From 7bcc10585bd9b762820fd8ffdc389aafa56ad689 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Sun, 19 Apr 2020 14:38:00 -0400 Subject: NFS/pnfs: Ensure that _pnfs_return_layout() waits for layoutreturn completion We require that any outstanding layout return completes before we can free up the inode so that the layout itself can be freed. Signed-off-by: Trond Myklebust --- fs/nfs/pnfs.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'fs/nfs') diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c index b8d78f393365..3bf6899cba95 100644 --- a/fs/nfs/pnfs.c +++ b/fs/nfs/pnfs.c @@ -1332,13 +1332,15 @@ _pnfs_return_layout(struct inode *ino) !valid_layout) { spin_unlock(&ino->i_lock); dprintk("NFS: %s no layout segments to return\n", __func__); - goto out_put_layout_hdr; + goto out_wait_layoutreturn; } send = pnfs_prepare_layoutreturn(lo, &stateid, &cred, NULL); spin_unlock(&ino->i_lock); if (send) status = pnfs_send_layoutreturn(lo, &stateid, &cred, IOMODE_ANY, true); +out_wait_layoutreturn: + wait_on_bit(&lo->plh_flags, NFS_LAYOUT_RETURN, TASK_UNINTERRUPTIBLE); out_put_layout_hdr: pnfs_free_lseg_list(&tmp_list); pnfs_put_layout_hdr(lo); -- cgit v1.2.3 From 4d8948c7334a970eab113a4bcf45b1920a5a1625 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Sun, 19 Apr 2020 23:53:52 -0400 Subject: NFS/pnfs: Fix a credential use-after-free issue in pnfs_roc() If the credential returned by pnfs_prepare_layoutreturn() does not match the credential of the RPC call, then we do end up calling pnfs_send_layoutreturn() with that credential, so don't free it! Fixes: 44ea8dfce021 ("NFS/pnfs: Reference the layout cred in pnfs_prepare_layoutreturn()") Signed-off-by: Trond Myklebust --- fs/nfs/pnfs.c | 7 ++----- 1 file changed, 2 insertions(+), 5 deletions(-) (limited to 'fs/nfs') diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c index 3bf6899cba95..dd2e14f5875d 100644 --- a/fs/nfs/pnfs.c +++ b/fs/nfs/pnfs.c @@ -1458,18 +1458,15 @@ retry: /* lo ref dropped in pnfs_roc_release() */ layoutreturn = pnfs_prepare_layoutreturn(lo, &stateid, &lc_cred, &iomode); /* If the creds don't match, we can't compound the layoutreturn */ - if (!layoutreturn) + if (!layoutreturn || cred_fscmp(cred, lc_cred) != 0) goto out_noroc; - if (cred_fscmp(cred, lc_cred) != 0) - goto out_noroc_put_cred; roc = layoutreturn; pnfs_init_layoutreturn_args(args, lo, &stateid, iomode); res->lrs_present = 0; layoutreturn = false; - -out_noroc_put_cred: put_cred(lc_cred); + out_noroc: spin_unlock(&ino->i_lock); rcu_read_unlock(); -- cgit v1.2.3 From 7648f939cb919b9d15c21fff8cd9eba908d595dc Mon Sep 17 00:00:00 2001 From: Andreas Gruenbacher Date: Mon, 20 Apr 2020 15:51:47 +0200 Subject: nfs: Fix potential posix_acl refcnt leak in nfs3_set_acl nfs3_set_acl keeps track of the acl it allocated locally to determine if an acl needs to be released at the end. This results in a memory leak when the function allocates an acl as well as a default acl. Fix by releasing acls that differ from the acl originally passed into nfs3_set_acl. Fixes: b7fa0554cf1b ("[PATCH] NFS: Add support for NFSv3 ACLs") Reported-by: Xiyu Yang Signed-off-by: Andreas Gruenbacher Signed-off-by: Trond Myklebust --- fs/nfs/nfs3acl.c | 22 +++++++++++++++------- 1 file changed, 15 insertions(+), 7 deletions(-) (limited to 'fs/nfs') diff --git a/fs/nfs/nfs3acl.c b/fs/nfs/nfs3acl.c index c5c3fc6e6c60..26c94b32d6f4 100644 --- a/fs/nfs/nfs3acl.c +++ b/fs/nfs/nfs3acl.c @@ -253,37 +253,45 @@ int nfs3_proc_setacls(struct inode *inode, struct posix_acl *acl, int nfs3_set_acl(struct inode *inode, struct posix_acl *acl, int type) { - struct posix_acl *alloc = NULL, *dfacl = NULL; + struct posix_acl *orig = acl, *dfacl = NULL, *alloc; int status; if (S_ISDIR(inode->i_mode)) { switch(type) { case ACL_TYPE_ACCESS: - alloc = dfacl = get_acl(inode, ACL_TYPE_DEFAULT); + alloc = get_acl(inode, ACL_TYPE_DEFAULT); if (IS_ERR(alloc)) goto fail; + dfacl = alloc; break; case ACL_TYPE_DEFAULT: - dfacl = acl; - alloc = acl = get_acl(inode, ACL_TYPE_ACCESS); + alloc = get_acl(inode, ACL_TYPE_ACCESS); if (IS_ERR(alloc)) goto fail; + dfacl = acl; + acl = alloc; break; } } if (acl == NULL) { - alloc = acl = posix_acl_from_mode(inode->i_mode, GFP_KERNEL); + alloc = posix_acl_from_mode(inode->i_mode, GFP_KERNEL); if (IS_ERR(alloc)) goto fail; + acl = alloc; } status = __nfs3_proc_setacls(inode, acl, dfacl); - posix_acl_release(alloc); +out: + if (acl != orig) + posix_acl_release(acl); + if (dfacl != orig) + posix_acl_release(dfacl); return status; fail: - return PTR_ERR(alloc); + status = PTR_ERR(alloc); + goto out; } const struct xattr_handler *nfs3_xattr_handlers[] = { -- cgit v1.2.3 From 6e47666ef93dc9c4011407df8d2de1dd1ed39f25 Mon Sep 17 00:00:00 2001 From: Xiyu Yang Date: Sat, 25 Apr 2020 21:04:40 +0800 Subject: NFSv4: Remove unreachable error condition due to rpc_run_task() nfs4_proc_layoutget() invokes rpc_run_task(), which return the value to "task". Since rpc_run_task() is impossible to return an ERR pointer, there is no need to add the IS_ERR() condition on "task" here. So we need to remove it. Signed-off-by: Xiyu Yang Signed-off-by: Xin Tan Signed-off-by: Trond Myklebust --- fs/nfs/nfs4proc.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'fs/nfs') diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 512afb1c7867..1c710a7834c2 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -9191,8 +9191,7 @@ nfs4_proc_layoutget(struct nfs4_layoutget *lgp, long *timeout) nfs4_init_sequence(&lgp->args.seq_args, &lgp->res.seq_res, 0, 0); task = rpc_run_task(&task_setup_data); - if (IS_ERR(task)) - return ERR_CAST(task); + status = rpc_wait_for_completion_task(task); if (status != 0) goto out; -- cgit v1.2.3 From dff58530c4ca8ce7ee5a74db431c6e35362cf682 Mon Sep 17 00:00:00 2001 From: Olga Kornievskaia Date: Fri, 24 Apr 2020 17:45:50 -0400 Subject: NFSv4.1: fix handling of backchannel binding in BIND_CONN_TO_SESSION Currently, if the client sends BIND_CONN_TO_SESSION with NFS4_CDFC4_FORE_OR_BOTH but only gets NFS4_CDFS4_FORE back it ignores that it wasn't able to enable a backchannel. To make sure, the client sends BIND_CONN_TO_SESSION as the first operation on the connections (ie., no other session compounds haven't been sent before), and if the client's request to bind the backchannel is not satisfied, then reset the connection and retry. Cc: stable@vger.kernel.org Signed-off-by: Olga Kornievskaia Signed-off-by: Trond Myklebust --- fs/nfs/nfs4proc.c | 8 ++++++++ 1 file changed, 8 insertions(+) (limited to 'fs/nfs') diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 1c710a7834c2..a0c1e653a935 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -7891,6 +7891,7 @@ static void nfs4_bind_one_conn_to_session_done(struct rpc_task *task, void *calldata) { struct nfs41_bind_conn_to_session_args *args = task->tk_msg.rpc_argp; + struct nfs41_bind_conn_to_session_res *res = task->tk_msg.rpc_resp; struct nfs_client *clp = args->client; switch (task->tk_status) { @@ -7899,6 +7900,12 @@ nfs4_bind_one_conn_to_session_done(struct rpc_task *task, void *calldata) nfs4_schedule_session_recovery(clp->cl_session, task->tk_status); } + if (args->dir == NFS4_CDFC4_FORE_OR_BOTH && + res->dir != NFS4_CDFS4_BOTH) { + rpc_task_close_connection(task); + if (args->retries++ < MAX_BIND_CONN_TO_SESSION_RETRIES) + rpc_restart_call(task); + } } static const struct rpc_call_ops nfs4_bind_one_conn_to_session_ops = { @@ -7921,6 +7928,7 @@ int nfs4_proc_bind_one_conn_to_session(struct rpc_clnt *clnt, struct nfs41_bind_conn_to_session_args args = { .client = clp, .dir = NFS4_CDFC4_FORE_OR_BOTH, + .retries = 0, }; struct nfs41_bind_conn_to_session_res res; struct rpc_message msg = { -- cgit v1.2.3 From 9c07b75b80eeff714420fb6a4c880b284e529d0f Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Thu, 30 Apr 2020 15:08:26 -0400 Subject: NFS: Fix a race in __nfs_list_for_each_server() The struct nfs_server gets put on the cl_superblocks list before the server->super field has been initialised, in which case the call to nfs_sb_active() will Oops. Add a check to ensure that we skip such a list entry. Fixes: 3c9e502b59fb ("NFS: Add a helper nfs_client_for_each_server()") Signed-off-by: Trond Myklebust --- fs/nfs/super.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs/nfs') diff --git a/fs/nfs/super.c b/fs/nfs/super.c index 59ef3b13ccca..bdb6d0c2e755 100644 --- a/fs/nfs/super.c +++ b/fs/nfs/super.c @@ -185,7 +185,7 @@ static int __nfs_list_for_each_server(struct list_head *head, rcu_read_lock(); list_for_each_entry_rcu(server, head, client_link) { - if (!nfs_sb_active(server->super)) + if (!(server->super && nfs_sb_active(server->super))) continue; rcu_read_unlock(); if (last) -- cgit v1.2.3 From d9bfced1fbcb35b28d8fbed4e785d2807055ed2b Mon Sep 17 00:00:00 2001 From: Dave Wysochanski Date: Wed, 15 Apr 2020 16:14:41 -0400 Subject: NFS: Fix fscache super_cookie index_key from changing after umount Commit 402cb8dda949 ("fscache: Attach the index key and aux data to the cookie") added the index_key and index_key_len parameters to fscache_acquire_cookie(), and updated the callers in the NFS client. One of the callers was inside nfs_fscache_get_super_cookie() and was changed to use the full struct nfs_fscache_key as the index_key. However, a couple members of this structure contain pointers and thus will change each time the same NFS share is remounted. Since index_key is used for fscache_cookie->key_hash and this subsequently is used to compare cookies, the effectiveness of fscache with NFS is reduced to the point at which a umount occurs. Any subsequent remount of the same share will cause a unique NFS super_block index_key and key_hash to be generated for the same data, rendering any prior fscache data unable to be found. A simple reproducer demonstrates the problem. 1. Mount share with 'fsc', create a file, drop page cache systemctl start cachefilesd mount -o vers=3,fsc 127.0.0.1:/export /mnt dd if=/dev/zero of=/mnt/file1.bin bs=4096 count=1 echo 3 > /proc/sys/vm/drop_caches 2. Read file into page cache and fscache, then unmount dd if=/mnt/file1.bin of=/dev/null bs=4096 count=1 umount /mnt 3. Remount and re-read which should come from fscache mount -o vers=3,fsc 127.0.0.1:/export /mnt echo 3 > /proc/sys/vm/drop_caches dd if=/mnt/file1.bin of=/dev/null bs=4096 count=1 4. Check for READ ops in mountstats - there should be none grep READ: /proc/self/mountstats Looking at the history and the removed function, nfs_super_get_key(), we should only use nfs_fscache_key.key plus any uniquifier, for the fscache index_key. Fixes: 402cb8dda949 ("fscache: Attach the index key and aux data to the cookie") Signed-off-by: Dave Wysochanski Signed-off-by: David Howells --- fs/nfs/fscache.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'fs/nfs') diff --git a/fs/nfs/fscache.c b/fs/nfs/fscache.c index 1abf126c2df4..8eff1fd806b1 100644 --- a/fs/nfs/fscache.c +++ b/fs/nfs/fscache.c @@ -188,7 +188,8 @@ void nfs_fscache_get_super_cookie(struct super_block *sb, const char *uniq, int /* create a cache index for looking up filehandles */ nfss->fscache = fscache_acquire_cookie(nfss->nfs_client->fscache, &nfs_fscache_super_index_def, - key, sizeof(*key) + ulen, + &key->key, + sizeof(key->key) + ulen, NULL, 0, nfss, 0, true); dfprintk(FSCACHE, "NFS: get superblock cookie (0x%p/0x%p)\n", -- cgit v1.2.3 From 15751612734ca0c419ac43ce986c9badcb5e2829 Mon Sep 17 00:00:00 2001 From: Dave Wysochanski Date: Wed, 15 Apr 2020 16:14:42 -0400 Subject: NFS: Fix fscache super_cookie allocation Commit f2aedb713c28 ("NFS: Add fs_context support.") reworked NFS mount code paths for fs_context support which included super_block initialization. In the process there was an extra return left in the code and so we never call nfs_fscache_get_super_cookie even if 'fsc' is given on as mount option. In addition, there is an extra check inside nfs_fscache_get_super_cookie for the NFS_OPTION_FSCACHE which is unnecessary since the only caller nfs_get_cache_cookie checks this flag. Fixes: f2aedb713c28 ("NFS: Add fs_context support.") Signed-off-by: Dave Wysochanski Signed-off-by: David Howells --- fs/nfs/fscache.c | 2 -- fs/nfs/super.c | 1 - 2 files changed, 3 deletions(-) (limited to 'fs/nfs') diff --git a/fs/nfs/fscache.c b/fs/nfs/fscache.c index 8eff1fd806b1..f51718415606 100644 --- a/fs/nfs/fscache.c +++ b/fs/nfs/fscache.c @@ -118,8 +118,6 @@ void nfs_fscache_get_super_cookie(struct super_block *sb, const char *uniq, int nfss->fscache_key = NULL; nfss->fscache = NULL; - if (!(nfss->options & NFS_OPTION_FSCACHE)) - return; if (!uniq) { uniq = ""; ulen = 1; diff --git a/fs/nfs/super.c b/fs/nfs/super.c index 59ef3b13ccca..cc34aa3a8ba4 100644 --- a/fs/nfs/super.c +++ b/fs/nfs/super.c @@ -1189,7 +1189,6 @@ static void nfs_get_cache_cookie(struct super_block *sb, uniq = ctx->fscache_uniq; ulen = strlen(ctx->fscache_uniq); } - return; } nfs_fscache_get_super_cookie(sb, uniq, ulen); -- cgit v1.2.3 From 50eaa652b54df1e2b48dc398d9e6114c9ed080eb Mon Sep 17 00:00:00 2001 From: Dave Wysochanski Date: Thu, 16 Apr 2020 06:06:08 -0400 Subject: NFSv4: Fix fscache cookie aux_data to ensure change_attr is included Commit 402cb8dda949 ("fscache: Attach the index key and aux data to the cookie") added the aux_data and aux_data_len to parameters to fscache_acquire_cookie(), and updated the callers in the NFS client. In the process it modified the aux_data to include the change_attr, but missed adding change_attr to a couple places where aux_data was used. Specifically, when opening a file and the change_attr is not added, the following attempt to lookup an object will fail inside cachefiles_check_object_xattr() = -116 due to nfs_fscache_inode_check_aux() failing memcmp on auxdata and returning FSCACHE_CHECKAUX_OBSOLETE. Fix this by adding nfs_fscache_update_auxdata() to set the auxdata from all relevant fields in the inode, including the change_attr. Fixes: 402cb8dda949 ("fscache: Attach the index key and aux data to the cookie") Signed-off-by: Dave Wysochanski Signed-off-by: David Howells --- fs/nfs/fscache.c | 34 ++++++++++++++++------------------ 1 file changed, 16 insertions(+), 18 deletions(-) (limited to 'fs/nfs') diff --git a/fs/nfs/fscache.c b/fs/nfs/fscache.c index f51718415606..a60df88efc40 100644 --- a/fs/nfs/fscache.c +++ b/fs/nfs/fscache.c @@ -225,6 +225,19 @@ void nfs_fscache_release_super_cookie(struct super_block *sb) } } +static void nfs_fscache_update_auxdata(struct nfs_fscache_inode_auxdata *auxdata, + struct nfs_inode *nfsi) +{ + memset(auxdata, 0, sizeof(*auxdata)); + auxdata->mtime_sec = nfsi->vfs_inode.i_mtime.tv_sec; + auxdata->mtime_nsec = nfsi->vfs_inode.i_mtime.tv_nsec; + auxdata->ctime_sec = nfsi->vfs_inode.i_ctime.tv_sec; + auxdata->ctime_nsec = nfsi->vfs_inode.i_ctime.tv_nsec; + + if (NFS_SERVER(&nfsi->vfs_inode)->nfs_client->rpc_ops->version == 4) + auxdata->change_attr = inode_peek_iversion_raw(&nfsi->vfs_inode); +} + /* * Initialise the per-inode cache cookie pointer for an NFS inode. */ @@ -238,14 +251,7 @@ void nfs_fscache_init_inode(struct inode *inode) if (!(nfss->fscache && S_ISREG(inode->i_mode))) return; - memset(&auxdata, 0, sizeof(auxdata)); - auxdata.mtime_sec = nfsi->vfs_inode.i_mtime.tv_sec; - auxdata.mtime_nsec = nfsi->vfs_inode.i_mtime.tv_nsec; - auxdata.ctime_sec = nfsi->vfs_inode.i_ctime.tv_sec; - auxdata.ctime_nsec = nfsi->vfs_inode.i_ctime.tv_nsec; - - if (NFS_SERVER(&nfsi->vfs_inode)->nfs_client->rpc_ops->version == 4) - auxdata.change_attr = inode_peek_iversion_raw(&nfsi->vfs_inode); + nfs_fscache_update_auxdata(&auxdata, nfsi); nfsi->fscache = fscache_acquire_cookie(NFS_SB(inode->i_sb)->fscache, &nfs_fscache_inode_object_def, @@ -265,11 +271,7 @@ void nfs_fscache_clear_inode(struct inode *inode) dfprintk(FSCACHE, "NFS: clear cookie (0x%p/0x%p)\n", nfsi, cookie); - memset(&auxdata, 0, sizeof(auxdata)); - auxdata.mtime_sec = nfsi->vfs_inode.i_mtime.tv_sec; - auxdata.mtime_nsec = nfsi->vfs_inode.i_mtime.tv_nsec; - auxdata.ctime_sec = nfsi->vfs_inode.i_ctime.tv_sec; - auxdata.ctime_nsec = nfsi->vfs_inode.i_ctime.tv_nsec; + nfs_fscache_update_auxdata(&auxdata, nfsi); fscache_relinquish_cookie(cookie, &auxdata, false); nfsi->fscache = NULL; } @@ -309,11 +311,7 @@ void nfs_fscache_open_file(struct inode *inode, struct file *filp) if (!fscache_cookie_valid(cookie)) return; - memset(&auxdata, 0, sizeof(auxdata)); - auxdata.mtime_sec = nfsi->vfs_inode.i_mtime.tv_sec; - auxdata.mtime_nsec = nfsi->vfs_inode.i_mtime.tv_nsec; - auxdata.ctime_sec = nfsi->vfs_inode.i_ctime.tv_sec; - auxdata.ctime_nsec = nfsi->vfs_inode.i_ctime.tv_nsec; + nfs_fscache_update_auxdata(&auxdata, nfsi); if (inode_is_open_for_write(inode)) { dfprintk(FSCACHE, "NFS: nfsi 0x%p disabling cache\n", nfsi); -- cgit v1.2.3 From 29fe839976266bc7c55b927360a1daae57477723 Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Mon, 11 May 2020 10:02:48 -0400 Subject: nfs: fix NULL deference in nfs4_get_valid_delegation We add the new state to the nfsi->open_states list, making it potentially visible to other threads, before we've finished initializing it. That wasn't a problem when all the readers were also taking the i_lock (as we do here), but since we switched to RCU, there's now a possibility that a reader could see the partially initialized state. Symptoms observed were a crash when another thread called nfs4_get_valid_delegation() on a NULL inode, resulting in an oops like: BUG: unable to handle page fault for address: ffffffffffffffb0 ... RIP: 0010:nfs4_get_valid_delegation+0x6/0x30 [nfsv4] ... Call Trace: nfs4_open_prepare+0x80/0x1c0 [nfsv4] __rpc_execute+0x75/0x390 [sunrpc] ? finish_task_switch+0x75/0x260 rpc_async_schedule+0x29/0x40 [sunrpc] process_one_work+0x1ad/0x370 worker_thread+0x30/0x390 ? create_worker+0x1a0/0x1a0 kthread+0x10c/0x130 ? kthread_park+0x80/0x80 ret_from_fork+0x22/0x30 Fixes: 9ae075fdd190 "NFSv4: Convert open state lookup to use RCU" Reviewed-by: Seiichi Ikarashi Tested-by: Daisuke Matsuda Tested-by: Masayoshi Mizuma Signed-off-by: J. Bruce Fields Cc: stable@vger.kernel.org # v4.20+ Signed-off-by: Trond Myklebust --- fs/nfs/nfs4state.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs/nfs') diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c index ac93715c05a4..a8dc25ce48bb 100644 --- a/fs/nfs/nfs4state.c +++ b/fs/nfs/nfs4state.c @@ -734,9 +734,9 @@ nfs4_get_open_state(struct inode *inode, struct nfs4_state_owner *owner) state = new; state->owner = owner; atomic_inc(&owner->so_count); - list_add_rcu(&state->inode_states, &nfsi->open_states); ihold(inode); state->inode = inode; + list_add_rcu(&state->inode_states, &nfsi->open_states); spin_unlock(&inode->i_lock); /* Note: The reclaim code dictates that we add stateless * and read-only stateids to the end of the list */ -- cgit v1.2.3 From f304a809a9f1eec23e023f3aae6c768a08495244 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Mon, 11 May 2020 10:42:04 -0400 Subject: NFS: Don't use RPC_TASK_CRED_NOREF with delegreturn We are not guaranteed that the credential will remain pinned. Fixes: 612965072020 ("NFSv4: Avoid referencing the cred unnecessarily during NFSv4 I/O") Signed-off-by: Trond Myklebust --- fs/nfs/nfs4proc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs/nfs') diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index a0c1e653a935..9056f3dd380e 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -6347,7 +6347,7 @@ static int _nfs4_proc_delegreturn(struct inode *inode, const struct cred *cred, .rpc_client = server->client, .rpc_message = &msg, .callback_ops = &nfs4_delegreturn_ops, - .flags = RPC_TASK_ASYNC | RPC_TASK_CRED_NOREF | RPC_TASK_TIMEOUT, + .flags = RPC_TASK_ASYNC | RPC_TASK_TIMEOUT, }; int status = 0; -- cgit v1.2.3 From 4fa7ef69e2918eb89c95e57978d839c0a31f332c Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 13 May 2020 09:55:36 -0400 Subject: NFS/pnfs: Don't use RPC_TASK_CRED_NOREF with pnfs When we're doing pnfs then the credential being used for the RPC call is not necessarily the same as the one used in the open context, so don't use RPC_TASK_CRED_NOREF. Fixes: 612965072020 ("NFSv4: Avoid referencing the cred unnecessarily during NFSv4 I/O") Signed-off-by: Trond Myklebust --- fs/nfs/pagelist.c | 5 +++-- fs/nfs/pnfs_nfs.c | 3 ++- fs/nfs/write.c | 4 ++-- 3 files changed, 7 insertions(+), 5 deletions(-) (limited to 'fs/nfs') diff --git a/fs/nfs/pagelist.c b/fs/nfs/pagelist.c index f61f96603df7..6ca421cbe19c 100644 --- a/fs/nfs/pagelist.c +++ b/fs/nfs/pagelist.c @@ -752,7 +752,7 @@ int nfs_initiate_pgio(struct rpc_clnt *clnt, struct nfs_pgio_header *hdr, .callback_ops = call_ops, .callback_data = hdr, .workqueue = nfsiod_workqueue, - .flags = RPC_TASK_ASYNC | RPC_TASK_CRED_NOREF | flags, + .flags = RPC_TASK_ASYNC | flags, }; hdr->rw_ops->rw_initiate(hdr, &msg, rpc_ops, &task_setup_data, how); @@ -950,7 +950,8 @@ static int nfs_generic_pg_pgios(struct nfs_pageio_descriptor *desc) hdr->cred, NFS_PROTO(hdr->inode), desc->pg_rpc_callops, - desc->pg_ioflags, 0); + desc->pg_ioflags, + RPC_TASK_CRED_NOREF); return ret; } diff --git a/fs/nfs/pnfs_nfs.c b/fs/nfs/pnfs_nfs.c index e7ddbce48321..679767ac258d 100644 --- a/fs/nfs/pnfs_nfs.c +++ b/fs/nfs/pnfs_nfs.c @@ -536,7 +536,8 @@ pnfs_generic_commit_pagelist(struct inode *inode, struct list_head *mds_pages, nfs_init_commit(data, NULL, NULL, cinfo); nfs_initiate_commit(NFS_CLIENT(inode), data, NFS_PROTO(data->inode), - data->mds_ops, how, 0); + data->mds_ops, how, + RPC_TASK_CRED_NOREF); } else { nfs_init_commit(data, NULL, data->lseg, cinfo); initiate_commit(data, how); diff --git a/fs/nfs/write.c b/fs/nfs/write.c index df4b87c30ac9..1e767f779c49 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -1695,7 +1695,7 @@ int nfs_initiate_commit(struct rpc_clnt *clnt, struct nfs_commit_data *data, .callback_ops = call_ops, .callback_data = data, .workqueue = nfsiod_workqueue, - .flags = RPC_TASK_ASYNC | RPC_TASK_CRED_NOREF | flags, + .flags = RPC_TASK_ASYNC | flags, .priority = priority, }; /* Set up the initial task struct. */ @@ -1813,7 +1813,7 @@ nfs_commit_list(struct inode *inode, struct list_head *head, int how, nfs_init_commit(data, head, NULL, cinfo); atomic_inc(&cinfo->mds->rpcs_out); return nfs_initiate_commit(NFS_CLIENT(inode), data, NFS_PROTO(inode), - data->mds_ops, how, 0); + data->mds_ops, how, RPC_TASK_CRED_NOREF); } /* -- cgit v1.2.3 From 8eed292bc8cbf737e46fb1c119d4c8f6dcb00650 Mon Sep 17 00:00:00 2001 From: Olga Kornievskaia Date: Thu, 14 May 2020 16:09:40 -0400 Subject: NFSv3: fix rpc receive buffer size for MOUNT call Prior to commit e3d3ab64dd66 ("SUNRPC: Use au_rslack when computing reply buffer size"), there was enough slack in the reply buffer to commodate filehandles of size 60bytes. However, the real problem was that the reply buffer size for the MOUNT operation was not correctly calculated. Received buffer size used the filehandle size for NFSv2 (32bytes) which is much smaller than the allowed filehandle size for the v3 mounts. Fix the reply buffer size (decode arguments size) for the MNT command. Fixes: 2c94b8eca1a2 ("SUNRPC: Use au_rslack when computing reply buffer size") Signed-off-by: Olga Kornievskaia Signed-off-by: Trond Myklebust --- fs/nfs/mount_clnt.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'fs/nfs') diff --git a/fs/nfs/mount_clnt.c b/fs/nfs/mount_clnt.c index 35c8cb2d7637..dda5c3e65d8d 100644 --- a/fs/nfs/mount_clnt.c +++ b/fs/nfs/mount_clnt.c @@ -30,6 +30,7 @@ #define encode_dirpath_sz (1 + XDR_QUADLEN(MNTPATHLEN)) #define MNT_status_sz (1) #define MNT_fhandle_sz XDR_QUADLEN(NFS2_FHSIZE) +#define MNT_fhandlev3_sz XDR_QUADLEN(NFS3_FHSIZE) #define MNT_authflav3_sz (1 + NFS_MAX_SECFLAVORS) /* @@ -37,7 +38,7 @@ */ #define MNT_enc_dirpath_sz encode_dirpath_sz #define MNT_dec_mountres_sz (MNT_status_sz + MNT_fhandle_sz) -#define MNT_dec_mountres3_sz (MNT_status_sz + MNT_fhandle_sz + \ +#define MNT_dec_mountres3_sz (MNT_status_sz + MNT_fhandlev3_sz + \ MNT_authflav3_sz) /* -- cgit v1.2.3