BATADV_BLA_CRC_INIT; entry->bat_priv = bat_priv; spin_lock_init(&entry->crc_lock); - atomic_set(&entry->request_sent, 0); - atomic_set(&entry->wait_periods, 0); + entry->state = BATADV_BLA_BACKBONE_GW_SYNCED; + entry->wait_periods = 0; ether_addr_copy(entry->orig, orig); INIT_WORK(&entry->report_work, batadv_bla_loopdetect_report); kref_init(&entry->refcount); @@ -544,9 +544,13 @@ batadv_bla_get_backbone_gw(struct batadv_priv *bat_priv, const u8 *orig, batadv_bla_send_announce(bat_priv, entry); /* this will be decreased in the worker thread */ - atomic_inc(&entry->request_sent); - atomic_set(&entry->wait_periods, BATADV_BLA_WAIT_PERIODS); - atomic_inc(&bat_priv->bla.num_requests); + spin_lock_bh(&bat_priv->bla.num_requests_lock); + if (entry->state == BATADV_BLA_BACKBONE_GW_SYNCED) { + entry->state = BATADV_BLA_BACKBONE_GW_UNSYNCED; + entry->wait_periods = BATADV_BLA_WAIT_PERIODS; + atomic_inc(&bat_priv->bla.num_requests); + } + spin_unlock_bh(&bat_priv->bla.num_requests_lock); } return entry; @@ -649,10 +653,12 @@ static void batadv_bla_send_request(struct batadv_bla_backbone_gw *backbone_gw) backbone_gw->vid, BATADV_CLAIM_TYPE_REQUEST); /* no local broadcasts should be sent or received, for now. */ - if (!atomic_read(&backbone_gw->request_sent)) { + spin_lock_bh(&backbone_gw->bat_priv->bla.num_requests_lock); + if (backbone_gw->state == BATADV_BLA_BACKBONE_GW_SYNCED) { + backbone_gw->state = BATADV_BLA_BACKBONE_GW_UNSYNCED; atomic_inc(&backbone_gw->bat_priv->bla.num_requests); - atomic_set(&backbone_gw->request_sent, 1); } + spin_unlock_bh(&backbone_gw->bat_priv->bla.num_requests_lock); } /** @@ -873,10 +879,12 @@ static bool batadv_handle_announce(struct batadv_priv *bat_priv, u8 *an_addr, /* if we have sent a request and the crc was OK, * we can allow traffic again. */ - if (atomic_read(&backbone_gw->request_sent)) { + spin_lock_bh(&bat_priv->bla.num_requests_lock); + if (backbone_gw->state == BATADV_BLA_BACKBONE_GW_UNSYNCED) { + backbone_gw->state = BATADV_BLA_BACKBONE_GW_SYNCED; atomic_dec(&backbone_gw->bat_priv->bla.num_requests); - atomic_set(&backbone_gw->request_sent, 0); } + spin_unlock_bh(&bat_priv->bla.num_requests_lock); } batadv_backbone_gw_put(backbone_gw); @@ -1255,9 +1263,13 @@ purge_now: purged = true; /* don't wait for the pending request anymore */ - if (atomic_read(&backbone_gw->request_sent)) + spin_lock_bh(&bat_priv->bla.num_requests_lock); + if (backbone_gw->state == BATADV_BLA_BACKBONE_GW_UNSYNCED) atomic_dec(&bat_priv->bla.num_requests); + backbone_gw->state = BATADV_BLA_BACKBONE_GW_STOPPED; + spin_unlock_bh(&bat_priv->bla.num_requests_lock); + batadv_bla_del_backbone_claims(backbone_gw); hlist_del_rcu(&backbone_gw->hash_entry); @@ -1508,7 +1520,7 @@ static void batadv_bla_periodic_work(struct work_struct *work) batadv_bla_send_loopdetect(bat_priv, backbone_gw); - /* request_sent is only set after creation to avoid + /* state is only set to unsynced after creation to avoid * problems when we are not yet known as backbone gw * in the backbone. * @@ -1517,14 +1529,21 @@ static void batadv_bla_periodic_work(struct work_struct *work) * some grace time. */ - if (atomic_read(&backbone_gw->request_sent) == 0) - continue; + spin_lock_bh(&bat_priv->bla.num_requests_lock); + if (backbone_gw->state != BATADV_BLA_BACKBONE_GW_UNSYNCED) + goto unlock_next; - if (!atomic_dec_and_test(&backbone_gw->wait_periods)) - continue; + if (backbone_gw->wait_periods > 0) + backbone_gw->wait_periods--; + + if (backbone_gw->wait_periods > 0) + goto unlock_next; + backbone_gw->state = BATADV_BLA_BACKBONE_GW_SYNCED; atomic_dec(&backbone_gw->bat_priv->bla.num_requests); - atomic_set(&backbone_gw->request_sent, 0); + +unlock_next: + spin_unlock_bh(&bat_priv->bla.num_requests_lock); } rcu_read_unlock(); } diff --git a/net/batman-adv/mesh-interface.c b/net/batman-adv/mesh-interface.c index 56ca1c1b83f2..e7aa45bc6b7a 100644 --- a/net/batman-adv/mesh-interface.c +++ b/net/batman-adv/mesh-interface.c @@ -787,6 +787,7 @@ static int batadv_meshif_init_late(struct net_device *dev) atomic_set(&bat_priv->tt.ogm_append_cnt, 0); #ifdef CONFIG_BATMAN_ADV_BLA atomic_set(&bat_priv->bla.num_requests, 0); + spin_lock_init(&bat_priv->bla.num_requests_lock); #endif atomic_set(&bat_priv->tp_num, 0); diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h index 821ada05d86a..a01ee46d97f3 100644 --- a/net/batman-adv/types.h +++ b/net/batman-adv/types.h @@ -1026,6 +1026,12 @@ struct batadv_priv_bla { /** @num_requests: number of bla requests in flight */ atomic_t num_requests; + /** + * @num_requests_lock: locks update num_requests + + * batadv_backbone_gw::state + batadv_backbone_gw::wait_periods update + */ + spinlock_t num_requests_lock; + /** * @claim_hash: hash table containing mesh nodes this host has claimed */ @@ -1672,6 +1678,27 @@ struct batadv_priv { #ifdef CONFIG_BATMAN_ADV_BLA +enum batadv_bla_backbone_gw_state { + /** + * @BATADV_BLA_BACKBONE_GW_STOPPED: backbone gw is being removed + * and it must not longer work on requests + */ + BATADV_BLA_BACKBONE_GW_STOPPED, + + /** + * @BATADV_BLA_BACKBONE_GW_UNSYNCED: backbone was detected out + * of sync and a request was send. No traffic is forwarded until the + * situation is resolved + */ + BATADV_BLA_BACKBONE_GW_UNSYNCED, + + /** + * @BATADV_BLA_BACKBONE_GW_SYNCED: backbone is consider to be in + * sync. traffic can be forwarded + */ + BATADV_BLA_BACKBONE_GW_SYNCED, +}; + /** * struct batadv_bla_backbone_gw - batman-adv gateway bridged into the LAN */ @@ -1697,16 +1724,12 @@ struct batadv_bla_backbone_gw { /** * @wait_periods: grace time for bridge forward delays and bla group * forming at bootup phase - no bcast traffic is formwared until it has - * elapsed + * elapsed. Must only be access with num_requests_lock. */ - atomic_t wait_periods; + u8 wait_periods; - /** - * @request_sent: if this bool is set to true we are out of sync with - * this backbone gateway - no bcast traffic is formwared until the - * situation was resolved - */ - atomic_t request_sent; + /** @state: sync state. Must only be access with num_requests_lock. */ + enum batadv_bla_backbone_gw_state state; /** @crc: crc16 checksum over all claims */ u16 crc; -- cgit v1.2.3 From f80d3d98d2ff78d9e2fe5d68b1f45948c4f7bd24 Mon Sep 17 00:00:00 2001 From: Sven Eckelmann Date: Tue, 19 May 2026 09:23:49 +0200 Subject: batman-adv: bla: avoid NULL-ptr deref for claim via dropped interface Without rtnl_lock held, a hardif might be retrieved as primary interface of a meshif, but then (while operating on this interface) getting decoupled from the mesh interface. In this case, the meshif still exists but the pointer from the primary hardif to the meshif is set to NULL. The mesh_iface must be checked first to be non-NULL before continuing to send an ARP request using meshif. Cc: stable@kernel.org Fixes: 23721387c409 ("batman-adv: add basic bridge loop avoidance code") Reported-by: Ido Schimmel Reported-by: syzbot+9fdcc9f05a98a540b816@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=9fdcc9f05a98a540b816 Signed-off-by: Sven Eckelmann --- net/batman-adv/bridge_loop_avoidance.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/batman-adv/bridge_loop_avoidance.c b/net/batman-adv/bridge_loop_avoidance.c index 1bef12e659cb..ffe854018bd3 100644 --- a/net/batman-adv/bridge_loop_avoidance.c +++ b/net/batman-adv/bridge_loop_avoidance.c @@ -356,12 +356,14 @@ static void batadv_bla_send_claim(struct batadv_priv *bat_priv, const u8 *mac, sizeof(local_claim_dest)); local_claim_dest.type = claimtype; - mesh_iface = primary_if->mesh_iface; + mesh_iface = READ_ONCE(primary_if->mesh_iface); + if (!mesh_iface) + goto out; skb = arp_create(ARPOP_REPLY, ETH_P_ARP, /* IP DST: 0.0.0.0 */ zeroip, - primary_if->mesh_iface, + mesh_iface, /* IP SRC: 0.0.0.0 */ zeroip, /* Ethernet DST: Broadcast */ -- cgit v1.2.3 From 50c2d91c5dfa0e465826ec1f8dbad9cdc254bd85 Mon Sep 17 00:00:00 2001 From: Shardul Bankar Date: Fri, 15 May 2026 06:27:32 +0200 Subject: mptcp: do not drop partial packets When a packet arrives with map_seq < ack_seq < end_seq, the beginning of the packet has already been acknowledged but the end contains new data. Currently the entire packet is dropped as "old data," forcing the sender to retransmit. Instead, skip the already-acked bytes by adjusting the skb offset and enqueue only the new portion. Update bytes_received and ack_seq to reflect the new data consumed. A previous attempt at this fix has been sent by Paolo Abeni [1], but had issues [2]: it also added a zero-window check and changed rcv_wnd_sent initialization, which caused test regressions. This version addresses only the partial packet handling without modifying receive window accounting. Fixes: ab174ad8ef76 ("mptcp: move ooo skbs into msk out of order queue.") Cc: stable@vger.kernel.org Link: https://lore.kernel.org/c9b426a4e163aa3c4fe8b80c79f1a610f47ae7d8.1763075056.git.pabeni@redhat.com [1] Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/600 [2] Signed-off-by: Shardul Bankar [pabeni@redhat.com: update map] Signed-off-by: Paolo Abeni Reviewed-by: Matthieu Baerts (NGI0) Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20260515-net-mptcp-misc-fixes-7-1-rc4-v2-1-701e96419f2f@kernel.org Signed-off-by: Paolo Abeni --- net/mptcp/protocol.c | 24 +++++++++++++++++++----- 1 file changed, 19 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index 4546a8b09884..859df49e16dc 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -397,12 +397,26 @@ static bool __mptcp_move_skb(struct sock *sk, struct sk_buff *skb) return false; } - /* old data, keep it simple and drop the whole pkt, sender - * will retransmit as needed, if needed. + /* Completely old data? */ + if (!after64(MPTCP_SKB_CB(skb)->end_seq, msk->ack_seq)) { + MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_DUPDATA); + mptcp_drop(sk, skb); + return false; + } + + /* Partial packet: map_seq < ack_seq < end_seq. + * Skip the already-acked bytes and enqueue the new data. */ - MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_DUPDATA); - mptcp_drop(sk, skb); - return false; + copy_len = MPTCP_SKB_CB(skb)->end_seq - msk->ack_seq; + MPTCP_SKB_CB(skb)->offset += msk->ack_seq - MPTCP_SKB_CB(skb)->map_seq; + MPTCP_SKB_CB(skb)->map_seq += msk->ack_seq - + MPTCP_SKB_CB(skb)->map_seq; + msk->bytes_received += copy_len; + WRITE_ONCE(msk->ack_seq, msk->ack_seq + copy_len); + + skb_set_owner_r(skb, sk); + __skb_queue_tail(&sk->sk_receive_queue, skb); + return true; } static void mptcp_stop_rtx_timer(struct sock *sk) -- cgit v1.2.3 From 51e398a3b8961b26a8c0a4ba9a777c5339791707 Mon Sep 17 00:00:00 2001 From: Li Xiasong Date: Fri, 15 May 2026 06:27:33 +0200 Subject: mptcp: pm: fix ADD_ADDR timer infinite retry on option space insufficient When TCP option space is insufficient (e.g., when sending ADD_ADDR with an IPv6 address and port while tcp_timestamps is enabled), the original code jumped to out_unlock without clearing the addr_signal flag. This caused mptcp_pm_add_timer to keep rescheduling indefinitely, not sending ADD_ADDR, preventing subsequent addresses in the endpoint list from being announced. Handle this case by clearing the ADD_ADDR signal and skipping the matching ADD_ADDR retransmission entry. The skip path cancels the matching timer (with id check) and advances PM state progression, preserving forward progress to subsequent PM work. This cancellation is inherently best-effort. A concurrent add_timer callback may already be running and may acquire pm.lock before the cancel path updates entry state. In that case, one final ADD_ADDR transmit attempt can still be executed. Once the cancel path sets entry->retrans_times to ADD_ADDR_RETRANS_MAX, the callback-side retrans_times check suppresses further ADD_ADDR retransmissions. Note that when an ADD_ADDR is being prepared, a pure-ACK is queued. On the output side, it means that it is fine to skip non-pure-ACK packets, when drop_other_suboptions is set: a pure-ACK will be processed soon after. Fixes: 00cfd77b9063 ("mptcp: retransmit ADD_ADDR when timeout") Cc: stable@vger.kernel.org Signed-off-by: Li Xiasong Reviewed-by: Matthieu Baerts (NGI0) Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20260515-net-mptcp-misc-fixes-7-1-rc4-v2-2-701e96419f2f@kernel.org Signed-off-by: Paolo Abeni --- net/mptcp/pm.c | 56 ++++++++++++++++++++++++++++++++++++++++++++++---------- 1 file changed, 46 insertions(+), 10 deletions(-) (limited to 'net') diff --git a/net/mptcp/pm.c b/net/mptcp/pm.c index 3c152bf66cd5..3e770c7407e1 100644 --- a/net/mptcp/pm.c +++ b/net/mptcp/pm.c @@ -364,7 +364,13 @@ static void mptcp_pm_add_timer(struct timer_list *timer) spin_lock_bh(&msk->pm.lock); - if (!mptcp_pm_should_add_signal_addr(msk)) { + /* The cancel path (mptcp_pm_del_add_timer()) can race with this + * callback. Once cancel updates retrans_times to MAX, suppress further + * retransmissions here. If this callback acquires pm.lock first, one + * final transmit attempt is still possible. + */ + if (entry->retrans_times < ADD_ADDR_RETRANS_MAX && + !mptcp_pm_should_add_signal_addr(msk)) { pr_debug("retransmit ADD_ADDR id=%d\n", entry->addr.id); mptcp_pm_announce_addr(msk, &entry->addr, false); mptcp_pm_add_addr_send_ack(msk); @@ -414,8 +420,12 @@ mptcp_pm_del_add_timer(struct mptcp_sock *msk, /* Note: entry might have been removed by another thread. * We hold rcu_read_lock() to ensure it is not freed under us. */ - if (stop_timer) - sk_stop_timer_sync(sk, &entry->add_timer); + if (stop_timer) { + if (check_id) + sk_stop_timer(sk, &entry->add_timer); + else + sk_stop_timer_sync(sk, &entry->add_timer); + } rcu_read_unlock(); return entry; @@ -882,6 +892,7 @@ bool mptcp_pm_add_addr_signal(struct mptcp_sock *msk, const struct sk_buff *skb, struct mptcp_addr_info *addr, bool *echo, bool *drop_other_suboptions) { + bool skip_add_addr = false; int ret = false; u8 add_addr; u8 family; @@ -903,24 +914,49 @@ bool mptcp_pm_add_addr_signal(struct mptcp_sock *msk, const struct sk_buff *skb, } *echo = mptcp_pm_should_add_signal_echo(msk); - port = !!(*echo ? msk->pm.remote.port : msk->pm.local.port); - - family = *echo ? msk->pm.remote.family : msk->pm.local.family; - if (remaining < mptcp_add_addr_len(family, *echo, port)) - goto out_unlock; - if (*echo) { *addr = msk->pm.remote; add_addr = msk->pm.addr_signal & ~BIT(MPTCP_ADD_ADDR_ECHO); + port = !!msk->pm.remote.port; + family = msk->pm.remote.family; } else { *addr = msk->pm.local; add_addr = msk->pm.addr_signal & ~BIT(MPTCP_ADD_ADDR_SIGNAL); + port = !!msk->pm.local.port; + family = msk->pm.local.family; } - WRITE_ONCE(msk->pm.addr_signal, add_addr); + + if (remaining < mptcp_add_addr_len(family, *echo, port)) { + struct net *net = sock_net((struct sock *)msk); + + if (!*drop_other_suboptions) + goto out_unlock; + + if (*echo) { + MPTCP_INC_STATS(net, MPTCP_MIB_ECHOADDTXDROP); + } else { + skip_add_addr = true; + MPTCP_INC_STATS(net, MPTCP_MIB_ADDADDRTXDROP); + } + goto drop_signal_mark; + } + ret = true; +drop_signal_mark: + WRITE_ONCE(msk->pm.addr_signal, add_addr); + out_unlock: spin_unlock_bh(&msk->pm.lock); + + /* On pure-ACK option-space exhaustion, stop retrying this ADD_ADDR: + * clear the signal bit, cancel the matching retransmission timer, and + * let the PM state machine progress. + */ + if (skip_add_addr) { + mptcp_pm_del_add_timer(msk, addr, true); + mptcp_pm_subflow_established(msk); + } return ret; } -- cgit v1.2.3 From 0981f90e1a05773a4c29c6e720f5ea1e3c8f1876 Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Fri, 15 May 2026 06:27:35 +0200 Subject: mptcp: reset rcv wnd on disconnect If the MPTCP socket fallback to TCP before the MP handshake completion, the IASN remain 0, and the rcv_wnd_sent field is not explicitly initialized, just incremented over time with the data transfer. At disconnect time such value is not cleared. If the next connection falls back to TCP before the MP handshake completion, the data transfer will keep incrementing the receive window end sequence starting from the last value used in the previous connection: the announced window will be unrelated from the actual receiver buffer size and likely too big. Address the issue zeroing the field at disconnect time. Fixes: b29fcfb54cd7 ("mptcp: full disconnect implementation") Cc: stable@vger.kernel.org Signed-off-by: Paolo Abeni Reviewed-by: Matthieu Baerts (NGI0) Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20260515-net-mptcp-misc-fixes-7-1-rc4-v2-4-701e96419f2f@kernel.org Signed-off-by: Paolo Abeni --- net/mptcp/protocol.c | 1 + 1 file changed, 1 insertion(+) (limited to 'net') diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index 859df49e16dc..a72a6ad6ee8b 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -3487,6 +3487,7 @@ static int mptcp_disconnect(struct sock *sk, int flags) /* for fallback's sake */ WRITE_ONCE(msk->ack_seq, 0); + atomic64_set(&msk->rcv_wnd_sent, 0); WRITE_ONCE(sk->sk_shutdown, 0); sk_error_report(sk); -- cgit v1.2.3 From 3a543ae0e2092d5c2085d5f21f7a7dbafdffea3c Mon Sep 17 00:00:00 2001 From: Gang Yan Date: Fri, 15 May 2026 06:27:36 +0200 Subject: mptcp: update window_clamp on subflows when SO_RCVBUF is set Add __mptcp_subflow_set_rcvbuf() helper to write the subflow sk_rcvbuf, but also to call the recently added tcp_set_rcvbuf() helper to update window_clamp. This is needed because the window clap is updated when scaling_ratio changes, in tcp_measure_rcv_mss(). Until scaling_ratio changes, the subflow is stuck with the old window clamp which may be based on a small initial buffer. Use this new helper in both mptcp_sol_socket_sync_intval() (setsockopt path) and sync_socket_options() (new subflow creation path). Note that this patch depends on commit b025461303d8 ("tcp: update window_clamp when SO_RCVBUF is set"): it fixes the issue on TCP side, but the same fix is needed on MPTCP side as well. Fixes: a2cbb1603943 ("tcp: Update window clamping condition") Cc: stable@vger.kernel.org Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/619 Signed-off-by: Gang Yan Reviewed-by: Matthieu Baerts (NGI0) Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20260515-net-mptcp-misc-fixes-7-1-rc4-v2-5-701e96419f2f@kernel.org Signed-off-by: Paolo Abeni --- net/mptcp/sockopt.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/mptcp/sockopt.c b/net/mptcp/sockopt.c index 1cf608e7357b..87b5796d0135 100644 --- a/net/mptcp/sockopt.c +++ b/net/mptcp/sockopt.c @@ -67,6 +67,12 @@ static int mptcp_get_int_option(struct mptcp_sock *msk, sockptr_t optval, return 0; } +static void __mptcp_subflow_set_rcvbuf(struct sock *ssk, int val) +{ + WRITE_ONCE(ssk->sk_rcvbuf, val); + tcp_set_rcvbuf(ssk, val); +} + static void mptcp_sol_socket_sync_intval(struct mptcp_sock *msk, int optname, int val) { struct mptcp_subflow_context *subflow; @@ -100,7 +106,7 @@ static void mptcp_sol_socket_sync_intval(struct mptcp_sock *msk, int optname, in case SO_RCVBUF: case SO_RCVBUFFORCE: ssk->sk_userlocks |= SOCK_RCVBUF_LOCK; - WRITE_ONCE(ssk->sk_rcvbuf, sk->sk_rcvbuf); + __mptcp_subflow_set_rcvbuf(ssk, sk->sk_rcvbuf); break; case SO_MARK: if (READ_ONCE(ssk->sk_mark) != sk->sk_mark) { @@ -1560,7 +1566,7 @@ static void sync_socket_options(struct mptcp_sock *msk, struct sock *ssk) mptcp_subflow_ctx(ssk)->cached_sndbuf = sk->sk_sndbuf; } if (sk->sk_userlocks & SOCK_RCVBUF_LOCK) - WRITE_ONCE(ssk->sk_rcvbuf, sk->sk_rcvbuf); + __mptcp_subflow_set_rcvbuf(ssk, sk->sk_rcvbuf); } if (sock_flag(sk, SOCK_LINGER)) { -- cgit v1.2.3 From 4df78ff02629c7729168f0696a7a2123c389818d Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Sun, 17 May 2026 15:11:21 +0300 Subject: bridge: mcast: Fix a possible use-after-free when removing a bridge port When per-VLAN multicast snooping is enabled, the bridge iterates over all the bridge ports, disables the per-port multicast context on each port and enables the per-{port, VLAN} multicast contexts instead. The reverse happens when per-VLAN multicast snooping is disabled. When global multicast snooping is enabled, the bridge iterates over all the bridge ports and enables the per-port multicast context on each port. The reverse happens when multicast snooping is disabled. The above scheme can result in a situation where both types of contexts (per-port and per-{port, VLAN}) are enabled on a single bridge port: # ip link add name br1 up type bridge mcast_snooping 1 mcast_querier 1 vlan_filtering 1 # ip link add name dummy1 up master br1 type dummy # ip link set dev br1 type bridge mcast_vlan_snooping 1 # ip link set dev br1 type bridge mcast_snooping 0 # ip link set dev br1 type bridge mcast_snooping 1 This is not intended and it is a problem since the commit cited below. Prior to this commit, when removing a bridge port, br_multicast_disable_port() would disable the per-port multicast context and the per-{port, VLAN} multicast contexts would get disabled when flushing VLANs. After this commit, br_multicast_disable_port() only disables the per-port multicast context if per-VLAN multicast snooping is disabled. If both types of contexts were enabled on the port when it was removed, the per-port multicast context would remain enabled when freeing the bridge port, leading to a use-after-free [1]. Fix by preventing the bridge from enabling / disabling the per-port multicast contexts when toggling global multicast snooping if per-VLAN multicast snooping is enabled. [1] ODEBUG: free active (active state 0) object: ffff88810f8bda78 object type: timer_list hint: br_ip6_multicast_port_query_expired (net/bridge/br_multicast.c:1927) WARNING: lib/debugobjects.c:629 at debug_print_object+0x1b1/0x3e0, CPU#5: swapper/5/0 [...] Call Trace: __debug_check_no_obj_freed (lib/debugobjects.c:1116) kfree (mm/slub.c:2620 mm/slub.c:6250 mm/slub.c:6565) kobject_cleanup (lib/kobject.c:689) rcu_do_batch (kernel/rcu/tree.c:2617) rcu_core (kernel/rcu/tree.c:2869) handle_softirqs (kernel/softirq.c:622) __irq_exit_rcu (kernel/softirq.c:656 kernel/softirq.c:496 kernel/softirq.c:735) irq_exit_rcu (kernel/softirq.c:752) sysvec_apic_timer_interrupt (arch/x86/kernel/apic/apic.c:1061 (discriminator 47) arch/x86/kernel/apic/apic.c:1061 (discriminator 47)) Fixes: 4b30ae9adb04 ("net: bridge: mcast: re-implement br_multicast_{enable, disable}_port functions") Reported-by: syzbot+ae231e0552fa77b26ea1@syzkaller.appspotmail.com Closes: https://lore.kernel.org/netdev/87qznowlfs.ffs@tglx/ Reported-by: Thomas Gleixner Acked-by: Nikolay Aleksandrov Signed-off-by: Ido Schimmel Link: https://patch.msgid.link/20260517121122.188333-2-idosch@nvidia.com Signed-off-by: Jakub Kicinski --- net/bridge/br_multicast.c | 22 +++++++++++++++++----- 1 file changed, 17 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/bridge/br_multicast.c b/net/bridge/br_multicast.c index 881d866d687a..2eef4f3345cd 100644 --- a/net/bridge/br_multicast.c +++ b/net/bridge/br_multicast.c @@ -4640,10 +4640,24 @@ static void br_multicast_start_querier(struct net_bridge_mcast *brmctx, rcu_read_unlock(); } -static void br_multicast_del_grps(struct net_bridge *br) +static void br_multicast_enable_all_ports(struct net_bridge *br) { struct net_bridge_port *port; + if (br_opt_get(br, BROPT_MCAST_VLAN_SNOOPING_ENABLED)) + return; + + list_for_each_entry(port, &br->port_list, list) + __br_multicast_enable_port_ctx(&port->multicast_ctx); +} + +static void br_multicast_disable_all_ports(struct net_bridge *br) +{ + struct net_bridge_port *port; + + if (br_opt_get(br, BROPT_MCAST_VLAN_SNOOPING_ENABLED)) + return; + list_for_each_entry(port, &br->port_list, list) __br_multicast_disable_port_ctx(&port->multicast_ctx); } @@ -4651,7 +4665,6 @@ static void br_multicast_del_grps(struct net_bridge *br) int br_multicast_toggle(struct net_bridge *br, unsigned long val, struct netlink_ext_ack *extack) { - struct net_bridge_port *port; bool change_snoopers = false; int err = 0; @@ -4668,7 +4681,7 @@ int br_multicast_toggle(struct net_bridge *br, unsigned long val, br_opt_toggle(br, BROPT_MULTICAST_ENABLED, !!val); if (!br_opt_get(br, BROPT_MULTICAST_ENABLED)) { change_snoopers = true; - br_multicast_del_grps(br); + br_multicast_disable_all_ports(br); goto unlock; } @@ -4676,8 +4689,7 @@ int br_multicast_toggle(struct net_bridge *br, unsigned long val, goto unlock; br_multicast_open(br); - list_for_each_entry(port, &br->port_list, list) - __br_multicast_enable_port_ctx(&port->multicast_ctx); + br_multicast_enable_all_ports(br); change_snoopers = true; -- cgit v1.2.3 From d4ea0dfd75011b78cebf3808f98ac4c4f51a6fb9 Mon Sep 17 00:00:00 2001 From: Justin Iurman Date: Sun, 17 May 2026 20:30:59 +0200 Subject: ipv6: ioam: add NULL check for idev in ipv6_hop_ioam() Reported by Sashiko: The function ipv6_hop_ioam() accesses __in6_dev_get(skb->dev)->cnf.ioam6_enabled without validating the returned idev pointer. Because addrconf_ifdown() can concurrently clear dev->ip6_ptr via RCU, __in6_dev_get() can return NULL during interface teardown, which could cause a NULL pointer dereference when processing an IOAM Hop-by-Hop option. Let's add a check and use SKB_DROP_REASON_IPV6DISABLED accordingly. Fixes: 9ee11f0fff20 ("ipv6: ioam: Data plane support for Pre-allocated Trace") Cc: stable@vger.kernel.org Signed-off-by: Justin Iurman Reviewed-by: Ido Schimmel Link: https://patch.msgid.link/20260517183059.29140-1-justin.iurman@gmail.com Signed-off-by: Jakub Kicinski --- net/ipv6/exthdrs.c | 15 +++++++++++++-- 1 file changed, 13 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/ipv6/exthdrs.c b/net/ipv6/exthdrs.c index 03cbce842c1a..47c5502a34a2 100644 --- a/net/ipv6/exthdrs.c +++ b/net/ipv6/exthdrs.c @@ -910,16 +910,27 @@ static bool ipv6_hop_ra(struct sk_buff *skb, int optoff) static bool ipv6_hop_ioam(struct sk_buff *skb, int optoff) { + enum skb_drop_reason drop_reason; struct ioam6_trace_hdr *trace; struct ioam6_namespace *ns; + struct inet6_dev *idev; struct ioam6_hdr *hdr; + drop_reason = SKB_DROP_REASON_IP_INHDR; + /* Bad alignment (must be 4n-aligned) */ if (optoff & 3) goto drop; + /* Does the device still have IPv6 configuration? */ + idev = __in6_dev_get(skb->dev); + if (!idev) { + drop_reason = SKB_DROP_REASON_IPV6DISABLED; + goto drop; + } + /* Ignore if IOAM is not enabled on ingress */ - if (!READ_ONCE(__in6_dev_get(skb->dev)->cnf.ioam6_enabled)) + if (!READ_ONCE(idev->cnf.ioam6_enabled)) goto ignore; /* Truncated Option header */ @@ -972,7 +983,7 @@ ignore: return true; drop: - kfree_skb_reason(skb, SKB_DROP_REASON_IP_INHDR); + kfree_skb_reason(skb, drop_reason); return false; } -- cgit v1.2.3 From be309f8eae8b474a4a617eaae01324da996fc719 Mon Sep 17 00:00:00 2001 From: Jann Horn Date: Mon, 18 May 2026 18:51:30 +0200 Subject: af_unix: Fix UAF read of tail->len in unix_stream_data_wait() unix_stream_data_wait() does skb_peek_tail(&sk->sk_receive_queue) without holding any lock that prevents SKBs on that queue from being dequeued and freed. This has been the case since commit 79f632c71bea ("unix/stream: fix peeking with an offset larger than data in queue"). The first consequence of this is that the pointer comparison `tail != last` can be false even if `last` semantically refers to an already-freed SKB while `tail` is a new SKB allocated at the same address; which can cause unix_stream_data_wait() to wrongly keep blocking after new data has arrived, but only in a weird scenario where a peeking recv() and a normal recv() on the same socket are racing, which is probably not a real problem. But since commit 2b514574f7e8 ("net: af_unix: implement splice for stream af_unix sockets"), `tail` is actually dereferenced, which can cause UAF in the following race scenario (where test_setup() runs single-threaded, and afterwards, test_thread1() and test_thread2() run concurrently in two threads: ``` static int socks[2]; void test_setup(void) { socketpair(AF_UNIX, SOCK_STREAM, 0, socks); send(socks[1], "A", 1, 0); int peekoff = 1; setsockopt(socks[0], SOL_SOCKET, SO_PEEK_OFF, &peekoff, sizeof(peekoff)); } void test_thread1(void) { char dummy; recv(socks[0], &dummy, 1, MSG_PEEK); } void test_thread2(void) { char dummy; recv(socks[0], &dummy, 1, 0); shutdown(socks[1], SHUT_WR); } ``` when racing like this: ``` thread1 thread2 unix_stream_read_generic mutex_lock(&u->iolock) skb_peek(&sk->sk_receive_queue) skb_peek_next(skb, &sk->sk_receive_queue) mutex_unlock(&u->iolock) unix_stream_read_generic unix_state_lock(sk) skb_peek(&sk->sk_receive_queue) unix_state_unlock(sk) unix_stream_data_wait unix_state_lock(sk) tail = skb_peek_tail(&sk->sk_receive_queue) spin_lock(&sk->sk_receive_queue.lock) __skb_unlink(skb, &sk->sk_receive_queue) spin_unlock(&sk->sk_receive_queue.lock) consume_skb(skb) [frees the SKB] `tail != last`: false `tail`: true `tail->len != last_len` ***UAF*** ``` Fix the UAF by removing the read of tail->len; checking tail->len would only make sense if SKBs in the receive queue of a UNIX socket could grow, which can no longer happen. Kuniyuki explained: > When commit 869e7c62486e ("net: af_unix: implement stream sendpage > support") added sendpage() support, data could be appended to the last > skb in the receiver's queue. > > That's why we needed to check if the length of the last skb was changed > while waiting for new data in unix_stream_data_wait(). > > However, commit a0dbf5f818f9 ("af_unix: Support MSG_SPLICE_PAGES") and > commit 57d44a354a43 ("unix: Convert unix_stream_sendpage() to use > MSG_SPLICE_PAGES") refactored sendmsg(), and now data is always added > to a new skb. That means this fix is not suitable for kernels before 6.5. Fixes: 2b514574f7e8 ("net: af_unix: implement splice for stream af_unix sockets") Cc: stable@vger.kernel.org # 6.5.x Signed-off-by: Jann Horn Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20260518-b4-unix-recv-wait-hotfix-v2-1-83e29ce8ad31@google.com Signed-off-by: Jakub Kicinski --- net/unix/af_unix.c | 11 ++--------- 1 file changed, 2 insertions(+), 9 deletions(-) (limited to 'net') diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index 1cbf36ea043b..dc71ed79be4a 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -2711,8 +2711,7 @@ static int unix_read_skb(struct sock *sk, skb_read_actor_t recv_actor) * Sleep until more data has arrived. But check for races.. */ static long unix_stream_data_wait(struct sock *sk, long timeo, - struct sk_buff *last, unsigned int last_len, - bool freezable) + struct sk_buff *last, bool freezable) { unsigned int state = TASK_INTERRUPTIBLE | freezable * TASK_FREEZABLE; struct sk_buff *tail; @@ -2725,7 +2724,6 @@ static long unix_stream_data_wait(struct sock *sk, long timeo, tail = skb_peek_tail(&sk->sk_receive_queue); if (tail != last || - (tail && tail->len != last_len) || sk->sk_err || (sk->sk_shutdown & RCV_SHUTDOWN) || signal_pending(current) || @@ -2921,7 +2919,6 @@ static int unix_stream_read_generic(struct unix_stream_read_state *state, int flags = state->flags; bool check_creds = false; struct scm_cookie scm; - unsigned int last_len; struct unix_sock *u; int copied = 0; int err = 0; @@ -2967,7 +2964,6 @@ redo: goto unlock; } last = skb = skb_peek(&sk->sk_receive_queue); - last_len = last ? last->len : 0; again: #if IS_ENABLED(CONFIG_AF_UNIX_OOB) @@ -3001,8 +2997,7 @@ again: mutex_unlock(&u->iolock); - timeo = unix_stream_data_wait(sk, timeo, last, - last_len, freezable); + timeo = unix_stream_data_wait(sk, timeo, last, freezable); if (signal_pending(current)) { err = sock_intr_errno(timeo); @@ -3019,7 +3014,6 @@ unlock: while (skip >= unix_skb_len(skb)) { skip -= unix_skb_len(skb); last = skb; - last_len = skb->len; skb = skb_peek_next(skb, &sk->sk_receive_queue); if (!skb) goto again; @@ -3094,7 +3088,6 @@ unlock: skip = 0; last = skb; - last_len = skb->len; unix_state_lock(sk); skb = skb_peek_next(skb, &sk->sk_receive_queue); if (skb) -- cgit v1.2.3 From f718506edd2d9c6a308ded9d13c632bf7b7d5a2c Mon Sep 17 00:00:00 2001 From: Alexandru Hossu Date: Fri, 15 May 2026 12:29:08 +0200 Subject: wifi: mac80211: bounds-check link_id in ieee80211_ml_epcs IEEE80211_MLE_STA_EPCS_CONTROL_LINK_ID is 0x000f, so link_id extracted from a PRIO_ACCESS ML element PER_STA_PROFILE subelement can be 0..15. sdata->link[] has IEEE80211_MLD_MAX_NUM_LINKS (15) entries (indices 0..14), making index 15 out-of-bounds. A connected WiFi 7 AP can trigger this by sending an EPCS Enable Response action frame with a PER_STA_PROFILE subelement where link_id = 15. The unsolicited-notification path (dialog_token = 0) is reachable any time EPCS is already enabled, without any prior client request. sdata->link[15] reads into the first word of sdata->activate_links_work (a wiphy_work whose embedded list_head is non-NULL after INIT_LIST_HEAD), so the NULL check on the result does not catch the invalid access. The garbage pointer is then passed to ieee80211_sta_wmm_params(), which dereferences link->sdata and crashes the kernel. The same class of bug was fixed for ieee80211_ml_reconfiguration() by commit 162d331d833d ("wifi: mac80211: bounds-check link_id in ieee80211_ml_reconfiguration"). Fixes: de86c5f60839 ("wifi: mac80211: Add support for EPCS configuration") Signed-off-by: Alexandru Hossu Link: https://patch.msgid.link/20260515102908.1653088-1-hossu.alexandru@gmail.com Signed-off-by: Johannes Berg --- net/mac80211/mlme.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'net') diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c index 0a0f27836d57..ca1d29daf019 100644 --- a/net/mac80211/mlme.c +++ b/net/mac80211/mlme.c @@ -11232,6 +11232,9 @@ static void ieee80211_ml_epcs(struct ieee80211_sub_if_data *sdata, control = get_unaligned_le16(pos); link_id = control & IEEE80211_MLE_STA_EPCS_CONTROL_LINK_ID; + if (link_id >= IEEE80211_MLD_MAX_NUM_LINKS) + continue; + link = sdata_dereference(sdata->link[link_id], sdata); if (!link) continue; -- cgit v1.2.3 From e1e83feb8eae82cc9cc676db4c70f52fedc4735d Mon Sep 17 00:00:00 2001 From: Emmanuel Grumbach Date: Wed, 13 May 2026 17:06:27 +0300 Subject: wifi: mac80211: don't override max_amsdu_subframes In client mode, the extended capabilities are handled by the kernel looking at the association frame. When the supplicant installs the keys it calls sta_apply_parameters and it doesn't include the extended capabilities since those can't change after association. As a result, we overrode the max_amsdu_subframes that we set after association. Check that the ext_capa coming from the user space is valid before looking at it. If the ext_capa is NULL, it really means that the extended capabilities are not changed (as opposed to cleared). The default value for max_amsdu_subframes is 0, which means there is no limit. This value is valid and in case the association response frame does not have extended capabilities, this is the value we should use. Closes: https://bugzilla.kernel.org/show_bug.cgi?id=221079 Signed-off-by: Emmanuel Grumbach Reviewed-by: Johannes Berg Signed-off-by: Miri Korenblit Link: https://patch.msgid.link/20260513170623.828dbb58c782.Ifd2bfc190c26140e919127adb02ffddd7b551499@changeid Signed-off-by: Johannes Berg --- net/mac80211/cfg.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c index 7b77d57c9f96..f9ee9947a94d 100644 --- a/net/mac80211/cfg.c +++ b/net/mac80211/cfg.c @@ -2344,8 +2344,9 @@ static int sta_apply_parameters(struct ieee80211_local *local, sta->sta.max_sp = params->max_sp; } - ieee80211_sta_set_max_amsdu_subframes(sta, params->ext_capab, - params->ext_capab_len); + if (params->ext_capab) + ieee80211_sta_set_max_amsdu_subframes(sta, params->ext_capab, + params->ext_capab_len); /* * cfg80211 validates this (1-2007) and allows setting the AID -- cgit v1.2.3 From a74e893f30db64cdce0fc7a96d3baa417bcd55f5 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Fri, 8 May 2026 09:10:31 +0200 Subject: wifi: mac80211: fix MLE defragmentation If either reconf or EPCS multi-link element (MLE) is contained in a non-transmitted profile, the defragmentation routine is called with a pointer to the defragmented copy, but the original elements. This is incorrect for two reasons: - if the original defragmentation was needed, it will not find the correct data - if the original frame is at a higher address, the parsing will potentially overrun the heap data (though given the layout of the buffers, only into the new defragmentation buffer, and then it has to stop and fail once that's filled with copied data. Fix it by tracking the container along with the pointer and in doing so also unify the two almost identical defragmentation routines. Fixes: 4d70e9c5488d ("wifi: mac80211: defragment reconfiguration MLE when parsing") Reviewed-by: Miriam Rachel Korenblit Reviewed-by: Ilan Peer Link: https://patch.msgid.link/20260508091031.8a6c34613178.I4de16ebbce2d27f2f8f98fc49949c7a376c2fe8d@changeid Signed-off-by: Johannes Berg --- net/mac80211/parse.c | 71 +++++++++++++++++++++++----------------------------- 1 file changed, 31 insertions(+), 40 deletions(-) (limited to 'net') diff --git a/net/mac80211/parse.c b/net/mac80211/parse.c index 2b3632c6008a..666cdd5fd0ea 100644 --- a/net/mac80211/parse.c +++ b/net/mac80211/parse.c @@ -34,6 +34,13 @@ #include "led.h" #include "wep.h" +struct ieee80211_elem_defrag { + const struct element *elem; + /* container start/len */ + const u8 *start; + size_t len; +}; + struct ieee80211_elems_parse { /* must be first for kfree to work */ struct ieee802_11_elems elems; @@ -41,11 +48,7 @@ struct ieee80211_elems_parse { /* The basic Multi-Link element in the original elements */ const struct element *ml_basic_elem; - /* The reconfiguration Multi-Link element in the original elements */ - const struct element *ml_reconf_elem; - - /* The EPCS Multi-Link element in the original elements */ - const struct element *ml_epcs_elem; + struct ieee80211_elem_defrag ml_reconf, ml_epcs; bool multi_link_inner; bool skip_vendor; @@ -162,10 +165,14 @@ ieee80211_parse_extension_element(u32 *crc, } break; case IEEE80211_ML_CONTROL_TYPE_RECONF: - elems_parse->ml_reconf_elem = elem; + elems_parse->ml_reconf.elem = elem; + elems_parse->ml_reconf.start = params->start; + elems_parse->ml_reconf.len = params->len; break; case IEEE80211_ML_CONTROL_TYPE_PRIO_ACCESS: - elems_parse->ml_epcs_elem = elem; + elems_parse->ml_epcs.elem = elem; + elems_parse->ml_epcs.start = params->start; + elems_parse->ml_epcs.len = params->len; break; default: break; @@ -990,46 +997,27 @@ ieee80211_prep_mle_link_parse(struct ieee80211_elems_parse *elems_parse, sub->start, sub->len); } -static void -ieee80211_mle_defrag_reconf(struct ieee80211_elems_parse *elems_parse) -{ - struct ieee802_11_elems *elems = &elems_parse->elems; - ssize_t ml_len; - - ml_len = cfg80211_defragment_element(elems_parse->ml_reconf_elem, - elems->ie_start, - elems->total_len, - elems_parse->scratch_pos, - elems_parse->scratch + - elems_parse->scratch_len - - elems_parse->scratch_pos, - WLAN_EID_FRAGMENT); - if (ml_len < 0) - return; - elems->ml_reconf = (void *)elems_parse->scratch_pos; - elems->ml_reconf_len = ml_len; - elems_parse->scratch_pos += ml_len; -} - -static void -ieee80211_mle_defrag_epcs(struct ieee80211_elems_parse *elems_parse) +static const void * +ieee80211_mle_defrag(struct ieee80211_elems_parse *elems_parse, + struct ieee80211_elem_defrag *defrag, + size_t *out_len) { - struct ieee802_11_elems *elems = &elems_parse->elems; + const void *ret; ssize_t ml_len; - ml_len = cfg80211_defragment_element(elems_parse->ml_epcs_elem, - elems->ie_start, - elems->total_len, + ml_len = cfg80211_defragment_element(defrag->elem, + defrag->start, defrag->len, elems_parse->scratch_pos, elems_parse->scratch + elems_parse->scratch_len - elems_parse->scratch_pos, WLAN_EID_FRAGMENT); if (ml_len < 0) - return; - elems->ml_epcs = (void *)elems_parse->scratch_pos; - elems->ml_epcs_len = ml_len; + return NULL; + ret = elems_parse->scratch_pos; + *out_len = ml_len; elems_parse->scratch_pos += ml_len; + return ret; } struct ieee802_11_elems * @@ -1109,9 +1097,12 @@ ieee802_11_parse_elems_full(struct ieee80211_elems_parse_params *params) _ieee802_11_parse_elems_full(&sub, elems_parse, NULL); } - ieee80211_mle_defrag_reconf(elems_parse); - - ieee80211_mle_defrag_epcs(elems_parse); + elems->ml_reconf = ieee80211_mle_defrag(elems_parse, + &elems_parse->ml_reconf, + &elems->ml_reconf_len); + elems->ml_epcs = ieee80211_mle_defrag(elems_parse, + &elems_parse->ml_epcs, + &elems->ml_epcs_len); if (elems->tim && !elems->parse_error) { const struct ieee80211_tim_ie *tim_ie = elems->tim; -- cgit v1.2.3 From fe2d61a5d2849ee75dd4deeb2fe35f78d80721f8 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Fri, 8 May 2026 09:10:32 +0200 Subject: wifi: mac80211: fix multi-link element inheritance When parsing a beacon, mac80211 erroneously inherits any reconfiguration or EPCS multi-link elements from the outer elements into the multi-BSSID profile that's requested, if connected to a non-transmitted BSS, unless that profile has a non-inheritance element. This also happens if parsing a multi-BSSID profile that doesn't have a non-inheritance element. Fix this by having an empty non-inheritance element so cfg80211_is_element_inherited() is invoked in these cases and causes the parser to skip the elements that should never be inherited. Fixes: cf36cdef10e2 ("wifi: mac80211: Add support for parsing Reconfiguration Multi Link element") Fixes: 24711d60f849 ("wifi: mac80211: Support parsing EPCS ML element") Reviewed-by: Ilan Peer Reviewed-by: Benjamin Berg Link: https://patch.msgid.link/20260508091032.92184c0a3f08.I3c43b0b63d2cef8a4ddddaef1c2faaeb1de711ad@changeid Signed-off-by: Johannes Berg --- net/mac80211/parse.c | 36 +++++++++++++++++++++++++++++++++--- 1 file changed, 33 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/mac80211/parse.c b/net/mac80211/parse.c index 666cdd5fd0ea..77894d997113 100644 --- a/net/mac80211/parse.c +++ b/net/mac80211/parse.c @@ -34,6 +34,15 @@ #include "led.h" #include "wep.h" +static const u8 empty_non_inheritance[] = { + WLAN_EID_EXTENSION, 1, WLAN_EID_EXT_NON_INHERITANCE, + /* + * cfg80211_is_element_inherited() hardcodes elements that + * cannot be inherited, so we just need an empty one to be + * calling it at all. + */ +}; + struct ieee80211_elem_defrag { const struct element *elem; /* container start/len */ @@ -923,7 +932,7 @@ ieee80211_prep_mle_link_parse(struct ieee80211_elems_parse *elems_parse, { struct ieee802_11_elems *elems = &elems_parse->elems; struct ieee80211_mle_per_sta_profile *prof; - const struct element *tmp; + const struct element *tmp, *ret; ssize_t ml_len; const u8 *end; @@ -993,8 +1002,17 @@ ieee80211_prep_mle_link_parse(struct ieee80211_elems_parse *elems_parse, sub->from_ap = params->from_ap; sub->link_id = -1; - return cfg80211_find_ext_elem(WLAN_EID_EXT_NON_INHERITANCE, - sub->start, sub->len); + ret = cfg80211_find_ext_elem(WLAN_EID_EXT_NON_INHERITANCE, + sub->start, sub->len); + if (ret) + return ret; + + /* + * Since we know we want and found a profile, apply an empty + * non-inheritance if the profile didn't have one, so that any + * element that shouldn't be inherited by spec isn't. + */ + return (const void *)empty_non_inheritance; } static const void * @@ -1030,6 +1048,7 @@ ieee802_11_parse_elems_full(struct ieee80211_elems_parse_params *params) size_t scratch_len = 3 * params->len; bool multi_link_inner = false; + BUILD_BUG_ON(sizeof(empty_non_inheritance) != empty_non_inheritance[1] + 2); BUILD_BUG_ON(offsetof(typeof(*elems_parse), elems) != 0); /* cannot parse for both a specific link and non-transmitted BSS */ @@ -1077,6 +1096,17 @@ ieee802_11_parse_elems_full(struct ieee80211_elems_parse_params *params) non_inherit = cfg80211_find_ext_elem(WLAN_EID_EXT_NON_INHERITANCE, sub.start, nontx_len); + /* + * If it's a non-transmitted BSS, we shouldn't pick + * any elements in the outer parsing that shouldn't + * be inherited. If the profile has a non-inheritance + * element this automatically happens, but if not then + * provide an empty one so that the hard-coded elements + * in cfg80211_is_element_inherited() are ignored, but + * it must be called. + */ + if (params->bss->transmitted_bss && !non_inherit) + non_inherit = (const void *)empty_non_inheritance; } else { /* must always parse to get elems_parse->ml_basic_elem */ non_inherit = ieee80211_prep_mle_link_parse(elems_parse, params, -- cgit v1.2.3 From d71c841be5d9e586ee7f36c0dc8ed4db0d9a1349 Mon Sep 17 00:00:00 2001 From: Zhao Li Date: Sat, 9 May 2026 12:34:28 +0800 Subject: wifi: mac80211: capture fast-RX rate before mesh reuses skb->cb ieee80211_invoke_fast_rx() reads RX status through IEEE80211_SKB_RXCB(skb), which aliases the same skb->cb storage that ieee80211_rx_mesh_data() reuses as IEEE80211_TX_INFO. In the unicast forward path, mesh_data does: info = IEEE80211_SKB_CB(fwd_skb); memset(info, 0, sizeof(*info)); on the same skb the caller still names via rx->skb, then either queues the skb for TX (success) or kfree_skb()'s it (no-route) before returning RX_QUEUED. The caller's RX_QUEUED arm then calls sta_stats_encode_rate(status) on memory that is either zeroed (success path) or freed (no-route path). The latter is KASAN slab-use-after-free in ieee80211_prepare_and_rx_handle. Fix by encoding the rate from status before invoking ieee80211_rx_mesh_data(), so the RX_QUEUED arm consumes a value captured while status was still backed by valid memory. Fixes: 3468e1e0c639 ("wifi: mac80211: add mesh fast-rx support") Cc: stable@vger.kernel.org Signed-off-by: Zhao Li Link: https://patch.msgid.link/20260509043427.60322-2-enderaoelyther@gmail.com Signed-off-by: Johannes Berg --- net/mac80211/rx.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c index d18e962126ce..3fb40449c6c5 100644 --- a/net/mac80211/rx.c +++ b/net/mac80211/rx.c @@ -4984,6 +4984,7 @@ static bool ieee80211_invoke_fast_rx(struct ieee80211_rx_data *rx, u8 sa[ETH_ALEN]; } addrs __aligned(2); struct ieee80211_sta_rx_stats *stats; + u32 encoded_rate; /* for parallel-rx, we need to have DUP_VALIDATED, otherwise we write * to a common data structure; drivers can implement that per queue @@ -5091,11 +5092,14 @@ static bool ieee80211_invoke_fast_rx(struct ieee80211_rx_data *rx, /* push the addresses in front */ memcpy(skb_push(skb, sizeof(addrs)), &addrs, sizeof(addrs)); + /* capture before mesh forward may memset or free skb->cb */ + encoded_rate = sta_stats_encode_rate(status); + res = ieee80211_rx_mesh_data(rx->sdata, rx->sta, rx->skb); switch (res) { case RX_QUEUED: stats->last_rx = jiffies; - stats->last_rate = sta_stats_encode_rate(status); + stats->last_rate = encoded_rate; return true; case RX_CONTINUE: break; -- cgit v1.2.3 From a6e6ccd5bd07155c2add6c74ce1a5e68ad3b95ea Mon Sep 17 00:00:00 2001 From: Michael Bommarito Date: Fri, 15 May 2026 11:17:18 -0400 Subject: wifi: mac80211: consume only present negotiated TTLM maps ieee80211_tid_to_link_map_size_ok() validates negotiated TTLM elements against the number of link-map entries indicated by link_map_presence. ieee80211_parse_neg_ttlm() must consume the same layout. The parser advanced its cursor for every TID, including TIDs whose presence bit is clear and therefore have no map bytes in the element. A sparse map can then make a later present TID read past the validated element. The bad bytes land in neg_ttlm->{up,down}link[tid] but are gated by valid_links before being applied to driver state, so a peer cannot turn the read into a policy change. Under KUnit + KASAN with an exact-sized element allocation the OOB read is reported as a slab-out-of-bounds; whether the same trigger fires under the production RX path depends on surrounding allocator state. Advance the cursor only when the current TID has a map present. Fixes: 8f500fbc6c65 ("wifi: mac80211: process and save negotiated TID to Link mapping request") Cc: stable@vger.kernel.org Assisted-by: Claude:claude-opus-4-7 Signed-off-by: Michael Bommarito Link: https://patch.msgid.link/20260515151719.1317659-2-michael.bommarito@gmail.com Signed-off-by: Johannes Berg --- net/mac80211/mlme.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c index ca1d29daf019..b98ddfa3003e 100644 --- a/net/mac80211/mlme.c +++ b/net/mac80211/mlme.c @@ -8164,6 +8164,7 @@ ieee80211_parse_neg_ttlm(struct ieee80211_sub_if_data *sdata, "No active links for TID %d", tid); return -EINVAL; } + pos += map_size; } else { map = 0; } @@ -8182,7 +8183,6 @@ ieee80211_parse_neg_ttlm(struct ieee80211_sub_if_data *sdata, default: return -EINVAL; } - pos += map_size; } return 0; } -- cgit v1.2.3 From dc14686f27df6454b13b16ad1c9203ab3e9b0375 Mon Sep 17 00:00:00 2001 From: Kartik Nair Date: Mon, 11 May 2026 01:54:37 +0530 Subject: wifi: cfg80211: wext: validate chandef in monitor mode cfg80211_wext_siwfreq() constructs a channel definition for monitor mode but passes it to cfg80211_set_monitor_channel() without first validating it with cfg80211_chandef_valid(). This causes a WARN_ON in cfg80211_chandef_dfs_required() when it receives an invalid chandef. Add the missing cfg80211_chandef_valid() check before calling cfg80211_set_monitor_channel() to return -EINVAL early on invalid channel definitions, consistent with how other callers handle this. Reported-by: syzbot+02a1a03b8622d3c7d1c9@syzkaller.appspotmail.com Signed-off-by: Kartik Nair Link: https://patch.msgid.link/20260510202437.7857-1-contact.kartikn@gmail.com [clarify subject] Signed-off-by: Johannes Berg --- net/wireless/wext-compat.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'net') diff --git a/net/wireless/wext-compat.c b/net/wireless/wext-compat.c index 22d9d9bae8f5..63d145b524c9 100644 --- a/net/wireless/wext-compat.c +++ b/net/wireless/wext-compat.c @@ -789,6 +789,8 @@ static int cfg80211_wext_siwfreq(struct net_device *dev, chandef.chan = ieee80211_get_channel(&rdev->wiphy, freq); if (!chandef.chan) return -EINVAL; + if (!cfg80211_chandef_valid(&chandef)) + return -EINVAL; return cfg80211_set_monitor_channel(rdev, dev, &chandef); case NL80211_IFTYPE_MESH_POINT: freq = cfg80211_wext_freq(wextfreq); -- cgit v1.2.3 From 59e932ded949fa6f0340bf7c6d7818f962fa4fd2 Mon Sep 17 00:00:00 2001 From: Jann Horn Date: Tue, 12 May 2026 22:15:39 +0200 Subject: Bluetooth: bnep: Fix UAF read of dev->name bnep_add_connection() needs to keep holding the bnep_session_sem while reading dev->name (just like bnep_get_connlist() does); otherwise the bnep_session() thread can concurrently free the net_device, which can for example be triggered by a concurrent bnep_del_connection(). (This UAF is fairly uninteresting from a security perspective; calling bnep_add_connection() requires passing a capable(CAP_NET_ADMIN) check. It also requires completely tearing down a netdev during a fairly tight race window.) Cc: stable@vger.kernel.org Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: Jann Horn Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/bnep/core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/bluetooth/bnep/core.c b/net/bluetooth/bnep/core.c index 853c8d7644b5..0de5df690bd0 100644 --- a/net/bluetooth/bnep/core.c +++ b/net/bluetooth/bnep/core.c @@ -645,8 +645,8 @@ int bnep_add_connection(struct bnep_connadd_req *req, struct socket *sock) goto failed; } - up_write(&bnep_session_sem); strcpy(req->device, dev->name); + up_write(&bnep_session_sem); return 0; failed: -- cgit v1.2.3 From 23d528d817a485fe9800a66c9411bd9e3d8a6f63 Mon Sep 17 00:00:00 2001 From: Luiz Augusto von Dentz Date: Thu, 14 May 2026 09:42:24 -0400 Subject: Bluetooth: hci_sync: Fix not setting mask for HCI_EVT_LE_ALL_REMOTE_FEATURES_COMPLETE This fixes not setting the bit for HCI_EVT_LE_ALL_REMOTE_FEATURES_COMPLETE when extended features bit is set otherwise the controller may not generate HCI_EVT_LE_ALL_REMOTE_FEATURES_COMPLETE causing hci_le_read_all_remote_features_sync to timeout waiting for it. Also remove dead code. Fixes: a106e50be74b ("Bluetooth: HCI: Add support for LL Extended Feature Set") Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/hci_sync.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c index fd3aacdea512..aff8562a8690 100644 --- a/net/bluetooth/hci_sync.c +++ b/net/bluetooth/hci_sync.c @@ -4438,6 +4438,9 @@ static int hci_le_set_event_mask_sync(struct hci_dev *hdev) events[4] |= 0x02; /* LE BIG Info Advertising Report */ } + if (ll_ext_feature_capable(hdev)) + events[5] |= BIT(2); + if (le_cs_capable(hdev)) { /* Channel Sounding events */ events[5] |= 0x08; /* LE CS Read Remote Supported Cap Complete event */ @@ -7413,9 +7416,6 @@ static int hci_le_read_all_remote_features_sync(struct hci_dev *hdev, sizeof(cp), &cp, HCI_EVT_LE_ALL_REMOTE_FEATURES_COMPLETE, HCI_CMD_TIMEOUT, NULL); - - return __hci_cmd_sync_status(hdev, HCI_OP_LE_READ_ALL_REMOTE_FEATURES, - sizeof(cp), &cp, HCI_CMD_TIMEOUT); } static int hci_le_read_remote_features_sync(struct hci_dev *hdev, void *data) -- cgit v1.2.3 From 84c24fb151fc1179355296d7ff29129ac7c42129 Mon Sep 17 00:00:00 2001 From: David Carlier Date: Fri, 15 May 2026 07:25:25 +0100 Subject: Bluetooth: ISO: drop ISO_END frames received without prior ISO_START ISO data PDUs carry a packet-boundary flag indicating START, CONT, END or SINGLE. The ISO_CONT branch of iso_recv() guards against a missing ISO_START by checking conn->rx_len before touching conn->rx_skb, but ISO_END does not. If a peer sends an ISO_END as the first packet on a fresh ISO connection, conn->rx_skb is still NULL and conn->rx_len is zero, so skb_put(conn->rx_skb, ...) dereferences NULL and oopses. For BIS, where receivers sync to a broadcaster without pairing, any broadcaster on the air can trigger this. Mirror the ISO_CONT check at the top of ISO_END so a stray end fragment is logged and dropped instead of crashing the host. Fixes: ccf74f2390d6 ("Bluetooth: Add BTPROTO_ISO socket type") Cc: stable@vger.kernel.org Assisted-by: Claude:claude-opus-4-7 Signed-off-by: David Carlier Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/iso.c | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'net') diff --git a/net/bluetooth/iso.c b/net/bluetooth/iso.c index 7cb2864fe872..b971281f0a2b 100644 --- a/net/bluetooth/iso.c +++ b/net/bluetooth/iso.c @@ -2593,6 +2593,11 @@ int iso_recv(struct hci_dev *hdev, u16 handle, struct sk_buff *skb, u16 flags) break; case ISO_END: + if (!conn->rx_len) { + BT_ERR("Unexpected end frame (len %d)", skb->len); + goto drop; + } + skb_copy_from_linear_data(skb, skb_put(conn->rx_skb, skb->len), skb->len); conn->rx_len -= skb->len; -- cgit v1.2.3 From d3f7d17960ed50df3a6709c5158caff989c8c905 Mon Sep 17 00:00:00 2001 From: Michael Bommarito Date: Fri, 15 May 2026 10:38:19 -0400 Subject: Bluetooth: MGMT: validate Add Extended Advertising Data length MGMT_OP_ADD_EXT_ADV_DATA is registered as a variable-length command, with MGMT_ADD_EXT_ADV_DATA_SIZE as the fixed header size. The handler then uses cp->adv_data_len and cp->scan_rsp_len to validate and copy cp->data, but it never checks that those bytes are part of the mgmt command payload. A short command can therefore make add_ext_adv_data() pass an out-of-bounds pointer into tlv_data_is_valid(). If the bytes beyond the command buffer are addressable, they can also be copied into the advertising instance as scan response data, where the caller can read them back via MGMT_OP_GET_ADV_INSTANCE. The trigger requires CAP_NET_ADMIN in the initial user namespace; KASAN reports an 8-byte slab-out-of-bounds read. Reject commands whose length does not match the fixed header plus both advertising data lengths before parsing cp->data. Fixes: 12410572833a ("Bluetooth: Break add adv into two mgmt commands") Cc: stable@vger.kernel.org Assisted-by: Claude:claude-opus-4-7 Signed-off-by: Michael Bommarito Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/mgmt.c | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'net') diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c index b05bb380e5f8..de5bd6b637b2 100644 --- a/net/bluetooth/mgmt.c +++ b/net/bluetooth/mgmt.c @@ -9110,9 +9110,15 @@ static int add_ext_adv_data(struct sock *sk, struct hci_dev *hdev, void *data, struct adv_info *adv_instance; int err = 0; struct mgmt_pending_cmd *cmd; + u16 expected_len; BT_DBG("%s", hdev->name); + expected_len = struct_size(cp, data, cp->adv_data_len + cp->scan_rsp_len); + if (expected_len != data_len) + return mgmt_cmd_status(sk, hdev->id, MGMT_OP_ADD_EXT_ADV_DATA, + MGMT_STATUS_INVALID_PARAMS); + hci_dev_lock(hdev); adv_instance = hci_find_adv_instance(hdev, cp->instance); -- cgit v1.2.3 From ab1513597c6cf17cd1ad2a21e3b045421b48e022 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Safa=20Karaku=C5=9F?= Date: Sat, 16 May 2026 21:15:04 +0300 Subject: Bluetooth: fix UAF in l2cap_sock_cleanup_listen() vs l2cap_conn_del() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit bt_accept_dequeue() unlinks a not-yet-accepted child from the parent accept queue and release_sock()s it before returning, so the returned sk has no caller reference and is unlocked. l2cap_sock_cleanup_listen() walks these children on listening-socket close. A concurrent HCI disconnect drives hci_rx_work -> l2cap_conn_del() which runs l2cap_chan_del() + l2cap_sock_kill() and frees the child sk and its l2cap_chan; cleanup_listen() then uses both: BUG: KASAN: slab-use-after-free in l2cap_sock_kill l2cap_sock_kill / l2cap_sock_cleanup_listen / __x64_sys_close Freed by: l2cap_conn_del -> l2cap_sock_close_cb -> l2cap_sock_kill This is distinct from the two fixes already in this area: commit e83f5e24da741 ("Bluetooth: serialize accept_q access") serialises the accept_q list/poll and takes temporary refs inside bt_accept_dequeue(), and CVE-2025-39860 serialises the userspace close()/accept() race by calling cleanup_listen() under lock_sock() in l2cap_sock_release(). Neither covers l2cap_conn_del() running from hci_rx_work, so this UAF still reproduces on current bluetooth/master. Take the reference at the source: bt_accept_dequeue() does sock_hold() while sk is still locked, before release_sock(); callers sock_put(). cleanup_listen() pins the chan with l2cap_chan_hold_unless_zero() under a brief child sk lock (serialising vs l2cap_sock_teardown_cb()), drops it before l2cap_chan_lock(), and skips a duplicate l2cap_sock_kill() on SOCK_DEAD. conn->lock is not taken here: cleanup_listen() runs under the parent sk lock and that would invert conn->lock -> chan->lock -> sk_lock (lockdep). KASAN/SMP: an unprivileged listen/close vs HCI-disconnect race produced 12 use-after-free reports per run before this change; 0, and no lockdep report, over 1600+ raced iterations after it on bluetooth/master. Fixes: 15f02b910562 ("Bluetooth: L2CAP: Add initial code for Enhanced Credit Based Mode") Cc: stable@vger.kernel.org Reported-by: Siwei Zhang Reviewed-by: Siwei Zhang Signed-off-by: Safa Karakuş Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/af_bluetooth.c | 10 +++++++++ net/bluetooth/iso.c | 9 +++++++- net/bluetooth/l2cap_sock.c | 51 ++++++++++++++++++++++++++++++++++++++------ net/bluetooth/rfcomm/sock.c | 9 +++++++- net/bluetooth/sco.c | 9 +++++++- 5 files changed, 78 insertions(+), 10 deletions(-) (limited to 'net') diff --git a/net/bluetooth/af_bluetooth.c b/net/bluetooth/af_bluetooth.c index 9d68dd86023c..1a6aa3f8d4d6 100644 --- a/net/bluetooth/af_bluetooth.c +++ b/net/bluetooth/af_bluetooth.c @@ -340,6 +340,16 @@ restart: if (newsock) sock_graft(sk, newsock); + /* Hand the caller a reference taken while sk is + * still locked. bt_accept_unlink() just dropped + * the accept-queue reference; without this hold a + * concurrent teardown (e.g. l2cap_conn_del() -> + * l2cap_sock_kill()) could free sk between + * release_sock() and the caller using it. Every + * caller drops this with sock_put() when done. + */ + sock_hold(sk); + release_sock(sk); if (next) sock_put(next); diff --git a/net/bluetooth/iso.c b/net/bluetooth/iso.c index b971281f0a2b..d7af617cda45 100644 --- a/net/bluetooth/iso.c +++ b/net/bluetooth/iso.c @@ -751,6 +751,8 @@ static void iso_sock_cleanup_listen(struct sock *parent) while ((sk = bt_accept_dequeue(parent, NULL))) { iso_sock_close(sk); iso_sock_kill(sk); + /* Drop the reference handed back by bt_accept_dequeue(). */ + sock_put(sk); } /* If listening socket has a hcon, properly disconnect it */ @@ -1356,8 +1358,13 @@ static int iso_sock_accept(struct socket *sock, struct socket *newsock, } ch = bt_accept_dequeue(sk, newsock); - if (ch) + if (ch) { + /* Drop the bridging ref from bt_accept_dequeue(); + * the grafted socket keeps ch alive from here. + */ + sock_put(ch); break; + } if (!timeo) { err = -EAGAIN; diff --git a/net/bluetooth/l2cap_sock.c b/net/bluetooth/l2cap_sock.c index cf590a67d364..b34e7da8d906 100644 --- a/net/bluetooth/l2cap_sock.c +++ b/net/bluetooth/l2cap_sock.c @@ -349,8 +349,13 @@ static int l2cap_sock_accept(struct socket *sock, struct socket *newsock, } nsk = bt_accept_dequeue(sk, newsock); - if (nsk) + if (nsk) { + /* Drop the bridging ref from bt_accept_dequeue(); + * the grafted socket keeps nsk alive from here. + */ + sock_put(nsk); break; + } if (!timeo) { err = -EAGAIN; @@ -1475,22 +1480,54 @@ static void l2cap_sock_cleanup_listen(struct sock *parent) BT_DBG("parent %p state %s", parent, state_to_string(parent->sk_state)); - /* Close not yet accepted channels */ + /* Close not yet accepted channels. + * + * bt_accept_dequeue() now returns sk with an extra reference held + * (taken while sk was still locked) so a concurrent l2cap_conn_del() + * -> l2cap_sock_kill() cannot free sk under us. + * + * cleanup_listen() runs under the parent sk lock, so unlike + * l2cap_sock_shutdown() we must NOT take conn->lock here: that would + * establish sk_lock -> conn->lock and invert the established + * conn->lock -> chan->lock -> sk_lock order (lockdep deadlock). + * + * Instead, briefly take the child sk lock to fetch and pin its chan. + * l2cap_conn_del() reaches the chan free only via + * l2cap_chan_del() -> l2cap_sock_teardown_cb(), which itself takes + * the child sk lock; holding it across l2cap_chan_hold_unless_zero() + * therefore guarantees the chan cannot be freed while we read and + * pin it (hold_unless_zero() additionally skips a chan already past + * its last reference). We then drop the sk lock before taking + * chan->lock, so sk and chan locks are never held together. + */ while ((sk = bt_accept_dequeue(parent, NULL))) { - struct l2cap_chan *chan = l2cap_pi(sk)->chan; + struct l2cap_chan *chan; + + lock_sock_nested(sk, L2CAP_NESTING_NORMAL); + chan = l2cap_chan_hold_unless_zero(l2cap_pi(sk)->chan); + release_sock(sk); + if (!chan) { + /* l2cap_conn_del() already tearing this child down */ + sock_put(sk); + continue; + } BT_DBG("child chan %p state %s", chan, state_to_string(chan->state)); - l2cap_chan_hold(chan); l2cap_chan_lock(chan); - __clear_chan_timer(chan); l2cap_chan_close(chan, ECONNRESET); - l2cap_sock_kill(sk); - + /* l2cap_conn_del() may already have killed this socket + * (it sets SOCK_DEAD); skip the duplicate to avoid a + * double sock_put()/l2cap_chan_put(). + */ + if (!sock_flag(sk, SOCK_DEAD)) + l2cap_sock_kill(sk); l2cap_chan_unlock(chan); + l2cap_chan_put(chan); + sock_put(sk); } } diff --git a/net/bluetooth/rfcomm/sock.c b/net/bluetooth/rfcomm/sock.c index be6639cd6f59..bd7d959c6e9e 100644 --- a/net/bluetooth/rfcomm/sock.c +++ b/net/bluetooth/rfcomm/sock.c @@ -180,6 +180,8 @@ static void rfcomm_sock_cleanup_listen(struct sock *parent) while ((sk = bt_accept_dequeue(parent, NULL))) { rfcomm_sock_close(sk); rfcomm_sock_kill(sk); + /* Drop the reference handed back by bt_accept_dequeue(). */ + sock_put(sk); } parent->sk_state = BT_CLOSED; @@ -497,8 +499,13 @@ static int rfcomm_sock_accept(struct socket *sock, struct socket *newsock, } nsk = bt_accept_dequeue(sk, newsock); - if (nsk) + if (nsk) { + /* Drop the bridging ref from bt_accept_dequeue(); + * the grafted socket keeps nsk alive from here. + */ + sock_put(nsk); break; + } if (!timeo) { err = -EAGAIN; diff --git a/net/bluetooth/sco.c b/net/bluetooth/sco.c index eba44525d41d..f1799c6a6f87 100644 --- a/net/bluetooth/sco.c +++ b/net/bluetooth/sco.c @@ -502,6 +502,8 @@ static void sco_sock_cleanup_listen(struct sock *parent) while ((sk = bt_accept_dequeue(parent, NULL))) { sco_sock_close(sk); sco_sock_kill(sk); + /* Drop the reference handed back by bt_accept_dequeue(). */ + sock_put(sk); } parent->sk_state = BT_CLOSED; @@ -765,8 +767,13 @@ static int sco_sock_accept(struct socket *sock, struct socket *newsock, } ch = bt_accept_dequeue(sk, newsock); - if (ch) + if (ch) { + /* Drop the bridging ref from bt_accept_dequeue(); + * the grafted socket keeps ch alive from here. + */ + sock_put(ch); break; + } if (!timeo) { err = -EAGAIN; -- cgit v1.2.3 From 5f17ae0f595aeb560155ce98edbe44d3eacc7e40 Mon Sep 17 00:00:00 2001 From: Alice Mikityanska Date: Mon, 18 May 2026 09:22:49 +0300 Subject: udp: gso: Fix handling checksum in __udp_gso_segment The cited commit started using msslen for uh->len, but still uses newlen to adjust uh->check. Although the checksum is ignored in most cases due to the hardware offload, __udp_gso_segment attempts to maintain the correct one. Fix uh->check and adjust it by the right value. Additionally, after the fix, newlen becomes assigned and unused before the loop. The code can be simplified a bit if mss adjustment is dropped, so that newlen becomes equal to msslen before the loop, and msslen can be also dropped, saving a few lines of code. This brings us back to one variable, drops an unneeded arithmetic for mss, and fixes the UDP checksum. Fixes: b10b446ce7ad ("udp: gso: Use single MSS length in UDP header for GSO_PARTIAL") Signed-off-by: Alice Mikityanska Reviewed-by: Willem de Bruijn Signed-off-by: Gal Pressman Link: https://patch.msgid.link/20260518062250.3019914-2-gal@nvidia.com Signed-off-by: Jakub Kicinski --- net/ipv4/udp_offload.c | 13 ++----------- 1 file changed, 2 insertions(+), 11 deletions(-) (limited to 'net') diff --git a/net/ipv4/udp_offload.c b/net/ipv4/udp_offload.c index a0813d425b71..2578aa7f9ff9 100644 --- a/net/ipv4/udp_offload.c +++ b/net/ipv4/udp_offload.c @@ -482,11 +482,11 @@ struct sk_buff *__udp_gso_segment(struct sk_buff *gso_skb, struct sock *sk = gso_skb->sk; unsigned int sum_truesize = 0; struct sk_buff *segs, *seg; - __be16 newlen, msslen; struct udphdr *uh; unsigned int mss; bool copy_dtor; __sum16 check; + __be16 newlen; int ret = 0; mss = skb_shinfo(gso_skb)->gso_size; @@ -555,15 +555,6 @@ struct sk_buff *__udp_gso_segment(struct sk_buff *gso_skb, return segs; } - msslen = htons(sizeof(*uh) + mss); - - /* GSO partial and frag_list segmentation only requires splitting - * the frame into an MSS multiple and possibly a remainder, both - * cases return a GSO skb. So update the mss now. - */ - if (skb_is_gso(segs)) - mss *= skb_shinfo(segs)->gso_segs; - seg = segs; uh = udp_hdr(seg); @@ -586,7 +577,7 @@ struct sk_buff *__udp_gso_segment(struct sk_buff *gso_skb, if (!seg->next) break; - uh->len = msslen; + uh->len = newlen; uh->check = check; if (seg->ip_summed == CHECKSUM_PARTIAL) -- cgit v1.2.3 From 78effd896eee11ac9db9bcbb53e7bbcad96073d7 Mon Sep 17 00:00:00 2001 From: Gal Pressman Date: Mon, 18 May 2026 09:22:50 +0300 Subject: udp: Fix UDP length on last GSO_PARTIAL segment Following the cited commit, __udp_gso_segment() writes single MSS length in the UDP header. The cited patch doesn't account for the fact that the last segment could be a GSO skb by itself. This could happen when the size of the packet is a multiple of MSS, hence the first segment is also the last one (there is no need for a remainder skb). When the post-loop segment is a GSO skb, assign the single MSS length in the UDP header. Fixes: b10b446ce7ad ("udp: gso: Use single MSS length in UDP header for GSO_PARTIAL") Reported-by: Matthew Schwartz Closes: https://lore.kernel.org/all/6c3fb15e-711d-4b8d-b152-e03d9b05293f@linux.dev/ Tested-by: Matthew Schwartz Reviewed-by: Dragos Tatulea Signed-off-by: Gal Pressman Link: https://patch.msgid.link/20260518062250.3019914-3-gal@nvidia.com Signed-off-by: Jakub Kicinski --- net/ipv4/udp_offload.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/ipv4/udp_offload.c b/net/ipv4/udp_offload.c index 2578aa7f9ff9..29651b1a0bc7 100644 --- a/net/ipv4/udp_offload.c +++ b/net/ipv4/udp_offload.c @@ -590,9 +590,12 @@ struct sk_buff *__udp_gso_segment(struct sk_buff *gso_skb, uh = udp_hdr(seg); } - /* last packet can be partial gso_size, account for that in checksum */ - newlen = htons(skb_tail_pointer(seg) - skb_transport_header(seg) + - seg->data_len); + /* Unless skb fits perfectly as GSO_PARTIAL, the trailing + * segment may not be full MSS, account for that in the checksum + */ + if (!skb_is_gso(seg)) + newlen = htons(skb_tail_pointer(seg) - + skb_transport_header(seg) + seg->data_len); check = csum16_add(csum16_sub(uh->check, uh->len), newlen); uh->len = newlen; -- cgit v1.2.3 From a3442936dd0523277e20aaf86207c574e755c634 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 15 May 2026 15:13:24 -0700 Subject: net: shaper: annotate the data races As previously discussed we don't care about making the shaper state fully RCU-compliant because the hierarchy itself can't be dumped in one go over Netlink. Let's annotate the reads and writes to make that clear. The field-by-field assignments will also be useful for the next commit which adds explicit "valid" field (which we don't want to override with the current full struct assignment). Reviewed-by: Simon Horman Link: https://patch.msgid.link/20260515221325.1685455-2-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/shaper/shaper.c | 53 ++++++++++++++++++++++++++++++++++++++--------------- 1 file changed, 38 insertions(+), 15 deletions(-) (limited to 'net') diff --git a/net/shaper/shaper.c b/net/shaper/shaper.c index b1c65110f04d..520cefdc3d90 100644 --- a/net/shaper/shaper.c +++ b/net/shaper/shaper.c @@ -138,35 +138,58 @@ handle_nest_cancel: return -EMSGSIZE; } +static void net_shaper_copy(struct net_shaper *dst, + const struct net_shaper *src) +{ + WRITE_ONCE(dst->parent.scope, READ_ONCE(src->parent.scope)); + WRITE_ONCE(dst->parent.id, READ_ONCE(src->parent.id)); + WRITE_ONCE(dst->handle.scope, READ_ONCE(src->handle.scope)); + WRITE_ONCE(dst->handle.id, READ_ONCE(src->handle.id)); + + WRITE_ONCE(dst->metric, READ_ONCE(src->metric)); + WRITE_ONCE(dst->bw_min, READ_ONCE(src->bw_min)); + WRITE_ONCE(dst->bw_max, READ_ONCE(src->bw_max)); + WRITE_ONCE(dst->burst, READ_ONCE(src->burst)); + WRITE_ONCE(dst->priority, READ_ONCE(src->priority)); + WRITE_ONCE(dst->weight, READ_ONCE(src->weight)); + + /* private fields are only used on the write path under the lock */ + data_race(dst->leaves = src->leaves); +} + static int net_shaper_fill_one(struct sk_buff *msg, const struct net_shaper_binding *binding, const struct net_shaper *shaper, const struct genl_info *info) { + struct net_shaper cur; void *hdr; hdr = genlmsg_iput(msg, info); if (!hdr) return -EMSGSIZE; + /* Make a copy to avoid data races */ + net_shaper_copy(&cur, shaper); + if (net_shaper_fill_binding(msg, binding, NET_SHAPER_A_IFINDEX) || - net_shaper_fill_handle(msg, &shaper->parent, + net_shaper_fill_handle(msg, &cur.parent, NET_SHAPER_A_PARENT) || - net_shaper_fill_handle(msg, &shaper->handle, + net_shaper_fill_handle(msg, &cur.handle, NET_SHAPER_A_HANDLE) || - ((shaper->bw_min || shaper->bw_max || shaper->burst) && - nla_put_u32(msg, NET_SHAPER_A_METRIC, shaper->metric)) || - (shaper->bw_min && - nla_put_uint(msg, NET_SHAPER_A_BW_MIN, shaper->bw_min)) || - (shaper->bw_max && - nla_put_uint(msg, NET_SHAPER_A_BW_MAX, shaper->bw_max)) || - (shaper->burst && - nla_put_uint(msg, NET_SHAPER_A_BURST, shaper->burst)) || - (shaper->priority && - nla_put_u32(msg, NET_SHAPER_A_PRIORITY, shaper->priority)) || - (shaper->weight && - nla_put_u32(msg, NET_SHAPER_A_WEIGHT, shaper->weight))) + ((cur.bw_min || cur.bw_max || cur.burst) && + nla_put_u32(msg, NET_SHAPER_A_METRIC, cur.metric)) || + (cur.bw_min && + nla_put_uint(msg, NET_SHAPER_A_BW_MIN, cur.bw_min)) || + (cur.bw_max && + nla_put_uint(msg, NET_SHAPER_A_BW_MAX, cur.bw_max)) || + (cur.burst && + nla_put_uint(msg, NET_SHAPER_A_BURST, cur.burst)) || + (cur.priority && + nla_put_u32(msg, NET_SHAPER_A_PRIORITY, cur.priority)) || + (cur.weight && + nla_put_u32(msg, NET_SHAPER_A_WEIGHT, cur.weight))) goto nla_put_failure; genlmsg_end(msg, hdr); @@ -424,7 +447,7 @@ static void net_shaper_commit(struct net_shaper_binding *binding, /* Successful update: drop the tentative mark * and update the hierarchy container. */ - *cur = shapers[i]; + net_shaper_copy(cur, &shapers[i]); smp_wmb(); __xa_set_mark(&hierarchy->shapers, index, NET_SHAPER_VALID); } -- cgit v1.2.3 From b8d7519352ba8c6df83259295d4a3bad093cae90 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 15 May 2026 15:13:25 -0700 Subject: net: shaper: rework the VALID marking (again) Recent commit changed the semantics from NOT_VALID to VALID. I didn't realize that the flags are not stored atomically with the entry in XArray. There's still a race of reader observing a VALID mark for a slot, getting interrupted, writer replacing the entry with a different one, reader continuing, fetching the entry which is now a different pointer than the pointer for which VALID was meant. The biggest consequence of this is that we may see a UAF since net_shaper_rollback() assumed that entries without VALID can be freed without observing RCU. Looks like the XArray marks are buying us nothing at this point. Let's convert the code to an explicit valid field. The smp_load_acquire() / smp_store_release() barriers are marginally cleaner. Reported-by: Sashiko Fixes: 93954b40f6a4 ("net-shapers: implement NL set and delete operations") Reviewed-by: Simon Horman Link: https://patch.msgid.link/20260515221325.1685455-3-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/shaper/shaper.c | 45 ++++++++++++++++++--------------------------- 1 file changed, 18 insertions(+), 27 deletions(-) (limited to 'net') diff --git a/net/shaper/shaper.c b/net/shaper/shaper.c index 520cefdc3d90..dea9270f3e57 100644 --- a/net/shaper/shaper.c +++ b/net/shaper/shaper.c @@ -306,31 +306,24 @@ static void net_shaper_default_parent(const struct net_shaper_handle *handle, parent->id = 0; } -/* MARK_0 is already in use due to XA_FLAGS_ALLOC. The VALID mark is set on - * an entry only after the device-side configuration has completed - * successfully (see net_shaper_commit()). Lookups and dumps must filter on - * this mark to avoid exposing tentative entries inserted by - * net_shaper_pre_insert() while the driver call is still in flight. - */ -#define NET_SHAPER_VALID XA_MARK_1 - static struct net_shaper * net_shaper_lookup(struct net_shaper_binding *binding, const struct net_shaper_handle *handle) { u32 index = net_shaper_handle_to_index(handle); struct net_shaper_hierarchy *hierarchy; + struct net_shaper *cur; hierarchy = net_shaper_hierarchy_rcu(binding); - if (!hierarchy || !xa_get_mark(&hierarchy->shapers, index, - NET_SHAPER_VALID)) + if (!hierarchy) return NULL; - /* Pairs with smp_wmb() in net_shaper_commit(): if the entry is - * valid, its contents must be visible too. - */ - smp_rmb(); - return xa_load(&hierarchy->shapers, index); + cur = xa_load(&hierarchy->shapers, index); + /* Check valid before reading fields */ + if (!cur || !smp_load_acquire(&cur->valid)) + return NULL; + + return cur; } /* Allocate on demand the per device shaper's hierarchy container. @@ -444,12 +437,10 @@ static void net_shaper_commit(struct net_shaper_binding *binding, if (WARN_ON_ONCE(!cur)) continue; - /* Successful update: drop the tentative mark - * and update the hierarchy container. - */ + /* Successful update: update the hierarchy container... */ net_shaper_copy(cur, &shapers[i]); - smp_wmb(); - __xa_set_mark(&hierarchy->shapers, index, NET_SHAPER_VALID); + /* ... publish to lockless readers. */ + smp_store_release(&cur->valid, true); } xa_unlock(&hierarchy->shapers); } @@ -466,10 +457,10 @@ static void net_shaper_rollback(struct net_shaper_binding *binding) xa_lock(&hierarchy->shapers); xa_for_each(&hierarchy->shapers, index, cur) { - if (xa_get_mark(&hierarchy->shapers, index, NET_SHAPER_VALID)) + if (cur->valid) continue; __xa_erase(&hierarchy->shapers, index); - kfree(cur); + kfree_rcu(cur, rcu); } xa_unlock(&hierarchy->shapers); } @@ -882,12 +873,12 @@ int net_shaper_nl_get_dumpit(struct sk_buff *skb, goto out_unlock; for (; (shaper = xa_find(&hierarchy->shapers, &ctx->start_index, - U32_MAX, NET_SHAPER_VALID)); + U32_MAX, XA_PRESENT)); ctx->start_index++) { - /* Pairs with smp_wmb() in net_shaper_commit(): the entry - * is marked VALID, so its contents must be visible too. - */ - smp_rmb(); + /* Check valid before reading fields */ + if (!smp_load_acquire(&shaper->valid)) + continue; + ret = net_shaper_fill_one(skb, binding, shaper, info); if (ret) break; -- cgit v1.2.3 From 2b50aceafe6606ea52ed42aadd1b4d44a188aade Mon Sep 17 00:00:00 2001 From: David Howells Date: Sat, 16 May 2026 00:05:13 +0100 Subject: crypto/krb5, rxrpc: Fix lack of pre-decrypt/pre-verify length checks Change the krb5 crypto library to provide facilities to precheck the length of the message about to be decrypted or verified. Fix AF_RXRPC to make use of this to validate DATA packets secured with RxGK. Fixes: 9d1d2b59341f ("rxrpc: rxgk: Implement the yfs-rxgk security class (GSSAPI)") Closes: https://sashiko.dev/#/patchset/20260511160753.607296-1-dhowells%40redhat.com Signed-off-by: David Howells cc: Herbert Xu cc: Simon Horman cc: Chuck Lever cc: linux-afs@lists.infradead.org Reviewed-by: Jeffrey Altman Tested-by: Marc Dionne Link: https://patch.msgid.link/20260515230516.2718212-2-dhowells@redhat.com Signed-off-by: Jakub Kicinski --- net/rxrpc/rxgk.c | 15 +++++++++++++-- 1 file changed, 13 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/rxrpc/rxgk.c b/net/rxrpc/rxgk.c index 0d5e654da918..26e723052a37 100644 --- a/net/rxrpc/rxgk.c +++ b/net/rxrpc/rxgk.c @@ -480,8 +480,12 @@ static int rxgk_verify_packet_integrity(struct rxrpc_call *call, _enter(""); - crypto_krb5_where_is_the_data(gk->krb5, KRB5_CHECKSUM_MODE, - &data_offset, &data_len); + if (crypto_krb5_where_is_the_data(gk->krb5, KRB5_CHECKSUM_MODE, + &data_offset, &data_len) < 0) { + ret = rxrpc_abort_eproto(call, skb, RXGK_PACKETSHORT, + rxgk_abort_1_short_header); + goto put_gk; + } hdr = kzalloc_obj(*hdr, GFP_NOFS); if (!hdr) @@ -529,6 +533,13 @@ static int rxgk_verify_packet_encrypted(struct rxrpc_call *call, _enter(""); + if (crypto_krb5_check_data_len(gk->krb5, KRB5_ENCRYPT_MODE, + len, sizeof(hdr)) < 0) { + ret = rxrpc_abort_eproto(call, skb, RXGK_PACKETSHORT, + rxgk_abort_2_short_header); + goto error; + } + ret = rxgk_decrypt_skb(gk->krb5, gk->rx_enc, skb, &offset, &len, &ac); if (ret < 0) { if (ret != -ENOMEM) -- cgit v1.2.3 From d2bc90cf6c75cb96d2ce549be6c35efa3099d25b Mon Sep 17 00:00:00 2001 From: David Howells Date: Sat, 16 May 2026 00:05:14 +0100 Subject: rxrpc: Fix DATA decrypt vs splice() by copying data to buffer in recvmsg This improves the fix for CVE-2026-43500. Fix the pagecache corruption from in-place decryption of a DATA packet transmitted locally by splice() by getting rid of the packet sharing in the I/O thread and unconditionally extracting the packet content into a bounce buffer in which the buffer is decrypted. recvmsg() (or the kernel equivalent) then copies the data from the bounce buffer to the destination buffer. The sk_buff then remains unmodified. This has an additional advantage in that the packet is then arranged in the buffer with the correct alignment required for the crypto algorithms to process directly. The performance of the crypto does seem to be a little faster and, surprisingly, the unencrypted performance doesn't seem to change much - possibly due to removing complexity from the I/O thread. Yet another advantage is that the I/O thread doesn't have to copy packets which would slow down packet distribution, ACK generation, etc.. The buffer belongs to the call and is allocated initially at 2K, sufficiently large to hold a whole jumbo subpacket, but the buffer will be increased in size if needed. However, to take this work, MSG_PEEK may cause a later packet to be decrypted into the buffer, in which case the earlier one will need re-decrypting for a subsequent recvmsg(). Note that rx_pkt_offset may legitimately see 0 as a valid offset now, so switch to using USHRT_MAX to indicate an invalid offset. Note also that I would generally prefer to replace the buffers of the current sk_buff with a new kmalloc'd buffer of the right size, ditching the old data and frags as this makes the handling of MSG_PEEK easier and removes the re-decryption issue, but this looks like quite a complicated thing to achieve. skb_morph() looks half way to what I want, but I don't want to have to allocate a new sk_buff. Fixes: d0d5c0cd1e71 ("rxrpc: Use skb_unshare() rather than skb_cow_data()") Reported-by: Hyunwoo Kim Closes: https://lore.kernel.org/r/afKV2zGR6rrelPC7@v4bel/ Signed-off-by: David Howells cc: Simon Horman cc: Jiayuan Chen cc: linux-afs@lists.infradead.org Reviewed-by: Jeffrey Altman Tested-by: Marc Dionne Link: https://patch.msgid.link/20260515230516.2718212-3-dhowells@redhat.com Signed-off-by: Jakub Kicinski --- net/rxrpc/ar-internal.h | 7 ++-- net/rxrpc/call_event.c | 22 +------------ net/rxrpc/call_object.c | 2 ++ net/rxrpc/insecure.c | 3 -- net/rxrpc/recvmsg.c | 68 ++++++++++++++++++++++++++++++-------- net/rxrpc/rxgk.c | 51 ++++++++++++++--------------- net/rxrpc/rxgk_common.h | 82 ++++++++++++++++++++++++++++++++++++++++++++++ net/rxrpc/rxkad.c | 86 ++++++++++++++++++------------------------------- 8 files changed, 201 insertions(+), 120 deletions(-) (limited to 'net') diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h index 27c2aa2dd023..783367eea798 100644 --- a/net/rxrpc/ar-internal.h +++ b/net/rxrpc/ar-internal.h @@ -213,8 +213,6 @@ struct rxrpc_skb_priv { struct { u16 offset; /* Offset of data */ u16 len; /* Length of data */ - u8 flags; -#define RXRPC_RX_VERIFIED 0x01 }; struct { rxrpc_seq_t first_ack; /* First packet in acks table */ @@ -774,6 +772,11 @@ struct rxrpc_call { struct sk_buff_head recvmsg_queue; /* Queue of packets ready for recvmsg() */ struct sk_buff_head rx_queue; /* Queue of packets for this call to receive */ struct sk_buff_head rx_oos_queue; /* Queue of out of sequence packets */ + void *rx_dec_buffer; /* Decryption buffer */ + unsigned short rx_dec_bsize; /* rx_dec_buffer size */ + unsigned short rx_dec_offset; /* Decrypted packet data offset */ + unsigned short rx_dec_len; /* Decrypted packet data len */ + rxrpc_seq_t rx_dec_seq; /* Packet in decryption buffer */ rxrpc_seq_t rx_highest_seq; /* Higest sequence number received */ rxrpc_seq_t rx_consumed; /* Highest packet consumed */ diff --git a/net/rxrpc/call_event.c b/net/rxrpc/call_event.c index 2b19b252225e..fec59d9338b9 100644 --- a/net/rxrpc/call_event.c +++ b/net/rxrpc/call_event.c @@ -332,27 +332,7 @@ bool rxrpc_input_call_event(struct rxrpc_call *call) saw_ack |= sp->hdr.type == RXRPC_PACKET_TYPE_ACK; - if (sp->hdr.type == RXRPC_PACKET_TYPE_DATA && - sp->hdr.securityIndex != 0 && - (skb_cloned(skb) || - skb_has_frag_list(skb) || - skb_has_shared_frag(skb))) { - /* Unshare the packet so that it can be - * modified by in-place decryption. - */ - struct sk_buff *nskb = skb_copy(skb, GFP_ATOMIC); - - if (nskb) { - rxrpc_new_skb(nskb, rxrpc_skb_new_unshared); - rxrpc_input_call_packet(call, nskb); - rxrpc_free_skb(nskb, rxrpc_skb_put_call_rx); - } else { - /* OOM - Drop the packet. */ - rxrpc_see_skb(skb, rxrpc_skb_see_unshare_nomem); - } - } else { - rxrpc_input_call_packet(call, skb); - } + rxrpc_input_call_packet(call, skb); rxrpc_free_skb(skb, rxrpc_skb_put_call_rx); did_receive = true; } diff --git a/net/rxrpc/call_object.c b/net/rxrpc/call_object.c index f035f486c139..fcb9d38bb521 100644 --- a/net/rxrpc/call_object.c +++ b/net/rxrpc/call_object.c @@ -152,6 +152,7 @@ struct rxrpc_call *rxrpc_alloc_call(struct rxrpc_sock *rx, gfp_t gfp, spin_lock_init(&call->notify_lock); refcount_set(&call->ref, 1); call->debug_id = debug_id; + call->rx_pkt_offset = USHRT_MAX; call->tx_total_len = -1; call->tx_jumbo_max = 1; call->next_rx_timo = 20 * HZ; @@ -553,6 +554,7 @@ static void rxrpc_cleanup_rx_buffers(struct rxrpc_call *call) rxrpc_purge_queue(&call->recvmsg_queue); rxrpc_purge_queue(&call->rx_queue); rxrpc_purge_queue(&call->rx_oos_queue); + kfree(call->rx_dec_buffer); } /* diff --git a/net/rxrpc/insecure.c b/net/rxrpc/insecure.c index 0a260df45d25..7a26c6097d03 100644 --- a/net/rxrpc/insecure.c +++ b/net/rxrpc/insecure.c @@ -32,9 +32,6 @@ static int none_secure_packet(struct rxrpc_call *call, struct rxrpc_txbuf *txb) static int none_verify_packet(struct rxrpc_call *call, struct sk_buff *skb) { - struct rxrpc_skb_priv *sp = rxrpc_skb(skb); - - sp->flags |= RXRPC_RX_VERIFIED; return 0; } diff --git a/net/rxrpc/recvmsg.c b/net/rxrpc/recvmsg.c index e1f7513a46db..c940600117a4 100644 --- a/net/rxrpc/recvmsg.c +++ b/net/rxrpc/recvmsg.c @@ -147,15 +147,52 @@ static void rxrpc_rotate_rx_window(struct rxrpc_call *call) } /* - * Decrypt and verify a DATA packet. + * Decrypt and verify a DATA packet. The content of the packet is pulled out + * into a flat buffer rather than decrypting in place in the skbuff. This also + * has the advantage of aligning the buffer correctly for the crypto routines. + * + * We keep track of the sequence number of the packet currently decrypted into + * the buffer in ->rx_dec_seq. If MSG_PEEK is used and steps onto a new + * packet, subsequent recvmsg() calls will have to go back and re-decrypt the + * current packet. */ static int rxrpc_verify_data(struct rxrpc_call *call, struct sk_buff *skb) { struct rxrpc_skb_priv *sp = rxrpc_skb(skb); + int ret; - if (sp->flags & RXRPC_RX_VERIFIED) - return 0; - return call->security->verify_packet(call, skb); + if (sp->len > call->rx_dec_bsize) { + /* Make sure we can hold a 1412-byte jumbo subpacket and make + * sure that the buffer size is aligned to a crypto blocksize. + */ + size_t size = clamp(round_up(sp->len, 32), 2048, 65535); + void *buffer = krealloc(call->rx_dec_buffer, size, GFP_NOFS); + + if (!buffer) + return -ENOMEM; + call->rx_dec_buffer = buffer; + call->rx_dec_bsize = size; + } + + ret = -EFAULT; + if (skb_copy_bits(skb, sp->offset, call->rx_dec_buffer, sp->len) < 0) + goto err; + + call->rx_dec_offset = 0; + call->rx_dec_len = sp->len; + call->rx_dec_seq = sp->hdr.seq; + ret = call->security->verify_packet(call, skb); + if (ret < 0) + goto err; + return 0; + +err: + kfree(call->rx_dec_buffer); + call->rx_dec_buffer = NULL; + call->rx_dec_bsize = 0; + call->rx_dec_offset = 0; + call->rx_dec_len = 0; + return ret; } /* @@ -283,16 +320,21 @@ static int rxrpc_recvmsg_data(struct socket *sock, struct rxrpc_call *call, if (msg) sock_recv_timestamp(msg, sock->sk, skb); - if (rx_pkt_offset == 0) { + if (call->rx_dec_seq != sp->hdr.seq || + !call->rx_dec_buffer) { ret2 = rxrpc_verify_data(call, skb); trace_rxrpc_recvdata(call, rxrpc_recvmsg_next, seq, - sp->offset, sp->len, ret2); + call->rx_dec_offset, + call->rx_dec_len, ret2); if (ret2 < 0) { ret = ret2; goto out; } - rx_pkt_offset = sp->offset; - rx_pkt_len = sp->len; + } + + if (rx_pkt_offset == USHRT_MAX) { + rx_pkt_offset = call->rx_dec_offset; + rx_pkt_len = call->rx_dec_len; } else { trace_rxrpc_recvdata(call, rxrpc_recvmsg_cont, seq, rx_pkt_offset, rx_pkt_len, 0); @@ -304,10 +346,10 @@ static int rxrpc_recvmsg_data(struct socket *sock, struct rxrpc_call *call, if (copy > remain) copy = remain; if (copy > 0) { - ret2 = skb_copy_datagram_iter(skb, rx_pkt_offset, iter, - copy); - if (ret2 < 0) { - ret = ret2; + ret2 = copy_to_iter(call->rx_dec_buffer + rx_pkt_offset, + copy, iter); + if (ret2 != copy) { + ret = -EFAULT; goto out; } @@ -328,7 +370,7 @@ static int rxrpc_recvmsg_data(struct socket *sock, struct rxrpc_call *call, /* The whole packet has been transferred. */ if (sp->hdr.flags & RXRPC_LAST_PACKET) ret = 1; - rx_pkt_offset = 0; + rx_pkt_offset = USHRT_MAX; rx_pkt_len = 0; skb = skb_peek_next(skb, &call->recvmsg_queue); diff --git a/net/rxrpc/rxgk.c b/net/rxrpc/rxgk.c index 26e723052a37..f81703ee7ac3 100644 --- a/net/rxrpc/rxgk.c +++ b/net/rxrpc/rxgk.c @@ -473,8 +473,9 @@ static int rxgk_verify_packet_integrity(struct rxrpc_call *call, struct rxrpc_skb_priv *sp = rxrpc_skb(skb); struct rxgk_header *hdr; struct krb5_buffer metadata; - unsigned int offset = sp->offset, len = sp->len; + unsigned int len = call->rx_dec_len; size_t data_offset = 0, data_len = len; + void *data = call->rx_dec_buffer, *p = data; u32 ac = 0; int ret = -ENOMEM; @@ -500,16 +501,15 @@ static int rxgk_verify_packet_integrity(struct rxrpc_call *call, metadata.len = sizeof(*hdr); metadata.data = hdr; - ret = rxgk_verify_mic_skb(gk->krb5, gk->rx_Kc, &metadata, - skb, &offset, &len, &ac); + ret = rxgk_verify_mic(gk->krb5, gk->rx_Kc, &metadata, &p, &len, &ac); kfree(hdr); if (ret < 0) { if (ret != -ENOMEM) rxrpc_abort_eproto(call, skb, ac, rxgk_abort_1_verify_mic_eproto); } else { - sp->offset = offset; - sp->len = len; + call->rx_dec_offset = p - data; + call->rx_dec_len = len; } put_gk: @@ -526,56 +526,53 @@ static int rxgk_verify_packet_encrypted(struct rxrpc_call *call, struct sk_buff *skb) { struct rxrpc_skb_priv *sp = rxrpc_skb(skb); - struct rxgk_header hdr; - unsigned int offset = sp->offset, len = sp->len; + struct rxgk_header *hdr; + unsigned int offset = 0, len = call->rx_dec_len; + void *data = call->rx_dec_buffer, *p = data; int ret; u32 ac = 0; _enter(""); if (crypto_krb5_check_data_len(gk->krb5, KRB5_ENCRYPT_MODE, - len, sizeof(hdr)) < 0) { + len, sizeof(*hdr)) < 0) { ret = rxrpc_abort_eproto(call, skb, RXGK_PACKETSHORT, rxgk_abort_2_short_header); goto error; } - ret = rxgk_decrypt_skb(gk->krb5, gk->rx_enc, skb, &offset, &len, &ac); + ret = rxgk_decrypt(gk->krb5, gk->rx_enc, &p, &len, &ac); if (ret < 0) { if (ret != -ENOMEM) rxrpc_abort_eproto(call, skb, ac, rxgk_abort_2_decrypt_eproto); goto error; } + offset = p - data; - if (len < sizeof(hdr)) { + if (len < sizeof(*hdr)) { ret = rxrpc_abort_eproto(call, skb, RXGK_PACKETSHORT, rxgk_abort_2_short_header); goto error; } /* Extract the header from the skb */ - ret = skb_copy_bits(skb, offset, &hdr, sizeof(hdr)); - if (ret < 0) { - ret = rxrpc_abort_eproto(call, skb, RXGK_PACKETSHORT, - rxgk_abort_2_short_encdata); - goto error; - } - offset += sizeof(hdr); - len -= sizeof(hdr); - - if (ntohl(hdr.epoch) != call->conn->proto.epoch || - ntohl(hdr.cid) != call->cid || - ntohl(hdr.call_number) != call->call_id || - ntohl(hdr.seq) != sp->hdr.seq || - ntohl(hdr.sec_index) != call->security_ix || - ntohl(hdr.data_len) > len) { + hdr = data + offset; + offset += sizeof(*hdr); + len -= sizeof(*hdr); + + if (ntohl(hdr->epoch) != call->conn->proto.epoch || + ntohl(hdr->cid) != call->cid || + ntohl(hdr->call_number) != call->call_id || + ntohl(hdr->seq) != sp->hdr.seq || + ntohl(hdr->sec_index) != call->security_ix || + ntohl(hdr->data_len) > len) { ret = rxrpc_abort_eproto(call, skb, RXGK_SEALEDINCON, rxgk_abort_2_short_data); goto error; } - sp->offset = offset; - sp->len = ntohl(hdr.data_len); + call->rx_dec_offset = offset; + call->rx_dec_len = ntohl(hdr->data_len); ret = 0; error: rxgk_put(gk); diff --git a/net/rxrpc/rxgk_common.h b/net/rxrpc/rxgk_common.h index 1e257d7ab8ec..112b5366ce11 100644 --- a/net/rxrpc/rxgk_common.h +++ b/net/rxrpc/rxgk_common.h @@ -105,6 +105,49 @@ int rxgk_decrypt_skb(const struct krb5_enctype *krb5, return ret; } +/* + * Apply decryption and checksumming functions a flat data buffer. The data + * point and length are updated to reflect the actual content of the encrypted + * region. + */ +static inline int rxgk_decrypt(const struct krb5_enctype *krb5, + struct crypto_aead *aead, + void **_data, unsigned int *_len, + int *_error_code) +{ + struct scatterlist sg[1]; + size_t offset = 0, len = *_len; + int ret; + + sg_init_one(sg, *_data, len); + + ret = crypto_krb5_decrypt(krb5, aead, sg, 1, &offset, &len); + switch (ret) { + case 0: + if (offset & 3) { + *_error_code = RXGK_INCONSISTENCY; + ret = -EPROTO; + break; + } + *_data += offset; + *_len = len; + break; + case -EBADMSG: /* Checksum mismatch. */ + case -EPROTO: + *_error_code = RXGK_SEALEDINCON; + break; + case -EMSGSIZE: + *_error_code = RXGK_PACKETSHORT; + break; + case -ENOPKG: /* Would prefer RXGK_BADETYPE, but not available for YFS. */ + default: + *_error_code = RXGK_INCONSISTENCY; + break; + } + + return ret; +} + /* * Check the MIC on a region of an skbuff. The offset and length are updated * to reflect the actual content of the secure region. @@ -148,3 +191,42 @@ int rxgk_verify_mic_skb(const struct krb5_enctype *krb5, return ret; } + +/* + * Check the MIC on a flat buffer. The data pointer and length are updated to + * reflect the actual content of the secure region. + */ +static inline +int rxgk_verify_mic(const struct krb5_enctype *krb5, + struct crypto_shash *shash, + const struct krb5_buffer *metadata, + void **_data, unsigned int *_len, + u32 *_error_code) +{ + struct scatterlist sg[1]; + size_t offset = 0, len = *_len; + int ret; + + sg_init_one(sg, *_data, len); + + ret = crypto_krb5_verify_mic(krb5, shash, metadata, sg, 1, &offset, &len); + switch (ret) { + case 0: + *_data += offset; + *_len = len; + break; + case -EBADMSG: /* Checksum mismatch */ + case -EPROTO: + *_error_code = RXGK_SEALEDINCON; + break; + case -EMSGSIZE: + *_error_code = RXGK_PACKETSHORT; + break; + case -ENOPKG: /* Would prefer RXGK_BADETYPE, but not available for YFS. */ + default: + *_error_code = RXGK_INCONSISTENCY; + break; + } + + return ret; +} diff --git a/net/rxrpc/rxkad.c b/net/rxrpc/rxkad.c index cba7935977f0..075936337836 100644 --- a/net/rxrpc/rxkad.c +++ b/net/rxrpc/rxkad.c @@ -430,27 +430,25 @@ static int rxkad_verify_packet_1(struct rxrpc_call *call, struct sk_buff *skb, rxrpc_seq_t seq, struct skcipher_request *req) { - struct rxkad_level1_hdr sechdr; + struct rxkad_level1_hdr *sechdr; struct rxrpc_skb_priv *sp = rxrpc_skb(skb); struct rxrpc_crypt iv; - struct scatterlist sg[16]; - u32 data_size, buf; + struct scatterlist sg[1]; + void *data = call->rx_dec_buffer; + u32 len = sp->len, data_size, buf; u16 check; int ret; _enter(""); - if (sp->len < 8) + if (len < 8) return rxrpc_abort_eproto(call, skb, RXKADSEALEDINCON, rxkad_abort_1_short_header); /* Decrypt the skbuff in-place. TODO: We really want to decrypt * directly into the target buffer. */ - sg_init_table(sg, ARRAY_SIZE(sg)); - ret = skb_to_sgvec(skb, sg, sp->offset, 8); - if (unlikely(ret < 0)) - return ret; + sg_init_one(sg, data, len); /* start the decryption afresh */ memset(&iv, 0, sizeof(iv)); @@ -464,13 +462,11 @@ static int rxkad_verify_packet_1(struct rxrpc_call *call, struct sk_buff *skb, return ret; /* Extract the decrypted packet length */ - if (skb_copy_bits(skb, sp->offset, &sechdr, sizeof(sechdr)) < 0) - return rxrpc_abort_eproto(call, skb, RXKADDATALEN, - rxkad_abort_1_short_encdata); - sp->offset += sizeof(sechdr); - sp->len -= sizeof(sechdr); + sechdr = data; + call->rx_dec_offset = sizeof(*sechdr); + len -= sizeof(*sechdr); - buf = ntohl(sechdr.data_size); + buf = ntohl(sechdr->data_size); data_size = buf & 0xffff; check = buf >> 16; @@ -479,10 +475,10 @@ static int rxkad_verify_packet_1(struct rxrpc_call *call, struct sk_buff *skb, if (check != 0) return rxrpc_abort_eproto(call, skb, RXKADSEALEDINCON, rxkad_abort_1_short_check); - if (data_size > sp->len) + if (data_size > len) return rxrpc_abort_eproto(call, skb, RXKADDATALEN, rxkad_abort_1_short_data); - sp->len = data_size; + call->rx_dec_len = data_size; _leave(" = 0 [dlen=%x]", data_size); return 0; @@ -496,43 +492,28 @@ static int rxkad_verify_packet_2(struct rxrpc_call *call, struct sk_buff *skb, struct skcipher_request *req) { const struct rxrpc_key_token *token; - struct rxkad_level2_hdr sechdr; + struct rxkad_level2_hdr *sechdr; struct rxrpc_skb_priv *sp = rxrpc_skb(skb); struct rxrpc_crypt iv; - struct scatterlist _sg[4], *sg; - u32 data_size, buf; + struct scatterlist sg[1]; + void *data = call->rx_dec_buffer; + u32 len = sp->len, data_size, buf; u16 check; - int nsg, ret; + int ret; - _enter(",{%d}", sp->len); + _enter(",{%d}", len); - if (sp->len < 8) + if (len < 8) return rxrpc_abort_eproto(call, skb, RXKADSEALEDINCON, rxkad_abort_2_short_header); /* Don't let the crypto algo see a misaligned length. */ - sp->len = round_down(sp->len, 8); + len = round_down(len, 8); - /* Decrypt the skbuff in-place. TODO: We really want to decrypt - * directly into the target buffer. + /* Decrypt in place in the call's decryption buffer. TODO: We really + * want to decrypt directly into the target buffer. */ - sg = _sg; - nsg = skb_shinfo(skb)->nr_frags + 1; - if (nsg <= 4) { - nsg = 4; - } else { - sg = kmalloc_objs(*sg, nsg, GFP_NOIO); - if (!sg) - return -ENOMEM; - } - - sg_init_table(sg, nsg); - ret = skb_to_sgvec(skb, sg, sp->offset, sp->len); - if (unlikely(ret < 0)) { - if (sg != _sg) - kfree(sg); - return ret; - } + sg_init_one(sg, data, len); /* decrypt from the session key */ token = call->conn->key->payload.data[0]; @@ -540,11 +521,9 @@ static int rxkad_verify_packet_2(struct rxrpc_call *call, struct sk_buff *skb, skcipher_request_set_sync_tfm(req, call->conn->rxkad.cipher); skcipher_request_set_callback(req, 0, NULL, NULL); - skcipher_request_set_crypt(req, sg, sg, sp->len, iv.x); + skcipher_request_set_crypt(req, sg, sg, len, iv.x); ret = crypto_skcipher_decrypt(req); skcipher_request_zero(req); - if (sg != _sg) - kfree(sg); if (ret < 0) { if (ret == -ENOMEM) return ret; @@ -553,13 +532,11 @@ static int rxkad_verify_packet_2(struct rxrpc_call *call, struct sk_buff *skb, } /* Extract the decrypted packet length */ - if (skb_copy_bits(skb, sp->offset, &sechdr, sizeof(sechdr)) < 0) - return rxrpc_abort_eproto(call, skb, RXKADDATALEN, - rxkad_abort_2_short_len); - sp->offset += sizeof(sechdr); - sp->len -= sizeof(sechdr); + sechdr = data; + call->rx_dec_offset = sizeof(*sechdr); + len -= sizeof(*sechdr); - buf = ntohl(sechdr.data_size); + buf = ntohl(sechdr->data_size); data_size = buf & 0xffff; check = buf >> 16; @@ -569,17 +546,18 @@ static int rxkad_verify_packet_2(struct rxrpc_call *call, struct sk_buff *skb, return rxrpc_abort_eproto(call, skb, RXKADSEALEDINCON, rxkad_abort_2_short_check); - if (data_size > sp->len) + if (data_size > len) return rxrpc_abort_eproto(call, skb, RXKADDATALEN, rxkad_abort_2_short_data); - sp->len = data_size; + call->rx_dec_len = data_size; _leave(" = 0 [dlen=%x]", data_size); return 0; } /* - * Verify the security on a received packet and the subpackets therein. + * Verify the security on a received (sub)packet. If the packet needs + * modifying (e.g. decrypting), it must be copied. */ static int rxkad_verify_packet(struct rxrpc_call *call, struct sk_buff *skb) { -- cgit v1.2.3 From 8bfab4b6ffc2fe92da86300728fc8c3c7ebffb56 Mon Sep 17 00:00:00 2001 From: David Howells Date: Sat, 16 May 2026 00:05:15 +0100 Subject: rxrpc: Fix RESPONSE packet verification to extract skb to a linear buffer This improves the fix for CVE-2026-43500. Fix the verification of RESPONSE packets to avoid the problem of overwriting a RESPONSE packet sent via splice to a local address by extracting the contents of the UDP packet into a kmalloc'd linear buffer rather than decrypting the data in place in the sk_buff (which may corrupt the original buffer). Fixes: 24481a7f5733 ("rxrpc: Fix conn-level packet handling to unshare RESPONSE packets") Reported-by: Hyunwoo Kim Closes: https://lore.kernel.org/r/afKV2zGR6rrelPC7@v4bel/ Signed-off-by: David Howells cc: Simon Horman cc: Jiayuan Chen cc: linux-afs@lists.infradead.org cc: stable@kernel.org Reviewed-by: Jeffrey Altman Tested-by: Marc Dionne Link: https://patch.msgid.link/20260515230516.2718212-4-dhowells@redhat.com Signed-off-by: Jakub Kicinski --- net/rxrpc/ar-internal.h | 7 ++-- net/rxrpc/conn_event.c | 32 +++++++---------- net/rxrpc/insecure.c | 5 +-- net/rxrpc/rxgk.c | 96 ++++++++++++++++--------------------------------- net/rxrpc/rxgk_app.c | 46 ++++++++++-------------- net/rxrpc/rxgk_common.h | 92 ++--------------------------------------------- net/rxrpc/rxkad.c | 29 ++++++--------- 7 files changed, 82 insertions(+), 225 deletions(-) (limited to 'net') diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h index 783367eea798..98f2165159d7 100644 --- a/net/rxrpc/ar-internal.h +++ b/net/rxrpc/ar-internal.h @@ -307,15 +307,16 @@ struct rxrpc_security { struct sk_buff *challenge); /* verify a response */ - int (*verify_response)(struct rxrpc_connection *, - struct sk_buff *); + int (*verify_response)(struct rxrpc_connection *conn, + struct sk_buff *response_skb, + void *response, unsigned int len); /* clear connection security */ void (*clear)(struct rxrpc_connection *); /* Default ticket -> key decoder */ int (*default_decode_ticket)(struct rxrpc_connection *conn, struct sk_buff *skb, - unsigned int ticket_offset, unsigned int ticket_len, + void *ticket, unsigned int ticket_len, struct key **_key); }; diff --git a/net/rxrpc/conn_event.c b/net/rxrpc/conn_event.c index 442414d90ba1..c96ca615b787 100644 --- a/net/rxrpc/conn_event.c +++ b/net/rxrpc/conn_event.c @@ -243,28 +243,22 @@ static void rxrpc_call_is_secure(struct rxrpc_call *call) static int rxrpc_verify_response(struct rxrpc_connection *conn, struct sk_buff *skb) { + unsigned int len = skb->len - sizeof(struct rxrpc_wire_header); + void *buffer; int ret; - if (skb_cloned(skb) || skb_has_frag_list(skb) || - skb_has_shared_frag(skb)) { - /* Copy the packet if shared so that we can do in-place - * decryption. - */ - struct sk_buff *nskb = skb_copy(skb, GFP_NOFS); - - if (nskb) { - rxrpc_new_skb(nskb, rxrpc_skb_new_unshared); - ret = conn->security->verify_response(conn, nskb); - rxrpc_free_skb(nskb, rxrpc_skb_put_response_copy); - } else { - /* OOM - Drop the packet. */ - rxrpc_see_skb(skb, rxrpc_skb_see_unshare_nomem); - ret = -ENOMEM; - } - } else { - ret = conn->security->verify_response(conn, skb); - } + buffer = kmalloc(len, GFP_NOFS); + if (!buffer) + return -ENOMEM; + + ret = skb_copy_bits(skb, sizeof(struct rxrpc_wire_header), buffer, len); + if (ret < 0) + goto out; + + ret = conn->security->verify_response(conn, skb, buffer, len); +out: + kfree(buffer); return ret; } diff --git a/net/rxrpc/insecure.c b/net/rxrpc/insecure.c index 7a26c6097d03..0b39046bdc61 100644 --- a/net/rxrpc/insecure.c +++ b/net/rxrpc/insecure.c @@ -54,9 +54,10 @@ static int none_sendmsg_respond_to_challenge(struct sk_buff *challenge, } static int none_verify_response(struct rxrpc_connection *conn, - struct sk_buff *skb) + struct sk_buff *response_skb, + void *response, unsigned int len) { - return rxrpc_abort_conn(conn, skb, RX_PROTOCOL_ERROR, -EPROTO, + return rxrpc_abort_conn(conn, response_skb, RX_PROTOCOL_ERROR, -EPROTO, rxrpc_eproto_rxnull_response); } diff --git a/net/rxrpc/rxgk.c b/net/rxrpc/rxgk.c index f81703ee7ac3..a1ee102abae1 100644 --- a/net/rxrpc/rxgk.c +++ b/net/rxrpc/rxgk.c @@ -1084,11 +1084,12 @@ static int rxgk_sendmsg_respond_to_challenge(struct sk_buff *challenge, * unsigned int call_numbers<>; * }; */ -static int rxgk_do_verify_authenticator(struct rxrpc_connection *conn, - const struct krb5_enctype *krb5, - struct sk_buff *skb, - __be32 *p, __be32 *end) +static int rxgk_verify_authenticator(struct rxrpc_connection *conn, + const struct krb5_enctype *krb5, + struct sk_buff *skb, + void *auth, unsigned int auth_len) { + __be32 *p = auth, *end = auth + auth_len; u32 app_len, call_count, level, epoch, cid, i; _enter(""); @@ -1151,37 +1152,6 @@ static int rxgk_do_verify_authenticator(struct rxrpc_connection *conn, return 0; } -/* - * Extract the authenticator and verify it. - */ -static int rxgk_verify_authenticator(struct rxrpc_connection *conn, - const struct krb5_enctype *krb5, - struct sk_buff *skb, - unsigned int auth_offset, unsigned int auth_len) -{ - void *auth; - __be32 *p; - int ret; - - auth = kmalloc(auth_len, GFP_NOFS); - if (!auth) - return -ENOMEM; - - ret = skb_copy_bits(skb, auth_offset, auth, auth_len); - if (ret < 0) { - ret = rxrpc_abort_conn(conn, skb, RXGK_NOTAUTH, -EPROTO, - rxgk_abort_resp_short_auth); - goto error; - } - - p = auth; - ret = rxgk_do_verify_authenticator(conn, krb5, skb, p, - p + auth_len / sizeof(*p)); -error: - kfree(auth); - return ret; -} - /* * Verify a response. * @@ -1192,49 +1162,45 @@ error: * }; */ static int rxgk_verify_response(struct rxrpc_connection *conn, - struct sk_buff *skb) + struct sk_buff *skb, + void *buffer, unsigned int len) { const struct krb5_enctype *krb5; struct rxrpc_key_token *token; struct rxrpc_skb_priv *sp = rxrpc_skb(skb); - struct rxgk_response rhdr; + struct rxgk_response *rhdr; struct rxgk_context *gk; struct key *key = NULL; - unsigned int offset = sizeof(struct rxrpc_wire_header); - unsigned int len = skb->len - sizeof(struct rxrpc_wire_header); - unsigned int token_offset, token_len; - unsigned int auth_offset, auth_len; + unsigned int resp_token_len, auth_len; + void *resp_token, *auth; __be32 xauth_len; int ret, ec; _enter("{%d}", conn->debug_id); /* Parse the RXGK_Response object */ - if (sizeof(rhdr) + sizeof(__be32) > len) - goto short_packet; - - if (skb_copy_bits(skb, offset, &rhdr, sizeof(rhdr)) < 0) + if (len < sizeof(*rhdr) + sizeof(__be32)) goto short_packet; - offset += sizeof(rhdr); - len -= sizeof(rhdr); - - token_offset = offset; - token_len = ntohl(rhdr.token_len); - if (token_len > len || - xdr_round_up(token_len) + sizeof(__be32) > len) + rhdr = buffer; + buffer += sizeof(*rhdr); + len -= sizeof(*rhdr); + + resp_token = buffer; + resp_token_len = ntohl(rhdr->token_len); + if (resp_token_len > len || + xdr_round_up(resp_token_len) + sizeof(__be32) > len) goto short_packet; - trace_rxrpc_rx_response(conn, sp->hdr.serial, 0, sp->hdr.cksum, token_len); + trace_rxrpc_rx_response(conn, sp->hdr.serial, 0, sp->hdr.cksum, resp_token_len); - offset += xdr_round_up(token_len); - len -= xdr_round_up(token_len); + buffer += xdr_round_up(resp_token_len); + len -= xdr_round_up(resp_token_len); - if (skb_copy_bits(skb, offset, &xauth_len, sizeof(xauth_len)) < 0) - goto short_packet; - offset += sizeof(xauth_len); + xauth_len = *(__be32 *)buffer; + buffer += sizeof(xauth_len); len -= sizeof(xauth_len); - auth_offset = offset; + auth = buffer; auth_len = ntohl(xauth_len); if (auth_len > len) goto short_packet; @@ -1249,7 +1215,7 @@ static int rxgk_verify_response(struct rxrpc_connection *conn, * to the app to deal with - which might mean a round trip to * userspace. */ - ret = rxgk_extract_token(conn, skb, token_offset, token_len, &key); + ret = rxgk_extract_token(conn, skb, resp_token, resp_token_len, &key); if (ret < 0) goto out; @@ -1263,7 +1229,7 @@ static int rxgk_verify_response(struct rxrpc_connection *conn, */ token = key->payload.data[0]; conn->security_level = token->rxgk->level; - conn->rxgk.start_time = __be64_to_cpu(rhdr.start_time); + conn->rxgk.start_time = __be64_to_cpu(rhdr->start_time); gk = rxgk_generate_transport_key(conn, token->rxgk, sp->hdr.cksum, GFP_NOFS); if (IS_ERR(gk)) { @@ -1273,18 +1239,18 @@ static int rxgk_verify_response(struct rxrpc_connection *conn, krb5 = gk->krb5; - trace_rxrpc_rx_response(conn, sp->hdr.serial, krb5->etype, sp->hdr.cksum, token_len); + trace_rxrpc_rx_response(conn, sp->hdr.serial, krb5->etype, sp->hdr.cksum, + resp_token_len); /* Decrypt, parse and verify the authenticator. */ - ret = rxgk_decrypt_skb(krb5, gk->resp_enc, skb, - &auth_offset, &auth_len, &ec); + ret = rxgk_decrypt(krb5, gk->resp_enc, &auth, &auth_len, &ec); if (ret < 0) { rxrpc_abort_conn(conn, skb, RXGK_SEALEDINCON, ret, rxgk_abort_resp_auth_dec); goto out_gk; } - ret = rxgk_verify_authenticator(conn, krb5, skb, auth_offset, auth_len); + ret = rxgk_verify_authenticator(conn, krb5, skb, auth, auth_len); if (ret < 0) goto out_gk; diff --git a/net/rxrpc/rxgk_app.c b/net/rxrpc/rxgk_app.c index 0ef2a29eb695..200a30064fae 100644 --- a/net/rxrpc/rxgk_app.c +++ b/net/rxrpc/rxgk_app.c @@ -40,7 +40,7 @@ * }; */ int rxgk_yfs_decode_ticket(struct rxrpc_connection *conn, struct sk_buff *skb, - unsigned int ticket_offset, unsigned int ticket_len, + void *buffer, unsigned int ticket_len, struct key **_key) { struct rxrpc_key_token *token; @@ -49,7 +49,7 @@ int rxgk_yfs_decode_ticket(struct rxrpc_connection *conn, struct sk_buff *skb, size_t pre_ticket_len, payload_len; unsigned int klen, enctype; void *payload, *ticket; - __be32 *t, *p, *q, tmp[2]; + __be32 *t, *p, *q, *tmp; int ret; _enter(""); @@ -59,10 +59,7 @@ int rxgk_yfs_decode_ticket(struct rxrpc_connection *conn, struct sk_buff *skb, rxgk_abort_resp_short_yfs_tkt); /* Get the session key length */ - ret = skb_copy_bits(skb, ticket_offset, tmp, sizeof(tmp)); - if (ret < 0) - return rxrpc_abort_conn(conn, skb, RXGK_INCONSISTENCY, -EPROTO, - rxgk_abort_resp_short_yfs_klen); + tmp = buffer; enctype = ntohl(tmp[0]); klen = ntohl(tmp[1]); @@ -84,12 +81,7 @@ int rxgk_yfs_decode_ticket(struct rxrpc_connection *conn, struct sk_buff *skb, * it. */ ticket = payload + pre_ticket_len; - ret = skb_copy_bits(skb, ticket_offset, ticket, ticket_len); - if (ret < 0) { - ret = rxrpc_abort_conn(conn, skb, RXGK_INCONSISTENCY, -EPROTO, - rxgk_abort_resp_short_yfs_tkt); - goto error; - } + memcpy(ticket, buffer, ticket_len); /* Fill out the form header. */ p = payload; @@ -131,7 +123,7 @@ int rxgk_yfs_decode_ticket(struct rxrpc_connection *conn, struct sk_buff *skb, goto error; } - /* Ticket read in with skb_copy_bits above */ + /* Ticket appended above. */ q += xdr_round_up(ticket_len) / 4; if (WARN_ON((unsigned long)q - (unsigned long)payload != payload_len)) { ret = -EIO; @@ -182,14 +174,15 @@ error: * [tools.ietf.org/html/draft-wilkinson-afs3-rxgk-afs-08 sec 6.1] */ int rxgk_extract_token(struct rxrpc_connection *conn, struct sk_buff *skb, - unsigned int token_offset, unsigned int token_len, + void *token, unsigned int token_len, struct key **_key) { const struct krb5_enctype *krb5; const struct krb5_buffer *server_secret; struct crypto_aead *token_enc = NULL; struct key *server_key; - unsigned int ticket_offset, ticket_len; + unsigned int ticket_len; + void *ticket; u32 kvno, enctype; int ret, ec = 0; @@ -197,24 +190,23 @@ int rxgk_extract_token(struct rxrpc_connection *conn, struct sk_buff *skb, __be32 kvno; __be32 enctype; __be32 token_len; - } container; + } *container; - if (token_len < sizeof(container)) + if (token_len < sizeof(*container)) goto short_packet; /* Decode the RXGK_TokenContainer object. This tells us which server * key we should be using. We can then fetch the key, get the secret * and set up the crypto to extract the token. */ - if (skb_copy_bits(skb, token_offset, &container, sizeof(container)) < 0) - goto short_packet; + container = token; + token += sizeof(*container); - kvno = ntohl(container.kvno); - enctype = ntohl(container.enctype); - ticket_len = ntohl(container.token_len); - ticket_offset = token_offset + sizeof(container); + kvno = ntohl(container->kvno); + enctype = ntohl(container->enctype); + ticket_len = ntohl(container->token_len); - if (ticket_len > xdr_round_down(token_len - sizeof(container))) + if (ticket_len > xdr_round_down(token_len - sizeof(*container))) goto short_packet; _debug("KVNO %u", kvno); @@ -237,8 +229,8 @@ int rxgk_extract_token(struct rxrpc_connection *conn, struct sk_buff *skb, * gain access to K0, from which we can derive the transport key and * thence decode the authenticator. */ - ret = rxgk_decrypt_skb(krb5, token_enc, skb, - &ticket_offset, &ticket_len, &ec); + ticket = token; + ret = rxgk_decrypt(krb5, token_enc, &ticket, &ticket_len, &ec); crypto_free_aead(token_enc); token_enc = NULL; if (ret < 0) { @@ -248,7 +240,7 @@ int rxgk_extract_token(struct rxrpc_connection *conn, struct sk_buff *skb, return ret; } - ret = conn->security->default_decode_ticket(conn, skb, ticket_offset, + ret = conn->security->default_decode_ticket(conn, skb, ticket, ticket_len, _key); if (ret < 0) goto cant_get_token; diff --git a/net/rxrpc/rxgk_common.h b/net/rxrpc/rxgk_common.h index 112b5366ce11..3deed5863f5a 100644 --- a/net/rxrpc/rxgk_common.h +++ b/net/rxrpc/rxgk_common.h @@ -41,10 +41,10 @@ struct rxgk_context { * rxgk_app.c */ int rxgk_yfs_decode_ticket(struct rxrpc_connection *conn, struct sk_buff *skb, - unsigned int ticket_offset, unsigned int ticket_len, + void *ticket, unsigned int ticket_len, struct key **_key); int rxgk_extract_token(struct rxrpc_connection *conn, struct sk_buff *skb, - unsigned int token_offset, unsigned int token_len, + void *token, unsigned int token_len, struct key **_key); /* @@ -61,50 +61,6 @@ int rxgk_set_up_token_cipher(const struct krb5_buffer *server_key, const struct krb5_enctype **_krb5, gfp_t gfp); -/* - * Apply decryption and checksumming functions to part of an skbuff. The - * offset and length are updated to reflect the actual content of the encrypted - * region. - */ -static inline -int rxgk_decrypt_skb(const struct krb5_enctype *krb5, - struct crypto_aead *aead, - struct sk_buff *skb, - unsigned int *_offset, unsigned int *_len, - int *_error_code) -{ - struct scatterlist sg[16]; - size_t offset = 0, len = *_len; - int nr_sg, ret; - - sg_init_table(sg, ARRAY_SIZE(sg)); - nr_sg = skb_to_sgvec(skb, sg, *_offset, len); - if (unlikely(nr_sg < 0)) - return nr_sg; - - ret = crypto_krb5_decrypt(krb5, aead, sg, nr_sg, - &offset, &len); - switch (ret) { - case 0: - *_offset += offset; - *_len = len; - break; - case -EBADMSG: /* Checksum mismatch. */ - case -EPROTO: - *_error_code = RXGK_SEALEDINCON; - break; - case -EMSGSIZE: - *_error_code = RXGK_PACKETSHORT; - break; - case -ENOPKG: /* Would prefer RXGK_BADETYPE, but not available for YFS. */ - default: - *_error_code = RXGK_INCONSISTENCY; - break; - } - - return ret; -} - /* * Apply decryption and checksumming functions a flat data buffer. The data * point and length are updated to reflect the actual content of the encrypted @@ -148,50 +104,6 @@ static inline int rxgk_decrypt(const struct krb5_enctype *krb5, return ret; } -/* - * Check the MIC on a region of an skbuff. The offset and length are updated - * to reflect the actual content of the secure region. - */ -static inline -int rxgk_verify_mic_skb(const struct krb5_enctype *krb5, - struct crypto_shash *shash, - const struct krb5_buffer *metadata, - struct sk_buff *skb, - unsigned int *_offset, unsigned int *_len, - u32 *_error_code) -{ - struct scatterlist sg[16]; - size_t offset = 0, len = *_len; - int nr_sg, ret; - - sg_init_table(sg, ARRAY_SIZE(sg)); - nr_sg = skb_to_sgvec(skb, sg, *_offset, len); - if (unlikely(nr_sg < 0)) - return nr_sg; - - ret = crypto_krb5_verify_mic(krb5, shash, metadata, sg, nr_sg, - &offset, &len); - switch (ret) { - case 0: - *_offset += offset; - *_len = len; - break; - case -EBADMSG: /* Checksum mismatch */ - case -EPROTO: - *_error_code = RXGK_SEALEDINCON; - break; - case -EMSGSIZE: - *_error_code = RXGK_PACKETSHORT; - break; - case -ENOPKG: /* Would prefer RXGK_BADETYPE, but not available for YFS. */ - default: - *_error_code = RXGK_INCONSISTENCY; - break; - } - - return ret; -} - /* * Check the MIC on a flat buffer. The data pointer and length are updated to * reflect the actual content of the secure region. diff --git a/net/rxrpc/rxkad.c b/net/rxrpc/rxkad.c index 075936337836..6fbd883401ac 100644 --- a/net/rxrpc/rxkad.c +++ b/net/rxrpc/rxkad.c @@ -963,7 +963,6 @@ static int rxkad_decrypt_ticket(struct rxrpc_connection *conn, *_expiry = 0; ASSERT(server_key->payload.data[0] != NULL); - ASSERTCMP((unsigned long) ticket & 7UL, ==, 0); memcpy(&iv, &server_key->payload.data[2], sizeof(iv)); @@ -1112,14 +1111,15 @@ unlock: * verify a response */ static int rxkad_verify_response(struct rxrpc_connection *conn, - struct sk_buff *skb) + struct sk_buff *skb, + void *buffer, unsigned int len) { struct rxkad_response *response; struct rxrpc_skb_priv *sp = rxrpc_skb(skb); struct rxrpc_crypt session_key; struct key *server_key; time64_t expiry; - void *ticket = NULL; + void *ticket; u32 version, kvno, ticket_len, level; __be32 csum; int ret, i; @@ -1142,13 +1142,8 @@ static int rxkad_verify_response(struct rxrpc_connection *conn, } } - ret = -ENOMEM; - response = kzalloc_obj(struct rxkad_response, GFP_NOFS); - if (!response) - goto error; - - if (skb_copy_bits(skb, sizeof(struct rxrpc_wire_header), - response, sizeof(*response)) < 0) { + response = buffer; + if (len < sizeof(*response)) { ret = rxrpc_abort_conn(conn, skb, RXKADPACKETSHORT, -EPROTO, rxkad_abort_resp_short); goto error; @@ -1160,6 +1155,9 @@ static int rxkad_verify_response(struct rxrpc_connection *conn, trace_rxrpc_rx_response(conn, sp->hdr.serial, version, kvno, ticket_len); + buffer += sizeof(*response); + len -= sizeof(*response); + if (version != RXKAD_VERSION) { ret = rxrpc_abort_conn(conn, skb, RXKADINCONSISTENCY, -EPROTO, rxkad_abort_resp_version); @@ -1179,13 +1177,8 @@ static int rxkad_verify_response(struct rxrpc_connection *conn, } /* extract the kerberos ticket and decrypt and decode it */ - ret = -ENOMEM; - ticket = kmalloc(ticket_len, GFP_NOFS); - if (!ticket) - goto error; - - if (skb_copy_bits(skb, sizeof(struct rxrpc_wire_header) + sizeof(*response), - ticket, ticket_len) < 0) { + ticket = buffer; + if (ticket_len > len) { ret = rxrpc_abort_conn(conn, skb, RXKADPACKETSHORT, -EPROTO, rxkad_abort_resp_short_tkt); goto error; @@ -1265,8 +1258,6 @@ static int rxkad_verify_response(struct rxrpc_connection *conn, ret = rxrpc_get_server_data_key(conn, &session_key, expiry, kvno); error: - kfree(ticket); - kfree(response); key_put(server_key); _leave(" = %d", ret); return ret; -- cgit v1.2.3 From ddf8029623a1af20e984c040e89ff918158397ab Mon Sep 17 00:00:00 2001 From: Xingwang Xiang Date: Sun, 17 May 2026 23:56:26 +0900 Subject: bpf, skmsg: fix verdict sk_data_ready racing with ktls rx sk_psock_strp_data_ready() already checks tls_sw_has_ctx_rx() and defers to psock->saved_data_ready when a TLS RX context is present, avoiding a conflict with the TLS strparser's ownership of the receive queue (commit e91de6afa81c, "bpf: Fix running sk_skb program types with ktls"). sk_psock_verdict_data_ready() has no equivalent guard. When a socket is inserted into a sockmap (BPF_SK_SKB_VERDICT) before TLS RX is configured, tls_sw_strparser_arm() saves sk_psock_verdict_data_ready as rx_ctx->saved_data_ready. On data arrival: tls_data_ready -> tls_strp_data_ready -> tls_rx_msg_ready -> saved_data_ready() = sk_psock_verdict_data_ready() -> tcp_read_skb() drains sk_receive_queue via __skb_unlink() without calling tcp_eat_skb(), so copied_seq is not advanced. tls_strp_msg_load() then finds tcp_inq() >= full_len (stale), calls tcp_recv_skb() on the now-empty queue, hits WARN_ON_ONCE(!first), and returns with rx_ctx->strp.anchor.frag_list pointing at a psock-owned (potentially freed) skb. tls_decrypt_sg() subsequently walks that frag_list: use-after-free. Apply the same fix as sk_psock_strp_data_ready(): if a TLS RX context is present, call psock->saved_data_ready (sock_def_readable) to wake recv() waiters and return immediately, leaving the receive queue untouched. TLS retains sole ownership of the queue and decrypts the record normally through tls_sw_recvmsg(). Fixes: ef5659280eb1 ("bpf, sockmap: Allow skipping sk_skb parser program") Signed-off-by: Xingwang Xiang Link: https://patch.msgid.link/20260517145630.20521-2-v3rdant.xiang@gmail.com Signed-off-by: Jakub Kicinski --- net/core/skmsg.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/core/skmsg.c b/net/core/skmsg.c index 6187a83bd741..e1850caf1a71 100644 --- a/net/core/skmsg.c +++ b/net/core/skmsg.c @@ -1268,12 +1268,19 @@ out: static void sk_psock_verdict_data_ready(struct sock *sk) { const struct proto_ops *ops = NULL; + struct sk_psock *psock; struct socket *sock; int copied; trace_sk_data_ready(sk); rcu_read_lock(); + psock = sk_psock(sk); + if (psock && tls_sw_has_ctx_rx(sk)) { + psock->saved_data_ready(sk); + rcu_read_unlock(); + return; + } sock = READ_ONCE(sk->sk_socket); if (likely(sock)) ops = READ_ONCE(sock->ops); @@ -1283,8 +1290,6 @@ static void sk_psock_verdict_data_ready(struct sock *sk) copied = ops->read_skb(sk, sk_psock_verdict_recv); if (copied >= 0) { - struct sk_psock *psock; - rcu_read_lock(); psock = sk_psock(sk); if (psock) -- cgit v1.2.3 From 4eb82ba543421e9e38cc14e4e82058b78850df50 Mon Sep 17 00:00:00 2001 From: David Carlier Date: Tue, 19 May 2026 21:35:30 +0100 Subject: net: devmem: reject dma-buf bind with non-page-aligned size or SG length net_devmem_bind_dmabuf() trusts dmabuf->size and sg_dma_len() to be PAGE_SIZE multiples without checking: - tx_vec is sized dmabuf->size / PAGE_SIZE, and net_devmem_get_niov_at() only bounds-checks virt_addr < dmabuf->size before indexing tx_vec[virt_addr / PAGE_SIZE]. With size = N*PAGE_SIZE + r (1 <= r < PAGE_SIZE), sendmsg() at iov_base = N*PAGE_SIZE passes the bound check and reads tx_vec[N] -- one past. - owner->area.num_niovs = len / PAGE_SIZE while gen_pool_add_owner() covers the full byte len, so a non-page-multiple non-final sg desyncs num_niovs from the gen_pool region for every later sg, on both RX and TX. dma-buf does not require page-aligned sizes, so the bind path has to enforce what its own indexing assumes. Reject both with -EINVAL. The size check is TX-only (only tx_vec is sized off dmabuf->size); the SG-length check covers both directions. Fixes: bd61848900bf ("net: devmem: Implement TX path") Cc: stable@vger.kernel.org Signed-off-by: David Carlier Reviewed-by: Bobby Eshleman Acked-by: Stanislav Fomichev Reviewed-by: Mina Almasry Link: https://patch.msgid.link/20260519203530.66310-1-devnexen@gmail.com Signed-off-by: Jakub Kicinski --- net/core/devmem.c | 11 +++++++++++ 1 file changed, 11 insertions(+) (limited to 'net') diff --git a/net/core/devmem.c b/net/core/devmem.c index 468344739db2..4f71de44c0fb 100644 --- a/net/core/devmem.c +++ b/net/core/devmem.c @@ -241,6 +241,11 @@ net_devmem_bind_dmabuf(struct net_device *dev, } if (direction == DMA_TO_DEVICE) { + if (!IS_ALIGNED(dmabuf->size, PAGE_SIZE)) { + err = -EINVAL; + NL_SET_ERR_MSG(extack, "TX dma-buf size must be a multiple of PAGE_SIZE"); + goto err_unmap; + } binding->tx_vec = kvmalloc_objs(struct net_iov *, dmabuf->size / PAGE_SIZE); if (!binding->tx_vec) { @@ -267,6 +272,12 @@ net_devmem_bind_dmabuf(struct net_device *dev, size_t len = sg_dma_len(sg); struct net_iov *niov; + if (!IS_ALIGNED(len, PAGE_SIZE)) { + err = -EINVAL; + NL_SET_ERR_MSG(extack, "dma-buf SG length must be PAGE_SIZE aligned"); + goto err_free_chunks; + } + owner = kzalloc_node(sizeof(*owner), GFP_KERNEL, dev_to_node(&dev->dev)); if (!owner) { -- cgit v1.2.3 From 7eb72c1e3984150c45f77aa4299f7c2598a68e9b Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 19 May 2026 20:08:36 +0000 Subject: ipv4: icmp: reject broadcast/multicast routes syzbot was able to trigger ip_rt_bug() in a loop, using an IPv4 packet with a crafted IPOPT_SSRR option: options: ipv4_options { options: array[ipv4_option] { union ipv4_option { ssrr: ipv4_option_route[IPOPT_SSRR] { type: const = 0x89 (1 bytes) length: len = 0x7 (1 bytes) pointer: int8 = 0xa2 (1 bytes) data: array[ipv4_addr] { union ipv4_addr { broadcast: const = 0xffffffff (4 bytes) } } } } Change __icmp_send() to not send ICMP to broadcast/multicast destinations. Fixes: c378a9c019cf ("ipv4: Give backtrace in ip_rt_bug().") Reported-by: syzbot+c13a57c2639c2c0d03a6@syzkaller.appspotmail.com Closes: https://lore.kernel.org/netdev/6a0cc169.170a0220.1f6c2d.0004.GAE@google.com/T/#u Signed-off-by: Eric Dumazet Reviewed-by: David Ahern Link: https://patch.msgid.link/20260519200836.4141061-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/ipv4/icmp.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'net') diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c index 7eeff658b467..23e921d313b3 100644 --- a/net/ipv4/icmp.c +++ b/net/ipv4/icmp.c @@ -961,6 +961,9 @@ void __icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info, if (IS_ERR(rt)) goto out_unlock; + if (rt->rt_flags & (RTCF_BROADCAST | RTCF_MULTICAST)) + goto ende; + /* peer icmp_ratelimit */ if (!icmpv4_xrlim_allow(net, rt, &fl4, type, code, apply_ratelimit)) goto ende; -- cgit v1.2.3 From e4bdef4d320b2fe73b8ebfc0cc0507fa9dc4a3b7 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 19 May 2026 19:32:48 +0000 Subject: ipv4: use WARN_ON_ONCE() in ip_rt_bug() It turns out ip_rt_bug() can be called more than expected. syzbot will still panic (because of panic_on_warn=1), but non debug kernels will no longer die while repeating stack traces on the console. Fixes: c378a9c019cf ("ipv4: Give backtrace in ip_rt_bug().") Signed-off-by: Eric Dumazet Reviewed-by: David Ahern Reviewed-by: Jiayuan Chen Link: https://patch.msgid.link/20260519193248.4018872-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/ipv4/route.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv4/route.c b/net/ipv4/route.c index bc1296f0ea69..3d62d45d84bd 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -1272,7 +1272,7 @@ static int ip_rt_bug(struct net *net, struct sock *sk, struct sk_buff *skb) __func__, &ip_hdr(skb)->saddr, &ip_hdr(skb)->daddr, skb->dev ? skb->dev->name : "?"); kfree_skb(skb); - WARN_ON(1); + WARN_ON_ONCE(1); return 0; } -- cgit v1.2.3 From 99e22ddf4edb63dc8382bc028af928056d3450cf Mon Sep 17 00:00:00 2001 From: Minh Nguyen Date: Tue, 19 May 2026 17:23:10 +0700 Subject: vsock/vmci: fix UAF when peer resets connection during handshake vmci_transport_recv_connecting_server() returned err = 0 for a peer RST in its default switch arm: err = pkt->type == VMCI_TRANSPORT_PACKET_TYPE_RST ? 0 : -EINVAL; That made vmci_transport_recv_listen() skip vsock_remove_pending(), leaving the pending socket on the listener's pending_links with sk_state = TCP_CLOSE while destroy: still dropped the explicit reference taken before schedule_delayed_work(). One second later vsock_pending_work() observed is_pending=true and performed full cleanup: vsock_remove_pending() then the two trailing sock_put(sk) calls -- the first reached refcount 0 and __sk_freed the socket, and the second wrote into the freed object: BUG: KASAN: slab-use-after-free in refcount_warn_saturate Write of size 4 at addr ffff88800b1cac80 by task kworker Workqueue: events vsock_pending_work Treat peer RST like any other unexpected packet type (err = -EINVAL). All destroy: arms now return err < 0, so vmci_transport_recv_listen() removes pending from pending_links synchronously and vsock_pending_work() takes the is_pending=false / !rejected branch, dropping only its own work reference. This also closes the multi-packet race Sashiko reported on v2: pending is removed from the list before any subsequent packet can find it. The pre-existing sk_acceptq_removed() gap on the err < 0 path of vmci_transport_recv_listen() that Sashiko also noted is not introduced or changed by this patch. Tested on lts-6.12.79 with KASAN: 52/100 unpatched -> 0/100 patched. Fixes: d021c344051a ("VSOCK: Introduce VM Sockets") Cc: stable@vger.kernel.org Signed-off-by: Minh Nguyen Acked-by: Bryan Tan Link: https://patch.msgid.link/20260519102310.237181-1-minhnguyen.080505@gmail.com Signed-off-by: Jakub Kicinski --- net/vmw_vsock/vmci_transport.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/vmw_vsock/vmci_transport.c b/net/vmw_vsock/vmci_transport.c index 4296ca1183f1..d2579380f51e 100644 --- a/net/vmw_vsock/vmci_transport.c +++ b/net/vmw_vsock/vmci_transport.c @@ -1164,7 +1164,7 @@ vmci_transport_recv_connecting_server(struct sock *listener, /* Close and cleanup the connection. */ vmci_transport_send_reset(pending, pkt); skerr = EPROTO; - err = pkt->type == VMCI_TRANSPORT_PACKET_TYPE_RST ? 0 : -EINVAL; + err = -EINVAL; goto destroy; } -- cgit v1.2.3 From 1bbf0ced1d9db73ac7893c2187f3459288603e0d Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 19 May 2026 08:46:11 +0000 Subject: tcp: fix stale per-CPU tcp_tw_isn leak enabling ISN prediction Blamed commit moved the TIME_WAIT-derived ISN from the skb control block to a per-CPU variable, assuming the value would always be consumed by tcp_conn_request() for the same packet that wrote it. That assumption is violated by multiple drop paths between the producer (__this_cpu_write(tcp_tw_isn, isn) in tcp_v{4,6}_rcv()) and the consumer (tcp_conn_request()): - min_ttl / min_hopcount check - xfrm policy check - tcp_inbound_hash() MD5/AO mismatch - tcp_filter() eBPF/SO_ATTACH_FILTER drop - th->syn && th->fin discard in tcp_rcv_state_process() TCP_LISTEN - psp_sk_rx_policy_check() in tcp_v{4,6}_do_rcv() - tcp_checksum_complete() in tcp_v{4,6}_do_rcv() - tcp_v{4,6}_cookie_check() returning NULL When a packet is dropped on any of these paths, tcp_tw_isn is left set. The next SYN processed on the same CPU then consumes the non zero value in tcp_conn_request(), receiving a potentially predictable ISN. This patch moves back tcp_tw_isn to skb->cb[], getting rid of the per-cpu variable. Note that tcp_v{4,6}_fill_cb() do not set it. Very litle impact on overall code size/complexity: $ scripts/bloat-o-meter -t vmlinux.old vmlinux.new add/remove: 0/0 grow/shrink: 2/1 up/down: 8/-15 (-7) Function old new delta tcp_v6_rcv 3038 3042 +4 tcp_v4_rcv 3035 3039 +4 tcp_conn_request 2938 2923 -15 Total: Before=24436060, After=24436053, chg -0.00% Fixes: 41eecbd712b7 ("tcp: replace TCP_SKB_CB(skb)->tcp_tw_isn with a per-cpu field") Reported-by: Chris Mason Signed-off-by: Eric Dumazet Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20260519084611.2485277-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/ipv4/tcp.c | 3 --- net/ipv4/tcp_input.c | 15 ++++++--------- net/ipv4/tcp_ipv4.c | 3 ++- net/ipv6/tcp_ipv6.c | 3 ++- 4 files changed, 10 insertions(+), 14 deletions(-) (limited to 'net') diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 432fa28e47d4..389a7cc17110 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -299,9 +299,6 @@ enum { DEFINE_PER_CPU(unsigned int, tcp_orphan_count); EXPORT_PER_CPU_SYMBOL_GPL(tcp_orphan_count); -DEFINE_PER_CPU(u32, tcp_tw_isn); -EXPORT_PER_CPU_SYMBOL_GPL(tcp_tw_isn); - long sysctl_tcp_mem[3] __read_mostly; DEFINE_PER_CPU(int, tcp_memory_per_cpu_fw_alloc); diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index d5c9e65d9760..de9f68a9c0cf 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -7589,6 +7589,7 @@ int tcp_conn_request(struct request_sock_ops *rsk_ops, struct sock *sk, struct sk_buff *skb) { struct tcp_fastopen_cookie foc = { .len = -1 }; + u32 isn = TCP_SKB_CB(skb)->tcp_tw_isn; struct tcp_options_received tmp_opt; const struct tcp_sock *tp = tcp_sk(sk); struct net *net = sock_net(sk); @@ -7599,20 +7600,16 @@ int tcp_conn_request(struct request_sock_ops *rsk_ops, struct dst_entry *dst; struct flowi fl; u8 syncookies; - u32 isn; #ifdef CONFIG_TCP_AO const struct tcp_ao_hdr *aoh; #endif - isn = __this_cpu_read(tcp_tw_isn); - if (isn) { - /* TW buckets are converted to open requests without - * limitations, they conserve resources and peer is - * evidently real one. - */ - __this_cpu_write(tcp_tw_isn, 0); - } else { + /* If isn is non-zero, this SYN originally matched a TIME_WAIT socket. + * TW sockets are converted to open requests without limitations, + * we skip the queue limits and syncookie checks in the block below. + */ + if (!isn) { syncookies = READ_ONCE(net->ipv4.sysctl_tcp_syncookies); if (syncookies == 2 || inet_csk_reqsk_queue_is_full(sk)) { diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index c0526cc03980..fdc81150ff6c 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2198,6 +2198,7 @@ lookup: } } + isn = 0; process: if (static_branch_unlikely(&ip4_min_ttl)) { /* min_ttl can be changed concurrently from do_ip_setsockopt() */ @@ -2227,6 +2228,7 @@ process: th = (const struct tcphdr *)skb->data; iph = ip_hdr(skb); tcp_v4_fill_cb(skb, iph, th); + TCP_SKB_CB(skb)->tcp_tw_isn = isn; skb->dev = NULL; @@ -2313,7 +2315,6 @@ do_time_wait: sk = sk2; tcp_v4_restore_cb(skb); refcounted = false; - __this_cpu_write(tcp_tw_isn, isn); goto process; } diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index d13d49bfef19..36d75fb50a70 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -1839,6 +1839,7 @@ lookup: } } + isn = 0; process: if (static_branch_unlikely(&ip6_min_hopcount)) { /* min_hopcount can be changed concurrently from do_ipv6_setsockopt() */ @@ -1868,6 +1869,7 @@ process: th = (const struct tcphdr *)skb->data; hdr = ipv6_hdr(skb); tcp_v6_fill_cb(skb, hdr, th); + TCP_SKB_CB(skb)->tcp_tw_isn = isn; skb->dev = NULL; @@ -1956,7 +1958,6 @@ do_time_wait: sk = sk2; tcp_v6_restore_cb(skb); refcounted = false; - __this_cpu_write(tcp_tw_isn, isn); goto process; } -- cgit v1.2.3 From 48f6a5356a33dd78e7144ae1faef95ffc990aae0 Mon Sep 17 00:00:00 2001 From: Hyunwoo Kim Date: Sat, 16 May 2026 07:28:53 +0900 Subject: net: skbuff: propagate shared-frag marker through frag-transfer helpers Two frag-transfer helpers (__pskb_copy_fclone() and skb_shift()) fail to propagate the SKBFL_SHARED_FRAG bit in skb_shinfo()->flags when moving frags from source to destination. __pskb_copy_fclone() defers the rest of the shinfo metadata to skb_copy_header() after copying frag descriptors, but that helper only carries over gso_{size,segs, type} and never touches skb_shinfo()->flags; skb_shift() moves frag descriptors directly and leaves flags untouched. As a result, the destination skb keeps a reference to the same externally-owned or page-cache-backed pages while reporting skb_has_shared_frag() as false. The mismatch is harmful in any in-place writer that uses skb_has_shared_frag() to decide whether shared pages must be detoured through skb_cow_data(). ESP input is one such writer (esp4.c, esp6.c), and a single nft 'dup to ' rule -- or any other nf_dup_ipv4() / xt_TEE caller -- is enough to land a pskb_copy()'d skb in esp_input() with the marker stripped, letting an unprivileged user write into the page cache of a root-owned read-only file via authencesn-ESN stray writes. Set SKBFL_SHARED_FRAG on the destination whenever frag descriptors were actually moved from the source. skb_copy() and skb_copy_expand() share skb_copy_header() too but linearize all paged data into freshly allocated head storage and emerge with nr_frags == 0, so skb_has_shared_frag() returns false on its own; they need no change. The same omission exists in skb_gro_receive() and skb_gro_receive_list(). The former moves the incoming skb's frag descriptors into the accumulator's last sub-skb via two paths (a direct frag-move loop and the head_frag + memcpy path); the latter chains the incoming skb whole onto p's frag_list. Downstream skb_segment() reads only skb_shinfo(p)->flags, and skb_segment_list() reuses each sub-skb's shinfo as the nskb -- both p and lp must carry the marker. The same omission also exists in tcp_clone_payload(), which builds an MTU probe skb by moving frag descriptors from skbs on sk_write_queue into a freshly allocated nskb. The helper falls into the same family and warrants the same fix for consistency; no TCP TX-side in-place writer is currently known to reach a user page through this gap, but a future consumer depending on the marker would regress silently. The same omission exists in skb_segment(): the per-iteration flag merge takes only head_skb's flag, and the inner switch that rebinds frag_skb to list_skb on head_skb-frags exhaustion does not fold the new frag_skb's flag into nskb. Fold frag_skb's flag at both sites so segments drawing frags from frag_list members carry the marker. Fixes: cef401de7be8 ("net: fix possible wrong checksum generation") Fixes: f4c50a4034e6 ("xfrm: esp: avoid in-place decrypt on shared skb frags") Suggested-by: Sabrina Dubroca Suggested-by: Sultan Alsawaf Suggested-by: Ben Hutchings Suggested-by: Lin Ma Suggested-by: Jingguo Tan Suggested-by: Aaron Esau Cc: stable@vger.kernel.org Signed-off-by: Hyunwoo Kim Tested-by: Rajat Gupta Link: https://patch.msgid.link/ageeJfJHwgzmKXbh@v4bel Signed-off-by: Paolo Abeni --- net/core/gro.c | 4 ++++ net/core/skbuff.c | 9 ++++++++- net/ipv4/tcp_output.c | 1 + 3 files changed, 13 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/core/gro.c b/net/core/gro.c index 31d21de5b15a..9f8960789b2c 100644 --- a/net/core/gro.c +++ b/net/core/gro.c @@ -213,10 +213,12 @@ done: p->data_len += len; p->truesize += delta_truesize; p->len += len; + skb_shinfo(p)->flags |= skbinfo->flags & SKBFL_SHARED_FRAG; if (lp != p) { lp->data_len += len; lp->truesize += delta_truesize; lp->len += len; + skb_shinfo(lp)->flags |= skbinfo->flags & SKBFL_SHARED_FRAG; } NAPI_GRO_CB(skb)->same_flow = 1; return 0; @@ -244,6 +246,8 @@ int skb_gro_receive_list(struct sk_buff *p, struct sk_buff *skb) p->truesize += skb->truesize; p->len += skb->len; + skb_shinfo(p)->flags |= skb_shinfo(skb)->flags & SKBFL_SHARED_FRAG; + NAPI_GRO_CB(skb)->same_flow = 1; return 0; diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 9c4e8d331d6d..44ac121cfccb 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -2248,6 +2248,7 @@ struct sk_buff *__pskb_copy_fclone(struct sk_buff *skb, int headroom, skb_frag_ref(skb, i); } skb_shinfo(n)->nr_frags = i; + skb_shinfo(n)->flags |= skb_shinfo(skb)->flags & SKBFL_SHARED_FRAG; } if (skb_has_frag_list(skb)) { @@ -4349,6 +4350,8 @@ onlymerged: tgt->ip_summed = CHECKSUM_PARTIAL; skb->ip_summed = CHECKSUM_PARTIAL; + skb_shinfo(tgt)->flags |= skb_shinfo(skb)->flags & SKBFL_SHARED_FRAG; + skb_len_add(skb, -shiftlen); skb_len_add(tgt, shiftlen); @@ -4959,7 +4962,8 @@ normal: skb_copy_from_linear_data_offset(head_skb, offset, skb_put(nskb, hsize), hsize); - skb_shinfo(nskb)->flags |= skb_shinfo(head_skb)->flags & + skb_shinfo(nskb)->flags |= (skb_shinfo(head_skb)->flags | + skb_shinfo(frag_skb)->flags) & SKBFL_SHARED_FRAG; if (skb_zerocopy_clone(nskb, frag_skb, GFP_ATOMIC)) @@ -4976,6 +4980,9 @@ normal: nfrags = skb_shinfo(list_skb)->nr_frags; frag = skb_shinfo(list_skb)->frags; frag_skb = list_skb; + + skb_shinfo(nskb)->flags |= skb_shinfo(frag_skb)->flags & SKBFL_SHARED_FRAG; + if (!skb_headlen(list_skb)) { BUG_ON(!nfrags); } else { diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index f9d8755705f7..6e4bb411dc04 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -2626,6 +2626,7 @@ static int tcp_clone_payload(struct sock *sk, struct sk_buff *to, todo = min_t(int, skb_frag_size(fragfrom), probe_size - len); len += todo; + skb_shinfo(to)->flags |= skb_shinfo(skb)->flags & SKBFL_SHARED_FRAG; if (lastfrag && skb_frag_page(fragfrom) == skb_frag_page(lastfrag) && skb_frag_off(fragfrom) == skb_frag_off(lastfrag) + -- cgit v1.2.3 From a4f0b001782b21663d10df983b4b208195bec66c Mon Sep 17 00:00:00 2001 From: Stefano Garzarella Date: Mon, 18 May 2026 11:06:55 +0200 Subject: vsock/virtio: reset connection on receiving queue overflow When there is no more space to queue an incoming packet, the packet is silently dropped. This causes data loss without any notification to either peer, since there is no retransmission. Under normal circumstances, this should never happen. However, it could happen if the other peer doesn't respect the credit, or if the skb overhead, which we recently began to take into account with commit 059b7dbd20a6 ("vsock/virtio: fix potential unbounded skb queue"), is too high. Fix this by resetting the connection and setting the local socket error to ENOBUFS when virtio_transport_recv_enqueue() can no longer queue a packet, so both peers are explicitly notified of the failure rather than silently losing data. Fixes: ae6fcfbf5f03 ("vsock/virtio: discard packets if credit is not respected") Cc: stable@vger.kernel.org Signed-off-by: Stefano Garzarella Link: https://patch.msgid.link/20260518090656.134588-2-sgarzare@redhat.com Signed-off-by: Paolo Abeni --- net/vmw_vsock/virtio_transport_common.c | 20 +++++++++++++++----- 1 file changed, 15 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/vmw_vsock/virtio_transport_common.c b/net/vmw_vsock/virtio_transport_common.c index 1e3409d28164..5028ff534888 100644 --- a/net/vmw_vsock/virtio_transport_common.c +++ b/net/vmw_vsock/virtio_transport_common.c @@ -1335,7 +1335,7 @@ destroy: return err; } -static void +static bool virtio_transport_recv_enqueue(struct vsock_sock *vsk, struct sk_buff *skb) { @@ -1350,10 +1350,8 @@ virtio_transport_recv_enqueue(struct vsock_sock *vsk, spin_lock_bh(&vvs->rx_lock); can_enqueue = virtio_transport_inc_rx_pkt(vvs, len); - if (!can_enqueue) { - free_pkt = true; + if (!can_enqueue) goto out; - } if (le32_to_cpu(hdr->flags) & VIRTIO_VSOCK_SEQ_EOM) vvs->msg_count++; @@ -1393,6 +1391,8 @@ out: spin_unlock_bh(&vvs->rx_lock); if (free_pkt) kfree_skb(skb); + + return can_enqueue; } static int @@ -1405,7 +1405,17 @@ virtio_transport_recv_connected(struct sock *sk, switch (le16_to_cpu(hdr->op)) { case VIRTIO_VSOCK_OP_RW: - virtio_transport_recv_enqueue(vsk, skb); + if (!virtio_transport_recv_enqueue(vsk, skb)) { + /* There is no more space to queue the packet, so let's + * close the connection; otherwise, we'll lose data. + */ + (void)virtio_transport_reset(vsk, skb); + virtio_transport_do_close(vsk, true); + sk->sk_err = ENOBUFS; + sk_error_report(sk); + vsock_remove_sock(vsk); + break; + } vsock_data_ready(sk); return err; case VIRTIO_VSOCK_OP_CREDIT_REQUEST: -- cgit v1.2.3 From c6087c5aaad6d1b8be1a1a641e0a422218ade911 Mon Sep 17 00:00:00 2001 From: Stefano Garzarella Date: Mon, 18 May 2026 11:06:56 +0200 Subject: vsock/virtio: fix skb overhead accounting to preserve full buf_alloc After commit 059b7dbd20a6 ("vsock/virtio: fix potential unbounded skb queue"), virtio_transport_inc_rx_pkt() subtracts per-skb overhead from buf_alloc when checking whether a new packet fits. This reduces the effective receive buffer below what the user configured via SO_VM_SOCKETS_BUFFER_SIZE, causing legitimate data packets to be silently dropped and applications that rely on the full buffer size to deadlock. Also, the reduced space is not communicated to the remote peer, so its credit calculation accounts more credit than the receiver will actually accept, causing data loss (there is no retransmission). With this approach we currently have failures in tools/testing/vsock/vsock_test.c. Test 18 sometimes fails, while test 22 always fails in this way: 18 - SOCK_STREAM MSG_ZEROCOPY...hash mismatch 22 - SOCK_STREAM virtio credit update + SO_RCVLOWAT...send failed: Resource temporarily unavailable Fix by allowing at most `buf_alloc * 2` as the total budget for payload plus skb overhead in virtio_transport_inc_rx_pkt(), similar to how SO_RCVBUF is doubled to reserve space for sk_buff metadata. This preserves the full buf_alloc for payload under normal operation, while still bounding the skb queue growth. With this patch, all tests in tools/testing/vsock/vsock_test.c are now passing again. Fixes: 059b7dbd20a6 ("vsock/virtio: fix potential unbounded skb queue") Cc: stable@vger.kernel.org Signed-off-by: Stefano Garzarella Link: https://patch.msgid.link/20260518090656.134588-3-sgarzare@redhat.com Signed-off-by: Paolo Abeni --- net/vmw_vsock/virtio_transport_common.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/vmw_vsock/virtio_transport_common.c b/net/vmw_vsock/virtio_transport_common.c index 5028ff534888..df3b418e0392 100644 --- a/net/vmw_vsock/virtio_transport_common.c +++ b/net/vmw_vsock/virtio_transport_common.c @@ -419,7 +419,14 @@ static bool virtio_transport_inc_rx_pkt(struct virtio_vsock_sock *vvs, { u64 skb_overhead = (skb_queue_len(&vvs->rx_queue) + 1) * SKB_TRUESIZE(0); - if (skb_overhead + vvs->buf_used + len > vvs->buf_alloc) + /* Allow at most buf_alloc * 2 total budget (payload + overhead), + * similar to how SO_RCVBUF is doubled to reserve space for sk_buff + * metadata. Check payload against buf_alloc to be sure the other + * peer is respecting the credit, and sk_buff overhead to bound + * queue growth. + */ + if ((u64)vvs->buf_used + len > vvs->buf_alloc || + skb_overhead > vvs->buf_alloc) return false; vvs->rx_bytes += len; -- cgit v1.2.3 From dbc81608e3a653dea6cf403f20cae35468b8ab9c Mon Sep 17 00:00:00 2001 From: Zijing Yin Date: Tue, 19 May 2026 10:26:33 -0700 Subject: phonet/pep: disable BH around forwarded sk_receive_skb() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The networking receive path is usually run from softirq context, but protocols that take the socket lock may have packets stored in the backlog and processed later from process context. In that case release_sock() -> __release_sock() drops the slock with spin_unlock_bh() and then calls sk->sk_backlog_rcv() with bottom halves enabled. Typical sk_backlog_rcv handlers process the socket whose backlog is being drained, so the BH state at entry is irrelevant for the slocks they touch. pep_do_rcv() is different: when the inbound skb targets an existing PEP pipe, it forwards the skb to a different *child* socket via sk_receive_skb(). That helper takes the child slock with bh_lock_sock_nested(), which is just spin_lock_nested() and assumes BH is already off. The same child slock therefore ends up acquired with BH on (process path) and with BH off (softirq path): process context softirq context --------------- --------------- release_sock(listener) __netif_receive_skb() __release_sock() phonet_rcv() spin_unlock_bh() __sk_receive_skb(listener) [BH now ENABLED] [BH already disabled] sk_backlog_rcv: sk_backlog_rcv: pep_do_rcv() pep_do_rcv() sk_receive_skb(child) sk_receive_skb(child) bh_lock_sock_nested(child) bh_lock_sock_nested(child) => SOFTIRQ-ON-W => IN-SOFTIRQ-W Lockdep flags this as inconsistent lock state, and it can become a real self-deadlock if a softirq on the same CPU tries to receive to the same child socket while its slock is held in the BH-enabled path: WARNING: inconsistent lock state inconsistent {SOFTIRQ-ON-W} -> {IN-SOFTIRQ-W} usage. (slock-AF_PHONET/1){+.?.}-{3:3}, at: __sk_receive_skb+0x1cf/0x900 __sk_receive_skb net/core/sock.c:563 sk_receive_skb include/net/sock.h:2022 [inline] pep_do_rcv net/phonet/pep.c:675 sk_backlog_rcv include/net/sock.h:1190 __release_sock net/core/sock.c:3216 release_sock net/core/sock.c:3815 pep_sock_accept net/phonet/pep.c:879 Wrap the forwarded sk_receive_skb() in local_bh_disable() / local_bh_enable() so the child slock is always acquired with BH off. local_bh_disable() nests safely on the softirq path. Discovered via in-house syzkaller fuzzing; the same root cause also on the linux-6.1.y syzbot dashboard as extid 44f0626dd6284f02663c. Reproduced under KASAN + LOCKDEP + PROVE_LOCKING, reproducer: https://pastebin.com/A3t8xzCR Fixes: 9641458d3ec4 ("Phonet: Pipe End Point for Phonet Pipes protocol") Link: https://syzkaller.appspot.com/bug?extid=44f0626dd6284f02663c Cc: stable@vger.kernel.org Signed-off-by: Zijing Yin Acked-by: Rémi Denis-Courmont Reported-by: syzbot+9f4a135646b66c509935@syzkaller.appspotmail.com Reviewed-by: Eric Dumazet Link: https://patch.msgid.link/20260519172635.86304-1-yzjaurora@gmail.com Signed-off-by: Jakub Kicinski --- net/phonet/pep.c | 19 +++++++++++++++++-- 1 file changed, 17 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/phonet/pep.c b/net/phonet/pep.c index 4dbf0914df7d..706927139393 100644 --- a/net/phonet/pep.c +++ b/net/phonet/pep.c @@ -671,8 +671,23 @@ static int pep_do_rcv(struct sock *sk, struct sk_buff *skb) /* Look for an existing pipe handle */ sknode = pep_find_pipe(&pn->hlist, &dst, pipe_handle); - if (sknode) - return sk_receive_skb(sknode, skb, 1); + if (sknode) { + int rc; + + /* pep_do_rcv() runs from two contexts: from softirq via + * phonet_rcv() -> __sk_receive_skb() with BH disabled, + * and from process context via + * release_sock() -> __release_sock(), which drops + * the listener slock with spin_unlock_bh() before draining + * the backlog. The child pipe slock is taken below via + * bh_lock_sock_nested(), which does not itself disable BH, so + * disable BH here to keep both acquire contexts consistent. + */ + local_bh_disable(); + rc = sk_receive_skb(sknode, skb, 1); + local_bh_enable(); + return rc; + } switch (hdr->message_id) { case PNS_PEP_CONNECT_REQ: -- cgit v1.2.3 From 1341db322417266fb5845df81d28305b83a37324 Mon Sep 17 00:00:00 2001 From: Yuho Choi Date: Tue, 19 May 2026 23:03:28 -0400 Subject: ipv6: route: Unregister netdevice notifier on BPF init failure ip6_route_init() registers ip6_route_dev_notifier before registering the IPv6 route BPF iterator target. If bpf_iter_register() fails after the notifier has been registered, the error path currently jumps to out_register_late_subsys and unwinds the RTNL handlers and pernet route state without removing the notifier from the netdevice notifier chain. This leaves ip6_route_dev_notify() callable after the IPv6 route state it uses has been torn down. Add a separate unwind label for the BPF iterator failure path and unregister the netdevice notifier before continuing with the existing cleanup. Fixes: 138d0be35b14 ("net: bpf: Add netlink and ipv6_route bpf_iter targets") Signed-off-by: Yuho Choi Reviewed-by: Ido Schimmel Link: https://patch.msgid.link/20260520030329.1061183-1-dbgh9129@gmail.com Signed-off-by: Jakub Kicinski --- net/ipv6/route.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv6/route.c b/net/ipv6/route.c index e3d355d1fbd6..b106e5fef9cb 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -6933,7 +6933,7 @@ int __init ip6_route_init(void) #if defined(CONFIG_BPF_SYSCALL) && defined(CONFIG_PROC_FS) ret = bpf_iter_register(); if (ret) - goto out_register_late_subsys; + goto out_register_notifier; #endif for_each_possible_cpu(cpu) { @@ -6946,6 +6946,10 @@ int __init ip6_route_init(void) out: return ret; +#if defined(CONFIG_BPF_SYSCALL) && defined(CONFIG_PROC_FS) +out_register_notifier: + unregister_netdevice_notifier(&ip6_route_dev_notifier); +#endif out_register_late_subsys: rtnl_unregister_all(PF_INET6); unregister_pernet_subsys(&ip6_route_net_late_ops); -- cgit v1.2.3 From c367b9082194d01cb38bdefac6e887ebf1ab017d Mon Sep 17 00:00:00 2001 From: Zhang Cen Date: Tue, 19 May 2026 18:46:47 +0800 Subject: netpoll: normalize skb->dev to the netpoll device __netpoll_send_skb() always transmits through np->dev and queues busy packets on np->dev->npinfo->txq, but it leaves skb->dev unchanged. Stacked callers such as DSA and macvlan can reach netpoll with skb->dev still naming the upper device while np->dev is the lower device that owns the netpoll state. If the skb has to be deferred, queue_process() later dequeues it from the lower device's txq but retries it through skb->dev. That can re-enter the upper ndo_start_xmit path on an already transformed skb, and if the upper device disappears before the lower txq drains the workqueue can dereference a stale skb->dev pointer. The buggy scenario involves two paths, with each column showing the order within that path: path A label: netpoll enqueue path path B label: upper-device teardown 1. Stacked xmit calls netpoll 1. Teardown unregisters the upper with lower np->dev and upper net_device while lower npinfo skb->dev. stays alive. 2. __netpoll_send_skb() uses 2. netdev_release() runs for the np->dev->npinfo as the txq upper net_device. owner. 3. Busy transmit queues the skb 3. The lower txq still owns the on that lower txq with upper deferred skb. skb->dev. 4. queue_process() drains the 4. queue_process() dereferences lower txq and reads skb->dev. that stale upper skb->dev. Normalize skb->dev to np->dev after loading np->dev from the netpoll instance, before either the direct transmit path or the fallback enqueue. This keeps the queued skb in the same device and txq domain as the netpoll state that owns it. KASAN report as below: KASAN slab-use-after-free in queue_process+0x7c/0x480 Workqueue: events queue_process The buggy address belongs to the object at ffff88810906c000 which belongs to the cache kmalloc-4k of size 4096 The buggy address is located 168 bytes inside of freed 4096-byte region [ffff88810906c000, ffff88810906d000) Read of size 8 Call trace: dump_stack_lvl+0x73/0xb0 (?:?) print_report+0xd1/0x620 (?:?) srso_alias_return_thunk+0x5/0xfbef5 (?:?) __virt_addr_valid+0x215/0x420 (?:?) kasan_complete_mode_report_info+0x64/0x200 (?:?) kasan_report+0xf7/0x130 (?:?) queue_process+0x7c/0x480 (net/core/netpoll.c:88) kasan_check_range+0x10c/0x1c0 (?:?) __kasan_check_read+0x15/0x20 (?:?) process_one_work+0x8b7/0x1af0 (kernel/workqueue.c:3200) assign_work+0x170/0x3f0 (?:?) worker_thread+0x574/0xf10 (?:?) _raw_spin_unlock_irqrestore+0x4b/0x60 (?:?) trace_hardirqs_on+0x2a/0x180 (?:?) kthread+0x2fc/0x3f0 (?:?) ret_from_fork+0x58b/0x830 (?:?) __switch_to+0x58e/0xe90 (?:?) __switch_to_asm+0x39/0x70 (?:?) ret_from_fork_asm+0x1a/0x30 (?:?) Freed by task stack: kasan_save_stack+0x3d/0x60 (?:?) kasan_save_track+0x18/0x40 (?:?) kasan_save_free_info+0x3f/0x60 (?:?) __kasan_slab_free+0x48/0x70 (?:?) kfree+0x20e/0x4e0 (?:?) kvfree+0x31/0x40 (?:?) netdev_release+0x71/0x90 (net/core/net-sysfs.c:2227) device_release+0xd2/0x250 (?:?) kobject_put+0x181/0x4c0 (lib/kobject.c:730) netdev_run_todo+0x700/0x1000 (net/core/dev.c:11666) rtnl_dellink+0x396/0xc00 (net/core/rtnetlink.c:3558) rtnetlink_rcv_msg+0x740/0xc20 (net/core/rtnetlink.c:6897) netlink_rcv_skb+0x147/0x3a0 (?:?) rtnetlink_rcv+0x19/0x20 (net/core/rtnetlink.c:7021) netlink_unicast+0x4d1/0x830 (net/netlink/af_netlink.c:1327) netlink_sendmsg+0x840/0xe10 (net/netlink/af_netlink.c:1812) ____sys_sendmsg+0x8a7/0xb50 (?:?) ___sys_sendmsg+0x104/0x190 (?:?) __sys_sendmsg+0x135/0x1d0 (?:?) __x64_sys_sendmsg+0x7b/0xc0 (?:?) x64_sys_call+0x205c/0x2130 (?:?) do_syscall_64+0x115/0x6a0 (arch/x86/entry/syscall_64.c:87) entry_SYSCALL_64_after_hwframe+0x77/0x7f (?:?) Fixes: 5de4a473bda4 ("netpoll queue cleanup") Signed-off-by: Zhang Cen Link: https://patch.msgid.link/20260519104647.3517990-1-rollkingzzc@gmail.com Signed-off-by: Jakub Kicinski --- net/core/netpoll.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'net') diff --git a/net/core/netpoll.c b/net/core/netpoll.c index 84faace50ac2..3f4a17fa5713 100644 --- a/net/core/netpoll.c +++ b/net/core/netpoll.c @@ -319,6 +319,8 @@ static netdev_tx_t __netpoll_send_skb(struct netpoll *np, struct sk_buff *skb) lockdep_assert_irqs_disabled(); dev = np->dev; + /* npinfo->txq belongs to np->dev, so retries must stay bound to it. */ + skb->dev = dev; rcu_read_lock(); npinfo = rcu_dereference_bh(dev->npinfo); -- cgit v1.2.3 From e46e6bc97fb1f339730ff1ba74267fbf48e7a422 Mon Sep 17 00:00:00 2001 From: Justin Iurman Date: Wed, 20 May 2026 14:42:42 +0200 Subject: ipv6: ioam: refresh hdr pointer before ioam6_event() Reported by Sashiko: In ipv6_hop_ioam(), the hdr pointer is initialized to point into the skb's linear data buffer. Later, the code calls skb_ensure_writable(), which might reallocate the buffer: if (skb_ensure_writable(skb, optoff + 2 + hdr->opt_len)) goto drop; /* Trace pointer may have changed */ trace = (struct ioam6_trace_hdr *)(skb_network_header(skb) + optoff + sizeof(*hdr)); ioam6_fill_trace_data(skb, ns, trace, true); ioam6_event(IOAM6_EVENT_TRACE, dev_net(skb->dev), GFP_ATOMIC, (void *)trace, hdr->opt_len - 2); If the skb is cloned or lacks sufficient linear headroom, skb_ensure_writable() will invoke pskb_expand_head(), which reallocates the skb's data buffer and frees the old one, invalidating pointers to it. While the code recalculates the trace pointer immediately after the call to skb_ensure_writable(), it fails to recalculate the hdr pointer. This patch fixes the above by recalculating the hdr pointer before passing hdr->opt_len to ioam6_event(), so that we avoid any UaF. Fixes: f655c78d6225 ("net: exthdrs: ioam6: send trace event") Cc: stable@vger.kernel.org Signed-off-by: Justin Iurman Reviewed-by: Ido Schimmel Link: https://patch.msgid.link/20260520124242.32320-1-justin.iurman@gmail.com Signed-off-by: Jakub Kicinski --- net/ipv6/exthdrs.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/ipv6/exthdrs.c b/net/ipv6/exthdrs.c index 47c5502a34a2..cf90f933ca1a 100644 --- a/net/ipv6/exthdrs.c +++ b/net/ipv6/exthdrs.c @@ -966,9 +966,9 @@ static bool ipv6_hop_ioam(struct sk_buff *skb, int optoff) if (skb_ensure_writable(skb, optoff + 2 + hdr->opt_len)) goto drop; - /* Trace pointer may have changed */ - trace = (struct ioam6_trace_hdr *)(skb_network_header(skb) - + optoff + sizeof(*hdr)); + /* Trace and hdr pointers may have changed */ + hdr = (struct ioam6_hdr *)(skb_network_header(skb) + optoff); + trace = (struct ioam6_trace_hdr *)((u8 *)hdr + sizeof(*hdr)); ioam6_fill_trace_data(skb, ns, trace, true); -- cgit v1.2.3 From 4db79a322db8c97f7b73b8a347395ef4d685eb40 Mon Sep 17 00:00:00 2001 From: Sabrina Dubroca Date: Wed, 20 May 2026 22:44:42 +0200 Subject: net: gro: don't merge zcopy skbs skb_gro_receive() can currently copy frags between the source and GRO skb, without checking the zerocopy status, and in particular the SKBFL_MANAGED_FRAG_REFS flag. When SKBFL_MANAGED_FRAG_REFS is set, the skb doesn't hold a reference on the pages in shinfo->frags. Appending those frags to another skb's frags without fixing up the page refcount can lead to UAF. When either the last skb in the GRO chain (the one we would append frags to) or the source skb is zerocopy, don't merge the skbs. Fixes: 753f1ca4e1e5 ("net: introduce managed frags infrastructure") Reported-by: Huzaifa Sidhpurwala Signed-off-by: Sabrina Dubroca Reviewed-by: Willem de Bruijn Link: https://patch.msgid.link/c3b7f906bbfcbdfd7b4fa9d6c18a438870df85be.1779307748.git.sd@queasysnail.net Signed-off-by: Jakub Kicinski --- net/core/gro.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'net') diff --git a/net/core/gro.c b/net/core/gro.c index 9f8960789b2c..a84753983467 100644 --- a/net/core/gro.c +++ b/net/core/gro.c @@ -109,6 +109,9 @@ int skb_gro_receive(struct sk_buff *p, struct sk_buff *skb) if (p->pp_recycle != skb->pp_recycle) return -ETOOMANYREFS; + if (skb_zcopy(p) || skb_zcopy(skb)) + return -ETOOMANYREFS; + if (unlikely(p->len + len >= netif_get_gro_max_size(p->dev, p) || NAPI_GRO_CB(skb)->flush)) return -E2BIG; -- cgit v1.2.3 From 979c017803c40829b03acd9e5236e354b7622360 Mon Sep 17 00:00:00 2001 From: Michael Bommarito Date: Mon, 18 May 2026 14:34:47 -0400 Subject: l2tp: use list_del_rcu in l2tp_session_unhash An unprivileged local user can pin a host CPU indefinitely in l2tp_session_get_by_ifname() by issuing L2TP_CMD_SESSION_GET on L2TP_ATTR_IFNAME concurrently with L2TP_CMD_SESSION_CREATE and L2TP_CMD_SESSION_DELETE on the same tunnel. All three commands take GENL_UNS_ADMIN_PERM, so CAP_NET_ADMIN in the netns user namespace suffices; on any host that has l2tp_core loaded the trigger is reachable from a standard `unshare -Urn` sandbox. l2tp_session_unhash() removes a session from tunnel->session_list with list_del_init(), but that list is walked by l2tp_session_get_by_ifname() with list_for_each_entry_rcu() under rcu_read_lock_bh(). list_del_init() leaves the deleted entry's next/prev self-pointing; a reader that has loaded the entry and then advances pos->list.next reads &session->list, container_of()s back to the same session, and list_for_each_entry_rcu() never reaches the list head. The CPU stays in strcmp() inside the walker, with BH and preemption disabled, so RCU grace periods on the host stall behind it and the wedged thread cannot be killed (SIGKILL is delivered on syscall return). Use list_del_rcu() to match the existing list_add_rcu() in l2tp_session_register(); the deleted session remains visible to in-flight walkers with consistent next/prev pointers until kfree_rcu() in l2tp_session_free() releases it. tunnel->session_list has exactly one list_del_init() call site; the list_del_init (&session->clist) at l2tp_core.c:533 operates on the per-collision list, which is not walked under RCU. list_empty(&session->list) is not used anywhere in net/l2tp/ after the unhash point, so dropping the post-delete self-init is safe; the fix has no userspace-visible behavior change. Fixes: 89b768ec2dfef ("l2tp: use rcu list add/del when updating lists") Cc: stable@vger.kernel.org # 6.11+ Signed-off-by: Michael Bommarito Link: https://patch.msgid.link/20260518183447.64078-1-michael.bommarito@gmail.com Signed-off-by: Jakub Kicinski --- net/l2tp/l2tp_core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/l2tp/l2tp_core.c b/net/l2tp/l2tp_core.c index 157fc23ce4e1..1455f67e01dd 100644 --- a/net/l2tp/l2tp_core.c +++ b/net/l2tp/l2tp_core.c @@ -1360,7 +1360,7 @@ static void l2tp_session_unhash(struct l2tp_session *session) spin_lock_bh(&pn->l2tp_session_idr_lock); /* Remove from the per-tunnel list */ - list_del_init(&session->list); + list_del_rcu(&session->list); /* Remove from per-net IDR */ if (tunnel->version == L2TP_HDR_VER_3) { -- cgit v1.2.3 From bdd39576bf50a50bdafe3da968fd271bc674a48f Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 20 May 2026 11:42:07 +0000 Subject: net: bridge: prevent too big nested attributes in br_fill_linkxstats() After commit ff205bf8c554 ("netlink: add one debug check in nla_nest_end()") syzbot found that br_fill_linkxstats() can send corrupted netlink packets. Make sure the nested attribute size is bounded. Fixes: a60c090361ea ("bridge: netlink: export per-vlan stats") Reported-by: syzbot+a35f9259d08f907c06e6@syzkaller.appspotmail.com Closes: https://lore.kernel.org/netdev/6a0b0da3.050a0220.175f0c.0000.GAE@google.com/ Signed-off-by: Eric Dumazet Reviewed-by: Ido Schimmel Acked-by: Nikolay Aleksandrov Link: https://patch.msgid.link/20260520114207.1394241-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/bridge/br_netlink.c | 10 ++++++++++ net/core/rtnetlink.c | 5 +++-- 2 files changed, 13 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/bridge/br_netlink.c b/net/bridge/br_netlink.c index 6fd5386a1d64..c04a4d0889ae 100644 --- a/net/bridge/br_netlink.c +++ b/net/bridge/br_netlink.c @@ -1824,6 +1824,7 @@ static int br_fill_linkxstats(struct sk_buff *skb, const struct net_device *dev, int *prividx, int attr) { + unsigned int limit = U16_MAX - nla_total_size(0); struct nlattr *nla __maybe_unused; struct net_bridge_port *p = NULL; struct net_bridge_vlan_group *vg; @@ -1841,6 +1842,7 @@ static int br_fill_linkxstats(struct sk_buff *skb, p = br_port_get_rtnl(dev); if (!p) return 0; + limit -= nla_total_size_64bit(sizeof(p->stp_xstats)); br = p->br; vg = nbp_vlan_group(p); break; @@ -1855,6 +1857,9 @@ static int br_fill_linkxstats(struct sk_buff *skb, if (vg) { u16 pvid; +#ifdef CONFIG_BRIDGE_IGMP_SNOOPING + limit -= nla_total_size_64bit(sizeof(struct br_mcast_stats)); +#endif pvid = br_get_pvid(vg); list_for_each_entry(v, &vg->vlan_list, vlist) { struct bridge_vlan_xstats vxi; @@ -1862,6 +1867,11 @@ static int br_fill_linkxstats(struct sk_buff *skb, if (++vl_idx < *prividx) continue; + + if (skb_tail_pointer(skb) - (unsigned char *)nest + + nla_total_size(sizeof(vxi)) >= limit) + goto nla_put_failure; + memset(&vxi, 0, sizeof(vxi)); vxi.vid = v->vid; vxi.flags = v->flags; diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index df042da422ef..511c25bf6f2a 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -6328,8 +6328,9 @@ static int rtnl_stats_get(struct sk_buff *skb, struct nlmsghdr *nlh, NETLINK_CB(skb).portid, nlh->nlmsg_seq, 0, 0, &filters, &idxattr, &prividx, extack); if (err < 0) { - /* -EMSGSIZE implies BUG in if_nlmsg_stats_size */ - WARN_ON(err == -EMSGSIZE); + /* -EMSGSIZE implies BUG in if_nlmsg_stats_size + * or a too big nested attribute. + */ kfree_skb(nskb); } else { err = rtnl_unicast(nskb, net, NETLINK_CB(skb).portid); -- cgit v1.2.3 From ea25e3c7915b24e0ef93ee85190f3fada037dfb1 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Tue, 21 Apr 2026 12:11:25 -0400 Subject: sunrpc: prevent out-of-bounds read in __cache_seq_start() Commit 7b546bd89975 ("sunrpc/cache: improve RCU safety in cache_list walking.") changed the tail of __cache_seq_start() to unconditionally store *pos = ((long long)hash << 32) + 1 before returning, dropping a prior "hash >= cd->hash_size" guard. When the while loop exits because every remaining bucket was empty, hash equals cd->hash_size, so the stored *pos is one position past the table's last valid bucket. seq_read_iter() caches that index in m->index. A subsequent pread(2) at the same file offset skips traverse() and hands the stored index back to __cache_seq_start(), which decodes hash = cd->hash_size and dereferences cd->hash_table[cd->hash_size] -- one hlist_head past the end of the kzalloc'd table. KASAN reports an eight-byte slab-out-of-bounds read at the tail of the 2048-byte hash_table allocation for the NFSD export cache (EXPORT_HASHMAX * sizeof(struct hlist_head) == 256 * 8). Reject an input hash that is out of range before touching the hash table. cache_seq_next() already bounds-checks its own loop; the start routine needs to be symmetric. Reported-by: syzbot+60cfa08822470bbebe44@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=60cfa08822470bbebe44 Fixes: 7b546bd89975 ("sunrpc/cache: improve RCU safety in cache_list walking.") Reviewed-by: Benjamin Coddington Reviewed-by: NeilBrown Signed-off-by: Chuck Lever --- net/sunrpc/cache.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'net') diff --git a/net/sunrpc/cache.c b/net/sunrpc/cache.c index b5474ce534fb..27dd6b58b8ff 100644 --- a/net/sunrpc/cache.c +++ b/net/sunrpc/cache.c @@ -1348,6 +1348,9 @@ static void *__cache_seq_start(struct seq_file *m, loff_t *pos) hash = n >> 32; entry = n & ((1LL<<32) - 1); + if (hash >= cd->hash_size) + return NULL; + hlist_for_each_entry_rcu(ch, &cd->hash_table[hash], cache_list) if (!entry--) return ch; -- cgit v1.2.3 From 54cf41c969da6637cce790b7400da1451609db9b Mon Sep 17 00:00:00 2001 From: Byungchul Park Date: Fri, 15 May 2026 12:47:01 +0900 Subject: Revert "mm: introduce a new page type for page pool in page type" This reverts commit db359fccf212 ("mm: introduce a new page type for page pool in page type") and a part of 735a309b4bfb9e ("net: add net_iov_init() and use it to initialize ->page_type"). Netpp page_type'ed pages might be used in mapping so as to use @_mapcount. However, since @page_type and @_mapcount are union'ed in struct page, these two can't be used at the same time. Revert the commit introducing page_type for Netpp for now. The patch will be retried once @page_type and @_mapcount get allowed to be used at the same time. The revert also includes removal of @page_type initialization part introduced by commit 735a309b4bfb9e ("net: add net_iov_init() and use it to initialize ->page_type"), which will be restored on the retry. Link: https://lore.kernel.org/20260515034701.17027-1-byungchul@sk.com Fixes: db359fccf212 ("mm: introduce a new page type for page pool in page type") Signed-off-by: Byungchul Park Reported-by: Dragos Tatulea Closes: https://lore.kernel.org/all/982b9bc1-0a0a-4fc5-8e3a-3672db2b29a1@nvidia.com Acked-by: Jakub Kicinski Acked-by: David Hildenbrand (Arm) Acked-by: Harry Yoo (Oracle) Reviewed-by: Lorenzo Stoakes Cc: Alexei Starovoitov Cc: Baolin Wang Cc: Brendan Jackman Cc: David S. Miller Cc: Eric Dumazet Cc: Ilias Apalodimas Cc: Jesper Dangaard Brouer Cc: Johannes Weiner Cc: John Fastabend Cc: Leon Romanovsky Cc: Liam R. Howlett Cc: Mark Bloch Cc: Matthew Wilcox (Oracle) Cc: Michal Hocko Cc: Mike Rapoport Cc: Paolo Abeni Cc: Pavel Begunkov Cc: Saeed Mahameed Cc: Simon Horman Cc: Stanislav Fomichev Cc: Suren Baghdasaryan Cc: Tariq Toukan Cc: Toke Hoiland-Jorgensen Cc: Vlastimil Babka Cc: Zi Yan Signed-off-by: Andrew Morton --- net/core/netmem_priv.h | 23 +++++++++++++---------- net/core/page_pool.c | 24 ++---------------------- 2 files changed, 15 insertions(+), 32 deletions(-) (limited to 'net') diff --git a/net/core/netmem_priv.h b/net/core/netmem_priv.h index 3e6fde8f1726..23175cb2bd86 100644 --- a/net/core/netmem_priv.h +++ b/net/core/netmem_priv.h @@ -8,18 +8,21 @@ static inline unsigned long netmem_get_pp_magic(netmem_ref netmem) return netmem_to_nmdesc(netmem)->pp_magic & ~PP_DMA_INDEX_MASK; } -static inline bool netmem_is_pp(netmem_ref netmem) +static inline void netmem_or_pp_magic(netmem_ref netmem, unsigned long pp_magic) +{ + netmem_to_nmdesc(netmem)->pp_magic |= pp_magic; +} + +static inline void netmem_clear_pp_magic(netmem_ref netmem) { - struct page *page; + WARN_ON_ONCE(netmem_to_nmdesc(netmem)->pp_magic & PP_DMA_INDEX_MASK); - /* XXX: Now that the offset of page_type is shared between - * struct page and net_iov, just cast the netmem to struct page - * unconditionally by clearing NET_IOV if any, no matter whether - * it comes from struct net_iov or struct page. This should be - * adjusted once the offset is no longer shared. - */ - page = (struct page *)((__force unsigned long)netmem & ~NET_IOV); - return PageNetpp(page); + netmem_to_nmdesc(netmem)->pp_magic = 0; +} + +static inline bool netmem_is_pp(netmem_ref netmem) +{ + return (netmem_get_pp_magic(netmem) & PP_MAGIC_MASK) == PP_SIGNATURE; } static inline void netmem_set_pp(netmem_ref netmem, struct page_pool *pool) diff --git a/net/core/page_pool.c b/net/core/page_pool.c index 6e576dec80db..8171d1173221 100644 --- a/net/core/page_pool.c +++ b/net/core/page_pool.c @@ -707,18 +707,8 @@ s32 page_pool_inflight(const struct page_pool *pool, bool strict) void page_pool_set_pp_info(struct page_pool *pool, netmem_ref netmem) { - struct page *page; - netmem_set_pp(netmem, pool); - - /* XXX: Now that the offset of page_type is shared between - * struct page and net_iov, just cast the netmem to struct page - * unconditionally by clearing NET_IOV if any, no matter whether - * it comes from struct net_iov or struct page. This should be - * adjusted once the offset is no longer shared. - */ - page = (struct page *)((__force unsigned long)netmem & ~NET_IOV); - __SetPageNetpp(page); + netmem_or_pp_magic(netmem, PP_SIGNATURE); /* Ensuring all pages have been split into one fragment initially: * page_pool_set_pp_info() is only called once for every page when it @@ -733,17 +723,7 @@ void page_pool_set_pp_info(struct page_pool *pool, netmem_ref netmem) void page_pool_clear_pp_info(netmem_ref netmem) { - struct page *page; - - /* XXX: Now that the offset of page_type is shared between - * struct page and net_iov, just cast the netmem to struct page - * unconditionally by clearing NET_IOV if any, no matter whether - * it comes from struct net_iov or struct page. This should be - * adjusted once the offset is no longer shared. - */ - page = (struct page *)((__force unsigned long)netmem & ~NET_IOV); - __ClearPageNetpp(page); - + netmem_clear_pp_magic(netmem); netmem_set_pp(netmem, NULL); } -- cgit v1.2.3 From 2982e599fff6faa21c8df147d96fc7af6c1a2f24 Mon Sep 17 00:00:00 2001 From: e521588 Date: Wed, 20 May 2026 09:27:17 +0200 Subject: esp: fix page frag reference leak on skb_to_sgvec failure In esp_output_tail(), when esp->inplace is false, the old skb page frags are replaced with a new page from the xfrm page_frag cache. The source scatterlist (sg) is built from the old frags before the replacement, and esp_ssg_unref() is responsible for releasing the old page references after the crypto operation completes. However, if the second skb_to_sgvec() call (which builds the destination scatterlist from the new page) fails, the code jumps to error_free which only calls kfree(tmp). The old page frag references captured in the source scatterlist are never released: 1. sg[] is built from old frags via skb_to_sgvec() (no extra get_page) 2. nr_frags is set to 1 and frag[0] is replaced with the new page 3. Second skb_to_sgvec() fails -> goto error_free 4. kfree(tmp) frees the sg[] memory but old frags are not unref'd 5. kfree_skb() only releases frag[0] (the new page), not the old ones Fix this by adding a bool parameter to esp_ssg_unref() that, when true, unconditionally unrefs the source scatterlist frags without checking req->src and req->dst, since those fields are not yet initialized by aead_request_set_crypt() at the point of the error. Existing callers pass false to preserve the original behavior. The same issue exists in both esp4 and esp6 as the code is identical. Fixes: cac2661c53f3 ("esp4: Avoid skb_cow_data whenever possible") Fixes: 03e2a30f6a27 ("esp6: Avoid skb_cow_data whenever possible") Signed-off-by: Alessandro Schino <7991aleschino@gmail.com> Signed-off-by: Steffen Klassert --- net/ipv4/esp4.c | 12 +++++++----- net/ipv6/esp6.c | 12 +++++++----- 2 files changed, 14 insertions(+), 10 deletions(-) (limited to 'net') diff --git a/net/ipv4/esp4.c b/net/ipv4/esp4.c index 6a5febbdbee4..8314d7bddcb7 100644 --- a/net/ipv4/esp4.c +++ b/net/ipv4/esp4.c @@ -96,7 +96,7 @@ static inline struct scatterlist *esp_req_sg(struct crypto_aead *aead, __alignof__(struct scatterlist)); } -static void esp_ssg_unref(struct xfrm_state *x, void *tmp, struct sk_buff *skb) +static void esp_ssg_unref(struct xfrm_state *x, void *tmp, struct sk_buff *skb, bool already_unref) { struct crypto_aead *aead = x->data; int extralen = 0; @@ -113,7 +113,7 @@ static void esp_ssg_unref(struct xfrm_state *x, void *tmp, struct sk_buff *skb) /* Unref skb_frag_pages in the src scatterlist if necessary. * Skip the first sg which comes from skb->data. */ - if (req->src != req->dst) + if (already_unref || req->src != req->dst) for (sg = sg_next(req->src); sg; sg = sg_next(sg)) skb_page_unref(page_to_netmem(sg_page(sg)), skb->pp_recycle); @@ -220,7 +220,7 @@ static void esp_output_done(void *data, int err) } tmp = ESP_SKB_CB(skb)->tmp; - esp_ssg_unref(x, tmp, skb); + esp_ssg_unref(x, tmp, skb, false); kfree(tmp); if (xo && (xo->flags & XFRM_DEV_RESUME)) { @@ -569,8 +569,10 @@ int esp_output_tail(struct xfrm_state *x, struct sk_buff *skb, struct esp_info * err = skb_to_sgvec(skb, dsg, (unsigned char *)esph - skb->data, assoclen + ivlen + esp->clen + alen); - if (unlikely(err < 0)) + if (unlikely(err < 0)) { + esp_ssg_unref(x, tmp, skb, true); goto error_free; + } } if ((x->props.flags & XFRM_STATE_ESN)) @@ -602,7 +604,7 @@ int esp_output_tail(struct xfrm_state *x, struct sk_buff *skb, struct esp_info * } if (sg != dsg) - esp_ssg_unref(x, tmp, skb); + esp_ssg_unref(x, tmp, skb, false); if (!err && x->encap && x->encap->encap_type == TCP_ENCAP_ESPINTCP) err = esp_output_tail_tcp(x, skb); diff --git a/net/ipv6/esp6.c b/net/ipv6/esp6.c index 9c06c5a1419d..9d0c4957ac62 100644 --- a/net/ipv6/esp6.c +++ b/net/ipv6/esp6.c @@ -113,7 +113,7 @@ static inline struct scatterlist *esp_req_sg(struct crypto_aead *aead, __alignof__(struct scatterlist)); } -static void esp_ssg_unref(struct xfrm_state *x, void *tmp, struct sk_buff *skb) +static void esp_ssg_unref(struct xfrm_state *x, void *tmp, struct sk_buff *skb, bool already_unref) { struct crypto_aead *aead = x->data; int extralen = 0; @@ -130,7 +130,7 @@ static void esp_ssg_unref(struct xfrm_state *x, void *tmp, struct sk_buff *skb) /* Unref skb_frag_pages in the src scatterlist if necessary. * Skip the first sg which comes from skb->data. */ - if (req->src != req->dst) + if (already_unref || req->src != req->dst) for (sg = sg_next(req->src); sg; sg = sg_next(sg)) skb_page_unref(page_to_netmem(sg_page(sg)), skb->pp_recycle); @@ -254,7 +254,7 @@ static void esp_output_done(void *data, int err) } tmp = ESP_SKB_CB(skb)->tmp; - esp_ssg_unref(x, tmp, skb); + esp_ssg_unref(x, tmp, skb, false); kfree(tmp); esp_output_encap_csum(skb); @@ -600,8 +600,10 @@ int esp6_output_tail(struct xfrm_state *x, struct sk_buff *skb, struct esp_info err = skb_to_sgvec(skb, dsg, (unsigned char *)esph - skb->data, assoclen + ivlen + esp->clen + alen); - if (unlikely(err < 0)) + if (unlikely(err < 0)) { + esp_ssg_unref(x, tmp, skb, true); goto error_free; + } } if ((x->props.flags & XFRM_STATE_ESN)) @@ -634,7 +636,7 @@ int esp6_output_tail(struct xfrm_state *x, struct sk_buff *skb, struct esp_info } if (sg != dsg) - esp_ssg_unref(x, tmp, skb); + esp_ssg_unref(x, tmp, skb, false); if (!err && x->encap && x->encap->encap_type == TCP_ENCAP_ESPINTCP) err = esp_output_tail_tcp(x, skb); -- cgit v1.2.3 From dfa0d7b0ff1eb6b2c416b8fdb9b4f2cefba57a40 Mon Sep 17 00:00:00 2001 From: Jingguo Tan Date: Mon, 18 May 2026 17:06:48 +0800 Subject: xfrm: esp: restore combined single-frag length gate The ESP out-of-place fast path appends the trailer in esp_output_head() before esp_output_tail() allocates the destination page frag. The head-side gate currently checks skb->data_len and tailen separately, but the tail code allocates a single destination frag from the combined post-trailer skb->data_len. Reject the page-frag fast path when the combined aligned length exceeds a page. Otherwise skb_page_frag_refill() may fall back to a single page while the destination sg still spans the combined skb->data_len. Restore this combined-length page gate for both IPv4 and IPv6. Fixes: 5bd8baab087d ("esp: limit skb_page_frag_refill use to a single page") Cc: stable@vger.kernel.org Signed-off-by: Lin Ma Signed-off-by: Chenyuan Mi Signed-off-by: Jingguo Tan Reviewed-by: Sabrina Dubroca Signed-off-by: Steffen Klassert --- net/ipv4/esp4.c | 4 ++-- net/ipv6/esp6.c | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/ipv4/esp4.c b/net/ipv4/esp4.c index 8314d7bddcb7..5d3a8656687e 100644 --- a/net/ipv4/esp4.c +++ b/net/ipv4/esp4.c @@ -419,8 +419,8 @@ int esp_output_head(struct xfrm_state *x, struct sk_buff *skb, struct esp_info * return err; } - if (ALIGN(tailen, L1_CACHE_BYTES) > PAGE_SIZE || - ALIGN(skb->data_len, L1_CACHE_BYTES) > PAGE_SIZE) + if (ALIGN(skb->data_len + tailen, L1_CACHE_BYTES) > + PAGE_SIZE) goto cow; if (!skb_cloned(skb)) { diff --git a/net/ipv6/esp6.c b/net/ipv6/esp6.c index 9d0c4957ac62..b963b8e72604 100644 --- a/net/ipv6/esp6.c +++ b/net/ipv6/esp6.c @@ -448,8 +448,8 @@ int esp6_output_head(struct xfrm_state *x, struct sk_buff *skb, struct esp_info return err; } - if (ALIGN(tailen, L1_CACHE_BYTES) > PAGE_SIZE || - ALIGN(skb->data_len, L1_CACHE_BYTES) > PAGE_SIZE) + if (ALIGN(skb->data_len + tailen, L1_CACHE_BYTES) > + PAGE_SIZE) goto cow; if (!skb_cloned(skb)) { -- cgit v1.2.3 From bed6e04be8e6b9133d8b16d5a42d0e0ce674fa9a Mon Sep 17 00:00:00 2001 From: Hamza Mahfooz Date: Mon, 11 May 2026 10:43:14 -0400 Subject: netfilter: conntrack: tcp: do not force CLOSE on invalid-seq RST without direction check An unintended behavior in the TCP conntrack state machine allows a connection to be forced into the CLOSE state using an RST packet with an invalid sequence number. Specifically, after a SYN packet is observed, an RST with an invalid SEQ can transition the conntrack entry to TCP_CONNTRACK_CLOSE, regardless of whether the RST corresponds to the expected reply direction. The relevant code path assumes the RST is a response to an outgoing SYN, but does not validate packet direction or ensure that a matching SYN was actually sent in the opposite direction. As a result, a crafted packet sequence consisting of a SYN followed by an invalid-sequence RST can prematurely terminate an active NAT entry. This makes connection teardown easier than intended. So, tighten the state transition logic to ensure that RST-triggered CLOSE transitions only occur when the RST is a valid response to a previously observed SYN in the correct direction. Cc: stable@vger.kernel.org Fixes: 9fb9cbb1082d ("[NETFILTER]: Add nf_conntrack subsystem.") Signed-off-by: Hamza Mahfooz Signed-off-by: Florian Westphal --- net/netfilter/nf_conntrack_proto_tcp.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/netfilter/nf_conntrack_proto_tcp.c b/net/netfilter/nf_conntrack_proto_tcp.c index b67426c2189b..e99ab1e88e9f 100644 --- a/net/netfilter/nf_conntrack_proto_tcp.c +++ b/net/netfilter/nf_conntrack_proto_tcp.c @@ -1221,7 +1221,8 @@ int nf_conntrack_tcp_packet(struct nf_conn *ct, new_state = old_state; } if (((test_bit(IPS_SEEN_REPLY_BIT, &ct->status) - && ct->proto.tcp.last_index == TCP_SYN_SET) + && ct->proto.tcp.last_index == TCP_SYN_SET + && ct->proto.tcp.last_dir != dir) || (!test_bit(IPS_ASSURED_BIT, &ct->status) && ct->proto.tcp.last_index == TCP_ACK_SET)) && ntohl(th->ack_seq) == ct->proto.tcp.last_end) { -- cgit v1.2.3 From 92170e6afe927ab2792a3f71902845789c8e31b1 Mon Sep 17 00:00:00 2001 From: Chris Mason Date: Tue, 19 May 2026 12:36:14 -0700 Subject: netfilter: synproxy: refresh tcphdr after skb_ensure_writable synproxy_tstamp_adjust() rewrites the TCP timestamp option in place and then patches the TCP checksum via inet_proto_csum_replace4() on the caller-supplied tcphdr pointer. Both ipv4_synproxy_hook() and ipv6_synproxy_hook() obtain that pointer with skb_header_pointer() before calling in, so it may either alias skb->head directly or point at the caller's on-stack _tcph buffer. Between obtaining the pointer and using it, the function calls skb_ensure_writable(skb, optend), which on a cloned or non-linear skb invokes pskb_expand_head() and frees the old skb->head. After that point the cached th is stale: caller (ipv[46]_synproxy_hook) th = skb_header_pointer(skb, ..., &_tcph) synproxy_tstamp_adjust(skb, protoff, th, ...) skb_ensure_writable(skb, optend) pskb_expand_head() /* kfree(old skb->head) */ ... inet_proto_csum_replace4(&th->check, ...) /* writes into freed head, or into the caller's stack copy leaving the on-wire checksum stale */ The option bytes are written through skb->data and are fine; only the checksum update goes through th and so lands in the wrong place. The result is either a write into freed slab memory or a packet leaving with a checksum that does not match its payload. Fix by re-deriving th from skb->data + protoff immediately after skb_ensure_writable() succeeds, so the subsequent checksum update targets the linear, writable header. Fixes: 48b1de4c110a ("netfilter: add SYNPROXY core/target") Assisted-by: kres (claude-opus-4-7) Signed-off-by: Chris Mason Reviewed-by: Fernando Fernandez Mancera Signed-off-by: Florian Westphal --- net/netfilter/nf_synproxy_core.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'net') diff --git a/net/netfilter/nf_synproxy_core.c b/net/netfilter/nf_synproxy_core.c index 57f57e2fc80a..036c8586f49b 100644 --- a/net/netfilter/nf_synproxy_core.c +++ b/net/netfilter/nf_synproxy_core.c @@ -200,6 +200,8 @@ synproxy_tstamp_adjust(struct sk_buff *skb, unsigned int protoff, if (skb_ensure_writable(skb, optend)) return 0; + th = (struct tcphdr *)(skb->data + protoff); + while (optoff < optend) { unsigned char *op = skb->data + optoff; -- cgit v1.2.3 From 47980b6dbf83961eec1c1363ea986e9c06ff8054 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 14 May 2026 14:21:57 +0200 Subject: netfilter: nf_conntrack_gre: fix gre keymap list corruption Quoting reporter: A race between GRE keymap insertion and destruction can corrupt the kernel list or use a freed object. `nf_ct_gre_keymap_add()` publishes a new keymap pointer before the embedded `list_head` is linked, while `nf_ct_gre_keymap_destroy()` can concurrently delete and free that same object. An unprivileged user can reach this through the PPTP conntrack helper by racing PPTP control messages or helper teardown, leading to KASAN-detectable list corruption/UAF in kernel context. ## Root Cause Analysis `exp_gre()` installs GRE expectations for a PPTP control flow and then adds two GRE keymap entries [..] The add path publishes `ct_pptp_info->keymap[dir]` before linking the embedded list node [..] Concurrent teardown deletes that partially initialized object. Make add/destroy symmetric: install both, destroy both while under lock. Furthermore, we should refuse to publish a new mapping in case ct is going away, else we may leak the allocation. The "retrans" detection is strange: existing mapping is checked for key equality with the new mapping, then for "is on the list" via list walk. But I can't see how an existing keymap entry can be NOT on list. Change this to only check if we're asked to map same tuple again -- if so, skip re-install, else signal failure. Last, add a bug trap for the keymap list; it has to be empty when namespace is going away. Reported-by: Leo Lin Signed-off-by: Florian Westphal --- net/netfilter/nf_conntrack_core.c | 8 +++ net/netfilter/nf_conntrack_pptp.c | 8 +-- net/netfilter/nf_conntrack_proto_gre.c | 106 +++++++++++++++++++++++++-------- 3 files changed, 91 insertions(+), 31 deletions(-) (limited to 'net') diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index 8ba5b22a1eef..b521b5ebd664 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -568,6 +568,13 @@ static void destroy_gre_conntrack(struct nf_conn *ct) #endif } +static void warn_on_keymap_list_leak(const struct net *net) +{ +#ifdef CONFIG_NF_CT_PROTO_GRE + WARN_ON_ONCE(!list_empty(&net->ct.nf_ct_proto.gre.keymap_list)); +#endif +} + void nf_ct_destroy(struct nf_conntrack *nfct) { struct nf_conn *ct = (struct nf_conn *)nfct; @@ -2510,6 +2517,7 @@ i_see_dead_people: } list_for_each_entry(net, net_exit_list, exit_list) { + warn_on_keymap_list_leak(net); nf_conntrack_ecache_pernet_fini(net); nf_conntrack_expect_pernet_fini(net); free_percpu(net->ct.stat); diff --git a/net/netfilter/nf_conntrack_pptp.c b/net/netfilter/nf_conntrack_pptp.c index 4c679638df06..dc23e4181618 100644 --- a/net/netfilter/nf_conntrack_pptp.c +++ b/net/netfilter/nf_conntrack_pptp.c @@ -225,13 +225,9 @@ static int exp_gre(struct nf_conn *ct, __be16 callid, __be16 peer_callid) if (nf_ct_expect_related(exp_reply, 0) != 0) goto out_unexpect_orig; - /* Add GRE keymap entries */ - if (nf_ct_gre_keymap_add(ct, IP_CT_DIR_ORIGINAL, &exp_orig->tuple) != 0) + if (!nf_ct_gre_keymap_add(ct, &exp_orig->tuple, + &exp_reply->tuple)) goto out_unexpect_both; - if (nf_ct_gre_keymap_add(ct, IP_CT_DIR_REPLY, &exp_reply->tuple) != 0) { - nf_ct_gre_keymap_destroy(ct); - goto out_unexpect_both; - } ret = 0; out_put_both: diff --git a/net/netfilter/nf_conntrack_proto_gre.c b/net/netfilter/nf_conntrack_proto_gre.c index 94c19bc4edc5..35e22082d65a 100644 --- a/net/netfilter/nf_conntrack_proto_gre.c +++ b/net/netfilter/nf_conntrack_proto_gre.c @@ -87,41 +87,97 @@ static __be16 gre_keymap_lookup(struct net *net, struct nf_conntrack_tuple *t) return key; } -/* add a single keymap entry, associate with specified master ct */ -int nf_ct_gre_keymap_add(struct nf_conn *ct, enum ip_conntrack_dir dir, - struct nf_conntrack_tuple *t) +enum nf_ct_gre_km_act { + NF_CT_GRE_KM_NEW, + NF_CT_GRE_KM_BAD, + NF_CT_GRE_KM_DUP +}; + +static enum nf_ct_gre_km_act +nf_ct_gre_km_acceptable(const struct nf_ct_pptp_master *ct_pptp_info, + const struct nf_conntrack_tuple *orig, + const struct nf_conntrack_tuple *repl) +{ + struct nf_ct_gre_keymap *km_orig, *km_repl; + + lockdep_assert_held(&keymap_lock); + + km_orig = ct_pptp_info->keymap[IP_CT_DIR_ORIGINAL]; + km_repl = ct_pptp_info->keymap[IP_CT_DIR_REPLY]; + + if (km_orig && km_repl) { + if (!gre_key_cmpfn(km_orig, orig)) + return NF_CT_GRE_KM_BAD; + + if (!gre_key_cmpfn(km_repl, repl)) + return NF_CT_GRE_KM_BAD; + + return NF_CT_GRE_KM_DUP; + } + + DEBUG_NET_WARN_ON_ONCE(km_orig); + DEBUG_NET_WARN_ON_ONCE(km_repl); + return NF_CT_GRE_KM_NEW; +} + +/* add keymap entries, associate with specified master ct */ +bool nf_ct_gre_keymap_add(struct nf_conn *ct, + const struct nf_conntrack_tuple *orig, + const struct nf_conntrack_tuple *repl) { struct net *net = nf_ct_net(ct); struct nf_gre_net *net_gre = gre_pernet(net); struct nf_ct_pptp_master *ct_pptp_info = nfct_help_data(ct); - struct nf_ct_gre_keymap **kmp, *km; - - kmp = &ct_pptp_info->keymap[dir]; - if (*kmp) { - /* check whether it's a retransmission */ - list_for_each_entry_rcu(km, &net_gre->keymap_list, list) { - if (gre_key_cmpfn(km, t) && km == *kmp) - return 0; - } - pr_debug("trying to override keymap_%s for ct %p\n", - dir == IP_CT_DIR_REPLY ? "reply" : "orig", ct); - return -EEXIST; - } + struct nf_ct_gre_keymap *km_orig, *km_repl; + bool ret = false; - km = kmalloc_obj(*km, GFP_ATOMIC); - if (!km) - return -ENOMEM; - memcpy(&km->tuple, t, sizeof(*t)); - *kmp = km; + km_orig = kmalloc_obj(*km_orig, GFP_ATOMIC); + if (!km_orig) + return false; + km_repl = kmalloc_obj(*km_repl, GFP_ATOMIC); + if (!km_repl) + goto km_free; - pr_debug("adding new entry %p: ", km); - nf_ct_dump_tuple(&km->tuple); + memcpy(&km_orig->tuple, orig, sizeof(*orig)); + memcpy(&km_repl->tuple, repl, sizeof(*repl)); spin_lock_bh(&keymap_lock); - list_add_tail(&km->list, &net_gre->keymap_list); + if (nf_ct_is_dying(ct)) + goto unlock_free; + + switch (nf_ct_gre_km_acceptable(ct_pptp_info, orig, repl)) { + case NF_CT_GRE_KM_NEW: + break; + case NF_CT_GRE_KM_DUP: + ret = true; + goto unlock_free; + case NF_CT_GRE_KM_BAD: + pr_debug("trying to override keymap for ct %p\n", ct); + goto unlock_free; + } + + if (ct_pptp_info->keymap[IP_CT_DIR_ORIGINAL] || + ct_pptp_info->keymap[IP_CT_DIR_REPLY]) + goto unlock_free; + + pr_debug("adding new entries %p,%p: ", km_orig, km_repl); + nf_ct_dump_tuple(&km_orig->tuple); + nf_ct_dump_tuple(&km_repl->tuple); + + list_add_tail_rcu(&km_orig->list, &net_gre->keymap_list); + list_add_tail_rcu(&km_repl->list, &net_gre->keymap_list); + ct_pptp_info->keymap[IP_CT_DIR_ORIGINAL] = km_orig; + ct_pptp_info->keymap[IP_CT_DIR_REPLY] = km_repl; spin_unlock_bh(&keymap_lock); - return 0; + return true; + +unlock_free: + spin_unlock_bh(&keymap_lock); +km_free: + kfree(km_orig); + kfree(km_repl); + return ret; } EXPORT_SYMBOL_GPL(nf_ct_gre_keymap_add); -- cgit v1.2.3 From c376f07e16c02239ed44cabb97145d03f65b4d15 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Tue, 19 May 2026 20:10:08 +0200 Subject: netfilter: xt_cpu: prefer raw_smp_processor_id With PREEMPT_RCU we get splat: BUG: using smp_processor_id() in preemptible [..] caller is cpu_mt+0x53/0xd0 net/netfilter/xt_cpu.c:37 CPU: 1 .. Comm: syz.3.1377 #0 PREEMPT(full) Call Trace: dump_stack_lvl+0xe8/0x150 lib/dump_stack.c:120 check_preemption_disabled+0xd3/0xe0 lib/smp_processor_id.c:47 cpu_mt+0x53/0xd0 net/netfilter/xt_cpu.c:37 [..] Just use raw version instead. This is similar to 14d14a5d2957 ("netfilter: nft_meta: use raw_smp_processor_id()"). Fixes: 0ca743a55991 ("netfilter: nf_tables: add compatibility layer for x_tables") Reported-by: syzbot+690d3e3ffa7335ac10eb@syzkaller.appspotmail.com Signed-off-by: Florian Westphal --- net/netfilter/xt_cpu.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/netfilter/xt_cpu.c b/net/netfilter/xt_cpu.c index 3bdc302a0f91..9cb259902a58 100644 --- a/net/netfilter/xt_cpu.c +++ b/net/netfilter/xt_cpu.c @@ -34,7 +34,7 @@ static bool cpu_mt(const struct sk_buff *skb, struct xt_action_param *par) { const struct xt_cpu_info *info = par->matchinfo; - return (info->cpu == smp_processor_id()) ^ info->invert; + return (info->cpu == raw_smp_processor_id()) ^ info->invert; } static struct xt_match cpu_mt_reg __read_mostly = { -- cgit v1.2.3 From 968cc2c96390f06e56ed6a43f935bfebdefed28f Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Sat, 16 May 2026 23:23:21 +0800 Subject: netfilter: disable payload mangling in userns Several parts of network stack rely on iph->ihl validation done by network stack before PRE_ROUTING. Disable this feature for user namespaces for now. tcp option handling is likely safe even for LOCAL_IN, so this this leaves tcp option mangling via nft_exthdr.c as-is. I don't think these are the only means to alter packets, but these appear to be relatively prominent. This could be relaxed later. Example: - allow userns for ingress hook. - allow userns if base is transport header. Also, we should revalidate or restrict generally: - Don't allow linklayer writes to spill into network header - restrict ipv4 and ipv6 to 'known safe' writes, e.g. saddr/daddr/check/tos Reported-by: Qi Tang Reported-by: Tong Liu Tested-by: Qi Tang Link: https://lore.kernel.org/netfilter-devel/20260515100411.3141-1-fw@strlen.de/ Signed-off-by: Florian Westphal --- net/netfilter/nfnetlink_queue.c | 6 ++++-- net/netfilter/nft_payload.c | 3 +++ 2 files changed, 7 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/netfilter/nfnetlink_queue.c b/net/netfilter/nfnetlink_queue.c index 984a0eb9e149..60ab88d45096 100644 --- a/net/netfilter/nfnetlink_queue.c +++ b/net/netfilter/nfnetlink_queue.c @@ -1141,6 +1141,9 @@ nfqnl_mangle(void *data, unsigned int data_len, struct nf_queue_entry *e, int di { struct sk_buff *nskb; + if (e->state.net->user_ns != &init_user_ns) + return -EPERM; + if (diff < 0) { unsigned int min_len = skb_transport_offset(e->skb); @@ -1537,8 +1540,7 @@ static int nfqnl_recv_verdict(struct sk_buff *skb, const struct nfnl_info *info, if (nfqnl_mangle(nla_data(nfqa[NFQA_PAYLOAD]), payload_len, entry, diff) < 0) verdict = NF_DROP; - - if (ct && diff) + else if (ct && diff) nfnl_ct->seq_adjust(entry->skb, ct, ctinfo, diff); } diff --git a/net/netfilter/nft_payload.c b/net/netfilter/nft_payload.c index 01e13e5255a9..484a5490832e 100644 --- a/net/netfilter/nft_payload.c +++ b/net/netfilter/nft_payload.c @@ -917,6 +917,9 @@ static int nft_payload_set_init(const struct nft_ctx *ctx, struct nft_payload_set *priv = nft_expr_priv(expr); int err; + if (ctx->net->user_ns != &init_user_ns) + return -EPERM; + priv->base = ntohl(nla_get_be32(tb[NFTA_PAYLOAD_BASE])); priv->len = ntohl(nla_get_be32(tb[NFTA_PAYLOAD_LEN])); -- cgit v1.2.3 From f438d1786d657d57790c5d138d6db3fc9fdac392 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Tue, 19 May 2026 22:52:07 +0200 Subject: netfilter: ebtables: fix OOB read in compat_mtw_from_user Luxiao Xu says: The function compat_mtw_from_user() converts ebtables extensions from 32-bit user structures to kernel native structures. However, it lacks proper validation of the user-supplied match_size/target_size. When certain extensions are processed, the kernel-side translation logic may perform memory accesses based on the extension's expected size. If the user provides a size smaller than what the extension requires, it results in an out-of-bounds read as reported by KASAN. This fix introduces a check to ensure match_size is at least as large as the extension's required compatsize. This covers matches, watchers, and targets, while maintaining compatibility with standard targets. AFAIU this is relevant for matches that need to go though match->compat_from_user() call. Those that use plain memcpy with the user-provided size are ok because the caller checks that size vs the start of the next rule entry offset (which itself is checked vs. total size copied from userspace). The ->compat_from_user() callbacks assume they can read compatsize bytes, so they need this extra check. Based on an earlier patch from Luxiao Xu. Fixes: 81e675c227ec ("netfilter: ebtables: add CONFIG_COMPAT support") Reported-by: Yuan Tan Reported-by: Yifan Wu Reported-by: Juefei Pu Reported-by: Xin Liu Signed-off-by: Luxiao Xu Signed-off-by: Ren Wei Reviewed-by: Fernando Fernandez Mancera Signed-off-by: Florian Westphal --- net/bridge/netfilter/ebtables.c | 30 ++++++++++++++++++++++++++++++ 1 file changed, 30 insertions(+) (limited to 'net') diff --git a/net/bridge/netfilter/ebtables.c b/net/bridge/netfilter/ebtables.c index b9f4daac09af..8a6a069329d2 100644 --- a/net/bridge/netfilter/ebtables.c +++ b/net/bridge/netfilter/ebtables.c @@ -1956,6 +1956,25 @@ enum compat_mwt { EBT_COMPAT_TARGET, }; +static bool match_size_ok(const struct xt_match *match, unsigned int match_size) +{ + u16 csize; + + if (match->matchsize == -1) /* cannot validate ebt_among */ + return true; + + csize = match->compatsize ? : match->matchsize; + + return match_size >= csize; +} + +static bool tgt_size_ok(const struct xt_target *tgt, unsigned int tgt_size) +{ + u16 csize = tgt->compatsize ? : tgt->targetsize; + + return tgt_size >= csize; +} + static int compat_mtw_from_user(const struct compat_ebt_entry_mwt *mwt, enum compat_mwt compat_mwt, struct ebt_entries_buf_state *state, @@ -1981,6 +2000,11 @@ static int compat_mtw_from_user(const struct compat_ebt_entry_mwt *mwt, if (IS_ERR(match)) return PTR_ERR(match); + if (!match_size_ok(match, match_size)) { + module_put(match->me); + return -EINVAL; + } + off = ebt_compat_match_offset(match, match_size); if (dst) { if (match->compat_from_user) @@ -2000,6 +2024,12 @@ static int compat_mtw_from_user(const struct compat_ebt_entry_mwt *mwt, mwt->u.revision); if (IS_ERR(wt)) return PTR_ERR(wt); + + if (!tgt_size_ok(wt, match_size)) { + module_put(wt->me); + return -EINVAL; + } + off = xt_compat_target_offset(wt); if (dst) { -- cgit v1.2.3 From 1d001b0a6182b0d2f41a8d687f7522b6f1e94280 Mon Sep 17 00:00:00 2001 From: Jiayuan Chen Date: Wed, 20 May 2026 10:34:09 +0800 Subject: netfilter: nft_fib_ipv6: walk fib6_siblings under RCU nft_fib6_info_nh_uses_dev() runs from nft_fib6_eval() in softirq under rcu_read_lock(). fib6_siblings is modified by writers that hold tb6_lock but do not wait for RCU readers, so the sibling walk should use list_for_each_entry_rcu(): it adds READ_ONCE() on the ->next pointer and lets CONFIG_PROVE_RCU_LIST validate the locking. No functional change for non-debug builds. Fixes: 1c32b24c234b ("netfilter: nft_fib_ipv6: switch to fib6_lookup") Signed-off-by: Jiayuan Chen Signed-off-by: Florian Westphal --- net/ipv6/netfilter/nft_fib_ipv6.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv6/netfilter/nft_fib_ipv6.c b/net/ipv6/netfilter/nft_fib_ipv6.c index 8b2dba88ee96..5e192a446ec8 100644 --- a/net/ipv6/netfilter/nft_fib_ipv6.c +++ b/net/ipv6/netfilter/nft_fib_ipv6.c @@ -170,7 +170,7 @@ static bool nft_fib6_info_nh_uses_dev(struct fib6_info *rt, if (nft_fib6_info_nh_dev_match(nh_dev, dev)) return true; - list_for_each_entry(iter, &rt->fib6_siblings, fib6_siblings) { + list_for_each_entry_rcu(iter, &rt->fib6_siblings, fib6_siblings) { nh_dev = fib6_info_nh_dev(iter); if (nft_fib6_info_nh_dev_match(nh_dev, dev)) -- cgit v1.2.3 From f81b0c2d281faa93e4c2b7247047922aaf3e4ba6 Mon Sep 17 00:00:00 2001 From: Jiayuan Chen Date: Wed, 20 May 2026 10:34:10 +0800 Subject: netfilter: nft_fib_ipv6: handle routes via external nexthop fib6_info has a union: union { struct list_head fib6_siblings; struct list_head nh_list; }; Old-style multipath (ip -6 route add ... nexthop ... nexthop ...) uses fib6_siblings. External nexthop (ip -6 route add ... nhid N) uses nh_list, linked into &nh->f6i_list. nft_fib6_info_nh_uses_dev() blindly walks &rt->fib6_siblings, causing an OOB read past the struct nexthop slab when rt->nh is set: ================================================================== BUG: KASAN: slab-out-of-bounds in nft_fib6_eval+0x1362/0x16c0 Read of size 8 at addr ffff888103a099d0 by task ping/386 CPU: 2 UID: 0 PID: 386 Comm: ping Not tainted 7.1.0-rc3+ #251 PREEMPT Call Trace: dump_stack_lvl+0x76/0xa0 print_report+0xd1/0x5f0 kasan_report+0xe7/0x130 __asan_report_load8_noabort+0x14/0x30 nft_fib6_eval+0x1362/0x16c0 nft_do_chain+0x279/0x18c0 nft_do_chain_ipv6+0x1a8/0x230 nf_hook_slow+0xad/0x200 ipv6_rcv+0x152/0x380 __netif_receive_skb_one_core+0x118/0x1c0 ================================================================== Branch by route shape: when rt->nh is set, walk via nexthop_for_each_fib6_nh() (also covers nh groups, which the original code missed); otherwise walk fib6_siblings, guarded by READ_ONCE() of rt->fib6_nsiblings as required by commit 31d7d67ba127 ("ipv6: annotate data-races around rt->fib6_nsiblings"). Fixes: 1c32b24c234b ("netfilter: nft_fib_ipv6: switch to fib6_lookup") Signed-off-by: Jiayuan Chen Signed-off-by: Florian Westphal --- net/ipv6/netfilter/nft_fib_ipv6.c | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) (limited to 'net') diff --git a/net/ipv6/netfilter/nft_fib_ipv6.c b/net/ipv6/netfilter/nft_fib_ipv6.c index 5e192a446ec8..c0a0075e2590 100644 --- a/net/ipv6/netfilter/nft_fib_ipv6.c +++ b/net/ipv6/netfilter/nft_fib_ipv6.c @@ -160,16 +160,32 @@ static bool nft_fib6_info_nh_dev_match(const struct net_device *nh_dev, l3mdev_master_ifindex_rcu(nh_dev) == dev->ifindex; } +static int nft_fib6_nh_match_dev_cb(struct fib6_nh *nh, void *arg) +{ + const struct net_device *dev = arg; + + return nft_fib6_info_nh_dev_match(nh->fib_nh_dev, dev); +} + static bool nft_fib6_info_nh_uses_dev(struct fib6_info *rt, const struct net_device *dev) { const struct net_device *nh_dev; struct fib6_info *iter; + /* External nexthop: fib6_siblings slot aliases nh_list, walk via nh. */ + if (rt->nh) + return nexthop_for_each_fib6_nh(rt->nh, + nft_fib6_nh_match_dev_cb, + (void *)dev); + nh_dev = fib6_info_nh_dev(rt); if (nft_fib6_info_nh_dev_match(nh_dev, dev)) return true; + if (!READ_ONCE(rt->fib6_nsiblings)) + return false; + list_for_each_entry_rcu(iter, &rt->fib6_siblings, fib6_siblings) { nh_dev = fib6_info_nh_dev(iter); -- cgit v1.2.3 From 18014147d3ee7831dce53fe65d7fc8d428b02552 Mon Sep 17 00:00:00 2001 From: Fernando Fernandez Mancera Date: Mon, 11 May 2026 16:37:56 +0200 Subject: netfilter: nf_tables: fix dst corruption in same register operation For lshift and rshift, the shift operations are performed in a loop over 32-bit words. The loop calculates the shifted value and write it to dst, and then immediately reads from src to calculate the carry for the next iteration. Because src and dst could point to the same memory location, the carry is incorrectly calculated using the newly modified dst value instead of the original src value. Adding a temporary local variable to cache the original value before writing to dst and using it for the carry calculation solves the problem. In addition, partial overlap is rejected from control plane for all kind of operations including byteorder. This was tested with the following bytecode: table test_table ip flags 0 use 1 handle 1 ip test_table test_chain use 3 type filter hook input prio 0 policy accept packets 0 bytes 0 flags 1 ip test_table test_chain 2 [ immediate reg 1 0x44332211 0x88776655 ] [ bitwise reg 1 = ( reg 1 << 0x08000000 ) ] [ cmp eq reg 1 0x66443322 0x00887766 ] [ counter pkts 0 bytes 0 ] ip test_table test_chain 4 3 [ immediate reg 1 0x44332211 0x88776655 ] [ bitwise reg 1 = ( reg 1 << 0x08000000 ) ] [ cmp eq reg 1 0x55443322 0x00887766 ] [ counter pkts 21794 bytes 1917798 ] Fixes: 567d746b55bc ("netfilter: bitwise: add support for shifts.") Acked-by: Jeremy Sowden Signed-off-by: Fernando Fernandez Mancera Signed-off-by: Florian Westphal --- net/netfilter/nft_bitwise.c | 18 ++++++++++++++---- net/netfilter/nft_byteorder.c | 13 ++++++++++--- 2 files changed, 24 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/net/netfilter/nft_bitwise.c b/net/netfilter/nft_bitwise.c index 94dccdcfa06b..785b8e9731d1 100644 --- a/net/netfilter/nft_bitwise.c +++ b/net/netfilter/nft_bitwise.c @@ -43,8 +43,10 @@ static void nft_bitwise_eval_lshift(u32 *dst, const u32 *src, u32 carry = 0; for (i = DIV_ROUND_UP(priv->len, sizeof(u32)); i > 0; i--) { - dst[i - 1] = (src[i - 1] << shift) | carry; - carry = src[i - 1] >> (BITS_PER_TYPE(u32) - shift); + u32 tmp_src = src[i - 1]; + + dst[i - 1] = (tmp_src << shift) | carry; + carry = tmp_src >> (BITS_PER_TYPE(u32) - shift); } } @@ -56,8 +58,10 @@ static void nft_bitwise_eval_rshift(u32 *dst, const u32 *src, u32 carry = 0; for (i = 0; i < DIV_ROUND_UP(priv->len, sizeof(u32)); i++) { - dst[i] = carry | (src[i] >> shift); - carry = src[i] << (BITS_PER_TYPE(u32) - shift); + u32 tmp_src = src[i]; + + dst[i] = carry | (tmp_src >> shift); + carry = tmp_src << (BITS_PER_TYPE(u32) - shift); } } @@ -235,6 +239,9 @@ static int nft_bitwise_init_bool(const struct nft_ctx *ctx, &priv->sreg2, priv->len); if (err < 0) return err; + + if (nft_reg_overlap(priv->sreg2, priv->dreg, priv->len)) + return -EINVAL; } return 0; @@ -265,6 +272,9 @@ static int nft_bitwise_init(const struct nft_ctx *ctx, if (err < 0) return err; + if (nft_reg_overlap(priv->sreg, priv->dreg, priv->len)) + return -EINVAL; + if (tb[NFTA_BITWISE_OP]) { priv->op = ntohl(nla_get_be32(tb[NFTA_BITWISE_OP])); switch (priv->op) { diff --git a/net/netfilter/nft_byteorder.c b/net/netfilter/nft_byteorder.c index e00dddfa2fc0..2316c77f4228 100644 --- a/net/netfilter/nft_byteorder.c +++ b/net/netfilter/nft_byteorder.c @@ -144,9 +144,16 @@ static int nft_byteorder_init(const struct nft_ctx *ctx, if (err < 0) return err; - return nft_parse_register_store(ctx, tb[NFTA_BYTEORDER_DREG], - &priv->dreg, NULL, NFT_DATA_VALUE, - priv->len); + err = nft_parse_register_store(ctx, tb[NFTA_BYTEORDER_DREG], + &priv->dreg, NULL, NFT_DATA_VALUE, + priv->len); + if (err < 0) + return err; + + if (nft_reg_overlap(priv->sreg, priv->dreg, priv->len)) + return -EINVAL; + + return 0; } static int nft_byteorder_dump(struct sk_buff *skb, -- cgit v1.2.3 From aae9d8a5528b8ee9ff8dc5d3558b8a9f852a724a Mon Sep 17 00:00:00 2001 From: Ziyu Zhang Date: Wed, 20 May 2026 00:56:36 +0800 Subject: vsock: keep poll shutdown state consistent vsock_poll() reads vsk->peer_shutdown before taking the socket lock to set EPOLLHUP and EPOLLRDHUP, then reads it again after taking the lock to report EOF readability. A shutdown packet can update peer_shutdown while poll is waiting for the lock, so one poll invocation can report EOF readability without the corresponding HUP/RDHUP bits. For connectible sockets, take one peer_shutdown snapshot after lock_sock() and use it for all peer-shutdown-derived poll bits. For datagram sockets, which do not take lock_sock() in poll(), take one lockless READ_ONCE() snapshot and pair it with WRITE_ONCE() on the writer side. This keeps the peer-shutdown-derived bits internally consistent for each poll pass. Fixes: d021c344051a ("VSOCK: Introduce VM Sockets") Signed-off-by: Ziyu Zhang Link: https://patch.msgid.link/20260519165636.62542-1-ziyuzhang201@gmail.com Signed-off-by: Jakub Kicinski --- net/vmw_vsock/af_vsock.c | 49 +++++++++++++++++++++------------ net/vmw_vsock/hyperv_transport.c | 9 ++++-- net/vmw_vsock/virtio_transport_common.c | 14 ++++++---- net/vmw_vsock/vmci_transport.c | 8 ++++-- 4 files changed, 52 insertions(+), 28 deletions(-) (limited to 'net') diff --git a/net/vmw_vsock/af_vsock.c b/net/vmw_vsock/af_vsock.c index 44037b066a5f..2ce1063d4a67 100644 --- a/net/vmw_vsock/af_vsock.c +++ b/net/vmw_vsock/af_vsock.c @@ -642,7 +642,7 @@ int vsock_assign_transport(struct vsock_sock *vsk, struct vsock_sock *psk) */ sock_reset_flag(sk, SOCK_DONE); sk->sk_state = TCP_CLOSE; - vsk->peer_shutdown = 0; + WRITE_ONCE(vsk->peer_shutdown, 0); } if (sk->sk_type == SOCK_SEQPACKET) { @@ -933,7 +933,7 @@ static struct sock *__vsock_create(struct net *net, vsk->rejected = false; vsk->sent_request = false; vsk->ignore_connecting_rst = false; - vsk->peer_shutdown = 0; + WRITE_ONCE(vsk->peer_shutdown, 0); INIT_DELAYED_WORK(&vsk->connect_work, vsock_connect_timeout); INIT_DELAYED_WORK(&vsk->pending_work, vsock_pending_work); @@ -1241,6 +1241,25 @@ out: return err; } +static __poll_t vsock_poll_shutdown(struct sock *sk, u32 peer_shutdown) +{ + __poll_t mask = 0; + + /* INET sockets treat local write shutdown and peer write shutdown as a + * case of EPOLLHUP set. + */ + if (sk->sk_shutdown == SHUTDOWN_MASK || + ((sk->sk_shutdown & SEND_SHUTDOWN) && + (peer_shutdown & SEND_SHUTDOWN))) + mask |= EPOLLHUP; + + if (sk->sk_shutdown & RCV_SHUTDOWN || + peer_shutdown & SEND_SHUTDOWN) + mask |= EPOLLRDHUP; + + return mask; +} + static __poll_t vsock_poll(struct file *file, struct socket *sock, poll_table *wait) { @@ -1258,24 +1277,17 @@ static __poll_t vsock_poll(struct file *file, struct socket *sock, /* Signify that there has been an error on this socket. */ mask |= EPOLLERR; - /* INET sockets treat local write shutdown and peer write shutdown as a - * case of EPOLLHUP set. - */ - if ((sk->sk_shutdown == SHUTDOWN_MASK) || - ((sk->sk_shutdown & SEND_SHUTDOWN) && - (vsk->peer_shutdown & SEND_SHUTDOWN))) { - mask |= EPOLLHUP; - } - - if (sk->sk_shutdown & RCV_SHUTDOWN || - vsk->peer_shutdown & SEND_SHUTDOWN) { - mask |= EPOLLRDHUP; - } - if (sk_is_readable(sk)) mask |= EPOLLIN | EPOLLRDNORM; if (sock->type == SOCK_DGRAM) { + u32 peer_shutdown = READ_ONCE(vsk->peer_shutdown); + + /* DGRAM sockets do not take lock_sock() in poll(), so use one + * lockless snapshot for all shutdown-derived mask bits. + */ + mask |= vsock_poll_shutdown(sk, peer_shutdown); + /* For datagram sockets we can read if there is something in * the queue and write as long as the socket isn't shutdown for * sending. @@ -1290,6 +1302,7 @@ static __poll_t vsock_poll(struct file *file, struct socket *sock, } else if (sock_type_connectible(sk->sk_type)) { const struct vsock_transport *transport; + u32 peer_shutdown; lock_sock(sk); @@ -1322,8 +1335,10 @@ static __poll_t vsock_poll(struct file *file, struct socket *sock, * terminated should also be considered read, and we check the * shutdown flag for that. */ + peer_shutdown = READ_ONCE(vsk->peer_shutdown); + mask |= vsock_poll_shutdown(sk, peer_shutdown); if (sk->sk_shutdown & RCV_SHUTDOWN || - vsk->peer_shutdown & SEND_SHUTDOWN) { + peer_shutdown & SEND_SHUTDOWN) { mask |= EPOLLIN | EPOLLRDNORM; } diff --git a/net/vmw_vsock/hyperv_transport.c b/net/vmw_vsock/hyperv_transport.c index 7a8963595bf9..b3394946b2ed 100644 --- a/net/vmw_vsock/hyperv_transport.c +++ b/net/vmw_vsock/hyperv_transport.c @@ -264,7 +264,7 @@ static void hvs_do_close_lock_held(struct vsock_sock *vsk, struct sock *sk = sk_vsock(vsk); sock_set_flag(sk, SOCK_DONE); - vsk->peer_shutdown = SHUTDOWN_MASK; + WRITE_ONCE(vsk->peer_shutdown, SHUTDOWN_MASK); if (vsock_stream_has_data(vsk) <= 0) sk->sk_state = TCP_CLOSING; sk->sk_state_change(sk); @@ -593,7 +593,9 @@ static int hvs_update_recv_data(struct hvsock *hvs) return -EIO; if (payload_len == 0) - hvs->vsk->peer_shutdown |= SEND_SHUTDOWN; + WRITE_ONCE(hvs->vsk->peer_shutdown, + READ_ONCE(hvs->vsk->peer_shutdown) | + SEND_SHUTDOWN); hvs->recv_data_len = payload_len; hvs->recv_data_off = 0; @@ -736,7 +738,8 @@ static s64 hvs_stream_has_data(struct vsock_sock *vsk) return ret; return hvs->recv_data_len; case 0: - vsk->peer_shutdown |= SEND_SHUTDOWN; + WRITE_ONCE(vsk->peer_shutdown, + READ_ONCE(vsk->peer_shutdown) | SEND_SHUTDOWN); ret = 0; break; default: /* -1 */ diff --git a/net/vmw_vsock/virtio_transport_common.c b/net/vmw_vsock/virtio_transport_common.c index df3b418e0392..d4d26fba9e37 100644 --- a/net/vmw_vsock/virtio_transport_common.c +++ b/net/vmw_vsock/virtio_transport_common.c @@ -1228,7 +1228,7 @@ static void virtio_transport_do_close(struct vsock_sock *vsk, struct sock *sk = sk_vsock(vsk); sock_set_flag(sk, SOCK_DONE); - vsk->peer_shutdown = SHUTDOWN_MASK; + WRITE_ONCE(vsk->peer_shutdown, SHUTDOWN_MASK); if (vsock_stream_has_data(vsk) <= 0) sk->sk_state = TCP_CLOSING; sk->sk_state_change(sk); @@ -1431,12 +1431,15 @@ virtio_transport_recv_connected(struct sock *sk, case VIRTIO_VSOCK_OP_CREDIT_UPDATE: sk->sk_write_space(sk); break; - case VIRTIO_VSOCK_OP_SHUTDOWN: + case VIRTIO_VSOCK_OP_SHUTDOWN: { + u32 peer_shutdown = READ_ONCE(vsk->peer_shutdown); + if (le32_to_cpu(hdr->flags) & VIRTIO_VSOCK_SHUTDOWN_RCV) - vsk->peer_shutdown |= RCV_SHUTDOWN; + peer_shutdown |= RCV_SHUTDOWN; if (le32_to_cpu(hdr->flags) & VIRTIO_VSOCK_SHUTDOWN_SEND) - vsk->peer_shutdown |= SEND_SHUTDOWN; - if (vsk->peer_shutdown == SHUTDOWN_MASK) { + peer_shutdown |= SEND_SHUTDOWN; + WRITE_ONCE(vsk->peer_shutdown, peer_shutdown); + if (peer_shutdown == SHUTDOWN_MASK) { if (vsock_stream_has_data(vsk) <= 0 && !sock_flag(sk, SOCK_DONE)) { (void)virtio_transport_reset(vsk, NULL); virtio_transport_do_close(vsk, true); @@ -1451,6 +1454,7 @@ virtio_transport_recv_connected(struct sock *sk, if (le32_to_cpu(virtio_vsock_hdr(skb)->flags)) sk->sk_state_change(sk); break; + } case VIRTIO_VSOCK_OP_RST: virtio_transport_do_close(vsk, true); break; diff --git a/net/vmw_vsock/vmci_transport.c b/net/vmw_vsock/vmci_transport.c index d2579380f51e..5c1ecd5bfdbc 100644 --- a/net/vmw_vsock/vmci_transport.c +++ b/net/vmw_vsock/vmci_transport.c @@ -819,7 +819,7 @@ static void vmci_transport_handle_detach(struct sock *sk) /* On a detach the peer will not be sending or receiving * anymore. */ - vsk->peer_shutdown = SHUTDOWN_MASK; + WRITE_ONCE(vsk->peer_shutdown, SHUTDOWN_MASK); /* We should not be sending anymore since the peer won't be * there to receive, but we can still receive if there is data @@ -1542,7 +1542,9 @@ static int vmci_transport_recv_connected(struct sock *sk, if (pkt->u.mode) { vsk = vsock_sk(sk); - vsk->peer_shutdown |= pkt->u.mode; + WRITE_ONCE(vsk->peer_shutdown, + READ_ONCE(vsk->peer_shutdown) | + pkt->u.mode); sk->sk_state_change(sk); } break; @@ -1559,7 +1561,7 @@ static int vmci_transport_recv_connected(struct sock *sk, * a clean shutdown. */ sock_set_flag(sk, SOCK_DONE); - vsk->peer_shutdown = SHUTDOWN_MASK; + WRITE_ONCE(vsk->peer_shutdown, SHUTDOWN_MASK); if (vsock_stream_has_data(vsk) <= 0) sk->sk_state = TCP_CLOSING; -- cgit v1.2.3 From 70f8592ee90585272018a725054b6eb2ab7e99ca Mon Sep 17 00:00:00 2001 From: Ilya Maximets Date: Wed, 20 May 2026 19:22:35 +0200 Subject: net: netlink: fix sending unassigned nsid after assigned one If the current skb is not shared, it is re-used directly for all the sockets subscribed to the notification. If we have remote all-nsid socket receiving a message first, then the 'nsid_is_set' will be set to 'true'. If the nsid is NOT_ASSIGNED for the next socket in the list, the 'nsid_is_set' will remain 'true' and the negative value is be delivered to the user space. All subsequent nsid values will be delivered as well, since there is no code path that sets the flag back to 'false'. Fix that by always dropping the flag to 'false' first. Fixes: 7212462fa6fd ("netlink: don't send unknown nsid") Signed-off-by: Ilya Maximets Acked-by: Nicolas Dichtel Link: https://patch.msgid.link/20260520172317.175168-2-i.maximets@ovn.org Signed-off-by: Jakub Kicinski --- net/netlink/af_netlink.c | 1 + 1 file changed, 1 insertion(+) (limited to 'net') diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index 2aeb0680807d..0742e97f256e 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -1482,6 +1482,7 @@ static void do_one_broadcast(struct sock *sk, p->skb2 = NULL; goto out; } + NETLINK_CB(p->skb2).nsid_is_set = false; NETLINK_CB(p->skb2).nsid = peernet2id(sock_net(sk), p->net); if (NETLINK_CB(p->skb2).nsid != NETNSA_NSID_NOT_ASSIGNED) NETLINK_CB(p->skb2).nsid_is_set = true; -- cgit v1.2.3 From 88b126b39f9757e9debc322d4679239e9af089c7 Mon Sep 17 00:00:00 2001 From: Ilya Maximets Date: Wed, 20 May 2026 19:22:36 +0200 Subject: net: netlink: don't set nsid on local notifications In most cases, notifications on sockets with NETLINK_LISTEN_ALL_NSID do not contain NSID in their ancillary data in case the event is local to the listener. However, when a self-referential NSID is allocated for a namespace, every local notification starts sending this ID to the user space. This is problematic, because the listener cannot tell if those notifications are local or not anymore without making extra requests to figure out if the provided NSID is local or not. The listener can also not figure out the local NSID beforehand as it can be allocated at any point in time by other processes, changing the structure of the future notifications for everyone. The value is practically not useful, since it's the namespace's own ID that the application has to obtain from other sources in order to figure out if it's the same or not. So, for the application it's just an extra busy work with no benefits. Moreover, applications that do not know about this quirk may be mishandling notifications with NSID set as notifications from remote namespaces. This is the case for ovs-vswitchd and the iproute2's 'ip monitor' that stops printing 'current' and starts printing the nsid number mid-session. Lack of clear documentation for this behavior is also not helping. A search though open-source projects doesn't reveal any projects that use NETNSA_NSID_NOT_ASSIGNED and rely on metadata to contain self-referential NSIDs (expected, since the value is not useful). Quite the opposite, as already mentioned, there are few applications that rely on NSID to not be present in local events. Since the value is not useful and actively harmful in some cases, let's not report it for local events, making the notifications more consistent. Also adding some blank lines for readability. Fixes: 59324cf35aba ("netlink: allow to listen "all" netns") Reported-by: Matteo Perin Signed-off-by: Ilya Maximets Acked-by: Nicolas Dichtel Link: https://patch.msgid.link/20260520172317.175168-3-i.maximets@ovn.org Signed-off-by: Jakub Kicinski --- net/netlink/af_netlink.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index 0742e97f256e..7269e23b578d 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -1482,10 +1482,14 @@ static void do_one_broadcast(struct sock *sk, p->skb2 = NULL; goto out; } + NETLINK_CB(p->skb2).nsid_is_set = false; - NETLINK_CB(p->skb2).nsid = peernet2id(sock_net(sk), p->net); - if (NETLINK_CB(p->skb2).nsid != NETNSA_NSID_NOT_ASSIGNED) - NETLINK_CB(p->skb2).nsid_is_set = true; + if (!net_eq(sock_net(sk), p->net)) { + NETLINK_CB(p->skb2).nsid = peernet2id(sock_net(sk), p->net); + if (NETLINK_CB(p->skb2).nsid != NETNSA_NSID_NOT_ASSIGNED) + NETLINK_CB(p->skb2).nsid_is_set = true; + } + val = netlink_broadcast_deliver(sk, p->skb2); if (val < 0) { netlink_overrun(sk); -- cgit v1.2.3 From 9e4389b0038781f19f97895186ed941ff8ac1678 Mon Sep 17 00:00:00 2001 From: Alexandra Winter Date: Thu, 21 May 2026 16:56:39 +0200 Subject: net/smc: Do not re-initialize smc hashtables INIT_HLIST_HEAD(&smc_v*_hashinfo.ht) are called after smc_nl_init(), proto_register() and sock_register(). This can lead to smc_v*_hashinfo.ht being reset even though hash entries already exist and are being used, possibly resulting in a corrupted list. Remove unnecessary and dangerous re-initialisation of smc_v*_hashinfo.ht in smc_init(); it is implicitly initialised to zero anyhow. Add HLIST_HEAD_INIT to the definitions for clarity. Fixes: f16a7dd5cf27 ("smc: netlink interface for SMC sockets") Suggested-by: Halil Pasic Signed-off-by: Alexandra Winter Acked-by: Halil Pasic Reviewed-by: Mahanta Jambigi Link: https://patch.msgid.link/20260521145639.10317-1-wintera@linux.ibm.com Signed-off-by: Jakub Kicinski --- net/smc/af_smc.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/smc/af_smc.c b/net/smc/af_smc.c index dffbd529762d..b5db69073e20 100644 --- a/net/smc/af_smc.c +++ b/net/smc/af_smc.c @@ -188,10 +188,12 @@ static bool smc_hs_congested(const struct sock *sk) struct smc_hashinfo smc_v4_hashinfo = { .lock = __RW_LOCK_UNLOCKED(smc_v4_hashinfo.lock), + .ht = HLIST_HEAD_INIT, }; struct smc_hashinfo smc_v6_hashinfo = { .lock = __RW_LOCK_UNLOCKED(smc_v6_hashinfo.lock), + .ht = HLIST_HEAD_INIT, }; int smc_hash_sk(struct sock *sk) @@ -3517,8 +3519,6 @@ static int __init smc_init(void) pr_err("%s: sock_register fails with %d\n", __func__, rc); goto out_proto6; } - INIT_HLIST_HEAD(&smc_v4_hashinfo.ht); - INIT_HLIST_HEAD(&smc_v6_hashinfo.ht); rc = smc_ib_register_client(); if (rc) { -- cgit v1.2.3 From 3589d20a666caf30ad100c960a2de7de390fce88 Mon Sep 17 00:00:00 2001 From: Breno Leitao Date: Thu, 21 May 2026 07:11:45 -0700 Subject: net/iucv: fix locking in .getsockopt Mirror iucv_sock_setsockopt() and wrap the whole switch in lock_sock()/release_sock(). The pre-existing SO_MSGLIMIT-only lock becomes redundant and is removed. Any AF_IUCV HIPER user can potentially crash the kernel by racing recvmsg() with getsockopt(SO_MSGSIZE): the SO_MSGSIZE arm dereferences iucv->hs_dev->mtu after iucv_sock_close() (called from the racing recvmsg()) has set hs_dev to NULL, producing a NULL pointer dereference oops. Suggested-by: Stanislav Fomichev Fixes: 51363b8751a6 ("af_iucv: allow retrieval of maximum message size") Signed-off-by: Breno Leitao Reviewed-by: Alexandra Winter Tested-by: Alexandra Winter Link: https://patch.msgid.link/20260521-af_iucv_fix2-v1-1-f16b1c510aa9@debian.org Signed-off-by: Jakub Kicinski --- net/iucv/af_iucv.c | 20 ++++++++++++++------ 1 file changed, 14 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/net/iucv/af_iucv.c b/net/iucv/af_iucv.c index 72dfccd4e3d5..c2dc3338670e 100644 --- a/net/iucv/af_iucv.c +++ b/net/iucv/af_iucv.c @@ -1540,7 +1540,7 @@ static int iucv_sock_getsockopt(struct socket *sock, int level, int optname, struct sock *sk = sock->sk; struct iucv_sock *iucv = iucv_sk(sk); unsigned int val; - int len; + int len, rc; if (level != SOL_IUCV) return -ENOPROTOOPT; @@ -1553,26 +1553,34 @@ static int iucv_sock_getsockopt(struct socket *sock, int level, int optname, len = min_t(unsigned int, len, sizeof(int)); + rc = 0; + + lock_sock(sk); switch (optname) { case SO_IPRMDATA_MSG: val = (iucv->flags & IUCV_IPRMDATA) ? 1 : 0; break; case SO_MSGLIMIT: - lock_sock(sk); val = (iucv->path != NULL) ? iucv->path->msglim /* connected */ : iucv->msglimit; /* default */ - release_sock(sk); break; case SO_MSGSIZE: - if (sk->sk_state == IUCV_OPEN) - return -EBADFD; + if (sk->sk_state == IUCV_OPEN) { + rc = -EBADFD; + break; + } val = (iucv->hs_dev) ? iucv->hs_dev->mtu - sizeof(struct af_iucv_trans_hdr) - ETH_HLEN : 0x7fffffff; break; default: - return -ENOPROTOOPT; + rc = -ENOPROTOOPT; + break; } + release_sock(sk); + + if (rc) + return rc; if (put_user(len, optlen)) return -EFAULT; -- cgit v1.2.3 From 4157501b9a8ff1bbe32ff5a7d8aece7ab18eff40 Mon Sep 17 00:00:00 2001 From: Stefano Garzarella Date: Thu, 21 May 2026 14:47:32 +0200 Subject: vsock/virtio: fix skb overhead overflow on 32-bit builds On 32-bit architectures, both skb_queue_len() and SKB_TRUESIZE(0) evaluate to 32-bit values. The multiplication can overflow before being assigned to the u64 skb_overhead variable, making the skb overhead check ineffective. Cast skb_queue_len() to u64 so the multiplication is always performed in 64-bit arithmetic. This issue was reported by Sashiko while reviewing another patch. Fixes: 059b7dbd20a6 ("vsock/virtio: fix potential unbounded skb queue") Closes: https://sashiko.dev/#/patchset/20260518090656.134588-1-sgarzare%40redhat.com Cc: stable@vger.kernel.org Signed-off-by: Stefano Garzarella Acked-by: Michael S. Tsirkin Link: https://patch.msgid.link/20260521124732.125771-1-sgarzare@redhat.com Signed-off-by: Jakub Kicinski --- net/vmw_vsock/virtio_transport_common.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/vmw_vsock/virtio_transport_common.c b/net/vmw_vsock/virtio_transport_common.c index d4d26fba9e37..b143290a311d 100644 --- a/net/vmw_vsock/virtio_transport_common.c +++ b/net/vmw_vsock/virtio_transport_common.c @@ -417,7 +417,7 @@ static int virtio_transport_send_pkt_info(struct vsock_sock *vsk, static bool virtio_transport_inc_rx_pkt(struct virtio_vsock_sock *vvs, u32 len) { - u64 skb_overhead = (skb_queue_len(&vvs->rx_queue) + 1) * SKB_TRUESIZE(0); + u64 skb_overhead = ((u64)skb_queue_len(&vvs->rx_queue) + 1) * SKB_TRUESIZE(0); /* Allow at most buf_alloc * 2 total budget (payload + overhead), * similar to how SO_RCVBUF is doubled to reserve space for sk_buff -- cgit v1.2.3 From 87a1e0fe7776da7ab411be332b4be58ac8840d10 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 21 May 2026 12:21:47 +0000 Subject: ipv4: free net->ipv4.sysctl_local_reserved_ports after unregister_net_sysctl_table() ipv4_sysctl_exit_net() is currently freeing net->ipv4.sysctl_local_reserved_ports too soon. Only after unregister_net_sysctl_table() we can be sure no threads can possibly use the sysctls, including /proc/sys/net/ipv4/ip_local_reserved_ports. Fixes: 122ff243f5f1 ("ipv4: make ip_local_reserved_ports per netns") Reported-by: Ji'an Zhou Signed-off-by: Eric Dumazet Cc: Cong Wang Reviewed-by: Jason Xing Reviewed-by: Jiayuan Chen Link: https://patch.msgid.link/20260521122147.3584624-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/ipv4/sysctl_net_ipv4.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index d8bdb1bdbff1..c0e85cc171ae 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -1705,10 +1705,10 @@ static __net_exit void ipv4_sysctl_exit_net(struct net *net) { const struct ctl_table *table; - kfree(net->ipv4.sysctl_local_reserved_ports); table = net->ipv4.ipv4_hdr->ctl_table_arg; unregister_net_sysctl_table(net->ipv4.ipv4_hdr); kfree(table); + kfree(net->ipv4.sysctl_local_reserved_ports); } static __net_initdata struct pernet_operations ipv4_sysctl_ops = { -- cgit v1.2.3 From f7b52afe3592eae66e160586b45a3f2242972c63 Mon Sep 17 00:00:00 2001 From: Zhengchuan Liang Date: Fri, 22 May 2026 17:42:26 +0800 Subject: ipv6: exthdrs: refresh nh after handling HAO option ip6_parse_tlv() caches skb_network_header(skb) in nh while walking IPv6 TLVs. ipv6_dest_hao() may call pskb_expand_head() for a cloned skb, which can move the skb head and invalidate the cached network header pointer. Refresh nh after ipv6_dest_hao() returns so any trailing padding or TLVs are parsed from the current skb head. This matches the existing pattern used in ip6_parse_tlv() after helpers that can modify skb header storage. Fixes: a831f5bbc89a ("[IPV6] MIP6: Add inbound interface of home address option.") Cc: stable@kernel.org Reported-by: Yuan Tan Reported-by: Xin Liu Co-developed-by: Luxing Yin Signed-off-by: Luxing Yin Signed-off-by: Zhengchuan Liang Signed-off-by: Ren Wei Reviewed-by: Justin Iurman Reviewed-by: Ido Schimmel Link: https://patch.msgid.link/7aba1debc2196189172499e5769802b026f8caf8.1779247873.git.zcliangcn@gmail.com Signed-off-by: Jakub Kicinski --- net/ipv6/exthdrs.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'net') diff --git a/net/ipv6/exthdrs.c b/net/ipv6/exthdrs.c index cf90f933ca1a..6d92c02d0e3d 100644 --- a/net/ipv6/exthdrs.c +++ b/net/ipv6/exthdrs.c @@ -201,6 +201,8 @@ static bool ip6_parse_tlv(bool hopbyhop, case IPV6_TLV_HAO: if (!ipv6_dest_hao(skb, off)) return false; + + nh = skb_network_header(skb); break; #endif default: -- cgit v1.2.3 From d47548a36639095939f4747d4c43f2271366f565 Mon Sep 17 00:00:00 2001 From: Justin Iurman Date: Fri, 22 May 2026 13:20:13 +0200 Subject: ipv6: exthdrs: refresh nh pointer after ipv6_hop_jumbo() ipv6_hop_jumbo() calls pskb_trim_rcsum(), which can change skb pointers. Let's recompute nh pointer to make sure any change won't mess things up. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Cc: stable@vger.kernel.org Signed-off-by: Justin Iurman Reviewed-by: Ido Schimmel Link: https://patch.msgid.link/20260522112013.12342-1-justin.iurman@gmail.com Signed-off-by: Jakub Kicinski --- net/ipv6/exthdrs.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'net') diff --git a/net/ipv6/exthdrs.c b/net/ipv6/exthdrs.c index 6d92c02d0e3d..aca2a2abd2df 100644 --- a/net/ipv6/exthdrs.c +++ b/net/ipv6/exthdrs.c @@ -184,6 +184,8 @@ static bool ip6_parse_tlv(bool hopbyhop, case IPV6_TLV_JUMBO: if (!ipv6_hop_jumbo(skb, off)) return false; + + nh = skb_network_header(skb); break; case IPV6_TLV_CALIPSO: if (!ipv6_hop_calipso(skb, off)) -- cgit v1.2.3 From f229426072fc865654a60978bb7fda790a051ff3 Mon Sep 17 00:00:00 2001 From: Luka Gejak Date: Sat, 23 May 2026 15:03:30 +0200 Subject: net: hsr: fix potential OOB access in supervision frame handling Ensure the entire TLV header is linearized before access by adding sizeof(struct hsr_sup_tlv) to the pskb_may_pull() calls. Without this, a truncated frame could cause an out-of-bounds access. Fixes: eafaa88b3eb7 ("net: hsr: Add support for redbox supervision frames") Signed-off-by: Luka Gejak Reviewed-by: Fernando Fernandez Mancera Link: https://patch.msgid.link/20260523130330.61880-1-luka.gejak@linux.dev Signed-off-by: Jakub Kicinski --- net/hsr/hsr_forward.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/hsr/hsr_forward.c b/net/hsr/hsr_forward.c index 0aca859c88cb..f669a226d728 100644 --- a/net/hsr/hsr_forward.c +++ b/net/hsr/hsr_forward.c @@ -84,7 +84,7 @@ static bool is_supervision_frame(struct hsr_priv *hsr, struct sk_buff *skb) /* Get next tlv */ total_length += hsr_sup_tag->tlv.HSR_TLV_length; - if (!pskb_may_pull(skb, total_length)) + if (!pskb_may_pull(skb, total_length + sizeof(struct hsr_sup_tlv))) return false; skb_pull(skb, total_length); hsr_sup_tlv = (struct hsr_sup_tlv *)skb->data; @@ -100,7 +100,7 @@ static bool is_supervision_frame(struct hsr_priv *hsr, struct sk_buff *skb) /* make sure another tlv follows */ total_length += sizeof(struct hsr_sup_tlv) + hsr_sup_tlv->HSR_TLV_length; - if (!pskb_may_pull(skb, total_length)) + if (!pskb_may_pull(skb, total_length + sizeof(struct hsr_sup_tlv))) return false; /* get next tlv */ -- cgit v1.2.3 From 7f83d174073234839aea176f265e517e0d50a1d2 Mon Sep 17 00:00:00 2001 From: Shaomin Chen Date: Thu, 21 May 2026 02:07:23 +0800 Subject: xfrm: iptfs: reset runtime state when cloning SAs iptfs_clone_state() clones the IPTFS mode data with kmemdup(). This copies runtime objects which must not be shared with the original SA, including the embedded sk_buff_head, hrtimers, spinlock, and in-flight reassembly/reorder state. If xfrm_state_migrate() fails after clone_state() but before the later init_state() call has reinitialized those fields, the cloned state can be destroyed by xfrm_state_gc_task() with list and timer state copied from the original SA. With queued packets this lets the clone splice and free skbs owned by the original IPTFS queue, leading to use-after-free and double-free reports in iptfs_destroy_state() and skb release paths. Reinitialize the clone's runtime state before publishing it through x->mode_data. Because clone_state() now publishes a destroyable mode_data object before init_state(), take the mode callback module reference there. Avoid taking it again from __iptfs_init_state() for the same object. Fixes: 0e4fbf013fa5 ("xfrm: iptfs: add user packet (tunnel ingress) handling") Cc: stable@vger.kernel.org Signed-off-by: Shaomin Chen Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_iptfs.c | 28 +++++++++++++++++++++++----- 1 file changed, 23 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/xfrm/xfrm_iptfs.c b/net/xfrm/xfrm_iptfs.c index 97bc979e55ba..6c6bbc040517 100644 --- a/net/xfrm/xfrm_iptfs.c +++ b/net/xfrm/xfrm_iptfs.c @@ -2650,7 +2650,8 @@ static void __iptfs_init_state(struct xfrm_state *x, x->props.enc_hdr_len = sizeof(struct ip_iptfs_hdr); /* Always keep a module reference when x->mode_data is set */ - __module_get(x->mode_cbs->owner); + if (x->mode_data != xtfs) + __module_get(x->mode_cbs->owner); x->mode_data = xtfs; xtfs->x = x; @@ -2658,22 +2659,39 @@ static void __iptfs_init_state(struct xfrm_state *x, static int iptfs_clone_state(struct xfrm_state *x, struct xfrm_state *orig) { + struct skb_wseq *w_saved = NULL; struct xfrm_iptfs_data *xtfs; xtfs = kmemdup(orig->mode_data, sizeof(*xtfs), GFP_KERNEL); if (!xtfs) return -ENOMEM; - xtfs->ra_newskb = NULL; if (xtfs->cfg.reorder_win_size) { - xtfs->w_saved = kzalloc_objs(*xtfs->w_saved, - xtfs->cfg.reorder_win_size); - if (!xtfs->w_saved) { + w_saved = kzalloc_objs(*w_saved, xtfs->cfg.reorder_win_size); + if (!w_saved) { kfree_sensitive(xtfs); return -ENOMEM; } } + xtfs->w_saved = w_saved; + + __skb_queue_head_init(&xtfs->queue); + xtfs->queue_size = 0; + hrtimer_setup(&xtfs->iptfs_timer, iptfs_delay_timer, CLOCK_MONOTONIC, + IPTFS_HRTIMER_MODE); + + spin_lock_init(&xtfs->drop_lock); + hrtimer_setup(&xtfs->drop_timer, iptfs_drop_timer, CLOCK_MONOTONIC, + IPTFS_HRTIMER_MODE); + xtfs->w_seq_set = false; + xtfs->w_wantseq = 0; + xtfs->w_savedlen = 0; + xtfs->ra_newskb = NULL; + xtfs->ra_wantseq = 0; + xtfs->ra_runtlen = 0; + + __module_get(x->mode_cbs->owner); x->mode_data = xtfs; xtfs->x = x; -- cgit v1.2.3 From 3e52417318473782012b236d0325bf7d2266a597 Mon Sep 17 00:00:00 2001 From: Usama Arif Date: Thu, 21 May 2026 03:29:26 -0700 Subject: xfrm: move policy_bydst RCU sync from per-netns .exit to .pre_exit The struct pernet_operations docstring in include/net/net_namespace.h explicitly warns against blocking RCU primitives in .exit handlers: Exit methods using blocking RCU primitives, such as synchronize_rcu(), should be implemented via exit_batch. [...] Please, avoid synchronize_rcu() at all, where it's possible. Note that a combination of pre_exit() and exit() can be used, since a synchronize_rcu() is guaranteed between the calls. xfrm_policy_fini() violates this: it calls synchronize_rcu() before freeing the policy_bydst hash tables (so no RCU reader is mid- traversal at free time), but runs from xfrm_net_ops.exit -- once per namespace -- so a cleanup_net() of N namespaces pays N full RCU grace periods serially. Use the documented pre_exit/exit split. Move the policy flush (and the workqueue drains it depends on) into a new .pre_exit handler; xfrm_policy_fini() then runs in .exit and frees the hash tables after the synchronize_rcu_expedited() that cleanup_net() guarantees between the two phases. Providing O(1) RCU grace periods per batch instead of O(N). Observed on Linux 6.18 with a workload doing unshare(CLONE_NEWNET) at ~13/sec sustained: cleanup_net() and the netns_wq rescuer kthread both stuck in xfrm_policy_fini()'s synchronize_rcu(), >300k struct net accumulated in the cleanup queue, Percpu in /proc/meminfo climbed to 130+ GB on 256-CPU hosts, and memcg OOMs followed. setup_net and __put_net counts were balanced, ruling out a refcount leak. Fixes: 069daad4f2ae ("xfrm: Wait for RCU readers during policy netns exit") Signed-off-by: Usama Arif Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_policy.c | 15 ++++++++------- 1 file changed, 8 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index 59968dcbafe1..dd09d2063da2 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -4276,21 +4276,21 @@ out_byidx: return -ENOMEM; } -static void xfrm_policy_fini(struct net *net) +static void __net_exit xfrm_net_pre_exit(struct net *net) { - struct xfrm_pol_inexact_bin *b, *t; - unsigned int sz; - int dir; - disable_work_sync(&net->xfrm.policy_hthresh.work); - flush_work(&net->xfrm.policy_hash_work); #ifdef CONFIG_XFRM_SUB_POLICY xfrm_policy_flush(net, XFRM_POLICY_TYPE_SUB, false); #endif xfrm_policy_flush(net, XFRM_POLICY_TYPE_MAIN, false); +} - synchronize_rcu(); +static void xfrm_policy_fini(struct net *net) +{ + struct xfrm_pol_inexact_bin *b, *t; + unsigned int sz; + int dir; WARN_ON(!list_empty(&net->xfrm.policy_all)); @@ -4368,6 +4368,7 @@ static void __net_exit xfrm_net_exit(struct net *net) static struct pernet_operations __net_initdata xfrm_net_ops = { .init = xfrm_net_init, + .pre_exit = xfrm_net_pre_exit, .exit = xfrm_net_exit, }; -- cgit v1.2.3 From c16f74dc1d75d0e2e7670076d5375deda110ebeb Mon Sep 17 00:00:00 2001 From: Zhengchuan Liang Date: Fri, 22 May 2026 17:31:55 +0800 Subject: xfrm: input: hold netns during deferred transport reinjection Transport-mode reinjection stores a struct net pointer in skb->cb and uses it later from xfrm_trans_reinject(). That pointer must stay valid until the deferred callback runs. Take a netns reference when queueing deferred reinjection work and drop it after the callback completes. Use maybe_get_net() so the queueing path does not revive a namespace that is already being torn down. This keeps the existing workqueue design and fixes the netns lifetime handling in one place for all users of xfrm_trans_queue_net(). Fixes: 7b3801927e52 ("xfrm: introduce xfrm_trans_queue_net") Cc: stable@kernel.org Reported-by: Yuan Tan Reported-by: Xin Liu Co-developed-by: Luxing Yin Signed-off-by: Luxing Yin Signed-off-by: Zhengchuan Liang Signed-off-by: Ren Wei Assisted-by: Codex:gpt-5.4 Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_input.c | 16 ++++++++++++---- 1 file changed, 12 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/xfrm/xfrm_input.c b/net/xfrm/xfrm_input.c index f65291eba1f6..e4c2cd24936d 100644 --- a/net/xfrm/xfrm_input.c +++ b/net/xfrm/xfrm_input.c @@ -797,9 +797,12 @@ static void xfrm_trans_reinject(struct work_struct *work) spin_unlock_bh(&trans->queue_lock); local_bh_disable(); - while ((skb = __skb_dequeue(&queue))) - XFRM_TRANS_SKB_CB(skb)->finish(XFRM_TRANS_SKB_CB(skb)->net, - NULL, skb); + while ((skb = __skb_dequeue(&queue))) { + struct net *net = XFRM_TRANS_SKB_CB(skb)->net; + + XFRM_TRANS_SKB_CB(skb)->finish(net, NULL, skb); + put_net(net); + } local_bh_enable(); } @@ -808,6 +811,7 @@ int xfrm_trans_queue_net(struct net *net, struct sk_buff *skb, struct sk_buff *)) { struct xfrm_trans_tasklet *trans; + struct net *hold_net; trans = this_cpu_ptr(&xfrm_trans_tasklet); @@ -816,8 +820,12 @@ int xfrm_trans_queue_net(struct net *net, struct sk_buff *skb, BUILD_BUG_ON(sizeof(struct xfrm_trans_cb) > sizeof(skb->cb)); + hold_net = maybe_get_net(net); + if (!hold_net) + return -ENODEV; + XFRM_TRANS_SKB_CB(skb)->finish = finish; - XFRM_TRANS_SKB_CB(skb)->net = net; + XFRM_TRANS_SKB_CB(skb)->net = hold_net; spin_lock_bh(&trans->queue_lock); __skb_queue_tail(&trans->queue, skb); spin_unlock_bh(&trans->queue_lock); -- cgit v1.2.3 From 11b326fb0a374f4654f9be22d0f0f7abd9f7d3fe Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Thu, 21 May 2026 21:05:54 +0800 Subject: ip6: vti: Use ip6_tnl.net in vti6_changelink(). ip netns add ns1 ip netns add ns2 ip -n ns1 link add vti6_test type vti6 remote ::1 local ::2 key 7 ip -n ns1 link set vti6_test netns ns2 ip -n ns2 link set vti6_test type vti6 remote ::3 local ::4 key 9 ip netns del ns2 ip netns del ns1 [ 132.495484] ------------[ cut here ]------------ [ 132.497609] kernel BUG at net/core/dev.c:12376! Commit 61220ab34948 ("vti6: Enable namespace changing") dropped NETIF_F_NETNS_LOCAL from vti6 devices. A vti6 tunnel can then move through IFLA_NET_NS_FD. After the move dev_net(dev) points at the new netns while t->net stays at the creation netns. vti6_changelink() and vti6_update() still use dev_net(dev) and dev_net(t->dev). They unlink from one per netns hash and relink into another. The creation netns is left with a stale entry. cleanup_net() of that netns later walks freed memory. Reachable from an unprivileged user namespace (unshare --user --map-root-user --net). Cross tenant scope on container hosts. Fixes: 61220ab34948 ("vti6: Enable namespace changing") Reported-by: Maoyi Xie Reviewed-by: Eric Dumazet Cc: stable@vger.kernel.org # v5.15+ Signed-off-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20260521130555.3421684-2-maoyixie.tju@gmail.com Signed-off-by: Paolo Abeni --- net/ipv6/ip6_vti.c | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/ipv6/ip6_vti.c b/net/ipv6/ip6_vti.c index ad5290be4dd6..dcb257411d6e 100644 --- a/net/ipv6/ip6_vti.c +++ b/net/ipv6/ip6_vti.c @@ -722,10 +722,11 @@ vti6_tnl_change(struct ip6_tnl *t, const struct __ip6_tnl_parm *p, static int vti6_update(struct ip6_tnl *t, struct __ip6_tnl_parm *p, bool keep_mtu) { - struct net *net = dev_net(t->dev); - struct vti6_net *ip6n = net_generic(net, vti6_net_id); + struct net *net = t->net; + struct vti6_net *ip6n; int err; + ip6n = net_generic(net, vti6_net_id); vti6_tnl_unlink(ip6n, t); synchronize_net(); err = vti6_tnl_change(t, p, keep_mtu); @@ -1031,11 +1032,12 @@ static int vti6_changelink(struct net_device *dev, struct nlattr *tb[], struct nlattr *data[], struct netlink_ext_ack *extack) { - struct ip6_tnl *t; + struct ip6_tnl *t = netdev_priv(dev); + struct net *net = t->net; struct __ip6_tnl_parm p; - struct net *net = dev_net(dev); - struct vti6_net *ip6n = net_generic(net, vti6_net_id); + struct vti6_net *ip6n; + ip6n = net_generic(net, vti6_net_id); if (dev == ip6n->fb_tnl_dev) return -EINVAL; -- cgit v1.2.3 From 8b484efd5cb4eeef9021a661e198edc5349dacf6 Mon Sep 17 00:00:00 2001 From: Maoyi Xie Date: Thu, 21 May 2026 21:05:55 +0800 Subject: ip6: vti: Use ip6_tnl.net in vti6_siocdevprivate(). After patch 1/2 in this series, vti6_update() unlinks and relinks the tunnel through t->net. vti6_siocdevprivate() still uses dev_net(dev) for the collision lookup. For a tunnel moved through IFLA_NET_NS_FD, dev_net(dev) is the new netns, not t->net. SIOCCHGTUNNEL on a migrated tunnel then runs: net = dev_net(dev) /* migrated netns */ t = vti6_locate(net, &p1, false) /* misses target in t->net */ ... t = netdev_priv(dev) vti6_update(t, &p1, false) /* mutates t->net's hash */ A caller in the migrated netns picks params that match a tunnel in the creation netns. The lookup in dev_net(dev) finds nothing. vti6_update() prepends the migrated tunnel at the head of the creation netns hash bucket for those params. Later lookups in the creation netns resolve to the migrated device. xfrm receive delivers the matched packets through a device the caller controls. Reachable from an unprivileged user namespace (unshare --user --map-root-user --net). Cross tenant scope on container hosts. Switch the SIOCCHGTUNNEL path on a non fallback device to use t->net for the lookup. The lookup now matches the netns vti6_update() operates on. Also add ns_capable(self->net->user_ns, CAP_NET_ADMIN) before the lookup. The check at the top of the case is against dev_net(dev)->user_ns, which after migration is the attacker's netns. A caller there can pick params absent from self->net, the lookup returns NULL, t becomes self, and vti6_update() inserts the device into the creation netns hash. The new check requires CAP_NET_ADMIN in the creation netns user_ns too. SIOCADDTUNNEL and SIOCCHGTUNNEL on the fallback device keep dev_net(dev), which equals init_net there. Fixes: 61220ab34948 ("vti6: Enable namespace changing") Suggested-by: Jakub Kicinski Suggested-by: Xiao Liang Cc: stable@vger.kernel.org # v5.15+ Signed-off-by: Maoyi Xie Link: https://patch.msgid.link/20260521130555.3421684-3-maoyixie.tju@gmail.com Signed-off-by: Paolo Abeni --- net/ipv6/ip6_vti.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/ipv6/ip6_vti.c b/net/ipv6/ip6_vti.c index dcb257411d6e..df793c8bfffb 100644 --- a/net/ipv6/ip6_vti.c +++ b/net/ipv6/ip6_vti.c @@ -835,17 +835,24 @@ vti6_siocdevprivate(struct net_device *dev, struct ifreq *ifr, void __user *data if (p.proto != IPPROTO_IPV6 && p.proto != 0) break; vti6_parm_from_user(&p1, &p); - t = vti6_locate(net, &p1, cmd == SIOCADDTUNNEL); if (dev != ip6n->fb_tnl_dev && cmd == SIOCCHGTUNNEL) { + struct ip6_tnl *self = netdev_priv(dev); + + err = -EPERM; + if (!ns_capable(self->net->user_ns, CAP_NET_ADMIN)) + break; + t = vti6_locate(self->net, &p1, false); if (t) { if (t->dev != dev) { err = -EEXIST; break; } } else - t = netdev_priv(dev); + t = self; err = vti6_update(t, &p1, false); + } else { + t = vti6_locate(net, &p1, cmd == SIOCADDTUNNEL); } if (t) { err = 0; -- cgit v1.2.3 From 2e357f002c61fd76fd8f12468744a06a5ec48eaa Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Bj=C3=B6rn=20T=C3=B6pel?= Date: Fri, 22 May 2026 14:06:40 +0200 Subject: net: Avoid checksumming unreadable skb tail on trim MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit pskb_trim_rcsum_slow() keeps CHECKSUM_COMPLETE valid by subtracting the checksum of the bytes removed from the skb tail. That assumes the removed bytes can be read. io_uring zcrx skbs may contain unreadable net_iov frags. With fbnic header/data split, small TCP/IPv4 packets can carry Ethernet padding in such a frag. ip_rcv_core() trims the skb to iph->tot_len before TCP sees it, and the CHECKSUM_COMPLETE adjustment then calls skb_checksum() on the padding. This is exposed by IPv4 because small TCP/IPv4 frames can be shorter than the Ethernet minimum payload. TCP/IPv6 frames are large enough in the normal zcrx path, so they do not hit the same padding trim. Keep the existing checksum adjustment for readable skbs. If the remaining packet is fully linear, drop CHECKSUM_COMPLETE and let the stack validate the packet after trimming. If unreadable payload would remain, fail the trim; the checksum cannot be adjusted without reading the trimmed tail. Also clear skb->unreadable when trimming removes all frags. Fixes: 65249feb6b3d ("net: add support for skbs with unreadable frags") Signed-off-by: Björn Töpel Reviewed-by: Breno Leitao Link: https://patch.msgid.link/20260522120643.242974-1-bjorn@kernel.org Signed-off-by: Paolo Abeni --- net/core/skbuff.c | 31 +++++++++++++++++++++++++++---- 1 file changed, 27 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 44ac121cfccb..d247acd447e4 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -2787,6 +2787,8 @@ done: skb->data_len = 0; skb_set_tail_pointer(skb, len); } + if (!skb_shinfo(skb)->nr_frags && !skb_has_frag_list(skb)) + skb->unreadable = 0; if (!skb->sk || skb->destructor == sock_edemux) skb_condense(skb); @@ -2794,16 +2796,37 @@ done: } EXPORT_SYMBOL(___pskb_trim); +static int pskb_trim_rcsum_complete(struct sk_buff *skb, unsigned int len) +{ + int delta = skb->len - len; + + if (skb_frags_readable(skb)) { + skb->csum = csum_block_sub(skb->csum, + skb_checksum(skb, len, delta, 0), + len); + return 0; + } + + if (len > skb_headlen(skb)) + return -EFAULT; + + /* The trimmed bytes are unreadable, but the remaining packet can be + * checksummed by software after trimming. + */ + skb->ip_summed = CHECKSUM_NONE; + return 0; +} + /* Note : use pskb_trim_rcsum() instead of calling this directly */ int pskb_trim_rcsum_slow(struct sk_buff *skb, unsigned int len) { if (skb->ip_summed == CHECKSUM_COMPLETE) { - int delta = skb->len - len; + int err; - skb->csum = csum_block_sub(skb->csum, - skb_checksum(skb, len, delta, 0), - len); + err = pskb_trim_rcsum_complete(skb, len); + if (err) + return err; } else if (skb->ip_summed == CHECKSUM_PARTIAL) { int hdlen = (len > skb_headlen(skb)) ? skb_headlen(skb) : len; int offset = skb_checksum_start_offset(skb) + skb->csum_offset; -- cgit v1.2.3 From c75b6f6eaacd0b74b832414cc3b9289c3686e941 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 22 May 2026 16:06:42 -0700 Subject: ethtool: rss: avoid modifying the RSS context response Gemini says that we're modifying the RSS_CREATE response skb. I think it's right, the comment says that unicast() should unshare the skb but I'm not entirely sure what I meant there. netlink_trim() does a copy but only if skb is not well sized (it's at least 2x larger than necessary for the payload). Fixes: a166ab7816c5 ("ethtool: rss: support creating contexts via Netlink") Link: https://patch.msgid.link/20260522230647.1705600-2-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/rss.c | 18 ++++++++++-------- 1 file changed, 10 insertions(+), 8 deletions(-) (limited to 'net') diff --git a/net/ethtool/rss.c b/net/ethtool/rss.c index 353110b862ab..8ffec9785efa 100644 --- a/net/ethtool/rss.c +++ b/net/ethtool/rss.c @@ -981,11 +981,17 @@ ethnl_rss_create_validate(struct net_device *dev, struct genl_info *info) } static void -ethnl_rss_create_send_ntf(struct sk_buff *rsp, struct net_device *dev) +ethnl_rss_create_send_ntf(const struct sk_buff *rsp, struct net_device *dev) { - struct nlmsghdr *nlh = (void *)rsp->data; struct genlmsghdr *genl_hdr; + struct nlmsghdr *nlh; + struct sk_buff *ntf; + + ntf = skb_copy_expand(rsp, 0, 0, GFP_KERNEL); + if (!ntf) + return; + nlh = nlmsg_hdr(ntf); /* Convert the reply into a notification */ nlh->nlmsg_pid = 0; nlh->nlmsg_seq = ethnl_bcast_seq_next(); @@ -993,7 +999,7 @@ ethnl_rss_create_send_ntf(struct sk_buff *rsp, struct net_device *dev) genl_hdr = nlmsg_data(nlh); genl_hdr->cmd = ETHTOOL_MSG_RSS_CREATE_NTF; - ethnl_multicast(rsp, dev); + ethnl_multicast(ntf, dev); } int ethnl_rss_create_doit(struct sk_buff *skb, struct genl_info *info) @@ -1104,12 +1110,8 @@ int ethnl_rss_create_doit(struct sk_buff *skb, struct genl_info *info) genlmsg_end(rsp, hdr); - /* Use the same skb for the response and the notification, - * genlmsg_reply() will copy the skb if it has elevated user count. - */ - skb_get(rsp); - ret = genlmsg_reply(rsp, info); ethnl_rss_create_send_ntf(rsp, dev); + ret = genlmsg_reply(rsp, info); rsp = NULL; exit_unlock: -- cgit v1.2.3 From 3e6c6e9782ff8a8d8ded774b07ad4590cd61d04c Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 22 May 2026 16:06:43 -0700 Subject: ethtool: rss: add missing errno on RSS context delete Remember to set ret before jumping out if someone tries to delete a context on a device which doesn't support contexts. Fixes: fbe09277fa63 ("ethtool: rss: support removing contexts via Netlink") Link: https://patch.msgid.link/20260522230647.1705600-3-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/rss.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/ethtool/rss.c b/net/ethtool/rss.c index 8ffec9785efa..a16ee1e8e640 100644 --- a/net/ethtool/rss.c +++ b/net/ethtool/rss.c @@ -1170,8 +1170,10 @@ int ethnl_rss_delete_doit(struct sk_buff *skb, struct genl_info *info) dev = req.dev; ops = dev->ethtool_ops; - if (!ops->create_rxfh_context) + if (!ops->create_rxfh_context) { + ret = -EOPNOTSUPP; goto exit_free_dev; + } rtnl_lock(); netdev_lock_ops(dev); -- cgit v1.2.3 From 8d60141a32875248ef71d49c9920fa5e2aa40b29 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 22 May 2026 16:06:44 -0700 Subject: ethtool: rss: fix falsely ignoring indir table updates rss_set_prep_indir() compares the new indirection table against the current one to determine whether any update is needed. The memcmp call passes data->indir_size as the length argument, but indir_size is the number of u32 entries, not the byte count. Fixes: c0ae03588bbb ("ethtool: rss: initial RSS_SET (indirection table handling)") Link: https://patch.msgid.link/20260522230647.1705600-4-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/rss.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/ethtool/rss.c b/net/ethtool/rss.c index a16ee1e8e640..458a4a7907e4 100644 --- a/net/ethtool/rss.c +++ b/net/ethtool/rss.c @@ -686,7 +686,7 @@ rss_set_prep_indir(struct net_device *dev, struct genl_info *info, ethtool_rxfh_indir_default(i, num_rx_rings); } - *mod |= memcmp(rxfh->indir, data->indir_table, data->indir_size); + *mod |= memcmp(rxfh->indir, data->indir_table, alloc_size); return user_size; -- cgit v1.2.3 From 266297692f97008ca48bc311775c087c59bd7fe3 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 22 May 2026 16:06:45 -0700 Subject: ethtool: rss: fix indir_table and hkey leak on get_rxfh failure rss_prepare_get() allocates the indirection table and hash key buffer via rss_get_data_alloc(), then calls ops->get_rxfh() to populate them. If get_rxfh() fails, the function returns an error without freeing the allocation. Fixes: 4f038a6a02d2 ("net: ethtool: Don't call .cleanup_data when prepare_data fails") Link: https://patch.msgid.link/20260522230647.1705600-5-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/rss.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/ethtool/rss.c b/net/ethtool/rss.c index 458a4a7907e4..9fb675d29232 100644 --- a/net/ethtool/rss.c +++ b/net/ethtool/rss.c @@ -170,8 +170,10 @@ rss_prepare_get(const struct rss_req_info *request, struct net_device *dev, rxfh.key = data->hkey; ret = ops->get_rxfh(dev, &rxfh); - if (ret) + if (ret) { + rss_get_data_free(data); goto out_unlock; + } data->hfunc = rxfh.hfunc; data->input_xfrm = rxfh.input_xfrm; -- cgit v1.2.3 From 78ccf1a70c6378e1f5073a8c2209b5129067b925 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 22 May 2026 16:06:46 -0700 Subject: ethtool: rss: fix hkey leak when indir_size is 0 rss_get_data_alloc() allocates a single buffer that backs both the indirection table and the hash key, but only assigned data->indir_table when indir_size was nonzero. The expectation was that no driver implements RSS without supporting indirection table but apparently enic does just that (it's the only such in-tree driver). enic has get_rxfh_key_size but no get_rxfh_indir_size. data->indir_table stays as NULL, hkey gets set but rss_get_data_free() kfree(data->indir_table) is a nop and the allocation leaks. Always store the allocation base in data->indir_table so the free path is unambiguous. No caller treats indir_table as a sentinel; everything keys off indir_size. Fixes: 7112a04664bf ("ethtool: add netlink based get rss support") Link: https://patch.msgid.link/20260522230647.1705600-6-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/rss.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'net') diff --git a/net/ethtool/rss.c b/net/ethtool/rss.c index 9fb675d29232..f5cf214f8f85 100644 --- a/net/ethtool/rss.c +++ b/net/ethtool/rss.c @@ -134,8 +134,7 @@ rss_get_data_alloc(struct net_device *dev, struct rss_reply_data *data) if (!rss_config) return -ENOMEM; - if (data->indir_size) - data->indir_table = (u32 *)rss_config; + data->indir_table = (u32 *)rss_config; if (data->hkey_size) data->hkey = rss_config + indir_bytes; -- cgit v1.2.3 From 32a9ecde62731c9f7412507709192c84dafc38d1 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 22 May 2026 16:06:47 -0700 Subject: ethtool: rss: avoid device context leak on reply-build failure We wait with filling the reply for new RSS context creation until after the driver ->create_rxfh_context call. The driver needs to fill some of the defaults in the context. The failure of rss_fill_reply() is somewhat theoretical, but doesn't take much effort to handle it properly. Call ->remove_rxfh_context(). If the driver's remove callback fails (some implementations like sfc can return real command errors from firmware RPCs) - skip the xa_erase and kfree, leaving the context in the xarray. This matches how ethnl_rss_delete_doit() behaves. Fixes: a166ab7816c5 ("ethtool: rss: support creating contexts via Netlink") Link: https://patch.msgid.link/20260522230647.1705600-7-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/rss.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/ethtool/rss.c b/net/ethtool/rss.c index f5cf214f8f85..53792f53f922 100644 --- a/net/ethtool/rss.c +++ b/net/ethtool/rss.c @@ -1106,7 +1106,7 @@ int ethnl_rss_create_doit(struct sk_buff *skb, struct genl_info *info) ntf_fail |= rss_fill_reply(rsp, &req.base, &data.base); if (WARN_ON(!hdr || ntf_fail)) { ret = -EMSGSIZE; - goto exit_unlock; + goto err_remove_ctx; } genlmsg_end(rsp, hdr); @@ -1134,6 +1134,10 @@ exit_free_rsp: nlmsg_free(rsp); return ret; +err_remove_ctx: + if (ops->remove_rxfh_context(dev, ctx, req.rss_context, NULL)) + /* leave the context on failure, like ethnl_rss_delete_doit() */ + goto exit_unlock; err_ctx_id_free: xa_erase(&dev->ethtool->rss_ctx, req.rss_context); err_unlock_free_ctx: -- cgit v1.2.3 From 84371fb58423f997939aacdcbc02d128d76a54e5 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 22 May 2026 16:13:04 -0700 Subject: ethtool: module: call ethnl_ops_complete() on module flash errors When validate() fails we are skipping over ethnl_ops_complete() even tho we already called ethnl_ops_begin(). Fixes: 32b4c8b53ee7 ("ethtool: Add ability to flash transceiver modules' firmware") Reviewed-by: Maxime Chevallier Reviewed-by: Danielle Ratson Link: https://patch.msgid.link/20260522231312.1710836-2-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/module.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/ethtool/module.c b/net/ethtool/module.c index cad2eb25b5a4..741f6fb25d45 100644 --- a/net/ethtool/module.c +++ b/net/ethtool/module.c @@ -427,10 +427,11 @@ int ethnl_act_module_fw_flash(struct sk_buff *skb, struct genl_info *info) ret = ethnl_module_fw_flash_validate(dev, info->extack); if (ret < 0) - goto out_unlock; + goto out_complete; ret = module_flash_fw(dev, tb, skb, info); +out_complete: ethnl_ops_complete(dev); out_unlock: -- cgit v1.2.3 From fb7f511d62692661846c47f199e0afe25c2982db Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 22 May 2026 16:13:05 -0700 Subject: ethtool: module: avoid leaking a netdev ref on module flash errors module_flash_fw_schedule() is missing undo for setting the "in_progress" flag and taking the netdev reference. Delay taking these, the device can't disappear while we are holding rtnl_lock. Fixes: 32b4c8b53ee7 ("ethtool: Add ability to flash transceiver modules' firmware") Reviewed-by: Maxime Chevallier Reviewed-by: Danielle Ratson Link: https://patch.msgid.link/20260522231312.1710836-3-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/module.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/ethtool/module.c b/net/ethtool/module.c index 741f6fb25d45..392c03935e5e 100644 --- a/net/ethtool/module.c +++ b/net/ethtool/module.c @@ -319,8 +319,6 @@ module_flash_fw_schedule(struct net_device *dev, const char *file_name, if (err < 0) goto err_release_firmware; - dev->ethtool->module_fw_flash_in_progress = true; - netdev_hold(dev, &module_fw->dev_tracker, GFP_KERNEL); fw_update->dev = dev; fw_update->ntf_params.portid = info->snd_portid; fw_update->ntf_params.seq = info->snd_seq; @@ -335,6 +333,9 @@ module_flash_fw_schedule(struct net_device *dev, const char *file_name, if (err < 0) goto err_release_firmware; + dev->ethtool->module_fw_flash_in_progress = true; + netdev_hold(dev, &module_fw->dev_tracker, GFP_KERNEL); + schedule_work(&module_fw->work); return 0; -- cgit v1.2.3 From 7a84b965ffc12030af63cd10a8f3a1123ff39b7a Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 22 May 2026 16:13:06 -0700 Subject: ethtool: module: avoid racy updates to dev->ethtool bitfield When reviewing other changes Gemini points out that we currently update module_fw_flash_in_progress without holding any locks. Since module_fw_flash_in_progress is part of a bitfield this is not great, updates to other fields may be lost. We could use a bool and sprinkle some READ_ONCE/WRITE_ONCE here but seems like the issue is rather than the work is an unusual writer. The other writers already hold the right locks. So just very briefly take these locks when the work completes. Note that nothing ever cancels the FW update work, so there's no concern with deadlocks vs cancel. Fixes: 32b4c8b53ee7 ("ethtool: Add ability to flash transceiver modules' firmware") Reviewed-by: Maxime Chevallier Reviewed-by: Danielle Ratson Link: https://patch.msgid.link/20260522231312.1710836-4-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/module.c | 12 ++++++++++-- 1 file changed, 10 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/ethtool/module.c b/net/ethtool/module.c index 392c03935e5e..cdb85e19a23b 100644 --- a/net/ethtool/module.c +++ b/net/ethtool/module.c @@ -221,14 +221,22 @@ static void module_flash_fw_work_list_del(struct list_head *list) static void module_flash_fw_work(struct work_struct *work) { struct ethtool_module_fw_flash *module_fw; + struct net_device *dev; module_fw = container_of(work, struct ethtool_module_fw_flash, work); + dev = module_fw->fw_update.dev; ethtool_cmis_fw_update(&module_fw->fw_update); module_flash_fw_work_list_del(&module_fw->list); - module_fw->fw_update.dev->ethtool->module_fw_flash_in_progress = false; - netdev_put(module_fw->fw_update.dev, &module_fw->dev_tracker); + + rtnl_lock(); + netdev_lock_ops(dev); + dev->ethtool->module_fw_flash_in_progress = false; + netdev_unlock_ops(dev); + rtnl_unlock(); + + netdev_put(dev, &module_fw->dev_tracker); release_firmware(module_fw->fw_update.fw); kfree(module_fw); } -- cgit v1.2.3 From 504eaefa44c8dec50f7499edcb36d24f3aefab2a Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 22 May 2026 16:13:07 -0700 Subject: ethtool: module: check fw_flash_in_progress under rtnl_lock ethnl_set_module_validate() inspects module_fw_flash_in_progress but validate is meant for _input_ validation, not state validation. rtnl_lock is not held, yet. Move the check into ethnl_set_module(). Fixes: 32b4c8b53ee7 ("ethtool: Add ability to flash transceiver modules' firmware") Reviewed-by: Maxime Chevallier Reviewed-by: Danielle Ratson Link: https://patch.msgid.link/20260522231312.1710836-5-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/module.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/net/ethtool/module.c b/net/ethtool/module.c index cdb85e19a23b..5b49004ddf60 100644 --- a/net/ethtool/module.c +++ b/net/ethtool/module.c @@ -120,12 +120,6 @@ ethnl_set_module_validate(struct ethnl_req_info *req_info, if (!tb[ETHTOOL_A_MODULE_POWER_MODE_POLICY]) return 0; - if (req_info->dev->ethtool->module_fw_flash_in_progress) { - NL_SET_ERR_MSG(info->extack, - "Module firmware flashing is in progress"); - return -EBUSY; - } - if (!ops->get_module_power_mode || !ops->set_module_power_mode) { NL_SET_ERR_MSG_ATTR(info->extack, tb[ETHTOOL_A_MODULE_POWER_MODE_POLICY], @@ -148,6 +142,12 @@ ethnl_set_module(struct ethnl_req_info *req_info, struct genl_info *info) ops = dev->ethtool_ops; + if (dev->ethtool->module_fw_flash_in_progress) { + NL_SET_ERR_MSG(info->extack, + "Module firmware flashing is in progress"); + return -EBUSY; + } + power_new.policy = nla_get_u8(tb[ETHTOOL_A_MODULE_POWER_MODE_POLICY]); ret = ops->get_module_power_mode(dev, &power, info->extack); if (ret < 0) -- cgit v1.2.3 From 760d04ebad5c4304f22c0d2251c9623b87a117c8 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 22 May 2026 16:13:08 -0700 Subject: ethtool: module: fix cleanup if socket used for flashing multiple devices When a single Netlink socket issues MODULE_FW_FLASH_ACT against multiple devices, ethnl_sock_priv_set() overwrites sk_priv->dev on each call, retaining only the last one. The socket priv is used on socket close, to walk the global work list and mark the uncompleted flashing work as "orphaned". Otherwise if another socket reuses the PID it will unexpectedly receive the flashing notifications. Don't record the device, record net pointer instead. The purpose of the dev is to scope the work to a netns, anyway. If we store netns the overrides are safe/a nop since all flashed devices must be in the same netns as the socket. Fixes: 32b4c8b53ee7 ("ethtool: Add ability to flash transceiver modules' firmware") Reviewed-by: Danielle Ratson Link: https://patch.msgid.link/20260522231312.1710836-6-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/module.c | 9 ++++----- net/ethtool/netlink.c | 4 ++-- net/ethtool/netlink.h | 4 ++-- 3 files changed, 8 insertions(+), 9 deletions(-) (limited to 'net') diff --git a/net/ethtool/module.c b/net/ethtool/module.c index 5b49004ddf60..ea4fb2a76650 100644 --- a/net/ethtool/module.c +++ b/net/ethtool/module.c @@ -291,11 +291,9 @@ void ethnl_module_fw_flash_sock_destroy(struct ethnl_sock_priv *sk_priv) spin_lock(&module_fw_flash_work_list_lock); list_for_each_entry(work, &module_fw_flash_work_list, list) { - if (work->fw_update.dev == sk_priv->dev && - work->fw_update.ntf_params.portid == sk_priv->portid) { + if (work->fw_update.ntf_params.portid == sk_priv->portid && + dev_net(work->fw_update.dev) == sk_priv->net) work->fw_update.ntf_params.closed_sock = true; - break; - } } spin_unlock(&module_fw_flash_work_list_lock); } @@ -332,7 +330,8 @@ module_flash_fw_schedule(struct net_device *dev, const char *file_name, fw_update->ntf_params.seq = info->snd_seq; fw_update->ntf_params.closed_sock = false; - err = ethnl_sock_priv_set(skb, dev, fw_update->ntf_params.portid, + err = ethnl_sock_priv_set(skb, dev_net(dev), + fw_update->ntf_params.portid, ETHTOOL_SOCK_TYPE_MODULE_FW_FLASH); if (err < 0) goto err_release_firmware; diff --git a/net/ethtool/netlink.c b/net/ethtool/netlink.c index 5046023a30b1..7d45f9a884e5 100644 --- a/net/ethtool/netlink.c +++ b/net/ethtool/netlink.c @@ -53,7 +53,7 @@ const struct nla_policy ethnl_header_policy_phy_stats[] = { [ETHTOOL_A_HEADER_PHY_INDEX] = NLA_POLICY_MIN(NLA_U32, 1), }; -int ethnl_sock_priv_set(struct sk_buff *skb, struct net_device *dev, u32 portid, +int ethnl_sock_priv_set(struct sk_buff *skb, struct net *net, u32 portid, enum ethnl_sock_type type) { struct ethnl_sock_priv *sk_priv; @@ -62,7 +62,7 @@ int ethnl_sock_priv_set(struct sk_buff *skb, struct net_device *dev, u32 portid, if (IS_ERR(sk_priv)) return PTR_ERR(sk_priv); - sk_priv->dev = dev; + sk_priv->net = net; sk_priv->portid = portid; sk_priv->type = type; diff --git a/net/ethtool/netlink.h b/net/ethtool/netlink.h index aaf6f2468768..fd2198e45d2b 100644 --- a/net/ethtool/netlink.h +++ b/net/ethtool/netlink.h @@ -318,12 +318,12 @@ enum ethnl_sock_type { }; struct ethnl_sock_priv { - struct net_device *dev; + struct net *net; u32 portid; enum ethnl_sock_type type; }; -int ethnl_sock_priv_set(struct sk_buff *skb, struct net_device *dev, u32 portid, +int ethnl_sock_priv_set(struct sk_buff *skb, struct net *net, u32 portid, enum ethnl_sock_type type); /** -- cgit v1.2.3 From 6c3f999a9d1338c6c89a9ff4549eafe72bc2e7b1 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 22 May 2026 16:13:09 -0700 Subject: ethtool: cmis: require exact CDB reply length Malicious SFP module could respond with rpl_len longer than what cmis_cdb_process_reply() expected, leading to OOB writes. Malicious HW is a bit theoretical but some modules may just be buggy and/or the reads may occasionally get corrupted, so let's protect the kernel. The existing check protects from short replies. We need to protect from long ones, too. All callers that pass a non-zero rpl_exp_len cast the reply payload to a fixed-layout struct and read fields at fixed offsets, with no version negotiation or short-reply handling: - cmis_cdb_validate_password() - cmis_cdb_module_features_get() - cmis_fw_update_fw_mng_features_get() so let's assume that responses longer than expected do not have to be handled gracefully here. Add a warning message to make the debug easier in case my understanding is wrong... Note that page_data->length (argument of kmalloc) comes from last arg to ethtool_cmis_page_init() which is rpl_exp_len. Note2 that AIs also like to point out overflows in args->req.payload itself (which is a fixed-size 120 B buffer, on the stack), but callers should be reading structs defined by the standard, so protecting from requests for more data than max seem like defensive programming. Fixes: a39c84d79625 ("ethtool: cmis_cdb: Add a layer for supporting CDB commands") Reviewed-by: Danielle Ratson Link: https://patch.msgid.link/20260522231312.1710836-7-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/cmis_cdb.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/ethtool/cmis_cdb.c b/net/ethtool/cmis_cdb.c index 3670ca42dd40..f3a53a984460 100644 --- a/net/ethtool/cmis_cdb.c +++ b/net/ethtool/cmis_cdb.c @@ -513,8 +513,13 @@ static int cmis_cdb_process_reply(struct net_device *dev, } rpl = (struct ethtool_cmis_cdb_rpl *)page_data->data; - if ((args->rpl_exp_len > rpl->hdr.rpl_len + rpl_hdr_len) || - !rpl->hdr.rpl_chk_code) { + if (rpl->hdr.rpl_len != args->rpl_exp_len) { + netdev_warn(dev, "CDB reply length mismatch, expected %u got %u\n", + args->rpl_exp_len, rpl->hdr.rpl_len); + err = -EIO; + goto out; + } + if (!rpl->hdr.rpl_chk_code) { err = -EIO; goto out; } -- cgit v1.2.3 From 3e8c3d464c36bb342fe377b026577c7ec27fdbb4 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 22 May 2026 16:13:10 -0700 Subject: ethtool: cmis: fix u16-to-u8 truncation of msleep_pre_rpl ethtool_cmis_cdb_compose_args() accepts msleep_pre_rpl as u16 but stores it into the u8 field ethtool_cmis_cdb_cmd_args::msleep_pre_rpl, silently truncating values >= 256. Seven of the nine call sites pass 1000 ms (it's the third argument from the end). Fixes: a39c84d79625 ("ethtool: cmis_cdb: Add a layer for supporting CDB commands") Reviewed-by: Maxime Chevallier Reviewed-by: Danielle Ratson Link: https://patch.msgid.link/20260522231312.1710836-8-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/cmis.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/ethtool/cmis.h b/net/ethtool/cmis.h index 4a9a946cabf0..778783a0f23c 100644 --- a/net/ethtool/cmis.h +++ b/net/ethtool/cmis.h @@ -63,9 +63,9 @@ struct ethtool_cmis_cdb_request { * struct ethtool_cmis_cdb_cmd_args - CDB commands execution arguments * @req: CDB command fields as described in the CMIS standard. * @max_duration: Maximum duration time for command completion in msec. + * @msleep_pre_rpl: Waiting time before checking reply in msec. * @read_write_len_ext: Allowable additional number of byte octets to the LPL * in a READ or a WRITE commands. - * @msleep_pre_rpl: Waiting time before checking reply in msec. * @rpl_exp_len: Expected reply length in bytes. * @flags: Validation flags for CDB commands. * @err_msg: Error message to be sent to user space. @@ -73,8 +73,8 @@ struct ethtool_cmis_cdb_request { struct ethtool_cmis_cdb_cmd_args { struct ethtool_cmis_cdb_request req; u16 max_duration; + u16 msleep_pre_rpl; u8 read_write_len_ext; - u8 msleep_pre_rpl; u8 rpl_exp_len; u8 flags; char *err_msg; -- cgit v1.2.3 From 12c2496a71f82f63617971ca9b730dffa05cf58b Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 22 May 2026 16:13:11 -0700 Subject: ethtool: cmis: validate start_cmd_payload_size from module The CMIS firmware update code reads start_cmd_payload_size from the module's FW Management Features CDB reply and uses it directly as the byte count for memcpy. The destination buffer is 112 bytes (ETHTOOL_CMIS_CDB_LPL_MAX_PL_LENGTH - 8). So a malicious module (or corrupted response) can cause a OOB write later on in cmis_fw_update_start_download(). Let's error out. If modules that expect longer LPL writes actually exist we should revisit. struct cmis_cdb_start_fw_download_pl's definition has to move, no change there. Fixes: c4f78134d45c ("ethtool: cmis_fw_update: add a layer for supporting firmware update using CDB") Reviewed-by: Maxime Chevallier Reviewed-by: Danielle Ratson Link: https://patch.msgid.link/20260522231312.1710836-9-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/cmis_fw_update.c | 36 ++++++++++++++++++++++-------------- 1 file changed, 22 insertions(+), 14 deletions(-) (limited to 'net') diff --git a/net/ethtool/cmis_fw_update.c b/net/ethtool/cmis_fw_update.c index df5f344209c4..16190c97e1f7 100644 --- a/net/ethtool/cmis_fw_update.c +++ b/net/ethtool/cmis_fw_update.c @@ -44,6 +44,20 @@ enum cmis_cdb_fw_write_mechanism { CMIS_CDB_FW_WRITE_MECHANISM_BOTH = 0x11, }; +/* See section 9.7.2 "CMD 0101h: Start Firmware Download" in CMIS standard + * revision 5.2. + * struct cmis_cdb_start_fw_download_pl is a structured layout of the + * flat array, ethtool_cmis_cdb_request::payload. + */ +struct cmis_cdb_start_fw_download_pl { + __struct_group(cmis_cdb_start_fw_download_pl_h, head, /* no attrs */, + __be32 image_size; + __be32 resv1; + ); + u8 vendor_data[ETHTOOL_CMIS_CDB_LPL_MAX_PL_LENGTH - + sizeof(struct cmis_cdb_start_fw_download_pl_h)]; +}; + static int cmis_fw_update_fw_mng_features_get(struct ethtool_cmis_cdb *cdb, struct net_device *dev, @@ -86,6 +100,14 @@ cmis_fw_update_fw_mng_features_get(struct ethtool_cmis_cdb *cdb, */ cdb->read_write_len_ext = rpl->read_write_len_ext; fw_mng->start_cmd_payload_size = rpl->start_cmd_payload_size; + if (fw_mng->start_cmd_payload_size > + sizeof_field(struct cmis_cdb_start_fw_download_pl, vendor_data)) { + ethnl_module_fw_flash_ntf_err(dev, ntf_params, + "Start cmd payload size exceeds max LPL payload", + NULL); + return -EINVAL; + } + fw_mng->write_mechanism = rpl->write_mechanism == CMIS_CDB_FW_WRITE_MECHANISM_LPL ? CMIS_CDB_FW_WRITE_MECHANISM_LPL : @@ -97,20 +119,6 @@ cmis_fw_update_fw_mng_features_get(struct ethtool_cmis_cdb *cdb, return 0; } -/* See section 9.7.2 "CMD 0101h: Start Firmware Download" in CMIS standard - * revision 5.2. - * struct cmis_cdb_start_fw_download_pl is a structured layout of the - * flat array, ethtool_cmis_cdb_request::payload. - */ -struct cmis_cdb_start_fw_download_pl { - __struct_group(cmis_cdb_start_fw_download_pl_h, head, /* no attrs */, - __be32 image_size; - __be32 resv1; - ); - u8 vendor_data[ETHTOOL_CMIS_CDB_LPL_MAX_PL_LENGTH - - sizeof(struct cmis_cdb_start_fw_download_pl_h)]; -}; - static int cmis_fw_update_start_download(struct ethtool_cmis_cdb *cdb, struct ethtool_cmis_fw_update_params *fw_update, -- cgit v1.2.3 From d5551f4c1800dc714cec86647bdd651ae0de923e Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 22 May 2026 16:13:12 -0700 Subject: ethtool: cmis: validate fw->size against start_cmd_payload_size cmis_fw_update_start_download() copies start_cmd_payload_size bytes from the firmware blob into the CDB LPL vendor_data[] payload without validating that the FW has enough data. Since the start_cmd_payload_size can only be ~120B an image too short is most likely corrupted, so reject it. Fixes: c4f78134d45c ("ethtool: cmis_fw_update: add a layer for supporting firmware update using CDB") Reviewed-by: Maxime Chevallier Reviewed-by: Danielle Ratson Link: https://patch.msgid.link/20260522231312.1710836-10-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/cmis_fw_update.c | 8 ++++++++ 1 file changed, 8 insertions(+) (limited to 'net') diff --git a/net/ethtool/cmis_fw_update.c b/net/ethtool/cmis_fw_update.c index 16190c97e1f7..291d04d2776a 100644 --- a/net/ethtool/cmis_fw_update.c +++ b/net/ethtool/cmis_fw_update.c @@ -130,6 +130,14 @@ cmis_fw_update_start_download(struct ethtool_cmis_cdb *cdb, u8 lpl_len; int err; + if (fw_update->fw->size < vendor_data_size) { + ethnl_module_fw_flash_ntf_err(fw_update->dev, + &fw_update->ntf_params, + "Firmware image too small for module's start payload", + NULL); + return -EINVAL; + } + pl.image_size = cpu_to_be32(fw_update->fw->size); memcpy(pl.vendor_data, fw_update->fw->data, vendor_data_size); -- cgit v1.2.3 From 05f95729ca844704d15e49ce14868af4b403b32b Mon Sep 17 00:00:00 2001 From: Michael Bommarito Date: Fri, 22 May 2026 22:34:23 -0400 Subject: l2tp: use refcount_inc_not_zero in l2tp_session_get_by_ifname A reader in l2tp_session_get_by_ifname() can return a pointer to a session whose refcount has reached zero. The getter takes its reference with plain refcount_inc(), but every other session getter in the same file (l2tp_v2_session_get, l2tp_v3_session_get, and the corresponding _get_next variants) uses refcount_inc_not_zero() because the IDR/RCU lookup can race with refcount_dec_and_test() -> l2tp_session_free() -> kfree_rcu(). The ifname getter is the only outlier; the inconsistency was raised on-list after 979c017803c4 ("l2tp: use list_del_rcu in l2tp_session_unhash"). A reader inside rcu_read_lock_bh() that matches session->ifname can be preempted between the strcmp() and the refcount_inc(). If the last reference drops on another CPU in that window, the reader's refcount_inc() runs on a counter that has reached zero. refcount_t catches the addition-on-zero, prints "refcount_t: addition on 0; use-after-free", saturates the counter, and returns the saturated pointer to the caller. Session memory is held live by the in-flight RCU read section, but the kfree_rcu() callback queued from l2tp_session_free() will free it once the grace period closes; a caller that dereferences the returned session past that point hits a slab-use-after-free. On PREEMPT_RT local_bh_disable() is a per-CPU sleeping lock and the preemption window is real; on stock PREEMPT kernels local_bh_disable() is a preempt_count increment that closes the cross-CPU race in practice (see below). Use refcount_inc_not_zero() and continue the list walk on failure, matching the other session getters in the file. The ifname getter is the only session getter in net/l2tp/ that still uses the bare refcount_inc() pattern; this change restores file-internal consistency. The success path is unchanged. Fixes: abe7a1a7d0b6 ("l2tp: improve tunnel/session refcount helpers") Cc: stable@vger.kernel.org Signed-off-by: Michael Bommarito Reviewed-by: James Chapman Reviewed-by: Simon Horman Link: https://patch.msgid.link/20260523023423.2568972-1-michael.bommarito@gmail.com Signed-off-by: Jakub Kicinski --- net/l2tp/l2tp_core.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/l2tp/l2tp_core.c b/net/l2tp/l2tp_core.c index 1455f67e01dd..9419c8555d22 100644 --- a/net/l2tp/l2tp_core.c +++ b/net/l2tp/l2tp_core.c @@ -441,12 +441,13 @@ struct l2tp_session *l2tp_session_get_by_ifname(const struct net *net, idr_for_each_entry_ul(&pn->l2tp_tunnel_idr, tunnel, tmp, tunnel_id) { if (tunnel) { list_for_each_entry_rcu(session, &tunnel->session_list, list) { - if (!strcmp(session->ifname, ifname)) { - refcount_inc(&session->ref_count); - rcu_read_unlock_bh(); + if (strcmp(session->ifname, ifname)) + continue; + if (!refcount_inc_not_zero(&session->ref_count)) + continue; + rcu_read_unlock_bh(); - return session; - } + return session; } } } -- cgit v1.2.3 From b4bc94353050b1fa7b702bd4c6600710dd926cff Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 25 May 2026 20:13:35 +0000 Subject: tunnels: load network headers after skb_cow() in iptunnel_pmtud_build_icmp[v6]() Sashiko found that iptunnel_pmtud_build_icmp() and iptunnel_pmtud_build_icmpv6() were caching ip_hdr() and ipv6_hdr() before an skb_cow() call which can reallocate skb->head. Fix this possible UAF by initializing the local variables after the skb_cow() call. Remove skb_reset_network_header() calls which were not needed. Fixes: 4cb47a8644cc ("tunnels: PMTU discovery support for directly bridged IP packets") Signed-off-by: Eric Dumazet Reviewed-by: Stefano Brivio Link: https://patch.msgid.link/20260525201335.2361845-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/ipv4/ip_tunnel_core.c | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/ipv4/ip_tunnel_core.c b/net/ipv4/ip_tunnel_core.c index 2667f53482bd..c77a4c3fbe75 100644 --- a/net/ipv4/ip_tunnel_core.c +++ b/net/ipv4/ip_tunnel_core.c @@ -212,7 +212,7 @@ EXPORT_SYMBOL_GPL(iptunnel_handle_offloads); */ static int iptunnel_pmtud_build_icmp(struct sk_buff *skb, int mtu) { - const struct iphdr *iph = ip_hdr(skb); + const struct iphdr *iph; struct icmphdr *icmph; struct iphdr *niph; struct ethhdr eh; @@ -226,7 +226,6 @@ static int iptunnel_pmtud_build_icmp(struct sk_buff *skb, int mtu) skb_copy_bits(skb, skb_mac_offset(skb), &eh, ETH_HLEN); pskb_pull(skb, ETH_HLEN); - skb_reset_network_header(skb); err = pskb_trim(skb, 576 - sizeof(*niph) - sizeof(*icmph)); if (err) @@ -236,7 +235,7 @@ static int iptunnel_pmtud_build_icmp(struct sk_buff *skb, int mtu) err = skb_cow(skb, sizeof(*niph) + sizeof(*icmph) + ETH_HLEN); if (err) return err; - + iph = ip_hdr(skb); icmph = skb_push(skb, sizeof(*icmph)); *icmph = (struct icmphdr) { .type = ICMP_DEST_UNREACH, @@ -308,7 +307,7 @@ static int iptunnel_pmtud_check_icmp(struct sk_buff *skb, int mtu) */ static int iptunnel_pmtud_build_icmpv6(struct sk_buff *skb, int mtu) { - const struct ipv6hdr *ip6h = ipv6_hdr(skb); + const struct ipv6hdr *ip6h; struct icmp6hdr *icmp6h; struct ipv6hdr *nip6h; struct ethhdr eh; @@ -323,7 +322,6 @@ static int iptunnel_pmtud_build_icmpv6(struct sk_buff *skb, int mtu) skb_copy_bits(skb, skb_mac_offset(skb), &eh, ETH_HLEN); pskb_pull(skb, ETH_HLEN); - skb_reset_network_header(skb); err = pskb_trim(skb, IPV6_MIN_MTU - sizeof(*nip6h) - sizeof(*icmp6h)); if (err) @@ -334,6 +332,7 @@ static int iptunnel_pmtud_build_icmpv6(struct sk_buff *skb, int mtu) if (err) return err; + ip6h = ipv6_hdr(skb); icmp6h = skb_push(skb, sizeof(*icmp6h)); *icmp6h = (struct icmp6hdr) { .icmp6_type = ICMPV6_PKT_TOOBIG, -- cgit v1.2.3 From 509323077ef79a26ba0c60bb556e45c12c398b2d Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 22 May 2026 11:55:12 +0000 Subject: tunnels: do not assume transport header in iptunnel_pmtud_check_icmp() In some cases, iptunnel_pmtud_check_icmp() can be called while skb transport header is not set. This triggers an out-of-bound access, because (typeof(skb->transport_header))~0U is 65535. Access the icmp header based on IPv4 network header, after making sure icmp->type is present in skb linear part. Note that iptunnel_pmtud_check_icmpv6()) is fine. Fixes: 4cb47a8644cc ("tunnels: PMTU discovery support for directly bridged IP packets") Reported-by: Damiano Melotti Signed-off-by: Eric Dumazet Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20260522115512.1519110-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/ipv4/ip_tunnel_core.c | 13 ++++++++++--- 1 file changed, 10 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/ipv4/ip_tunnel_core.c b/net/ipv4/ip_tunnel_core.c index c77a4c3fbe75..d3c677e9bff2 100644 --- a/net/ipv4/ip_tunnel_core.c +++ b/net/ipv4/ip_tunnel_core.c @@ -280,7 +280,6 @@ static int iptunnel_pmtud_build_icmp(struct sk_buff *skb, int mtu) */ static int iptunnel_pmtud_check_icmp(struct sk_buff *skb, int mtu) { - const struct icmphdr *icmph = icmp_hdr(skb); const struct iphdr *iph = ip_hdr(skb); if (mtu < 576 || iph->frag_off != htons(IP_DF)) @@ -291,9 +290,17 @@ static int iptunnel_pmtud_check_icmp(struct sk_buff *skb, int mtu) ipv4_is_lbcast(iph->saddr) || ipv4_is_multicast(iph->saddr)) return 0; - if (iph->protocol == IPPROTO_ICMP && icmp_is_err(icmph->type)) - return 0; + if (iph->protocol == IPPROTO_ICMP) { + const struct icmphdr *icmph; + if (!pskb_network_may_pull(skb, iph->ihl * 4 + + offsetofend(struct icmphdr, type))) + return 0; + iph = ip_hdr(skb); + icmph = (void *)iph + iph->ihl * 4; + if (icmp_is_err(icmph->type)) + return 0; + } return iptunnel_pmtud_build_icmp(skb, mtu); } -- cgit v1.2.3 From dd433671fef381fdaf7b530c631e6b782d66e224 Mon Sep 17 00:00:00 2001 From: Qi Tang Date: Sat, 23 May 2026 22:32:45 +0800 Subject: ipv6: validate extension header length before copying to cmsg ip6_datagram_recv_specific_ctl() builds IPV6_{HOPOPTS,DSTOPTS,RTHDR} cmsgs (and their IPV6_2292* legacy counterparts) by trusting the on-wire hdrlen byte (ptr[1]) when computing the put_cmsg() length. The length was validated only at parse time (ipv6_parse_hopopts(), etc.). An nftables payload-write expression can rewrite hdrlen after parsing and before the skb reaches recvmsg; the write itself is in-bounds but put_cmsg() then reads up to ((hdrlen+1) << 3) = 2040 bytes from an 8-byte header. nftables is reachable from an unprivileged user namespace, so this is an unprivileged slab-out-of-bounds read: BUG: KASAN: slab-out-of-bounds in put_cmsg+0x3ac/0x540 put_cmsg+0x3ac/0x540 udpv6_recvmsg+0xca0/0x1250 sock_recvmsg+0xdf/0x190 ____sys_recvmsg+0x1b1/0x620 Add ipv6_get_exthdr_len() which validates that at least two bytes are accessible before reading the hdrlen field, then checks the computed length against skb_tail_pointer(skb), returning 0 on failure. Extension headers are kept in the linear skb area by pskb_may_pull() during input, so skb_tail_pointer() is the correct bound. Use ipv6_get_exthdr_len() at all non-AH call sites: the five standalone cmsg blocks (HbH, 2292HbH, 2292DSTOPTS x2, 2292RTHDR) and the three standard cases in the extension-header walk loop (DSTOPTS, ROUTING, default). AH retains an inline bounds check because its length formula differs ((ptr[1]+2)<<2). The walk loop also gets a pre-read bounds check at the top to validate ptr before any case accesses ptr[0] or ptr[1]. When the walk loop detects a corrupted header, return from the function instead of continuing to process later socket options. Cc: stable@vger.kernel.org Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: Qi Tang Reviewed-by: Willem de Bruijn Link: https://patch.msgid.link/20260523143245.2281415-1-tpluszz77@gmail.com Signed-off-by: Jakub Kicinski --- net/ipv6/datagram.c | 54 +++++++++++++++++++++++++++++++++++++++++++++-------- 1 file changed, 46 insertions(+), 8 deletions(-) (limited to 'net') diff --git a/net/ipv6/datagram.c b/net/ipv6/datagram.c index ca3605acb433..38d7b4845281 100644 --- a/net/ipv6/datagram.c +++ b/net/ipv6/datagram.c @@ -617,6 +617,18 @@ void ip6_datagram_recv_common_ctl(struct sock *sk, struct msghdr *msg, } } +static u16 ipv6_get_exthdr_len(const struct sk_buff *skb, const u8 *ptr) +{ + u16 len; + + if (ptr + 2 > skb_tail_pointer(skb)) + return 0; + + len = (ptr[1] + 1) << 3; + + return (len <= skb_tail_pointer(skb) - ptr) ? len : 0; +} + void ip6_datagram_recv_specific_ctl(struct sock *sk, struct msghdr *msg, struct sk_buff *skb) { @@ -643,7 +655,10 @@ void ip6_datagram_recv_specific_ctl(struct sock *sk, struct msghdr *msg, /* HbH is allowed only once */ if (np->rxopt.bits.hopopts && (opt->flags & IP6SKB_HOPBYHOP)) { u8 *ptr = nh + sizeof(struct ipv6hdr); - put_cmsg(msg, SOL_IPV6, IPV6_HOPOPTS, (ptr[1]+1)<<3, ptr); + u16 len = ipv6_get_exthdr_len(skb, ptr); + + if (len) + put_cmsg(msg, SOL_IPV6, IPV6_HOPOPTS, len, ptr); } if (opt->lastopt && @@ -664,26 +679,37 @@ void ip6_datagram_recv_specific_ctl(struct sock *sk, struct msghdr *msg, unsigned int len; u8 *ptr = nh + off; + if (ptr + 2 > skb_tail_pointer(skb)) + return; + switch (nexthdr) { case IPPROTO_DSTOPTS: nexthdr = ptr[0]; - len = (ptr[1] + 1) << 3; + len = ipv6_get_exthdr_len(skb, ptr); + if (!len) + return; if (np->rxopt.bits.dstopts) put_cmsg(msg, SOL_IPV6, IPV6_DSTOPTS, len, ptr); break; case IPPROTO_ROUTING: nexthdr = ptr[0]; - len = (ptr[1] + 1) << 3; + len = ipv6_get_exthdr_len(skb, ptr); + if (!len) + return; if (np->rxopt.bits.srcrt) put_cmsg(msg, SOL_IPV6, IPV6_RTHDR, len, ptr); break; case IPPROTO_AH: nexthdr = ptr[0]; len = (ptr[1] + 2) << 2; + if (ptr + len > skb_tail_pointer(skb)) + return; break; default: nexthdr = ptr[0]; - len = (ptr[1] + 1) << 3; + len = ipv6_get_exthdr_len(skb, ptr); + if (!len) + return; break; } @@ -705,19 +731,31 @@ void ip6_datagram_recv_specific_ctl(struct sock *sk, struct msghdr *msg, } if (np->rxopt.bits.ohopopts && (opt->flags & IP6SKB_HOPBYHOP)) { u8 *ptr = nh + sizeof(struct ipv6hdr); - put_cmsg(msg, SOL_IPV6, IPV6_2292HOPOPTS, (ptr[1]+1)<<3, ptr); + u16 len = ipv6_get_exthdr_len(skb, ptr); + + if (len) + put_cmsg(msg, SOL_IPV6, IPV6_2292HOPOPTS, len, ptr); } if (np->rxopt.bits.odstopts && opt->dst0) { u8 *ptr = nh + opt->dst0; - put_cmsg(msg, SOL_IPV6, IPV6_2292DSTOPTS, (ptr[1]+1)<<3, ptr); + u16 len = ipv6_get_exthdr_len(skb, ptr); + + if (len) + put_cmsg(msg, SOL_IPV6, IPV6_2292DSTOPTS, len, ptr); } if (np->rxopt.bits.osrcrt && opt->srcrt) { struct ipv6_rt_hdr *rthdr = (struct ipv6_rt_hdr *)(nh + opt->srcrt); - put_cmsg(msg, SOL_IPV6, IPV6_2292RTHDR, (rthdr->hdrlen+1) << 3, rthdr); + u16 len = ipv6_get_exthdr_len(skb, (u8 *)rthdr); + + if (len) + put_cmsg(msg, SOL_IPV6, IPV6_2292RTHDR, len, rthdr); } if (np->rxopt.bits.odstopts && opt->dst1) { u8 *ptr = nh + opt->dst1; - put_cmsg(msg, SOL_IPV6, IPV6_2292DSTOPTS, (ptr[1]+1)<<3, ptr); + u16 len = ipv6_get_exthdr_len(skb, ptr); + + if (len) + put_cmsg(msg, SOL_IPV6, IPV6_2292DSTOPTS, len, ptr); } if (np->rxopt.bits.rxorigdstaddr) { struct sockaddr_in6 sin6; -- cgit v1.2.3 From bfa9d28960ed677d556bdf097073bc3129686229 Mon Sep 17 00:00:00 2001 From: Pavitra Jha Date: Thu, 21 May 2026 04:04:14 -0400 Subject: Bluetooth: hci_conn: Fix memory leak in hci_le_big_terminate() hci_le_big_terminate() allocates iso_list_data via kzalloc_obj but returns 0 without freeing it when neither pa_sync_term nor big_sync_term flags are set after evaluating the PA and BIG sync connection state. This early-return path was introduced when hci_le_big_terminate() was refactored to take struct hci_conn instead of raw u8 parameters, adding PA/BIG flag evaluation logic. The existing kfree() on hci_cmd_sync_queue failure does not cover this path. Fixes: a7bcffc673de ("Bluetooth: Add PA_LINK to distinguish BIG sync and PA sync connections") Cc: stable@vger.kernel.org Signed-off-by: Pavitra Jha Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/hci_conn.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c index 17b46ad6a349..54eabaa46960 100644 --- a/net/bluetooth/hci_conn.c +++ b/net/bluetooth/hci_conn.c @@ -870,8 +870,10 @@ static int hci_le_big_terminate(struct hci_dev *hdev, struct hci_conn *conn) d->big_sync_term = true; } - if (!d->pa_sync_term && !d->big_sync_term) + if (!d->pa_sync_term && !d->big_sync_term) { + kfree(d); return 0; + } ret = hci_cmd_sync_queue(hdev, big_terminate_sync, d, terminate_big_destroy); -- cgit v1.2.3 From 9dbd84990394c51f5cee1e8871bb5ff8af5ed939 Mon Sep 17 00:00:00 2001 From: Siwei Zhang Date: Wed, 20 May 2026 22:30:36 -0400 Subject: Bluetooth: L2CAP: fix chan ref leak in l2cap_chan_timeout() on !conn __set_chan_timer() takes a l2cap_chan reference via l2cap_chan_hold() before scheduling the delayed work. The normal path in l2cap_chan_timeout() drops this reference with l2cap_chan_put() at the end, but the early return when chan->conn is NULL skips the put, leaking the reference. Add the missing l2cap_chan_put() before the early return. Fixes: adf0398cee86 ("Bluetooth: l2cap: fix null-ptr-deref in l2cap_chan_timeout") Cc: stable@vger.kernel.org Signed-off-by: Siwei Zhang Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/l2cap_core.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c index fdccd62ccca8..5668c92b3f58 100644 --- a/net/bluetooth/l2cap_core.c +++ b/net/bluetooth/l2cap_core.c @@ -411,8 +411,10 @@ static void l2cap_chan_timeout(struct work_struct *work) BT_DBG("chan %p state %s", chan, state_to_string(chan->state)); - if (!conn) + if (!conn) { + l2cap_chan_put(chan); return; + } mutex_lock(&conn->lock); /* __set_chan_timer() calls l2cap_chan_hold(chan) while scheduling -- cgit v1.2.3 From 8c8e620467a7b51562dbcefbd1f09f288d7d710d Mon Sep 17 00:00:00 2001 From: Siwei Zhang Date: Wed, 20 May 2026 22:12:20 -0400 Subject: Bluetooth: L2CAP: use chan timer to close channels in cleanup_listen() l2cap_chan_close() removes the channel from conn->chan_l, which must be done under conn->lock. cleanup_listen() runs under the parent sk_lock, so acquiring conn->lock would invert the established conn->lock -> chan->lock -> sk_lock order. Instead of calling l2cap_chan_close() directly, schedule l2cap_chan_timeout with delay 0 to close the channel asynchronously. The timeout handler already acquires conn->lock and chan->lock in the correct order. The timer is only armed when chan->conn is still set: if it is already NULL, l2cap_conn_del() has already processed this channel (l2cap_chan_del + l2cap_sock_teardown_cb + l2cap_sock_close_cb), so there is nothing left to do. If l2cap_conn_del() races in after the timer is armed, __clear_chan_timer() inside l2cap_chan_del() cancels it; if the timer has already fired, the handler returns harmlessly because chan->conn was cleared. Fixes: 3df91ea20e74 ("Bluetooth: Revert to mutexes from RCU list") Cc: # 0b58004: Bluetooth: fix UAF in l2cap_sock_cleanup_listen() vs l2cap_conn_del() Signed-off-by: Siwei Zhang Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/l2cap_sock.c | 16 +++++++++------- 1 file changed, 9 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/net/bluetooth/l2cap_sock.c b/net/bluetooth/l2cap_sock.c index b34e7da8d906..c138aa4ae266 100644 --- a/net/bluetooth/l2cap_sock.c +++ b/net/bluetooth/l2cap_sock.c @@ -1499,6 +1499,10 @@ static void l2cap_sock_cleanup_listen(struct sock *parent) * pin it (hold_unless_zero() additionally skips a chan already past * its last reference). We then drop the sk lock before taking * chan->lock, so sk and chan locks are never held together. + * + * Since we cannot call l2cap_chan_close() without conn->lock, + * schedule l2cap_chan_timeout to close the channel; it already + * acquires conn->lock -> chan->lock in the correct order. */ while ((sk = bt_accept_dequeue(parent, NULL))) { struct l2cap_chan *chan; @@ -1516,14 +1520,12 @@ static void l2cap_sock_cleanup_listen(struct sock *parent) state_to_string(chan->state)); l2cap_chan_lock(chan); - __clear_chan_timer(chan); - l2cap_chan_close(chan, ECONNRESET); - /* l2cap_conn_del() may already have killed this socket - * (it sets SOCK_DEAD); skip the duplicate to avoid a - * double sock_put()/l2cap_chan_put(). + /* Since we cannot call l2cap_chan_close() without + * conn->lock, schedule its timer to trigger the close + * and cleanup of this channel. */ - if (!sock_flag(sk, SOCK_DEAD)) - l2cap_sock_kill(sk); + if (chan->conn) + __set_chan_timer(chan, 0); l2cap_chan_unlock(chan); l2cap_chan_put(chan); -- cgit v1.2.3 From 2a3ac9ee11dbb9845f3947cef4a79dba658cf6f6 Mon Sep 17 00:00:00 2001 From: Muhammad Bilal Date: Wed, 20 May 2026 18:56:43 -0400 Subject: Bluetooth: HIDP: fix missing length checks in hidp_input_report() hidp_input_report() reads keyboard and mouse payload data from an skb without first verifying that skb->len contains enough data. hidp_recv_intr_frame() pulls the 1-byte HIDP header before dispatching to hidp_input_report(). If a paired device sends a truncated packet, the handler reads beyond the valid skb data, resulting in an out-of-bounds read of skb data. The OOB bytes may be interpreted as phantom key presses or spurious mouse movement. Replace the open-coded length tracking and pointer arithmetic with skb_pull_data() calls. skb_pull_data() returns NULL if the requested bytes are not present, eliminating the need for a manual size variable and the separate skb->len guard. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Cc: stable@vger.kernel.org Signed-off-by: Muhammad Bilal Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/hidp/core.c | 23 ++++++++++++++++++----- 1 file changed, 18 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/bluetooth/hidp/core.c b/net/bluetooth/hidp/core.c index 976f91eeb745..70344bd3248a 100644 --- a/net/bluetooth/hidp/core.c +++ b/net/bluetooth/hidp/core.c @@ -179,12 +179,21 @@ static void hidp_input_report(struct hidp_session *session, struct sk_buff *skb) { struct input_dev *dev = session->input; unsigned char *keys = session->keys; - unsigned char *udata = skb->data + 1; - signed char *sdata = skb->data + 1; - int i, size = skb->len - 1; + unsigned char *udata; + signed char *sdata; + u8 *hdr; + int i; + + hdr = skb_pull_data(skb, 1); + if (!hdr) + return; - switch (skb->data[0]) { + switch (*hdr) { case 0x01: /* Keyboard report */ + udata = skb_pull_data(skb, 8); + if (!udata) + break; + for (i = 0; i < 8; i++) input_report_key(dev, hidp_keycode[i + 224], (udata[0] >> i) & 1); @@ -213,6 +222,10 @@ static void hidp_input_report(struct hidp_session *session, struct sk_buff *skb) break; case 0x02: /* Mouse report */ + sdata = skb_pull_data(skb, 3); + if (!sdata) + break; + input_report_key(dev, BTN_LEFT, sdata[0] & 0x01); input_report_key(dev, BTN_RIGHT, sdata[0] & 0x02); input_report_key(dev, BTN_MIDDLE, sdata[0] & 0x04); @@ -222,7 +235,7 @@ static void hidp_input_report(struct hidp_session *session, struct sk_buff *skb) input_report_rel(dev, REL_X, sdata[1]); input_report_rel(dev, REL_Y, sdata[2]); - if (size > 3) + if (skb->len > 0) input_report_rel(dev, REL_WHEEL, sdata[3]); break; } -- cgit v1.2.3 From 3c40d381ce04f9575a5d8b542898183c3b4b38dc Mon Sep 17 00:00:00 2001 From: Zhao Dongdong Date: Tue, 26 May 2026 11:21:39 +0800 Subject: Bluetooth: 6lowpan: check skb_clone() return value in send_mcast_pkt() The skb_clone() function can return NULL if memory allocation fails. send_mcast_pkt() calls skb_clone() without checking the return value, which can lead to a NULL pointer dereference in send_pkt() when it dereferences skb->data. Add a NULL check after skb_clone() and skip the peer if the clone fails. Fixes: 18722c247023 ("Bluetooth: Enable 6LoWPAN support for BT LE devices") Signed-off-by: Zhao Dongdong Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/6lowpan.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'net') diff --git a/net/bluetooth/6lowpan.c b/net/bluetooth/6lowpan.c index 2f03b780b40d..960a19b3e26d 100644 --- a/net/bluetooth/6lowpan.c +++ b/net/bluetooth/6lowpan.c @@ -486,6 +486,8 @@ static int send_mcast_pkt(struct sk_buff *skb, struct net_device *netdev) int ret; local_skb = skb_clone(skb, GFP_ATOMIC); + if (!local_skb) + continue; BT_DBG("xmit %s to %pMR type %u IP %pI6c chan %p", netdev->name, -- cgit v1.2.3 From bfea6091e0fffb270c20e74384b660910277eb6c Mon Sep 17 00:00:00 2001 From: Doruk Tan Ozturk Date: Mon, 25 May 2026 18:24:38 +0200 Subject: Bluetooth: hci_sync: fix UAF in hci_le_create_cis_sync hci_le_create_cis_sync() dereferences conn->conn_timeout after releasing both rcu_read_lock() and hci_dev_lock(hdev). The conn pointer was obtained from an RCU-protected iteration over hdev->conn_hash.list and is not valid once these locks are dropped. A concurrent disconnect can free the hci_conn between the unlock and the dereference, causing a use-after-free read. The cancellation mechanism in hci_conn_del() cannot prevent this because hci_le_create_cis_pending() queues hci_create_cis_sync with data=NULL: hci_cmd_sync_queue(hdev, hci_create_cis_sync, NULL, NULL); While hci_conn_del() dequeues with data=conn: hci_cmd_sync_dequeue(hdev, NULL, conn, NULL); Since NULL != conn, the lookup in _hci_cmd_sync_lookup_entry() never matches, and the pending work item is not cancelled. Fix this by saving conn->conn_timeout into a local variable while the locks are still held, so the stale conn pointer is never dereferenced after unlock. This is the same class of bug as the one fixed by commit 035c25007c9e ("Bluetooth: hci_sync: Fix UAF on le_read_features_complete") which addressed the identical pattern in a different function. This vulnerability was identified using 0sec.ai, an open-source automated security auditing platform (https://github.com/0sec-labs). Fixes: c09b80be6ffc ("Bluetooth: hci_conn: Fix not waiting for HCI_EVT_LE_CIS_ESTABLISHED") Cc: stable@vger.kernel.org Reported-by: Doruk Tan Ozturk Signed-off-by: Doruk Tan Ozturk Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/hci_sync.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c index aff8562a8690..1faf8df6d159 100644 --- a/net/bluetooth/hci_sync.c +++ b/net/bluetooth/hci_sync.c @@ -6699,6 +6699,7 @@ int hci_le_create_cis_sync(struct hci_dev *hdev) DEFINE_FLEX(struct hci_cp_le_create_cis, cmd, cis, num_cis, 0x1f); size_t aux_num_cis = 0; struct hci_conn *conn; + u16 timeout = 0; u8 cig = BT_ISO_QOS_CIG_UNSET; /* The spec allows only one pending LE Create CIS command at a time. If @@ -6769,6 +6770,7 @@ int hci_le_create_cis_sync(struct hci_dev *hdev) set_bit(HCI_CONN_CREATE_CIS, &conn->flags); cis->acl_handle = cpu_to_le16(conn->parent->handle); cis->cis_handle = cpu_to_le16(conn->handle); + timeout = conn->conn_timeout; aux_num_cis++; if (aux_num_cis >= cmd->num_cis) @@ -6788,7 +6790,7 @@ done: return __hci_cmd_sync_status_sk(hdev, HCI_OP_LE_CREATE_CIS, struct_size(cmd, cis, cmd->num_cis), cmd, HCI_EVT_LE_CIS_ESTABLISHED, - conn->conn_timeout, NULL); + timeout, NULL); } int hci_le_remove_cig_sync(struct hci_dev *hdev, u8 handle) -- cgit v1.2.3 From 5eec4427b89c2fb2beac54920101e55a2f1c0c21 Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Tue, 26 May 2026 09:48:16 +0300 Subject: bridge: Fix sleep in atomic context in netlink path Since the introduction of the netlink configuration path for bridge ports in commit 25c71c75ac87 ("bridge: bridge port parameters over netlink"), br_setport() was always called with the bridge lock held around it. Back then this decision made sense: The bridge lock protects the STP state of the bridge and its ports and at that time the function only processed three STP related netlink attributes (cost, priority and state). Nowadays, br_setport() processes a lot more attributes and most of them do not need the bridge lock: * Bridge flags: Only require RTNL. Read locklessly by the data path. Annotations can be added in net-next. * FDB port flushing: Only requires the FDB lock. * Multicast attributes: Only require the multicast lock. * Group forward mask: Only requires RTNL. Read locklessly by the data path. Annotations can be added in net-next. * Backup port and NHID: Only require RTNL. Read locklessly by the data path. This is a problem as the bridge calls dev_set_promiscuity() when certain bridge port flags change and this function can sleep since the commit cited below, resulting in a splat such as [1]. Fix this by reducing the scope of the bridge lock and only take it when processing the three STP related attributes that require it. This is consistent with the multicast attributes where each attribute acquires the multicast lock instead of having one critical section for all relevant attributes. [1] BUG: sleeping function called from invalid context at net/core/dev_addr_lists.c:1262 in_atomic(): 1, irqs_disabled(): 0, non_block: 0, pid: 356, name: bridge preempt_count: 201, expected: 0 RCU nest depth: 0, expected: 0 2 locks held by bridge/356: #0: ffffffff919473a0 (rtnl_mutex){+.+.}-{4:4}, at: rtnetlink_rcv_msg (net/core/rtnetlink.c:80 net/core/rtnetlink.c:7002) #1: ffff888115072d58 (&br->lock){+...}-{3:3}, at: br_setlink (./include/linux/spinlock.h:348 net/bridge/br_netlink.c:1117) Preemption disabled at: 0x0 Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011 Call Trace: dump_stack_lvl (lib/dump_stack.c:94 lib/dump_stack.c:120) __might_resched.cold (kernel/sched/core.c:9163) netif_rx_mode_run (net/core/dev_addr_lists.c:1262) netif_rx_mode_sync (net/core/dev_addr_lists.c:1428) dev_set_promiscuity (net/core/dev_api.c:289) br_manage_promisc (net/bridge/br_if.c:135 net/bridge/br_if.c:172) br_port_flags_change (net/bridge/br_if.c:242 net/bridge/br_if.c:747) br_setport (net/bridge/br_netlink.c:1000) br_setlink (net/bridge/br_netlink.c:1118) rtnl_bridge_setlink (net/core/rtnetlink.c:5572) rtnetlink_rcv_msg (net/core/rtnetlink.c:7005) netlink_rcv_skb (net/netlink/af_netlink.c:2550) netlink_unicast (net/netlink/af_netlink.c:1318 net/netlink/af_netlink.c:1344) netlink_sendmsg (net/netlink/af_netlink.c:1894) __sock_sendmsg (net/socket.c:787 (discriminator 4) net/socket.c:802 (discriminator 4)) ____sys_sendmsg (net/socket.c:2698) ___sys_sendmsg (net/socket.c:2752) __sys_sendmsg (net/socket.c:2784) do_syscall_64 (arch/x86/entry/syscall_64.c:63 arch/x86/entry/syscall_64.c:94) entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:121) Fixes: 78cd408356fe ("net: add missing instance lock to dev_set_promiscuity") Reviewed-by: Nikolay Aleksandrov Signed-off-by: Ido Schimmel Link: https://patch.msgid.link/20260526064818.272516-2-idosch@nvidia.com Signed-off-by: Jakub Kicinski --- net/bridge/br_netlink.c | 17 +++++++---------- 1 file changed, 7 insertions(+), 10 deletions(-) (limited to 'net') diff --git a/net/bridge/br_netlink.c b/net/bridge/br_netlink.c index c04a4d0889ae..b9591dd755f9 100644 --- a/net/bridge/br_netlink.c +++ b/net/bridge/br_netlink.c @@ -1000,19 +1000,25 @@ static int br_setport(struct net_bridge_port *p, struct nlattr *tb[], br_port_flags_change(p, changed_mask); if (tb[IFLA_BRPORT_COST]) { + spin_lock_bh(&p->br->lock); err = br_stp_set_path_cost(p, nla_get_u32(tb[IFLA_BRPORT_COST])); + spin_unlock_bh(&p->br->lock); if (err) return err; } if (tb[IFLA_BRPORT_PRIORITY]) { + spin_lock_bh(&p->br->lock); err = br_stp_set_port_priority(p, nla_get_u16(tb[IFLA_BRPORT_PRIORITY])); + spin_unlock_bh(&p->br->lock); if (err) return err; } if (tb[IFLA_BRPORT_STATE]) { + spin_lock_bh(&p->br->lock); err = br_set_port_state(p, nla_get_u8(tb[IFLA_BRPORT_STATE])); + spin_unlock_bh(&p->br->lock); if (err) return err; } @@ -1114,9 +1120,7 @@ int br_setlink(struct net_device *dev, struct nlmsghdr *nlh, u16 flags, if (err) return err; - spin_lock_bh(&p->br->lock); err = br_setport(p, tb, extack); - spin_unlock_bh(&p->br->lock); } else { /* Binary compatibility with old RSTP */ if (nla_len(protinfo) < sizeof(u8)) @@ -1203,17 +1207,10 @@ static int br_port_slave_changelink(struct net_device *brdev, struct nlattr *data[], struct netlink_ext_ack *extack) { - struct net_bridge *br = netdev_priv(brdev); - int ret; - if (!data) return 0; - spin_lock_bh(&br->lock); - ret = br_setport(br_port_get_rtnl(dev), data, extack); - spin_unlock_bh(&br->lock); - - return ret; + return br_setport(br_port_get_rtnl(dev), data, extack); } static int br_port_fill_slave_info(struct sk_buff *skb, -- cgit v1.2.3 From 6d34594cc619d0d4b07d5afcad8b5984f3526dcf Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Tue, 26 May 2026 09:48:17 +0300 Subject: bridge: Fix sleep in atomic context in sysfs path Since the start of the git history, brport_store() always acquired the bridge lock. Back then this decision made sense: The bridge lock protects the STP state of the bridge and its ports and at that time the function was only used by two STP related attributes (cost and priority). Nowadays, brport_store() processes a lot more attributes and most of them do not need the bridge lock: * Bridge flags: Only require RTNL. Read locklessly by the data path. Annotations can be added in net-next. * FDB port flushing: Only requires the FDB lock. * Multicast attributes: Only require the multicast lock. * Group forward mask: Only requires RTNL. Read locklessly by the data path. Annotations can be added in net-next. * Backup port: Only requires RTNL. Read locklessly by the data path. This is a problem as the bridge calls dev_set_promiscuity() when certain bridge port flags change and this function can sleep since the commit cited below, resulting in a splat such as [1]. Fix this by reducing the scope of the bridge lock and only take it when processing the two STP related attributes that require it. Remove the now stale comment from br_switchdev_set_port_flag(). The SWITCHDEV_F_DEFER flag can be removed in net-next. [1] BUG: sleeping function called from invalid context at net/core/dev_addr_lists.c:1262 in_atomic(): 1, irqs_disabled(): 0, non_block: 0, pid: 372, name: bash preempt_count: 201, expected: 0 RCU nest depth: 0, expected: 0 5 locks held by bash/372: #0: ffff88810c51c3f0 (sb_writers#7){.+.+}-{0:0}, at: ksys_write (fs/read_write.c:740) #1: ffff888115ce9480 (&of->mutex){+.+.}-{4:4}, at: kernfs_fop_write_iter (fs/kernfs/file.c:343) #2: ffff88810b9fd330 (kn->active#37){.+.+}-{0:0}, at: kernfs_fop_write_iter (fs/kernfs/file.c:80 fs/kernfs/file.c:344) #3: ffffffffa59473a0 (rtnl_mutex){+.+.}-{4:4}, at: brport_store (net/bridge/br_sysfs_if.c:326) #4: ffff8881099d2d58 (&br->lock){+...}-{3:3}, at: brport_store (./include/linux/spinlock.h:348 net/bridge/br_sysfs_if.c:345) Preemption disabled at: 0x0 Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011 Call Trace: dump_stack_lvl (lib/dump_stack.c:94 lib/dump_stack.c:120) __might_resched.cold (kernel/sched/core.c:9163) netif_rx_mode_run (net/core/dev_addr_lists.c:1262) netif_rx_mode_sync (net/core/dev_addr_lists.c:1428) dev_set_promiscuity (net/core/dev_api.c:289) br_manage_promisc (net/bridge/br_if.c:135 net/bridge/br_if.c:172) br_port_flags_change (net/bridge/br_if.c:242 net/bridge/br_if.c:747) store_learning (net/bridge/br_sysfs_if.c:79 net/bridge/br_sysfs_if.c:235) brport_store (net/bridge/br_sysfs_if.c:346) kernfs_fop_write_iter (fs/kernfs/file.c:352) new_sync_write (fs/read_write.c:595) vfs_write (fs/read_write.c:688) ksys_write (fs/read_write.c:740) do_syscall_64 (arch/x86/entry/syscall_64.c:63 arch/x86/entry/syscall_64.c:94) entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:121) Fixes: 78cd408356fe ("net: add missing instance lock to dev_set_promiscuity") Reviewed-by: Nikolay Aleksandrov Signed-off-by: Ido Schimmel Link: https://patch.msgid.link/20260526064818.272516-3-idosch@nvidia.com Signed-off-by: Jakub Kicinski --- net/bridge/br_switchdev.c | 1 - net/bridge/br_sysfs_if.c | 30 ++++++++++++++++++++++-------- 2 files changed, 22 insertions(+), 9 deletions(-) (limited to 'net') diff --git a/net/bridge/br_switchdev.c b/net/bridge/br_switchdev.c index 18b558a931ad..ee3ad9dfbab9 100644 --- a/net/bridge/br_switchdev.c +++ b/net/bridge/br_switchdev.c @@ -99,7 +99,6 @@ int br_switchdev_set_port_flag(struct net_bridge_port *p, attr.u.brport_flags.val = flags; attr.u.brport_flags.mask = mask; - /* We run from atomic context here */ err = call_switchdev_notifiers(SWITCHDEV_PORT_ATTR_SET, p->dev, &info.info, extack); err = notifier_to_errno(err); diff --git a/net/bridge/br_sysfs_if.c b/net/bridge/br_sysfs_if.c index 1f57c36a7fc0..d6df81fa0d13 100644 --- a/net/bridge/br_sysfs_if.c +++ b/net/bridge/br_sysfs_if.c @@ -86,16 +86,34 @@ static ssize_t show_path_cost(struct net_bridge_port *p, char *buf) return sysfs_emit(buf, "%d\n", p->path_cost); } -static BRPORT_ATTR(path_cost, 0644, - show_path_cost, br_stp_set_path_cost); +static int store_path_cost(struct net_bridge_port *p, unsigned long v) +{ + int ret; + + spin_lock_bh(&p->br->lock); + ret = br_stp_set_path_cost(p, v); + spin_unlock_bh(&p->br->lock); + return ret; +} + +static BRPORT_ATTR(path_cost, 0644, show_path_cost, store_path_cost); static ssize_t show_priority(struct net_bridge_port *p, char *buf) { return sysfs_emit(buf, "%d\n", p->priority); } -static BRPORT_ATTR(priority, 0644, - show_priority, br_stp_set_port_priority); +static int store_priority(struct net_bridge_port *p, unsigned long v) +{ + int ret; + + spin_lock_bh(&p->br->lock); + ret = br_stp_set_port_priority(p, v); + spin_unlock_bh(&p->br->lock); + return ret; +} + +static BRPORT_ATTR(priority, 0644, show_priority, store_priority); static ssize_t show_designated_root(struct net_bridge_port *p, char *buf) { @@ -334,17 +352,13 @@ static ssize_t brport_store(struct kobject *kobj, ret = -ENOMEM; goto out_unlock; } - spin_lock_bh(&p->br->lock); ret = brport_attr->store_raw(p, buf_copy); - spin_unlock_bh(&p->br->lock); kfree(buf_copy); } else if (brport_attr->store) { val = simple_strtoul(buf, &endp, 0); if (endp == buf) goto out_unlock; - spin_lock_bh(&p->br->lock); ret = brport_attr->store(p, val); - spin_unlock_bh(&p->br->lock); } if (!ret) { -- cgit v1.2.3 From 7281b096b072f6c6e30420e3467d738f2e4c4b57 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 26 May 2026 08:35:24 -0700 Subject: ethtool: coalesce: cap profile updates at NET_DIM_PARAMS_NUM_PROFILES ethnl_update_profile() walks the ETHTOOL_A_PROFILE_IRQ_MODERATION nest list with an index 'i' and writes new_profile[i++] without bounding i. The destination is kmemdup()'d at NET_DIM_PARAMS_NUM_PROFILES entries (5), but the Netlink nest count is entirely user-controlled. Netlink policies do not have support for constraining the number of nested entries (or number of multi-attr entries). Fixes: f750dfe825b9 ("ethtool: provide customized dim profile management") Reviewed-by: Maxime Chevallier Link: https://patch.msgid.link/20260526153533.2779187-2-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/coalesce.c | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'net') diff --git a/net/ethtool/coalesce.c b/net/ethtool/coalesce.c index 1e2c5c7048a8..e73fc3e5a02b 100644 --- a/net/ethtool/coalesce.c +++ b/net/ethtool/coalesce.c @@ -472,6 +472,12 @@ static int ethnl_update_profile(struct net_device *dev, nla_for_each_nested_type(nest, ETHTOOL_A_PROFILE_IRQ_MODERATION, nests, rem) { + if (i >= NET_DIM_PARAMS_NUM_PROFILES) { + NL_SET_BAD_ATTR(extack, nest); + ret = -E2BIG; + goto err_out; + } + ret = nla_parse_nested(tb, len_irq_moder - 1, nest, coalesce_irq_moderation_policy, extack); -- cgit v1.2.3 From a888bbd43940cada72f7686337741ce86d1cf869 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 26 May 2026 08:35:25 -0700 Subject: ethtool: tsconfig: fix reply error handling A couple of trivial bugs in error handling in tsconfig_send_reply(). If we failed to allocate rskb we need to set the error. If we did allocate it but failed to send it - we need to remember to free it. Fixes: 6e9e2eed4f39 ("net: ethtool: Add support for tsconfig command to get/set hwtstamp config") Reviewed-by: Vadim Fedorenko Reviewed-by: Kory Maincent Link: https://patch.msgid.link/20260526153533.2779187-3-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/tsconfig.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/ethtool/tsconfig.c b/net/ethtool/tsconfig.c index e4f518e49d4c..e9db4ee2299d 100644 --- a/net/ethtool/tsconfig.c +++ b/net/ethtool/tsconfig.c @@ -224,16 +224,21 @@ static int tsconfig_send_reply(struct net_device *dev, struct genl_info *info) reply_len = ret + ethnl_reply_header_size(); rskb = ethnl_reply_init(reply_len, dev, ETHTOOL_MSG_TSCONFIG_SET_REPLY, ETHTOOL_A_TSCONFIG_HEADER, info, &reply_payload); - if (!rskb) + if (!rskb) { + ret = -ENOMEM; goto err_cleanup; + } ret = tsconfig_fill_reply(rskb, &req_info->base, &reply_data->base); if (ret < 0) - goto err_cleanup; + goto err_free_msg; genlmsg_end(rskb, reply_payload); ret = genlmsg_reply(rskb, info); + rskb = NULL; +err_free_msg: + nlmsg_free(rskb); err_cleanup: kfree(reply_data); kfree(req_info); -- cgit v1.2.3 From 596c51ed9e125b12c4d85b4530dfd4c7847634b7 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 26 May 2026 08:35:26 -0700 Subject: ethtool: linkstate: fix unbalanced ethnl_ops_complete() on PHY lookup error linkstate_prepare_data() calls ethnl_req_get_phydev() before ethnl_ops_begin(), but routes its error path through "goto out" which calls ethnl_ops_complete(). Fixes: fe55b1d401c6 ("ethtool: linkstate: migrate linkstate functions to support multi-PHY setups") Reviewed-by: Maxime Chevallier Link: https://patch.msgid.link/20260526153533.2779187-4-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/linkstate.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/ethtool/linkstate.c b/net/ethtool/linkstate.c index 8a5985fd7712..24569e92942c 100644 --- a/net/ethtool/linkstate.c +++ b/net/ethtool/linkstate.c @@ -106,10 +106,8 @@ static int linkstate_prepare_data(const struct ethnl_req_info *req_base, phydev = ethnl_req_get_phydev(req_base, tb, ETHTOOL_A_LINKSTATE_HEADER, info->extack); - if (IS_ERR(phydev)) { - ret = PTR_ERR(phydev); - goto out; - } + if (IS_ERR(phydev)) + return PTR_ERR(phydev); ret = ethnl_ops_begin(dev); if (ret < 0) -- cgit v1.2.3 From ab5bf428fb6bd361163c7247b92750d1d24ca2ed Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 26 May 2026 08:35:27 -0700 Subject: ethtool: pse-pd: fix missing ethnl_ops_complete() pse_prepare_data() is missing ethnl_ops_complete() if ethnl_req_get_phydev() returned an error. Move getting phydev up so that we don't have to worry about this (similar order to linkstate_prepare_data()). Note that phydev may still be NULL (this is checked in pse_get_pse_attributes()), the goal isn't really to avoid the _begin() / _complete() calls, only to simplify the error handling. While at it propagate the original error. Why this code overrides the error with -ENODEV but !phydev generates -EOPNOTSUPP is unclear to me... Fixes: 31748765bed3 ("net: ethtool: pse-pd: Target the command to the requested PHY") Reviewed-by: Maxime Chevallier Link: https://patch.msgid.link/20260526153533.2779187-5-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/pse-pd.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/ethtool/pse-pd.c b/net/ethtool/pse-pd.c index 2eb9bdc2dcb9..757c9e0cc856 100644 --- a/net/ethtool/pse-pd.c +++ b/net/ethtool/pse-pd.c @@ -62,14 +62,14 @@ static int pse_prepare_data(const struct ethnl_req_info *req_base, struct phy_device *phydev; int ret; - ret = ethnl_ops_begin(dev); - if (ret < 0) - return ret; - phydev = ethnl_req_get_phydev(req_base, tb, ETHTOOL_A_PSE_HEADER, info->extack); if (IS_ERR(phydev)) - return -ENODEV; + return PTR_ERR(phydev); + + ret = ethnl_ops_begin(dev); + if (ret < 0) + return ret; ret = pse_get_pse_attributes(phydev, info->extack, data); -- cgit v1.2.3 From 6386bd772de64e6760306eb91c7e86163af6c22f Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 26 May 2026 08:35:28 -0700 Subject: ethtool: tsconfig: fix missing ethnl_ops_complete() tsconfig_prepare_data() calls ethnl_ops_begin(), we need to call ethnl_ops_complete() before returning the error. Fixes: 6e9e2eed4f39 ("net: ethtool: Add support for tsconfig command to get/set hwtstamp config") Reviewed-by: Vadim Fedorenko Reviewed-by: Kory Maincent Link: https://patch.msgid.link/20260526153533.2779187-6-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/tsconfig.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/ethtool/tsconfig.c b/net/ethtool/tsconfig.c index e9db4ee2299d..fc4f93cfa459 100644 --- a/net/ethtool/tsconfig.c +++ b/net/ethtool/tsconfig.c @@ -69,8 +69,10 @@ static int tsconfig_prepare_data(const struct ethnl_req_info *req_base, if (ret) goto out; - if (ts_info.phc_index == -1) - return -ENODEV; + if (ts_info.phc_index == -1) { + ret = -ENODEV; + goto out; + } data->hwprov_desc.index = ts_info.phc_index; data->hwprov_desc.qualifier = ts_info.phc_qualifier; -- cgit v1.2.3 From 1de405699c62c3a9544bcdcfb9eff8a01cfc7582 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 26 May 2026 08:35:29 -0700 Subject: ethtool: tsinfo: fix uninitialized stats on the by-PHC path tsinfo_prepare_data() has two code paths: a "by-PHC" path for user-specified hardware timestamping providers, and the old path. Commit 89e281ebff72 ("ethtool: init tsinfo stats if requested") added ethtool_stats_init() to mark stat slots as ETHTOOL_STAT_NOT_SET before the driver callback populates them, but placed the call inside the old-path block. When commit b9e3f7dc9ed9 ("net: ethtool: tsinfo: Enhance tsinfo to support several hwtstamp by net topology") added the by-PHC early return, it landed above the stats initialization. On that path the stats array retains the zero-fill from ethnl_init_reply_data()'s zalloc. This leads to the reply including a stats nest with four zero-valued attributes that should have been absent. Reject GET requests for stats with HWTSTAMP_PROVIDER or dump. Fixes: b9e3f7dc9ed9 ("net: ethtool: tsinfo: Enhance tsinfo to support several hwtstamp by net topology") Reviewed-by: Maxime Chevallier Link: https://patch.msgid.link/20260526153533.2779187-7-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/tsinfo.c | 13 +++++++++++++ 1 file changed, 13 insertions(+) (limited to 'net') diff --git a/net/ethtool/tsinfo.c b/net/ethtool/tsinfo.c index a865f0fdd26b..f54fe6b662b2 100644 --- a/net/ethtool/tsinfo.c +++ b/net/ethtool/tsinfo.c @@ -83,6 +83,11 @@ tsinfo_parse_request(struct ethnl_req_info *req_base, if (!tb[ETHTOOL_A_TSINFO_HWTSTAMP_PROVIDER]) return 0; + if (req_base->flags & ETHTOOL_FLAG_STATS) { + NL_SET_ERR_MSG(extack, "can't query statistics for a provider"); + return -EOPNOTSUPP; + } + return ts_parse_hwtst_provider(tb[ETHTOOL_A_TSINFO_HWTSTAMP_PROVIDER], &req->hwprov_desc, extack, &mod); } @@ -523,6 +528,12 @@ int ethnl_tsinfo_start(struct netlink_callback *cb) if (ret < 0) goto free_reply_data; + if (req_info->base.flags & ETHTOOL_FLAG_STATS) { + NL_SET_ERR_MSG(cb->extack, "stats not supported in dump"); + ret = -EOPNOTSUPP; + goto err_dev_put; + } + ctx->req_info = req_info; ctx->reply_data = reply_data; ctx->pos_ifindex = 0; @@ -532,6 +543,8 @@ int ethnl_tsinfo_start(struct netlink_callback *cb) return 0; +err_dev_put: + ethnl_parse_header_dev_put(&req_info->base); free_reply_data: kfree(reply_data); free_req_info: -- cgit v1.2.3 From c3fc9976f686f9a95baf87db9d387f218fd65394 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 26 May 2026 08:35:30 -0700 Subject: ethtool: tsinfo: don't pass ERR_PTR to genlmsg_cancel on prepare failure The goto err label leads to: genlmsg_cancel(skb, ehdr); return ret; If ethnl_tsinfo_prepare_dump() failed, it has not started a genlmsg. There's nothing to cancel, and passing an error pointer to genlmsg_cancel() would cause a crash. Fixes: b9e3f7dc9ed9 ("net: ethtool: tsinfo: Enhance tsinfo to support several hwtstamp by net topology") Reviewed-by: Maxime Chevallier Reviewed-by: Kory Maincent Link: https://patch.msgid.link/20260526153533.2779187-8-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/tsinfo.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/ethtool/tsinfo.c b/net/ethtool/tsinfo.c index f54fe6b662b2..14bf01e3b55c 100644 --- a/net/ethtool/tsinfo.c +++ b/net/ethtool/tsinfo.c @@ -407,10 +407,8 @@ static int ethnl_tsinfo_dump_one_netdev(struct sk_buff *skb, continue; ehdr = ethnl_tsinfo_prepare_dump(skb, dev, reply_data, cb); - if (IS_ERR(ehdr)) { - ret = PTR_ERR(ehdr); - goto err; - } + if (IS_ERR(ehdr)) + return PTR_ERR(ehdr); reply_data->ts_info.phc_qualifier = ctx->pos_phcqualifier; ret = ops->get_ts_info(dev, &reply_data->ts_info); -- cgit v1.2.3 From a8d8bef6b45bf7cc0b1f6110c5cd8d0160a9bad7 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 26 May 2026 08:35:31 -0700 Subject: ethtool: strset: fix header attribute index in ethnl_req_get_phydev() strset_prepare_data() passes ETHTOOL_A_HEADER_FLAGS (3) as the header attribute to ethnl_req_get_phydev(). This is incorrect, in the main attr space 3 is ETHTOOL_A_STRSET_COUNTS_ONLY, not the request header attr. The correct constant is ETHTOOL_A_STRSET_HEADER (1). ethnl_req_get_phydev() only uses this value for the extack, so this is not a "functionally visible"(?) bug. Fixes: e96c93aa4be9 ("net: ethtool: strset: Allow querying phy stats by index") Reviewed-by: Maxime Chevallier Link: https://patch.msgid.link/20260526153533.2779187-9-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/strset.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/ethtool/strset.c b/net/ethtool/strset.c index bb1e829ba099..94c4718d31ae 100644 --- a/net/ethtool/strset.c +++ b/net/ethtool/strset.c @@ -311,7 +311,7 @@ static int strset_prepare_data(const struct ethnl_req_info *req_base, return 0; } - phydev = ethnl_req_get_phydev(req_base, tb, ETHTOOL_A_HEADER_FLAGS, + phydev = ethnl_req_get_phydev(req_base, tb, ETHTOOL_A_STRSET_HEADER, info->extack); /* phydev can be NULL, check for errors only */ -- cgit v1.2.3 From 2376586f85f972fefe701f095bb37dcfe7405d21 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 26 May 2026 08:35:32 -0700 Subject: ethtool: eeprom: add missing ethnl_ops_begin() / _complete() during fallback All ethtool driver op calls should be sandwiched between ethnl_ops_begin() / ethnl_ops_complete(). In Netlink eeprom code, if the paged access failed we fall back to old API, but we first call _complete() and the fallback never does its own ethnl_ops_begin(). Move the fallback into the _begin() / _complete() section. Fixes: 96d971e307cc ("ethtool: Add fallback to get_module_eeprom from netlink command") Reviewed-by: Maxime Chevallier Link: https://patch.msgid.link/20260526153533.2779187-10-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/eeprom.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/ethtool/eeprom.c b/net/ethtool/eeprom.c index a557e3996c85..836316df3092 100644 --- a/net/ethtool/eeprom.c +++ b/net/ethtool/eeprom.c @@ -141,12 +141,11 @@ static int eeprom_prepare_data(const struct ethnl_req_info *req_base, return 0; err_ops: + if (ret == -EOPNOTSUPP) + ret = eeprom_fallback(request, reply); ethnl_ops_complete(dev); err_free: kfree(page_data.data); - - if (ret == -EOPNOTSUPP) - return eeprom_fallback(request, reply); return ret; } -- cgit v1.2.3 From 67cfdd9210b99f260b3e0afeb9525e0acc7be31e Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 26 May 2026 08:35:33 -0700 Subject: ethtool: eeprom: add more safeties to EEPROM Netlink fallback The Netlink fallback path for reading module EEPROM (fallback_set_params()) validates that offset < eeprom_len, but does not check that offset + length stays within eeprom_len. The ioctl equivalent (ethtool_get_any_eeprom() in ioctl.c) has always enforced both bounds: if (eeprom.offset + eeprom.len > total_len) return -EINVAL; This could lead to surprises in both drivers and device FW. Add the missing offset + length validation to fallback_set_params(), mirroring the ioctl. Similarly - ethtool core in general, and ethtool_get_any_eeprom() in particular tries to zero-init all buffers passed to the drivers to avoid any extra work of zeroing things out. eeprom_fallback() uses a plain kmalloc(), change it to zalloc. Fixes: 96d971e307cc ("ethtool: Add fallback to get_module_eeprom from netlink command") Reviewed-by: Maxime Chevallier Link: https://patch.msgid.link/20260526153533.2779187-11-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/eeprom.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/ethtool/eeprom.c b/net/ethtool/eeprom.c index 836316df3092..0b8cfeddb014 100644 --- a/net/ethtool/eeprom.c +++ b/net/ethtool/eeprom.c @@ -44,6 +44,9 @@ static int fallback_set_params(struct eeprom_req_info *request, if (offset >= modinfo->eeprom_len) return -EINVAL; + if (length > modinfo->eeprom_len - offset) + return -EINVAL; + eeprom->cmd = ETHTOOL_GMODULEEEPROM; eeprom->len = length; eeprom->offset = offset; @@ -69,7 +72,7 @@ static int eeprom_fallback(struct eeprom_req_info *request, if (err < 0) return err; - data = kmalloc(eeprom.len, GFP_KERNEL); + data = kzalloc(eeprom.len, GFP_KERNEL); if (!data) return -ENOMEM; err = ethtool_get_module_eeprom_call(dev, &eeprom, data); -- cgit v1.2.3 From 9d5e7a46a9f6d8f503b41bfefef70659845f1679 Mon Sep 17 00:00:00 2001 From: Rahul Chandelkar Date: Mon, 25 May 2026 21:10:31 +0530 Subject: ipv6: rpl: fix hdrlen overflow in ipv6_rpl_srh_decompress() ipv6_rpl_srh_decompress() computes: outhdr->hdrlen = (((n + 1) * sizeof(struct in6_addr)) >> 3); hdrlen is __u8. For n >= 127 the result exceeds 255 and silently truncates. With n=127 (cmpri=15, cmpre=15, pad=0, hdrlen=16): (128 * 16) >> 3 = 256, truncated to 0 as __u8 The caller in ipv6_rpl_srh_rcv() then places the compressed header at buf + ((ohdr->hdrlen + 1) << 3). With hdrlen=0 this is buf + 8, but the decompressed region occupies buf[0..2055] (8-byte header plus 128 full addresses). The compressed header overlaps the decompressed data, and ipv6_rpl_srh_compress() writes into this overlap, corrupting the routing header of the forwarded packet. The existing guard at exthdrs.c:546 checks (n + 1) > 255, which prevents n+1 from overflowing unsigned char (the segments_left field), but does not prevent the computed hdrlen from overflowing __u8. n=127 passes because 128 <= 255, yet hdrlen=256 does not fit. Tighten the bound to (n + 1) > 127. This caps n at 126, giving hdrlen = (127 * 16) >> 3 = 254, which fits in __u8. The compressed header then lands at buf + ((254 + 1) << 3) = buf + 2040, exactly past the decompressed region (buf[0..2039]). No overlap. 127 segments is well beyond any realistic RPL deployment. Fixes: 8610c7c6e3bd ("net: ipv6: add support for rpl sr exthdr") Signed-off-by: Rahul Chandelkar Link: https://patch.msgid.link/20260525154031.2290876-1-rc@rexion.ai Signed-off-by: Jakub Kicinski --- net/ipv6/exthdrs.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv6/exthdrs.c b/net/ipv6/exthdrs.c index aca2a2abd2df..43f46ef9c53b 100644 --- a/net/ipv6/exthdrs.c +++ b/net/ipv6/exthdrs.c @@ -548,7 +548,7 @@ looped_back: * unsigned char which is segments_left field. Should not be * higher than that. */ - if (r || (n + 1) > 255) { + if (r || (n + 1) > 127) { kfree_skb(skb); return -1; } -- cgit v1.2.3 From eda0b7f203bb166c98d1418b204135bd566ac83b Mon Sep 17 00:00:00 2001 From: Jamal Hadi Salim Date: Mon, 25 May 2026 08:25:49 -0400 Subject: net/sched: Revert "net/sched: Restrict conditions for adding duplicating netems to qdisc tree" This reverts commit ec8e0e3d7adef940cdf9475e2352c0680189d14e. The original patch rejects any tree containing two netems when either has duplication set, even when they sit on unrelated classes of the same classful parent. That broke configurations that have worked since netem was introduced. The re-entrancy problem the original commit was trying to solve is handled by later patch using tc_depth flag. Doing this revert will (re)expose the original bug with multiple netem duplication. When this patch is backported make sure and get the full series. Fixes: ec8e0e3d7ade ("net/sched: Restrict conditions for adding duplicating netems to qdisc tree") Reported-by: Ji-Soo Chung Reported-by: Gerlinde Closes: https://bugzilla.kernel.org/show_bug.cgi?id=220774 Reported-by: zyc zyc Closes: https://lore.kernel.org/all/19adda5a1e2.12410b78222774.9191120410578703463@zohomail.cn/ Reported-by: Manas Ghandat Closes: https://lore.kernel.org/netdev/f69b2c8f-8325-4c2e-a011-6dbc089f30e4@gmail.com/ Reviewed-by: Stephen Hemminger Signed-off-by: Jamal Hadi Salim Link: https://patch.msgid.link/20260525122556.973584-3-jhs@mojatatu.com Signed-off-by: Paolo Abeni --- net/sched/sch_netem.c | 40 ---------------------------------------- 1 file changed, 40 deletions(-) (limited to 'net') diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c index bc18e1976b6e..d97acd2f3923 100644 --- a/net/sched/sch_netem.c +++ b/net/sched/sch_netem.c @@ -1007,41 +1007,6 @@ static int parse_attr(struct nlattr *tb[], int maxtype, struct nlattr *nla, return 0; } -static const struct Qdisc_class_ops netem_class_ops; - -static int check_netem_in_tree(struct Qdisc *sch, bool duplicates, - struct netlink_ext_ack *extack) -{ - struct Qdisc *root, *q; - unsigned int i; - - root = qdisc_root_sleeping(sch); - - if (sch != root && root->ops->cl_ops == &netem_class_ops) { - if (duplicates || - ((struct netem_sched_data *)qdisc_priv(root))->duplicate) - goto err; - } - - if (!qdisc_dev(root)) - return 0; - - hash_for_each(qdisc_dev(root)->qdisc_hash, i, q, hash) { - if (sch != q && q->ops->cl_ops == &netem_class_ops) { - if (duplicates || - ((struct netem_sched_data *)qdisc_priv(q))->duplicate) - goto err; - } - } - - return 0; - -err: - NL_SET_ERR_MSG(extack, - "netem: cannot mix duplicating netems with other netems in tree"); - return -EINVAL; -} - /* Parse netlink message to set options */ static int netem_change(struct Qdisc *sch, struct nlattr *opt, struct netlink_ext_ack *extack) @@ -1118,11 +1083,6 @@ static int netem_change(struct Qdisc *sch, struct nlattr *opt, q->gap = qopt->gap; q->counter = 0; q->loss = qopt->loss; - - ret = check_netem_in_tree(sch, qopt->duplicate, extack); - if (ret) - goto unlock; - q->duplicate = qopt->duplicate; /* for compatibility with earlier versions. -- cgit v1.2.3 From 9552b11e3edabc97cfcd9f29103d5afbce7ae183 Mon Sep 17 00:00:00 2001 From: Jamal Hadi Salim Date: Mon, 25 May 2026 08:25:51 -0400 Subject: net/sched: fix packet loop on netem when duplicate is on When netem duplicates a packet it re-enqueues the copy at the root qdisc. If another netem sits in the tree the copy can be duplicated again, recursing until the stack or memory is exhausted. The original duplication guard temporarily zeroed q->duplicate around the re-enqueue, but that does not cover all cases because it is per-qdisc state shared across all concurrent enqueue paths and is not safe without additional locking. Use the skb tc_depth field introduced in an earlier patch: - increment it on the duplicate before re-enqueue - skip duplication for any skb whose tc_depth is already non-zero. This marks the packet itself rather than mutating qdisc state, therefore it is safe regardless of tree topology or concurrency. Fixes: 0afb51e72855 ("[PKT_SCHED]: netem: reinsert for duplication") Reported-by: William Liu Reported-by: Savino Dicanosa Closes: https://lore.kernel.org/netdev/8DuRWwfqjoRDLDmBMlIfbrsZg9Gx50DHJc1ilxsEBNe2D6NMoigR_eIRIG0LOjMc3r10nUUZtArXx4oZBIdUfZQrwjcQhdinnMis_0G7VEk=@willsroot.io/ Co-developed-by: Victor Nogueira Signed-off-by: Victor Nogueira Reviewed-by: William Liu Reviewed-by: Stephen Hemminger Signed-off-by: Jamal Hadi Salim Link: https://patch.msgid.link/20260525122556.973584-5-jhs@mojatatu.com Signed-off-by: Paolo Abeni --- net/sched/sch_netem.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c index d97acd2f3923..17a79fe2f091 100644 --- a/net/sched/sch_netem.c +++ b/net/sched/sch_netem.c @@ -461,7 +461,8 @@ static int netem_enqueue(struct sk_buff *skb, struct Qdisc *sch, skb->prev = NULL; /* Random duplication */ - if (q->duplicate && q->duplicate >= get_crandom(&q->dup_cor, &q->prng)) + if (q->duplicate && skb->tc_depth == 0 && + q->duplicate >= get_crandom(&q->dup_cor, &q->prng)) ++count; /* Drop packet? */ @@ -540,11 +541,9 @@ static int netem_enqueue(struct sk_buff *skb, struct Qdisc *sch, */ if (skb2) { struct Qdisc *rootq = qdisc_root_bh(sch); - u32 dupsave = q->duplicate; /* prevent duplicating a dup... */ - q->duplicate = 0; + skb2->tc_depth++; /* prevent duplicating a dup... */ rootq->enqueue(skb2, rootq, to_free); - q->duplicate = dupsave; skb2 = NULL; } -- cgit v1.2.3 From db875221ab08d213a83bf30196ae8b64d55a3403 Mon Sep 17 00:00:00 2001 From: Jamal Hadi Salim Date: Mon, 25 May 2026 08:25:52 -0400 Subject: net/sched: Fix ethx:ingress -> ethy:egress -> ethx:ingress mirred loop When mirred redirects to ingress (from either ingress or egress) the loop state from sched_mirred_dev array dev is lost because of 1) the packet deferral into the backlog and 2) the fact the sched_mirred_dev array is cleared. In such cases, if there was a loop we won't discover it. Here's a simple test to reproduce: ip a add dev port0 10.10.10.11/24 tc qdisc add dev port0 clsact tc filter add dev port0 egress protocol ip \ prio 10 matchall action mirred ingress redirect dev port1 tc qdisc add dev port1 clsact tc filter add dev port1 ingress protocol ip \ prio 10 matchall action mirred egress redirect dev port0 ping -c 1 -W0.01 10.10.10.10 Fixes: fe946a751d9b ("net/sched: act_mirred: add loop detection") Tested-by: Victor Nogueira Reviewed-by: Stephen Hemminger Signed-off-by: Jamal Hadi Salim Link: https://patch.msgid.link/20260525122556.973584-6-jhs@mojatatu.com Signed-off-by: Paolo Abeni --- net/sched/act_mirred.c | 47 ++++++++++++++++++++++++++++++----------------- 1 file changed, 30 insertions(+), 17 deletions(-) (limited to 'net') diff --git a/net/sched/act_mirred.c b/net/sched/act_mirred.c index 2c5a7a321a94..dd5e7ea7ef26 100644 --- a/net/sched/act_mirred.c +++ b/net/sched/act_mirred.c @@ -26,6 +26,10 @@ #include #include +#define MIRRED_DEFER_LIMIT 3 +_Static_assert(MIRRED_DEFER_LIMIT <= 3, + "MIRRED_DEFER_LIMIT exceeds tc_depth bitfield width"); + static LIST_HEAD(mirred_list); static DEFINE_SPINLOCK(mirred_list_lock); @@ -234,12 +238,15 @@ tcf_mirred_forward(bool at_ingress, bool want_ingress, struct sk_buff *skb) { int err; - if (!want_ingress) + if (!want_ingress) { err = tcf_dev_queue_xmit(skb, dev_queue_xmit); - else if (!at_ingress) - err = netif_rx(skb); - else - err = netif_receive_skb(skb); + } else { + skb->tc_depth++; + if (!at_ingress) + err = netif_rx(skb); + else + err = netif_receive_skb(skb); + } return err; } @@ -426,6 +433,7 @@ TC_INDIRECT_SCOPE int tcf_mirred_act(struct sk_buff *skb, struct netdev_xmit *xmit; bool m_mac_header_xmit; struct net_device *dev; + bool want_ingress; int i, m_eaction; u32 blockid; @@ -434,7 +442,8 @@ TC_INDIRECT_SCOPE int tcf_mirred_act(struct sk_buff *skb, #else xmit = this_cpu_ptr(&softnet_data.xmit); #endif - if (unlikely(xmit->sched_mirred_nest >= MIRRED_NEST_LIMIT)) { + if (unlikely(xmit->sched_mirred_nest >= MIRRED_NEST_LIMIT || + skb->tc_depth >= MIRRED_DEFER_LIMIT)) { net_warn_ratelimited("Packet exceeded mirred recursion limit on dev %s\n", netdev_name(skb->dev)); return TC_ACT_SHOT; @@ -453,23 +462,27 @@ TC_INDIRECT_SCOPE int tcf_mirred_act(struct sk_buff *skb, tcf_action_inc_overlimit_qstats(&m->common); return retval; } - for (i = 0; i < xmit->sched_mirred_nest; i++) { - if (xmit->sched_mirred_dev[i] != dev) - continue; - pr_notice_once("tc mirred: loop on device %s\n", - netdev_name(dev)); - tcf_action_inc_overlimit_qstats(&m->common); - return retval; - } - xmit->sched_mirred_dev[xmit->sched_mirred_nest++] = dev; + m_eaction = READ_ONCE(m->tcfm_eaction); + want_ingress = tcf_mirred_act_wants_ingress(m_eaction); + if (!want_ingress) { + for (i = 0; i < xmit->sched_mirred_nest; i++) { + if (xmit->sched_mirred_dev[i] != dev) + continue; + pr_notice_once("tc mirred: loop on device %s\n", + netdev_name(dev)); + tcf_action_inc_overlimit_qstats(&m->common); + return retval; + } + xmit->sched_mirred_dev[xmit->sched_mirred_nest++] = dev; + } m_mac_header_xmit = READ_ONCE(m->tcfm_mac_header_xmit); - m_eaction = READ_ONCE(m->tcfm_eaction); retval = tcf_mirred_to_dev(skb, m, dev, m_mac_header_xmit, m_eaction, retval); - xmit->sched_mirred_nest--; + if (!want_ingress) + xmit->sched_mirred_nest--; return retval; } -- cgit v1.2.3 From a005fa5d7502eefec7ee6e1c01adadc06de2f9ad Mon Sep 17 00:00:00 2001 From: "Kito Xu (veritas501)" Date: Mon, 25 May 2026 08:25:53 -0400 Subject: net/sched: act_mirred: Fix blockcast recursion bypass leading to stack overflow tcf_mirred_act() checks sched_mirred_nest against MIRRED_NEST_LIMIT (4) to prevent deep recursion. However, when the action uses blockcast (tcfm_blockid != 0), the function returns at the tcf_blockcast() call BEFORE reaching the counter increment. As a result, the recursion counter never advances and the limit check is entirely bypassed. When two devices share a TC egress block with a mirred blockcast rule, a packet egressing on device A is mirrored to device B via blockcast; device B's egress TC re-enters tcf_mirred_act() via blockcast and mirrors back to A, creating an unbounded recursion loop: tcf_mirred_act -> tcf_blockcast -> tcf_mirred_to_dev -> dev_queue_xmit -> sch_handle_egress -> tcf_classify -> tcf_mirred_act -> (repeat) This recursion continues until the kernel stack overflows. The bug is reachable from an unprivileged user via unshare(CLONE_NEWUSER | CLONE_NEWNET): user namespaces grant CAP_NET_ADMIN in the new network namespace, which is sufficient to create dummy devices, attach clsact qdiscs with shared blocks, and install mirred blockcast filters. BUG: TASK stack guard page was hit at ffffc90000b7fff8 Oops: stack guard page: 0000 [#1] SMP KASAN NOPTI CPU: 2 UID: 1000 PID: 169 Comm: poc Not tainted 7.0.0-rc7-next-20260410 RIP: 0010:xas_find+0x17/0x480 Call Trace: xa_find+0x17b/0x1d0 tcf_mirred_act+0x640/0x1060 tcf_action_exec+0x400/0x530 basic_classify+0x128/0x1d0 tcf_classify+0xd83/0x1150 tc_run+0x328/0x620 __dev_queue_xmit+0x797/0x3100 tcf_mirred_to_dev+0x7b1/0xf70 tcf_mirred_act+0x68a/0x1060 [repeating ~30+ times until stack overflow] Kernel panic - not syncing: Fatal exception in interrupt Fix this by incrementing sched_mirred_nest before calling tcf_blockcast() and decrementing it on return, mirroring the non-blockcast path. This ensures subsequent recursive entries see the updated counter and are correctly limited by MIRRED_NEST_LIMIT. Fixes: fe946a751d9b ("net/sched: act_mirred: add loop detection") Signed-off-by: Kito Xu (veritas501) Link: https://patch.msgid.link/20260525122556.973584-7-jhs@mojatatu.com Signed-off-by: Paolo Abeni --- net/sched/act_mirred.c | 18 +++++++++++------- 1 file changed, 11 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/net/sched/act_mirred.c b/net/sched/act_mirred.c index dd5e7ea7ef26..dbe4a4ff3e08 100644 --- a/net/sched/act_mirred.c +++ b/net/sched/act_mirred.c @@ -396,14 +396,12 @@ static int tcf_blockcast_mirror(struct sk_buff *skb, struct tcf_mirred *m, static int tcf_blockcast(struct sk_buff *skb, struct tcf_mirred *m, const u32 blockid, struct tcf_result *res, - int retval) + int m_eaction, int retval) { const u32 exception_ifindex = skb->dev->ifindex; struct tcf_block *block; bool is_redirect; - int m_eaction; - m_eaction = READ_ONCE(m->tcfm_eaction); is_redirect = tcf_mirred_is_act_redirect(m_eaction); /* we are already under rcu protection, so can call block lookup @@ -453,8 +451,16 @@ TC_INDIRECT_SCOPE int tcf_mirred_act(struct sk_buff *skb, tcf_action_update_bstats(&m->common, skb); blockid = READ_ONCE(m->tcfm_blockid); - if (blockid) - return tcf_blockcast(skb, m, blockid, res, retval); + m_eaction = READ_ONCE(m->tcfm_eaction); + want_ingress = tcf_mirred_act_wants_ingress(m_eaction); + if (blockid) { + if (!want_ingress) + xmit->sched_mirred_dev[xmit->sched_mirred_nest++] = NULL; + retval = tcf_blockcast(skb, m, blockid, res, m_eaction, retval); + if (!want_ingress) + xmit->sched_mirred_nest--; + return retval; + } dev = rcu_dereference_bh(m->tcfm_dev); if (unlikely(!dev)) { @@ -463,8 +469,6 @@ TC_INDIRECT_SCOPE int tcf_mirred_act(struct sk_buff *skb, return retval; } - m_eaction = READ_ONCE(m->tcfm_eaction); - want_ingress = tcf_mirred_act_wants_ingress(m_eaction); if (!want_ingress) { for (i = 0; i < xmit->sched_mirred_nest; i++) { if (xmit->sched_mirred_dev[i] != dev) -- cgit v1.2.3 From e80ad525fc7e8c933ad78478c5dda286cfd55c60 Mon Sep 17 00:00:00 2001 From: Victor Nogueira Date: Mon, 25 May 2026 08:25:54 -0400 Subject: net/sched: act_mirred: Fix return code in early mirred redirect error paths Since retval is set as TC_ACT_STOLEN in the mirred redirect case, returning retval in cases where redirect failed will make the callers not register the skb as being dropped. Fix this by returning TC_ACT_SHOT instead in such scenarios. Fixes: 16085e48cb48 ("net/sched: act_mirred: Create function tcf_mirred_to_dev and improve readability") Reported-by: Sashiko Closes: https://sashiko.dev/#/patchset/20260413082027.2244884-1-hxzene%40gmail.com Signed-off-by: Victor Nogueira Link: https://patch.msgid.link/20260525122556.973584-8-jhs@mojatatu.com Signed-off-by: Paolo Abeni --- net/sched/act_mirred.c | 18 +++++++++++++----- 1 file changed, 13 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/sched/act_mirred.c b/net/sched/act_mirred.c index dbe4a4ff3e08..553342c55cf7 100644 --- a/net/sched/act_mirred.c +++ b/net/sched/act_mirred.c @@ -372,7 +372,8 @@ assign_prev: dev_is_mac_header_xmit(dev_prev), m_eaction, retval); - return retval; + /* If the packet wasn't redirected, we have to register as a drop */ + return TC_ACT_SHOT; } static int tcf_blockcast_mirror(struct sk_buff *skb, struct tcf_mirred *m, @@ -410,7 +411,7 @@ static int tcf_blockcast(struct sk_buff *skb, struct tcf_mirred *m, block = tcf_block_lookup(dev_net(skb->dev), blockid); if (!block || xa_empty(&block->ports)) { tcf_action_inc_overlimit_qstats(&m->common); - return retval; + return is_redirect ? TC_ACT_SHOT : retval; } if (is_redirect) @@ -428,8 +429,8 @@ TC_INDIRECT_SCOPE int tcf_mirred_act(struct sk_buff *skb, { struct tcf_mirred *m = to_mirred(a); int retval = READ_ONCE(m->tcf_action); + bool m_mac_header_xmit, is_redirect; struct netdev_xmit *xmit; - bool m_mac_header_xmit; struct net_device *dev; bool want_ingress; int i, m_eaction; @@ -462,11 +463,13 @@ TC_INDIRECT_SCOPE int tcf_mirred_act(struct sk_buff *skb, return retval; } + is_redirect = tcf_mirred_is_act_redirect(m_eaction); + dev = rcu_dereference_bh(m->tcfm_dev); if (unlikely(!dev)) { pr_notice_once("tc mirred: target device is gone\n"); tcf_action_inc_overlimit_qstats(&m->common); - return retval; + goto err_out; } if (!want_ingress) { @@ -476,7 +479,7 @@ TC_INDIRECT_SCOPE int tcf_mirred_act(struct sk_buff *skb, pr_notice_once("tc mirred: loop on device %s\n", netdev_name(dev)); tcf_action_inc_overlimit_qstats(&m->common); - return retval; + goto err_out; } xmit->sched_mirred_dev[xmit->sched_mirred_nest++] = dev; } @@ -489,6 +492,11 @@ TC_INDIRECT_SCOPE int tcf_mirred_act(struct sk_buff *skb, xmit->sched_mirred_nest--; return retval; + +err_out: + if (is_redirect) + retval = TC_ACT_SHOT; + return retval; } static void tcf_stats_update(struct tc_action *a, u64 bytes, u64 packets, -- cgit v1.2.3 From 98d0912e9f841e5529a5b89a972805f34cb1c69d Mon Sep 17 00:00:00 2001 From: Minh Nguyen Date: Tue, 26 May 2026 11:12:39 +0700 Subject: net: skbuff: fix missing zerocopy reference in pskb_carve helpers pskb_carve_inside_header() and pskb_carve_inside_nonlinear() both copy the old skb_shared_info header into a new buffer via memcpy(), which includes the destructor_arg pointer (uarg) for MSG_ZEROCOPY skbs. Neither function calls net_zcopy_get() for the new shinfo, creating an unaccounted holder: every skb_shared_info with destructor_arg set will call skb_zcopy_clear() once when freed, but the corresponding net_zcopy_get() was never called for the new copy. Repeated calls drive uarg->refcnt to zero prematurely, freeing ubuf_info_msgzc while TX skbs still hold live destructor_arg pointers. KASAN reports use-after-free on a freed ubuf_info_msgzc: BUG: KASAN: slab-use-after-free in skb_release_data+0x77b/0x810 Read of size 8 at addr ffff88801574d3e8 by task poc/220 Call Trace: skb_release_data+0x77b/0x810 kfree_skb_list_reason+0x13e/0x610 skb_release_data+0x4cd/0x810 sk_skb_reason_drop+0xf3/0x340 skb_queue_purge_reason+0x282/0x440 rds_tcp_inc_free+0x1e/0x30 rds_recvmsg+0x354/0x1780 __sys_recvmsg+0xdf/0x180 Allocated by task 219: msg_zerocopy_realloc+0x157/0x7b0 tcp_sendmsg_locked+0x2892/0x3ba0 Freed by task 219: ip_recv_error+0x74a/0xb10 tcp_recvmsg+0x475/0x530 The skb consuming the late access still referenced the same uarg via shinfo->destructor_arg copied by pskb_carve_inside_nonlinear() without a refcount bump. This has been verified to be reliably exploitable: a working proof-of-concept achieves full root privilege escalation from an unprivileged local user on a default kernel configuration. The fix follows the pattern of pskb_expand_head() which has the same memcpy/cloned structure. For pskb_carve_inside_header(), net_zcopy_get() is placed after skb_orphan_frags() succeeds, so the orphan error path needs no cleanup. For pskb_carve_inside_nonlinear(), net_zcopy_get() is placed after all failure points and just before skb_release_data(), so no error path needs cleanup at all -- matching pskb_expand_head() more closely and avoiding the need for a balancing net_zcopy_put(). Fixes: 6fa01ccd8830 ("skbuff: Add pskb_extract() helper function") Cc: stable@vger.kernel.org Assisted-by: Claude:claude-sonnet-4-6 Signed-off-by: Minh Nguyen Reviewed-by: Willem de Bruijn Link: https://patch.msgid.link/20260526041240.329462-1-minhnguyen.080505@gmail.com Signed-off-by: Paolo Abeni --- net/core/skbuff.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'net') diff --git a/net/core/skbuff.c b/net/core/skbuff.c index d247acd447e4..0d3cc115f2e7 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -6833,6 +6833,8 @@ static int pskb_carve_inside_header(struct sk_buff *skb, const u32 off, skb_kfree_head(data); return -ENOMEM; } + if (skb_zcopy(skb)) + net_zcopy_get(skb_zcopy(skb)); for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) skb_frag_ref(skb, i); if (skb_has_frag_list(skb)) @@ -6976,6 +6978,8 @@ static int pskb_carve_inside_nonlinear(struct sk_buff *skb, const u32 off, skb_kfree_head(data); return -ENOMEM; } + if (skb_zcopy(skb)) + net_zcopy_get(skb_zcopy(skb)); skb_release_data(skb, SKB_CONSUMED); skb->head = data; -- cgit v1.2.3 From cc993e0927ec8bd98ea33377ada03295fcda0f24 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Mon, 25 May 2026 12:51:15 -0400 Subject: net/handshake: Use spin_lock_bh for hn_lock nvmet_tcp_state_change(), a socket callback that runs in BH context, can reach handshake_req_cancel() via nvmet_tcp_schedule_release_queue() and tls_handshake_cancel(). handshake_req_cancel() acquires hn->hn_lock with plain spin_lock(). If a process-context thread on the same CPU holds hn->hn_lock when a softirq invokes the cancel path, the lock attempt deadlocks. This is the only caller that invokes tls_handshake_cancel() from BH context; every other consumer calls it from process context. Deferring the cancel to process context in the NVMe target is not straightforward: nvmet_tcp_schedule_release_queue() must call tls_handshake_cancel() atomically with its state transition to DISCONNECTING. If the cancel were deferred, the handshake completion callback could fire in the window before the cancel runs, observe the unexpected state, and return without dropping its kref on the queue. Reworking that interlock is considerably more invasive than hardening the handshake lock. Convert all hn->hn_lock acquisitions from spin_lock/spin_unlock to spin_lock_bh/spin_unlock_bh so the lock is never taken with softirqs enabled. Fixes: 675b453e0241 ("nvmet-tcp: enable TLS handshake upcall") Signed-off-by: Chuck Lever Reviewed-by: Hannes Reinecke Link: https://patch.msgid.link/20260525-handshake-file-pin-v3-1-66c616906ead@oracle.com Signed-off-by: Paolo Abeni --- net/handshake/netlink.c | 4 ++-- net/handshake/request.c | 14 +++++++------- net/handshake/tlshd.c | 2 ++ 3 files changed, 11 insertions(+), 9 deletions(-) (limited to 'net') diff --git a/net/handshake/netlink.c b/net/handshake/netlink.c index b989456fc4c5..97114ec8027a 100644 --- a/net/handshake/netlink.c +++ b/net/handshake/netlink.c @@ -202,10 +202,10 @@ static void __net_exit handshake_net_exit(struct net *net) * accepted and are in progress will be destroyed when * the socket is closed. */ - spin_lock(&hn->hn_lock); + spin_lock_bh(&hn->hn_lock); set_bit(HANDSHAKE_F_NET_DRAINING, &hn->hn_flags); list_splice_init(&requests, &hn->hn_requests); - spin_unlock(&hn->hn_lock); + spin_unlock_bh(&hn->hn_lock); while (!list_empty(&requests)) { req = list_first_entry(&requests, struct handshake_req, hr_list); diff --git a/net/handshake/request.c b/net/handshake/request.c index 2829adbeb149..5d4a17f902d2 100644 --- a/net/handshake/request.c +++ b/net/handshake/request.c @@ -167,12 +167,12 @@ static bool remove_pending(struct handshake_net *hn, struct handshake_req *req) { bool ret = false; - spin_lock(&hn->hn_lock); + spin_lock_bh(&hn->hn_lock); if (!list_empty(&req->hr_list)) { __remove_pending_locked(hn, req); ret = true; } - spin_unlock(&hn->hn_lock); + spin_unlock_bh(&hn->hn_lock); return ret; } @@ -182,7 +182,7 @@ struct handshake_req *handshake_req_next(struct handshake_net *hn, int class) struct handshake_req *req, *pos; req = NULL; - spin_lock(&hn->hn_lock); + spin_lock_bh(&hn->hn_lock); list_for_each_entry(pos, &hn->hn_requests, hr_list) { if (pos->hr_proto->hp_handler_class != class) continue; @@ -190,7 +190,7 @@ struct handshake_req *handshake_req_next(struct handshake_net *hn, int class) req = pos; break; } - spin_unlock(&hn->hn_lock); + spin_unlock_bh(&hn->hn_lock); return req; } @@ -249,7 +249,7 @@ int handshake_req_submit(struct socket *sock, struct handshake_req *req, if (READ_ONCE(hn->hn_pending) >= hn->hn_pending_max) goto out_err; - spin_lock(&hn->hn_lock); + spin_lock_bh(&hn->hn_lock); ret = -EOPNOTSUPP; if (test_bit(HANDSHAKE_F_NET_DRAINING, &hn->hn_flags)) goto out_unlock; @@ -258,7 +258,7 @@ int handshake_req_submit(struct socket *sock, struct handshake_req *req, goto out_unlock; if (!__add_pending_locked(hn, req)) goto out_unlock; - spin_unlock(&hn->hn_lock); + spin_unlock_bh(&hn->hn_lock); ret = handshake_genl_notify(net, req->hr_proto, flags); if (ret) { @@ -274,7 +274,7 @@ int handshake_req_submit(struct socket *sock, struct handshake_req *req, return 0; out_unlock: - spin_unlock(&hn->hn_lock); + spin_unlock_bh(&hn->hn_lock); out_err: /* Restore original destructor so socket teardown still runs on failure */ req->hr_sk->sk_destruct = req->hr_odestruct; diff --git a/net/handshake/tlshd.c b/net/handshake/tlshd.c index 8f9532a15f43..af294c6cc717 100644 --- a/net/handshake/tlshd.c +++ b/net/handshake/tlshd.c @@ -425,6 +425,8 @@ EXPORT_SYMBOL(tls_server_hello_psk); * Request cancellation races with request completion. To determine * who won, callers examine the return value from this function. * + * Context: May be called from process or softirq context. + * * Return values: * %true - Uncompleted handshake request was canceled * %false - Handshake request already completed or not found -- cgit v1.2.3 From 6b22d433aa13f68e3cd9534ca9a5f4277bfa01c2 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Mon, 25 May 2026 12:51:17 -0400 Subject: net/handshake: Pass negative errno through handshake_complete() handshake_complete() declares status as unsigned int and tls_handshake_done() negates that value (-status) before handing it to the TLS consumer. Consumers match on negative errno constants -- xs_tls_handshake_done() has switch (status) { case 0: case -EACCES: case -ETIMEDOUT: lower_transport->xprt_err = status; break; default: lower_transport->xprt_err = -EACCES; } so the API as designed expects callers to pass positive errno values that the tlshd shim then negates. Three internal callers in handshake_nl_accept_doit(), the net-exit drain, and a kunit test follow kernel convention and pass negative errnos -- -EIO, -ETIMEDOUT, -ETIMEDOUT. The implicit conversion to unsigned int turns -ETIMEDOUT into 0xFFFFFF92; the subsequent -status in tls_handshake_done() wraps back to 110, the consumer's switch falls through, and the xprt reports -EACCES on what should be -ETIMEDOUT or -EIO. Fix the API rather than the call sites. The natural kernel convention is negative errno in, negative errno out. Change handshake_complete() and hp_done to take int status, drop the negation in tls_handshake_done(), and negate once in handshake_nl_done_doit() where status arrives from the wire as an unsigned netlink attribute. The three internal callers were already correct under that convention and need no change. At the same wire boundary, declare MAX_ERRNO as the netlink policy upper bound for HANDSHAKE_A_DONE_STATUS. Attribute validation rejects out-of-range values before handshake_nl_done_doit() runs, and negating a bounded u32 there stays within int range -- closing the UBSAN-visible signed- integer overflow that an unconstrained u32 would invoke. Fixes: 3b3009ea8abb ("net/handshake: Create a NETLINK service for handling handshake requests") Signed-off-by: Chuck Lever Reviewed-by: Hannes Reinecke Link: https://patch.msgid.link/20260525-handshake-file-pin-v3-3-66c616906ead@oracle.com Signed-off-by: Paolo Abeni --- net/handshake/genl.c | 3 ++- net/handshake/genl.h | 1 + net/handshake/handshake-test.c | 2 +- net/handshake/handshake.h | 4 ++-- net/handshake/netlink.c | 2 +- net/handshake/request.c | 2 +- net/handshake/tlshd.c | 4 ++-- 7 files changed, 10 insertions(+), 8 deletions(-) (limited to 'net') diff --git a/net/handshake/genl.c b/net/handshake/genl.c index 870612609491..4b20cd9cdd0e 100644 --- a/net/handshake/genl.c +++ b/net/handshake/genl.c @@ -10,6 +10,7 @@ #include "genl.h" #include +#include /* HANDSHAKE_CMD_ACCEPT - do */ static const struct nla_policy handshake_accept_nl_policy[HANDSHAKE_A_ACCEPT_HANDLER_CLASS + 1] = { @@ -18,7 +19,7 @@ static const struct nla_policy handshake_accept_nl_policy[HANDSHAKE_A_ACCEPT_HAN /* HANDSHAKE_CMD_DONE - do */ static const struct nla_policy handshake_done_nl_policy[HANDSHAKE_A_DONE_REMOTE_AUTH + 1] = { - [HANDSHAKE_A_DONE_STATUS] = { .type = NLA_U32, }, + [HANDSHAKE_A_DONE_STATUS] = NLA_POLICY_MAX(NLA_U32, MAX_ERRNO), [HANDSHAKE_A_DONE_SOCKFD] = { .type = NLA_S32, }, [HANDSHAKE_A_DONE_REMOTE_AUTH] = { .type = NLA_U32, }, }; diff --git a/net/handshake/genl.h b/net/handshake/genl.h index 8d3e18672daf..46b65f131669 100644 --- a/net/handshake/genl.h +++ b/net/handshake/genl.h @@ -11,6 +11,7 @@ #include #include +#include int handshake_nl_accept_doit(struct sk_buff *skb, struct genl_info *info); int handshake_nl_done_doit(struct sk_buff *skb, struct genl_info *info); diff --git a/net/handshake/handshake-test.c b/net/handshake/handshake-test.c index 55442b2f518a..df3948e807a0 100644 --- a/net/handshake/handshake-test.c +++ b/net/handshake/handshake-test.c @@ -25,7 +25,7 @@ static int test_accept_func(struct handshake_req *req, struct genl_info *info, return 0; } -static void test_done_func(struct handshake_req *req, unsigned int status, +static void test_done_func(struct handshake_req *req, int status, struct genl_info *info) { } diff --git a/net/handshake/handshake.h b/net/handshake/handshake.h index a48163765a7a..2289b0e274f4 100644 --- a/net/handshake/handshake.h +++ b/net/handshake/handshake.h @@ -57,7 +57,7 @@ struct handshake_proto { int (*hp_accept)(struct handshake_req *req, struct genl_info *info, int fd); void (*hp_done)(struct handshake_req *req, - unsigned int status, + int status, struct genl_info *info); void (*hp_destroy)(struct handshake_req *req); }; @@ -86,7 +86,7 @@ struct handshake_req *handshake_req_hash_lookup(struct sock *sk); struct handshake_req *handshake_req_next(struct handshake_net *hn, int class); int handshake_req_submit(struct socket *sock, struct handshake_req *req, gfp_t flags); -void handshake_complete(struct handshake_req *req, unsigned int status, +void handshake_complete(struct handshake_req *req, int status, struct genl_info *info); bool handshake_req_cancel(struct sock *sk); diff --git a/net/handshake/netlink.c b/net/handshake/netlink.c index 97114ec8027a..039344979de9 100644 --- a/net/handshake/netlink.c +++ b/net/handshake/netlink.c @@ -160,7 +160,7 @@ int handshake_nl_done_doit(struct sk_buff *skb, struct genl_info *info) status = -EIO; if (info->attrs[HANDSHAKE_A_DONE_STATUS]) - status = nla_get_u32(info->attrs[HANDSHAKE_A_DONE_STATUS]); + status = -(int)nla_get_u32(info->attrs[HANDSHAKE_A_DONE_STATUS]); handshake_complete(req, status, info); sockfd_put(sock); diff --git a/net/handshake/request.c b/net/handshake/request.c index 5d4a17f902d2..97f9f8239949 100644 --- a/net/handshake/request.c +++ b/net/handshake/request.c @@ -284,7 +284,7 @@ out_err: } EXPORT_SYMBOL(handshake_req_submit); -void handshake_complete(struct handshake_req *req, unsigned int status, +void handshake_complete(struct handshake_req *req, int status, struct genl_info *info) { struct sock *sk = req->hr_sk; diff --git a/net/handshake/tlshd.c b/net/handshake/tlshd.c index af294c6cc717..7567150c2a4f 100644 --- a/net/handshake/tlshd.c +++ b/net/handshake/tlshd.c @@ -93,7 +93,7 @@ static void tls_handshake_remote_peerids(struct tls_handshake_req *treq, * */ static void tls_handshake_done(struct handshake_req *req, - unsigned int status, struct genl_info *info) + int status, struct genl_info *info) { struct tls_handshake_req *treq = handshake_req_private(req); @@ -104,7 +104,7 @@ static void tls_handshake_done(struct handshake_req *req, if (!status) set_bit(HANDSHAKE_F_REQ_SESSION, &req->hr_flags); - treq->th_consumer_done(treq->th_consumer_data, -status, + treq->th_consumer_done(treq->th_consumer_data, status, treq->th_peerid[0]); } -- cgit v1.2.3 From 09dba37eee70d0596e26645015f1aa95a9848e9d Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Mon, 25 May 2026 12:51:18 -0400 Subject: net/handshake: Take a long-lived file reference at submit handshake_nl_accept_doit() needs the file pointer backing req->hr_sk->sk_socket to survive the window between handshake_req_next() and the subsequent FD_PREPARE() and get_file(). The submit-side sock_hold() does not provide that. sk_refcnt keeps struct sock alive, but struct socket is owned by sock->file: when the consumer fputs the last file reference, sock_release() tears the socket down regardless of any sock_hold. Add an hr_file pointer to struct handshake_req and acquire an explicit reference on sock->file during handshake_req_submit(). handshake_complete() and handshake_req_cancel() release the reference on the completion-bit-winning path. The submit error path must also release the file reference, but after rhashtable insertion a concurrent handshake_req_cancel() can discover the request and race the error path. Gate the error-path cleanup -- sk_destruct restoration, fput, and request destruction -- with test_and_set_bit(HANDSHAKE_F_REQ_COMPLETED), the same serialization handshake_complete() and handshake_req_cancel() already use. When cancel has already claimed ownership, the submit error path returns without touching the request; socket teardown handles final destruction. The accept-side dereferences are not yet retargeted; that change comes in the next patch. Signed-off-by: Chuck Lever Link: https://patch.msgid.link/20260525-handshake-file-pin-v3-4-66c616906ead@oracle.com Signed-off-by: Paolo Abeni --- net/handshake/handshake.h | 2 ++ net/handshake/netlink.c | 6 ------ net/handshake/request.c | 42 +++++++++++++++++++++++++++++++++++------- 3 files changed, 37 insertions(+), 13 deletions(-) (limited to 'net') diff --git a/net/handshake/handshake.h b/net/handshake/handshake.h index 2289b0e274f4..da61cadd1ad3 100644 --- a/net/handshake/handshake.h +++ b/net/handshake/handshake.h @@ -24,6 +24,7 @@ enum hn_flags_bits { HANDSHAKE_F_NET_DRAINING, }; +struct file; struct handshake_proto; /* One handshake request */ @@ -32,6 +33,7 @@ struct handshake_req { struct rhash_head hr_rhash; unsigned long hr_flags; const struct handshake_proto *hr_proto; + struct file *hr_file; struct sock *hr_sk; void (*hr_odestruct)(struct sock *sk); diff --git a/net/handshake/netlink.c b/net/handshake/netlink.c index 039344979de9..1a5821eb7184 100644 --- a/net/handshake/netlink.c +++ b/net/handshake/netlink.c @@ -210,12 +210,6 @@ static void __net_exit handshake_net_exit(struct net *net) while (!list_empty(&requests)) { req = list_first_entry(&requests, struct handshake_req, hr_list); list_del(&req->hr_list); - - /* - * Requests on this list have not yet been - * accepted, so they do not have an fd to put. - */ - handshake_complete(req, -ETIMEDOUT, NULL); } } diff --git a/net/handshake/request.c b/net/handshake/request.c index 97f9f8239949..da064511ab86 100644 --- a/net/handshake/request.c +++ b/net/handshake/request.c @@ -13,6 +13,7 @@ #include #include #include +#include #include #include @@ -215,9 +216,16 @@ EXPORT_SYMBOL_IF_KUNIT(handshake_req_next); * A zero return value from handshake_req_submit() means that * exactly one subsequent completion callback is guaranteed. * - * A negative return value from handshake_req_submit() means that - * no completion callback will be done and that @req has been - * destroyed. + * A negative return value from handshake_req_submit() guarantees that + * no completion callback will occur and that @req is no longer owned by + * the caller. If cancellation wins the completion race after the request + * has been published, final destruction is deferred until socket teardown. + * + * The caller must hold a reference on @sock->file for the duration + * of this call. Once the request is published to the accept side, a + * concurrent completion or cancellation may release the request's pin on + * @sock->file; the caller's reference is what keeps @sock->sk valid until + * handshake_req_submit() returns. */ int handshake_req_submit(struct socket *sock, struct handshake_req *req, gfp_t flags) @@ -236,6 +244,14 @@ int handshake_req_submit(struct socket *sock, struct handshake_req *req, kfree(req); return -EINVAL; } + + /* + * Pin sock->file for the lifetime of the request so the + * accept side does not race a consumer that releases the + * socket while a handshake is pending. + */ + req->hr_file = get_file(sock->file); + req->hr_odestruct = req->hr_sk->sk_destruct; req->hr_sk->sk_destruct = handshake_sk_destruct; @@ -267,7 +283,11 @@ int handshake_req_submit(struct socket *sock, struct handshake_req *req, goto out_err; } - /* Prevent socket release while a handshake request is pending */ + /* + * Pin struct sock so sk_destruct does not run until the + * handshake completion path releases it; struct socket is + * held separately via hr_file above. + */ sock_hold(req->hr_sk); trace_handshake_submit(net, req, req->hr_sk); @@ -276,10 +296,13 @@ int handshake_req_submit(struct socket *sock, struct handshake_req *req, out_unlock: spin_unlock_bh(&hn->hn_lock); out_err: - /* Restore original destructor so socket teardown still runs on failure */ - req->hr_sk->sk_destruct = req->hr_odestruct; trace_handshake_submit_err(net, req, req->hr_sk, ret); - handshake_req_destroy(req); + if (!test_and_set_bit(HANDSHAKE_F_REQ_COMPLETED, &req->hr_flags)) { + /* Restore original destructor so socket teardown still runs. */ + req->hr_sk->sk_destruct = req->hr_odestruct; + fput(req->hr_file); + handshake_req_destroy(req); + } return ret; } EXPORT_SYMBOL(handshake_req_submit); @@ -291,11 +314,15 @@ void handshake_complete(struct handshake_req *req, int status, struct net *net = sock_net(sk); if (!test_and_set_bit(HANDSHAKE_F_REQ_COMPLETED, &req->hr_flags)) { + struct file *file = req->hr_file; + trace_handshake_complete(net, req, sk, status); req->hr_proto->hp_done(req, status, info); /* Handshake request is no longer pending */ sock_put(sk); + + fput(file); } } EXPORT_SYMBOL_IF_KUNIT(handshake_complete); @@ -344,6 +371,7 @@ out_true: /* Handshake request is no longer pending */ sock_put(sk); + fput(req->hr_file); return true; } EXPORT_SYMBOL(handshake_req_cancel); -- cgit v1.2.3 From f4251190e58b209999c1ba9e6d2976136a1be055 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Mon, 25 May 2026 12:51:19 -0400 Subject: net/handshake: hand off the pinned file reference to accept_doit handshake_req_next() removes the request from the per-net pending list and drops hn_lock before handshake_nl_accept_doit() reads req->hr_sk->sk_socket and dereferences sock->file (once in FD_PREPARE() and again in get_file()). In that window a consumer running tls_handshake_cancel() followed by sockfd_put() (svc_sock_free) or __fput_sync() (xs_reset_transport) releases sock->file. sock_release() then runs sock_orphan(), zeroing sk_socket, and frees the struct socket. The accept-side code either reads NULL through sk_socket or chases freed memory. The submit-side sock_hold() does not prevent this. sk_refcnt protects struct sock, but struct socket and sock->file are independently refcounted via the file descriptor the consumer owns. Pinning sk leaves sock and sock->file unprotected. Retarget the accept-side dereferences at req->hr_file, which was pinned at submit time, instead of req->hr_sk->sk_socket->file. Pinning on its own is not sufficient: a consumer that cancels between handshake_req_next() returning and accept_doit reaching FD_PREPARE() takes the !remove_pending() branch in handshake_req_cancel() and drops hr_file before the accept side takes its own reference. Hand off an additional file reference inside handshake_req_next(), under hn_lock, so the accept side operates on a reference that no concurrent handshake_req_cancel() can revoke. FD_PREPARE() consumes that handed-off reference, either by transferring it to the new fd in fd_publish() or by dropping it in the cleanup destructor on error; the explicit get_file() that previously balanced FD_PREPARE() is therefore redundant and goes away. Update handshake_req_cancel_test2 and _test3 to simulate the FD_PREPARE() consumption with an fput() so the kunit file-count assertions stay balanced. Reported-by: Chris Mason Fixes: 3b3009ea8abb ("net/handshake: Create a NETLINK service for handling handshake requests") Signed-off-by: Chuck Lever Reviewed-by: Hannes Reinecke Link: https://patch.msgid.link/20260525-handshake-file-pin-v3-5-66c616906ead@oracle.com Signed-off-by: Paolo Abeni --- net/handshake/handshake-test.c | 8 ++++++++ net/handshake/netlink.c | 7 ++----- net/handshake/request.c | 18 ++++++++++++++++++ 3 files changed, 28 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/handshake/handshake-test.c b/net/handshake/handshake-test.c index df3948e807a0..9cc7a95f4120 100644 --- a/net/handshake/handshake-test.c +++ b/net/handshake/handshake-test.c @@ -375,6 +375,10 @@ static void handshake_req_cancel_test2(struct kunit *test) /* Pretend to accept this request */ next = handshake_req_next(hn, HANDSHAKE_HANDLER_CLASS_TLSHD); KUNIT_ASSERT_PTR_EQ(test, req, next); + /* Simulate FD_PREPARE() consuming the file reference handed + * off by handshake_req_next(); see handshake_nl_accept_doit(). + */ + fput(filp); /* Act */ result = handshake_req_cancel(sock->sk); @@ -417,6 +421,10 @@ static void handshake_req_cancel_test3(struct kunit *test) /* Pretend to accept this request */ next = handshake_req_next(hn, HANDSHAKE_HANDLER_CLASS_TLSHD); KUNIT_ASSERT_PTR_EQ(test, req, next); + /* Simulate FD_PREPARE() consuming the file reference handed + * off by handshake_req_next(); see handshake_nl_accept_doit(). + */ + fput(filp); /* Pretend to complete this request */ handshake_complete(next, -ETIMEDOUT, NULL); diff --git a/net/handshake/netlink.c b/net/handshake/netlink.c index 1a5821eb7184..21d6cbd52fcd 100644 --- a/net/handshake/netlink.c +++ b/net/handshake/netlink.c @@ -92,7 +92,6 @@ int handshake_nl_accept_doit(struct sk_buff *skb, struct genl_info *info) struct net *net = sock_net(skb->sk); struct handshake_net *hn = handshake_pernet(net); struct handshake_req *req = NULL; - struct socket *sock; int class, err; err = -EOPNOTSUPP; @@ -107,15 +106,13 @@ int handshake_nl_accept_doit(struct sk_buff *skb, struct genl_info *info) err = -EAGAIN; req = handshake_req_next(hn, class); if (req) { - sock = req->hr_sk->sk_socket; - - FD_PREPARE(fdf, O_CLOEXEC, sock->file); + FD_PREPARE(fdf, O_CLOEXEC, req->hr_file); if (fdf.err) { + fput(req->hr_file); /* drop ref from handshake_req_next() */ err = fdf.err; goto out_complete; } - get_file(sock->file); /* FD_PREPARE() consumes a reference. */ err = req->hr_proto->hp_accept(req, info, fd_prepare_fd(fdf)); if (err) goto out_complete; /* Automatic cleanup handles fput */ diff --git a/net/handshake/request.c b/net/handshake/request.c index da064511ab86..e2d7ee7ce6e0 100644 --- a/net/handshake/request.c +++ b/net/handshake/request.c @@ -178,6 +178,17 @@ static bool remove_pending(struct handshake_net *hn, struct handshake_req *req) return ret; } +/** + * handshake_req_next - Return the next queued handshake request + * @hn: per-net handshake state + * @class: handler class to match + * + * On a non-NULL return, the caller owns an extra reference + * on @req->hr_file. FD_PREPARE() consumes it on success; on + * the FD_PREPARE() failure path the caller must fput() it. + * + * Return: pointer to a removed handshake_req, or NULL. + */ struct handshake_req *handshake_req_next(struct handshake_net *hn, int class) { struct handshake_req *req, *pos; @@ -188,6 +199,13 @@ struct handshake_req *handshake_req_next(struct handshake_net *hn, int class) if (pos->hr_proto->hp_handler_class != class) continue; __remove_pending_locked(hn, pos); + /* Hand off a file reference to the accept side under + * hn_lock. A concurrent handshake_req_cancel() can drop + * hr_file before accept reaches FD_PREPARE(); this extra + * reference keeps the file alive until FD_PREPARE() takes + * ownership. + */ + get_file(pos->hr_file); req = pos; break; } -- cgit v1.2.3 From 5da98f55b13173c08f003011b76531b25c821c07 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Mon, 25 May 2026 12:51:20 -0400 Subject: net/handshake: Close the submit-side sock_hold race handshake_req_submit() publishes the request via handshake_req_hash_add() and __add_pending_locked(), drops hn_lock, and calls handshake_genl_notify() (which can sleep) before taking sock_hold() on req->hr_sk. A fast tlshd ACCEPT followed by DONE can drive handshake_complete()'s sock_put() into the window between the spin_unlock and the late sock_hold(); on a system where the consumer's fd held the only sk reference, the late sock_hold() then operates on an sk whose refcount has reached zero. The preceding two patches install an explicit file reference on struct handshake_req. That file pins sock->file, which pins the embedded struct socket, which defers inet_release()'s sock_put(). As long as hr_file is held, sk cannot reach refcount zero from the consumer side, and the submit-side sock_hold() with its matching sock_put() calls in handshake_complete() and handshake_req_cancel() is now redundant. Drop all three. The file reference already keeps each request's socket alive, and the lifetime story is contained in a single get_file()/fput() pair. Fixes: 3b3009ea8abb ("net/handshake: Create a NETLINK service for handling handshake requests") Signed-off-by: Chuck Lever Reviewed-by: Hannes Reinecke Link: https://patch.msgid.link/20260525-handshake-file-pin-v3-6-66c616906ead@oracle.com Signed-off-by: Paolo Abeni --- net/handshake/request.c | 12 ------------ 1 file changed, 12 deletions(-) (limited to 'net') diff --git a/net/handshake/request.c b/net/handshake/request.c index e2d7ee7ce6e0..bd3d9467ab91 100644 --- a/net/handshake/request.c +++ b/net/handshake/request.c @@ -301,13 +301,6 @@ int handshake_req_submit(struct socket *sock, struct handshake_req *req, goto out_err; } - /* - * Pin struct sock so sk_destruct does not run until the - * handshake completion path releases it; struct socket is - * held separately via hr_file above. - */ - sock_hold(req->hr_sk); - trace_handshake_submit(net, req, req->hr_sk); return 0; @@ -337,9 +330,6 @@ void handshake_complete(struct handshake_req *req, int status, trace_handshake_complete(net, req, sk, status); req->hr_proto->hp_done(req, status, info); - /* Handshake request is no longer pending */ - sock_put(sk); - fput(file); } } @@ -387,8 +377,6 @@ bool handshake_req_cancel(struct sock *sk) out_true: trace_handshake_cancel(net, req, sk); - /* Handshake request is no longer pending */ - sock_put(sk); fput(req->hr_file); return true; } -- cgit v1.2.3 From 204a5efde5ed52932840ee1d15d3b581cfda48e2 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Mon, 25 May 2026 12:51:21 -0400 Subject: net/handshake: Verify file-reference balance in submit paths The new file-reference contract on struct handshake_req is silently breakable: a missing get_file() at submit or a missing fput() on an error path leaves the file leaked but does not crash the test, so the existing absence-of-crash checks pass either way. Snapshot file_count(filp) before each handshake_req_submit() in the submit-success, EAGAIN, EBUSY, and cancel tests, and assert the expected balance after submit and again after cancel. The already-completed cancel test also asserts the post-complete balance, which pins down that handshake_complete() drops the reference and that the subsequent cancel does not double-fput. The destroy test gets the same treatment before __fput_sync(), which double-checks that cancel's fput() ran and the only remaining reference is the one sock_alloc_file() established. Signed-off-by: Chuck Lever Reviewed-by: Hannes Reinecke Link: https://patch.msgid.link/20260525-handshake-file-pin-v3-7-66c616906ead@oracle.com Signed-off-by: Paolo Abeni --- net/handshake/handshake-test.c | 28 ++++++++++++++++++++++++++++ 1 file changed, 28 insertions(+) (limited to 'net') diff --git a/net/handshake/handshake-test.c b/net/handshake/handshake-test.c index 9cc7a95f4120..3dd507470d5f 100644 --- a/net/handshake/handshake-test.c +++ b/net/handshake/handshake-test.c @@ -208,6 +208,7 @@ static void handshake_req_submit_test3(struct kunit *test) static void handshake_req_submit_test4(struct kunit *test) { struct handshake_req *req, *result; + unsigned long fcount_before; struct socket *sock; struct file *filp; int err; @@ -224,8 +225,10 @@ static void handshake_req_submit_test4(struct kunit *test) KUNIT_ASSERT_NOT_NULL(test, sock->sk); sock->file = filp; + fcount_before = file_count(filp); err = handshake_req_submit(sock, req, GFP_KERNEL); KUNIT_ASSERT_EQ(test, err, 0); + KUNIT_EXPECT_EQ(test, file_count(filp), fcount_before + 1); /* Act */ result = handshake_req_hash_lookup(sock->sk); @@ -235,11 +238,13 @@ static void handshake_req_submit_test4(struct kunit *test) KUNIT_EXPECT_PTR_EQ(test, req, result); handshake_req_cancel(sock->sk); + KUNIT_EXPECT_EQ(test, file_count(filp), fcount_before); fput(filp); } static void handshake_req_submit_test5(struct kunit *test) { + unsigned long fcount_before; struct handshake_req *req; struct handshake_net *hn; struct socket *sock; @@ -265,12 +270,14 @@ static void handshake_req_submit_test5(struct kunit *test) saved = hn->hn_pending; hn->hn_pending = hn->hn_pending_max + 1; + fcount_before = file_count(filp); /* Act */ err = handshake_req_submit(sock, req, GFP_KERNEL); /* Assert */ KUNIT_EXPECT_EQ(test, err, -EAGAIN); + KUNIT_EXPECT_EQ(test, file_count(filp), fcount_before); fput(filp); hn->hn_pending = saved; @@ -279,6 +286,7 @@ static void handshake_req_submit_test5(struct kunit *test) static void handshake_req_submit_test6(struct kunit *test) { struct handshake_req *req1, *req2; + unsigned long fcount_before; struct socket *sock; struct file *filp; int err; @@ -296,21 +304,26 @@ static void handshake_req_submit_test6(struct kunit *test) KUNIT_ASSERT_NOT_ERR_OR_NULL(test, filp); KUNIT_ASSERT_NOT_NULL(test, sock->sk); sock->file = filp; + fcount_before = file_count(filp); /* Act */ err = handshake_req_submit(sock, req1, GFP_KERNEL); KUNIT_ASSERT_EQ(test, err, 0); + KUNIT_EXPECT_EQ(test, file_count(filp), fcount_before + 1); err = handshake_req_submit(sock, req2, GFP_KERNEL); /* Assert */ KUNIT_EXPECT_EQ(test, err, -EBUSY); + KUNIT_EXPECT_EQ(test, file_count(filp), fcount_before + 1); handshake_req_cancel(sock->sk); + KUNIT_EXPECT_EQ(test, file_count(filp), fcount_before); fput(filp); } static void handshake_req_cancel_test1(struct kunit *test) { + unsigned long fcount_before; struct handshake_req *req; struct socket *sock; struct file *filp; @@ -329,8 +342,10 @@ static void handshake_req_cancel_test1(struct kunit *test) KUNIT_ASSERT_NOT_ERR_OR_NULL(test, filp); sock->file = filp; + fcount_before = file_count(filp); err = handshake_req_submit(sock, req, GFP_KERNEL); KUNIT_ASSERT_EQ(test, err, 0); + KUNIT_EXPECT_EQ(test, file_count(filp), fcount_before + 1); /* NB: handshake_req hasn't been accepted */ @@ -339,12 +354,14 @@ static void handshake_req_cancel_test1(struct kunit *test) /* Assert */ KUNIT_EXPECT_TRUE(test, result); + KUNIT_EXPECT_EQ(test, file_count(filp), fcount_before); fput(filp); } static void handshake_req_cancel_test2(struct kunit *test) { + unsigned long fcount_before; struct handshake_req *req, *next; struct handshake_net *hn; struct socket *sock; @@ -365,8 +382,10 @@ static void handshake_req_cancel_test2(struct kunit *test) KUNIT_ASSERT_NOT_ERR_OR_NULL(test, filp); sock->file = filp; + fcount_before = file_count(filp); err = handshake_req_submit(sock, req, GFP_KERNEL); KUNIT_ASSERT_EQ(test, err, 0); + KUNIT_EXPECT_EQ(test, file_count(filp), fcount_before + 1); net = sock_net(sock->sk); hn = handshake_pernet(net); @@ -385,12 +404,14 @@ static void handshake_req_cancel_test2(struct kunit *test) /* Assert */ KUNIT_EXPECT_TRUE(test, result); + KUNIT_EXPECT_EQ(test, file_count(filp), fcount_before); fput(filp); } static void handshake_req_cancel_test3(struct kunit *test) { + unsigned long fcount_before; struct handshake_req *req, *next; struct handshake_net *hn; struct socket *sock; @@ -411,8 +432,10 @@ static void handshake_req_cancel_test3(struct kunit *test) KUNIT_ASSERT_NOT_ERR_OR_NULL(test, filp); sock->file = filp; + fcount_before = file_count(filp); err = handshake_req_submit(sock, req, GFP_KERNEL); KUNIT_ASSERT_EQ(test, err, 0); + KUNIT_EXPECT_EQ(test, file_count(filp), fcount_before + 1); net = sock_net(sock->sk); hn = handshake_pernet(net); @@ -428,12 +451,14 @@ static void handshake_req_cancel_test3(struct kunit *test) /* Pretend to complete this request */ handshake_complete(next, -ETIMEDOUT, NULL); + KUNIT_EXPECT_EQ(test, file_count(filp), fcount_before); /* Act */ result = handshake_req_cancel(sock->sk); /* Assert */ KUNIT_EXPECT_FALSE(test, result); + KUNIT_EXPECT_EQ(test, file_count(filp), fcount_before); fput(filp); } @@ -454,6 +479,7 @@ static struct handshake_proto handshake_req_alloc_proto_destroy = { static void handshake_req_destroy_test1(struct kunit *test) { + unsigned long fcount_before; struct handshake_req *req; struct socket *sock; struct file *filp; @@ -473,10 +499,12 @@ static void handshake_req_destroy_test1(struct kunit *test) KUNIT_ASSERT_NOT_ERR_OR_NULL(test, filp); sock->file = filp; + fcount_before = file_count(filp); err = handshake_req_submit(sock, req, GFP_KERNEL); KUNIT_ASSERT_EQ(test, err, 0); handshake_req_cancel(sock->sk); + KUNIT_EXPECT_EQ(test, file_count(filp), fcount_before); /* Act */ /* Ensure the close/release/put process has run to -- cgit v1.2.3 From ea5fe6a73ca57e5150b8a38b341aef2636eb72f0 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Mon, 25 May 2026 12:51:22 -0400 Subject: net/handshake: Drain pending requests at net namespace exit The arguments to list_splice_init() in handshake_net_exit() are reversed. The call moves the local empty "requests" list onto hn->hn_requests, leaving the local list empty, so the subsequent drain loop runs zero iterations. Pending handshake requests that had not yet been accepted are not torn down when the net namespace is destroyed; each one keeps a reference on a socket file and on the handshake_req allocation. Pass the source and destination in the documented order (list_splice_init(list, head) moves list onto head) so the pending list is transferred to the local scratch list and drained through handshake_complete(). Fixing the splice direction exposes a list-corruption race. After the splice each req->hr_list still has non-empty link pointers, threading the stack-local scratch list rather than hn_requests. A concurrent handshake_req_cancel() -- for example, from sunrpc's TLS timeout on a kernel socket whose netns reference was not taken -- finds the request through the rhashtable, calls remove_pending(), and sees !list_empty(&req->hr_list). __remove_pending_locked() then list_del_init()s an entry off the scratch list while the drain iterates, corrupting it. The same call arriving after the drain loop has run list_del() on an entry hits LIST_POISON instead. Have remove_pending() check HANDSHAKE_F_NET_DRAINING under hn_lock and report not-found when drain is in progress. The drain has already taken ownership; handshake_complete()'s existing test_and_set on HANDSHAKE_F_REQ_COMPLETED still arbitrates between drain and cancel for who calls the consumer's hp_done. Use list_del_init() rather than list_del() in the drain so req->hr_list does not carry LIST_POISON after drain releases the entry. The DRAINING guard in remove_pending() makes cancel return false, but cancel still falls through to test_and_set_bit on HANDSHAKE_F_REQ_COMPLETED and drops the request's hr_file reference. Without another pin, if that is the last reference, sk_destruct frees the request while it is still linked on the drain loop's local list. Pin each request's hr_file under hn_lock before releasing the list, and drop that drain pin after the loop finishes with the request. Fixes: 3b3009ea8abb ("net/handshake: Create a NETLINK service for handling handshake requests") Signed-off-by: Chuck Lever Reviewed-by: Hannes Reinecke Link: https://patch.msgid.link/20260525-handshake-file-pin-v3-8-66c616906ead@oracle.com Signed-off-by: Paolo Abeni --- net/handshake/netlink.c | 10 ++++++++-- net/handshake/request.c | 5 ++++- 2 files changed, 12 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/handshake/netlink.c b/net/handshake/netlink.c index 21d6cbd52fcd..3fd4fef9bab1 100644 --- a/net/handshake/netlink.c +++ b/net/handshake/netlink.c @@ -201,13 +201,19 @@ static void __net_exit handshake_net_exit(struct net *net) */ spin_lock_bh(&hn->hn_lock); set_bit(HANDSHAKE_F_NET_DRAINING, &hn->hn_flags); - list_splice_init(&requests, &hn->hn_requests); + list_splice_init(&hn->hn_requests, &requests); + list_for_each_entry(req, &requests, hr_list) + get_file(req->hr_file); spin_unlock_bh(&hn->hn_lock); while (!list_empty(&requests)) { + struct file *file; + req = list_first_entry(&requests, struct handshake_req, hr_list); - list_del(&req->hr_list); + file = req->hr_file; + list_del_init(&req->hr_list); handshake_complete(req, -ETIMEDOUT, NULL); + fput(file); } } diff --git a/net/handshake/request.c b/net/handshake/request.c index bd3d9467ab91..cd30d54d0501 100644 --- a/net/handshake/request.c +++ b/net/handshake/request.c @@ -163,13 +163,16 @@ static void __remove_pending_locked(struct handshake_net *hn, * otherwise %false. * * If @req was on a pending list, it has not yet been accepted. + * Returns %false when the net namespace is draining; the drain + * loop has taken ownership of the pending list. */ static bool remove_pending(struct handshake_net *hn, struct handshake_req *req) { bool ret = false; spin_lock_bh(&hn->hn_lock); - if (!list_empty(&req->hr_list)) { + if (!test_bit(HANDSHAKE_F_NET_DRAINING, &hn->hn_flags) && + !list_empty(&req->hr_list)) { __remove_pending_locked(hn, req); ret = true; } -- cgit v1.2.3 From 00e1950716c6ed67d74777b2db286b0fa23b4be9 Mon Sep 17 00:00:00 2001 From: Zhenghang Xiao Date: Tue, 26 May 2026 18:51:52 +0800 Subject: Bluetooth: l2cap: clear chan->ident on ECRED reconfiguration success l2cap_ecred_reconf_rsp() returns early on success without clearing chan->ident. Every other L2CAP response handler (l2cap_ecred_conn_rsp, l2cap_le_connect_rsp, l2cap_config_rsp) clears chan->ident after a successful transaction to prevent the channel from matching subsequent responses with the recycled ident value. A remote attacker that completed a reconfiguration as the peer can replay a failure response with the stale ident, causing the kernel to match and destroy the already-established channel via l2cap_chan_del(chan, ECONNRESET). Clear chan->ident for all matching channels on success, and harden the failure path by using l2cap_chan_hold_unless_zero() consistent with other L2CAP handlers (l2cap_le_command_rej, __l2cap_get_chan_by_ident). Fixes: 15f02b910562 ("Bluetooth: L2CAP: Add initial code for Enhanced Credit Based Mode") Signed-off-by: Zhenghang Xiao Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/l2cap_core.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c index 5668c92b3f58..ff13c43e3588 100644 --- a/net/bluetooth/l2cap_core.c +++ b/net/bluetooth/l2cap_core.c @@ -5460,14 +5460,20 @@ static inline int l2cap_ecred_reconf_rsp(struct l2cap_conn *conn, BT_DBG("result 0x%4.4x", result); - if (!result) + if (!result) { + list_for_each_entry(chan, &conn->chan_l, list) { + if (chan->ident == cmd->ident) + chan->ident = 0; + } return 0; + } list_for_each_entry_safe(chan, tmp, &conn->chan_l, list) { if (chan->ident != cmd->ident) continue; - l2cap_chan_hold(chan); + if (!l2cap_chan_hold_unless_zero(chan)) + continue; l2cap_chan_lock(chan); l2cap_chan_del(chan, ECONNRESET); -- cgit v1.2.3 From 41c2713b204e6cb6a94587bc6bf6935107df5479 Mon Sep 17 00:00:00 2001 From: Luiz Augusto von Dentz Date: Mon, 11 May 2026 12:09:42 -0400 Subject: Bluetooth: L2CAP: Fix possible crash on l2cap_ecred_conn_rsp If dcid is received for an already-assigned destination CID the spec requires that both channels to be discarded, but calling l2cap_chan_del may invalidate the tmp cursor created by list_for_each_entry_safe and in fact it is the wrong procedure as the chan->dcid may be assigned previously it really needs to be disconnected. Calling l2cap_chan_clone directly may still lead to l2cap_chan_del so instead schedule l2cap_chan_timeout with delay 0 to close the channel asynchronously. Fixes: 15f02b910562 ("Bluetooth: L2CAP: Add initial code for Enhanced Credit Based Mode") Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/l2cap_core.c | 27 ++++++++++++++++++++++----- 1 file changed, 22 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c index ff13c43e3588..45b175399e8d 100644 --- a/net/bluetooth/l2cap_core.c +++ b/net/bluetooth/l2cap_core.c @@ -5262,6 +5262,7 @@ static inline int l2cap_ecred_conn_rsp(struct l2cap_conn *conn, cmd_len -= sizeof(*rsp); list_for_each_entry_safe(chan, tmp, &conn->chan_l, list) { + struct l2cap_chan *orig; u16 dcid; if (chan->ident != cmd->ident || @@ -5283,8 +5284,10 @@ static inline int l2cap_ecred_conn_rsp(struct l2cap_conn *conn, BT_DBG("dcid[%d] 0x%4.4x", i, dcid); + orig = __l2cap_get_chan_by_dcid(conn, dcid); + /* Check if dcid is already in use */ - if (dcid && __l2cap_get_chan_by_dcid(conn, dcid)) { + if (dcid && orig) { /* If a device receives a * L2CAP_CREDIT_BASED_CONNECTION_RSP packet with an * already-assigned Destination CID, then both the @@ -5293,10 +5296,24 @@ static inline int l2cap_ecred_conn_rsp(struct l2cap_conn *conn, */ l2cap_chan_del(chan, ECONNREFUSED); l2cap_chan_unlock(chan); - chan = __l2cap_get_chan_by_dcid(conn, dcid); - l2cap_chan_lock(chan); - l2cap_chan_del(chan, ECONNRESET); - l2cap_chan_unlock(chan); + + /* Check that the dcid channel mode is + * L2CAP_MODE_EXT_FLOWCTL since this procedure is only + * valid for that mode and shouldn't disconnect a dcid + * in other modes. + */ + if (orig->mode == L2CAP_MODE_EXT_FLOWCTL) { + l2cap_chan_lock(orig); + /* Disconnect the original channel as it may be + * considered connected since dcid has already + * been assigned; don't call l2cap_chan_close + * directly since that could lead to + * l2cap_chan_del and then removing the channel + * from the list while we're iterating over it. + */ + __set_chan_timer(orig, 0); + l2cap_chan_unlock(orig); + } continue; } -- cgit v1.2.3 From 47f23a259517abbdb8032c057a1e8a6bf3734878 Mon Sep 17 00:00:00 2001 From: Muhammad Bilal Date: Wed, 27 May 2026 04:59:17 +0000 Subject: Bluetooth: ISO: fix UAF in iso_recv_frame iso_recv_frame reads conn->sk under iso_conn_lock but releases the lock before using sk, with no reference held. A concurrent iso_sock_kill() can free sk in that window, causing use-after-free on sk->sk_state and sock_queue_rcv_skb(). Fix by replacing the bare pointer read with iso_sock_hold(conn), which calls sock_hold() while the spinlock is held, atomically elevating the refcount before the lock drops. Add a drop_put label so sock_put() is called on all exit paths where the hold succeeded. Fixes: ccf74f2390d60a2f9a75ef496d2564abb478f46a ("Bluetooth: Add BTPROTO_ISO socket type") Cc: stable@vger.kernel.org Signed-off-by: Muhammad Bilal Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/iso.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/bluetooth/iso.c b/net/bluetooth/iso.c index d7af617cda45..f03b7fa5dccc 100644 --- a/net/bluetooth/iso.c +++ b/net/bluetooth/iso.c @@ -564,7 +564,7 @@ static void iso_recv_frame(struct iso_conn *conn, struct sk_buff *skb) struct sock *sk; iso_conn_lock(conn); - sk = conn->sk; + sk = iso_sock_hold(conn); iso_conn_unlock(conn); if (!sk) @@ -573,11 +573,15 @@ static void iso_recv_frame(struct iso_conn *conn, struct sk_buff *skb) BT_DBG("sk %p len %d", sk, skb->len); if (sk->sk_state != BT_CONNECTED) - goto drop; + goto drop_put; - if (!sock_queue_rcv_skb(sk, skb)) + if (!sock_queue_rcv_skb(sk, skb)) { + sock_put(sk); return; + } +drop_put: + sock_put(sk); drop: kfree_skb(skb); } -- cgit v1.2.3 From 4b5f8e608749b7e8fa386c6e4301cf9272595859 Mon Sep 17 00:00:00 2001 From: Muhammad Bilal Date: Wed, 27 May 2026 04:59:18 +0000 Subject: Bluetooth: ISO: serialize iso_sock_clear_timer with socket lock iso_sock_close() calls iso_sock_clear_timer() before acquiring lock_sock(sk). iso_sock_clear_timer() reads iso_pi(sk)->conn twice without the socket lock held: if (!iso_pi(sk)->conn) return; cancel_delayed_work(&iso_pi(sk)->conn->timeout_work); Concurrently, iso_conn_del() executes under lock_sock(sk) and calls iso_chan_del(), which sets iso_pi(sk)->conn to NULL and may result in the final reference to the connection being dropped: CPU0 CPU1 ---- ---- iso_sock_clear_timer() if (conn != NULL) ... lock_sock(sk) iso_chan_del() iso_pi(sk)->conn = NULL cancel_delayed_work(conn) /* NULL deref or UAF */ iso_pi(sk)->conn is not stable across the unlock window, causing a NULL pointer dereference or use-after-free. Serialize iso_sock_clear_timer() with the socket lock by moving it inside lock_sock()/release_sock(), matching the pattern used in iso_conn_del() and all other call sites. Fixes: ccf74f2390d60a2f9a75ef496d2564abb478f46a ("Bluetooth: Add BTPROTO_ISO socket type") Cc: stable@vger.kernel.org Signed-off-by: Muhammad Bilal Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/iso.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/bluetooth/iso.c b/net/bluetooth/iso.c index f03b7fa5dccc..876649556d3c 100644 --- a/net/bluetooth/iso.c +++ b/net/bluetooth/iso.c @@ -864,8 +864,8 @@ static void __iso_sock_close(struct sock *sk) /* Must be called on unlocked socket. */ static void iso_sock_close(struct sock *sk) { - iso_sock_clear_timer(sk); lock_sock(sk); + iso_sock_clear_timer(sk); __iso_sock_close(sk); release_sock(sk); iso_sock_kill(sk); -- cgit v1.2.3 From 40b87657200cfae93e48904fd9c9c8fc3e192cae Mon Sep 17 00:00:00 2001 From: Heitor Alves de Siqueira Date: Tue, 26 May 2026 10:50:57 -0300 Subject: Bluetooth: hci_core: Rework hci_dev_do_reset() to use hci_sync functions The current HCI reset function in hci_core.c duplicates most of the work done by hci_dev_close_sync(), and doesn't handle LE, advertising or discovery. Instead of porting these to hci_dev_do_reset(), directly call the close/open functions from hci_sync to reset the hdev. MGMT now notifies when a user performs a reset. Suggested-by: Luiz Augusto von Dentz Signed-off-by: Heitor Alves de Siqueira Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/hci_core.c | 43 +++---------------------------------------- 1 file changed, 3 insertions(+), 40 deletions(-) (limited to 'net') diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c index c46c1236ebfa..28d7929dc593 100644 --- a/net/bluetooth/hci_core.c +++ b/net/bluetooth/hci_core.c @@ -539,46 +539,9 @@ static int hci_dev_do_reset(struct hci_dev *hdev) hci_req_sync_lock(hdev); - /* Drop queues */ - skb_queue_purge(&hdev->rx_q); - skb_queue_purge(&hdev->cmd_q); - - /* Cancel these to avoid queueing non-chained pending work */ - hci_dev_set_flag(hdev, HCI_CMD_DRAIN_WORKQUEUE); - /* Wait for - * - * if (!hci_dev_test_flag(hdev, HCI_CMD_DRAIN_WORKQUEUE)) - * queue_delayed_work(&hdev->{cmd,ncmd}_timer) - * - * inside RCU section to see the flag or complete scheduling. - */ - synchronize_rcu(); - /* Explicitly cancel works in case scheduled after setting the flag. */ - cancel_delayed_work(&hdev->cmd_timer); - cancel_delayed_work(&hdev->ncmd_timer); - - /* Avoid potential lockdep warnings from the *_flush() calls by - * ensuring the workqueue is empty up front. - */ - drain_workqueue(hdev->workqueue); - - hci_dev_lock(hdev); - hci_inquiry_cache_flush(hdev); - hci_conn_hash_flush(hdev); - hci_dev_unlock(hdev); - - if (hdev->flush) - hdev->flush(hdev); - - hci_dev_clear_flag(hdev, HCI_CMD_DRAIN_WORKQUEUE); - - atomic_set(&hdev->cmd_cnt, 1); - hdev->acl_cnt = 0; - hdev->sco_cnt = 0; - hdev->le_cnt = 0; - hdev->iso_cnt = 0; - - ret = hci_reset_sync(hdev); + ret = hci_dev_close_sync(hdev); + if (!ret) + ret = hci_dev_open_sync(hdev); hci_req_sync_unlock(hdev); return ret; -- cgit v1.2.3 From 525daaea459fc215f432de1b8debbd9144bf97b0 Mon Sep 17 00:00:00 2001 From: Heitor Alves de Siqueira Date: Tue, 26 May 2026 10:50:58 -0300 Subject: Bluetooth: hci_sync: Set HCI_CMD_DRAIN_WORKQUEUE during device close Since hci_dev_close_sync() can now be called during the reset path, we should also set HCI_CMD_DRAIN_WORKQUEUE. This avoids queuing timeouts while the hdev workqueue is being drained. Fixes: 877afadad2dc ("Bluetooth: When HCI work queue is drained, only queue chained work") Signed-off-by: Heitor Alves de Siqueira Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/hci_sync.c | 8 ++++++++ 1 file changed, 8 insertions(+) (limited to 'net') diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c index 1faf8df6d159..0f016d269c62 100644 --- a/net/bluetooth/hci_sync.c +++ b/net/bluetooth/hci_sync.c @@ -5301,6 +5301,12 @@ int hci_dev_close_sync(struct hci_dev *hdev) bt_dev_dbg(hdev, ""); + /* Set HCI_DRAIN_WORKQUEUE flag to prevent queuing work during + * reset/close. See hci_cmd_work() and handle_cmd_cnt_and_timer(). + */ + hci_dev_set_flag(hdev, HCI_CMD_DRAIN_WORKQUEUE); + synchronize_rcu(); + if (hci_dev_test_flag(hdev, HCI_UNREGISTER)) { disable_delayed_work(&hdev->power_off); disable_delayed_work(&hdev->ncmd_timer); @@ -5324,6 +5330,7 @@ int hci_dev_close_sync(struct hci_dev *hdev) if (!test_and_clear_bit(HCI_UP, &hdev->flags)) { cancel_delayed_work_sync(&hdev->cmd_timer); + hci_dev_clear_flag(hdev, HCI_CMD_DRAIN_WORKQUEUE); return err; } @@ -5423,6 +5430,7 @@ int hci_dev_close_sync(struct hci_dev *hdev) /* Clear flags */ hdev->flags &= BIT(HCI_RAW); hci_dev_clear_volatile_flags(hdev); + hci_dev_clear_flag(hdev, HCI_CMD_DRAIN_WORKQUEUE); memset(hdev->eir, 0, sizeof(hdev->eir)); memset(hdev->dev_class, 0, sizeof(hdev->dev_class)); -- cgit v1.2.3 From cdf88b35e06f1b385f7f6228060ae541d44fbb72 Mon Sep 17 00:00:00 2001 From: Heitor Alves de Siqueira Date: Tue, 26 May 2026 10:50:59 -0300 Subject: Bluetooth: hci_sync: Reset device counters in hci_dev_close_sync() Before resetting or closing the device, protocol counters should also be zeroed. Fixes: d0b137062b2d ("Bluetooth: hci_sync: Rework init stages") Signed-off-by: Heitor Alves de Siqueira Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/hci_sync.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'net') diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c index 0f016d269c62..aeccd8084cba 100644 --- a/net/bluetooth/hci_sync.c +++ b/net/bluetooth/hci_sync.c @@ -5393,6 +5393,10 @@ int hci_dev_close_sync(struct hci_dev *hdev) /* Reset device */ skb_queue_purge(&hdev->cmd_q); atomic_set(&hdev->cmd_cnt, 1); + hdev->acl_cnt = 0; + hdev->sco_cnt = 0; + hdev->le_cnt = 0; + hdev->iso_cnt = 0; if (hci_test_quirk(hdev, HCI_QUIRK_RESET_ON_CLOSE) && !auto_off && !hci_dev_test_flag(hdev, HCI_UNCONFIGURED)) { set_bit(HCI_INIT, &hdev->flags); -- cgit v1.2.3 From f14fe6395a8b3d961a61e138ad7b36ba3626dd4e Mon Sep 17 00:00:00 2001 From: Zhenghang Xiao Date: Wed, 27 May 2026 11:24:11 +0800 Subject: sctp: fix race between sctp_wait_for_connect and peeloff MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit sctp_wait_for_connect() drops and re-acquires the socket lock while waiting for the association to reach ESTABLISHED state. During this window, another thread can peeloff the association to a new socket via getsockopt(SCTP_SOCKOPT_PEELOFF), changing asoc->base.sk. After re-acquiring the old socket lock, sctp_wait_for_connect() returns success without noticing the migration — the caller then accesses the association under the wrong lock in sctp_datamsg_from_user(). Add the same sk != asoc->base.sk check that sctp_wait_for_sndbuf() already has, returning an error if the association was migrated while we slept. Fixes: 668c9beb9020 ("sctp: implement assign_number for sctp_stream_interleave") Signed-off-by: Zhenghang Xiao Acked-by: Xin Long Link: https://patch.msgid.link/20260527032411.60959-1-kipreyyy@gmail.com Signed-off-by: Jakub Kicinski --- net/sctp/socket.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'net') diff --git a/net/sctp/socket.c b/net/sctp/socket.c index 1d2568bb6bc2..66e12fb0c646 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -9403,6 +9403,8 @@ static int sctp_wait_for_connect(struct sctp_association *asoc, long *timeo_p) release_sock(sk); current_timeo = schedule_timeout(current_timeo); lock_sock(sk); + if (sk != asoc->base.sk) + goto do_error; *timeo_p = current_timeo; } -- cgit v1.2.3 From 6851161feb01cea41358c9ec304bd2f981fc8505 Mon Sep 17 00:00:00 2001 From: Steffen Klassert Date: Fri, 29 May 2026 10:23:25 +0200 Subject: Revert "esp: fix page frag reference leak on skb_to_sgvec failure" This reverts commit 2982e599fff6faa21c8df147d96fc7af6c1a2f24. The patch does not fully fix the issue and the Author does not match the 'Signed-off-by:' tag, so revert it for now. Signed-off-by: Steffen Klassert --- net/ipv4/esp4.c | 12 +++++------- net/ipv6/esp6.c | 12 +++++------- 2 files changed, 10 insertions(+), 14 deletions(-) (limited to 'net') diff --git a/net/ipv4/esp4.c b/net/ipv4/esp4.c index 5d3a8656687e..513c8215c947 100644 --- a/net/ipv4/esp4.c +++ b/net/ipv4/esp4.c @@ -96,7 +96,7 @@ static inline struct scatterlist *esp_req_sg(struct crypto_aead *aead, __alignof__(struct scatterlist)); } -static void esp_ssg_unref(struct xfrm_state *x, void *tmp, struct sk_buff *skb, bool already_unref) +static void esp_ssg_unref(struct xfrm_state *x, void *tmp, struct sk_buff *skb) { struct crypto_aead *aead = x->data; int extralen = 0; @@ -113,7 +113,7 @@ static void esp_ssg_unref(struct xfrm_state *x, void *tmp, struct sk_buff *skb, /* Unref skb_frag_pages in the src scatterlist if necessary. * Skip the first sg which comes from skb->data. */ - if (already_unref || req->src != req->dst) + if (req->src != req->dst) for (sg = sg_next(req->src); sg; sg = sg_next(sg)) skb_page_unref(page_to_netmem(sg_page(sg)), skb->pp_recycle); @@ -220,7 +220,7 @@ static void esp_output_done(void *data, int err) } tmp = ESP_SKB_CB(skb)->tmp; - esp_ssg_unref(x, tmp, skb, false); + esp_ssg_unref(x, tmp, skb); kfree(tmp); if (xo && (xo->flags & XFRM_DEV_RESUME)) { @@ -569,10 +569,8 @@ int esp_output_tail(struct xfrm_state *x, struct sk_buff *skb, struct esp_info * err = skb_to_sgvec(skb, dsg, (unsigned char *)esph - skb->data, assoclen + ivlen + esp->clen + alen); - if (unlikely(err < 0)) { - esp_ssg_unref(x, tmp, skb, true); + if (unlikely(err < 0)) goto error_free; - } } if ((x->props.flags & XFRM_STATE_ESN)) @@ -604,7 +602,7 @@ int esp_output_tail(struct xfrm_state *x, struct sk_buff *skb, struct esp_info * } if (sg != dsg) - esp_ssg_unref(x, tmp, skb, false); + esp_ssg_unref(x, tmp, skb); if (!err && x->encap && x->encap->encap_type == TCP_ENCAP_ESPINTCP) err = esp_output_tail_tcp(x, skb); diff --git a/net/ipv6/esp6.c b/net/ipv6/esp6.c index b963b8e72604..57481e423e59 100644 --- a/net/ipv6/esp6.c +++ b/net/ipv6/esp6.c @@ -113,7 +113,7 @@ static inline struct scatterlist *esp_req_sg(struct crypto_aead *aead, __alignof__(struct scatterlist)); } -static void esp_ssg_unref(struct xfrm_state *x, void *tmp, struct sk_buff *skb, bool already_unref) +static void esp_ssg_unref(struct xfrm_state *x, void *tmp, struct sk_buff *skb) { struct crypto_aead *aead = x->data; int extralen = 0; @@ -130,7 +130,7 @@ static void esp_ssg_unref(struct xfrm_state *x, void *tmp, struct sk_buff *skb, /* Unref skb_frag_pages in the src scatterlist if necessary. * Skip the first sg which comes from skb->data. */ - if (already_unref || req->src != req->dst) + if (req->src != req->dst) for (sg = sg_next(req->src); sg; sg = sg_next(sg)) skb_page_unref(page_to_netmem(sg_page(sg)), skb->pp_recycle); @@ -254,7 +254,7 @@ static void esp_output_done(void *data, int err) } tmp = ESP_SKB_CB(skb)->tmp; - esp_ssg_unref(x, tmp, skb, false); + esp_ssg_unref(x, tmp, skb); kfree(tmp); esp_output_encap_csum(skb); @@ -600,10 +600,8 @@ int esp6_output_tail(struct xfrm_state *x, struct sk_buff *skb, struct esp_info err = skb_to_sgvec(skb, dsg, (unsigned char *)esph - skb->data, assoclen + ivlen + esp->clen + alen); - if (unlikely(err < 0)) { - esp_ssg_unref(x, tmp, skb, true); + if (unlikely(err < 0)) goto error_free; - } } if ((x->props.flags & XFRM_STATE_ESN)) @@ -636,7 +634,7 @@ int esp6_output_tail(struct xfrm_state *x, struct sk_buff *skb, struct esp_info } if (sg != dsg) - esp_ssg_unref(x, tmp, skb, false); + esp_ssg_unref(x, tmp, skb); if (!err && x->encap && x->encap->encap_type == TCP_ENCAP_ESPINTCP) err = esp_output_tail_tcp(x, skb); -- cgit v1.2.3 From 1e584c304cfb94a759417130b1fc6d30b30c4cce Mon Sep 17 00:00:00 2001 From: Jingguo Tan Date: Wed, 27 May 2026 10:33:01 +0800 Subject: vsock/virtio: bind uarg before filling zerocopy skb virtio_transport_send_pkt_info() allocates or reuses the zerocopy uarg before entering the send loop, but virtio_transport_alloc_skb() still fills the skb before it inherits that uarg. When fixed-buffer vectored zerocopy hits MAX_SKB_FRAGS, io_sg_from_iter() may partially attach managed frags and return -EMSGSIZE. The rollback path call kfree_skb() to free an skb that carries SKBFL_MANAGED_FRAG_REFS but no uarg, so skb_release_data() falls through to ordinary frag unref. Pass the uarg into virtio_transport_alloc_skb() and bind it immediately before virtio_transport_fill_skb(). This keeps control or no-payload skbs untouched while ensuring success and rollback share one lifetime rule. Fixes: 581512a6dc93 ("vsock/virtio: MSG_ZEROCOPY flag support") Signed-off-by: Lin Ma Signed-off-by: Rongzhen Cui Signed-off-by: Jingguo Tan Acked-by: Arseniy Krasnov Acked-by: Michael S. Tsirkin Reviewed-by: Stefano Garzarella Link: https://patch.msgid.link/20260527023301.1075581-1-malin89@huawei.com Signed-off-by: Jakub Kicinski --- net/vmw_vsock/virtio_transport_common.c | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/vmw_vsock/virtio_transport_common.c b/net/vmw_vsock/virtio_transport_common.c index b143290a311d..b10666937c49 100644 --- a/net/vmw_vsock/virtio_transport_common.c +++ b/net/vmw_vsock/virtio_transport_common.c @@ -205,6 +205,7 @@ static u16 virtio_transport_get_type(struct sock *sk) static struct sk_buff *virtio_transport_alloc_skb(struct virtio_vsock_pkt_info *info, size_t payload_len, bool zcopy, + struct ubuf_info *uarg, u32 src_cid, u32 src_port, u32 dst_cid, @@ -245,6 +246,12 @@ static struct sk_buff *virtio_transport_alloc_skb(struct virtio_vsock_pkt_info * if (info->msg && payload_len > 0) { int err; + /* Bind the zerocopy lifetime before filling frags so error + * rollback frees managed fixed-buffer pages through + * the uarg-aware path. + */ + skb_zcopy_set(skb, uarg, NULL); + err = virtio_transport_fill_skb(skb, info, payload_len, zcopy); if (err) goto out; @@ -364,6 +371,7 @@ static int virtio_transport_send_pkt_info(struct vsock_sock *vsk, skb_len = min(max_skb_len, rest_len); skb = virtio_transport_alloc_skb(info, skb_len, can_zcopy, + uarg, src_cid, src_port, dst_cid, dst_port); if (!skb) { @@ -371,8 +379,6 @@ static int virtio_transport_send_pkt_info(struct vsock_sock *vsk, break; } - skb_zcopy_set(skb, uarg, NULL); - virtio_transport_inc_tx_pkt(vvs, skb); ret = t_ops->send_pkt(skb, info->net); @@ -1183,7 +1189,7 @@ static int virtio_transport_reset_no_sock(const struct virtio_transport *t, if (!t) return -ENOTCONN; - reply = virtio_transport_alloc_skb(&info, 0, false, + reply = virtio_transport_alloc_skb(&info, 0, false, NULL, le64_to_cpu(hdr->dst_cid), le32_to_cpu(hdr->dst_port), le64_to_cpu(hdr->src_cid), -- cgit v1.2.3 From f72eed9b84fb771019a955908132410a9ba9ea3f Mon Sep 17 00:00:00 2001 From: Yuqi Xu Date: Wed, 27 May 2026 11:48:15 +0800 Subject: bpf: sockmap: fix tail fragment offset in bpf_msg_push_data When bpf_msg_push_data() inserts data in the middle of a scatterlist entry, it splits the original entry into a left fragment and a right fragment. The right fragment offset is page-local, but the code advances it with `start`, which is the message-global insertion point. For inserts into a non-first SG entry, this over-advances the offset and leaves the split layout inconsistent. Advance the right fragment offset by the fragment-local delta, `start - offset`, which matches the length removed from the front of the original entry. Fixes: 6fff607e2f14 ("bpf: sk_msg program helper bpf_msg_push_data") Cc: stable@kernel.org Reported-by: Yuan Tan Reported-by: Zhengchuan Liang Reported-by: Xin Liu Signed-off-by: Yuqi Xu Signed-off-by: Ren Wei Link: https://patch.msgid.link/8b129d10566aa3eb43f61a8f9757bcf51707d324.1779636774.git.xuyq21@lenovo.com Signed-off-by: Jakub Kicinski --- net/core/filter.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/core/filter.c b/net/core/filter.c index 9590877b0714..80439767e0ee 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -2869,7 +2869,7 @@ BPF_CALL_4(bpf_msg_push_data, struct sk_msg *, msg, u32, start, psge->length = start - offset; rsge.length -= psge->length; - rsge.offset += start; + rsge.offset += start - offset; sk_msg_iter_var_next(i); sg_unmark_end(psge); -- cgit v1.2.3 From 9f72412bcf60144f252b0d6205106abf14344abc Mon Sep 17 00:00:00 2001 From: Jiayuan Chen Date: Wed, 27 May 2026 13:31:30 +0800 Subject: ipv6: fix possible infinite loop in rt6_fill_node() Sashiko reported this issue [1]. Apply the same fix as commit f8d8ce1b515a ("ipv6: fix possible infinite loop in fib6_info_uses_dev()"). Writers holding tb6_lock can list_del_rcu(&rt->fib6_siblings) without waiting for RCU readers; rt->fib6_siblings.next then still points into the old ring and this softirq-side walker never reaches &rt->fib6_siblings, causing a CPU stall. fib6_del_route() always WRITE_ONCE()s rt->fib6_nsiblings to 0 before list_del_rcu(), so an inside-loop check is a reliable detach signal. [1] https://sashiko.dev/#/patchset/20260526020227.4857-1-jiayuan.chen%40linux.dev Fixes: d9ccb18f83ea ("ipv6: Fix soft lockups in fib6_select_path under high next hop churn") Signed-off-by: Jiayuan Chen Reviewed-by: Ido Schimmel Link: https://patch.msgid.link/20260527053133.180695-1-jiayuan.chen@linux.dev Signed-off-by: Jakub Kicinski --- net/ipv6/route.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'net') diff --git a/net/ipv6/route.c b/net/ipv6/route.c index b106e5fef9cb..dad416fdc585 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -5902,6 +5902,8 @@ static int rt6_fill_node(struct net *net, struct sk_buff *skb, goto nla_put_failure; } + if (!READ_ONCE(rt->fib6_nsiblings)) + break; } rcu_read_unlock(); -- cgit v1.2.3 From 9c7da87c2dc860bb17ca1ece942495d28b1ce3b9 Mon Sep 17 00:00:00 2001 From: Jiayuan Chen Date: Wed, 27 May 2026 13:31:31 +0800 Subject: ipv6: fix possible infinite loop in fib6_select_path() Found while auditing the same pattern Sashiko reported in rt6_fill_node() [1]. Apply the same fix as commit f8d8ce1b515a ("ipv6: fix possible infinite loop in fib6_info_uses_dev()"). Writers holding tb6_lock can list_del_rcu(&first->fib6_siblings) without waiting for RCU readers; first->fib6_siblings.next then still points into the old ring and this softirq-side walker never reaches &first->fib6_siblings as its terminator. fib6_purge_rt() always WRITE_ONCE()s first->fib6_nsiblings to 0 before list_del_rcu(), so an inside-loop check is a reliable detach signal. [1] https://sashiko.dev/#/patchset/20260526020227.4857-1-jiayuan.chen%40linux.dev Fixes: d9ccb18f83ea ("ipv6: Fix soft lockups in fib6_select_path under high next hop churn") Signed-off-by: Jiayuan Chen Reviewed-by: Ido Schimmel Link: https://patch.msgid.link/20260527053133.180695-2-jiayuan.chen@linux.dev Signed-off-by: Jakub Kicinski --- net/ipv6/route.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'net') diff --git a/net/ipv6/route.c b/net/ipv6/route.c index dad416fdc585..636f0120d7e3 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -481,6 +481,9 @@ void fib6_select_path(const struct net *net, struct fib6_result *res, const struct fib6_nh *nh = sibling->fib6_nh; int nh_upper_bound; + if (!READ_ONCE(first->fib6_nsiblings)) + break; + nh_upper_bound = atomic_read(&nh->fib_nh_upper_bound); if (hash > nh_upper_bound) continue; -- cgit v1.2.3 From ff6e798c2eac3ebd0501ad7e796f583fab928de8 Mon Sep 17 00:00:00 2001 From: Pavel Begunkov Date: Thu, 28 May 2026 19:43:53 +0100 Subject: net: skbuff: fix pskb_carve leaking zcopy pages When SKBFL_MANAGED_FRAG_REFS is set, frag pages are not refcounted but their lifetime is controlled by the attached ubuf_info. To make a copy of the skb_shared_info, we either should clear the flag and reference the frags, or keep the flag and have frags unreferenced. pskb_carve_inside_header() and pskb_carve_inside_nonlinear() don't follow the rule and thus can leak page references. Let's clear SKBFL_MANAGED_FRAG_REFS from the original skb to fix it. It's the simplest way to address it, but there are more performant ways to do that if it ever becomes a problem. Link: https://lore.kernel.org/all/20260523085809.26331-1-nvminh232@clc.fitus.edu.vn/ Fixes: 753f1ca4e1e50 ("net: introduce managed frags infrastructure") Reported-by: Minh Nguyen Reported-by: Willem de Bruijn Signed-off-by: Pavel Begunkov Reviewed-by: Willem de Bruijn Link: https://patch.msgid.link/1e2086aa69217d7f9c8da3d38f5be7160f1b4cd1.1779993185.git.asml.silence@gmail.com Signed-off-by: Jakub Kicinski --- net/core/skbuff.c | 10 ++++++++++ 1 file changed, 10 insertions(+) (limited to 'net') diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 0d3cc115f2e7..c02f0a507ba8 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -6823,6 +6823,11 @@ static int pskb_carve_inside_header(struct sk_buff *skb, const u32 off, skb_copy_from_linear_data_offset(skb, off, data, new_hlen); skb->len -= off; + /* Remove SKBFL_MANAGED_FRAG_REFS instead of trying to honour it + * while refcounting frags below. + */ + skb_zcopy_downgrade_managed(skb); + memcpy((struct skb_shared_info *)(data + size), skb_shinfo(skb), offsetof(struct skb_shared_info, @@ -6936,6 +6941,11 @@ static int pskb_carve_inside_nonlinear(struct sk_buff *skb, const u32 off, return -ENOMEM; size = SKB_WITH_OVERHEAD(size); + /* Remove SKBFL_MANAGED_FRAG_REFS instead of trying to honour it + * while refcounting frags below. + */ + skb_zcopy_downgrade_managed(skb); + memcpy((struct skb_shared_info *)(data + size), skb_shinfo(skb), offsetof(struct skb_shared_info, frags[0])); if (skb_orphan_frags(skb, gfp_mask)) { -- cgit v1.2.3 From 072aa0f5c3d8f11f3159037418ec45edce7440b8 Mon Sep 17 00:00:00 2001 From: Fernando Fernandez Mancera Date: Fri, 29 May 2026 13:23:57 +0200 Subject: Revert "ipv6: preserve insertion order for same-scope addresses" Chris Adams reported that preserving insertion order for same-scope addresses is causing SSH connections to be dropped after stopping a VM while running NetworkManager. NetworkManager caches the IPv6 address configuration, when a RA arrives, it determines the list of addresses to configure and checks if the addresses are already in the right order in the kernel. If they aren't, NetworkManager removes and re-adds them to achieve the desired order. As the order changes, NetworkManager is confused and reconfigures the addresses on every update. In addition, this would also affect to cloud tooling that relies on IPv6 addresses order to identify primary and secondaries addresses. This reverts commit cb3de96eea66f5e4a580086c6a1be46e765f97f4. Fixes: cb3de96eea66 ("ipv6: preserve insertion order for same-scope addresses") Reported-by: Chris Adams Closes: https://lore.kernel.org/netdev/20260521135310.GC977@cmadams.net/ Signed-off-by: Fernando Fernandez Mancera Link: https://patch.msgid.link/20260529112357.5079-1-fmancera@suse.de Signed-off-by: Jakub Kicinski --- net/ipv6/addrconf.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 5476b6536eb7..bb84a78b80f6 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -1013,7 +1013,7 @@ ipv6_link_dev_addr(struct inet6_dev *idev, struct inet6_ifaddr *ifp) list_for_each(p, &idev->addr_list) { struct inet6_ifaddr *ifa = list_entry(p, struct inet6_ifaddr, if_list); - if (ifp_scope > ipv6_addr_src_scope(&ifa->addr)) + if (ifp_scope >= ipv6_addr_src_scope(&ifa->addr)) break; } -- cgit v1.2.3