diff options
| author | Eric Dumazet <edumazet@google.com> | 2026-01-29 15:34:58 +0000 |
|---|---|---|
| committer | Jakub Kicinski <kuba@kernel.org> | 2026-01-30 17:15:51 -0800 |
| commit | ed9b70040d7b22552f1392bed529ef0861f2a25c (patch) | |
| tree | 060312d9e749a0dbe4712785a9ae10f827409218 /net | |
| parent | 6ad42b24413630199cb9f35d073f71768788f8dc (diff) | |
tcp: reduce tcp sockets size by one cache line
By default, when a kmem_cache is created with SLAB_TYPESAFE_BY_RCU,
slub has to use extra storage for the freelist pointer after each
object, because slub assumes that any bit in the object
can be used by RCU readers.
Because proto_register() is also using SLAB_HWCACHE_ALIGN,
this forces slub to use one extra cache line per object.
We can instead put the slub freelist anywhere in the object,
granted the concurrent RCU readers are not supposed to
use the pointer value.
Add a new (struct sock)sk_freeptr field, in an union
with sk_rcu: No RCU readers would need to look at sk_rcu,
which is only used at free phase.
Tested:
grep . /sys/kernel/slab/TCP/{object_size,slab_size,objs_per_slab}
grep . /sys/kernel/slab/TCPv6/{object_size,slab_size,objs_per_slab}
Before:
/sys/kernel/slab/TCP/object_size:2368
/sys/kernel/slab/TCP/slab_size:2432
/sys/kernel/slab/TCP/objs_per_slab:13
/sys/kernel/slab/TCPv6/object_size:2496
/sys/kernel/slab/TCPv6/slab_size:2560
/sys/kernel/slab/TCPv6/objs_per_slab:12
After this patch, we can pack one more TCPv6 object per slab,
and object_size == slab_size.
/sys/kernel/slab/TCP/object_size:2368
/sys/kernel/slab/TCP/slab_size:2368
/sys/kernel/slab/TCP/objs_per_slab:13
/sys/kernel/slab/TCPv6/object_size:2496
/sys/kernel/slab/TCPv6/slab_size:2496
/sys/kernel/slab/TCPv6/objs_per_slab:13
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20260129153458.4163797-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Diffstat (limited to 'net')
| -rw-r--r-- | net/core/sock.c | 16 | ||||
| -rw-r--r-- | net/ipv4/tcp_ipv4.c | 2 | ||||
| -rw-r--r-- | net/ipv6/tcp_ipv6.c | 2 |
3 files changed, 14 insertions, 6 deletions
diff --git a/net/core/sock.c b/net/core/sock.c index a1c8b47b0d56..693e6d80f501 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -4193,13 +4193,17 @@ int proto_register(struct proto *prot, int alloc_slab) return -EINVAL; } if (alloc_slab) { - prot->slab = kmem_cache_create_usercopy(prot->name, - prot->obj_size, 0, - SLAB_HWCACHE_ALIGN | SLAB_ACCOUNT | - prot->slab_flags, - prot->useroffset, prot->usersize, - NULL); + struct kmem_cache_args args = { + .useroffset = prot->useroffset, + .usersize = prot->usersize, + .freeptr_offset = prot->freeptr_offset, + .use_freeptr_offset = !!prot->freeptr_offset, + }; + prot->slab = kmem_cache_create(prot->name, prot->obj_size, + &args, + SLAB_HWCACHE_ALIGN | SLAB_ACCOUNT | + prot->slab_flags); if (prot->slab == NULL) { pr_crit("%s: Can't create sock SLAB cache!\n", prot->name); diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index ffdf52fbf646..0fc8a42921aa 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -3460,6 +3460,8 @@ struct proto tcp_prot = { .sysctl_rmem_offset = offsetof(struct net, ipv4.sysctl_tcp_rmem), .max_header = MAX_TCP_HEADER, .obj_size = sizeof(struct tcp_sock), + .freeptr_offset = offsetof(struct tcp_sock, + inet_conn.icsk_inet.sk.sk_freeptr), .slab_flags = SLAB_TYPESAFE_BY_RCU, .twsk_prot = &tcp_timewait_sock_ops, .rsk_prot = &tcp_request_sock_ops, diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 4ae664b05fa9..8bf29186c15f 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -2332,6 +2332,8 @@ struct proto tcpv6_prot = { .sysctl_rmem_offset = offsetof(struct net, ipv4.sysctl_tcp_rmem), .max_header = MAX_TCP_HEADER, .obj_size = sizeof(struct tcp6_sock), + .freeptr_offset = offsetof(struct tcp6_sock, + tcp.inet_conn.icsk_inet.sk.sk_freeptr), .ipv6_pinfo_offset = offsetof(struct tcp6_sock, inet6), .slab_flags = SLAB_TYPESAFE_BY_RCU, .twsk_prot = &tcp6_timewait_sock_ops, |