From 342a67f0884293639bd17ea44df754ead799e669 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 7 Jul 2021 22:08:32 -0400 Subject: [PATCH 01/84] NFS: Label the dentry with a verifier in nfs_link(), nfs_symlink() After the success of an operation such as link(), or symlink(), we expect to add the dentry back to the dcache as an ordinary positive dentry. However in NFS, unless it is labelled with the appropriate verifier for the parent directory state, then nfs_lookup_revalidate will end up discarding that dentry and forcing a new lookup. The fix is to ensure that we relabel the dentry appropriately on success. Signed-off-by: Trond Myklebust --- fs/nfs/dir.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index 1a6d2867fba4..baca036f3890 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -2352,6 +2352,8 @@ int nfs_symlink(struct user_namespace *mnt_userns, struct inode *dir, return error; } + nfs_set_verifier(dentry, nfs_save_change_attribute(dir)); + /* * No big deal if we can't add this page to the page cache here. * READLINK will get the missing page from the server if needed. @@ -2385,6 +2387,7 @@ nfs_link(struct dentry *old_dentry, struct inode *dir, struct dentry *dentry) d_drop(dentry); error = NFS_PROTO(dir)->link(inode, dir, &dentry->d_name); if (error == 0) { + nfs_set_verifier(dentry, nfs_save_change_attribute(dir)); ihold(inode); d_add(dentry, inode); } From 9019fb391de02cbff422090768b73afe9f6174df Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 7 Jul 2021 21:43:09 -0400 Subject: [PATCH 02/84] NFS: Label the dentry with a verifier in nfs_rmdir() and nfs_unlink() After the success of an operation such as rmdir() or unlink(), we expect to add the dentry back to the dcache as an ordinary negative dentry. However in NFS, unless it is labelled with the appropriate verifier for the parent directory state, then nfs_lookup_revalidate will end up discarding that dentry and forcing a new lookup. The fix is to ensure that we relabel the dentry appropriately on success. Signed-off-by: Trond Myklebust --- fs/nfs/dir.c | 18 +++++++++++++++--- 1 file changed, 15 insertions(+), 3 deletions(-) diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index baca036f3890..1ce1fa0a5926 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -2197,6 +2197,18 @@ static void nfs_dentry_handle_enoent(struct dentry *dentry) d_delete(dentry); } +static void nfs_dentry_remove_handle_error(struct inode *dir, + struct dentry *dentry, int error) +{ + switch (error) { + case -ENOENT: + d_delete(dentry); + fallthrough; + case 0: + nfs_set_verifier(dentry, nfs_save_change_attribute(dir)); + } +} + int nfs_rmdir(struct inode *dir, struct dentry *dentry) { int error; @@ -2219,6 +2231,7 @@ int nfs_rmdir(struct inode *dir, struct dentry *dentry) up_write(&NFS_I(d_inode(dentry))->rmdir_sem); } else error = NFS_PROTO(dir)->rmdir(dir, &dentry->d_name); + nfs_dentry_remove_handle_error(dir, dentry, error); trace_nfs_rmdir_exit(dir, dentry, error); return error; @@ -2288,9 +2301,8 @@ int nfs_unlink(struct inode *dir, struct dentry *dentry) } spin_unlock(&dentry->d_lock); error = nfs_safe_remove(dentry); - if (!error || error == -ENOENT) { - nfs_set_verifier(dentry, nfs_save_change_attribute(dir)); - } else if (need_rehash) + nfs_dentry_remove_handle_error(dir, dentry, error); + if (need_rehash) d_rehash(dentry); out: trace_nfs_unlink_exit(dir, dentry, error); From ca05cbae2a0468e5d78e9b4605936a8bf5da328b Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Sat, 10 Jul 2021 18:07:14 -0400 Subject: [PATCH 03/84] NFS: Fix up nfs_ctx_key_to_expire() If the cached credential exists but doesn't have any expiration callback then exit early. Fix up atomicity issues when replacing the credential with a new one since the existing code could lead to refcount leaks. Signed-off-by: Trond Myklebust --- fs/nfs/inode.c | 4 ++-- fs/nfs/write.c | 41 ++++++++++++++++++++++++++++------------- include/linux/nfs_fs.h | 2 +- 3 files changed, 31 insertions(+), 16 deletions(-) diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index 853213b3a209..4f45281c47cf 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -1024,7 +1024,7 @@ struct nfs_open_context *alloc_nfs_open_context(struct dentry *dentry, ctx->cred = get_cred(filp->f_cred); else ctx->cred = get_current_cred(); - ctx->ll_cred = NULL; + rcu_assign_pointer(ctx->ll_cred, NULL); ctx->state = NULL; ctx->mode = f_mode; ctx->flags = 0; @@ -1063,7 +1063,7 @@ static void __put_nfs_open_context(struct nfs_open_context *ctx, int is_sync) put_cred(ctx->cred); dput(ctx->dentry); nfs_sb_deactive(sb); - put_rpccred(ctx->ll_cred); + put_rpccred(rcu_dereference_protected(ctx->ll_cred, 1)); kfree(ctx->mdsthreshold); kfree_rcu(ctx, rcu_head); } diff --git a/fs/nfs/write.c b/fs/nfs/write.c index eae9bf114041..773ea2c8504d 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -1246,7 +1246,7 @@ nfs_key_timeout_notify(struct file *filp, struct inode *inode) struct nfs_open_context *ctx = nfs_file_open_context(filp); if (nfs_ctx_key_to_expire(ctx, inode) && - !ctx->ll_cred) + !rcu_access_pointer(ctx->ll_cred)) /* Already expired! */ return -EACCES; return 0; @@ -1258,23 +1258,38 @@ nfs_key_timeout_notify(struct file *filp, struct inode *inode) bool nfs_ctx_key_to_expire(struct nfs_open_context *ctx, struct inode *inode) { struct rpc_auth *auth = NFS_SERVER(inode)->client->cl_auth; - struct rpc_cred *cred = ctx->ll_cred; + struct rpc_cred *cred, *new, *old = NULL; struct auth_cred acred = { .cred = ctx->cred, }; + bool ret = false; - if (cred && !cred->cr_ops->crmatch(&acred, cred, 0)) { - put_rpccred(cred); - ctx->ll_cred = NULL; - cred = NULL; - } - if (!cred) - cred = auth->au_ops->lookup_cred(auth, &acred, 0); - if (!cred || IS_ERR(cred)) + rcu_read_lock(); + cred = rcu_dereference(ctx->ll_cred); + if (cred && !(cred->cr_ops->crkey_timeout && + cred->cr_ops->crkey_timeout(cred))) + goto out; + rcu_read_unlock(); + + new = auth->au_ops->lookup_cred(auth, &acred, 0); + if (new == cred) { + put_rpccred(new); return true; - ctx->ll_cred = cred; - return !!(cred->cr_ops->crkey_timeout && - cred->cr_ops->crkey_timeout(cred)); + } + if (IS_ERR_OR_NULL(new)) { + new = NULL; + ret = true; + } else if (new->cr_ops->crkey_timeout && + new->cr_ops->crkey_timeout(new)) + ret = true; + + rcu_read_lock(); + old = rcu_dereference_protected(xchg(&ctx->ll_cred, + RCU_INITIALIZER(new)), 1); +out: + rcu_read_unlock(); + put_rpccred(old); + return ret; } /* diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index b9a8b925db43..9b75448ce0df 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -81,7 +81,7 @@ struct nfs_open_context { fl_owner_t flock_owner; struct dentry *dentry; const struct cred *cred; - struct rpc_cred *ll_cred; /* low-level cred - use to check for expiry */ + struct rpc_cred __rcu *ll_cred; /* low-level cred - use to check for expiry */ struct nfs4_state *state; fmode_t mode; From ea7a1019d8baf8503ecd6e3ec8436dec283569e6 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Mon, 12 Jul 2021 09:52:59 -0400 Subject: [PATCH 04/84] SUNRPC: Partial revert of commit 6f9f17287e78 The premise of commit 6f9f17287e78 ("SUNRPC: Mitigate cond_resched() in xprt_transmit()") was that cond_resched() is expensive and unnecessary when there has been just a single send. The point of cond_resched() is to ensure that tasks that should pre-empt this one get a chance to do so when it is safe to do so. The code prior to commit 6f9f17287e78 failed to take into account that it was keeping a rpc_task pinned for longer than it needed to, and so rather than doing a full revert, let's just move the cond_resched. Signed-off-by: Trond Myklebust --- net/sunrpc/xprt.c | 28 +++++++++++++++------------- 1 file changed, 15 insertions(+), 13 deletions(-) diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c index cfd681700d1a..d4aeee83763e 100644 --- a/net/sunrpc/xprt.c +++ b/net/sunrpc/xprt.c @@ -1603,15 +1603,14 @@ xprt_transmit(struct rpc_task *task) { struct rpc_rqst *next, *req = task->tk_rqstp; struct rpc_xprt *xprt = req->rq_xprt; - int counter, status; + int status; spin_lock(&xprt->queue_lock); - counter = 0; - while (!list_empty(&xprt->xmit_queue)) { - if (++counter == 20) + for (;;) { + next = list_first_entry_or_null(&xprt->xmit_queue, + struct rpc_rqst, rq_xmit); + if (!next) break; - next = list_first_entry(&xprt->xmit_queue, - struct rpc_rqst, rq_xmit); xprt_pin_rqst(next); spin_unlock(&xprt->queue_lock); status = xprt_request_transmit(next, task); @@ -1619,13 +1618,16 @@ xprt_transmit(struct rpc_task *task) status = 0; spin_lock(&xprt->queue_lock); xprt_unpin_rqst(next); - if (status == 0) { - if (!xprt_request_data_received(task) || - test_bit(RPC_TASK_NEED_XMIT, &task->tk_runstate)) - continue; - } else if (test_bit(RPC_TASK_NEED_XMIT, &task->tk_runstate)) - task->tk_status = status; - break; + if (status < 0) { + if (test_bit(RPC_TASK_NEED_XMIT, &task->tk_runstate)) + task->tk_status = status; + break; + } + /* Was @task transmitted, and has it received a reply? */ + if (xprt_request_data_received(task) && + !test_bit(RPC_TASK_NEED_XMIT, &task->tk_runstate)) + break; + cond_resched_lock(&xprt->queue_lock); } spin_unlock(&xprt->queue_lock); } From 47dd8796a31e132f9e2b93a4f558a9f924a7388f Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Mon, 12 Jul 2021 09:57:08 -0400 Subject: [PATCH 05/84] SUNRPC: Add cond_resched() at the appropriate point in __rpc_execute() Allow tasks that need to pre-empt rpciod/xprtiod to do so when it is safe. Signed-off-by: Trond Myklebust --- net/sunrpc/sched.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/net/sunrpc/sched.c b/net/sunrpc/sched.c index c045f63d11fa..dc46130d46a0 100644 --- a/net/sunrpc/sched.c +++ b/net/sunrpc/sched.c @@ -903,8 +903,10 @@ static void __rpc_execute(struct rpc_task *task) /* * Lockless check for whether task is sleeping or not. */ - if (!RPC_IS_QUEUED(task)) + if (!RPC_IS_QUEUED(task)) { + cond_resched(); continue; + } /* * Signalled tasks should exit rather than sleep. From 6dbcbe3f78bec62a4a96ac9cfddaf894a140b821 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Mon, 12 Jul 2021 11:57:15 -0400 Subject: [PATCH 06/84] SUNRPC: Remove WQ_HIGHPRI from xprtiod Don't let xprtiod pre-empt softirq. Signed-off-by: Trond Myklebust --- net/sunrpc/sched.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/net/sunrpc/sched.c b/net/sunrpc/sched.c index dc46130d46a0..b21457cec8a5 100644 --- a/net/sunrpc/sched.c +++ b/net/sunrpc/sched.c @@ -1232,8 +1232,7 @@ static int rpciod_start(void) if (!wq) goto out_failed; rpciod_workqueue = wq; - /* Note: highpri because network receive is latency sensitive */ - wq = alloc_workqueue("xprtiod", WQ_UNBOUND|WQ_MEM_RECLAIM|WQ_HIGHPRI, 0); + wq = alloc_workqueue("xprtiod", WQ_UNBOUND | WQ_MEM_RECLAIM, 0); if (!wq) goto free_rpciod; xprtiod_workqueue = wq; From b9f8713f42af11ae6d7f63075334ba5298436be6 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Mon, 12 Jul 2021 12:24:15 -0400 Subject: [PATCH 07/84] SUNRPC: Remove unnecessary memory barriers The only check for RPC_TASK_RUNNING is the one in rpc_make_runnable(), which happens under the same spin lock held when we call rpc_clear_running(). Ditto, the last check for RPC_TASK_QUEUED in rpc_execute() is performed under the same lock as the one held when we call rpc_clear_queued(). Signed-off-by: Trond Myklebust --- include/linux/sunrpc/sched.h | 16 ++-------------- 1 file changed, 2 insertions(+), 14 deletions(-) diff --git a/include/linux/sunrpc/sched.h b/include/linux/sunrpc/sched.h index a237b8dbf608..db964bb63912 100644 --- a/include/linux/sunrpc/sched.h +++ b/include/linux/sunrpc/sched.h @@ -150,25 +150,13 @@ struct rpc_task_setup { #define RPC_TASK_MSG_PIN_WAIT 5 #define RPC_TASK_SIGNALLED 6 -#define RPC_IS_RUNNING(t) test_bit(RPC_TASK_RUNNING, &(t)->tk_runstate) -#define rpc_set_running(t) set_bit(RPC_TASK_RUNNING, &(t)->tk_runstate) #define rpc_test_and_set_running(t) \ test_and_set_bit(RPC_TASK_RUNNING, &(t)->tk_runstate) -#define rpc_clear_running(t) \ - do { \ - smp_mb__before_atomic(); \ - clear_bit(RPC_TASK_RUNNING, &(t)->tk_runstate); \ - smp_mb__after_atomic(); \ - } while (0) +#define rpc_clear_running(t) clear_bit(RPC_TASK_RUNNING, &(t)->tk_runstate) #define RPC_IS_QUEUED(t) test_bit(RPC_TASK_QUEUED, &(t)->tk_runstate) #define rpc_set_queued(t) set_bit(RPC_TASK_QUEUED, &(t)->tk_runstate) -#define rpc_clear_queued(t) \ - do { \ - smp_mb__before_atomic(); \ - clear_bit(RPC_TASK_QUEUED, &(t)->tk_runstate); \ - smp_mb__after_atomic(); \ - } while (0) +#define rpc_clear_queued(t) clear_bit(RPC_TASK_QUEUED, &(t)->tk_runstate) #define RPC_IS_ACTIVATED(t) test_bit(RPC_TASK_ACTIVE, &(t)->tk_runstate) From 33c3214bf450051db99d352cfeef7e0ffcbb8614 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Mon, 12 Jul 2021 12:34:34 -0400 Subject: [PATCH 08/84] SUNRPC: xprt_clear_locked() only needs release memory semantics The clearing of the XPRT_LOCKED bit has to happen after we clear xprt->snd_task, but we don't require any extra memory barriers after that. Signed-off-by: Trond Myklebust --- net/sunrpc/xprt.c | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-) diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c index d4aeee83763e..48560188e84d 100644 --- a/net/sunrpc/xprt.c +++ b/net/sunrpc/xprt.c @@ -246,11 +246,9 @@ EXPORT_SYMBOL_GPL(xprt_find_transport_ident); static void xprt_clear_locked(struct rpc_xprt *xprt) { xprt->snd_task = NULL; - if (!test_bit(XPRT_CLOSE_WAIT, &xprt->state)) { - smp_mb__before_atomic(); - clear_bit(XPRT_LOCKED, &xprt->state); - smp_mb__after_atomic(); - } else + if (!test_bit(XPRT_CLOSE_WAIT, &xprt->state)) + clear_bit_unlock(XPRT_LOCKED, &xprt->state); + else queue_work(xprtiod_workqueue, &xprt->task_cleanup); } From 43d20e80e2880a1791d87d8b3fc062e91cd2ec4b Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Tue, 13 Jul 2021 12:28:22 -0400 Subject: [PATCH 09/84] NFS: Fix a few more clear_bit() instances that need release semantics All these bits are being used as bit locks. Signed-off-by: Trond Myklebust --- fs/nfs/nfs4state.c | 5 +---- fs/nfs/pagelist.c | 10 ++-------- 2 files changed, 3 insertions(+), 12 deletions(-) diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c index f22818a80c2c..ecc4594299d6 100644 --- a/fs/nfs/nfs4state.c +++ b/fs/nfs/nfs4state.c @@ -1194,10 +1194,7 @@ static int nfs4_run_state_manager(void *); static void nfs4_clear_state_manager_bit(struct nfs_client *clp) { - smp_mb__before_atomic(); - clear_bit(NFS4CLNT_MANAGER_RUNNING, &clp->cl_state); - smp_mb__after_atomic(); - wake_up_bit(&clp->cl_state, NFS4CLNT_MANAGER_RUNNING); + clear_and_wake_up_bit(NFS4CLNT_MANAGER_RUNNING, &clp->cl_state); rpc_wake_up(&clp->cl_rpcwaitq); } diff --git a/fs/nfs/pagelist.c b/fs/nfs/pagelist.c index cc232d1f16f2..72333bcaa4c4 100644 --- a/fs/nfs/pagelist.c +++ b/fs/nfs/pagelist.c @@ -271,8 +271,7 @@ nfs_page_set_headlock(struct nfs_page *req) void nfs_page_clear_headlock(struct nfs_page *req) { - smp_mb__before_atomic(); - clear_bit(PG_HEADLOCK, &req->wb_flags); + clear_bit_unlock(PG_HEADLOCK, &req->wb_flags); smp_mb__after_atomic(); if (!test_bit(PG_CONTENDED1, &req->wb_flags)) return; @@ -525,12 +524,7 @@ nfs_create_subreq(struct nfs_page *req, */ void nfs_unlock_request(struct nfs_page *req) { - if (!NFS_WBACK_BUSY(req)) { - printk(KERN_ERR "NFS: Invalid unlock attempted\n"); - BUG(); - } - smp_mb__before_atomic(); - clear_bit(PG_BUSY, &req->wb_flags); + clear_bit_unlock(PG_BUSY, &req->wb_flags); smp_mb__after_atomic(); if (!test_bit(PG_CONTENDED2, &req->wb_flags)) return; From a1e7f30a86062380ac804b50491fd24bb9dfb99f Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 14 Jul 2021 13:00:58 -0400 Subject: [PATCH 10/84] NFSv4: Retrieve ACCESS on open if we're not using NFS4_CREATE_EXCLUSIVE NFS4_CREATE_EXCLUSIVE does not allow the caller to set an access mode, so for most Linux filesystems, the access call ends up returning no permissions. However both NFS4_CREATE_EXCLUSIVE4_1 and NFS4_CREATE_GUARDED allow the client to set the access mode. Signed-off-by: Trond Myklebust --- fs/nfs/nfs4proc.c | 45 ++++++++++++++++++++++----------------------- 1 file changed, 22 insertions(+), 23 deletions(-) diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index e1214bb6b7ee..9e89198ea21f 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -1388,27 +1388,22 @@ static struct nfs4_opendata *nfs4_opendata_alloc(struct dentry *dentry, sizeof(p->o_arg.u.verifier.data)); } } - /* don't put an ACCESS op in OPEN compound if O_EXCL, because ACCESS - * will return permission denied for all bits until close */ - if (!(flags & O_EXCL)) { - /* ask server to check for all possible rights as results - * are cached */ - switch (p->o_arg.claim) { - default: - break; - case NFS4_OPEN_CLAIM_NULL: - case NFS4_OPEN_CLAIM_FH: - p->o_arg.access = NFS4_ACCESS_READ | - NFS4_ACCESS_MODIFY | - NFS4_ACCESS_EXTEND | - NFS4_ACCESS_EXECUTE; + /* ask server to check for all possible rights as results + * are cached */ + switch (p->o_arg.claim) { + default: + break; + case NFS4_OPEN_CLAIM_NULL: + case NFS4_OPEN_CLAIM_FH: + p->o_arg.access = NFS4_ACCESS_READ | NFS4_ACCESS_MODIFY | + NFS4_ACCESS_EXTEND | NFS4_ACCESS_DELETE | + NFS4_ACCESS_EXECUTE; #ifdef CONFIG_NFS_V4_2 - if (server->caps & NFS_CAP_XATTR) - p->o_arg.access |= NFS4_ACCESS_XAREAD | - NFS4_ACCESS_XAWRITE | - NFS4_ACCESS_XALIST; + if (!(server->caps & NFS_CAP_XATTR)) + break; + p->o_arg.access |= NFS4_ACCESS_XAREAD | NFS4_ACCESS_XAWRITE | + NFS4_ACCESS_XALIST; #endif - } } p->o_arg.clientid = server->nfs_client->cl_clientid; p->o_arg.id.create_time = ktime_to_ns(sp->so_seqid.create_time); @@ -2472,11 +2467,15 @@ static void nfs4_open_prepare(struct rpc_task *task, void *calldata) /* Set the create mode (note dependency on the session type) */ data->o_arg.createmode = NFS4_CREATE_UNCHECKED; if (data->o_arg.open_flags & O_EXCL) { - data->o_arg.createmode = NFS4_CREATE_EXCLUSIVE; - if (nfs4_has_persistent_session(clp)) + data->o_arg.createmode = NFS4_CREATE_EXCLUSIVE4_1; + if (clp->cl_mvops->minor_version == 0) { + data->o_arg.createmode = NFS4_CREATE_EXCLUSIVE; + /* don't put an ACCESS op in OPEN compound if O_EXCL, + * because ACCESS will return permission denied for + * all bits until close */ + data->o_res.access_request = data->o_arg.access = 0; + } else if (nfs4_has_persistent_session(clp)) data->o_arg.createmode = NFS4_CREATE_GUARDED; - else if (clp->cl_mvops->minor_version > 0) - data->o_arg.createmode = NFS4_CREATE_EXCLUSIVE4_1; } return; unlock_no_action: From eea413308f2e6deb00f061f18081a53f3ecc8cc6 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Sun, 26 Sep 2021 14:05:04 -0400 Subject: [PATCH 11/84] NFS: Default change_attr_type to NFS4_CHANGE_TYPE_IS_UNDEFINED Both NFSv3 and NFSv2 generate their change attribute from the ctime value that was supplied by the server. However the problem is that there are plenty of servers out there with ctime resolutions of 1ms or worse. In a modern performance system, this is insufficient when trying to decide which is the most recent set of attributes when, for instance, a READ or GETATTR call races with a WRITE or SETATTR. For this reason, let's revert to labelling the NFSv2/v3 change attributes as NFS4_CHANGE_TYPE_IS_UNDEFINED. This will ensure we protect against such races. Fixes: 7b24dacf0840 ("NFS: Another inode revalidation improvement") Signed-off-by: Trond Myklebust Tested-by: Chuck Lever --- fs/nfs/inode.c | 4 +++- fs/nfs/nfs3xdr.c | 2 +- fs/nfs/proc.c | 2 +- 3 files changed, 5 insertions(+), 3 deletions(-) diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index 4f45281c47cf..0f092ccb0ca1 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -1777,8 +1777,10 @@ static int nfs_inode_finish_partial_attr_update(const struct nfs_fattr *fattr, NFS_INO_INVALID_BLOCKS | NFS_INO_INVALID_OTHER | NFS_INO_INVALID_NLINK; unsigned long cache_validity = NFS_I(inode)->cache_validity; + enum nfs4_change_attr_type ctype = NFS_SERVER(inode)->change_attr_type; - if (!(cache_validity & NFS_INO_INVALID_CHANGE) && + if (ctype != NFS4_CHANGE_TYPE_IS_UNDEFINED && + !(cache_validity & NFS_INO_INVALID_CHANGE) && (cache_validity & check_valid) != 0 && (fattr->valid & NFS_ATTR_FATTR_CHANGE) != 0 && nfs_inode_attrs_cmp_monotonic(fattr, inode) == 0) diff --git a/fs/nfs/nfs3xdr.c b/fs/nfs/nfs3xdr.c index e6eca1d7481b..9274c9c5efea 100644 --- a/fs/nfs/nfs3xdr.c +++ b/fs/nfs/nfs3xdr.c @@ -2227,7 +2227,7 @@ static int decode_fsinfo3resok(struct xdr_stream *xdr, /* ignore properties */ result->lease_time = 0; - result->change_attr_type = NFS4_CHANGE_TYPE_IS_TIME_METADATA; + result->change_attr_type = NFS4_CHANGE_TYPE_IS_UNDEFINED; return 0; } diff --git a/fs/nfs/proc.c b/fs/nfs/proc.c index ea19dbf12301..ecc4e717808c 100644 --- a/fs/nfs/proc.c +++ b/fs/nfs/proc.c @@ -91,7 +91,7 @@ nfs_proc_get_root(struct nfs_server *server, struct nfs_fh *fhandle, info->dtpref = fsinfo.tsize; info->maxfilesize = 0x7FFFFFFF; info->lease_time = 0; - info->change_attr_type = NFS4_CHANGE_TYPE_IS_TIME_METADATA; + info->change_attr_type = NFS4_CHANGE_TYPE_IS_UNDEFINED; return 0; } From 488796ec1e39fb9194cc8175f770823d40fbf0ed Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Tue, 28 Sep 2021 11:15:53 -0400 Subject: [PATCH 12/84] NFS: Don't set NFS_INO_DATA_INVAL_DEFER and NFS_INO_INVALID_DATA NFS_INO_DATA_INVAL_DEFER and NFS_INO_INVALID_DATA should be considered mutually exclusive. Fixes: 1c341b777501 ("NFS: Add deferred cache invalidation for close-to-open consistency violations") Signed-off-by: Trond Myklebust Tested-by: Benjamin Coddington Reviewed-by: Benjamin Coddington --- fs/nfs/inode.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index 0f092ccb0ca1..dcb885b7ad73 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -210,10 +210,15 @@ void nfs_set_cache_invalid(struct inode *inode, unsigned long flags) flags &= ~NFS_INO_INVALID_XATTR; if (flags & NFS_INO_INVALID_DATA) nfs_fscache_invalidate(inode); - if (inode->i_mapping->nrpages == 0) - flags &= ~(NFS_INO_INVALID_DATA|NFS_INO_DATA_INVAL_DEFER); flags &= ~(NFS_INO_REVAL_PAGECACHE | NFS_INO_REVAL_FORCED); + nfsi->cache_validity |= flags; + + if (inode->i_mapping->nrpages == 0) + nfsi->cache_validity &= ~(NFS_INO_INVALID_DATA | + NFS_INO_DATA_INVAL_DEFER); + else if (nfsi->cache_validity & NFS_INO_INVALID_DATA) + nfsi->cache_validity &= ~NFS_INO_DATA_INVAL_DEFER; } EXPORT_SYMBOL_GPL(nfs_set_cache_invalid); From a6a361c4ca3cc3e6f3b39d1b6bca1de90f5f4b11 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Tue, 28 Sep 2021 11:24:57 -0400 Subject: [PATCH 13/84] NFS: Ignore the directory size when marking for revalidation If we want to revalidate the directory, then just mark the change attribute as invalid. Fixes: 13c0b082b6a9 ("NFS: Replace use of NFS_INO_REVAL_PAGECACHE when checking cache validity") Signed-off-by: Trond Myklebust Tested-by: Benjamin Coddington Reviewed-by: Benjamin Coddington --- fs/nfs/dir.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index 1ce1fa0a5926..f2df664db020 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -1413,7 +1413,7 @@ out_force: static void nfs_mark_dir_for_revalidate(struct inode *inode) { spin_lock(&inode->i_lock); - nfs_set_cache_invalid(inode, NFS_INO_REVAL_PAGECACHE); + nfs_set_cache_invalid(inode, NFS_INO_INVALID_CHANGE); spin_unlock(&inode->i_lock); } From 2929bc3329f4c7e4df400acca2b1844492650bfd Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Tue, 28 Sep 2021 12:37:05 -0400 Subject: [PATCH 14/84] NFS: Fix up nfs_readdir_inode_mapping_valid() The check for duplicate readdir cookies should only care if the change attribute is invalid or the data cache is invalid. Signed-off-by: Trond Myklebust Tested-by: Benjamin Coddington Reviewed-by: Benjamin Coddington --- fs/nfs/dir.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index f2df664db020..fa4d33687d2b 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -411,7 +411,8 @@ out_eof: static bool nfs_readdir_inode_mapping_valid(struct nfs_inode *nfsi) { - if (nfsi->cache_validity & (NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA)) + if (nfsi->cache_validity & (NFS_INO_INVALID_CHANGE | + NFS_INO_INVALID_DATA)) return false; smp_rmb(); return !test_bit(NFS_INO_INVALIDATING, &nfsi->flags); From ff81dfb5d721fff87bd516c558847f6effb70031 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Tue, 28 Sep 2021 14:33:44 -0400 Subject: [PATCH 15/84] NFS: Further optimisations for 'ls -l' If a user is doing 'ls -l', we have a heuristic in GETATTR that tells the readdir code to try to use READDIRPLUS in order to refresh the inode attributes. In certain cirumstances, we also try to invalidate the remaining directory entries in order to ensure this refresh. If there are multiple readers of the directory, we probably should avoid invalidating the page cache, since the heuristic breaks down in that situation anyway. Signed-off-by: Trond Myklebust Tested-by: Benjamin Coddington Reviewed-by: Benjamin Coddington --- fs/nfs/dir.c | 16 +++++++++++----- include/linux/nfs_fs.h | 5 ++--- 2 files changed, 13 insertions(+), 8 deletions(-) diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index fa4d33687d2b..33cfff8ea551 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -78,6 +78,7 @@ static struct nfs_open_dir_context *alloc_nfs_open_dir_context(struct inode *dir ctx->attr_gencount = nfsi->attr_gencount; ctx->dir_cookie = 0; ctx->dup_cookie = 0; + ctx->page_index = 0; spin_lock(&dir->i_lock); if (list_empty(&nfsi->open_files) && (nfsi->cache_validity & NFS_INO_DATA_INVAL_DEFER)) @@ -85,6 +86,7 @@ static struct nfs_open_dir_context *alloc_nfs_open_dir_context(struct inode *dir NFS_INO_INVALID_DATA | NFS_INO_REVAL_FORCED); list_add(&ctx->list, &nfsi->open_files); + clear_bit(NFS_INO_FORCE_READDIR, &nfsi->flags); spin_unlock(&dir->i_lock); return ctx; } @@ -627,8 +629,7 @@ void nfs_force_use_readdirplus(struct inode *dir) if (nfs_server_capable(dir, NFS_CAP_READDIRPLUS) && !list_empty(&nfsi->open_files)) { set_bit(NFS_INO_ADVISE_RDPLUS, &nfsi->flags); - invalidate_mapping_pages(dir->i_mapping, - nfsi->page_index + 1, -1); + set_bit(NFS_INO_FORCE_READDIR, &nfsi->flags); } } @@ -938,10 +939,8 @@ static int find_and_lock_cache_page(struct nfs_readdir_descriptor *desc) sizeof(nfsi->cookieverf)); } res = nfs_readdir_search_array(desc); - if (res == 0) { - nfsi->page_index = desc->page_index; + if (res == 0) return 0; - } nfs_readdir_page_unlock_and_put_cached(desc); return res; } @@ -1080,6 +1079,7 @@ static int nfs_readdir(struct file *file, struct dir_context *ctx) struct nfs_inode *nfsi = NFS_I(inode); struct nfs_open_dir_context *dir_ctx = file->private_data; struct nfs_readdir_descriptor *desc; + pgoff_t page_index; int res; dfprintk(FILE, "NFS: readdir(%pD2) starting at cookie %llu\n", @@ -1110,10 +1110,15 @@ static int nfs_readdir(struct file *file, struct dir_context *ctx) desc->dir_cookie = dir_ctx->dir_cookie; desc->dup_cookie = dir_ctx->dup_cookie; desc->duped = dir_ctx->duped; + page_index = dir_ctx->page_index; desc->attr_gencount = dir_ctx->attr_gencount; memcpy(desc->verf, dir_ctx->verf, sizeof(desc->verf)); spin_unlock(&file->f_lock); + if (test_and_clear_bit(NFS_INO_FORCE_READDIR, &nfsi->flags) && + list_is_singular(&nfsi->open_files)) + invalidate_mapping_pages(inode->i_mapping, page_index + 1, -1); + do { res = readdir_search_pagecache(desc); @@ -1150,6 +1155,7 @@ static int nfs_readdir(struct file *file, struct dir_context *ctx) dir_ctx->dup_cookie = desc->dup_cookie; dir_ctx->duped = desc->duped; dir_ctx->attr_gencount = desc->attr_gencount; + dir_ctx->page_index = desc->page_index; memcpy(dir_ctx->verf, desc->verf, sizeof(dir_ctx->verf)); spin_unlock(&file->f_lock); diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index 9b75448ce0df..ca547cc5458c 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -103,6 +103,7 @@ struct nfs_open_dir_context { __be32 verf[NFS_DIR_VERIFIER_SIZE]; __u64 dir_cookie; __u64 dup_cookie; + pgoff_t page_index; signed char duped; }; @@ -181,9 +182,6 @@ struct nfs_inode { struct rw_semaphore rmdir_sem; struct mutex commit_mutex; - /* track last access to cached pages */ - unsigned long page_index; - #if IS_ENABLED(CONFIG_NFS_V4) struct nfs4_cached_acl *nfs4_acl; /* NFSv4 state */ @@ -272,6 +270,7 @@ struct nfs4_copy_state { #define NFS_INO_INVALIDATING (3) /* inode is being invalidated */ #define NFS_INO_FSCACHE (5) /* inode can be cached by FS-Cache */ #define NFS_INO_FSCACHE_LOCK (6) /* FS-Cache cookie management lock */ +#define NFS_INO_FORCE_READDIR (7) /* force readdirplus */ #define NFS_INO_LAYOUTCOMMIT (9) /* layoutcommit required */ #define NFS_INO_LAYOUTCOMMITTING (10) /* layoutcommit inflight */ #define NFS_INO_LAYOUTSTATS (11) /* layoutstats inflight */ From cec08f452a687fce9dfdf47946d00a1d12a8bec5 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 29 Sep 2021 08:12:53 -0400 Subject: [PATCH 16/84] NFS: Fix dentry verifier races If the directory changed while we were revalidating the dentry, then don't update the dentry verifier. There is no value in setting the verifier to an older value, and we could end up overwriting a more up to date verifier from a parallel revalidation. Fixes: efeda80da38d ("NFSv4: Fix revalidation of dentries with delegations") Signed-off-by: Trond Myklebust Tested-by: Benjamin Coddington Reviewed-by: Benjamin Coddington --- fs/nfs/dir.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index 33cfff8ea551..3fafecdb2070 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -1276,13 +1276,12 @@ static bool nfs_verifier_is_delegated(struct dentry *dentry) static void nfs_set_verifier_locked(struct dentry *dentry, unsigned long verf) { struct inode *inode = d_inode(dentry); + struct inode *dir = d_inode(dentry->d_parent); - if (!nfs_verifier_is_delegated(dentry) && - !nfs_verify_change_attribute(d_inode(dentry->d_parent), verf)) - goto out; + if (!nfs_verify_change_attribute(dir, verf)) + return; if (inode && NFS_PROTO(inode)->have_delegation(inode, FMODE_READ)) nfs_set_verifier_delegated(&verf); -out: dentry->d_time = verf; } From b97583b26326ad559d1b1ba7dafec98712ffd834 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Sat, 2 Oct 2021 19:04:59 -0400 Subject: [PATCH 17/84] NFS: Do not flush the readdir cache in nfs_dentry_iput() The original premise in commit 83672d392f7b ("NFS: Fix directory caching problem - with test case and patch.") was that readdirplus was caching attribute information and replaying it later. This is no longer the case. Signed-off-by: Trond Myklebust --- fs/nfs/dir.c | 4 ---- 1 file changed, 4 deletions(-) diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index 3fafecdb2070..210c5945ac2b 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -1727,10 +1727,6 @@ static void nfs_drop_nlink(struct inode *inode) */ static void nfs_dentry_iput(struct dentry *dentry, struct inode *inode) { - if (S_ISDIR(inode->i_mode)) - /* drop any readdir cache as it could easily be old */ - nfs_set_cache_invalid(inode, NFS_INO_INVALID_DATA); - if (dentry->d_flags & DCACHE_NFSFS_RENAMED) { nfs_complete_unlink(dentry, inode); nfs_drop_nlink(inode); From 36a10a3c4cb6ea7c8bd895c16b3f59e1f0db2f6a Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Sat, 2 Oct 2021 19:21:49 -0400 Subject: [PATCH 18/84] NFS: Remove unnecessary page cache invalidations Remove cache invalidations that are already covered by change attribute updates. Signed-off-by: Trond Myklebust --- fs/nfs/inode.c | 7 +------ 1 file changed, 1 insertion(+), 6 deletions(-) diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index dcb885b7ad73..3bd0ae438663 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -1451,8 +1451,6 @@ static void nfs_wcc_update_inode(struct inode *inode, struct nfs_fattr *fattr) && (fattr->valid & NFS_ATTR_FATTR_MTIME) && timespec64_equal(&ts, &fattr->pre_mtime)) { inode->i_mtime = fattr->mtime; - if (S_ISDIR(inode->i_mode)) - nfs_set_cache_invalid(inode, NFS_INO_INVALID_DATA); } if ((fattr->valid & NFS_ATTR_FATTR_PRESIZE) && (fattr->valid & NFS_ATTR_FATTR_SIZE) @@ -2162,11 +2160,8 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) save_cache_validity & NFS_INO_INVALID_OTHER; if (fattr->valid & NFS_ATTR_FATTR_NLINK) { - if (inode->i_nlink != fattr->nlink) { - if (S_ISDIR(inode->i_mode)) - invalid |= NFS_INO_INVALID_DATA; + if (inode->i_nlink != fattr->nlink) set_nlink(inode, fattr->nlink); - } } else if (fattr_supported & NFS_ATTR_FATTR_NLINK) nfsi->cache_validity |= save_cache_validity & NFS_INO_INVALID_NLINK; From a2915fa06227b056a8f9b0d79b61dca08ad5cfc6 Mon Sep 17 00:00:00 2001 From: Baptiste Lepers Date: Mon, 6 Sep 2021 11:59:24 +1000 Subject: [PATCH 19/84] pnfs/flexfiles: Fix misplaced barrier in nfs4_ff_layout_prepare_ds _nfs4_pnfs_v3/v4_ds_connect do some work smp_wmb ds->ds_clp = clp; And nfs4_ff_layout_prepare_ds currently does smp_rmb if(ds->ds_clp) ... This patch places the smp_rmb after the if. This ensures that following reads only happen once nfs4_ff_layout_prepare_ds has checked that data has been properly initialized. Fixes: d67ae825a59d6 ("pnfs/flexfiles: Add the FlexFile Layout Driver") Signed-off-by: Baptiste Lepers Signed-off-by: Trond Myklebust --- fs/nfs/flexfilelayout/flexfilelayoutdev.c | 4 ++-- fs/nfs/pnfs_nfs.c | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/fs/nfs/flexfilelayout/flexfilelayoutdev.c b/fs/nfs/flexfilelayout/flexfilelayoutdev.c index c9b61b818ec1..bfa7202ca7be 100644 --- a/fs/nfs/flexfilelayout/flexfilelayoutdev.c +++ b/fs/nfs/flexfilelayout/flexfilelayoutdev.c @@ -378,10 +378,10 @@ nfs4_ff_layout_prepare_ds(struct pnfs_layout_segment *lseg, goto noconnect; ds = mirror->mirror_ds->ds; + if (READ_ONCE(ds->ds_clp)) + goto out; /* matching smp_wmb() in _nfs4_pnfs_v3/4_ds_connect */ smp_rmb(); - if (ds->ds_clp) - goto out; /* FIXME: For now we assume the server sent only one version of NFS * to use for the DS. diff --git a/fs/nfs/pnfs_nfs.c b/fs/nfs/pnfs_nfs.c index cf19914fec81..02bd6e83961d 100644 --- a/fs/nfs/pnfs_nfs.c +++ b/fs/nfs/pnfs_nfs.c @@ -895,7 +895,7 @@ static int _nfs4_pnfs_v3_ds_connect(struct nfs_server *mds_srv, } smp_wmb(); - ds->ds_clp = clp; + WRITE_ONCE(ds->ds_clp, clp); dprintk("%s [new] addr: %s\n", __func__, ds->ds_remotestr); out: return status; @@ -973,7 +973,7 @@ static int _nfs4_pnfs_v4_ds_connect(struct nfs_server *mds_srv, } smp_wmb(); - ds->ds_clp = clp; + WRITE_ONCE(ds->ds_clp, clp); dprintk("%s [new] addr: %s\n", __func__, ds->ds_remotestr); out: return status; From 8e09650f5ec68858f4b8b67cdef9e2ece9b208f3 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Mon, 4 Oct 2021 10:09:57 -0400 Subject: [PATCH 20/84] NFS: Remove unnecessary TRACE_DEFINE_ENUM()s Clean up: TRACE_DEFINE_ENUM is unnecessary because the target symbols are all C macros, not enums. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- fs/nfs/nfstrace.h | 68 ----------------------------------------------- 1 file changed, 68 deletions(-) diff --git a/fs/nfs/nfstrace.h b/fs/nfs/nfstrace.h index 8a224871be74..589f32fdbe63 100644 --- a/fs/nfs/nfstrace.h +++ b/fs/nfs/nfstrace.h @@ -11,16 +11,6 @@ #include #include -TRACE_DEFINE_ENUM(DT_UNKNOWN); -TRACE_DEFINE_ENUM(DT_FIFO); -TRACE_DEFINE_ENUM(DT_CHR); -TRACE_DEFINE_ENUM(DT_DIR); -TRACE_DEFINE_ENUM(DT_BLK); -TRACE_DEFINE_ENUM(DT_REG); -TRACE_DEFINE_ENUM(DT_LNK); -TRACE_DEFINE_ENUM(DT_SOCK); -TRACE_DEFINE_ENUM(DT_WHT); - #define nfs_show_file_type(ftype) \ __print_symbolic(ftype, \ { DT_UNKNOWN, "UNKNOWN" }, \ @@ -33,24 +23,6 @@ TRACE_DEFINE_ENUM(DT_WHT); { DT_SOCK, "SOCK" }, \ { DT_WHT, "WHT" }) -TRACE_DEFINE_ENUM(NFS_INO_INVALID_DATA); -TRACE_DEFINE_ENUM(NFS_INO_INVALID_ATIME); -TRACE_DEFINE_ENUM(NFS_INO_INVALID_ACCESS); -TRACE_DEFINE_ENUM(NFS_INO_INVALID_ACL); -TRACE_DEFINE_ENUM(NFS_INO_REVAL_PAGECACHE); -TRACE_DEFINE_ENUM(NFS_INO_REVAL_FORCED); -TRACE_DEFINE_ENUM(NFS_INO_INVALID_LABEL); -TRACE_DEFINE_ENUM(NFS_INO_INVALID_CHANGE); -TRACE_DEFINE_ENUM(NFS_INO_INVALID_CTIME); -TRACE_DEFINE_ENUM(NFS_INO_INVALID_MTIME); -TRACE_DEFINE_ENUM(NFS_INO_INVALID_SIZE); -TRACE_DEFINE_ENUM(NFS_INO_INVALID_OTHER); -TRACE_DEFINE_ENUM(NFS_INO_DATA_INVAL_DEFER); -TRACE_DEFINE_ENUM(NFS_INO_INVALID_BLOCKS); -TRACE_DEFINE_ENUM(NFS_INO_INVALID_XATTR); -TRACE_DEFINE_ENUM(NFS_INO_INVALID_NLINK); -TRACE_DEFINE_ENUM(NFS_INO_INVALID_MODE); - #define nfs_show_cache_validity(v) \ __print_flags(v, "|", \ { NFS_INO_INVALID_DATA, "INVALID_DATA" }, \ @@ -71,17 +43,6 @@ TRACE_DEFINE_ENUM(NFS_INO_INVALID_MODE); { NFS_INO_INVALID_NLINK, "INVALID_NLINK" }, \ { NFS_INO_INVALID_MODE, "INVALID_MODE" }) -TRACE_DEFINE_ENUM(NFS_INO_ADVISE_RDPLUS); -TRACE_DEFINE_ENUM(NFS_INO_STALE); -TRACE_DEFINE_ENUM(NFS_INO_ACL_LRU_SET); -TRACE_DEFINE_ENUM(NFS_INO_INVALIDATING); -TRACE_DEFINE_ENUM(NFS_INO_FSCACHE); -TRACE_DEFINE_ENUM(NFS_INO_FSCACHE_LOCK); -TRACE_DEFINE_ENUM(NFS_INO_LAYOUTCOMMIT); -TRACE_DEFINE_ENUM(NFS_INO_LAYOUTCOMMITTING); -TRACE_DEFINE_ENUM(NFS_INO_LAYOUTSTATS); -TRACE_DEFINE_ENUM(NFS_INO_ODIRECT); - #define nfs_show_nfsi_flags(v) \ __print_flags(v, "|", \ { BIT(NFS_INO_ADVISE_RDPLUS), "ADVISE_RDPLUS" }, \ @@ -270,19 +231,6 @@ TRACE_EVENT(nfs_access_exit, ) ); -TRACE_DEFINE_ENUM(LOOKUP_FOLLOW); -TRACE_DEFINE_ENUM(LOOKUP_DIRECTORY); -TRACE_DEFINE_ENUM(LOOKUP_AUTOMOUNT); -TRACE_DEFINE_ENUM(LOOKUP_PARENT); -TRACE_DEFINE_ENUM(LOOKUP_REVAL); -TRACE_DEFINE_ENUM(LOOKUP_RCU); -TRACE_DEFINE_ENUM(LOOKUP_OPEN); -TRACE_DEFINE_ENUM(LOOKUP_CREATE); -TRACE_DEFINE_ENUM(LOOKUP_EXCL); -TRACE_DEFINE_ENUM(LOOKUP_RENAME_TARGET); -TRACE_DEFINE_ENUM(LOOKUP_EMPTY); -TRACE_DEFINE_ENUM(LOOKUP_DOWN); - #define show_lookup_flags(flags) \ __print_flags(flags, "|", \ { LOOKUP_FOLLOW, "FOLLOW" }, \ @@ -392,22 +340,6 @@ DEFINE_NFS_LOOKUP_EVENT_DONE(nfs_lookup_exit); DEFINE_NFS_LOOKUP_EVENT(nfs_lookup_revalidate_enter); DEFINE_NFS_LOOKUP_EVENT_DONE(nfs_lookup_revalidate_exit); -TRACE_DEFINE_ENUM(O_WRONLY); -TRACE_DEFINE_ENUM(O_RDWR); -TRACE_DEFINE_ENUM(O_CREAT); -TRACE_DEFINE_ENUM(O_EXCL); -TRACE_DEFINE_ENUM(O_NOCTTY); -TRACE_DEFINE_ENUM(O_TRUNC); -TRACE_DEFINE_ENUM(O_APPEND); -TRACE_DEFINE_ENUM(O_NONBLOCK); -TRACE_DEFINE_ENUM(O_DSYNC); -TRACE_DEFINE_ENUM(O_DIRECT); -TRACE_DEFINE_ENUM(O_LARGEFILE); -TRACE_DEFINE_ENUM(O_DIRECTORY); -TRACE_DEFINE_ENUM(O_NOFOLLOW); -TRACE_DEFINE_ENUM(O_NOATIME); -TRACE_DEFINE_ENUM(O_CLOEXEC); - #define show_open_flags(flags) \ __print_flags(flags, "|", \ { O_WRONLY, "O_WRONLY" }, \ From 0392dd51f9c78d46109a408f27dc820300dcd8bd Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Mon, 4 Oct 2021 10:10:10 -0400 Subject: [PATCH 21/84] SUNRPC: Per-rpc_clnt task PIDs The current range of RPC task PIDs is 0..65535. That's not adequate for distinguishing tasks across multiple rpc_clnts running high throughput workloads. To help relieve this situation and to reduce the bottleneck of having a single atomic for assigning all RPC task PIDs, assign task PIDs per rpc_clnt. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/sunrpc/clnt.h | 1 + net/sunrpc/sched.c | 12 ++++++++++-- 2 files changed, 11 insertions(+), 2 deletions(-) diff --git a/include/linux/sunrpc/clnt.h b/include/linux/sunrpc/clnt.h index a4661646adc9..267b7aeaf1a6 100644 --- a/include/linux/sunrpc/clnt.h +++ b/include/linux/sunrpc/clnt.h @@ -40,6 +40,7 @@ struct rpc_clnt { unsigned int cl_clid; /* client id */ struct list_head cl_clients; /* Global list of clients */ struct list_head cl_tasks; /* List of tasks */ + atomic_t cl_pid; /* task PID counter */ spinlock_t cl_lock; /* spinlock */ struct rpc_xprt __rcu * cl_xprt; /* transport */ const struct rpc_procinfo *cl_procinfo; /* procedure info */ diff --git a/net/sunrpc/sched.c b/net/sunrpc/sched.c index b21457cec8a5..f4f311ea7a66 100644 --- a/net/sunrpc/sched.c +++ b/net/sunrpc/sched.c @@ -277,9 +277,17 @@ static int rpc_wait_bit_killable(struct wait_bit_key *key, int mode) #if IS_ENABLED(CONFIG_SUNRPC_DEBUG) || IS_ENABLED(CONFIG_TRACEPOINTS) static void rpc_task_set_debuginfo(struct rpc_task *task) { - static atomic_t rpc_pid; + struct rpc_clnt *clnt = task->tk_client; - task->tk_pid = atomic_inc_return(&rpc_pid); + /* Might be a task carrying a reverse-direction operation */ + if (!clnt) { + static atomic_t rpc_pid; + + task->tk_pid = atomic_inc_return(&rpc_pid); + return; + } + + task->tk_pid = atomic_inc_return(&clnt->cl_pid); } #else static inline void rpc_task_set_debuginfo(struct rpc_task *task) From 110cb2d2f9326030f13b7ec85d6d482934ea5462 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Mon, 4 Oct 2021 10:10:16 -0400 Subject: [PATCH 22/84] NFS: Instrument i_size_write() Generate a trace event whenever the NFS client modifies the size of a file. These new events aid troubleshooting workloads that trigger races around size updates. There are four new trace points, all named nfs_size_something so they are easy to grep for or enable as a group with a single glob. Size updated on the server: kworker/u24:10-194 [010] 369.939174: nfs_size_update: fileid=00:28:2 fhandle=0x36fbbe51 version=1752899344277980615 cursize=250471 newsize=172083 Server-side size update reported via NFSv3 WCC attributes: fsx-1387 [006] 380.760686: nfs_size_wcc: fileid=00:28:2 fhandle=0x36fbbe51 version=1752899355909932456 cursize=146792 newsize=171216 File has been truncated locally: fsx-1387 [007] 369.437421: nfs_size_truncate: fileid=00:28:2 fhandle=0x36fbbe51 version=1752899231200117272 cursize=215244 newsize=0 File has been extended locally: fsx-1387 [007] 369.439213: nfs_size_grow: fileid=00:28:2 fhandle=0x36fbbe51 version=1752899343704248410 cursize=258048 newsize=262144 Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- fs/nfs/inode.c | 9 +++------ fs/nfs/nfstrace.h | 50 +++++++++++++++++++++++++++++++++++++++++++++++ fs/nfs/write.c | 1 + 3 files changed, 54 insertions(+), 6 deletions(-) diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index 3bd0ae438663..a10572f278e6 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -671,6 +671,7 @@ static int nfs_vmtruncate(struct inode * inode, loff_t offset) if (err) goto out; + trace_nfs_size_truncate(inode, offset); i_size_write(inode, offset); /* Optimisation */ if (offset == 0) @@ -1456,6 +1457,7 @@ static void nfs_wcc_update_inode(struct inode *inode, struct nfs_fattr *fattr) && (fattr->valid & NFS_ATTR_FATTR_SIZE) && i_size_read(inode) == nfs_size_to_loff_t(fattr->pre_size) && !nfs_have_writebacks(inode)) { + trace_nfs_size_wcc(inode, fattr->size); i_size_write(inode, nfs_size_to_loff_t(fattr->size)); } } @@ -2100,16 +2102,11 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) /* Do we perhaps have any outstanding writes, or has * the file grown beyond our last write? */ if (!nfs_have_writebacks(inode) || new_isize > cur_isize) { + trace_nfs_size_update(inode, new_isize); i_size_write(inode, new_isize); if (!have_writers) invalid |= NFS_INO_INVALID_DATA; } - dprintk("NFS: isize change on server for file %s/%ld " - "(%Ld to %Ld)\n", - inode->i_sb->s_id, - inode->i_ino, - (long long)cur_isize, - (long long)new_isize); } if (new_isize == 0 && !(fattr->valid & (NFS_ATTR_FATTR_SPACE_USED | diff --git a/fs/nfs/nfstrace.h b/fs/nfs/nfstrace.h index 589f32fdbe63..44fd016a8e65 100644 --- a/fs/nfs/nfstrace.h +++ b/fs/nfs/nfstrace.h @@ -231,6 +231,56 @@ TRACE_EVENT(nfs_access_exit, ) ); +DECLARE_EVENT_CLASS(nfs_update_size_class, + TP_PROTO( + const struct inode *inode, + loff_t new_size + ), + + TP_ARGS(inode, new_size), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __field(u64, version) + __field(loff_t, cur_size) + __field(loff_t, new_size) + ), + + TP_fast_assign( + const struct nfs_inode *nfsi = NFS_I(inode); + + __entry->dev = inode->i_sb->s_dev; + __entry->fhandle = nfs_fhandle_hash(&nfsi->fh); + __entry->fileid = nfsi->fileid; + __entry->version = inode_peek_iversion_raw(inode); + __entry->cur_size = i_size_read(inode); + __entry->new_size = new_size; + ), + + TP_printk( + "fileid=%02x:%02x:%llu fhandle=0x%08x version=%llu cursize=%lld newsize=%lld", + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, __entry->version, + __entry->cur_size, __entry->new_size + ) +); + +#define DEFINE_NFS_UPDATE_SIZE_EVENT(name) \ + DEFINE_EVENT(nfs_update_size_class, nfs_size_##name, \ + TP_PROTO( \ + const struct inode *inode, \ + loff_t new_size \ + ), \ + TP_ARGS(inode, new_size)) + +DEFINE_NFS_UPDATE_SIZE_EVENT(truncate); +DEFINE_NFS_UPDATE_SIZE_EVENT(wcc); +DEFINE_NFS_UPDATE_SIZE_EVENT(update); +DEFINE_NFS_UPDATE_SIZE_EVENT(grow); + #define show_lookup_flags(flags) \ __print_flags(flags, "|", \ { LOOKUP_FOLLOW, "FOLLOW" }, \ diff --git a/fs/nfs/write.c b/fs/nfs/write.c index 773ea2c8504d..b89d5ef3af0e 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -288,6 +288,7 @@ static void nfs_grow_file(struct page *page, unsigned int offset, unsigned int c end = page_file_offset(page) + ((loff_t)offset+count); if (i_size >= end) goto out; + trace_nfs_size_grow(inode, end); i_size_write(inode, end); NFS_I(inode)->cache_validity &= ~NFS_INO_INVALID_SIZE; nfs_inc_stats(inode, NFSIOS_EXTENDWRITE); From 64a93dbf25d3a1368bb58ddf0f61d0a92d7479e3 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Mon, 4 Oct 2021 15:44:16 -0400 Subject: [PATCH 23/84] NFS: Fix deadlocks in nfs_scan_commit_list() Partially revert commit 2ce209c42c01 ("NFS: Wait for requests that are locked on the commit list"), since it can lead to deadlocks between commit requests and nfs_join_page_group(). For now we should assume that any locked requests on the commit list are either about to be removed and committed by another task, or the writes they describe are about to be retransmitted. In either case, we should not need to worry. Fixes: 2ce209c42c01 ("NFS: Wait for requests that are locked on the commit list") Signed-off-by: Trond Myklebust --- fs/nfs/write.c | 17 ++--------------- 1 file changed, 2 insertions(+), 15 deletions(-) diff --git a/fs/nfs/write.c b/fs/nfs/write.c index b89d5ef3af0e..38f181e1343a 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -1039,25 +1039,11 @@ nfs_scan_commit_list(struct list_head *src, struct list_head *dst, struct nfs_page *req, *tmp; int ret = 0; -restart: list_for_each_entry_safe(req, tmp, src, wb_list) { kref_get(&req->wb_kref); if (!nfs_lock_request(req)) { - int status; - - /* Prevent deadlock with nfs_lock_and_join_requests */ - if (!list_empty(dst)) { - nfs_release_request(req); - continue; - } - /* Ensure we make progress to prevent livelock */ - mutex_unlock(&NFS_I(cinfo->inode)->commit_mutex); - status = nfs_wait_on_request(req); nfs_release_request(req); - mutex_lock(&NFS_I(cinfo->inode)->commit_mutex); - if (status < 0) - break; - goto restart; + continue; } nfs_request_remove_commit_list(req, cinfo); clear_bit(PG_COMMIT_TO_DS, &req->wb_flags); @@ -1952,6 +1938,7 @@ static int __nfs_commit_inode(struct inode *inode, int how, int may_wait = how & FLUSH_SYNC; int ret, nscan; + how &= ~FLUSH_SYNC; nfs_init_cinfo_from_inode(&cinfo, inode); nfs_commit_begin(cinfo.mds); for (;;) { From 133a48abf6ecc535d7eddc6da1c3e4c972445882 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Mon, 4 Oct 2021 15:37:42 -0400 Subject: [PATCH 24/84] NFS: Fix up commit deadlocks If O_DIRECT bumps the commit_info rpcs_out field, then that could lead to fsync() hangs. The fix is to ensure that O_DIRECT calls nfs_commit_end(). Fixes: 723c921e7dfc ("sched/wait, fs/nfs: Convert wait_on_atomic_t() usage to the new wait_var_event() API") Signed-off-by: Trond Myklebust --- fs/nfs/direct.c | 2 +- fs/nfs/pnfs_nfs.c | 2 -- fs/nfs/write.c | 9 ++++++--- include/linux/nfs_fs.h | 1 + 4 files changed, 8 insertions(+), 6 deletions(-) diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c index 2e894fec036b..3c0335c15a73 100644 --- a/fs/nfs/direct.c +++ b/fs/nfs/direct.c @@ -620,7 +620,7 @@ static void nfs_direct_commit_complete(struct nfs_commit_data *data) nfs_unlock_and_release_request(req); } - if (atomic_dec_and_test(&cinfo.mds->rpcs_out)) + if (nfs_commit_end(cinfo.mds)) nfs_direct_write_complete(dreq); } diff --git a/fs/nfs/pnfs_nfs.c b/fs/nfs/pnfs_nfs.c index 02bd6e83961d..316f68f96e57 100644 --- a/fs/nfs/pnfs_nfs.c +++ b/fs/nfs/pnfs_nfs.c @@ -468,7 +468,6 @@ pnfs_bucket_alloc_ds_commits(struct list_head *list, goto out_error; data->ds_commit_index = i; list_add_tail(&data->list, list); - atomic_inc(&cinfo->mds->rpcs_out); nreq++; } mutex_unlock(&NFS_I(cinfo->inode)->commit_mutex); @@ -520,7 +519,6 @@ pnfs_generic_commit_pagelist(struct inode *inode, struct list_head *mds_pages, data->ds_commit_index = -1; list_splice_init(mds_pages, &data->pages); list_add_tail(&data->list, &list); - atomic_inc(&cinfo->mds->rpcs_out); nreq++; } diff --git a/fs/nfs/write.c b/fs/nfs/write.c index 38f181e1343a..465220f47142 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -1673,10 +1673,13 @@ static void nfs_commit_begin(struct nfs_mds_commit_info *cinfo) atomic_inc(&cinfo->rpcs_out); } -static void nfs_commit_end(struct nfs_mds_commit_info *cinfo) +bool nfs_commit_end(struct nfs_mds_commit_info *cinfo) { - if (atomic_dec_and_test(&cinfo->rpcs_out)) + if (atomic_dec_and_test(&cinfo->rpcs_out)) { wake_up_var(&cinfo->rpcs_out); + return true; + } + return false; } void nfs_commitdata_release(struct nfs_commit_data *data) @@ -1776,6 +1779,7 @@ void nfs_init_commit(struct nfs_commit_data *data, data->res.fattr = &data->fattr; data->res.verf = &data->verf; nfs_fattr_init(&data->fattr); + nfs_commit_begin(cinfo->mds); } EXPORT_SYMBOL_GPL(nfs_init_commit); @@ -1822,7 +1826,6 @@ nfs_commit_list(struct inode *inode, struct list_head *head, int how, /* Set up the argument struct */ nfs_init_commit(data, head, NULL, cinfo); - atomic_inc(&cinfo->mds->rpcs_out); if (NFS_SERVER(inode)->nfs_client->cl_minorversion) task_flags = RPC_TASK_MOVEABLE; return nfs_initiate_commit(NFS_CLIENT(inode), data, NFS_PROTO(inode), diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index ca547cc5458c..e58b78da7a98 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -568,6 +568,7 @@ extern int nfs_wb_page_cancel(struct inode *inode, struct page* page); extern int nfs_commit_inode(struct inode *, int); extern struct nfs_commit_data *nfs_commitdata_alloc(bool never_fail); extern void nfs_commit_free(struct nfs_commit_data *data); +bool nfs_commit_end(struct nfs_mds_commit_info *cinfo); static inline int nfs_have_writebacks(struct inode *inode) From f0caea8882a7412a2ad4d8274f0280cdf849c9e2 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Tue, 5 Oct 2021 14:05:02 -0400 Subject: [PATCH 25/84] NFS: Fix an Oops in pnfs_mark_request_commit() Olga reports seeing the following Oops when doing O_DIRECT writes to a pNFS flexfiles server: Oops: 0000 [#1] SMP PTI CPU: 1 PID: 234186 Comm: kworker/u8:1 Not tainted 5.15.0-rc4+ #4 Hardware name: Red Hat KVM/RHEL-AV, BIOS 1.13.0-2.module+el8.3.0+7353+9de0a3cc 04/01/2014 Workqueue: nfsiod rpc_async_release [sunrpc] RIP: 0010:nfs_mark_request_commit+0x12/0x30 [nfs] Code: ff ff be 03 00 00 00 e8 ac 34 83 eb e9 29 ff ff ff e8 22 bc d7 eb 66 90 0f 1f 44 00 00 48 85 f6 74 16 48 8b 42 10 48 8b 40 18 <48> 8b 40 18 48 85 c0 74 05 e9 70 fc 15 ec 48 89 d6 e9 68 ed ff ff RSP: 0018:ffffa82f0159fe00 EFLAGS: 00010286 RAX: 0000000000000000 RBX: ffff8f3393141880 RCX: 0000000000000000 RDX: ffffa82f0159fe08 RSI: ffff8f3381252500 RDI: ffff8f3393141880 RBP: ffff8f33ac317c00 R08: 0000000000000000 R09: ffff8f3487724cb0 R10: 0000000000000008 R11: 0000000000000001 R12: 0000000000000001 R13: ffff8f3485bccee0 R14: ffff8f33ac317c10 R15: ffff8f33ac317cd8 FS: 0000000000000000(0000) GS:ffff8f34fbc80000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000000000018 CR3: 0000000122120006 CR4: 0000000000770ee0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 PKRU: 55555554 Call Trace: nfs_direct_write_completion+0x13b/0x250 [nfs] rpc_free_task+0x39/0x60 [sunrpc] rpc_async_release+0x29/0x40 [sunrpc] process_one_work+0x1ce/0x370 worker_thread+0x30/0x380 ? process_one_work+0x370/0x370 kthread+0x11a/0x140 ? set_kthread_struct+0x40/0x40 ret_from_fork+0x22/0x30 Reported-by: Olga Kornievskaia Fixes: 9c455a8c1e14 ("NFS/pNFS: Clean up pNFS commit operations") Signed-off-by: Trond Myklebust --- fs/nfs/pnfs.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/nfs/pnfs.h b/fs/nfs/pnfs.h index d810ae674f4e..a0f6ff094b3a 100644 --- a/fs/nfs/pnfs.h +++ b/fs/nfs/pnfs.h @@ -517,7 +517,7 @@ pnfs_mark_request_commit(struct nfs_page *req, struct pnfs_layout_segment *lseg, { struct pnfs_ds_commit_info *fl_cinfo = cinfo->ds; - if (!lseg || !fl_cinfo->ops->mark_request_commit) + if (!lseg || !fl_cinfo->ops || !fl_cinfo->ops->mark_request_commit) return false; fl_cinfo->ops->mark_request_commit(req, lseg, cinfo, ds_commit_idx); return true; From 6e176d47160cec8bcaa28d9aa06926d72d54237c Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Sun, 10 Oct 2021 10:58:12 +0200 Subject: [PATCH 26/84] NFSv4: Fixes for nfs4_inode_return_delegation() We mustn't call nfs_wb_all() on anything other than a regular file. Furthermore, we can exit early when we don't hold a delegation. Reported-by: David Wysochanski Signed-off-by: Trond Myklebust --- fs/nfs/delegation.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) diff --git a/fs/nfs/delegation.c b/fs/nfs/delegation.c index 11118398f495..7c9eb679dbdb 100644 --- a/fs/nfs/delegation.c +++ b/fs/nfs/delegation.c @@ -755,11 +755,13 @@ int nfs4_inode_return_delegation(struct inode *inode) struct nfs_delegation *delegation; delegation = nfs_start_delegation_return(nfsi); - /* Synchronous recall of any application leases */ - break_lease(inode, O_WRONLY | O_RDWR); - nfs_wb_all(inode); - if (delegation != NULL) + if (delegation != NULL) { + /* Synchronous recall of any application leases */ + break_lease(inode, O_WRONLY | O_RDWR); + if (S_ISREG(inode->i_mode)) + nfs_wb_all(inode); return nfs_end_delegation_return(inode, delegation, 1); + } return 0; } From 0ebeebcf59601bcfa0284f4bb7abdec051eb856d Mon Sep 17 00:00:00 2001 From: Dave Wysochanski Date: Sun, 10 Oct 2021 18:23:13 -0400 Subject: [PATCH 27/84] NFS: Fix WARN_ON due to unionization of nfs_inode.nrequests Fixes the following WARN_ON WARNING: CPU: 2 PID: 18678 at fs/nfs/inode.c:123 nfs_clear_inode+0x3b/0x50 [nfs] ... Call Trace: nfs4_evict_inode+0x57/0x70 [nfsv4] evict+0xd1/0x180 dispose_list+0x48/0x60 evict_inodes+0x156/0x190 generic_shutdown_super+0x37/0x110 nfs_kill_super+0x1d/0x40 [nfs] deactivate_locked_super+0x36/0xa0 Signed-off-by: Dave Wysochanski Signed-off-by: Trond Myklebust --- include/linux/nfs_fs.h | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index e58b78da7a98..457b866a2d9e 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -573,7 +573,9 @@ bool nfs_commit_end(struct nfs_mds_commit_info *cinfo); static inline int nfs_have_writebacks(struct inode *inode) { - return atomic_long_read(&NFS_I(inode)->nrequests) != 0; + if (S_ISREG(inode->i_mode)) + return atomic_long_read(&NFS_I(inode)->nrequests) != 0; + return 0; } /* From e591b298d7ecb851e200f65946e3d53fe78a3c4f Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Tue, 28 Sep 2021 17:41:41 -0400 Subject: [PATCH 28/84] NFS: Save some space in the inode Save some space in the nfs_inode by setting up an anonymous union with the fields that are peculiar to a specific type of filesystem object. Signed-off-by: Trond Myklebust --- fs/nfs/inode.c | 26 ++++++++++++++++++-------- include/linux/nfs_fs.h | 42 ++++++++++++++++++++++++------------------ 2 files changed, 42 insertions(+), 26 deletions(-) diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index a10572f278e6..b81b2d2f47ad 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -431,6 +431,22 @@ nfs_ilookup(struct super_block *sb, struct nfs_fattr *fattr, struct nfs_fh *fh) return inode; } +static void nfs_inode_init_regular(struct nfs_inode *nfsi) +{ + atomic_long_set(&nfsi->nrequests, 0); + INIT_LIST_HEAD(&nfsi->commit_info.list); + atomic_long_set(&nfsi->commit_info.ncommit, 0); + atomic_set(&nfsi->commit_info.rpcs_out, 0); + mutex_init(&nfsi->commit_mutex); +} + +static void nfs_inode_init_dir(struct nfs_inode *nfsi) +{ + nfsi->cache_change_attribute = 0; + memset(nfsi->cookieverf, 0, sizeof(nfsi->cookieverf)); + init_rwsem(&nfsi->rmdir_sem); +} + /* * This is our front-end to iget that looks up inodes by file handle * instead of inode number. @@ -485,10 +501,12 @@ nfs_fhget(struct super_block *sb, struct nfs_fh *fh, struct nfs_fattr *fattr, st if (S_ISREG(inode->i_mode)) { inode->i_fop = NFS_SB(sb)->nfs_client->rpc_ops->file_ops; inode->i_data.a_ops = &nfs_file_aops; + nfs_inode_init_regular(nfsi); } else if (S_ISDIR(inode->i_mode)) { inode->i_op = NFS_SB(sb)->nfs_client->rpc_ops->dir_inode_ops; inode->i_fop = &nfs_dir_operations; inode->i_data.a_ops = &nfs_dir_aops; + nfs_inode_init_dir(nfsi); /* Deal with crossing mountpoints */ if (fattr->valid & NFS_ATTR_FATTR_MOUNTPOINT || fattr->valid & NFS_ATTR_FATTR_V4_REFERRAL) { @@ -514,7 +532,6 @@ nfs_fhget(struct super_block *sb, struct nfs_fh *fh, struct nfs_fattr *fattr, st inode->i_uid = make_kuid(&init_user_ns, -2); inode->i_gid = make_kgid(&init_user_ns, -2); inode->i_blocks = 0; - memset(nfsi->cookieverf, 0, sizeof(nfsi->cookieverf)); nfsi->write_io = 0; nfsi->read_io = 0; @@ -2259,14 +2276,7 @@ static void init_once(void *foo) INIT_LIST_HEAD(&nfsi->open_files); INIT_LIST_HEAD(&nfsi->access_cache_entry_lru); INIT_LIST_HEAD(&nfsi->access_cache_inode_lru); - INIT_LIST_HEAD(&nfsi->commit_info.list); - atomic_long_set(&nfsi->nrequests, 0); - atomic_long_set(&nfsi->commit_info.ncommit, 0); - atomic_set(&nfsi->commit_info.rpcs_out, 0); - init_rwsem(&nfsi->rmdir_sem); - mutex_init(&nfsi->commit_mutex); nfs4_init_once(nfsi); - nfsi->cache_change_attribute = 0; } static int __init nfs_init_inodecache(void) diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index 457b866a2d9e..739ca1ef934f 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -155,33 +155,39 @@ struct nfs_inode { unsigned long attrtimeo_timestamp; unsigned long attr_gencount; - /* "Generation counter" for the attribute cache. This is - * bumped whenever we update the metadata on the - * server. - */ - unsigned long cache_change_attribute; struct rb_root access_cache; struct list_head access_cache_entry_lru; struct list_head access_cache_inode_lru; - /* - * This is the cookie verifier used for NFSv3 readdir - * operations - */ - __be32 cookieverf[NFS_DIR_VERIFIER_SIZE]; - - atomic_long_t nrequests; - struct nfs_mds_commit_info commit_info; + union { + /* Directory */ + struct { + /* "Generation counter" for the attribute cache. + * This is bumped whenever we update the metadata + * on the server. + */ + unsigned long cache_change_attribute; + /* + * This is the cookie verifier used for NFSv3 readdir + * operations + */ + __be32 cookieverf[NFS_DIR_VERIFIER_SIZE]; + /* Readers: in-flight sillydelete RPC calls */ + /* Writers: rmdir */ + struct rw_semaphore rmdir_sem; + }; + /* Regular file */ + struct { + atomic_long_t nrequests; + struct nfs_mds_commit_info commit_info; + struct mutex commit_mutex; + }; + }; /* Open contexts for shared mmap writes */ struct list_head open_files; - /* Readers: in-flight sillydelete RPC calls */ - /* Writers: rmdir */ - struct rw_semaphore rmdir_sem; - struct mutex commit_mutex; - #if IS_ENABLED(CONFIG_NFS_V4) struct nfs4_cached_acl *nfs4_acl; /* NFSv4 state */ From d5f458a979650e5ed37212f6134e4ee2b28cb6ed Mon Sep 17 00:00:00 2001 From: Alexey Gladkov Date: Thu, 14 Oct 2021 18:02:30 +0200 Subject: [PATCH 29/84] Fix user namespace leak Fixes: 61ca2c4afd9d ("NFS: Only reference user namespace from nfs4idmap struct instead of cred") Signed-off-by: Alexey Gladkov Signed-off-by: Trond Myklebust --- fs/nfs/nfs4idmap.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/nfs/nfs4idmap.c b/fs/nfs/nfs4idmap.c index 8d8aba305ecc..f331866dd418 100644 --- a/fs/nfs/nfs4idmap.c +++ b/fs/nfs/nfs4idmap.c @@ -487,7 +487,7 @@ nfs_idmap_new(struct nfs_client *clp) err_destroy_pipe: rpc_destroy_pipe_data(idmap->idmap_pipe); err: - get_user_ns(idmap->user_ns); + put_user_ns(idmap->user_ns); kfree(idmap); return error; } From 21037b8c2258ec40de3b31be9ced43ceb3b784f7 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Tue, 5 Oct 2021 10:17:59 -0400 Subject: [PATCH 30/84] xprtrdma: Provide a buffer to pad Write chunks of unaligned length This is a buffer to be left persistently registered while a connection is up. Connection tear-down will automatically DMA-unmap, invalidate, and dereg the MR. A persistently registered buffer is lower in cost to provide, and it can never be coalesced into the RDMA segment that carries the data payload. An RPC that provisions a Write chunk with a non-aligned length now uses this MR rather than the tail buffer of the RPC's rq_rcv_buf. Reviewed-By: Tom Talpey Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/trace/events/rpcrdma.h | 13 +++++++++--- net/sunrpc/xprtrdma/frwr_ops.c | 35 +++++++++++++++++++++++++++++++++ net/sunrpc/xprtrdma/rpc_rdma.c | 23 +++++++++++++--------- net/sunrpc/xprtrdma/verbs.c | 1 + net/sunrpc/xprtrdma/xprt_rdma.h | 5 +++++ 5 files changed, 65 insertions(+), 12 deletions(-) diff --git a/include/trace/events/rpcrdma.h b/include/trace/events/rpcrdma.h index de4195499592..afb2e394797c 100644 --- a/include/trace/events/rpcrdma.h +++ b/include/trace/events/rpcrdma.h @@ -375,10 +375,16 @@ DECLARE_EVENT_CLASS(xprtrdma_mr_class, TP_fast_assign( const struct rpcrdma_req *req = mr->mr_req; - const struct rpc_task *task = req->rl_slot.rq_task; - __entry->task_id = task->tk_pid; - __entry->client_id = task->tk_client->cl_clid; + if (req) { + const struct rpc_task *task = req->rl_slot.rq_task; + + __entry->task_id = task->tk_pid; + __entry->client_id = task->tk_client->cl_clid; + } else { + __entry->task_id = 0; + __entry->client_id = -1; + } __entry->mr_id = mr->mr_ibmr->res.id; __entry->nents = mr->mr_nents; __entry->handle = mr->mr_handle; @@ -639,6 +645,7 @@ TRACE_EVENT(xprtrdma_nomrs_err, DEFINE_RDCH_EVENT(read); DEFINE_WRCH_EVENT(write); DEFINE_WRCH_EVENT(reply); +DEFINE_WRCH_EVENT(wp); TRACE_DEFINE_ENUM(rpcrdma_noch); TRACE_DEFINE_ENUM(rpcrdma_noch_pullup); diff --git a/net/sunrpc/xprtrdma/frwr_ops.c b/net/sunrpc/xprtrdma/frwr_ops.c index f700b34a5bfd..3eccf365fcb8 100644 --- a/net/sunrpc/xprtrdma/frwr_ops.c +++ b/net/sunrpc/xprtrdma/frwr_ops.c @@ -666,3 +666,38 @@ void frwr_unmap_async(struct rpcrdma_xprt *r_xprt, struct rpcrdma_req *req) */ rpcrdma_force_disconnect(ep); } + +/** + * frwr_wp_create - Create an MR for padding Write chunks + * @r_xprt: transport resources to use + * + * Return 0 on success, negative errno on failure. + */ +int frwr_wp_create(struct rpcrdma_xprt *r_xprt) +{ + struct rpcrdma_ep *ep = r_xprt->rx_ep; + struct rpcrdma_mr_seg seg; + struct rpcrdma_mr *mr; + + mr = rpcrdma_mr_get(r_xprt); + if (!mr) + return -EAGAIN; + mr->mr_req = NULL; + ep->re_write_pad_mr = mr; + + seg.mr_len = XDR_UNIT; + seg.mr_page = virt_to_page(ep->re_write_pad); + seg.mr_offset = offset_in_page(ep->re_write_pad); + if (IS_ERR(frwr_map(r_xprt, &seg, 1, true, xdr_zero, mr))) + return -EIO; + trace_xprtrdma_mr_fastreg(mr); + + mr->mr_cqe.done = frwr_wc_fastreg; + mr->mr_regwr.wr.next = NULL; + mr->mr_regwr.wr.wr_cqe = &mr->mr_cqe; + mr->mr_regwr.wr.num_sge = 0; + mr->mr_regwr.wr.opcode = IB_WR_REG_MR; + mr->mr_regwr.wr.send_flags = 0; + + return ib_post_send(ep->re_id->qp, &mr->mr_regwr.wr, NULL); +} diff --git a/net/sunrpc/xprtrdma/rpc_rdma.c b/net/sunrpc/xprtrdma/rpc_rdma.c index c335c1361564..8035a983c8ce 100644 --- a/net/sunrpc/xprtrdma/rpc_rdma.c +++ b/net/sunrpc/xprtrdma/rpc_rdma.c @@ -255,15 +255,7 @@ rpcrdma_convert_iovs(struct rpcrdma_xprt *r_xprt, struct xdr_buf *xdrbuf, page_base = 0; } - if (type == rpcrdma_readch) - goto out; - - /* When encoding a Write chunk, some servers need to see an - * extra segment for non-XDR-aligned Write chunks. The upper - * layer provides space in the tail iovec that may be used - * for this purpose. - */ - if (type == rpcrdma_writech && r_xprt->rx_ep->re_implicit_roundup) + if (type == rpcrdma_readch || type == rpcrdma_writech) goto out; if (xdrbuf->tail[0].iov_len) @@ -405,6 +397,7 @@ static int rpcrdma_encode_write_list(struct rpcrdma_xprt *r_xprt, enum rpcrdma_chunktype wtype) { struct xdr_stream *xdr = &req->rl_stream; + struct rpcrdma_ep *ep = r_xprt->rx_ep; struct rpcrdma_mr_seg *seg; struct rpcrdma_mr *mr; int nsegs, nchunks; @@ -443,6 +436,18 @@ static int rpcrdma_encode_write_list(struct rpcrdma_xprt *r_xprt, nsegs -= mr->mr_nents; } while (nsegs); + if (xdr_pad_size(rqst->rq_rcv_buf.page_len)) { + if (encode_rdma_segment(xdr, ep->re_write_pad_mr) < 0) + return -EMSGSIZE; + + trace_xprtrdma_chunk_wp(rqst->rq_task, ep->re_write_pad_mr, + nsegs); + r_xprt->rx_stats.write_chunk_count++; + r_xprt->rx_stats.total_rdma_request += mr->mr_length; + nchunks++; + nsegs -= mr->mr_nents; + } + /* Update count of segments in this Write chunk */ *segcount = cpu_to_be32(nchunks); diff --git a/net/sunrpc/xprtrdma/verbs.c b/net/sunrpc/xprtrdma/verbs.c index aaec3c9be8db..c3784b7b6855 100644 --- a/net/sunrpc/xprtrdma/verbs.c +++ b/net/sunrpc/xprtrdma/verbs.c @@ -551,6 +551,7 @@ int rpcrdma_xprt_connect(struct rpcrdma_xprt *r_xprt) goto out; } rpcrdma_mrs_create(r_xprt); + frwr_wp_create(r_xprt); out: trace_xprtrdma_connect(r_xprt, rc); diff --git a/net/sunrpc/xprtrdma/xprt_rdma.h b/net/sunrpc/xprtrdma/xprt_rdma.h index d91f54eae00b..b6d8b3e6356c 100644 --- a/net/sunrpc/xprtrdma/xprt_rdma.h +++ b/net/sunrpc/xprtrdma/xprt_rdma.h @@ -68,12 +68,14 @@ /* * RDMA Endpoint -- connection endpoint details */ +struct rpcrdma_mr; struct rpcrdma_ep { struct kref re_kref; struct rdma_cm_id *re_id; struct ib_pd *re_pd; unsigned int re_max_rdma_segs; unsigned int re_max_fr_depth; + struct rpcrdma_mr *re_write_pad_mr; bool re_implicit_roundup; enum ib_mr_type re_mrtype; struct completion re_done; @@ -97,6 +99,8 @@ struct rpcrdma_ep { unsigned int re_inline_recv; /* negotiated */ atomic_t re_completion_ids; + + char re_write_pad[XDR_UNIT]; }; /* Pre-allocate extra Work Requests for handling reverse-direction @@ -535,6 +539,7 @@ int frwr_send(struct rpcrdma_xprt *r_xprt, struct rpcrdma_req *req); void frwr_reminv(struct rpcrdma_rep *rep, struct list_head *mrs); void frwr_unmap_sync(struct rpcrdma_xprt *r_xprt, struct rpcrdma_req *req); void frwr_unmap_async(struct rpcrdma_xprt *r_xprt, struct rpcrdma_req *req); +int frwr_wp_create(struct rpcrdma_xprt *r_xprt); /* * RPC/RDMA protocol calls - xprtrdma/rpc_rdma.c From 7a3d524c4cf520aa4501b66eac4a7d2339b018e6 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Tue, 5 Oct 2021 10:18:06 -0400 Subject: [PATCH 31/84] xprtrdma: Remove rpcrdma_ep::re_implicit_roundup Clean up: this field is no longer used. xprt_rdma_pad_optimize is also no longer used, but is left in place because it is part of the kernel/userspace API. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- net/sunrpc/xprtrdma/verbs.c | 2 -- net/sunrpc/xprtrdma/xprt_rdma.h | 1 - 2 files changed, 3 deletions(-) diff --git a/net/sunrpc/xprtrdma/verbs.c b/net/sunrpc/xprtrdma/verbs.c index c3784b7b6855..3d3673ba9e1e 100644 --- a/net/sunrpc/xprtrdma/verbs.c +++ b/net/sunrpc/xprtrdma/verbs.c @@ -205,14 +205,12 @@ static void rpcrdma_update_cm_private(struct rpcrdma_ep *ep, unsigned int rsize, wsize; /* Default settings for RPC-over-RDMA Version One */ - ep->re_implicit_roundup = xprt_rdma_pad_optimize; rsize = RPCRDMA_V1_DEF_INLINE_SIZE; wsize = RPCRDMA_V1_DEF_INLINE_SIZE; if (pmsg && pmsg->cp_magic == rpcrdma_cmp_magic && pmsg->cp_version == RPCRDMA_CMP_VERSION) { - ep->re_implicit_roundup = true; rsize = rpcrdma_decode_buffer_size(pmsg->cp_send_size); wsize = rpcrdma_decode_buffer_size(pmsg->cp_recv_size); } diff --git a/net/sunrpc/xprtrdma/xprt_rdma.h b/net/sunrpc/xprtrdma/xprt_rdma.h index b6d8b3e6356c..c79f92eeda76 100644 --- a/net/sunrpc/xprtrdma/xprt_rdma.h +++ b/net/sunrpc/xprtrdma/xprt_rdma.h @@ -76,7 +76,6 @@ struct rpcrdma_ep { unsigned int re_max_rdma_segs; unsigned int re_max_fr_depth; struct rpcrdma_mr *re_write_pad_mr; - bool re_implicit_roundup; enum ib_mr_type re_mrtype; struct completion re_done; unsigned int re_send_count; From b4776a341ec05e809d21e98db5ed49dbdc81d5d8 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Sat, 16 Oct 2021 18:02:24 -0400 Subject: [PATCH 32/84] SUNRPC: Tracepoints should display tk_pid and cl_clid as a fixed-size field For certain special cases, RPC-related tracepoints record a -1 as the task ID or the client ID. It's ugly for a trace event to display 4 billion in these cases. To help keep SUNRPC tracepoints consistent, create a macro that defines the print format specifiers for tk_pid and cl_clid. At some point in the future we might try tk_pid with a wider range of values than 0..64K so this makes it easier to make that change. RPC tracepoints now look like this: <...>-1276 [009] 149.720358: rpc_clnt_new: client=00000005 peer=[192.168.2.55]:20049 program=nfs server=klimt.ib <...>-1342 [004] 149.921234: rpc_xdr_recvfrom: task:0000001a@00000005 head=[0xff1242d9ab6dc01c,144] page=0 tail=[(nil),0] len=144 <...>-1342 [004] 149.921235: xprt_release_cong: task:0000001a@00000005 snd_task:ffffffff cong=256 cwnd=16384 <...>-1342 [004] 149.921235: xprt_put_cong: task:0000001a@00000005 snd_task:ffffffff cong=0 cwnd=16384 Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- fs/nfs/nfs4trace.h | 9 ++-- fs/nfs/nfstrace.h | 6 ++- include/trace/events/rpcgss.h | 18 +++++--- include/trace/events/rpcrdma.h | 42 +++++++++-------- include/trace/events/sunrpc.h | 74 ++++++++++++++++++------------ include/trace/events/sunrpc_base.h | 18 ++++++++ 6 files changed, 108 insertions(+), 59 deletions(-) create mode 100644 include/trace/events/sunrpc_base.h diff --git a/fs/nfs/nfs4trace.h b/fs/nfs/nfs4trace.h index 7a2567aa2b86..d4f061046c09 100644 --- a/fs/nfs/nfs4trace.h +++ b/fs/nfs/nfs4trace.h @@ -9,6 +9,7 @@ #define _TRACE_NFS4_H #include +#include TRACE_DEFINE_ENUM(EPERM); TRACE_DEFINE_ENUM(ENOENT); @@ -694,8 +695,8 @@ TRACE_EVENT(nfs4_xdr_bad_operation, __entry->expected = expected; ), - TP_printk( - "task:%u@%d xid=0x%08x operation=%u, expected=%u", + TP_printk(SUNRPC_TRACE_TASK_SPECIFIER + " xid=0x%08x operation=%u, expected=%u", __entry->task_id, __entry->client_id, __entry->xid, __entry->op, __entry->expected ) @@ -729,8 +730,8 @@ DECLARE_EVENT_CLASS(nfs4_xdr_event, __entry->error = error; ), - TP_printk( - "task:%u@%d xid=0x%08x error=%ld (%s) operation=%u", + TP_printk(SUNRPC_TRACE_TASK_SPECIFIER + " xid=0x%08x error=%ld (%s) operation=%u", __entry->task_id, __entry->client_id, __entry->xid, -__entry->error, show_nfsv4_errors(__entry->error), __entry->op diff --git a/fs/nfs/nfstrace.h b/fs/nfs/nfstrace.h index 44fd016a8e65..82b51120450f 100644 --- a/fs/nfs/nfstrace.h +++ b/fs/nfs/nfstrace.h @@ -11,6 +11,8 @@ #include #include +#include + #define nfs_show_file_type(ftype) \ __print_symbolic(ftype, \ { DT_UNKNOWN, "UNKNOWN" }, \ @@ -1409,8 +1411,8 @@ DECLARE_EVENT_CLASS(nfs_xdr_event, __assign_str(procedure, task->tk_msg.rpc_proc->p_name); ), - TP_printk( - "task:%u@%d xid=0x%08x %sv%d %s error=%ld (%s)", + TP_printk(SUNRPC_TRACE_TASK_SPECIFIER + " xid=0x%08x %sv%d %s error=%ld (%s)", __entry->task_id, __entry->client_id, __entry->xid, __get_str(program), __entry->version, __get_str(procedure), -__entry->error, diff --git a/include/trace/events/rpcgss.h b/include/trace/events/rpcgss.h index b2a2672e6632..3ba63319af3c 100644 --- a/include/trace/events/rpcgss.h +++ b/include/trace/events/rpcgss.h @@ -13,6 +13,8 @@ #include +#include + /** ** GSS-API related trace events **/ @@ -99,7 +101,7 @@ DECLARE_EVENT_CLASS(rpcgss_gssapi_event, __entry->maj_stat = maj_stat; ), - TP_printk("task:%u@%u maj_stat=%s", + TP_printk(SUNRPC_TRACE_TASK_SPECIFIER " maj_stat=%s", __entry->task_id, __entry->client_id, __entry->maj_stat == 0 ? "GSS_S_COMPLETE" : show_gss_status(__entry->maj_stat)) @@ -332,7 +334,8 @@ TRACE_EVENT(rpcgss_unwrap_failed, __entry->client_id = task->tk_client->cl_clid; ), - TP_printk("task:%u@%u", __entry->task_id, __entry->client_id) + TP_printk(SUNRPC_TRACE_TASK_SPECIFIER, + __entry->task_id, __entry->client_id) ); TRACE_EVENT(rpcgss_bad_seqno, @@ -358,7 +361,8 @@ TRACE_EVENT(rpcgss_bad_seqno, __entry->received = received; ), - TP_printk("task:%u@%u expected seqno %u, received seqno %u", + TP_printk(SUNRPC_TRACE_TASK_SPECIFIER + " expected seqno %u, received seqno %u", __entry->task_id, __entry->client_id, __entry->expected, __entry->received) ); @@ -386,7 +390,7 @@ TRACE_EVENT(rpcgss_seqno, __entry->seqno = rqst->rq_seqno; ), - TP_printk("task:%u@%u xid=0x%08x seqno=%u", + TP_printk(SUNRPC_TRACE_TASK_SPECIFIER " xid=0x%08x seqno=%u", __entry->task_id, __entry->client_id, __entry->xid, __entry->seqno) ); @@ -418,7 +422,8 @@ TRACE_EVENT(rpcgss_need_reencode, __entry->ret = ret; ), - TP_printk("task:%u@%u xid=0x%08x rq_seqno=%u seq_xmit=%u reencode %sneeded", + TP_printk(SUNRPC_TRACE_TASK_SPECIFIER + " xid=0x%08x rq_seqno=%u seq_xmit=%u reencode %sneeded", __entry->task_id, __entry->client_id, __entry->xid, __entry->seqno, __entry->seq_xmit, __entry->ret ? "" : "un") @@ -452,7 +457,8 @@ TRACE_EVENT(rpcgss_update_slack, __entry->verfsize = auth->au_verfsize; ), - TP_printk("task:%u@%u xid=0x%08x auth=%p rslack=%u ralign=%u verfsize=%u\n", + TP_printk(SUNRPC_TRACE_TASK_SPECIFIER + " xid=0x%08x auth=%p rslack=%u ralign=%u verfsize=%u\n", __entry->task_id, __entry->client_id, __entry->xid, __entry->auth, __entry->rslack, __entry->ralign, __entry->verfsize) diff --git a/include/trace/events/rpcrdma.h b/include/trace/events/rpcrdma.h index afb2e394797c..7f46ef621c95 100644 --- a/include/trace/events/rpcrdma.h +++ b/include/trace/events/rpcrdma.h @@ -14,7 +14,9 @@ #include #include #include + #include +#include /** ** Event classes @@ -279,7 +281,8 @@ DECLARE_EVENT_CLASS(xprtrdma_rdch_event, __entry->nsegs = nsegs; ), - TP_printk("task:%u@%u pos=%u %u@0x%016llx:0x%08x (%s)", + TP_printk(SUNRPC_TRACE_TASK_SPECIFIER + " pos=%u %u@0x%016llx:0x%08x (%s)", __entry->task_id, __entry->client_id, __entry->pos, __entry->length, (unsigned long long)__entry->offset, __entry->handle, @@ -326,7 +329,8 @@ DECLARE_EVENT_CLASS(xprtrdma_wrch_event, __entry->nsegs = nsegs; ), - TP_printk("task:%u@%u %u@0x%016llx:0x%08x (%s)", + TP_printk(SUNRPC_TRACE_TASK_SPECIFIER + " %u@0x%016llx:0x%08x (%s)", __entry->task_id, __entry->client_id, __entry->length, (unsigned long long)__entry->offset, __entry->handle, @@ -393,7 +397,8 @@ DECLARE_EVENT_CLASS(xprtrdma_mr_class, __entry->dir = mr->mr_dir; ), - TP_printk("task:%u@%u mr.id=%u nents=%d %u@0x%016llx:0x%08x (%s)", + TP_printk(SUNRPC_TRACE_TASK_SPECIFIER + " mr.id=%u nents=%d %u@0x%016llx:0x%08x (%s)", __entry->task_id, __entry->client_id, __entry->mr_id, __entry->nents, __entry->length, (unsigned long long)__entry->offset, __entry->handle, @@ -636,9 +641,9 @@ TRACE_EVENT(xprtrdma_nomrs_err, __assign_str(port, rpcrdma_portstr(r_xprt)); ), - TP_printk("peer=[%s]:%s task:%u@%u", - __get_str(addr), __get_str(port), - __entry->task_id, __entry->client_id + TP_printk(SUNRPC_TRACE_TASK_SPECIFIER " peer=[%s]:%s", + __entry->task_id, __entry->client_id, + __get_str(addr), __get_str(port) ) ); @@ -700,7 +705,8 @@ TRACE_EVENT(xprtrdma_marshal, __entry->wtype = wtype; ), - TP_printk("task:%u@%u xid=0x%08x: hdr=%u xdr=%u/%u/%u %s/%s", + TP_printk(SUNRPC_TRACE_TASK_SPECIFIER + " xid=0x%08x hdr=%u xdr=%u/%u/%u %s/%s", __entry->task_id, __entry->client_id, __entry->xid, __entry->hdrlen, __entry->headlen, __entry->pagelen, __entry->taillen, @@ -730,7 +736,7 @@ TRACE_EVENT(xprtrdma_marshal_failed, __entry->ret = ret; ), - TP_printk("task:%u@%u xid=0x%08x: ret=%d", + TP_printk(SUNRPC_TRACE_TASK_SPECIFIER " xid=0x%08x ret=%d", __entry->task_id, __entry->client_id, __entry->xid, __entry->ret ) @@ -757,7 +763,7 @@ TRACE_EVENT(xprtrdma_prepsend_failed, __entry->ret = ret; ), - TP_printk("task:%u@%u xid=0x%08x: ret=%d", + TP_printk(SUNRPC_TRACE_TASK_SPECIFIER " xid=0x%08x ret=%d", __entry->task_id, __entry->client_id, __entry->xid, __entry->ret ) @@ -792,7 +798,7 @@ TRACE_EVENT(xprtrdma_post_send, __entry->signaled = req->rl_wr.send_flags & IB_SEND_SIGNALED; ), - TP_printk("task:%u@%u cq.id=%u cid=%d (%d SGE%s) %s", + TP_printk(SUNRPC_TRACE_TASK_SPECIFIER " cq.id=%u cid=%d (%d SGE%s) %s", __entry->task_id, __entry->client_id, __entry->cq_id, __entry->completion_id, __entry->num_sge, (__entry->num_sge == 1 ? "" : "s"), @@ -827,7 +833,7 @@ TRACE_EVENT(xprtrdma_post_send_err, __entry->rc = rc; ), - TP_printk("task:%u@%u cq.id=%u rc=%d", + TP_printk(SUNRPC_TRACE_TASK_SPECIFIER " cq.id=%u rc=%d", __entry->task_id, __entry->client_id, __entry->cq_id, __entry->rc ) @@ -939,7 +945,7 @@ TRACE_EVENT(xprtrdma_post_linv_err, __entry->status = status; ), - TP_printk("task:%u@%u status=%d", + TP_printk(SUNRPC_TRACE_TASK_SPECIFIER " status=%d", __entry->task_id, __entry->client_id, __entry->status ) ); @@ -1127,7 +1133,7 @@ TRACE_EVENT(xprtrdma_reply, __entry->credits = credits; ), - TP_printk("task:%u@%u xid=0x%08x credits=%u", + TP_printk(SUNRPC_TRACE_TASK_SPECIFIER " xid=0x%08x credits=%u", __entry->task_id, __entry->client_id, __entry->xid, __entry->credits ) @@ -1163,7 +1169,7 @@ TRACE_EVENT(xprtrdma_err_vers, __entry->max = be32_to_cpup(max); ), - TP_printk("task:%u@%u xid=0x%08x versions=[%u, %u]", + TP_printk(SUNRPC_TRACE_TASK_SPECIFIER " xid=0x%08x versions=[%u, %u]", __entry->task_id, __entry->client_id, __entry->xid, __entry->min, __entry->max ) @@ -1188,7 +1194,7 @@ TRACE_EVENT(xprtrdma_err_chunk, __entry->xid = be32_to_cpu(rqst->rq_xid); ), - TP_printk("task:%u@%u xid=0x%08x", + TP_printk(SUNRPC_TRACE_TASK_SPECIFIER " xid=0x%08x", __entry->task_id, __entry->client_id, __entry->xid ) ); @@ -1214,7 +1220,7 @@ TRACE_EVENT(xprtrdma_err_unrecognized, __entry->procedure = be32_to_cpup(procedure); ), - TP_printk("task:%u@%u xid=0x%08x procedure=%u", + TP_printk(SUNRPC_TRACE_TASK_SPECIFIER " xid=0x%08x procedure=%u", __entry->task_id, __entry->client_id, __entry->xid, __entry->procedure ) @@ -1246,7 +1252,7 @@ TRACE_EVENT(xprtrdma_fixup, __entry->taillen = rqst->rq_rcv_buf.tail[0].iov_len; ), - TP_printk("task:%u@%u fixup=%lu xdr=%zu/%u/%zu", + TP_printk(SUNRPC_TRACE_TASK_SPECIFIER " fixup=%lu xdr=%zu/%u/%zu", __entry->task_id, __entry->client_id, __entry->fixup, __entry->headlen, __entry->pagelen, __entry->taillen ) @@ -1296,7 +1302,7 @@ TRACE_EVENT(xprtrdma_mrs_zap, __entry->client_id = task->tk_client->cl_clid; ), - TP_printk("task:%u@%u", + TP_printk(SUNRPC_TRACE_TASK_SPECIFIER, __entry->task_id, __entry->client_id ) ); diff --git a/include/trace/events/sunrpc.h b/include/trace/events/sunrpc.h index 2d04eb96d418..92def7d6663e 100644 --- a/include/trace/events/sunrpc.h +++ b/include/trace/events/sunrpc.h @@ -14,6 +14,8 @@ #include #include +#include + TRACE_DEFINE_ENUM(SOCK_STREAM); TRACE_DEFINE_ENUM(SOCK_DGRAM); TRACE_DEFINE_ENUM(SOCK_RAW); @@ -78,7 +80,8 @@ DECLARE_EVENT_CLASS(rpc_xdr_buf_class, __entry->msg_len = xdr->len; ), - TP_printk("task:%u@%u head=[%p,%zu] page=%u tail=[%p,%zu] len=%u", + TP_printk(SUNRPC_TRACE_TASK_SPECIFIER + " head=[%p,%zu] page=%u tail=[%p,%zu] len=%u", __entry->task_id, __entry->client_id, __entry->head_base, __entry->head_len, __entry->page_len, __entry->tail_base, __entry->tail_len, __entry->msg_len @@ -114,7 +117,7 @@ DECLARE_EVENT_CLASS(rpc_clnt_class, __entry->client_id = clnt->cl_clid; ), - TP_printk("clid=%u", __entry->client_id) + TP_printk("client=" SUNRPC_TRACE_CLID_SPECIFIER, __entry->client_id) ); #define DEFINE_RPC_CLNT_EVENT(name) \ @@ -158,7 +161,8 @@ TRACE_EVENT(rpc_clnt_new, __assign_str(server, server); ), - TP_printk("client=%u peer=[%s]:%s program=%s server=%s", + TP_printk("client=" SUNRPC_TRACE_CLID_SPECIFIER + " peer=[%s]:%s program=%s server=%s", __entry->client_id, __get_str(addr), __get_str(port), __get_str(program), __get_str(server)) ); @@ -206,7 +210,8 @@ TRACE_EVENT(rpc_clnt_clone_err, __entry->error = error; ), - TP_printk("client=%u error=%d", __entry->client_id, __entry->error) + TP_printk("client=" SUNRPC_TRACE_CLID_SPECIFIER " error=%d", + __entry->client_id, __entry->error) ); @@ -248,7 +253,7 @@ DECLARE_EVENT_CLASS(rpc_task_status, __entry->status = task->tk_status; ), - TP_printk("task:%u@%u status=%d", + TP_printk(SUNRPC_TRACE_TASK_SPECIFIER " status=%d", __entry->task_id, __entry->client_id, __entry->status) ); @@ -288,7 +293,7 @@ TRACE_EVENT(rpc_request, __assign_str(procname, rpc_proc_name(task)); ), - TP_printk("task:%u@%u %sv%d %s (%ssync)", + TP_printk(SUNRPC_TRACE_TASK_SPECIFIER " %sv%d %s (%ssync)", __entry->task_id, __entry->client_id, __get_str(progname), __entry->version, __get_str(procname), __entry->async ? "a": "" @@ -348,7 +353,8 @@ DECLARE_EVENT_CLASS(rpc_task_running, __entry->flags = task->tk_flags; ), - TP_printk("task:%u@%d flags=%s runstate=%s status=%d action=%ps", + TP_printk(SUNRPC_TRACE_TASK_SPECIFIER + " flags=%s runstate=%s status=%d action=%ps", __entry->task_id, __entry->client_id, rpc_show_task_flags(__entry->flags), rpc_show_runstate(__entry->runstate), @@ -400,7 +406,8 @@ DECLARE_EVENT_CLASS(rpc_task_queued, __assign_str(q_name, rpc_qname(q)); ), - TP_printk("task:%u@%d flags=%s runstate=%s status=%d timeout=%lu queue=%s", + TP_printk(SUNRPC_TRACE_TASK_SPECIFIER + " flags=%s runstate=%s status=%d timeout=%lu queue=%s", __entry->task_id, __entry->client_id, rpc_show_task_flags(__entry->flags), rpc_show_runstate(__entry->runstate), @@ -436,7 +443,7 @@ DECLARE_EVENT_CLASS(rpc_failure, __entry->client_id = task->tk_client->cl_clid; ), - TP_printk("task:%u@%u", + TP_printk(SUNRPC_TRACE_TASK_SPECIFIER, __entry->task_id, __entry->client_id) ); @@ -478,7 +485,8 @@ DECLARE_EVENT_CLASS(rpc_reply_event, __assign_str(servername, task->tk_xprt->servername); ), - TP_printk("task:%u@%d server=%s xid=0x%08x %sv%d %s", + TP_printk(SUNRPC_TRACE_TASK_SPECIFIER + " server=%s xid=0x%08x %sv%d %s", __entry->task_id, __entry->client_id, __get_str(servername), __entry->xid, __get_str(progname), __entry->version, __get_str(procname)) @@ -538,7 +546,8 @@ TRACE_EVENT(rpc_buf_alloc, __entry->status = status; ), - TP_printk("task:%u@%u callsize=%zu recvsize=%zu status=%d", + TP_printk(SUNRPC_TRACE_TASK_SPECIFIER + " callsize=%zu recvsize=%zu status=%d", __entry->task_id, __entry->client_id, __entry->callsize, __entry->recvsize, __entry->status ) @@ -567,7 +576,8 @@ TRACE_EVENT(rpc_call_rpcerror, __entry->rpc_status = rpc_status; ), - TP_printk("task:%u@%u tk_status=%d rpc_status=%d", + TP_printk(SUNRPC_TRACE_TASK_SPECIFIER + " tk_status=%d rpc_status=%d", __entry->task_id, __entry->client_id, __entry->tk_status, __entry->rpc_status) ); @@ -607,7 +617,8 @@ TRACE_EVENT(rpc_stats_latency, __entry->execute = ktime_to_us(execute); ), - TP_printk("task:%u@%d xid=0x%08x %sv%d %s backlog=%lu rtt=%lu execute=%lu", + TP_printk(SUNRPC_TRACE_TASK_SPECIFIER + " xid=0x%08x %sv%d %s backlog=%lu rtt=%lu execute=%lu", __entry->task_id, __entry->client_id, __entry->xid, __get_str(progname), __entry->version, __get_str(procname), __entry->backlog, __entry->rtt, __entry->execute) @@ -651,8 +662,8 @@ TRACE_EVENT(rpc_xdr_overflow, __entry->version = task->tk_client->cl_vers; __assign_str(procedure, task->tk_msg.rpc_proc->p_name); } else { - __entry->task_id = 0; - __entry->client_id = 0; + __entry->task_id = -1; + __entry->client_id = -1; __assign_str(progname, "unknown"); __entry->version = 0; __assign_str(procedure, "unknown"); @@ -668,8 +679,8 @@ TRACE_EVENT(rpc_xdr_overflow, __entry->len = xdr->buf->len; ), - TP_printk( - "task:%u@%u %sv%d %s requested=%zu p=%p end=%p xdr=[%p,%zu]/%u/[%p,%zu]/%u\n", + TP_printk(SUNRPC_TRACE_TASK_SPECIFIER + " %sv%d %s requested=%zu p=%p end=%p xdr=[%p,%zu]/%u/[%p,%zu]/%u\n", __entry->task_id, __entry->client_id, __get_str(progname), __entry->version, __get_str(procedure), __entry->requested, __entry->p, __entry->end, @@ -727,8 +738,8 @@ TRACE_EVENT(rpc_xdr_alignment, __entry->len = xdr->buf->len; ), - TP_printk( - "task:%u@%u %sv%d %s offset=%zu copied=%u xdr=[%p,%zu]/%u/[%p,%zu]/%u\n", + TP_printk(SUNRPC_TRACE_TASK_SPECIFIER + " %sv%d %s offset=%zu copied=%u xdr=[%p,%zu]/%u/[%p,%zu]/%u\n", __entry->task_id, __entry->client_id, __get_str(progname), __entry->version, __get_str(procedure), __entry->offset, __entry->copied, @@ -917,7 +928,8 @@ TRACE_EVENT(rpc_socket_nospace, __entry->remaining = rqst->rq_slen - transport->xmit.offset; ), - TP_printk("task:%u@%u total=%u remaining=%u", + TP_printk(SUNRPC_TRACE_TASK_SPECIFIER + " total=%u remaining=%u", __entry->task_id, __entry->client_id, __entry->total, __entry->remaining ) @@ -1042,8 +1054,8 @@ TRACE_EVENT(xprt_transmit, __entry->status = status; ), - TP_printk( - "task:%u@%u xid=0x%08x seqno=%u status=%d", + TP_printk(SUNRPC_TRACE_TASK_SPECIFIER + " xid=0x%08x seqno=%u status=%d", __entry->task_id, __entry->client_id, __entry->xid, __entry->seqno, __entry->status) ); @@ -1082,8 +1094,8 @@ TRACE_EVENT(xprt_retransmit, __assign_str(procname, rpc_proc_name(task)); ), - TP_printk( - "task:%u@%u xid=0x%08x %sv%d %s ntrans=%d timeout=%lu", + TP_printk(SUNRPC_TRACE_TASK_SPECIFIER + " xid=0x%08x %sv%d %s ntrans=%d timeout=%lu", __entry->task_id, __entry->client_id, __entry->xid, __get_str(progname), __entry->version, __get_str(procname), __entry->ntrans, __entry->timeout @@ -1137,7 +1149,8 @@ DECLARE_EVENT_CLASS(xprt_writelock_event, xprt->snd_task->tk_pid : -1; ), - TP_printk("task:%u@%u snd_task:%u", + TP_printk(SUNRPC_TRACE_TASK_SPECIFIER + " snd_task:" SUNRPC_TRACE_PID_SPECIFIER, __entry->task_id, __entry->client_id, __entry->snd_task_id) ); @@ -1185,7 +1198,9 @@ DECLARE_EVENT_CLASS(xprt_cong_event, __entry->wait = test_bit(XPRT_CWND_WAIT, &xprt->state); ), - TP_printk("task:%u@%u snd_task:%u cong=%lu cwnd=%lu%s", + TP_printk(SUNRPC_TRACE_TASK_SPECIFIER + " snd_task:" SUNRPC_TRACE_PID_SPECIFIER + " cong=%lu cwnd=%lu%s", __entry->task_id, __entry->client_id, __entry->snd_task_id, __entry->cong, __entry->cwnd, __entry->wait ? " (wait)" : "") @@ -1223,7 +1238,7 @@ TRACE_EVENT(xprt_reserve, __entry->xid = be32_to_cpu(rqst->rq_xid); ), - TP_printk("task:%u@%u xid=0x%08x", + TP_printk(SUNRPC_TRACE_TASK_SPECIFIER " xid=0x%08x", __entry->task_id, __entry->client_id, __entry->xid ) ); @@ -1312,7 +1327,8 @@ TRACE_EVENT(rpcb_getport, __assign_str(servername, task->tk_xprt->servername); ), - TP_printk("task:%u@%u server=%s program=%u version=%u protocol=%d bind_version=%u", + TP_printk(SUNRPC_TRACE_TASK_SPECIFIER + " server=%s program=%u version=%u protocol=%d bind_version=%u", __entry->task_id, __entry->client_id, __get_str(servername), __entry->program, __entry->version, __entry->protocol, __entry->bind_version @@ -1342,7 +1358,7 @@ TRACE_EVENT(rpcb_setport, __entry->port = port; ), - TP_printk("task:%u@%u status=%d port=%u", + TP_printk(SUNRPC_TRACE_TASK_SPECIFIER " status=%d port=%u", __entry->task_id, __entry->client_id, __entry->status, __entry->port ) diff --git a/include/trace/events/sunrpc_base.h b/include/trace/events/sunrpc_base.h new file mode 100644 index 000000000000..588557d07ea8 --- /dev/null +++ b/include/trace/events/sunrpc_base.h @@ -0,0 +1,18 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Copyright (c) 2021 Oracle and/or its affiliates. + * + * Common types and format specifiers for sunrpc. + */ + +#if !defined(_TRACE_SUNRPC_BASE_H) +#define _TRACE_SUNRPC_BASE_H + +#include + +#define SUNRPC_TRACE_PID_SPECIFIER "%08x" +#define SUNRPC_TRACE_CLID_SPECIFIER "%08x" +#define SUNRPC_TRACE_TASK_SPECIFIER \ + "task:" SUNRPC_TRACE_PID_SPECIFIER "@" SUNRPC_TRACE_CLID_SPECIFIER + +#endif /* _TRACE_SUNRPC_BASE_H */ From 76497b1adb89175eee85afc437f08a68247314b3 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Sat, 16 Oct 2021 18:02:38 -0400 Subject: [PATCH 33/84] SUNRPC: Use BIT() macro in rpc_show_xprt_state() Clean up: BIT() is preferred over open-coding the shift. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/trace/events/sunrpc.h | 24 ++++++++++++------------ 1 file changed, 12 insertions(+), 12 deletions(-) diff --git a/include/trace/events/sunrpc.h b/include/trace/events/sunrpc.h index 92def7d6663e..f8b12eb07693 100644 --- a/include/trace/events/sunrpc.h +++ b/include/trace/events/sunrpc.h @@ -937,18 +937,18 @@ TRACE_EVENT(rpc_socket_nospace, #define rpc_show_xprt_state(x) \ __print_flags(x, "|", \ - { (1UL << XPRT_LOCKED), "LOCKED"}, \ - { (1UL << XPRT_CONNECTED), "CONNECTED"}, \ - { (1UL << XPRT_CONNECTING), "CONNECTING"}, \ - { (1UL << XPRT_CLOSE_WAIT), "CLOSE_WAIT"}, \ - { (1UL << XPRT_BOUND), "BOUND"}, \ - { (1UL << XPRT_BINDING), "BINDING"}, \ - { (1UL << XPRT_CLOSING), "CLOSING"}, \ - { (1UL << XPRT_OFFLINE), "OFFLINE"}, \ - { (1UL << XPRT_REMOVE), "REMOVE"}, \ - { (1UL << XPRT_CONGESTED), "CONGESTED"}, \ - { (1UL << XPRT_CWND_WAIT), "CWND_WAIT"}, \ - { (1UL << XPRT_WRITE_SPACE), "WRITE_SPACE"}) + { BIT(XPRT_LOCKED), "LOCKED" }, \ + { BIT(XPRT_CONNECTED), "CONNECTED" }, \ + { BIT(XPRT_CONNECTING), "CONNECTING" }, \ + { BIT(XPRT_CLOSE_WAIT), "CLOSE_WAIT" }, \ + { BIT(XPRT_BOUND), "BOUND" }, \ + { BIT(XPRT_BINDING), "BINDING" }, \ + { BIT(XPRT_CLOSING), "CLOSING" }, \ + { BIT(XPRT_OFFLINE), "OFFLINE" }, \ + { BIT(XPRT_REMOVE), "REMOVE" }, \ + { BIT(XPRT_CONGESTED), "CONGESTED" }, \ + { BIT(XPRT_CWND_WAIT), "CWND_WAIT" }, \ + { BIT(XPRT_WRITE_SPACE), "WRITE_SPACE" }) DECLARE_EVENT_CLASS(rpc_xprt_lifetime_class, TP_PROTO( From d9f877433ef8ba2a79d2abd921ddaf2d301e24bb Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Sat, 16 Oct 2021 18:02:51 -0400 Subject: [PATCH 34/84] NFS: Replace dprintk callsites in nfs_readpage(s) These new events report slightly different information for readpage and readpages/readahead. For readpage: fsx-1387 [006] 380.761896: nfs_aop_readpage: fileid=00:28:2 fhandle=0x36fbbe51 version=1752899355910932437 offset=131072 fsx-1387 [006] 380.761900: nfs_aop_readpage_done: fileid=00:28:2 fhandle=0x36fbbe51 version=1752899355910932437 offset=131072 ret=0 The index of a synchronous single-page read is reported. For readpages: fsx-1387 [006] 380.760847: nfs_aop_readahead: fileid=00:28:2 fhandle=0x36fbbe51 version=1752899355909932456 nr_pages=3 fsx-1387 [006] 380.760853: nfs_aop_readahead_done: fileid=00:28:2 fhandle=0x36fbbe51 version=1752899355909932456 nr_pages=3 ret=0 The count of pages requested is reported. nfs_readpages does not wait for the READ requests to complete. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- fs/nfs/nfstrace.h | 146 ++++++++++++++++++++++++++++++++++++++++++++++ fs/nfs/read.c | 11 ++-- 2 files changed, 151 insertions(+), 6 deletions(-) diff --git a/fs/nfs/nfstrace.h b/fs/nfs/nfstrace.h index 82b51120450f..83e9615c8b8c 100644 --- a/fs/nfs/nfstrace.h +++ b/fs/nfs/nfstrace.h @@ -862,6 +862,152 @@ TRACE_EVENT(nfs_sillyrename_unlink, ) ); +TRACE_EVENT(nfs_aop_readpage, + TP_PROTO( + const struct inode *inode, + struct page *page + ), + + TP_ARGS(inode, page), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __field(u64, version) + __field(loff_t, offset) + ), + + TP_fast_assign( + const struct nfs_inode *nfsi = NFS_I(inode); + + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = nfsi->fileid; + __entry->fhandle = nfs_fhandle_hash(&nfsi->fh); + __entry->version = inode_peek_iversion_raw(inode); + __entry->offset = page_index(page) << PAGE_SHIFT; + ), + + TP_printk( + "fileid=%02x:%02x:%llu fhandle=0x%08x version=%llu offset=%lld", + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, __entry->version, + __entry->offset + ) +); + +TRACE_EVENT(nfs_aop_readpage_done, + TP_PROTO( + const struct inode *inode, + struct page *page, + int ret + ), + + TP_ARGS(inode, page, ret), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u32, fhandle) + __field(int, ret) + __field(u64, fileid) + __field(u64, version) + __field(loff_t, offset) + ), + + TP_fast_assign( + const struct nfs_inode *nfsi = NFS_I(inode); + + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = nfsi->fileid; + __entry->fhandle = nfs_fhandle_hash(&nfsi->fh); + __entry->version = inode_peek_iversion_raw(inode); + __entry->offset = page_index(page) << PAGE_SHIFT; + __entry->ret = ret; + ), + + TP_printk( + "fileid=%02x:%02x:%llu fhandle=0x%08x version=%llu offset=%lld ret=%d", + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, __entry->version, + __entry->offset, __entry->ret + ) +); + +TRACE_EVENT(nfs_aop_readahead, + TP_PROTO( + const struct inode *inode, + unsigned int nr_pages + ), + + TP_ARGS(inode, nr_pages), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __field(u64, version) + __field(unsigned int, nr_pages) + ), + + TP_fast_assign( + const struct nfs_inode *nfsi = NFS_I(inode); + + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = nfsi->fileid; + __entry->fhandle = nfs_fhandle_hash(&nfsi->fh); + __entry->version = inode_peek_iversion_raw(inode); + __entry->nr_pages = nr_pages; + ), + + TP_printk( + "fileid=%02x:%02x:%llu fhandle=0x%08x version=%llu nr_pages=%u", + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, __entry->version, + __entry->nr_pages + ) +); + +TRACE_EVENT(nfs_aop_readahead_done, + TP_PROTO( + const struct inode *inode, + unsigned int nr_pages, + int ret + ), + + TP_ARGS(inode, nr_pages, ret), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u32, fhandle) + __field(int, ret) + __field(u64, fileid) + __field(u64, version) + __field(unsigned int, nr_pages) + ), + + TP_fast_assign( + const struct nfs_inode *nfsi = NFS_I(inode); + + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = nfsi->fileid; + __entry->fhandle = nfs_fhandle_hash(&nfsi->fh); + __entry->version = inode_peek_iversion_raw(inode); + __entry->nr_pages = nr_pages; + __entry->ret = ret; + ), + + TP_printk( + "fileid=%02x:%02x:%llu fhandle=0x%08x version=%llu nr_pages=%u ret=%d", + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, __entry->version, + __entry->nr_pages, __entry->ret + ) +); + TRACE_EVENT(nfs_initiate_read, TP_PROTO( const struct nfs_pgio_header *hdr diff --git a/fs/nfs/read.c b/fs/nfs/read.c index 08d6cc57cbc3..c8273d4b12ad 100644 --- a/fs/nfs/read.c +++ b/fs/nfs/read.c @@ -337,8 +337,7 @@ int nfs_readpage(struct file *file, struct page *page) struct inode *inode = page_file_mapping(page)->host; int ret; - dprintk("NFS: nfs_readpage (%p %ld@%lu)\n", - page, PAGE_SIZE, page_index(page)); + trace_nfs_aop_readpage(inode, page); nfs_inc_stats(inode, NFSIOS_VFSREADPAGE); /* @@ -390,9 +389,11 @@ out_wait: } out: put_nfs_open_context(desc.ctx); + trace_nfs_aop_readpage_done(inode, page, ret); return ret; out_unlock: unlock_page(page); + trace_nfs_aop_readpage_done(inode, page, ret); return ret; } @@ -403,10 +404,7 @@ int nfs_readpages(struct file *file, struct address_space *mapping, struct inode *inode = mapping->host; int ret; - dprintk("NFS: nfs_readpages (%s/%Lu %d)\n", - inode->i_sb->s_id, - (unsigned long long)NFS_FILEID(inode), - nr_pages); + trace_nfs_aop_readahead(inode, nr_pages); nfs_inc_stats(inode, NFSIOS_VFSREADPAGES); ret = -ESTALE; @@ -439,6 +437,7 @@ int nfs_readpages(struct file *file, struct address_space *mapping, read_complete: put_nfs_open_context(desc.ctx); out: + trace_nfs_aop_readahead_done(inode, nr_pages, ret); return ret; } From b40887e10dcacc5e8ae3c1a99dcba20877c4831b Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Sat, 16 Oct 2021 18:02:57 -0400 Subject: [PATCH 35/84] SUNRPC: Trace calls to .rpc_call_done Introduce a single tracepoint that can replace simple dprintk call sites in upper layer "rpc_call_done" callbacks. Example: kworker/u24:2-1254 [001] 771.026677: rpc_stats_latency: task:00000001@00000002 xid=0x16a6f3c0 rpcbindv2 GETPORT backlog=446 rtt=101 execute=555 kworker/u24:2-1254 [001] 771.026677: rpc_task_call_done: task:00000001@00000002 flags=ASYNC|DYNAMIC|SOFT|SOFTCONN|SENT runstate=RUNNING|ACTIVE status=0 action=rpcb_getport_done kworker/u24:2-1254 [001] 771.026678: rpcb_setport: task:00000001@00000002 status=0 port=20048 Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- fs/lockd/clntproc.c | 3 --- fs/lockd/svc4proc.c | 2 -- fs/lockd/svcproc.c | 2 -- fs/nfs/filelayout/filelayout.c | 2 -- fs/nfs/flexfilelayout/flexfilelayout.c | 2 -- fs/nfs/pagelist.c | 3 --- fs/nfs/write.c | 3 --- include/trace/events/sunrpc.h | 1 + net/sunrpc/sched.c | 1 + 9 files changed, 2 insertions(+), 17 deletions(-) diff --git a/fs/lockd/clntproc.c b/fs/lockd/clntproc.c index b11f2afa84f1..99fffc9cb958 100644 --- a/fs/lockd/clntproc.c +++ b/fs/lockd/clntproc.c @@ -794,9 +794,6 @@ static void nlmclnt_cancel_callback(struct rpc_task *task, void *data) goto retry_cancel; } - dprintk("lockd: cancel status %u (task %u)\n", - status, task->tk_pid); - switch (status) { case NLM_LCK_GRANTED: case NLM_LCK_DENIED_GRACE_PERIOD: diff --git a/fs/lockd/svc4proc.c b/fs/lockd/svc4proc.c index e10ae2c41279..176b468a61c7 100644 --- a/fs/lockd/svc4proc.c +++ b/fs/lockd/svc4proc.c @@ -269,8 +269,6 @@ nlm4svc_proc_granted(struct svc_rqst *rqstp) */ static void nlm4svc_callback_exit(struct rpc_task *task, void *data) { - dprintk("lockd: %5u callback returned %d\n", task->tk_pid, - -task->tk_status); } static void nlm4svc_callback_release(void *data) diff --git a/fs/lockd/svcproc.c b/fs/lockd/svcproc.c index 99696d3f6dd6..4dc1b40a489a 100644 --- a/fs/lockd/svcproc.c +++ b/fs/lockd/svcproc.c @@ -301,8 +301,6 @@ nlmsvc_proc_granted(struct svc_rqst *rqstp) */ static void nlmsvc_callback_exit(struct rpc_task *task, void *data) { - dprintk("lockd: %5u callback returned %d\n", task->tk_pid, - -task->tk_status); } void nlmsvc_release_call(struct nlm_rqst *call) diff --git a/fs/nfs/filelayout/filelayout.c b/fs/nfs/filelayout/filelayout.c index d2103852475f..9c96e3e5ed35 100644 --- a/fs/nfs/filelayout/filelayout.c +++ b/fs/nfs/filelayout/filelayout.c @@ -293,8 +293,6 @@ static void filelayout_read_call_done(struct rpc_task *task, void *data) { struct nfs_pgio_header *hdr = data; - dprintk("--> %s task->tk_status %d\n", __func__, task->tk_status); - if (test_bit(NFS_IOHDR_REDO, &hdr->flags) && task->tk_status == 0) { nfs41_sequence_done(task, &hdr->res.seq_res); diff --git a/fs/nfs/flexfilelayout/flexfilelayout.c b/fs/nfs/flexfilelayout/flexfilelayout.c index d383de00d486..a553d59afa8b 100644 --- a/fs/nfs/flexfilelayout/flexfilelayout.c +++ b/fs/nfs/flexfilelayout/flexfilelayout.c @@ -1414,8 +1414,6 @@ static void ff_layout_read_call_done(struct rpc_task *task, void *data) { struct nfs_pgio_header *hdr = data; - dprintk("--> %s task->tk_status %d\n", __func__, task->tk_status); - if (test_bit(NFS_IOHDR_REDO, &hdr->flags) && task->tk_status == 0) { nfs4_sequence_done(task, &hdr->res.seq_res); diff --git a/fs/nfs/pagelist.c b/fs/nfs/pagelist.c index 72333bcaa4c4..ad7f83dc9a2d 100644 --- a/fs/nfs/pagelist.c +++ b/fs/nfs/pagelist.c @@ -864,9 +864,6 @@ static void nfs_pgio_result(struct rpc_task *task, void *calldata) struct nfs_pgio_header *hdr = calldata; struct inode *inode = hdr->inode; - dprintk("NFS: %s: %5u, (status %d)\n", __func__, - task->tk_pid, task->tk_status); - if (hdr->rw_ops->rw_done(task, hdr, inode) != 0) return; if (task->tk_status < 0) diff --git a/fs/nfs/write.c b/fs/nfs/write.c index 465220f47142..82c5b89395f6 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -1840,9 +1840,6 @@ static void nfs_commit_done(struct rpc_task *task, void *calldata) { struct nfs_commit_data *data = calldata; - dprintk("NFS: %5u nfs_commit_done (status %d)\n", - task->tk_pid, task->tk_status); - /* Call the NFS version-specific code */ NFS_PROTO(data->inode)->commit_done(task, data); trace_nfs_commit_done(task, data); diff --git a/include/trace/events/sunrpc.h b/include/trace/events/sunrpc.h index f8b12eb07693..dc922e664820 100644 --- a/include/trace/events/sunrpc.h +++ b/include/trace/events/sunrpc.h @@ -378,6 +378,7 @@ DEFINE_RPC_RUNNING_EVENT(complete); DEFINE_RPC_RUNNING_EVENT(timeout); DEFINE_RPC_RUNNING_EVENT(signalled); DEFINE_RPC_RUNNING_EVENT(end); +DEFINE_RPC_RUNNING_EVENT(call_done); DECLARE_EVENT_CLASS(rpc_task_queued, diff --git a/net/sunrpc/sched.c b/net/sunrpc/sched.c index f4f311ea7a66..e2c835482791 100644 --- a/net/sunrpc/sched.c +++ b/net/sunrpc/sched.c @@ -837,6 +837,7 @@ void rpc_exit_task(struct rpc_task *task) else if (task->tk_client) rpc_count_iostats(task, task->tk_client->cl_metrics); if (task->tk_ops->rpc_call_done != NULL) { + trace_rpc_task_call_done(task, task->tk_ops->rpc_call_done); task->tk_ops->rpc_call_done(task, task->tk_calldata); if (task->tk_action != NULL) { /* Always release the RPC slot and buffer memory */ From 86882c75464920684d39b747d7f52a75200cc24f Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Sat, 16 Oct 2021 18:03:04 -0400 Subject: [PATCH 36/84] NFS: Remove --> and <-- dprintk call sites dprintk call sites that display no other information than the function name can be replaced with use of the trace "function" or "function_graph" plug-ins. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- fs/nfs/nfs4proc.c | 54 +++++------------------------------------------ 1 file changed, 5 insertions(+), 49 deletions(-) diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 9e89198ea21f..979665a91e80 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -3560,7 +3560,6 @@ static void nfs4_close_done(struct rpc_task *task, void *data) .stateid = &calldata->arg.stateid, }; - dprintk("%s: begin!\n", __func__); if (!nfs4_sequence_done(task, &calldata->res.seq_res)) return; trace_nfs4_close(state, &calldata->arg, &calldata->res, task->tk_status); @@ -3615,7 +3614,7 @@ out_release: task->tk_status = 0; nfs_release_seqid(calldata->arg.seqid); nfs_refresh_inode(calldata->inode, &calldata->fattr); - dprintk("%s: done, ret = %d!\n", __func__, task->tk_status); + dprintk("%s: ret = %d\n", __func__, task->tk_status); return; out_restart: task->tk_status = 0; @@ -3633,7 +3632,6 @@ static void nfs4_close_prepare(struct rpc_task *task, void *data) bool is_rdonly, is_wronly, is_rdwr; int call_close = 0; - dprintk("%s: begin!\n", __func__); if (nfs_wait_on_sequence(calldata->arg.seqid, task) != 0) goto out_wait; @@ -3707,7 +3705,6 @@ static void nfs4_close_prepare(struct rpc_task *task, void *data) &calldata->res.seq_res, task) != 0) nfs_release_seqid(calldata->arg.seqid); - dprintk("%s: done!\n", __func__); return; out_no_action: task->tk_action = NULL; @@ -5346,8 +5343,6 @@ static bool nfs4_read_plus_not_supported(struct rpc_task *task, static int nfs4_read_done(struct rpc_task *task, struct nfs_pgio_header *hdr) { - dprintk("--> %s\n", __func__); - if (!nfs4_sequence_done(task, &hdr->res.seq_res)) return -EAGAIN; if (nfs4_read_stateid_changed(task, &hdr->args)) @@ -7002,7 +6997,6 @@ static void nfs4_lock_prepare(struct rpc_task *task, void *calldata) struct nfs4_lockdata *data = calldata; struct nfs4_state *state = data->lsp->ls_state; - dprintk("%s: begin!\n", __func__); if (nfs_wait_on_sequence(data->arg.lock_seqid, task) != 0) goto out_wait; /* Do we need to do an open_to_lock_owner? */ @@ -7036,7 +7030,7 @@ out_release_lock_seqid: nfs_release_seqid(data->arg.lock_seqid); out_wait: nfs4_sequence_done(task, &data->res.seq_res); - dprintk("%s: done!, ret = %d\n", __func__, data->rpc_status); + dprintk("%s: ret = %d\n", __func__, data->rpc_status); } static void nfs4_lock_done(struct rpc_task *task, void *calldata) @@ -7044,8 +7038,6 @@ static void nfs4_lock_done(struct rpc_task *task, void *calldata) struct nfs4_lockdata *data = calldata; struct nfs4_lock_state *lsp = data->lsp; - dprintk("%s: begin!\n", __func__); - if (!nfs4_sequence_done(task, &data->res.seq_res)) return; @@ -7079,7 +7071,7 @@ static void nfs4_lock_done(struct rpc_task *task, void *calldata) goto out_restart; } out_done: - dprintk("%s: done, ret = %d!\n", __func__, data->rpc_status); + dprintk("%s: ret = %d!\n", __func__, data->rpc_status); return; out_restart: if (!data->cancelled) @@ -7091,7 +7083,6 @@ static void nfs4_lock_release(void *calldata) { struct nfs4_lockdata *data = calldata; - dprintk("%s: begin!\n", __func__); nfs_free_seqid(data->arg.open_seqid); if (data->cancelled && data->rpc_status == 0) { struct rpc_task *task; @@ -7105,7 +7096,6 @@ static void nfs4_lock_release(void *calldata) nfs4_put_lock_state(data->lsp); put_nfs_open_context(data->ctx); kfree(data); - dprintk("%s: done!\n", __func__); } static const struct rpc_call_ops nfs4_lock_ops = { @@ -7152,7 +7142,6 @@ static int _nfs4_do_setlk(struct nfs4_state *state, int cmd, struct file_lock *f if (client->cl_minorversion) task_setup_data.flags |= RPC_TASK_MOVEABLE; - dprintk("%s: begin!\n", __func__); data = nfs4_alloc_lockdata(fl, nfs_file_open_context(fl->fl_file), fl->fl_u.nfs4_fl.owner, recovery_type == NFS_LOCK_NEW ? GFP_KERNEL : GFP_NOFS); @@ -7183,7 +7172,7 @@ static int _nfs4_do_setlk(struct nfs4_state *state, int cmd, struct file_lock *f data->cancelled = true; trace_nfs4_set_lock(fl, state, &data->res.stateid, cmd, ret); rpc_put_task(task); - dprintk("%s: done, ret = %d!\n", __func__, ret); + dprintk("%s: ret = %d\n", __func__, ret); return ret; } @@ -8854,14 +8843,12 @@ static void nfs4_get_lease_time_prepare(struct rpc_task *task, struct nfs4_get_lease_time_data *data = (struct nfs4_get_lease_time_data *)calldata; - dprintk("--> %s\n", __func__); /* just setup sequence, do not trigger session recovery since we're invoked within one */ nfs4_setup_sequence(data->clp, &data->args->la_seq_args, &data->res->lr_seq_res, task); - dprintk("<-- %s\n", __func__); } /* @@ -8873,13 +8860,11 @@ static void nfs4_get_lease_time_done(struct rpc_task *task, void *calldata) struct nfs4_get_lease_time_data *data = (struct nfs4_get_lease_time_data *)calldata; - dprintk("--> %s\n", __func__); if (!nfs4_sequence_done(task, &data->res->lr_seq_res)) return; switch (task->tk_status) { case -NFS4ERR_DELAY: case -NFS4ERR_GRACE: - dprintk("%s Retry: tk_status %d\n", __func__, task->tk_status); rpc_delay(task, NFS4_POLL_RETRY_MIN); task->tk_status = 0; fallthrough; @@ -8887,7 +8872,6 @@ static void nfs4_get_lease_time_done(struct rpc_task *task, void *calldata) rpc_restart_call_prepare(task); return; } - dprintk("<-- %s\n", __func__); } static const struct rpc_call_ops nfs4_get_lease_time_ops = { @@ -9119,7 +9103,6 @@ int nfs4_proc_create_session(struct nfs_client *clp, const struct cred *cred) dprintk("%s client>seqid %d sessionid %u:%u:%u:%u\n", __func__, clp->cl_seqid, ptr[0], ptr[1], ptr[2], ptr[3]); out: - dprintk("<-- %s\n", __func__); return status; } @@ -9137,8 +9120,6 @@ int nfs4_proc_destroy_session(struct nfs4_session *session, }; int status = 0; - dprintk("--> nfs4_proc_destroy_session\n"); - /* session is still being setup */ if (!test_and_clear_bit(NFS4_SESSION_ESTABLISHED, &session->session_state)) return 0; @@ -9150,8 +9131,6 @@ int nfs4_proc_destroy_session(struct nfs4_session *session, if (status) dprintk("NFS: Got error %d from the server on DESTROY_SESSION. " "Session has been destroyed regardless...\n", status); - - dprintk("<-- nfs4_proc_destroy_session\n"); return status; } @@ -9199,7 +9178,7 @@ static void nfs41_sequence_call_done(struct rpc_task *task, void *data) if (task->tk_status < 0) { dprintk("%s ERROR %d\n", __func__, task->tk_status); if (refcount_read(&clp->cl_count) == 1) - goto out; + return; if (nfs41_sequence_handle_errors(task, clp) == -EAGAIN) { rpc_restart_call_prepare(task); @@ -9207,8 +9186,6 @@ static void nfs41_sequence_call_done(struct rpc_task *task, void *data) } } dprintk("%s rpc_cred %p\n", __func__, task->tk_msg.rpc_cred); -out: - dprintk("<-- %s\n", __func__); } static void nfs41_sequence_prepare(struct rpc_task *task, void *data) @@ -9355,7 +9332,6 @@ static void nfs4_reclaim_complete_done(struct rpc_task *task, void *data) struct nfs_client *clp = calldata->clp; struct nfs4_sequence_res *res = &calldata->res.seq_res; - dprintk("--> %s\n", __func__); if (!nfs41_sequence_done(task, res)) return; @@ -9364,7 +9340,6 @@ static void nfs4_reclaim_complete_done(struct rpc_task *task, void *data) rpc_restart_call_prepare(task); return; } - dprintk("<-- %s\n", __func__); } static void nfs4_free_reclaim_complete_data(void *data) @@ -9399,7 +9374,6 @@ static int nfs41_proc_reclaim_complete(struct nfs_client *clp, }; int status = -ENOMEM; - dprintk("--> %s\n", __func__); calldata = kzalloc(sizeof(*calldata), GFP_NOFS); if (calldata == NULL) goto out; @@ -9422,19 +9396,15 @@ nfs4_layoutget_prepare(struct rpc_task *task, void *calldata) struct nfs4_layoutget *lgp = calldata; struct nfs_server *server = NFS_SERVER(lgp->args.inode); - dprintk("--> %s\n", __func__); nfs4_setup_sequence(server->nfs_client, &lgp->args.seq_args, &lgp->res.seq_res, task); - dprintk("<-- %s\n", __func__); } static void nfs4_layoutget_done(struct rpc_task *task, void *calldata) { struct nfs4_layoutget *lgp = calldata; - dprintk("--> %s\n", __func__); nfs41_sequence_process(task, &lgp->res.seq_res); - dprintk("<-- %s\n", __func__); } static int @@ -9523,7 +9493,6 @@ nfs4_layoutget_handle_exception(struct rpc_task *task, status = err; } out: - dprintk("<-- %s\n", __func__); return status; } @@ -9537,10 +9506,8 @@ static void nfs4_layoutget_release(void *calldata) { struct nfs4_layoutget *lgp = calldata; - dprintk("--> %s\n", __func__); nfs4_sequence_free_slot(&lgp->res.seq_res); pnfs_layoutget_free(lgp); - dprintk("<-- %s\n", __func__); } static const struct rpc_call_ops nfs4_layoutget_call_ops = { @@ -9576,8 +9543,6 @@ nfs4_proc_layoutget(struct nfs4_layoutget *lgp, long *timeout) }; int status = 0; - dprintk("--> %s\n", __func__); - nfs4_init_sequence(&lgp->args.seq_args, &lgp->res.seq_res, 0, 0); task = rpc_run_task(&task_setup_data); @@ -9613,7 +9578,6 @@ nfs4_layoutreturn_prepare(struct rpc_task *task, void *calldata) { struct nfs4_layoutreturn *lrp = calldata; - dprintk("--> %s\n", __func__); nfs4_setup_sequence(lrp->clp, &lrp->args.seq_args, &lrp->res.seq_res, @@ -9627,8 +9591,6 @@ static void nfs4_layoutreturn_done(struct rpc_task *task, void *calldata) struct nfs4_layoutreturn *lrp = calldata; struct nfs_server *server; - dprintk("--> %s\n", __func__); - if (!nfs41_sequence_process(task, &lrp->res.seq_res)) return; @@ -9659,7 +9621,6 @@ static void nfs4_layoutreturn_done(struct rpc_task *task, void *calldata) break; goto out_restart; } - dprintk("<-- %s\n", __func__); return; out_restart: task->tk_status = 0; @@ -9672,7 +9633,6 @@ static void nfs4_layoutreturn_release(void *calldata) struct nfs4_layoutreturn *lrp = calldata; struct pnfs_layout_hdr *lo = lrp->args.layout; - dprintk("--> %s\n", __func__); pnfs_layoutreturn_free_lsegs(lo, &lrp->args.stateid, &lrp->args.range, lrp->res.lrs_present ? &lrp->res.stateid : NULL); nfs4_sequence_free_slot(&lrp->res.seq_res); @@ -9682,7 +9642,6 @@ static void nfs4_layoutreturn_release(void *calldata) nfs_iput_and_deactive(lrp->inode); put_cred(lrp->cred); kfree(calldata); - dprintk("<-- %s\n", __func__); } static const struct rpc_call_ops nfs4_layoutreturn_call_ops = { @@ -9713,7 +9672,6 @@ int nfs4_proc_layoutreturn(struct nfs4_layoutreturn *lrp, bool sync) NFS_SP4_MACH_CRED_PNFS_CLEANUP, &task_setup_data.rpc_client, &msg); - dprintk("--> %s\n", __func__); lrp->inode = nfs_igrab_and_active(lrp->args.inode); if (!sync) { if (!lrp->inode) { @@ -9760,7 +9718,6 @@ _nfs4_proc_getdeviceinfo(struct nfs_server *server, }; int status; - dprintk("--> %s\n", __func__); status = nfs4_call_sync(server->client, server, &msg, &args.seq_args, &res.seq_res, 0); if (res.notification & ~args.notify_types) dprintk("%s: unsupported notification\n", __func__); @@ -9932,7 +9889,6 @@ _nfs41_proc_secinfo_no_name(struct nfs_server *server, struct nfs_fh *fhandle, msg.rpc_cred = cred; } - dprintk("--> %s\n", __func__); nfs4_init_sequence(&args.seq_args, &res.seq_res, 0, 0); status = nfs4_call_sync_custom(&task_setup); dprintk("<-- %s status=%d\n", __func__, status); From 01dde76e471229e3437a2686c572f4980b2c483e Mon Sep 17 00:00:00 2001 From: Anna Schumaker Date: Thu, 14 Oct 2021 13:55:04 -0400 Subject: [PATCH 37/84] NFS: Create an nfs4_server_set_init_caps() function And call it before doing an FSINFO probe to reset to the baseline capabilities before probing. Signed-off-by: Anna Schumaker Signed-off-by: Trond Myklebust --- fs/nfs/internal.h | 1 + fs/nfs/nfs4client.c | 33 +++++++++++++++++++-------------- fs/nfs/nfs4proc.c | 2 ++ 3 files changed, 22 insertions(+), 14 deletions(-) diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h index 66fc936834f2..ee14d71da4a1 100644 --- a/fs/nfs/internal.h +++ b/fs/nfs/internal.h @@ -209,6 +209,7 @@ extern struct nfs_client * nfs4_find_client_sessionid(struct net *, const struct sockaddr *, struct nfs4_sessionid *, u32); extern struct nfs_server *nfs_create_server(struct fs_context *); +extern void nfs4_server_set_init_caps(struct nfs_server *); extern struct nfs_server *nfs4_create_server(struct fs_context *); extern struct nfs_server *nfs4_create_referral_server(struct fs_context *); extern int nfs4_update_server(struct nfs_server *server, const char *hostname, diff --git a/fs/nfs/nfs4client.c b/fs/nfs/nfs4client.c index af57332503be..3fb0ca92377c 100644 --- a/fs/nfs/nfs4client.c +++ b/fs/nfs/nfs4client.c @@ -1059,6 +1059,24 @@ static void nfs4_session_limit_xasize(struct nfs_server *server) #endif } +void nfs4_server_set_init_caps(struct nfs_server *server) +{ + /* Set the basic capabilities */ + server->caps |= server->nfs_client->cl_mvops->init_caps; + if (server->flags & NFS_MOUNT_NORDIRPLUS) + server->caps &= ~NFS_CAP_READDIRPLUS; + if (server->nfs_client->cl_proto == XPRT_TRANSPORT_RDMA) + server->caps &= ~NFS_CAP_READ_PLUS; + + /* + * Don't use NFS uid/gid mapping if we're using AUTH_SYS or lower + * authentication. + */ + if (nfs4_disable_idmapping && + server->client->cl_auth->au_flavor == RPC_AUTH_UNIX) + server->caps |= NFS_CAP_UIDGID_NOMAP; +} + static int nfs4_server_common_setup(struct nfs_server *server, struct nfs_fh *mntfh, bool auth_probe) { @@ -1078,20 +1096,7 @@ static int nfs4_server_common_setup(struct nfs_server *server, if (error < 0) goto out; - /* Set the basic capabilities */ - server->caps |= server->nfs_client->cl_mvops->init_caps; - if (server->flags & NFS_MOUNT_NORDIRPLUS) - server->caps &= ~NFS_CAP_READDIRPLUS; - if (server->nfs_client->cl_proto == XPRT_TRANSPORT_RDMA) - server->caps &= ~NFS_CAP_READ_PLUS; - /* - * Don't use NFS uid/gid mapping if we're using AUTH_SYS or lower - * authentication. - */ - if (nfs4_disable_idmapping && - server->client->cl_auth->au_flavor == RPC_AUTH_UNIX) - server->caps |= NFS_CAP_UIDGID_NOMAP; - + nfs4_server_set_init_caps(server); /* Probe the root fh to retrieve its FSID and filehandle */ error = nfs4_get_rootfh(server, mntfh, auth_probe); diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 979665a91e80..1c485edf1d07 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -3937,6 +3937,8 @@ int nfs4_server_capabilities(struct nfs_server *server, struct nfs_fh *fhandle) .interruptible = true, }; int err; + + nfs4_server_set_init_caps(server); do { err = nfs4_handle_exception(server, _nfs4_server_capabilities(server, fhandle), From e5731131fb6fefaa69064ca511b7c4971d6cf54f Mon Sep 17 00:00:00 2001 From: Anna Schumaker Date: Thu, 14 Oct 2021 13:55:05 -0400 Subject: [PATCH 38/84] NFS: Move nfs_probe_destination() into the generic client And rename it to nfs_probe_server(). I also change it to take the nfs_fh as an argument so callers can choose what filehandle to probe. Signed-off-by: Anna Schumaker Signed-off-by: Trond Myklebust --- fs/nfs/client.c | 24 ++++++++++++++++++++++++ fs/nfs/internal.h | 1 + fs/nfs/nfs4client.c | 26 +------------------------- 3 files changed, 26 insertions(+), 25 deletions(-) diff --git a/fs/nfs/client.c b/fs/nfs/client.c index 23e165d5ec9c..e867d9090386 100644 --- a/fs/nfs/client.c +++ b/fs/nfs/client.c @@ -864,6 +864,30 @@ int nfs_probe_fsinfo(struct nfs_server *server, struct nfs_fh *mntfh, struct nfs } EXPORT_SYMBOL_GPL(nfs_probe_fsinfo); +/* + * Grab the destination's particulars, including lease expiry time. + * + * Returns zero if probe succeeded and retrieved FSID matches the FSID + * we have cached. + */ +int nfs_probe_server(struct nfs_server *server, struct nfs_fh *mntfh) +{ + struct nfs_fattr *fattr; + int error; + + fattr = nfs_alloc_fattr(); + if (fattr == NULL) + return -ENOMEM; + + /* Sanity: the probe won't work if the destination server + * does not recognize the migrated FH. */ + error = nfs_probe_fsinfo(server, mntfh, fattr); + + nfs_free_fattr(fattr); + return error; +} +EXPORT_SYMBOL_GPL(nfs_probe_server); + /* * Copy useful information when duplicating a server record */ diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h index ee14d71da4a1..c1253dd60c32 100644 --- a/fs/nfs/internal.h +++ b/fs/nfs/internal.h @@ -194,6 +194,7 @@ extern struct nfs_client *nfs_alloc_client(const struct nfs_client_initdata *); int nfs_create_rpc_client(struct nfs_client *, const struct nfs_client_initdata *, rpc_authflavor_t); struct nfs_client *nfs_get_client(const struct nfs_client_initdata *); int nfs_probe_fsinfo(struct nfs_server *server, struct nfs_fh *, struct nfs_fattr *); +int nfs_probe_server(struct nfs_server *, struct nfs_fh *); void nfs_server_insert_lists(struct nfs_server *); void nfs_server_remove_lists(struct nfs_server *); void nfs_init_timeout_values(struct rpc_timeout *to, int proto, int timeo, int retrans); diff --git a/fs/nfs/nfs4client.c b/fs/nfs/nfs4client.c index 3fb0ca92377c..85978ecb727e 100644 --- a/fs/nfs/nfs4client.c +++ b/fs/nfs/nfs4client.c @@ -1293,30 +1293,6 @@ error: return ERR_PTR(error); } -/* - * Grab the destination's particulars, including lease expiry time. - * - * Returns zero if probe succeeded and retrieved FSID matches the FSID - * we have cached. - */ -static int nfs_probe_destination(struct nfs_server *server) -{ - struct inode *inode = d_inode(server->super->s_root); - struct nfs_fattr *fattr; - int error; - - fattr = nfs_alloc_fattr(); - if (fattr == NULL) - return -ENOMEM; - - /* Sanity: the probe won't work if the destination server - * does not recognize the migrated FH. */ - error = nfs_probe_fsinfo(server, NFS_FH(inode), fattr); - - nfs_free_fattr(fattr); - return error; -} - /** * nfs4_update_server - Move an nfs_server to a different nfs_client * @@ -1377,5 +1353,5 @@ int nfs4_update_server(struct nfs_server *server, const char *hostname, server->nfs_client->cl_hostname = kstrdup(hostname, GFP_KERNEL); nfs_server_insert_lists(server); - return nfs_probe_destination(server); + return nfs_probe_server(server, NFS_FH(d_inode(server->super->s_root))); } From 4d4cf8d2d6ccb43c68bc5925dc83500b81b50f9e Mon Sep 17 00:00:00 2001 From: Anna Schumaker Date: Thu, 14 Oct 2021 13:55:06 -0400 Subject: [PATCH 39/84] NFS: Replace calls to nfs_probe_fsinfo() with nfs_probe_server() Clean up. There are a few places where we want to probe the server, but don't actually care about the fsinfo result. Change these to use nfs_probe_server(), which handles the fattr allocation for us. Signed-off-by: Anna Schumaker Signed-off-by: Trond Myklebust --- fs/nfs/client.c | 10 +--------- fs/nfs/nfs4client.c | 8 +------- 2 files changed, 2 insertions(+), 16 deletions(-) diff --git a/fs/nfs/client.c b/fs/nfs/client.c index e867d9090386..1a882c78a794 100644 --- a/fs/nfs/client.c +++ b/fs/nfs/client.c @@ -1082,7 +1082,6 @@ struct nfs_server *nfs_clone_server(struct nfs_server *source, rpc_authflavor_t flavor) { struct nfs_server *server; - struct nfs_fattr *fattr_fsinfo; int error; server = nfs_alloc_server(); @@ -1091,11 +1090,6 @@ struct nfs_server *nfs_clone_server(struct nfs_server *source, server->cred = get_cred(source->cred); - error = -ENOMEM; - fattr_fsinfo = nfs_alloc_fattr(); - if (fattr_fsinfo == NULL) - goto out_free_server; - /* Copy data from the source */ server->nfs_client = source->nfs_client; server->destroy = source->destroy; @@ -1111,7 +1105,7 @@ struct nfs_server *nfs_clone_server(struct nfs_server *source, goto out_free_server; /* probe the filesystem info for this server filesystem */ - error = nfs_probe_fsinfo(server, fh, fattr_fsinfo); + error = nfs_probe_server(server, fh); if (error < 0) goto out_free_server; @@ -1125,11 +1119,9 @@ struct nfs_server *nfs_clone_server(struct nfs_server *source, nfs_server_insert_lists(server); server->mount_time = jiffies; - nfs_free_fattr(fattr_fsinfo); return server; out_free_server: - nfs_free_fattr(fattr_fsinfo); nfs_free_server(server); return ERR_PTR(error); } diff --git a/fs/nfs/nfs4client.c b/fs/nfs/nfs4client.c index 85978ecb727e..d8b5a250ca05 100644 --- a/fs/nfs/nfs4client.c +++ b/fs/nfs/nfs4client.c @@ -1080,17 +1080,12 @@ void nfs4_server_set_init_caps(struct nfs_server *server) static int nfs4_server_common_setup(struct nfs_server *server, struct nfs_fh *mntfh, bool auth_probe) { - struct nfs_fattr *fattr; int error; /* data servers support only a subset of NFSv4.1 */ if (is_ds_only_client(server->nfs_client)) return -EPROTONOSUPPORT; - fattr = nfs_alloc_fattr(); - if (fattr == NULL) - return -ENOMEM; - /* We must ensure the session is initialised first */ error = nfs4_init_session(server->nfs_client); if (error < 0) @@ -1108,7 +1103,7 @@ static int nfs4_server_common_setup(struct nfs_server *server, (unsigned long long) server->fsid.minor); nfs_display_fhandle(mntfh, "Pseudo-fs root FH"); - error = nfs_probe_fsinfo(server, mntfh, fattr); + error = nfs_probe_server(server, mntfh); if (error < 0) goto out; @@ -1122,7 +1117,6 @@ static int nfs4_server_common_setup(struct nfs_server *server, server->mount_time = jiffies; server->destroy = nfs4_destroy_server; out: - nfs_free_fattr(fattr); return error; } From 1301ba603ca519f9191eae68ee59a2d1165b3458 Mon Sep 17 00:00:00 2001 From: Anna Schumaker Date: Thu, 14 Oct 2021 13:55:07 -0400 Subject: [PATCH 40/84] NFS: Call nfs_probe_server() during a fscontext-reconfigure event This lets us update the server's attributes when the user does a "mount -o remount" on the filesystem. Signed-off-by: Anna Schumaker Signed-off-by: Trond Myklebust --- fs/nfs/super.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/fs/nfs/super.c b/fs/nfs/super.c index e65c83494c05..3aced401735c 100644 --- a/fs/nfs/super.c +++ b/fs/nfs/super.c @@ -1004,6 +1004,7 @@ int nfs_reconfigure(struct fs_context *fc) struct nfs_fs_context *ctx = nfs_fc2context(fc); struct super_block *sb = fc->root->d_sb; struct nfs_server *nfss = sb->s_fs_info; + int ret; sync_filesystem(sb); @@ -1028,7 +1029,11 @@ int nfs_reconfigure(struct fs_context *fc) } /* compare new mount options with old ones */ - return nfs_compare_remount_data(nfss, ctx); + ret = nfs_compare_remount_data(nfss, ctx); + if (ret) + return ret; + + return nfs_probe_server(nfss, NFS_FH(d_inode(fc->root))); } EXPORT_SYMBOL_GPL(nfs_reconfigure); From 5fe1210d259542f966bab130830ece08e97f68f5 Mon Sep 17 00:00:00 2001 From: Anna Schumaker Date: Thu, 14 Oct 2021 13:55:08 -0400 Subject: [PATCH 41/84] NFS: Unexport nfs_probe_fsinfo() All the callers are now in client.c so we can remove the EXPORT_SYMBOL_GPL() and make it static. Signed-off-by: Anna Schumaker Signed-off-by: Trond Myklebust --- fs/nfs/client.c | 3 +-- fs/nfs/internal.h | 1 - 2 files changed, 1 insertion(+), 3 deletions(-) diff --git a/fs/nfs/client.c b/fs/nfs/client.c index 1a882c78a794..960b9d87648e 100644 --- a/fs/nfs/client.c +++ b/fs/nfs/client.c @@ -828,7 +828,7 @@ static void nfs_server_set_fsinfo(struct nfs_server *server, /* * Probe filesystem information, including the FSID on v2/v3 */ -int nfs_probe_fsinfo(struct nfs_server *server, struct nfs_fh *mntfh, struct nfs_fattr *fattr) +static int nfs_probe_fsinfo(struct nfs_server *server, struct nfs_fh *mntfh, struct nfs_fattr *fattr) { struct nfs_fsinfo fsinfo; struct nfs_client *clp = server->nfs_client; @@ -862,7 +862,6 @@ int nfs_probe_fsinfo(struct nfs_server *server, struct nfs_fh *mntfh, struct nfs return 0; } -EXPORT_SYMBOL_GPL(nfs_probe_fsinfo); /* * Grab the destination's particulars, including lease expiry time. diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h index c1253dd60c32..123078c76495 100644 --- a/fs/nfs/internal.h +++ b/fs/nfs/internal.h @@ -193,7 +193,6 @@ extern void nfs_clients_exit(struct net *net); extern struct nfs_client *nfs_alloc_client(const struct nfs_client_initdata *); int nfs_create_rpc_client(struct nfs_client *, const struct nfs_client_initdata *, rpc_authflavor_t); struct nfs_client *nfs_get_client(const struct nfs_client_initdata *); -int nfs_probe_fsinfo(struct nfs_server *server, struct nfs_fh *, struct nfs_fattr *); int nfs_probe_server(struct nfs_server *, struct nfs_fh *); void nfs_server_insert_lists(struct nfs_server *); void nfs_server_remove_lists(struct nfs_server *); From 023859ce6f88f7cfc223752fb56ec453a147b852 Mon Sep 17 00:00:00 2001 From: Thiago Rafael Becker Date: Wed, 20 Oct 2021 18:04:28 -0300 Subject: [PATCH 42/84] sunrpc: remove unnecessary test in rpc_task_set_client() In rpc_task_set_client(), testing for a NULL clnt is not necessary, as clnt should always be a valid pointer to a rpc_client. Signed-off-by: Thiago Rafael Becker Signed-off-by: Trond Myklebust --- net/sunrpc/clnt.c | 33 +++++++++++++++------------------ 1 file changed, 15 insertions(+), 18 deletions(-) diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c index f056ff931444..a312ea2bc440 100644 --- a/net/sunrpc/clnt.c +++ b/net/sunrpc/clnt.c @@ -1076,24 +1076,21 @@ void rpc_task_set_transport(struct rpc_task *task, struct rpc_clnt *clnt) static void rpc_task_set_client(struct rpc_task *task, struct rpc_clnt *clnt) { - - if (clnt != NULL) { - rpc_task_set_transport(task, clnt); - task->tk_client = clnt; - refcount_inc(&clnt->cl_count); - if (clnt->cl_softrtry) - task->tk_flags |= RPC_TASK_SOFT; - if (clnt->cl_softerr) - task->tk_flags |= RPC_TASK_TIMEOUT; - if (clnt->cl_noretranstimeo) - task->tk_flags |= RPC_TASK_NO_RETRANS_TIMEOUT; - if (atomic_read(&clnt->cl_swapper)) - task->tk_flags |= RPC_TASK_SWAPPER; - /* Add to the client's list of all tasks */ - spin_lock(&clnt->cl_lock); - list_add_tail(&task->tk_task, &clnt->cl_tasks); - spin_unlock(&clnt->cl_lock); - } + rpc_task_set_transport(task, clnt); + task->tk_client = clnt; + refcount_inc(&clnt->cl_count); + if (clnt->cl_softrtry) + task->tk_flags |= RPC_TASK_SOFT; + if (clnt->cl_softerr) + task->tk_flags |= RPC_TASK_TIMEOUT; + if (clnt->cl_noretranstimeo) + task->tk_flags |= RPC_TASK_NO_RETRANS_TIMEOUT; + if (atomic_read(&clnt->cl_swapper)) + task->tk_flags |= RPC_TASK_SWAPPER; + /* Add to the client's list of all tasks */ + spin_lock(&clnt->cl_lock); + list_add_tail(&task->tk_task, &clnt->cl_tasks); + spin_unlock(&clnt->cl_lock); } static void From 4cd27df88af29929cda6e8eb4e0f5bb4e25812bf Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Thu, 21 Oct 2021 17:11:37 -0400 Subject: [PATCH 43/84] NFS: Remove redundant call to __set_page_dirty_nobuffers Remove a redundant call in nfs_updatepage(). nfs_writepage_setup() will have already called nfs_mark_request_dirty() on success. Signed-off-by: Trond Myklebust --- fs/nfs/write.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/fs/nfs/write.c b/fs/nfs/write.c index 82c5b89395f6..9b7619ce17a7 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -1384,8 +1384,6 @@ int nfs_updatepage(struct file *file, struct page *page, status = nfs_writepage_setup(ctx, page, offset, count); if (status < 0) nfs_set_pageerror(mapping); - else - __set_page_dirty_nobuffers(page); out: dprintk("NFS: nfs_updatepage returns %d (isize %lld)\n", status, (long long)i_size_read(inode)); From 01d29f87fcfef38d51ce2b473981a5c1e861ac0a Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Tue, 26 Oct 2021 21:56:40 -0400 Subject: [PATCH 44/84] NFSv4: Fix a regression in nfs_set_open_stateid_locked() If we already hold open state on the client, yet the server gives us a completely different stateid to the one we already hold, then we currently treat it as if it were an out-of-sequence update, and wait for 5 seconds for other updates to come in. This commit fixes the behaviour so that we immediately start processing of the new stateid, and then leave it to the call to nfs4_test_and_free_stateid() to decide what to do with the old stateid. Fixes: b4868b44c562 ("NFSv4: Wait for stateid updates after CLOSE/OPEN_DOWNGRADE") Signed-off-by: Trond Myklebust --- fs/nfs/nfs4proc.c | 15 ++++++++------- 1 file changed, 8 insertions(+), 7 deletions(-) diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 1c485edf1d07..1c94f54cab58 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -1604,15 +1604,16 @@ static bool nfs_stateid_is_sequential(struct nfs4_state *state, { if (test_bit(NFS_OPEN_STATE, &state->flags)) { /* The common case - we're updating to a new sequence number */ - if (nfs4_stateid_match_other(stateid, &state->open_stateid) && - nfs4_stateid_is_next(&state->open_stateid, stateid)) { - return true; + if (nfs4_stateid_match_other(stateid, &state->open_stateid)) { + if (nfs4_stateid_is_next(&state->open_stateid, stateid)) + return true; + return false; } - } else { - /* This is the first OPEN in this generation */ - if (stateid->seqid == cpu_to_be32(1)) - return true; + /* The server returned a new stateid */ } + /* This is the first OPEN in this generation */ + if (stateid->seqid == cpu_to_be32(1)) + return true; return false; } From ea9afca88bbea26f23697b3305789f77f0341d23 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Fri, 29 Oct 2021 11:02:20 -0400 Subject: [PATCH 45/84] SUNRPC: Replace use of socket sk_callback_lock with sock_lock Since we do things like setting flags, etc it really is more appropriate to use sock_lock(). Signed-off-by: Trond Myklebust --- net/sunrpc/xprtsock.c | 38 +++++++++++--------------------------- 1 file changed, 11 insertions(+), 27 deletions(-) diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index 04f1b78bcbca..1c42153025dd 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -1153,14 +1153,13 @@ static void xs_error_report(struct sock *sk) struct sock_xprt *transport; struct rpc_xprt *xprt; - read_lock_bh(&sk->sk_callback_lock); if (!(xprt = xprt_from_sock(sk))) - goto out; + return; transport = container_of(xprt, struct sock_xprt, xprt); transport->xprt_err = -sk->sk_err; if (transport->xprt_err == 0) - goto out; + return; dprintk("RPC: xs_error_report client %p, error=%d...\n", xprt, -transport->xprt_err); trace_rpc_socket_error(xprt, sk->sk_socket, transport->xprt_err); @@ -1168,8 +1167,6 @@ static void xs_error_report(struct sock *sk) /* barrier ensures xprt_err is set before XPRT_SOCK_WAKE_ERROR */ smp_mb__before_atomic(); xs_run_error_worker(transport, XPRT_SOCK_WAKE_ERROR); - out: - read_unlock_bh(&sk->sk_callback_lock); } static void xs_reset_transport(struct sock_xprt *transport) @@ -1188,7 +1185,7 @@ static void xs_reset_transport(struct sock_xprt *transport) kernel_sock_shutdown(sock, SHUT_RDWR); mutex_lock(&transport->recv_mutex); - write_lock_bh(&sk->sk_callback_lock); + lock_sock(sk); transport->inet = NULL; transport->sock = NULL; transport->file = NULL; @@ -1197,10 +1194,10 @@ static void xs_reset_transport(struct sock_xprt *transport) xs_restore_old_callbacks(transport, sk); xprt_clear_connected(xprt); - write_unlock_bh(&sk->sk_callback_lock); xs_sock_reset_connection_flags(xprt); /* Reset stream record info */ xs_stream_reset_connect(transport); + release_sock(sk); mutex_unlock(&transport->recv_mutex); trace_rpc_socket_close(xprt, sock); @@ -1364,7 +1361,6 @@ static void xs_data_ready(struct sock *sk) { struct rpc_xprt *xprt; - read_lock_bh(&sk->sk_callback_lock); dprintk("RPC: xs_data_ready...\n"); xprt = xprt_from_sock(sk); if (xprt != NULL) { @@ -1379,7 +1375,6 @@ static void xs_data_ready(struct sock *sk) if (!test_and_set_bit(XPRT_SOCK_DATA_READY, &transport->sock_state)) queue_work(xprtiod_workqueue, &transport->recv_worker); } - read_unlock_bh(&sk->sk_callback_lock); } /* @@ -1408,9 +1403,8 @@ static void xs_tcp_state_change(struct sock *sk) struct rpc_xprt *xprt; struct sock_xprt *transport; - read_lock_bh(&sk->sk_callback_lock); if (!(xprt = xprt_from_sock(sk))) - goto out; + return; dprintk("RPC: xs_tcp_state_change client %p...\n", xprt); dprintk("RPC: state %x conn %d dead %d zapped %d sk_shutdown %d\n", sk->sk_state, xprt_connected(xprt), @@ -1471,8 +1465,6 @@ static void xs_tcp_state_change(struct sock *sk) /* Trigger the socket release */ xs_run_error_worker(transport, XPRT_SOCK_WAKE_DISCONNECT); } - out: - read_unlock_bh(&sk->sk_callback_lock); } static void xs_write_space(struct sock *sk) @@ -1511,13 +1503,9 @@ out: */ static void xs_udp_write_space(struct sock *sk) { - read_lock_bh(&sk->sk_callback_lock); - /* from net/core/sock.c:sock_def_write_space */ if (sock_writeable(sk)) xs_write_space(sk); - - read_unlock_bh(&sk->sk_callback_lock); } /** @@ -1532,13 +1520,9 @@ static void xs_udp_write_space(struct sock *sk) */ static void xs_tcp_write_space(struct sock *sk) { - read_lock_bh(&sk->sk_callback_lock); - /* from net/core/stream.c:sk_stream_write_space */ if (sk_stream_is_writeable(sk)) xs_write_space(sk); - - read_unlock_bh(&sk->sk_callback_lock); } static void xs_udp_do_set_buffer_size(struct rpc_xprt *xprt) @@ -1833,7 +1817,7 @@ static int xs_local_finish_connecting(struct rpc_xprt *xprt, if (!transport->inet) { struct sock *sk = sock->sk; - write_lock_bh(&sk->sk_callback_lock); + lock_sock(sk); xs_save_old_callbacks(transport, sk); @@ -1849,7 +1833,7 @@ static int xs_local_finish_connecting(struct rpc_xprt *xprt, transport->sock = sock; transport->inet = sk; - write_unlock_bh(&sk->sk_callback_lock); + release_sock(sk); } xs_stream_start_connect(transport); @@ -2031,7 +2015,7 @@ static void xs_udp_finish_connecting(struct rpc_xprt *xprt, struct socket *sock) if (!transport->inet) { struct sock *sk = sock->sk; - write_lock_bh(&sk->sk_callback_lock); + lock_sock(sk); xs_save_old_callbacks(transport, sk); @@ -2048,7 +2032,7 @@ static void xs_udp_finish_connecting(struct rpc_xprt *xprt, struct socket *sock) xs_set_memalloc(xprt); - write_unlock_bh(&sk->sk_callback_lock); + release_sock(sk); } xs_udp_do_set_buffer_size(xprt); @@ -2194,7 +2178,7 @@ static int xs_tcp_finish_connecting(struct rpc_xprt *xprt, struct socket *sock) xs_tcp_set_socket_timeouts(xprt, sock); tcp_sock_set_nodelay(sk); - write_lock_bh(&sk->sk_callback_lock); + lock_sock(sk); xs_save_old_callbacks(transport, sk); @@ -2214,7 +2198,7 @@ static int xs_tcp_finish_connecting(struct rpc_xprt *xprt, struct socket *sock) transport->sock = sock; transport->inet = sk; - write_unlock_bh(&sk->sk_callback_lock); + release_sock(sk); } if (!xprt_bound(xprt)) From 280254b605ffb6ec88f33b43a360aa6b5247bef7 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Fri, 29 Oct 2021 12:05:48 -0400 Subject: [PATCH 46/84] SUNRPC: Clean up xs_tcp_setup_sock() Move the error handling into a single switch statement for clarity. Signed-off-by: Trond Myklebust --- net/sunrpc/xprtsock.c | 68 ++++++++++++++++++------------------------- 1 file changed, 28 insertions(+), 40 deletions(-) diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index 1c42153025dd..aa293e4a77fa 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -2158,7 +2158,6 @@ static void xs_tcp_set_connect_timeout(struct rpc_xprt *xprt, static int xs_tcp_finish_connecting(struct rpc_xprt *xprt, struct socket *sock) { struct sock_xprt *transport = container_of(xprt, struct sock_xprt, xprt); - int ret = -ENOTCONN; if (!transport->inet) { struct sock *sk = sock->sk; @@ -2202,7 +2201,7 @@ static int xs_tcp_finish_connecting(struct rpc_xprt *xprt, struct socket *sock) } if (!xprt_bound(xprt)) - goto out; + return -ENOTCONN; xs_set_memalloc(xprt); @@ -2210,22 +2209,7 @@ static int xs_tcp_finish_connecting(struct rpc_xprt *xprt, struct socket *sock) /* Tell the socket layer to start connecting... */ set_bit(XPRT_SOCK_CONNECTING, &transport->sock_state); - ret = kernel_connect(sock, xs_addr(xprt), xprt->addrlen, O_NONBLOCK); - switch (ret) { - case 0: - xs_set_srcport(transport, sock); - fallthrough; - case -EINPROGRESS: - /* SYN_SENT! */ - if (xprt->reestablish_timeout < XS_TCP_INIT_REEST_TO) - xprt->reestablish_timeout = XS_TCP_INIT_REEST_TO; - break; - case -EADDRNOTAVAIL: - /* Source port number is unavailable. Try a new one! */ - transport->srcport = 0; - } -out: - return ret; + return kernel_connect(sock, xs_addr(xprt), xprt->addrlen, O_NONBLOCK); } /** @@ -2240,14 +2224,14 @@ static void xs_tcp_setup_socket(struct work_struct *work) container_of(work, struct sock_xprt, connect_worker.work); struct socket *sock = transport->sock; struct rpc_xprt *xprt = &transport->xprt; - int status = -EIO; + int status; if (!sock) { sock = xs_create_sock(xprt, transport, xs_addr(xprt)->sa_family, SOCK_STREAM, IPPROTO_TCP, true); if (IS_ERR(sock)) { - status = PTR_ERR(sock); + xprt_wake_pending_tasks(xprt, PTR_ERR(sock)); goto out; } } @@ -2264,21 +2248,21 @@ static void xs_tcp_setup_socket(struct work_struct *work) xprt, -status, xprt_connected(xprt), sock->sk->sk_state); switch (status) { - default: - printk("%s: connect returned unhandled error %d\n", - __func__, status); - fallthrough; - case -EADDRNOTAVAIL: - /* We're probably in TIME_WAIT. Get rid of existing socket, - * and retry - */ - xs_tcp_force_close(xprt); - break; case 0: + xs_set_srcport(transport, sock); + fallthrough; case -EINPROGRESS: + /* SYN_SENT! */ + if (xprt->reestablish_timeout < XS_TCP_INIT_REEST_TO) + xprt->reestablish_timeout = XS_TCP_INIT_REEST_TO; + fallthrough; case -EALREADY: - xprt_unlock_connect(xprt, transport); - return; + goto out_unlock; + case -EADDRNOTAVAIL: + /* Source port number is unavailable. Try a new one! */ + transport->srcport = 0; + status = -EAGAIN; + break; case -EINVAL: /* Happens, for instance, if the user specified a link * local IPv6 address without a scope-id. @@ -2290,18 +2274,22 @@ static void xs_tcp_setup_socket(struct work_struct *work) case -EHOSTUNREACH: case -EADDRINUSE: case -ENOBUFS: - /* xs_tcp_force_close() wakes tasks with a fixed error code. - * We need to wake them first to ensure the correct error code. - */ - xprt_wake_pending_tasks(xprt, status); - xs_tcp_force_close(xprt); - goto out; + break; + default: + printk("%s: connect returned unhandled error %d\n", + __func__, status); + status = -EAGAIN; } - status = -EAGAIN; + + /* xs_tcp_force_close() wakes tasks with a fixed error code. + * We need to wake them first to ensure the correct error code. + */ + xprt_wake_pending_tasks(xprt, status); + xs_tcp_force_close(xprt); out: xprt_clear_connecting(xprt); +out_unlock: xprt_unlock_connect(xprt, transport); - xprt_wake_pending_tasks(xprt, status); } /** From 9d2d48bbbdabf7b2f029369c4f926d133c1d47ad Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Fri, 22 Oct 2021 16:16:56 -0400 Subject: [PATCH 47/84] NFS: Move generic FS show macros to global header Refactor: Surface useful show_ macros for use by other trace subsystems. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- fs/nfs/nfs4trace.h | 67 ++++++--------------- fs/nfs/nfstrace.h | 80 +++++-------------------- include/trace/events/fs.h | 122 ++++++++++++++++++++++++++++++++++++++ 3 files changed, 156 insertions(+), 113 deletions(-) create mode 100644 include/trace/events/fs.h diff --git a/fs/nfs/nfs4trace.h b/fs/nfs/nfs4trace.h index d4f061046c09..424d9cd4c196 100644 --- a/fs/nfs/nfs4trace.h +++ b/fs/nfs/nfs4trace.h @@ -11,6 +11,8 @@ #include #include +#include + TRACE_DEFINE_ENUM(EPERM); TRACE_DEFINE_ENUM(ENOENT); TRACE_DEFINE_ENUM(EIO); @@ -314,19 +316,6 @@ TRACE_DEFINE_ENUM(NFS4ERR_RESET_TO_PNFS); { NFS4ERR_RESET_TO_MDS, "RESET_TO_MDS" }, \ { NFS4ERR_RESET_TO_PNFS, "RESET_TO_PNFS" }) -#define show_open_flags(flags) \ - __print_flags(flags, "|", \ - { O_CREAT, "O_CREAT" }, \ - { O_EXCL, "O_EXCL" }, \ - { O_TRUNC, "O_TRUNC" }, \ - { O_DIRECT, "O_DIRECT" }) - -#define show_fmode_flags(mode) \ - __print_flags(mode, "|", \ - { ((__force unsigned long)FMODE_READ), "READ" }, \ - { ((__force unsigned long)FMODE_WRITE), "WRITE" }, \ - { ((__force unsigned long)FMODE_EXEC), "EXEC" }) - #define show_nfs_fattr_flags(valid) \ __print_flags((unsigned long)valid, "|", \ { NFS_ATTR_FATTR_TYPE, "TYPE" }, \ @@ -794,8 +783,8 @@ DECLARE_EVENT_CLASS(nfs4_open_event, TP_STRUCT__entry( __field(unsigned long, error) - __field(unsigned int, flags) - __field(unsigned int, fmode) + __field(unsigned long, flags) + __field(unsigned long, fmode) __field(dev_t, dev) __field(u32, fhandle) __field(u64, fileid) @@ -813,7 +802,7 @@ DECLARE_EVENT_CLASS(nfs4_open_event, __entry->error = -error; __entry->flags = flags; - __entry->fmode = (__force unsigned int)ctx->mode; + __entry->fmode = (__force unsigned long)ctx->mode; __entry->dev = ctx->dentry->d_sb->s_dev; if (!IS_ERR_OR_NULL(state)) { inode = state->inode; @@ -843,15 +832,15 @@ DECLARE_EVENT_CLASS(nfs4_open_event, ), TP_printk( - "error=%ld (%s) flags=%d (%s) fmode=%s " + "error=%ld (%s) flags=%lu (%s) fmode=%s " "fileid=%02x:%02x:%llu fhandle=0x%08x " "name=%02x:%02x:%llu/%s stateid=%d:0x%08x " "openstateid=%d:0x%08x", -__entry->error, show_nfsv4_errors(__entry->error), __entry->flags, - show_open_flags(__entry->flags), - show_fmode_flags(__entry->fmode), + show_fs_fcntl_open_flags(__entry->flags), + show_fs_fmode_flags(__entry->fmode), MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->fileid, __entry->fhandle, @@ -905,7 +894,7 @@ TRACE_EVENT(nfs4_cached_open, TP_printk( "fmode=%s fileid=%02x:%02x:%llu " "fhandle=0x%08x stateid=%d:0x%08x", - __entry->fmode ? show_fmode_flags(__entry->fmode) : + __entry->fmode ? show_fs_fmode_flags(__entry->fmode) : "closed", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->fileid, @@ -953,7 +942,7 @@ TRACE_EVENT(nfs4_close, "fhandle=0x%08x openstateid=%d:0x%08x", -__entry->error, show_nfsv4_errors(__entry->error), - __entry->fmode ? show_fmode_flags(__entry->fmode) : + __entry->fmode ? show_fs_fmode_flags(__entry->fmode) : "closed", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->fileid, @@ -962,24 +951,6 @@ TRACE_EVENT(nfs4_close, ) ); -TRACE_DEFINE_ENUM(F_GETLK); -TRACE_DEFINE_ENUM(F_SETLK); -TRACE_DEFINE_ENUM(F_SETLKW); -TRACE_DEFINE_ENUM(F_RDLCK); -TRACE_DEFINE_ENUM(F_WRLCK); -TRACE_DEFINE_ENUM(F_UNLCK); - -#define show_lock_cmd(type) \ - __print_symbolic((int)type, \ - { F_GETLK, "GETLK" }, \ - { F_SETLK, "SETLK" }, \ - { F_SETLKW, "SETLKW" }) -#define show_lock_type(type) \ - __print_symbolic((int)type, \ - { F_RDLCK, "RDLCK" }, \ - { F_WRLCK, "WRLCK" }, \ - { F_UNLCK, "UNLCK" }) - DECLARE_EVENT_CLASS(nfs4_lock_event, TP_PROTO( const struct file_lock *request, @@ -992,8 +963,8 @@ DECLARE_EVENT_CLASS(nfs4_lock_event, TP_STRUCT__entry( __field(unsigned long, error) - __field(int, cmd) - __field(char, type) + __field(unsigned long, cmd) + __field(unsigned long, type) __field(loff_t, start) __field(loff_t, end) __field(dev_t, dev) @@ -1026,8 +997,8 @@ DECLARE_EVENT_CLASS(nfs4_lock_event, "stateid=%d:0x%08x", -__entry->error, show_nfsv4_errors(__entry->error), - show_lock_cmd(__entry->cmd), - show_lock_type(__entry->type), + show_fs_fcntl_cmd(__entry->cmd), + show_fs_fcntl_lock_type(__entry->type), (long long)__entry->start, (long long)__entry->end, MAJOR(__entry->dev), MINOR(__entry->dev), @@ -1062,8 +1033,8 @@ TRACE_EVENT(nfs4_set_lock, TP_STRUCT__entry( __field(unsigned long, error) - __field(int, cmd) - __field(char, type) + __field(unsigned long, cmd) + __field(unsigned long, type) __field(loff_t, start) __field(loff_t, end) __field(dev_t, dev) @@ -1102,8 +1073,8 @@ TRACE_EVENT(nfs4_set_lock, "stateid=%d:0x%08x lockstateid=%d:0x%08x", -__entry->error, show_nfsv4_errors(__entry->error), - show_lock_cmd(__entry->cmd), - show_lock_type(__entry->type), + show_fs_fcntl_cmd(__entry->cmd), + show_fs_fcntl_lock_type(__entry->type), (long long)__entry->start, (long long)__entry->end, MAJOR(__entry->dev), MINOR(__entry->dev), @@ -1220,7 +1191,7 @@ DECLARE_EVENT_CLASS(nfs4_set_delegation_event, TP_printk( "fmode=%s fileid=%02x:%02x:%llu fhandle=0x%08x", - show_fmode_flags(__entry->fmode), + show_fs_fmode_flags(__entry->fmode), MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->fileid, __entry->fhandle diff --git a/fs/nfs/nfstrace.h b/fs/nfs/nfstrace.h index 83e9615c8b8c..331bcc0c0a75 100644 --- a/fs/nfs/nfstrace.h +++ b/fs/nfs/nfstrace.h @@ -11,20 +11,9 @@ #include #include +#include #include -#define nfs_show_file_type(ftype) \ - __print_symbolic(ftype, \ - { DT_UNKNOWN, "UNKNOWN" }, \ - { DT_FIFO, "FIFO" }, \ - { DT_CHR, "CHR" }, \ - { DT_DIR, "DIR" }, \ - { DT_BLK, "BLK" }, \ - { DT_REG, "REG" }, \ - { DT_LNK, "LNK" }, \ - { DT_SOCK, "SOCK" }, \ - { DT_WHT, "WHT" }) - #define nfs_show_cache_validity(v) \ __print_flags(v, "|", \ { NFS_INO_INVALID_DATA, "INVALID_DATA" }, \ @@ -131,7 +120,7 @@ DECLARE_EVENT_CLASS(nfs_inode_event_done, (unsigned long long)__entry->fileid, __entry->fhandle, __entry->type, - nfs_show_file_type(__entry->type), + show_fs_dirent_type(__entry->type), (unsigned long long)__entry->version, (long long)__entry->size, __entry->cache_validity, @@ -222,7 +211,7 @@ TRACE_EVENT(nfs_access_exit, (unsigned long long)__entry->fileid, __entry->fhandle, __entry->type, - nfs_show_file_type(__entry->type), + show_fs_dirent_type(__entry->type), (unsigned long long)__entry->version, (long long)__entry->size, __entry->cache_validity, @@ -283,21 +272,6 @@ DEFINE_NFS_UPDATE_SIZE_EVENT(wcc); DEFINE_NFS_UPDATE_SIZE_EVENT(update); DEFINE_NFS_UPDATE_SIZE_EVENT(grow); -#define show_lookup_flags(flags) \ - __print_flags(flags, "|", \ - { LOOKUP_FOLLOW, "FOLLOW" }, \ - { LOOKUP_DIRECTORY, "DIRECTORY" }, \ - { LOOKUP_AUTOMOUNT, "AUTOMOUNT" }, \ - { LOOKUP_PARENT, "PARENT" }, \ - { LOOKUP_REVAL, "REVAL" }, \ - { LOOKUP_RCU, "RCU" }, \ - { LOOKUP_OPEN, "OPEN" }, \ - { LOOKUP_CREATE, "CREATE" }, \ - { LOOKUP_EXCL, "EXCL" }, \ - { LOOKUP_RENAME_TARGET, "RENAME_TARGET" }, \ - { LOOKUP_EMPTY, "EMPTY" }, \ - { LOOKUP_DOWN, "DOWN" }) - DECLARE_EVENT_CLASS(nfs_lookup_event, TP_PROTO( const struct inode *dir, @@ -324,7 +298,7 @@ DECLARE_EVENT_CLASS(nfs_lookup_event, TP_printk( "flags=0x%lx (%s) name=%02x:%02x:%llu/%s", __entry->flags, - show_lookup_flags(__entry->flags), + show_fs_lookup_flags(__entry->flags), MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->dir, __get_str(name) @@ -370,7 +344,7 @@ DECLARE_EVENT_CLASS(nfs_lookup_event_done, "error=%ld (%s) flags=0x%lx (%s) name=%02x:%02x:%llu/%s", -__entry->error, nfs_show_status(__entry->error), __entry->flags, - show_lookup_flags(__entry->flags), + show_fs_lookup_flags(__entry->flags), MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->dir, __get_str(name) @@ -392,30 +366,6 @@ DEFINE_NFS_LOOKUP_EVENT_DONE(nfs_lookup_exit); DEFINE_NFS_LOOKUP_EVENT(nfs_lookup_revalidate_enter); DEFINE_NFS_LOOKUP_EVENT_DONE(nfs_lookup_revalidate_exit); -#define show_open_flags(flags) \ - __print_flags(flags, "|", \ - { O_WRONLY, "O_WRONLY" }, \ - { O_RDWR, "O_RDWR" }, \ - { O_CREAT, "O_CREAT" }, \ - { O_EXCL, "O_EXCL" }, \ - { O_NOCTTY, "O_NOCTTY" }, \ - { O_TRUNC, "O_TRUNC" }, \ - { O_APPEND, "O_APPEND" }, \ - { O_NONBLOCK, "O_NONBLOCK" }, \ - { O_DSYNC, "O_DSYNC" }, \ - { O_DIRECT, "O_DIRECT" }, \ - { O_LARGEFILE, "O_LARGEFILE" }, \ - { O_DIRECTORY, "O_DIRECTORY" }, \ - { O_NOFOLLOW, "O_NOFOLLOW" }, \ - { O_NOATIME, "O_NOATIME" }, \ - { O_CLOEXEC, "O_CLOEXEC" }) - -#define show_fmode_flags(mode) \ - __print_flags(mode, "|", \ - { ((__force unsigned long)FMODE_READ), "READ" }, \ - { ((__force unsigned long)FMODE_WRITE), "WRITE" }, \ - { ((__force unsigned long)FMODE_EXEC), "EXEC" }) - TRACE_EVENT(nfs_atomic_open_enter, TP_PROTO( const struct inode *dir, @@ -427,7 +377,7 @@ TRACE_EVENT(nfs_atomic_open_enter, TP_STRUCT__entry( __field(unsigned long, flags) - __field(unsigned int, fmode) + __field(unsigned long, fmode) __field(dev_t, dev) __field(u64, dir) __string(name, ctx->dentry->d_name.name) @@ -437,15 +387,15 @@ TRACE_EVENT(nfs_atomic_open_enter, __entry->dev = dir->i_sb->s_dev; __entry->dir = NFS_FILEID(dir); __entry->flags = flags; - __entry->fmode = (__force unsigned int)ctx->mode; + __entry->fmode = (__force unsigned long)ctx->mode; __assign_str(name, ctx->dentry->d_name.name); ), TP_printk( "flags=0x%lx (%s) fmode=%s name=%02x:%02x:%llu/%s", __entry->flags, - show_open_flags(__entry->flags), - show_fmode_flags(__entry->fmode), + show_fs_fcntl_open_flags(__entry->flags), + show_fs_fmode_flags(__entry->fmode), MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->dir, __get_str(name) @@ -465,7 +415,7 @@ TRACE_EVENT(nfs_atomic_open_exit, TP_STRUCT__entry( __field(unsigned long, error) __field(unsigned long, flags) - __field(unsigned int, fmode) + __field(unsigned long, fmode) __field(dev_t, dev) __field(u64, dir) __string(name, ctx->dentry->d_name.name) @@ -476,7 +426,7 @@ TRACE_EVENT(nfs_atomic_open_exit, __entry->dev = dir->i_sb->s_dev; __entry->dir = NFS_FILEID(dir); __entry->flags = flags; - __entry->fmode = (__force unsigned int)ctx->mode; + __entry->fmode = (__force unsigned long)ctx->mode; __assign_str(name, ctx->dentry->d_name.name); ), @@ -485,8 +435,8 @@ TRACE_EVENT(nfs_atomic_open_exit, "name=%02x:%02x:%llu/%s", -__entry->error, nfs_show_status(__entry->error), __entry->flags, - show_open_flags(__entry->flags), - show_fmode_flags(__entry->fmode), + show_fs_fcntl_open_flags(__entry->flags), + show_fs_fmode_flags(__entry->fmode), MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->dir, __get_str(name) @@ -519,7 +469,7 @@ TRACE_EVENT(nfs_create_enter, TP_printk( "flags=0x%lx (%s) name=%02x:%02x:%llu/%s", __entry->flags, - show_open_flags(__entry->flags), + show_fs_fcntl_open_flags(__entry->flags), MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->dir, __get_str(name) @@ -556,7 +506,7 @@ TRACE_EVENT(nfs_create_exit, "error=%ld (%s) flags=0x%lx (%s) name=%02x:%02x:%llu/%s", -__entry->error, nfs_show_status(__entry->error), __entry->flags, - show_open_flags(__entry->flags), + show_fs_fcntl_open_flags(__entry->flags), MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->dir, __get_str(name) diff --git a/include/trace/events/fs.h b/include/trace/events/fs.h new file mode 100644 index 000000000000..738b97f22f36 --- /dev/null +++ b/include/trace/events/fs.h @@ -0,0 +1,122 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Display helpers for generic filesystem items + * + * Author: Chuck Lever + * + * Copyright (c) 2020, Oracle and/or its affiliates. + */ + +#include + +#define show_fs_dirent_type(x) \ + __print_symbolic(x, \ + { DT_UNKNOWN, "UNKNOWN" }, \ + { DT_FIFO, "FIFO" }, \ + { DT_CHR, "CHR" }, \ + { DT_DIR, "DIR" }, \ + { DT_BLK, "BLK" }, \ + { DT_REG, "REG" }, \ + { DT_LNK, "LNK" }, \ + { DT_SOCK, "SOCK" }, \ + { DT_WHT, "WHT" }) + +#define show_fs_fcntl_open_flags(x) \ + __print_flags(x, "|", \ + { O_WRONLY, "O_WRONLY" }, \ + { O_RDWR, "O_RDWR" }, \ + { O_CREAT, "O_CREAT" }, \ + { O_EXCL, "O_EXCL" }, \ + { O_NOCTTY, "O_NOCTTY" }, \ + { O_TRUNC, "O_TRUNC" }, \ + { O_APPEND, "O_APPEND" }, \ + { O_NONBLOCK, "O_NONBLOCK" }, \ + { O_DSYNC, "O_DSYNC" }, \ + { O_DIRECT, "O_DIRECT" }, \ + { O_LARGEFILE, "O_LARGEFILE" }, \ + { O_DIRECTORY, "O_DIRECTORY" }, \ + { O_NOFOLLOW, "O_NOFOLLOW" }, \ + { O_NOATIME, "O_NOATIME" }, \ + { O_CLOEXEC, "O_CLOEXEC" }) + +#define __fmode_flag(x) { (__force unsigned long)FMODE_##x, #x } +#define show_fs_fmode_flags(x) \ + __print_flags(x, "|", \ + __fmode_flag(READ), \ + __fmode_flag(WRITE), \ + __fmode_flag(EXEC)) + +#ifdef CONFIG_64BIT +#define show_fs_fcntl_cmd(x) \ + __print_symbolic(x, \ + { F_DUPFD, "DUPFD" }, \ + { F_GETFD, "GETFD" }, \ + { F_SETFD, "SETFD" }, \ + { F_GETFL, "GETFL" }, \ + { F_SETFL, "SETFL" }, \ + { F_GETLK, "GETLK" }, \ + { F_SETLK, "SETLK" }, \ + { F_SETLKW, "SETLKW" }, \ + { F_SETOWN, "SETOWN" }, \ + { F_GETOWN, "GETOWN" }, \ + { F_SETSIG, "SETSIG" }, \ + { F_GETSIG, "GETSIG" }, \ + { F_SETOWN_EX, "SETOWN_EX" }, \ + { F_GETOWN_EX, "GETOWN_EX" }, \ + { F_GETOWNER_UIDS, "GETOWNER_UIDS" }, \ + { F_OFD_GETLK, "OFD_GETLK" }, \ + { F_OFD_SETLK, "OFD_SETLK" }, \ + { F_OFD_SETLKW, "OFD_SETLKW" }) +#else /* CONFIG_64BIT */ +#define show_fs_fcntl_cmd(x) \ + __print_symbolic(x, \ + { F_DUPFD, "DUPFD" }, \ + { F_GETFD, "GETFD" }, \ + { F_SETFD, "SETFD" }, \ + { F_GETFL, "GETFL" }, \ + { F_SETFL, "SETFL" }, \ + { F_GETLK, "GETLK" }, \ + { F_SETLK, "SETLK" }, \ + { F_SETLKW, "SETLKW" }, \ + { F_SETOWN, "SETOWN" }, \ + { F_GETOWN, "GETOWN" }, \ + { F_SETSIG, "SETSIG" }, \ + { F_GETSIG, "GETSIG" }, \ + { F_GETLK64, "GETLK64" }, \ + { F_SETLK64, "SETLK64" }, \ + { F_SETLKW64, "SETLKW64" }, \ + { F_SETOWN_EX, "SETOWN_EX" }, \ + { F_GETOWN_EX, "GETOWN_EX" }, \ + { F_GETOWNER_UIDS, "GETOWNER_UIDS" }, \ + { F_OFD_GETLK, "OFD_GETLK" }, \ + { F_OFD_SETLK, "OFD_SETLK" }, \ + { F_OFD_SETLKW, "OFD_SETLKW" }) +#endif /* CONFIG_64BIT */ + +#define show_fs_fcntl_lock_type(x) \ + __print_symbolic(x, \ + { F_RDLCK, "RDLCK" }, \ + { F_WRLCK, "WRLCK" }, \ + { F_UNLCK, "UNLCK" }) + +#define show_fs_lookup_flags(flags) \ + __print_flags(flags, "|", \ + { LOOKUP_FOLLOW, "FOLLOW" }, \ + { LOOKUP_DIRECTORY, "DIRECTORY" }, \ + { LOOKUP_AUTOMOUNT, "AUTOMOUNT" }, \ + { LOOKUP_EMPTY, "EMPTY" }, \ + { LOOKUP_DOWN, "DOWN" }, \ + { LOOKUP_MOUNTPOINT, "MOUNTPOINT" }, \ + { LOOKUP_REVAL, "REVAL" }, \ + { LOOKUP_RCU, "RCU" }, \ + { LOOKUP_OPEN, "OPEN" }, \ + { LOOKUP_CREATE, "CREATE" }, \ + { LOOKUP_EXCL, "EXCL" }, \ + { LOOKUP_RENAME_TARGET, "RENAME_TARGET" }, \ + { LOOKUP_PARENT, "PARENT" }, \ + { LOOKUP_NO_SYMLINKS, "NO_SYMLINKS" }, \ + { LOOKUP_NO_MAGICLINKS, "NO_MAGICLINKS" }, \ + { LOOKUP_NO_XDEV, "NO_XDEV" }, \ + { LOOKUP_BENEATH, "BENEATH" }, \ + { LOOKUP_IN_ROOT, "IN_ROOT" }, \ + { LOOKUP_CACHED, "CACHED" }) From 8791545eda52e8f3bc48e3cd902e38bf4ba4c9de Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Fri, 22 Oct 2021 16:17:03 -0400 Subject: [PATCH 48/84] NFS: Move NFS protocol display macros to global header Refactor: surface useful show_ macros so they can be shared between the client and server trace code. Additional clean up: - Housekeeping: ensure the correct #include files are pulled in and add proper TRACE_DEFINE_ENUM where they are missing - Use a consistent naming scheme for the helpers - Store values to be displayed symbolically as unsigned long, as that is the type that the __print_yada() functions take Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- fs/nfs/nfs4trace.h | 403 ++++--------------------------------- fs/nfs/nfstrace.h | 117 ++--------- fs/nfs/pnfs.h | 4 - fs/nfsd/trace.h | 1 + include/linux/nfs4.h | 4 + include/trace/events/nfs.h | 375 ++++++++++++++++++++++++++++++++++ 6 files changed, 433 insertions(+), 471 deletions(-) create mode 100644 include/trace/events/nfs.h diff --git a/fs/nfs/nfs4trace.h b/fs/nfs/nfs4trace.h index 424d9cd4c196..18f149f72160 100644 --- a/fs/nfs/nfs4trace.h +++ b/fs/nfs/nfs4trace.h @@ -12,309 +12,7 @@ #include #include - -TRACE_DEFINE_ENUM(EPERM); -TRACE_DEFINE_ENUM(ENOENT); -TRACE_DEFINE_ENUM(EIO); -TRACE_DEFINE_ENUM(ENXIO); -TRACE_DEFINE_ENUM(EACCES); -TRACE_DEFINE_ENUM(EEXIST); -TRACE_DEFINE_ENUM(EXDEV); -TRACE_DEFINE_ENUM(ENOTDIR); -TRACE_DEFINE_ENUM(EISDIR); -TRACE_DEFINE_ENUM(EFBIG); -TRACE_DEFINE_ENUM(ENOSPC); -TRACE_DEFINE_ENUM(EROFS); -TRACE_DEFINE_ENUM(EMLINK); -TRACE_DEFINE_ENUM(ENAMETOOLONG); -TRACE_DEFINE_ENUM(ENOTEMPTY); -TRACE_DEFINE_ENUM(EDQUOT); -TRACE_DEFINE_ENUM(ESTALE); -TRACE_DEFINE_ENUM(EBADHANDLE); -TRACE_DEFINE_ENUM(EBADCOOKIE); -TRACE_DEFINE_ENUM(ENOTSUPP); -TRACE_DEFINE_ENUM(ETOOSMALL); -TRACE_DEFINE_ENUM(EREMOTEIO); -TRACE_DEFINE_ENUM(EBADTYPE); -TRACE_DEFINE_ENUM(EAGAIN); -TRACE_DEFINE_ENUM(ELOOP); -TRACE_DEFINE_ENUM(EOPNOTSUPP); -TRACE_DEFINE_ENUM(EDEADLK); -TRACE_DEFINE_ENUM(ENOMEM); -TRACE_DEFINE_ENUM(EKEYEXPIRED); -TRACE_DEFINE_ENUM(ETIMEDOUT); -TRACE_DEFINE_ENUM(ERESTARTSYS); -TRACE_DEFINE_ENUM(ECONNREFUSED); -TRACE_DEFINE_ENUM(ECONNRESET); -TRACE_DEFINE_ENUM(ENETUNREACH); -TRACE_DEFINE_ENUM(EHOSTUNREACH); -TRACE_DEFINE_ENUM(EHOSTDOWN); -TRACE_DEFINE_ENUM(EPIPE); -TRACE_DEFINE_ENUM(EPFNOSUPPORT); -TRACE_DEFINE_ENUM(EPROTONOSUPPORT); - -TRACE_DEFINE_ENUM(NFS4_OK); -TRACE_DEFINE_ENUM(NFS4ERR_ACCESS); -TRACE_DEFINE_ENUM(NFS4ERR_ATTRNOTSUPP); -TRACE_DEFINE_ENUM(NFS4ERR_ADMIN_REVOKED); -TRACE_DEFINE_ENUM(NFS4ERR_BACK_CHAN_BUSY); -TRACE_DEFINE_ENUM(NFS4ERR_BADCHAR); -TRACE_DEFINE_ENUM(NFS4ERR_BADHANDLE); -TRACE_DEFINE_ENUM(NFS4ERR_BADIOMODE); -TRACE_DEFINE_ENUM(NFS4ERR_BADLAYOUT); -TRACE_DEFINE_ENUM(NFS4ERR_BADLABEL); -TRACE_DEFINE_ENUM(NFS4ERR_BADNAME); -TRACE_DEFINE_ENUM(NFS4ERR_BADOWNER); -TRACE_DEFINE_ENUM(NFS4ERR_BADSESSION); -TRACE_DEFINE_ENUM(NFS4ERR_BADSLOT); -TRACE_DEFINE_ENUM(NFS4ERR_BADTYPE); -TRACE_DEFINE_ENUM(NFS4ERR_BADXDR); -TRACE_DEFINE_ENUM(NFS4ERR_BAD_COOKIE); -TRACE_DEFINE_ENUM(NFS4ERR_BAD_HIGH_SLOT); -TRACE_DEFINE_ENUM(NFS4ERR_BAD_RANGE); -TRACE_DEFINE_ENUM(NFS4ERR_BAD_SEQID); -TRACE_DEFINE_ENUM(NFS4ERR_BAD_SESSION_DIGEST); -TRACE_DEFINE_ENUM(NFS4ERR_BAD_STATEID); -TRACE_DEFINE_ENUM(NFS4ERR_CB_PATH_DOWN); -TRACE_DEFINE_ENUM(NFS4ERR_CLID_INUSE); -TRACE_DEFINE_ENUM(NFS4ERR_CLIENTID_BUSY); -TRACE_DEFINE_ENUM(NFS4ERR_COMPLETE_ALREADY); -TRACE_DEFINE_ENUM(NFS4ERR_CONN_NOT_BOUND_TO_SESSION); -TRACE_DEFINE_ENUM(NFS4ERR_DEADLOCK); -TRACE_DEFINE_ENUM(NFS4ERR_DEADSESSION); -TRACE_DEFINE_ENUM(NFS4ERR_DELAY); -TRACE_DEFINE_ENUM(NFS4ERR_DELEG_ALREADY_WANTED); -TRACE_DEFINE_ENUM(NFS4ERR_DELEG_REVOKED); -TRACE_DEFINE_ENUM(NFS4ERR_DENIED); -TRACE_DEFINE_ENUM(NFS4ERR_DIRDELEG_UNAVAIL); -TRACE_DEFINE_ENUM(NFS4ERR_DQUOT); -TRACE_DEFINE_ENUM(NFS4ERR_ENCR_ALG_UNSUPP); -TRACE_DEFINE_ENUM(NFS4ERR_EXIST); -TRACE_DEFINE_ENUM(NFS4ERR_EXPIRED); -TRACE_DEFINE_ENUM(NFS4ERR_FBIG); -TRACE_DEFINE_ENUM(NFS4ERR_FHEXPIRED); -TRACE_DEFINE_ENUM(NFS4ERR_FILE_OPEN); -TRACE_DEFINE_ENUM(NFS4ERR_GRACE); -TRACE_DEFINE_ENUM(NFS4ERR_HASH_ALG_UNSUPP); -TRACE_DEFINE_ENUM(NFS4ERR_INVAL); -TRACE_DEFINE_ENUM(NFS4ERR_IO); -TRACE_DEFINE_ENUM(NFS4ERR_ISDIR); -TRACE_DEFINE_ENUM(NFS4ERR_LAYOUTTRYLATER); -TRACE_DEFINE_ENUM(NFS4ERR_LAYOUTUNAVAILABLE); -TRACE_DEFINE_ENUM(NFS4ERR_LEASE_MOVED); -TRACE_DEFINE_ENUM(NFS4ERR_LOCKED); -TRACE_DEFINE_ENUM(NFS4ERR_LOCKS_HELD); -TRACE_DEFINE_ENUM(NFS4ERR_LOCK_RANGE); -TRACE_DEFINE_ENUM(NFS4ERR_MINOR_VERS_MISMATCH); -TRACE_DEFINE_ENUM(NFS4ERR_MLINK); -TRACE_DEFINE_ENUM(NFS4ERR_MOVED); -TRACE_DEFINE_ENUM(NFS4ERR_NAMETOOLONG); -TRACE_DEFINE_ENUM(NFS4ERR_NOENT); -TRACE_DEFINE_ENUM(NFS4ERR_NOFILEHANDLE); -TRACE_DEFINE_ENUM(NFS4ERR_NOMATCHING_LAYOUT); -TRACE_DEFINE_ENUM(NFS4ERR_NOSPC); -TRACE_DEFINE_ENUM(NFS4ERR_NOTDIR); -TRACE_DEFINE_ENUM(NFS4ERR_NOTEMPTY); -TRACE_DEFINE_ENUM(NFS4ERR_NOTSUPP); -TRACE_DEFINE_ENUM(NFS4ERR_NOT_ONLY_OP); -TRACE_DEFINE_ENUM(NFS4ERR_NOT_SAME); -TRACE_DEFINE_ENUM(NFS4ERR_NO_GRACE); -TRACE_DEFINE_ENUM(NFS4ERR_NXIO); -TRACE_DEFINE_ENUM(NFS4ERR_OLD_STATEID); -TRACE_DEFINE_ENUM(NFS4ERR_OPENMODE); -TRACE_DEFINE_ENUM(NFS4ERR_OP_ILLEGAL); -TRACE_DEFINE_ENUM(NFS4ERR_OP_NOT_IN_SESSION); -TRACE_DEFINE_ENUM(NFS4ERR_PERM); -TRACE_DEFINE_ENUM(NFS4ERR_PNFS_IO_HOLE); -TRACE_DEFINE_ENUM(NFS4ERR_PNFS_NO_LAYOUT); -TRACE_DEFINE_ENUM(NFS4ERR_RECALLCONFLICT); -TRACE_DEFINE_ENUM(NFS4ERR_RECLAIM_BAD); -TRACE_DEFINE_ENUM(NFS4ERR_RECLAIM_CONFLICT); -TRACE_DEFINE_ENUM(NFS4ERR_REJECT_DELEG); -TRACE_DEFINE_ENUM(NFS4ERR_REP_TOO_BIG); -TRACE_DEFINE_ENUM(NFS4ERR_REP_TOO_BIG_TO_CACHE); -TRACE_DEFINE_ENUM(NFS4ERR_REQ_TOO_BIG); -TRACE_DEFINE_ENUM(NFS4ERR_RESOURCE); -TRACE_DEFINE_ENUM(NFS4ERR_RESTOREFH); -TRACE_DEFINE_ENUM(NFS4ERR_RETRY_UNCACHED_REP); -TRACE_DEFINE_ENUM(NFS4ERR_RETURNCONFLICT); -TRACE_DEFINE_ENUM(NFS4ERR_ROFS); -TRACE_DEFINE_ENUM(NFS4ERR_SAME); -TRACE_DEFINE_ENUM(NFS4ERR_SHARE_DENIED); -TRACE_DEFINE_ENUM(NFS4ERR_SEQUENCE_POS); -TRACE_DEFINE_ENUM(NFS4ERR_SEQ_FALSE_RETRY); -TRACE_DEFINE_ENUM(NFS4ERR_SEQ_MISORDERED); -TRACE_DEFINE_ENUM(NFS4ERR_SERVERFAULT); -TRACE_DEFINE_ENUM(NFS4ERR_STALE); -TRACE_DEFINE_ENUM(NFS4ERR_STALE_CLIENTID); -TRACE_DEFINE_ENUM(NFS4ERR_STALE_STATEID); -TRACE_DEFINE_ENUM(NFS4ERR_SYMLINK); -TRACE_DEFINE_ENUM(NFS4ERR_TOOSMALL); -TRACE_DEFINE_ENUM(NFS4ERR_TOO_MANY_OPS); -TRACE_DEFINE_ENUM(NFS4ERR_UNKNOWN_LAYOUTTYPE); -TRACE_DEFINE_ENUM(NFS4ERR_UNSAFE_COMPOUND); -TRACE_DEFINE_ENUM(NFS4ERR_WRONGSEC); -TRACE_DEFINE_ENUM(NFS4ERR_WRONG_CRED); -TRACE_DEFINE_ENUM(NFS4ERR_WRONG_TYPE); -TRACE_DEFINE_ENUM(NFS4ERR_XDEV); - -TRACE_DEFINE_ENUM(NFS4ERR_RESET_TO_MDS); -TRACE_DEFINE_ENUM(NFS4ERR_RESET_TO_PNFS); - -#define show_nfsv4_errors(error) \ - __print_symbolic(error, \ - { NFS4_OK, "OK" }, \ - /* Mapped by nfs4_stat_to_errno() */ \ - { EPERM, "EPERM" }, \ - { ENOENT, "ENOENT" }, \ - { EIO, "EIO" }, \ - { ENXIO, "ENXIO" }, \ - { EACCES, "EACCES" }, \ - { EEXIST, "EEXIST" }, \ - { EXDEV, "EXDEV" }, \ - { ENOTDIR, "ENOTDIR" }, \ - { EISDIR, "EISDIR" }, \ - { EFBIG, "EFBIG" }, \ - { ENOSPC, "ENOSPC" }, \ - { EROFS, "EROFS" }, \ - { EMLINK, "EMLINK" }, \ - { ENAMETOOLONG, "ENAMETOOLONG" }, \ - { ENOTEMPTY, "ENOTEMPTY" }, \ - { EDQUOT, "EDQUOT" }, \ - { ESTALE, "ESTALE" }, \ - { EBADHANDLE, "EBADHANDLE" }, \ - { EBADCOOKIE, "EBADCOOKIE" }, \ - { ENOTSUPP, "ENOTSUPP" }, \ - { ETOOSMALL, "ETOOSMALL" }, \ - { EREMOTEIO, "EREMOTEIO" }, \ - { EBADTYPE, "EBADTYPE" }, \ - { EAGAIN, "EAGAIN" }, \ - { ELOOP, "ELOOP" }, \ - { EOPNOTSUPP, "EOPNOTSUPP" }, \ - { EDEADLK, "EDEADLK" }, \ - /* RPC errors */ \ - { ENOMEM, "ENOMEM" }, \ - { EKEYEXPIRED, "EKEYEXPIRED" }, \ - { ETIMEDOUT, "ETIMEDOUT" }, \ - { ERESTARTSYS, "ERESTARTSYS" }, \ - { ECONNREFUSED, "ECONNREFUSED" }, \ - { ECONNRESET, "ECONNRESET" }, \ - { ENETUNREACH, "ENETUNREACH" }, \ - { EHOSTUNREACH, "EHOSTUNREACH" }, \ - { EHOSTDOWN, "EHOSTDOWN" }, \ - { EPIPE, "EPIPE" }, \ - { EPFNOSUPPORT, "EPFNOSUPPORT" }, \ - { EPROTONOSUPPORT, "EPROTONOSUPPORT" }, \ - /* NFSv4 native errors */ \ - { NFS4ERR_ACCESS, "ACCESS" }, \ - { NFS4ERR_ATTRNOTSUPP, "ATTRNOTSUPP" }, \ - { NFS4ERR_ADMIN_REVOKED, "ADMIN_REVOKED" }, \ - { NFS4ERR_BACK_CHAN_BUSY, "BACK_CHAN_BUSY" }, \ - { NFS4ERR_BADCHAR, "BADCHAR" }, \ - { NFS4ERR_BADHANDLE, "BADHANDLE" }, \ - { NFS4ERR_BADIOMODE, "BADIOMODE" }, \ - { NFS4ERR_BADLAYOUT, "BADLAYOUT" }, \ - { NFS4ERR_BADLABEL, "BADLABEL" }, \ - { NFS4ERR_BADNAME, "BADNAME" }, \ - { NFS4ERR_BADOWNER, "BADOWNER" }, \ - { NFS4ERR_BADSESSION, "BADSESSION" }, \ - { NFS4ERR_BADSLOT, "BADSLOT" }, \ - { NFS4ERR_BADTYPE, "BADTYPE" }, \ - { NFS4ERR_BADXDR, "BADXDR" }, \ - { NFS4ERR_BAD_COOKIE, "BAD_COOKIE" }, \ - { NFS4ERR_BAD_HIGH_SLOT, "BAD_HIGH_SLOT" }, \ - { NFS4ERR_BAD_RANGE, "BAD_RANGE" }, \ - { NFS4ERR_BAD_SEQID, "BAD_SEQID" }, \ - { NFS4ERR_BAD_SESSION_DIGEST, "BAD_SESSION_DIGEST" }, \ - { NFS4ERR_BAD_STATEID, "BAD_STATEID" }, \ - { NFS4ERR_CB_PATH_DOWN, "CB_PATH_DOWN" }, \ - { NFS4ERR_CLID_INUSE, "CLID_INUSE" }, \ - { NFS4ERR_CLIENTID_BUSY, "CLIENTID_BUSY" }, \ - { NFS4ERR_COMPLETE_ALREADY, "COMPLETE_ALREADY" }, \ - { NFS4ERR_CONN_NOT_BOUND_TO_SESSION, \ - "CONN_NOT_BOUND_TO_SESSION" }, \ - { NFS4ERR_DEADLOCK, "DEADLOCK" }, \ - { NFS4ERR_DEADSESSION, "DEAD_SESSION" }, \ - { NFS4ERR_DELAY, "DELAY" }, \ - { NFS4ERR_DELEG_ALREADY_WANTED, \ - "DELEG_ALREADY_WANTED" }, \ - { NFS4ERR_DELEG_REVOKED, "DELEG_REVOKED" }, \ - { NFS4ERR_DENIED, "DENIED" }, \ - { NFS4ERR_DIRDELEG_UNAVAIL, "DIRDELEG_UNAVAIL" }, \ - { NFS4ERR_DQUOT, "DQUOT" }, \ - { NFS4ERR_ENCR_ALG_UNSUPP, "ENCR_ALG_UNSUPP" }, \ - { NFS4ERR_EXIST, "EXIST" }, \ - { NFS4ERR_EXPIRED, "EXPIRED" }, \ - { NFS4ERR_FBIG, "FBIG" }, \ - { NFS4ERR_FHEXPIRED, "FHEXPIRED" }, \ - { NFS4ERR_FILE_OPEN, "FILE_OPEN" }, \ - { NFS4ERR_GRACE, "GRACE" }, \ - { NFS4ERR_HASH_ALG_UNSUPP, "HASH_ALG_UNSUPP" }, \ - { NFS4ERR_INVAL, "INVAL" }, \ - { NFS4ERR_IO, "IO" }, \ - { NFS4ERR_ISDIR, "ISDIR" }, \ - { NFS4ERR_LAYOUTTRYLATER, "LAYOUTTRYLATER" }, \ - { NFS4ERR_LAYOUTUNAVAILABLE, "LAYOUTUNAVAILABLE" }, \ - { NFS4ERR_LEASE_MOVED, "LEASE_MOVED" }, \ - { NFS4ERR_LOCKED, "LOCKED" }, \ - { NFS4ERR_LOCKS_HELD, "LOCKS_HELD" }, \ - { NFS4ERR_LOCK_RANGE, "LOCK_RANGE" }, \ - { NFS4ERR_MINOR_VERS_MISMATCH, "MINOR_VERS_MISMATCH" }, \ - { NFS4ERR_MLINK, "MLINK" }, \ - { NFS4ERR_MOVED, "MOVED" }, \ - { NFS4ERR_NAMETOOLONG, "NAMETOOLONG" }, \ - { NFS4ERR_NOENT, "NOENT" }, \ - { NFS4ERR_NOFILEHANDLE, "NOFILEHANDLE" }, \ - { NFS4ERR_NOMATCHING_LAYOUT, "NOMATCHING_LAYOUT" }, \ - { NFS4ERR_NOSPC, "NOSPC" }, \ - { NFS4ERR_NOTDIR, "NOTDIR" }, \ - { NFS4ERR_NOTEMPTY, "NOTEMPTY" }, \ - { NFS4ERR_NOTSUPP, "NOTSUPP" }, \ - { NFS4ERR_NOT_ONLY_OP, "NOT_ONLY_OP" }, \ - { NFS4ERR_NOT_SAME, "NOT_SAME" }, \ - { NFS4ERR_NO_GRACE, "NO_GRACE" }, \ - { NFS4ERR_NXIO, "NXIO" }, \ - { NFS4ERR_OLD_STATEID, "OLD_STATEID" }, \ - { NFS4ERR_OPENMODE, "OPENMODE" }, \ - { NFS4ERR_OP_ILLEGAL, "OP_ILLEGAL" }, \ - { NFS4ERR_OP_NOT_IN_SESSION, "OP_NOT_IN_SESSION" }, \ - { NFS4ERR_PERM, "PERM" }, \ - { NFS4ERR_PNFS_IO_HOLE, "PNFS_IO_HOLE" }, \ - { NFS4ERR_PNFS_NO_LAYOUT, "PNFS_NO_LAYOUT" }, \ - { NFS4ERR_RECALLCONFLICT, "RECALLCONFLICT" }, \ - { NFS4ERR_RECLAIM_BAD, "RECLAIM_BAD" }, \ - { NFS4ERR_RECLAIM_CONFLICT, "RECLAIM_CONFLICT" }, \ - { NFS4ERR_REJECT_DELEG, "REJECT_DELEG" }, \ - { NFS4ERR_REP_TOO_BIG, "REP_TOO_BIG" }, \ - { NFS4ERR_REP_TOO_BIG_TO_CACHE, \ - "REP_TOO_BIG_TO_CACHE" }, \ - { NFS4ERR_REQ_TOO_BIG, "REQ_TOO_BIG" }, \ - { NFS4ERR_RESOURCE, "RESOURCE" }, \ - { NFS4ERR_RESTOREFH, "RESTOREFH" }, \ - { NFS4ERR_RETRY_UNCACHED_REP, "RETRY_UNCACHED_REP" }, \ - { NFS4ERR_RETURNCONFLICT, "RETURNCONFLICT" }, \ - { NFS4ERR_ROFS, "ROFS" }, \ - { NFS4ERR_SAME, "SAME" }, \ - { NFS4ERR_SHARE_DENIED, "SHARE_DENIED" }, \ - { NFS4ERR_SEQUENCE_POS, "SEQUENCE_POS" }, \ - { NFS4ERR_SEQ_FALSE_RETRY, "SEQ_FALSE_RETRY" }, \ - { NFS4ERR_SEQ_MISORDERED, "SEQ_MISORDERED" }, \ - { NFS4ERR_SERVERFAULT, "SERVERFAULT" }, \ - { NFS4ERR_STALE, "STALE" }, \ - { NFS4ERR_STALE_CLIENTID, "STALE_CLIENTID" }, \ - { NFS4ERR_STALE_STATEID, "STALE_STATEID" }, \ - { NFS4ERR_SYMLINK, "SYMLINK" }, \ - { NFS4ERR_TOOSMALL, "TOOSMALL" }, \ - { NFS4ERR_TOO_MANY_OPS, "TOO_MANY_OPS" }, \ - { NFS4ERR_UNKNOWN_LAYOUTTYPE, "UNKNOWN_LAYOUTTYPE" }, \ - { NFS4ERR_UNSAFE_COMPOUND, "UNSAFE_COMPOUND" }, \ - { NFS4ERR_WRONGSEC, "WRONGSEC" }, \ - { NFS4ERR_WRONG_CRED, "WRONG_CRED" }, \ - { NFS4ERR_WRONG_TYPE, "WRONG_TYPE" }, \ - { NFS4ERR_XDEV, "XDEV" }, \ - /* ***** Internal to Linux NFS client ***** */ \ - { NFS4ERR_RESET_TO_MDS, "RESET_TO_MDS" }, \ - { NFS4ERR_RESET_TO_PNFS, "RESET_TO_PNFS" }) +#include #define show_nfs_fattr_flags(valid) \ __print_flags((unsigned long)valid, "|", \ @@ -355,7 +53,7 @@ DECLARE_EVENT_CLASS(nfs4_clientid_event, TP_printk( "error=%ld (%s) dstaddr=%s", -__entry->error, - show_nfsv4_errors(__entry->error), + show_nfs4_status(__entry->error), __get_str(dstaddr) ) ); @@ -379,29 +77,6 @@ DEFINE_NFS4_CLIENTID_EVENT(nfs4_bind_conn_to_session); DEFINE_NFS4_CLIENTID_EVENT(nfs4_sequence); DEFINE_NFS4_CLIENTID_EVENT(nfs4_reclaim_complete); -#define show_nfs4_sequence_status_flags(status) \ - __print_flags((unsigned long)status, "|", \ - { SEQ4_STATUS_CB_PATH_DOWN, "CB_PATH_DOWN" }, \ - { SEQ4_STATUS_CB_GSS_CONTEXTS_EXPIRING, \ - "CB_GSS_CONTEXTS_EXPIRING" }, \ - { SEQ4_STATUS_CB_GSS_CONTEXTS_EXPIRED, \ - "CB_GSS_CONTEXTS_EXPIRED" }, \ - { SEQ4_STATUS_EXPIRED_ALL_STATE_REVOKED, \ - "EXPIRED_ALL_STATE_REVOKED" }, \ - { SEQ4_STATUS_EXPIRED_SOME_STATE_REVOKED, \ - "EXPIRED_SOME_STATE_REVOKED" }, \ - { SEQ4_STATUS_ADMIN_STATE_REVOKED, \ - "ADMIN_STATE_REVOKED" }, \ - { SEQ4_STATUS_RECALLABLE_STATE_REVOKED, \ - "RECALLABLE_STATE_REVOKED" }, \ - { SEQ4_STATUS_LEASE_MOVED, "LEASE_MOVED" }, \ - { SEQ4_STATUS_RESTART_RECLAIM_NEEDED, \ - "RESTART_RECLAIM_NEEDED" }, \ - { SEQ4_STATUS_CB_PATH_DOWN_SESSION, \ - "CB_PATH_DOWN_SESSION" }, \ - { SEQ4_STATUS_BACKCHANNEL_FAULT, \ - "BACKCHANNEL_FAULT" }) - TRACE_EVENT(nfs4_sequence_done, TP_PROTO( const struct nfs4_session *session, @@ -415,7 +90,7 @@ TRACE_EVENT(nfs4_sequence_done, __field(unsigned int, seq_nr) __field(unsigned int, highest_slotid) __field(unsigned int, target_highest_slotid) - __field(unsigned int, status_flags) + __field(unsigned long, status_flags) __field(unsigned long, error) ), @@ -434,16 +109,16 @@ TRACE_EVENT(nfs4_sequence_done, TP_printk( "error=%ld (%s) session=0x%08x slot_nr=%u seq_nr=%u " "highest_slotid=%u target_highest_slotid=%u " - "status_flags=%u (%s)", + "status_flags=0x%lx (%s)", -__entry->error, - show_nfsv4_errors(__entry->error), + show_nfs4_status(__entry->error), __entry->session, __entry->slot_nr, __entry->seq_nr, __entry->highest_slotid, __entry->target_highest_slotid, __entry->status_flags, - show_nfs4_sequence_status_flags(__entry->status_flags) + show_nfs4_seq4_status(__entry->status_flags) ) ); @@ -480,7 +155,7 @@ TRACE_EVENT(nfs4_cb_sequence, "error=%ld (%s) session=0x%08x slot_nr=%u seq_nr=%u " "highest_slotid=%u", -__entry->error, - show_nfsv4_errors(__entry->error), + show_nfs4_status(__entry->error), __entry->session, __entry->slot_nr, __entry->seq_nr, @@ -517,7 +192,7 @@ TRACE_EVENT(nfs4_cb_seqid_err, "error=%ld (%s) session=0x%08x slot_nr=%u seq_nr=%u " "highest_slotid=%u", -__entry->error, - show_nfsv4_errors(__entry->error), + show_nfs4_status(__entry->error), __entry->session, __entry->slot_nr, __entry->seq_nr, @@ -651,7 +326,7 @@ TRACE_EVENT(nfs4_state_mgr_failed, "hostname=%s clp state=%s error=%ld (%s) section=%s", __get_str(hostname), show_nfs4_clp_state(__entry->state), -__entry->error, - show_nfsv4_errors(__entry->error), __get_str(section) + show_nfs4_status(__entry->error), __get_str(section) ) ) @@ -722,7 +397,7 @@ DECLARE_EVENT_CLASS(nfs4_xdr_event, TP_printk(SUNRPC_TRACE_TASK_SPECIFIER " xid=0x%08x error=%ld (%s) operation=%u", __entry->task_id, __entry->client_id, __entry->xid, - -__entry->error, show_nfsv4_errors(__entry->error), + -__entry->error, show_nfs4_status(__entry->error), __entry->op ) ); @@ -837,7 +512,7 @@ DECLARE_EVENT_CLASS(nfs4_open_event, "name=%02x:%02x:%llu/%s stateid=%d:0x%08x " "openstateid=%d:0x%08x", -__entry->error, - show_nfsv4_errors(__entry->error), + show_nfs4_status(__entry->error), __entry->flags, show_fs_fcntl_open_flags(__entry->flags), show_fs_fmode_flags(__entry->fmode), @@ -941,7 +616,7 @@ TRACE_EVENT(nfs4_close, "error=%ld (%s) fmode=%s fileid=%02x:%02x:%llu " "fhandle=0x%08x openstateid=%d:0x%08x", -__entry->error, - show_nfsv4_errors(__entry->error), + show_nfs4_status(__entry->error), __entry->fmode ? show_fs_fmode_flags(__entry->fmode) : "closed", MAJOR(__entry->dev), MINOR(__entry->dev), @@ -996,7 +671,7 @@ DECLARE_EVENT_CLASS(nfs4_lock_event, "fileid=%02x:%02x:%llu fhandle=0x%08x " "stateid=%d:0x%08x", -__entry->error, - show_nfsv4_errors(__entry->error), + show_nfs4_status(__entry->error), show_fs_fcntl_cmd(__entry->cmd), show_fs_fcntl_lock_type(__entry->type), (long long)__entry->start, @@ -1072,7 +747,7 @@ TRACE_EVENT(nfs4_set_lock, "fileid=%02x:%02x:%llu fhandle=0x%08x " "stateid=%d:0x%08x lockstateid=%d:0x%08x", -__entry->error, - show_nfsv4_errors(__entry->error), + show_nfs4_status(__entry->error), show_fs_fcntl_cmd(__entry->cmd), show_fs_fcntl_lock_type(__entry->type), (long long)__entry->start, @@ -1238,7 +913,7 @@ TRACE_EVENT(nfs4_delegreturn_exit, "error=%ld (%s) dev=%02x:%02x fhandle=0x%08x " "stateid=%d:0x%08x", -__entry->error, - show_nfsv4_errors(__entry->error), + show_nfs4_status(__entry->error), MAJOR(__entry->dev), MINOR(__entry->dev), __entry->fhandle, __entry->stateid_seq, __entry->stateid_hash @@ -1281,7 +956,7 @@ DECLARE_EVENT_CLASS(nfs4_test_stateid_event, "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " "stateid=%d:0x%08x", -__entry->error, - show_nfsv4_errors(__entry->error), + show_nfs4_status(__entry->error), MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->fileid, __entry->fhandle, @@ -1328,7 +1003,7 @@ DECLARE_EVENT_CLASS(nfs4_lookup_event, TP_printk( "error=%ld (%s) name=%02x:%02x:%llu/%s", -__entry->error, - show_nfsv4_errors(__entry->error), + show_nfs4_status(__entry->error), MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->dir, __get_str(name) @@ -1375,7 +1050,7 @@ TRACE_EVENT(nfs4_lookupp, TP_printk( "error=%ld (%s) inode=%02x:%02x:%llu", -__entry->error, - show_nfsv4_errors(__entry->error), + show_nfs4_status(__entry->error), MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->ino ) @@ -1414,7 +1089,7 @@ TRACE_EVENT(nfs4_rename, "error=%ld (%s) oldname=%02x:%02x:%llu/%s " "newname=%02x:%02x:%llu/%s", -__entry->error, - show_nfsv4_errors(__entry->error), + show_nfs4_status(__entry->error), MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->olddir, __get_str(oldname), @@ -1449,7 +1124,7 @@ DECLARE_EVENT_CLASS(nfs4_inode_event, TP_printk( "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x", -__entry->error, - show_nfsv4_errors(__entry->error), + show_nfs4_status(__entry->error), MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->fileid, __entry->fhandle @@ -1507,7 +1182,7 @@ DECLARE_EVENT_CLASS(nfs4_inode_stateid_event, "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " "stateid=%d:0x%08x", -__entry->error, - show_nfsv4_errors(__entry->error), + show_nfs4_status(__entry->error), MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->fileid, __entry->fhandle, @@ -1560,7 +1235,7 @@ DECLARE_EVENT_CLASS(nfs4_getattr_event, "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " "valid=%s", -__entry->error, - show_nfsv4_errors(__entry->error), + show_nfs4_status(__entry->error), MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->fileid, __entry->fhandle, @@ -1616,7 +1291,7 @@ DECLARE_EVENT_CLASS(nfs4_inode_callback_event, "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " "dstaddr=%s", -__entry->error, - show_nfsv4_errors(__entry->error), + show_nfs4_status(__entry->error), MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->fileid, __entry->fhandle, @@ -1677,7 +1352,7 @@ DECLARE_EVENT_CLASS(nfs4_inode_stateid_callback_event, "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " "stateid=%d:0x%08x dstaddr=%s", -__entry->error, - show_nfsv4_errors(__entry->error), + show_nfs4_status(__entry->error), MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->fileid, __entry->fhandle, @@ -1726,7 +1401,7 @@ DECLARE_EVENT_CLASS(nfs4_idmap_event, TP_printk( "error=%ld (%s) id=%u name=%s", - -__entry->error, show_nfsv4_errors(__entry->error), + -__entry->error, show_nfs4_status(__entry->error), __entry->id, __get_str(name) ) @@ -1804,7 +1479,7 @@ DECLARE_EVENT_CLASS(nfs4_read_event, "offset=%lld count=%u res=%u stateid=%d:0x%08x " "layoutstateid=%d:0x%08x", -__entry->error, - show_nfsv4_errors(__entry->error), + show_nfs4_status(__entry->error), MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->fileid, __entry->fhandle, @@ -1878,7 +1553,7 @@ DECLARE_EVENT_CLASS(nfs4_write_event, "offset=%lld count=%u res=%u stateid=%d:0x%08x " "layoutstateid=%d:0x%08x", -__entry->error, - show_nfsv4_errors(__entry->error), + show_nfs4_status(__entry->error), MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->fileid, __entry->fhandle, @@ -1942,7 +1617,7 @@ DECLARE_EVENT_CLASS(nfs4_commit_event, "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " "offset=%lld count=%u layoutstateid=%d:0x%08x", -__entry->error, - show_nfsv4_errors(__entry->error), + show_nfs4_status(__entry->error), MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->fileid, __entry->fhandle, @@ -1962,16 +1637,6 @@ DEFINE_NFS4_COMMIT_EVENT(nfs4_commit); #ifdef CONFIG_NFS_V4_1 DEFINE_NFS4_COMMIT_EVENT(nfs4_pnfs_commit_ds); -TRACE_DEFINE_ENUM(IOMODE_READ); -TRACE_DEFINE_ENUM(IOMODE_RW); -TRACE_DEFINE_ENUM(IOMODE_ANY); - -#define show_pnfs_iomode(iomode) \ - __print_symbolic(iomode, \ - { IOMODE_READ, "READ" }, \ - { IOMODE_RW, "RW" }, \ - { IOMODE_ANY, "ANY" }) - TRACE_EVENT(nfs4_layoutget, TP_PROTO( const struct nfs_open_context *ctx, @@ -2027,11 +1692,11 @@ TRACE_EVENT(nfs4_layoutget, "iomode=%s offset=%llu count=%llu stateid=%d:0x%08x " "layoutstateid=%d:0x%08x", -__entry->error, - show_nfsv4_errors(__entry->error), + show_nfs4_status(__entry->error), MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->fileid, __entry->fhandle, - show_pnfs_iomode(__entry->iomode), + show_pnfs_layout_iomode(__entry->iomode), (unsigned long long)__entry->offset, (unsigned long long)__entry->count, __entry->stateid_seq, __entry->stateid_hash, @@ -2125,7 +1790,7 @@ TRACE_EVENT(pnfs_update_layout, MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->fileid, __entry->fhandle, - show_pnfs_iomode(__entry->iomode), + show_pnfs_layout_iomode(__entry->iomode), (unsigned long long)__entry->pos, (unsigned long long)__entry->count, __entry->layoutstateid_seq, __entry->layoutstateid_hash, @@ -2179,7 +1844,7 @@ DECLARE_EVENT_CLASS(pnfs_layout_event, MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->fileid, __entry->fhandle, - show_pnfs_iomode(__entry->iomode), + show_pnfs_layout_iomode(__entry->iomode), (unsigned long long)__entry->pos, (unsigned long long)__entry->count, __entry->layoutstateid_seq, __entry->layoutstateid_hash, @@ -2324,7 +1989,7 @@ DECLARE_EVENT_CLASS(nfs4_flexfiles_io_event, "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " "offset=%llu count=%u stateid=%d:0x%08x dstaddr=%s", -__entry->error, - show_nfsv4_errors(__entry->error), + show_nfs4_status(__entry->error), MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->fileid, __entry->fhandle, @@ -2380,7 +2045,7 @@ TRACE_EVENT(ff_layout_commit_error, "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " "offset=%llu count=%u dstaddr=%s", -__entry->error, - show_nfsv4_errors(__entry->error), + show_nfs4_status(__entry->error), MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->fileid, __entry->fhandle, diff --git a/fs/nfs/nfstrace.h b/fs/nfs/nfstrace.h index 331bcc0c0a75..76f04aa3367c 100644 --- a/fs/nfs/nfstrace.h +++ b/fs/nfs/nfstrace.h @@ -12,6 +12,7 @@ #include #include +#include #include #define nfs_show_cache_validity(v) \ @@ -115,7 +116,7 @@ DECLARE_EVENT_CLASS(nfs_inode_event_done, "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " "type=%u (%s) version=%llu size=%lld " "cache_validity=0x%lx (%s) nfs_flags=0x%lx (%s)", - -__entry->error, nfs_show_status(__entry->error), + -__entry->error, show_nfs_status(__entry->error), MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->fileid, __entry->fhandle, @@ -206,7 +207,7 @@ TRACE_EVENT(nfs_access_exit, "type=%u (%s) version=%llu size=%lld " "cache_validity=0x%lx (%s) nfs_flags=0x%lx (%s) " "mask=0x%x permitted=0x%x", - -__entry->error, nfs_show_status(__entry->error), + -__entry->error, show_nfs_status(__entry->error), MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->fileid, __entry->fhandle, @@ -342,7 +343,7 @@ DECLARE_EVENT_CLASS(nfs_lookup_event_done, TP_printk( "error=%ld (%s) flags=0x%lx (%s) name=%02x:%02x:%llu/%s", - -__entry->error, nfs_show_status(__entry->error), + -__entry->error, show_nfs_status(__entry->error), __entry->flags, show_fs_lookup_flags(__entry->flags), MAJOR(__entry->dev), MINOR(__entry->dev), @@ -433,7 +434,7 @@ TRACE_EVENT(nfs_atomic_open_exit, TP_printk( "error=%ld (%s) flags=0x%lx (%s) fmode=%s " "name=%02x:%02x:%llu/%s", - -__entry->error, nfs_show_status(__entry->error), + -__entry->error, show_nfs_status(__entry->error), __entry->flags, show_fs_fcntl_open_flags(__entry->flags), show_fs_fmode_flags(__entry->fmode), @@ -504,7 +505,7 @@ TRACE_EVENT(nfs_create_exit, TP_printk( "error=%ld (%s) flags=0x%lx (%s) name=%02x:%02x:%llu/%s", - -__entry->error, nfs_show_status(__entry->error), + -__entry->error, show_nfs_status(__entry->error), __entry->flags, show_fs_fcntl_open_flags(__entry->flags), MAJOR(__entry->dev), MINOR(__entry->dev), @@ -574,7 +575,7 @@ DECLARE_EVENT_CLASS(nfs_directory_event_done, TP_printk( "error=%ld (%s) name=%02x:%02x:%llu/%s", - -__entry->error, nfs_show_status(__entry->error), + -__entry->error, show_nfs_status(__entry->error), MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->dir, __get_str(name) @@ -664,7 +665,7 @@ TRACE_EVENT(nfs_link_exit, TP_printk( "error=%ld (%s) fileid=%02x:%02x:%llu name=%02x:%02x:%llu/%s", - -__entry->error, nfs_show_status(__entry->error), + -__entry->error, show_nfs_status(__entry->error), MAJOR(__entry->dev), MINOR(__entry->dev), __entry->fileid, MAJOR(__entry->dev), MINOR(__entry->dev), @@ -751,7 +752,7 @@ DECLARE_EVENT_CLASS(nfs_rename_event_done, TP_printk( "error=%ld (%s) old_name=%02x:%02x:%llu/%s " "new_name=%02x:%02x:%llu/%s", - -__entry->error, nfs_show_status(__entry->error), + -__entry->error, show_nfs_status(__entry->error), MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->old_dir, __get_str(old_name), @@ -805,7 +806,7 @@ TRACE_EVENT(nfs_sillyrename_unlink, TP_printk( "error=%ld (%s) name=%02x:%02x:%llu/%s", - -__entry->error, nfs_show_status(__entry->error), + -__entry->error, show_nfs_status(__entry->error), MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->dir, __get_str(name) @@ -1134,16 +1135,6 @@ TRACE_EVENT(nfs_pgio_error, ) ); -TRACE_DEFINE_ENUM(NFS_UNSTABLE); -TRACE_DEFINE_ENUM(NFS_DATA_SYNC); -TRACE_DEFINE_ENUM(NFS_FILE_SYNC); - -#define nfs_show_stable(stable) \ - __print_symbolic(stable, \ - { NFS_UNSTABLE, "UNSTABLE" }, \ - { NFS_DATA_SYNC, "DATA_SYNC" }, \ - { NFS_FILE_SYNC, "FILE_SYNC" }) - TRACE_EVENT(nfs_initiate_write, TP_PROTO( const struct nfs_pgio_header *hdr @@ -1157,7 +1148,7 @@ TRACE_EVENT(nfs_initiate_write, __field(u64, fileid) __field(loff_t, offset) __field(u32, count) - __field(enum nfs3_stable_how, stable) + __field(unsigned long, stable) ), TP_fast_assign( @@ -1181,7 +1172,7 @@ TRACE_EVENT(nfs_initiate_write, (unsigned long long)__entry->fileid, __entry->fhandle, (long long)__entry->offset, __entry->count, - nfs_show_stable(__entry->stable) + show_nfs_stable_how(__entry->stable) ) ); @@ -1201,7 +1192,7 @@ TRACE_EVENT(nfs_writeback_done, __field(u32, arg_count) __field(u32, res_count) __field(int, status) - __field(enum nfs3_stable_how, stable) + __field(unsigned long, stable) __array(char, verifier, NFS4_VERIFIER_SIZE) ), @@ -1234,8 +1225,8 @@ TRACE_EVENT(nfs_writeback_done, __entry->fhandle, (long long)__entry->offset, __entry->arg_count, __entry->res_count, __entry->status, - nfs_show_stable(__entry->stable), - __print_hex_str(__entry->verifier, NFS4_VERIFIER_SIZE) + show_nfs_stable_how(__entry->stable), + show_nfs4_verifier(__entry->verifier) ) ); @@ -1336,7 +1327,7 @@ TRACE_EVENT(nfs_commit_done, __field(u64, fileid) __field(loff_t, offset) __field(int, status) - __field(enum nfs3_stable_how, stable) + __field(unsigned long, stable) __array(char, verifier, NFS4_VERIFIER_SIZE) ), @@ -1365,8 +1356,8 @@ TRACE_EVENT(nfs_commit_done, (unsigned long long)__entry->fileid, __entry->fhandle, (long long)__entry->offset, __entry->status, - nfs_show_stable(__entry->stable), - __print_hex_str(__entry->verifier, NFS4_VERIFIER_SIZE) + show_nfs_stable_how(__entry->stable), + show_nfs4_verifier(__entry->verifier) ) ); @@ -1403,76 +1394,6 @@ TRACE_EVENT(nfs_fh_to_dentry, ) ); -TRACE_DEFINE_ENUM(NFS_OK); -TRACE_DEFINE_ENUM(NFSERR_PERM); -TRACE_DEFINE_ENUM(NFSERR_NOENT); -TRACE_DEFINE_ENUM(NFSERR_IO); -TRACE_DEFINE_ENUM(NFSERR_NXIO); -TRACE_DEFINE_ENUM(ECHILD); -TRACE_DEFINE_ENUM(NFSERR_EAGAIN); -TRACE_DEFINE_ENUM(NFSERR_ACCES); -TRACE_DEFINE_ENUM(NFSERR_EXIST); -TRACE_DEFINE_ENUM(NFSERR_XDEV); -TRACE_DEFINE_ENUM(NFSERR_NODEV); -TRACE_DEFINE_ENUM(NFSERR_NOTDIR); -TRACE_DEFINE_ENUM(NFSERR_ISDIR); -TRACE_DEFINE_ENUM(NFSERR_INVAL); -TRACE_DEFINE_ENUM(NFSERR_FBIG); -TRACE_DEFINE_ENUM(NFSERR_NOSPC); -TRACE_DEFINE_ENUM(NFSERR_ROFS); -TRACE_DEFINE_ENUM(NFSERR_MLINK); -TRACE_DEFINE_ENUM(NFSERR_OPNOTSUPP); -TRACE_DEFINE_ENUM(NFSERR_NAMETOOLONG); -TRACE_DEFINE_ENUM(NFSERR_NOTEMPTY); -TRACE_DEFINE_ENUM(NFSERR_DQUOT); -TRACE_DEFINE_ENUM(NFSERR_STALE); -TRACE_DEFINE_ENUM(NFSERR_REMOTE); -TRACE_DEFINE_ENUM(NFSERR_WFLUSH); -TRACE_DEFINE_ENUM(NFSERR_BADHANDLE); -TRACE_DEFINE_ENUM(NFSERR_NOT_SYNC); -TRACE_DEFINE_ENUM(NFSERR_BAD_COOKIE); -TRACE_DEFINE_ENUM(NFSERR_NOTSUPP); -TRACE_DEFINE_ENUM(NFSERR_TOOSMALL); -TRACE_DEFINE_ENUM(NFSERR_SERVERFAULT); -TRACE_DEFINE_ENUM(NFSERR_BADTYPE); -TRACE_DEFINE_ENUM(NFSERR_JUKEBOX); - -#define nfs_show_status(x) \ - __print_symbolic(x, \ - { NFS_OK, "OK" }, \ - { NFSERR_PERM, "PERM" }, \ - { NFSERR_NOENT, "NOENT" }, \ - { NFSERR_IO, "IO" }, \ - { NFSERR_NXIO, "NXIO" }, \ - { ECHILD, "CHILD" }, \ - { NFSERR_EAGAIN, "AGAIN" }, \ - { NFSERR_ACCES, "ACCES" }, \ - { NFSERR_EXIST, "EXIST" }, \ - { NFSERR_XDEV, "XDEV" }, \ - { NFSERR_NODEV, "NODEV" }, \ - { NFSERR_NOTDIR, "NOTDIR" }, \ - { NFSERR_ISDIR, "ISDIR" }, \ - { NFSERR_INVAL, "INVAL" }, \ - { NFSERR_FBIG, "FBIG" }, \ - { NFSERR_NOSPC, "NOSPC" }, \ - { NFSERR_ROFS, "ROFS" }, \ - { NFSERR_MLINK, "MLINK" }, \ - { NFSERR_OPNOTSUPP, "OPNOTSUPP" }, \ - { NFSERR_NAMETOOLONG, "NAMETOOLONG" }, \ - { NFSERR_NOTEMPTY, "NOTEMPTY" }, \ - { NFSERR_DQUOT, "DQUOT" }, \ - { NFSERR_STALE, "STALE" }, \ - { NFSERR_REMOTE, "REMOTE" }, \ - { NFSERR_WFLUSH, "WFLUSH" }, \ - { NFSERR_BADHANDLE, "BADHANDLE" }, \ - { NFSERR_NOT_SYNC, "NOTSYNC" }, \ - { NFSERR_BAD_COOKIE, "BADCOOKIE" }, \ - { NFSERR_NOTSUPP, "NOTSUPP" }, \ - { NFSERR_TOOSMALL, "TOOSMALL" }, \ - { NFSERR_SERVERFAULT, "REMOTEIO" }, \ - { NFSERR_BADTYPE, "BADTYPE" }, \ - { NFSERR_JUKEBOX, "JUKEBOX" }) - DECLARE_EVENT_CLASS(nfs_xdr_event, TP_PROTO( const struct xdr_stream *xdr, @@ -1512,7 +1433,7 @@ DECLARE_EVENT_CLASS(nfs_xdr_event, __entry->task_id, __entry->client_id, __entry->xid, __get_str(program), __entry->version, __get_str(procedure), -__entry->error, - nfs_show_status(__entry->error) + show_nfs_status(__entry->error) ) ); #define DEFINE_NFS_XDR_EVENT(name) \ diff --git a/fs/nfs/pnfs.h b/fs/nfs/pnfs.h index a0f6ff094b3a..f4d7548d67b2 100644 --- a/fs/nfs/pnfs.h +++ b/fs/nfs/pnfs.h @@ -82,10 +82,6 @@ enum pnfs_try_status { PNFS_TRY_AGAIN = 2, }; -/* error codes for internal use */ -#define NFS4ERR_RESET_TO_MDS 12001 -#define NFS4ERR_RESET_TO_PNFS 12002 - #ifdef CONFIG_NFS_V4_1 #define LAYOUT_NFSV4_1_MODULE_PREFIX "nfs-layouttype4" diff --git a/fs/nfsd/trace.h b/fs/nfsd/trace.h index 538520957a81..f1e0d3c51bc2 100644 --- a/fs/nfsd/trace.h +++ b/fs/nfsd/trace.h @@ -9,6 +9,7 @@ #define _NFSD_TRACE_H #include + #include "export.h" #include "nfsfh.h" diff --git a/include/linux/nfs4.h b/include/linux/nfs4.h index 15004c469807..5662d8be04eb 100644 --- a/include/linux/nfs4.h +++ b/include/linux/nfs4.h @@ -292,6 +292,10 @@ enum nfsstat4 { NFS4ERR_XATTR2BIG = 10096, }; +/* error codes for internal client use */ +#define NFS4ERR_RESET_TO_MDS 12001 +#define NFS4ERR_RESET_TO_PNFS 12002 + static inline bool seqid_mutating_err(u32 err) { /* See RFC 7530, section 9.1.7 */ diff --git a/include/trace/events/nfs.h b/include/trace/events/nfs.h new file mode 100644 index 000000000000..09ffdbb04134 --- /dev/null +++ b/include/trace/events/nfs.h @@ -0,0 +1,375 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Display helpers for NFS protocol elements + * + * Author: Chuck Lever + * + * Copyright (c) 2020, Oracle and/or its affiliates. + */ + +#include +#include +#include + +TRACE_DEFINE_ENUM(NFS_OK); +TRACE_DEFINE_ENUM(NFSERR_PERM); +TRACE_DEFINE_ENUM(NFSERR_NOENT); +TRACE_DEFINE_ENUM(NFSERR_IO); +TRACE_DEFINE_ENUM(NFSERR_NXIO); +TRACE_DEFINE_ENUM(NFSERR_EAGAIN); +TRACE_DEFINE_ENUM(NFSERR_ACCES); +TRACE_DEFINE_ENUM(NFSERR_EXIST); +TRACE_DEFINE_ENUM(NFSERR_XDEV); +TRACE_DEFINE_ENUM(NFSERR_NODEV); +TRACE_DEFINE_ENUM(NFSERR_NOTDIR); +TRACE_DEFINE_ENUM(NFSERR_ISDIR); +TRACE_DEFINE_ENUM(NFSERR_INVAL); +TRACE_DEFINE_ENUM(NFSERR_FBIG); +TRACE_DEFINE_ENUM(NFSERR_NOSPC); +TRACE_DEFINE_ENUM(NFSERR_ROFS); +TRACE_DEFINE_ENUM(NFSERR_MLINK); +TRACE_DEFINE_ENUM(NFSERR_OPNOTSUPP); +TRACE_DEFINE_ENUM(NFSERR_NAMETOOLONG); +TRACE_DEFINE_ENUM(NFSERR_NOTEMPTY); +TRACE_DEFINE_ENUM(NFSERR_DQUOT); +TRACE_DEFINE_ENUM(NFSERR_STALE); +TRACE_DEFINE_ENUM(NFSERR_REMOTE); +TRACE_DEFINE_ENUM(NFSERR_WFLUSH); +TRACE_DEFINE_ENUM(NFSERR_BADHANDLE); +TRACE_DEFINE_ENUM(NFSERR_NOT_SYNC); +TRACE_DEFINE_ENUM(NFSERR_BAD_COOKIE); +TRACE_DEFINE_ENUM(NFSERR_NOTSUPP); +TRACE_DEFINE_ENUM(NFSERR_TOOSMALL); +TRACE_DEFINE_ENUM(NFSERR_SERVERFAULT); +TRACE_DEFINE_ENUM(NFSERR_BADTYPE); +TRACE_DEFINE_ENUM(NFSERR_JUKEBOX); + +#define show_nfs_status(x) \ + __print_symbolic(x, \ + { NFS_OK, "OK" }, \ + { NFSERR_PERM, "PERM" }, \ + { NFSERR_NOENT, "NOENT" }, \ + { NFSERR_IO, "IO" }, \ + { NFSERR_NXIO, "NXIO" }, \ + { ECHILD, "CHILD" }, \ + { NFSERR_EAGAIN, "AGAIN" }, \ + { NFSERR_ACCES, "ACCES" }, \ + { NFSERR_EXIST, "EXIST" }, \ + { NFSERR_XDEV, "XDEV" }, \ + { NFSERR_NODEV, "NODEV" }, \ + { NFSERR_NOTDIR, "NOTDIR" }, \ + { NFSERR_ISDIR, "ISDIR" }, \ + { NFSERR_INVAL, "INVAL" }, \ + { NFSERR_FBIG, "FBIG" }, \ + { NFSERR_NOSPC, "NOSPC" }, \ + { NFSERR_ROFS, "ROFS" }, \ + { NFSERR_MLINK, "MLINK" }, \ + { NFSERR_OPNOTSUPP, "OPNOTSUPP" }, \ + { NFSERR_NAMETOOLONG, "NAMETOOLONG" }, \ + { NFSERR_NOTEMPTY, "NOTEMPTY" }, \ + { NFSERR_DQUOT, "DQUOT" }, \ + { NFSERR_STALE, "STALE" }, \ + { NFSERR_REMOTE, "REMOTE" }, \ + { NFSERR_WFLUSH, "WFLUSH" }, \ + { NFSERR_BADHANDLE, "BADHANDLE" }, \ + { NFSERR_NOT_SYNC, "NOTSYNC" }, \ + { NFSERR_BAD_COOKIE, "BADCOOKIE" }, \ + { NFSERR_NOTSUPP, "NOTSUPP" }, \ + { NFSERR_TOOSMALL, "TOOSMALL" }, \ + { NFSERR_SERVERFAULT, "REMOTEIO" }, \ + { NFSERR_BADTYPE, "BADTYPE" }, \ + { NFSERR_JUKEBOX, "JUKEBOX" }) + +TRACE_DEFINE_ENUM(NFS_UNSTABLE); +TRACE_DEFINE_ENUM(NFS_DATA_SYNC); +TRACE_DEFINE_ENUM(NFS_FILE_SYNC); + +#define show_nfs_stable_how(x) \ + __print_symbolic(x, \ + { NFS_UNSTABLE, "UNSTABLE" }, \ + { NFS_DATA_SYNC, "DATA_SYNC" }, \ + { NFS_FILE_SYNC, "FILE_SYNC" }) + +TRACE_DEFINE_ENUM(NFS4_OK); +TRACE_DEFINE_ENUM(NFS4ERR_ACCESS); +TRACE_DEFINE_ENUM(NFS4ERR_ATTRNOTSUPP); +TRACE_DEFINE_ENUM(NFS4ERR_ADMIN_REVOKED); +TRACE_DEFINE_ENUM(NFS4ERR_BACK_CHAN_BUSY); +TRACE_DEFINE_ENUM(NFS4ERR_BADCHAR); +TRACE_DEFINE_ENUM(NFS4ERR_BADHANDLE); +TRACE_DEFINE_ENUM(NFS4ERR_BADIOMODE); +TRACE_DEFINE_ENUM(NFS4ERR_BADLAYOUT); +TRACE_DEFINE_ENUM(NFS4ERR_BADLABEL); +TRACE_DEFINE_ENUM(NFS4ERR_BADNAME); +TRACE_DEFINE_ENUM(NFS4ERR_BADOWNER); +TRACE_DEFINE_ENUM(NFS4ERR_BADSESSION); +TRACE_DEFINE_ENUM(NFS4ERR_BADSLOT); +TRACE_DEFINE_ENUM(NFS4ERR_BADTYPE); +TRACE_DEFINE_ENUM(NFS4ERR_BADXDR); +TRACE_DEFINE_ENUM(NFS4ERR_BAD_COOKIE); +TRACE_DEFINE_ENUM(NFS4ERR_BAD_HIGH_SLOT); +TRACE_DEFINE_ENUM(NFS4ERR_BAD_RANGE); +TRACE_DEFINE_ENUM(NFS4ERR_BAD_SEQID); +TRACE_DEFINE_ENUM(NFS4ERR_BAD_SESSION_DIGEST); +TRACE_DEFINE_ENUM(NFS4ERR_BAD_STATEID); +TRACE_DEFINE_ENUM(NFS4ERR_CB_PATH_DOWN); +TRACE_DEFINE_ENUM(NFS4ERR_CLID_INUSE); +TRACE_DEFINE_ENUM(NFS4ERR_CLIENTID_BUSY); +TRACE_DEFINE_ENUM(NFS4ERR_COMPLETE_ALREADY); +TRACE_DEFINE_ENUM(NFS4ERR_CONN_NOT_BOUND_TO_SESSION); +TRACE_DEFINE_ENUM(NFS4ERR_DEADLOCK); +TRACE_DEFINE_ENUM(NFS4ERR_DEADSESSION); +TRACE_DEFINE_ENUM(NFS4ERR_DELAY); +TRACE_DEFINE_ENUM(NFS4ERR_DELEG_ALREADY_WANTED); +TRACE_DEFINE_ENUM(NFS4ERR_DELEG_REVOKED); +TRACE_DEFINE_ENUM(NFS4ERR_DENIED); +TRACE_DEFINE_ENUM(NFS4ERR_DIRDELEG_UNAVAIL); +TRACE_DEFINE_ENUM(NFS4ERR_DQUOT); +TRACE_DEFINE_ENUM(NFS4ERR_ENCR_ALG_UNSUPP); +TRACE_DEFINE_ENUM(NFS4ERR_EXIST); +TRACE_DEFINE_ENUM(NFS4ERR_EXPIRED); +TRACE_DEFINE_ENUM(NFS4ERR_FBIG); +TRACE_DEFINE_ENUM(NFS4ERR_FHEXPIRED); +TRACE_DEFINE_ENUM(NFS4ERR_FILE_OPEN); +TRACE_DEFINE_ENUM(NFS4ERR_GRACE); +TRACE_DEFINE_ENUM(NFS4ERR_HASH_ALG_UNSUPP); +TRACE_DEFINE_ENUM(NFS4ERR_INVAL); +TRACE_DEFINE_ENUM(NFS4ERR_IO); +TRACE_DEFINE_ENUM(NFS4ERR_ISDIR); +TRACE_DEFINE_ENUM(NFS4ERR_LAYOUTTRYLATER); +TRACE_DEFINE_ENUM(NFS4ERR_LAYOUTUNAVAILABLE); +TRACE_DEFINE_ENUM(NFS4ERR_LEASE_MOVED); +TRACE_DEFINE_ENUM(NFS4ERR_LOCKED); +TRACE_DEFINE_ENUM(NFS4ERR_LOCKS_HELD); +TRACE_DEFINE_ENUM(NFS4ERR_LOCK_RANGE); +TRACE_DEFINE_ENUM(NFS4ERR_MINOR_VERS_MISMATCH); +TRACE_DEFINE_ENUM(NFS4ERR_MLINK); +TRACE_DEFINE_ENUM(NFS4ERR_MOVED); +TRACE_DEFINE_ENUM(NFS4ERR_NAMETOOLONG); +TRACE_DEFINE_ENUM(NFS4ERR_NOENT); +TRACE_DEFINE_ENUM(NFS4ERR_NOFILEHANDLE); +TRACE_DEFINE_ENUM(NFS4ERR_NOMATCHING_LAYOUT); +TRACE_DEFINE_ENUM(NFS4ERR_NOSPC); +TRACE_DEFINE_ENUM(NFS4ERR_NOTDIR); +TRACE_DEFINE_ENUM(NFS4ERR_NOTEMPTY); +TRACE_DEFINE_ENUM(NFS4ERR_NOTSUPP); +TRACE_DEFINE_ENUM(NFS4ERR_NOT_ONLY_OP); +TRACE_DEFINE_ENUM(NFS4ERR_NOT_SAME); +TRACE_DEFINE_ENUM(NFS4ERR_NO_GRACE); +TRACE_DEFINE_ENUM(NFS4ERR_NXIO); +TRACE_DEFINE_ENUM(NFS4ERR_OLD_STATEID); +TRACE_DEFINE_ENUM(NFS4ERR_OPENMODE); +TRACE_DEFINE_ENUM(NFS4ERR_OP_ILLEGAL); +TRACE_DEFINE_ENUM(NFS4ERR_OP_NOT_IN_SESSION); +TRACE_DEFINE_ENUM(NFS4ERR_PERM); +TRACE_DEFINE_ENUM(NFS4ERR_PNFS_IO_HOLE); +TRACE_DEFINE_ENUM(NFS4ERR_PNFS_NO_LAYOUT); +TRACE_DEFINE_ENUM(NFS4ERR_RECALLCONFLICT); +TRACE_DEFINE_ENUM(NFS4ERR_RECLAIM_BAD); +TRACE_DEFINE_ENUM(NFS4ERR_RECLAIM_CONFLICT); +TRACE_DEFINE_ENUM(NFS4ERR_REJECT_DELEG); +TRACE_DEFINE_ENUM(NFS4ERR_REP_TOO_BIG); +TRACE_DEFINE_ENUM(NFS4ERR_REP_TOO_BIG_TO_CACHE); +TRACE_DEFINE_ENUM(NFS4ERR_REQ_TOO_BIG); +TRACE_DEFINE_ENUM(NFS4ERR_RESOURCE); +TRACE_DEFINE_ENUM(NFS4ERR_RESTOREFH); +TRACE_DEFINE_ENUM(NFS4ERR_RETRY_UNCACHED_REP); +TRACE_DEFINE_ENUM(NFS4ERR_RETURNCONFLICT); +TRACE_DEFINE_ENUM(NFS4ERR_ROFS); +TRACE_DEFINE_ENUM(NFS4ERR_SAME); +TRACE_DEFINE_ENUM(NFS4ERR_SHARE_DENIED); +TRACE_DEFINE_ENUM(NFS4ERR_SEQUENCE_POS); +TRACE_DEFINE_ENUM(NFS4ERR_SEQ_FALSE_RETRY); +TRACE_DEFINE_ENUM(NFS4ERR_SEQ_MISORDERED); +TRACE_DEFINE_ENUM(NFS4ERR_SERVERFAULT); +TRACE_DEFINE_ENUM(NFS4ERR_STALE); +TRACE_DEFINE_ENUM(NFS4ERR_STALE_CLIENTID); +TRACE_DEFINE_ENUM(NFS4ERR_STALE_STATEID); +TRACE_DEFINE_ENUM(NFS4ERR_SYMLINK); +TRACE_DEFINE_ENUM(NFS4ERR_TOOSMALL); +TRACE_DEFINE_ENUM(NFS4ERR_TOO_MANY_OPS); +TRACE_DEFINE_ENUM(NFS4ERR_UNKNOWN_LAYOUTTYPE); +TRACE_DEFINE_ENUM(NFS4ERR_UNSAFE_COMPOUND); +TRACE_DEFINE_ENUM(NFS4ERR_WRONGSEC); +TRACE_DEFINE_ENUM(NFS4ERR_WRONG_CRED); +TRACE_DEFINE_ENUM(NFS4ERR_WRONG_TYPE); +TRACE_DEFINE_ENUM(NFS4ERR_XDEV); + +TRACE_DEFINE_ENUM(NFS4ERR_RESET_TO_MDS); +TRACE_DEFINE_ENUM(NFS4ERR_RESET_TO_PNFS); + +#define show_nfs4_status(x) \ + __print_symbolic(x, \ + { NFS4_OK, "OK" }, \ + { EPERM, "EPERM" }, \ + { ENOENT, "ENOENT" }, \ + { EIO, "EIO" }, \ + { ENXIO, "ENXIO" }, \ + { EACCES, "EACCES" }, \ + { EEXIST, "EEXIST" }, \ + { EXDEV, "EXDEV" }, \ + { ENOTDIR, "ENOTDIR" }, \ + { EISDIR, "EISDIR" }, \ + { EFBIG, "EFBIG" }, \ + { ENOSPC, "ENOSPC" }, \ + { EROFS, "EROFS" }, \ + { EMLINK, "EMLINK" }, \ + { ENAMETOOLONG, "ENAMETOOLONG" }, \ + { ENOTEMPTY, "ENOTEMPTY" }, \ + { EDQUOT, "EDQUOT" }, \ + { ESTALE, "ESTALE" }, \ + { EBADHANDLE, "EBADHANDLE" }, \ + { EBADCOOKIE, "EBADCOOKIE" }, \ + { ENOTSUPP, "ENOTSUPP" }, \ + { ETOOSMALL, "ETOOSMALL" }, \ + { EREMOTEIO, "EREMOTEIO" }, \ + { EBADTYPE, "EBADTYPE" }, \ + { EAGAIN, "EAGAIN" }, \ + { ELOOP, "ELOOP" }, \ + { EOPNOTSUPP, "EOPNOTSUPP" }, \ + { EDEADLK, "EDEADLK" }, \ + { ENOMEM, "ENOMEM" }, \ + { EKEYEXPIRED, "EKEYEXPIRED" }, \ + { ETIMEDOUT, "ETIMEDOUT" }, \ + { ERESTARTSYS, "ERESTARTSYS" }, \ + { ECONNREFUSED, "ECONNREFUSED" }, \ + { ECONNRESET, "ECONNRESET" }, \ + { ENETUNREACH, "ENETUNREACH" }, \ + { EHOSTUNREACH, "EHOSTUNREACH" }, \ + { EHOSTDOWN, "EHOSTDOWN" }, \ + { EPIPE, "EPIPE" }, \ + { EPFNOSUPPORT, "EPFNOSUPPORT" }, \ + { EPROTONOSUPPORT, "EPROTONOSUPPORT" }, \ + { NFS4ERR_ACCESS, "ACCESS" }, \ + { NFS4ERR_ATTRNOTSUPP, "ATTRNOTSUPP" }, \ + { NFS4ERR_ADMIN_REVOKED, "ADMIN_REVOKED" }, \ + { NFS4ERR_BACK_CHAN_BUSY, "BACK_CHAN_BUSY" }, \ + { NFS4ERR_BADCHAR, "BADCHAR" }, \ + { NFS4ERR_BADHANDLE, "BADHANDLE" }, \ + { NFS4ERR_BADIOMODE, "BADIOMODE" }, \ + { NFS4ERR_BADLAYOUT, "BADLAYOUT" }, \ + { NFS4ERR_BADLABEL, "BADLABEL" }, \ + { NFS4ERR_BADNAME, "BADNAME" }, \ + { NFS4ERR_BADOWNER, "BADOWNER" }, \ + { NFS4ERR_BADSESSION, "BADSESSION" }, \ + { NFS4ERR_BADSLOT, "BADSLOT" }, \ + { NFS4ERR_BADTYPE, "BADTYPE" }, \ + { NFS4ERR_BADXDR, "BADXDR" }, \ + { NFS4ERR_BAD_COOKIE, "BAD_COOKIE" }, \ + { NFS4ERR_BAD_HIGH_SLOT, "BAD_HIGH_SLOT" }, \ + { NFS4ERR_BAD_RANGE, "BAD_RANGE" }, \ + { NFS4ERR_BAD_SEQID, "BAD_SEQID" }, \ + { NFS4ERR_BAD_SESSION_DIGEST, "BAD_SESSION_DIGEST" }, \ + { NFS4ERR_BAD_STATEID, "BAD_STATEID" }, \ + { NFS4ERR_CB_PATH_DOWN, "CB_PATH_DOWN" }, \ + { NFS4ERR_CLID_INUSE, "CLID_INUSE" }, \ + { NFS4ERR_CLIENTID_BUSY, "CLIENTID_BUSY" }, \ + { NFS4ERR_COMPLETE_ALREADY, "COMPLETE_ALREADY" }, \ + { NFS4ERR_CONN_NOT_BOUND_TO_SESSION, "CONN_NOT_BOUND_TO_SESSION" }, \ + { NFS4ERR_DEADLOCK, "DEADLOCK" }, \ + { NFS4ERR_DEADSESSION, "DEAD_SESSION" }, \ + { NFS4ERR_DELAY, "DELAY" }, \ + { NFS4ERR_DELEG_ALREADY_WANTED, "DELEG_ALREADY_WANTED" }, \ + { NFS4ERR_DELEG_REVOKED, "DELEG_REVOKED" }, \ + { NFS4ERR_DENIED, "DENIED" }, \ + { NFS4ERR_DIRDELEG_UNAVAIL, "DIRDELEG_UNAVAIL" }, \ + { NFS4ERR_DQUOT, "DQUOT" }, \ + { NFS4ERR_ENCR_ALG_UNSUPP, "ENCR_ALG_UNSUPP" }, \ + { NFS4ERR_EXIST, "EXIST" }, \ + { NFS4ERR_EXPIRED, "EXPIRED" }, \ + { NFS4ERR_FBIG, "FBIG" }, \ + { NFS4ERR_FHEXPIRED, "FHEXPIRED" }, \ + { NFS4ERR_FILE_OPEN, "FILE_OPEN" }, \ + { NFS4ERR_GRACE, "GRACE" }, \ + { NFS4ERR_HASH_ALG_UNSUPP, "HASH_ALG_UNSUPP" }, \ + { NFS4ERR_INVAL, "INVAL" }, \ + { NFS4ERR_IO, "IO" }, \ + { NFS4ERR_ISDIR, "ISDIR" }, \ + { NFS4ERR_LAYOUTTRYLATER, "LAYOUTTRYLATER" }, \ + { NFS4ERR_LAYOUTUNAVAILABLE, "LAYOUTUNAVAILABLE" }, \ + { NFS4ERR_LEASE_MOVED, "LEASE_MOVED" }, \ + { NFS4ERR_LOCKED, "LOCKED" }, \ + { NFS4ERR_LOCKS_HELD, "LOCKS_HELD" }, \ + { NFS4ERR_LOCK_RANGE, "LOCK_RANGE" }, \ + { NFS4ERR_MINOR_VERS_MISMATCH, "MINOR_VERS_MISMATCH" }, \ + { NFS4ERR_MLINK, "MLINK" }, \ + { NFS4ERR_MOVED, "MOVED" }, \ + { NFS4ERR_NAMETOOLONG, "NAMETOOLONG" }, \ + { NFS4ERR_NOENT, "NOENT" }, \ + { NFS4ERR_NOFILEHANDLE, "NOFILEHANDLE" }, \ + { NFS4ERR_NOMATCHING_LAYOUT, "NOMATCHING_LAYOUT" }, \ + { NFS4ERR_NOSPC, "NOSPC" }, \ + { NFS4ERR_NOTDIR, "NOTDIR" }, \ + { NFS4ERR_NOTEMPTY, "NOTEMPTY" }, \ + { NFS4ERR_NOTSUPP, "NOTSUPP" }, \ + { NFS4ERR_NOT_ONLY_OP, "NOT_ONLY_OP" }, \ + { NFS4ERR_NOT_SAME, "NOT_SAME" }, \ + { NFS4ERR_NO_GRACE, "NO_GRACE" }, \ + { NFS4ERR_NXIO, "NXIO" }, \ + { NFS4ERR_OLD_STATEID, "OLD_STATEID" }, \ + { NFS4ERR_OPENMODE, "OPENMODE" }, \ + { NFS4ERR_OP_ILLEGAL, "OP_ILLEGAL" }, \ + { NFS4ERR_OP_NOT_IN_SESSION, "OP_NOT_IN_SESSION" }, \ + { NFS4ERR_PERM, "PERM" }, \ + { NFS4ERR_PNFS_IO_HOLE, "PNFS_IO_HOLE" }, \ + { NFS4ERR_PNFS_NO_LAYOUT, "PNFS_NO_LAYOUT" }, \ + { NFS4ERR_RECALLCONFLICT, "RECALLCONFLICT" }, \ + { NFS4ERR_RECLAIM_BAD, "RECLAIM_BAD" }, \ + { NFS4ERR_RECLAIM_CONFLICT, "RECLAIM_CONFLICT" }, \ + { NFS4ERR_REJECT_DELEG, "REJECT_DELEG" }, \ + { NFS4ERR_REP_TOO_BIG, "REP_TOO_BIG" }, \ + { NFS4ERR_REP_TOO_BIG_TO_CACHE, "REP_TOO_BIG_TO_CACHE" }, \ + { NFS4ERR_REQ_TOO_BIG, "REQ_TOO_BIG" }, \ + { NFS4ERR_RESOURCE, "RESOURCE" }, \ + { NFS4ERR_RESTOREFH, "RESTOREFH" }, \ + { NFS4ERR_RETRY_UNCACHED_REP, "RETRY_UNCACHED_REP" }, \ + { NFS4ERR_RETURNCONFLICT, "RETURNCONFLICT" }, \ + { NFS4ERR_ROFS, "ROFS" }, \ + { NFS4ERR_SAME, "SAME" }, \ + { NFS4ERR_SHARE_DENIED, "SHARE_DENIED" }, \ + { NFS4ERR_SEQUENCE_POS, "SEQUENCE_POS" }, \ + { NFS4ERR_SEQ_FALSE_RETRY, "SEQ_FALSE_RETRY" }, \ + { NFS4ERR_SEQ_MISORDERED, "SEQ_MISORDERED" }, \ + { NFS4ERR_SERVERFAULT, "SERVERFAULT" }, \ + { NFS4ERR_STALE, "STALE" }, \ + { NFS4ERR_STALE_CLIENTID, "STALE_CLIENTID" }, \ + { NFS4ERR_STALE_STATEID, "STALE_STATEID" }, \ + { NFS4ERR_SYMLINK, "SYMLINK" }, \ + { NFS4ERR_TOOSMALL, "TOOSMALL" }, \ + { NFS4ERR_TOO_MANY_OPS, "TOO_MANY_OPS" }, \ + { NFS4ERR_UNKNOWN_LAYOUTTYPE, "UNKNOWN_LAYOUTTYPE" }, \ + { NFS4ERR_UNSAFE_COMPOUND, "UNSAFE_COMPOUND" }, \ + { NFS4ERR_WRONGSEC, "WRONGSEC" }, \ + { NFS4ERR_WRONG_CRED, "WRONG_CRED" }, \ + { NFS4ERR_WRONG_TYPE, "WRONG_TYPE" }, \ + { NFS4ERR_XDEV, "XDEV" }, \ + /* ***** Internal to Linux NFS client ***** */ \ + { NFS4ERR_RESET_TO_MDS, "RESET_TO_MDS" }, \ + { NFS4ERR_RESET_TO_PNFS, "RESET_TO_PNFS" }) + +#define show_nfs4_verifier(x) \ + __print_hex_str(x, NFS4_VERIFIER_SIZE) + +TRACE_DEFINE_ENUM(IOMODE_READ); +TRACE_DEFINE_ENUM(IOMODE_RW); +TRACE_DEFINE_ENUM(IOMODE_ANY); + +#define show_pnfs_layout_iomode(x) \ + __print_symbolic(x, \ + { IOMODE_READ, "READ" }, \ + { IOMODE_RW, "RW" }, \ + { IOMODE_ANY, "ANY" }) + +#define show_nfs4_seq4_status(x) \ + __print_flags(x, "|", \ + { SEQ4_STATUS_CB_PATH_DOWN, "CB_PATH_DOWN" }, \ + { SEQ4_STATUS_CB_GSS_CONTEXTS_EXPIRING, "CB_GSS_CONTEXTS_EXPIRING" }, \ + { SEQ4_STATUS_CB_GSS_CONTEXTS_EXPIRED, "CB_GSS_CONTEXTS_EXPIRED" }, \ + { SEQ4_STATUS_EXPIRED_ALL_STATE_REVOKED, "EXPIRED_ALL_STATE_REVOKED" }, \ + { SEQ4_STATUS_EXPIRED_SOME_STATE_REVOKED, "EXPIRED_SOME_STATE_REVOKED" }, \ + { SEQ4_STATUS_ADMIN_STATE_REVOKED, "ADMIN_STATE_REVOKED" }, \ + { SEQ4_STATUS_RECALLABLE_STATE_REVOKED, "RECALLABLE_STATE_REVOKED" }, \ + { SEQ4_STATUS_LEASE_MOVED, "LEASE_MOVED" }, \ + { SEQ4_STATUS_RESTART_RECLAIM_NEEDED, "RESTART_RECLAIM_NEEDED" }, \ + { SEQ4_STATUS_CB_PATH_DOWN_SESSION, "CB_PATH_DOWN_SESSION" }, \ + { SEQ4_STATUS_BACKCHANNEL_FAULT, "BACKCHANNEL_FAULT" }) From cb5a967f7ce413d08cb86ab2285ed34f5ca54935 Mon Sep 17 00:00:00 2001 From: Benjamin Coddington Date: Tue, 2 Nov 2021 14:48:59 -0400 Subject: [PATCH 49/84] xprtrdma: Fix a maybe-uninitialized compiler warning This minor fix-up keeps GCC from complaining that "last' may be used uninitialized", which breaks some build workflows that have been running with all warnings treated as errors. Signed-off-by: Benjamin Coddington Signed-off-by: Trond Myklebust --- net/sunrpc/xprtrdma/frwr_ops.c | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) diff --git a/net/sunrpc/xprtrdma/frwr_ops.c b/net/sunrpc/xprtrdma/frwr_ops.c index 3eccf365fcb8..ff699307e820 100644 --- a/net/sunrpc/xprtrdma/frwr_ops.c +++ b/net/sunrpc/xprtrdma/frwr_ops.c @@ -515,8 +515,8 @@ void frwr_unmap_sync(struct rpcrdma_xprt *r_xprt, struct rpcrdma_req *req) * a single ib_post_send() call. */ prev = &first; - while ((mr = rpcrdma_mr_pop(&req->rl_registered))) { - + mr = rpcrdma_mr_pop(&req->rl_registered); + do { trace_xprtrdma_mr_localinv(mr); r_xprt->rx_stats.local_inv_needed++; @@ -533,7 +533,8 @@ void frwr_unmap_sync(struct rpcrdma_xprt *r_xprt, struct rpcrdma_req *req) *prev = last; prev = &last->next; - } + } while ((mr = rpcrdma_mr_pop(&req->rl_registered))); + mr = container_of(last, struct rpcrdma_mr, mr_invwr); /* Strong send queue ordering guarantees that when the @@ -617,8 +618,8 @@ void frwr_unmap_async(struct rpcrdma_xprt *r_xprt, struct rpcrdma_req *req) * a single ib_post_send() call. */ prev = &first; - while ((mr = rpcrdma_mr_pop(&req->rl_registered))) { - + mr = rpcrdma_mr_pop(&req->rl_registered); + do { trace_xprtrdma_mr_localinv(mr); r_xprt->rx_stats.local_inv_needed++; @@ -635,7 +636,7 @@ void frwr_unmap_async(struct rpcrdma_xprt *r_xprt, struct rpcrdma_req *req) *prev = last; prev = &last->next; - } + } while ((mr = rpcrdma_mr_pop(&req->rl_registered))); /* Strong send queue ordering guarantees that when the * last WR in the chain completes, all WRs in the chain From edfa0b16bf9eb1e2c93e7e846e0e02c51395ca7b Mon Sep 17 00:00:00 2001 From: Dave Wysochanski Date: Tue, 2 Nov 2021 15:51:55 -0400 Subject: [PATCH 50/84] NFS: Add offset to nfs_aop_readahead tracepoint Add the byte offset of the readahead request to the tracepoint output so we know where the read starts. Before this patch: cat-8104 [002] ..... 813.168775: nfs_aop_readahead: fileid=00:31:141 fhandle=0xe55807f6 version=1756509392533525500 nr_pages=256 cat-8104 [002] ..... 813.174973: nfs_aop_readahead_done: fileid=00:31:141 fhandle=0xe55807f6 version=1756509392533525500 nr_pages=256 ret=0 cat-8104 [002] ..... 813.175963: nfs_aop_readahead: fileid=00:31:141 fhandle=0xe55807f6 version=1756509392533525500 nr_pages=256 cat-8104 [002] ..... 813.183742: nfs_aop_readahead_done: fileid=00:31:141 fhandle=0xe55807f6 version=1756509392533525500 nr_pages=1 ret=0 After this patch: cat-6392 [001] ..... 73.107782: nfs_aop_readahead: fileid=00:31:141 fhandle=0xed22403f version=1756511950029502774 offset=5242880 nr_pages=256 cat-6392 [001] ..... 73.112466: nfs_aop_readahead_done: fileid=00:31:141 fhandle=0xed22403f version=1756511950029502774 nr_pages=256 ret=0 cat-6392 [001] ..... 73.115692: nfs_aop_readahead: fileid=00:31:141 fhandle=0xed22403f version=1756511950029502774 offset=6291456 nr_pages=256 cat-6392 [001] ..... 73.123283: nfs_aop_readahead_done: fileid=00:31:141 fhandle=0xed22403f version=1756511950029502774 nr_pages=256 ret=0 Signed-off-by: Dave Wysochanski Signed-off-by: Trond Myklebust --- fs/nfs/nfstrace.h | 10 +++++++--- fs/nfs/read.c | 2 +- 2 files changed, 8 insertions(+), 4 deletions(-) diff --git a/fs/nfs/nfstrace.h b/fs/nfs/nfstrace.h index 76f04aa3367c..21dac847f1e4 100644 --- a/fs/nfs/nfstrace.h +++ b/fs/nfs/nfstrace.h @@ -889,16 +889,18 @@ TRACE_EVENT(nfs_aop_readpage_done, TRACE_EVENT(nfs_aop_readahead, TP_PROTO( const struct inode *inode, + struct page *page, unsigned int nr_pages ), - TP_ARGS(inode, nr_pages), + TP_ARGS(inode, page, nr_pages), TP_STRUCT__entry( __field(dev_t, dev) __field(u32, fhandle) __field(u64, fileid) __field(u64, version) + __field(loff_t, offset) __field(unsigned int, nr_pages) ), @@ -909,15 +911,16 @@ TRACE_EVENT(nfs_aop_readahead, __entry->fileid = nfsi->fileid; __entry->fhandle = nfs_fhandle_hash(&nfsi->fh); __entry->version = inode_peek_iversion_raw(inode); + __entry->offset = page_index(page) << PAGE_SHIFT; __entry->nr_pages = nr_pages; ), TP_printk( - "fileid=%02x:%02x:%llu fhandle=0x%08x version=%llu nr_pages=%u", + "fileid=%02x:%02x:%llu fhandle=0x%08x version=%llu offset=%lld nr_pages=%u", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->fileid, __entry->fhandle, __entry->version, - __entry->nr_pages + __entry->offset, __entry->nr_pages ) ); @@ -936,6 +939,7 @@ TRACE_EVENT(nfs_aop_readahead_done, __field(int, ret) __field(u64, fileid) __field(u64, version) + __field(loff_t, offset) __field(unsigned int, nr_pages) ), diff --git a/fs/nfs/read.c b/fs/nfs/read.c index c8273d4b12ad..d11af2a9299c 100644 --- a/fs/nfs/read.c +++ b/fs/nfs/read.c @@ -404,7 +404,7 @@ int nfs_readpages(struct file *file, struct address_space *mapping, struct inode *inode = mapping->host; int ret; - trace_nfs_aop_readahead(inode, nr_pages); + trace_nfs_aop_readahead(inode, lru_to_page(pages), nr_pages); nfs_inc_stats(inode, NFSIOS_VFSREADPAGES); ret = -ESTALE; From 576acc259146af848cec0940f573f7125a116b9f Mon Sep 17 00:00:00 2001 From: Scott Mayhew Date: Wed, 3 Nov 2021 06:24:40 -0400 Subject: [PATCH 51/84] nfs4: take a reference on the nfs_client when running FREE_STATEID During umount, the session slot tables are freed. If there are outstanding FREE_STATEID tasks, a use-after-free and slab corruption can occur when rpc_exit_task calls rpc_call_done -> nfs41_sequence_done -> nfs4_sequence_process/nfs41_sequence_free_slot. Prevent that from happening by taking a reference on the nfs_client in nfs41_free_stateid and putting it in nfs41_free_stateid_release. Signed-off-by: Scott Mayhew Signed-off-by: Trond Myklebust --- fs/nfs/nfs4proc.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 1c94f54cab58..127388fabda8 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -10115,6 +10115,10 @@ static void nfs41_free_stateid_done(struct rpc_task *task, void *calldata) static void nfs41_free_stateid_release(void *calldata) { + struct nfs_free_stateid_data *data = calldata; + struct nfs_client *clp = data->server->nfs_client; + + nfs_put_client(clp); kfree(calldata); } @@ -10151,6 +10155,10 @@ static int nfs41_free_stateid(struct nfs_server *server, }; struct nfs_free_stateid_data *data; struct rpc_task *task; + struct nfs_client *clp = server->nfs_client; + + if (!refcount_inc_not_zero(&clp->cl_count)) + return -EIO; nfs4_state_protect(server->nfs_client, NFS_SP4_MACH_CRED_STATEID, &task_setup.rpc_client, &msg); From 4330fe35b8213e92ff51907b4cb6323be943a9ad Mon Sep 17 00:00:00 2001 From: Jonathan Corbet Date: Tue, 2 Nov 2021 16:01:56 -0600 Subject: [PATCH 52/84] nfs: remove unused header Commit 19fcae3d4f2dd ("scsi: remove the SCSI OSD library") deleted the last file that included but left that file behind. It's unused, get rid of it now. Cc: Christoph Hellwig Cc: Trond Myklebust Cc: Anna Schumaker Cc: linux-nfs@vger.kernel.org Signed-off-by: Jonathan Corbet Signed-off-by: Trond Myklebust --- include/linux/pnfs_osd_xdr.h | 317 ----------------------------------- 1 file changed, 317 deletions(-) delete mode 100644 include/linux/pnfs_osd_xdr.h diff --git a/include/linux/pnfs_osd_xdr.h b/include/linux/pnfs_osd_xdr.h deleted file mode 100644 index 17d7d0d20eca..000000000000 --- a/include/linux/pnfs_osd_xdr.h +++ /dev/null @@ -1,317 +0,0 @@ -/* - * pNFS-osd on-the-wire data structures - * - * Copyright (C) 2007 Panasas Inc. [year of first publication] - * All rights reserved. - * - * Benny Halevy - * Boaz Harrosh - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License version 2 - * See the file COPYING included with this distribution for more details. - * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions - * are met: - * - * 1. Redistributions of source code must retain the above copyright - * notice, this list of conditions and the following disclaimer. - * 2. Redistributions in binary form must reproduce the above copyright - * notice, this list of conditions and the following disclaimer in the - * documentation and/or other materials provided with the distribution. - * 3. Neither the name of the Panasas company nor the names of its - * contributors may be used to endorse or promote products derived - * from this software without specific prior written permission. - * - * THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESS OR IMPLIED - * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF - * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE - * DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE - * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR - * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF - * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR - * BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF - * LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING - * NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS - * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - */ -#ifndef __PNFS_OSD_XDR_H__ -#define __PNFS_OSD_XDR_H__ - -#include - -/* - * draft-ietf-nfsv4-minorversion-22 - * draft-ietf-nfsv4-pnfs-obj-12 - */ - -/* Layout Structure */ - -enum pnfs_osd_raid_algorithm4 { - PNFS_OSD_RAID_0 = 1, - PNFS_OSD_RAID_4 = 2, - PNFS_OSD_RAID_5 = 3, - PNFS_OSD_RAID_PQ = 4 /* Reed-Solomon P+Q */ -}; - -/* struct pnfs_osd_data_map4 { - * uint32_t odm_num_comps; - * length4 odm_stripe_unit; - * uint32_t odm_group_width; - * uint32_t odm_group_depth; - * uint32_t odm_mirror_cnt; - * pnfs_osd_raid_algorithm4 odm_raid_algorithm; - * }; - */ -struct pnfs_osd_data_map { - u32 odm_num_comps; - u64 odm_stripe_unit; - u32 odm_group_width; - u32 odm_group_depth; - u32 odm_mirror_cnt; - u32 odm_raid_algorithm; -}; - -/* struct pnfs_osd_objid4 { - * deviceid4 oid_device_id; - * uint64_t oid_partition_id; - * uint64_t oid_object_id; - * }; - */ -struct pnfs_osd_objid { - struct nfs4_deviceid oid_device_id; - u64 oid_partition_id; - u64 oid_object_id; -}; - -/* For printout. I use: - * kprint("dev(%llx:%llx)", _DEVID_LO(pointer), _DEVID_HI(pointer)); - * BE style - */ -#define _DEVID_LO(oid_device_id) \ - (unsigned long long)be64_to_cpup((__be64 *)(oid_device_id)->data) - -#define _DEVID_HI(oid_device_id) \ - (unsigned long long)be64_to_cpup(((__be64 *)(oid_device_id)->data) + 1) - -enum pnfs_osd_version { - PNFS_OSD_MISSING = 0, - PNFS_OSD_VERSION_1 = 1, - PNFS_OSD_VERSION_2 = 2 -}; - -struct pnfs_osd_opaque_cred { - u32 cred_len; - void *cred; -}; - -enum pnfs_osd_cap_key_sec { - PNFS_OSD_CAP_KEY_SEC_NONE = 0, - PNFS_OSD_CAP_KEY_SEC_SSV = 1, -}; - -/* struct pnfs_osd_object_cred4 { - * pnfs_osd_objid4 oc_object_id; - * pnfs_osd_version4 oc_osd_version; - * pnfs_osd_cap_key_sec4 oc_cap_key_sec; - * opaque oc_capability_key<>; - * opaque oc_capability<>; - * }; - */ -struct pnfs_osd_object_cred { - struct pnfs_osd_objid oc_object_id; - u32 oc_osd_version; - u32 oc_cap_key_sec; - struct pnfs_osd_opaque_cred oc_cap_key; - struct pnfs_osd_opaque_cred oc_cap; -}; - -/* struct pnfs_osd_layout4 { - * pnfs_osd_data_map4 olo_map; - * uint32_t olo_comps_index; - * pnfs_osd_object_cred4 olo_components<>; - * }; - */ -struct pnfs_osd_layout { - struct pnfs_osd_data_map olo_map; - u32 olo_comps_index; - u32 olo_num_comps; - struct pnfs_osd_object_cred *olo_comps; -}; - -/* Device Address */ -enum pnfs_osd_targetid_type { - OBJ_TARGET_ANON = 1, - OBJ_TARGET_SCSI_NAME = 2, - OBJ_TARGET_SCSI_DEVICE_ID = 3, -}; - -/* union pnfs_osd_targetid4 switch (pnfs_osd_targetid_type4 oti_type) { - * case OBJ_TARGET_SCSI_NAME: - * string oti_scsi_name<>; - * - * case OBJ_TARGET_SCSI_DEVICE_ID: - * opaque oti_scsi_device_id<>; - * - * default: - * void; - * }; - * - * union pnfs_osd_targetaddr4 switch (bool ota_available) { - * case TRUE: - * netaddr4 ota_netaddr; - * case FALSE: - * void; - * }; - * - * struct pnfs_osd_deviceaddr4 { - * pnfs_osd_targetid4 oda_targetid; - * pnfs_osd_targetaddr4 oda_targetaddr; - * uint64_t oda_lun; - * opaque oda_systemid<>; - * pnfs_osd_object_cred4 oda_root_obj_cred; - * opaque oda_osdname<>; - * }; - */ -struct pnfs_osd_targetid { - u32 oti_type; - struct nfs4_string oti_scsi_device_id; -}; - -/* struct netaddr4 { - * // see struct rpcb in RFC1833 - * string r_netid<>; // network id - * string r_addr<>; // universal address - * }; - */ -struct pnfs_osd_net_addr { - struct nfs4_string r_netid; - struct nfs4_string r_addr; -}; - -struct pnfs_osd_targetaddr { - u32 ota_available; - struct pnfs_osd_net_addr ota_netaddr; -}; - -struct pnfs_osd_deviceaddr { - struct pnfs_osd_targetid oda_targetid; - struct pnfs_osd_targetaddr oda_targetaddr; - u8 oda_lun[8]; - struct nfs4_string oda_systemid; - struct pnfs_osd_object_cred oda_root_obj_cred; - struct nfs4_string oda_osdname; -}; - -/* LAYOUTCOMMIT: layoutupdate */ - -/* union pnfs_osd_deltaspaceused4 switch (bool dsu_valid) { - * case TRUE: - * int64_t dsu_delta; - * case FALSE: - * void; - * }; - * - * struct pnfs_osd_layoutupdate4 { - * pnfs_osd_deltaspaceused4 olu_delta_space_used; - * bool olu_ioerr_flag; - * }; - */ -struct pnfs_osd_layoutupdate { - u32 dsu_valid; - s64 dsu_delta; - u32 olu_ioerr_flag; -}; - -/* LAYOUTRETURN: I/O Rrror Report */ - -enum pnfs_osd_errno { - PNFS_OSD_ERR_EIO = 1, - PNFS_OSD_ERR_NOT_FOUND = 2, - PNFS_OSD_ERR_NO_SPACE = 3, - PNFS_OSD_ERR_BAD_CRED = 4, - PNFS_OSD_ERR_NO_ACCESS = 5, - PNFS_OSD_ERR_UNREACHABLE = 6, - PNFS_OSD_ERR_RESOURCE = 7 -}; - -/* struct pnfs_osd_ioerr4 { - * pnfs_osd_objid4 oer_component; - * length4 oer_comp_offset; - * length4 oer_comp_length; - * bool oer_iswrite; - * pnfs_osd_errno4 oer_errno; - * }; - */ -struct pnfs_osd_ioerr { - struct pnfs_osd_objid oer_component; - u64 oer_comp_offset; - u64 oer_comp_length; - u32 oer_iswrite; - u32 oer_errno; -}; - -/* OSD XDR Client API */ -/* Layout helpers */ -/* Layout decoding is done in two parts: - * 1. First Call pnfs_osd_xdr_decode_layout_map to read in only the header part - * of the layout. @iter members need not be initialized. - * Returned: - * @layout members are set. (@layout->olo_comps set to NULL). - * - * Zero on success, or negative error if passed xdr is broken. - * - * 2. 2nd Call pnfs_osd_xdr_decode_layout_comp() in a loop until it returns - * false, to decode the next component. - * Returned: - * true if there is more to decode or false if we are done or error. - * - * Example: - * struct pnfs_osd_xdr_decode_layout_iter iter; - * struct pnfs_osd_layout layout; - * struct pnfs_osd_object_cred comp; - * int status; - * - * status = pnfs_osd_xdr_decode_layout_map(&layout, &iter, xdr); - * if (unlikely(status)) - * goto err; - * while(pnfs_osd_xdr_decode_layout_comp(&comp, &iter, xdr, &status)) { - * // All of @comp strings point to inside the xdr_buffer - * // or scrach buffer. Copy them out to user memory eg. - * copy_single_comp(dest_comp++, &comp); - * } - * if (unlikely(status)) - * goto err; - */ - -struct pnfs_osd_xdr_decode_layout_iter { - unsigned total_comps; - unsigned decoded_comps; -}; - -extern int pnfs_osd_xdr_decode_layout_map(struct pnfs_osd_layout *layout, - struct pnfs_osd_xdr_decode_layout_iter *iter, struct xdr_stream *xdr); - -extern bool pnfs_osd_xdr_decode_layout_comp(struct pnfs_osd_object_cred *comp, - struct pnfs_osd_xdr_decode_layout_iter *iter, struct xdr_stream *xdr, - int *err); - -/* Device Info helpers */ - -/* Note: All strings inside @deviceaddr point to space inside @p. - * @p should stay valid while @deviceaddr is in use. - */ -extern void pnfs_osd_xdr_decode_deviceaddr( - struct pnfs_osd_deviceaddr *deviceaddr, __be32 *p); - -/* layoutupdate (layout_commit) xdr helpers */ -extern int -pnfs_osd_xdr_encode_layoutupdate(struct xdr_stream *xdr, - struct pnfs_osd_layoutupdate *lou); - -/* osd_ioerror encoding (layout_return) */ -extern __be32 *pnfs_osd_xdr_ioerr_reserve_space(struct xdr_stream *xdr); -extern void pnfs_osd_xdr_encode_ioerr(__be32 *p, struct pnfs_osd_ioerr *ioerr); - -#endif /* __PNFS_OSD_XDR_H__ */ From 17f09d3f619a7ad2d2b021b4e5246f08225b1b0f Mon Sep 17 00:00:00 2001 From: Anna Schumaker Date: Thu, 28 Oct 2021 15:17:41 -0400 Subject: [PATCH 53/84] SUNRPC: Check if the xprt is connected before handling sysfs reads xprts don't immediately reconnect when changing the "dstaddr" property, instead this gets handled the next time an operation uses the transport. This could lead to NULL pointer dereferences when trying to read sysfs files between the disconnect and reconnect operations. Fix this by returning an error if the xprt is not connected. Signed-off-by: Anna Schumaker Signed-off-by: Trond Myklebust --- net/sunrpc/sysfs.c | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) diff --git a/net/sunrpc/sysfs.c b/net/sunrpc/sysfs.c index 9a6f17e18f73..2766dd21935b 100644 --- a/net/sunrpc/sysfs.c +++ b/net/sunrpc/sysfs.c @@ -109,8 +109,10 @@ static ssize_t rpc_sysfs_xprt_srcaddr_show(struct kobject *kobj, struct sock_xprt *sock; ssize_t ret = -1; - if (!xprt) - return 0; + if (!xprt || !xprt_connected(xprt)) { + xprt_put(xprt); + return -ENOTCONN; + } sock = container_of(xprt, struct sock_xprt, xprt); if (kernel_getsockname(sock->sock, (struct sockaddr *)&saddr) < 0) @@ -129,8 +131,10 @@ static ssize_t rpc_sysfs_xprt_info_show(struct kobject *kobj, struct rpc_xprt *xprt = rpc_sysfs_xprt_kobj_get_xprt(kobj); ssize_t ret; - if (!xprt) - return 0; + if (!xprt || !xprt_connected(xprt)) { + xprt_put(xprt); + return -ENOTCONN; + } ret = sprintf(buf, "last_used=%lu\ncur_cong=%lu\ncong_win=%lu\n" "max_num_slots=%u\nmin_num_slots=%u\nnum_reqs=%u\n" From f628d462b366a160c181e89279f15295fad62f59 Mon Sep 17 00:00:00 2001 From: Olga Kornievskaia Date: Thu, 4 Nov 2021 10:57:08 -0400 Subject: [PATCH 54/84] NFSv4.2 add tracepoint to SEEK Add a tracepoint to the SEEK operation. Signed-off-by: Olga Kornievskaia Signed-off-by: Trond Myklebust --- fs/nfs/nfs42proc.c | 1 + fs/nfs/nfs4trace.h | 74 ++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 75 insertions(+) diff --git a/fs/nfs/nfs42proc.c b/fs/nfs/nfs42proc.c index a24349512ffe..87c0dcb8823b 100644 --- a/fs/nfs/nfs42proc.c +++ b/fs/nfs/nfs42proc.c @@ -678,6 +678,7 @@ static loff_t _nfs42_proc_llseek(struct file *filep, status = nfs4_call_sync(server->client, server, &msg, &args.seq_args, &res.seq_res, 0); + trace_nfs4_llseek(inode, &args, &res, status); if (status == -ENOTSUPP) server->caps &= ~NFS_CAP_SEEK; if (status) diff --git a/fs/nfs/nfs4trace.h b/fs/nfs/nfs4trace.h index 18f149f72160..823ac436a1da 100644 --- a/fs/nfs/nfs4trace.h +++ b/fs/nfs/nfs4trace.h @@ -2054,6 +2054,80 @@ TRACE_EVENT(ff_layout_commit_error, ) ); +TRACE_DEFINE_ENUM(NFS4_CONTENT_DATA); +TRACE_DEFINE_ENUM(NFS4_CONTENT_HOLE); + +#define show_llseek_mode(what) \ + __print_symbolic(what, \ + { NFS4_CONTENT_DATA, "DATA" }, \ + { NFS4_CONTENT_HOLE, "HOLE" }) + +#ifdef CONFIG_NFS_V4_2 +TRACE_EVENT(nfs4_llseek, + TP_PROTO( + const struct inode *inode, + const struct nfs42_seek_args *args, + const struct nfs42_seek_res *res, + int error + ), + + TP_ARGS(inode, args, res, error), + + TP_STRUCT__entry( + __field(unsigned long, error) + __field(u32, fhandle) + __field(u32, fileid) + __field(dev_t, dev) + __field(int, stateid_seq) + __field(u32, stateid_hash) + __field(loff_t, offset_s) + __field(u32, what) + __field(loff_t, offset_r) + __field(u32, eof) + ), + + TP_fast_assign( + const struct nfs_inode *nfsi = NFS_I(inode); + const struct nfs_fh *fh = args->sa_fh; + + __entry->fileid = nfsi->fileid; + __entry->dev = inode->i_sb->s_dev; + __entry->fhandle = nfs_fhandle_hash(fh); + __entry->offset_s = args->sa_offset; + __entry->stateid_seq = + be32_to_cpu(args->sa_stateid.seqid); + __entry->stateid_hash = + nfs_stateid_hash(&args->sa_stateid); + __entry->what = args->sa_what; + if (error) { + __entry->error = -error; + __entry->offset_r = 0; + __entry->eof = 0; + } else { + __entry->error = 0; + __entry->offset_r = res->sr_offset; + __entry->eof = res->sr_eof; + } + ), + + TP_printk( + "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " + "stateid=%d:0x%08x offset_s=%llu what=%s " + "offset_r=%llu eof=%u", + -__entry->error, + show_nfs4_status(__entry->error), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, + __entry->stateid_seq, __entry->stateid_hash, + __entry->offset_s, + show_llseek_mode(__entry->what), + __entry->offset_r, + __entry->eof + ) +); + +#endif /* CONFIG_NFS_V4_2 */ #endif /* CONFIG_NFS_V4_1 */ From 40a8241771a77e479146e0205112f88241571822 Mon Sep 17 00:00:00 2001 From: Olga Kornievskaia Date: Thu, 4 Nov 2021 10:57:09 -0400 Subject: [PATCH 55/84] NFSv4.2 add tracepoints to FALLOCATE and DEALLOCATE Add a tracepoint to the FALLOCATE/DEALLOCATE operations. Signed-off-by: Olga Kornievskaia Signed-off-by: Trond Myklebust --- fs/nfs/nfs42proc.c | 4 ++++ fs/nfs/nfs4trace.h | 56 ++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 60 insertions(+) diff --git a/fs/nfs/nfs42proc.c b/fs/nfs/nfs42proc.c index 87c0dcb8823b..c36824888601 100644 --- a/fs/nfs/nfs42proc.c +++ b/fs/nfs/nfs42proc.c @@ -83,6 +83,10 @@ static int _nfs42_proc_fallocate(struct rpc_message *msg, struct file *filep, status = nfs_post_op_update_inode_force_wcc(inode, res.falloc_fattr); + if (msg->rpc_proc == &nfs4_procedures[NFSPROC4_CLNT_ALLOCATE]) + trace_nfs4_fallocate(inode, &args, status); + else + trace_nfs4_deallocate(inode, &args, status); kfree(res.falloc_fattr); return status; } diff --git a/fs/nfs/nfs4trace.h b/fs/nfs/nfs4trace.h index 823ac436a1da..a88464238b88 100644 --- a/fs/nfs/nfs4trace.h +++ b/fs/nfs/nfs4trace.h @@ -2127,6 +2127,62 @@ TRACE_EVENT(nfs4_llseek, ) ); +DECLARE_EVENT_CLASS(nfs4_sparse_event, + TP_PROTO( + const struct inode *inode, + const struct nfs42_falloc_args *args, + int error + ), + + TP_ARGS(inode, args, error), + + TP_STRUCT__entry( + __field(unsigned long, error) + __field(loff_t, offset) + __field(loff_t, len) + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __field(int, stateid_seq) + __field(u32, stateid_hash) + ), + + TP_fast_assign( + __entry->error = error < 0 ? -error : 0; + __entry->offset = args->falloc_offset; + __entry->len = args->falloc_length; + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = NFS_FILEID(inode); + __entry->fhandle = nfs_fhandle_hash(NFS_FH(inode)); + __entry->stateid_seq = + be32_to_cpu(args->falloc_stateid.seqid); + __entry->stateid_hash = + nfs_stateid_hash(&args->falloc_stateid); + ), + + TP_printk( + "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " + "stateid=%d:0x%08x offset=%llu len=%llu", + -__entry->error, + show_nfs4_status(__entry->error), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, + __entry->stateid_seq, __entry->stateid_hash, + (long long)__entry->offset, + (long long)__entry->len + ) +); +#define DEFINE_NFS4_SPARSE_EVENT(name) \ + DEFINE_EVENT(nfs4_sparse_event, name, \ + TP_PROTO( \ + const struct inode *inode, \ + const struct nfs42_falloc_args *args, \ + int error \ + ), \ + TP_ARGS(inode, args, error)) +DEFINE_NFS4_SPARSE_EVENT(nfs4_fallocate); +DEFINE_NFS4_SPARSE_EVENT(nfs4_deallocate); #endif /* CONFIG_NFS_V4_2 */ #endif /* CONFIG_NFS_V4_1 */ From ce7cea1ba72ef2ba91bcb28ab441597405d6d705 Mon Sep 17 00:00:00 2001 From: Olga Kornievskaia Date: Thu, 4 Nov 2021 10:57:10 -0400 Subject: [PATCH 56/84] NFSv4.2 add tracepoint to COPY Add a tracepoint to the COPY operation. Signed-off-by: Olga Kornievskaia Signed-off-by: Trond Myklebust --- fs/nfs/nfs42proc.c | 1 + fs/nfs/nfs4trace.h | 107 +++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 108 insertions(+) diff --git a/fs/nfs/nfs42proc.c b/fs/nfs/nfs42proc.c index c36824888601..a072cdaf7bdc 100644 --- a/fs/nfs/nfs42proc.c +++ b/fs/nfs/nfs42proc.c @@ -367,6 +367,7 @@ static ssize_t _nfs42_proc_copy(struct file *src, status = nfs4_call_sync(dst_server->client, dst_server, &msg, &args->seq_args, &res->seq_res, 0); + trace_nfs4_copy(src_inode, dst_inode, args, res, nss, status); if (status == -ENOTSUPP) dst_server->caps &= ~NFS_CAP_COPY; if (status) diff --git a/fs/nfs/nfs4trace.h b/fs/nfs/nfs4trace.h index a88464238b88..bdccaec504d8 100644 --- a/fs/nfs/nfs4trace.h +++ b/fs/nfs/nfs4trace.h @@ -2183,6 +2183,113 @@ DECLARE_EVENT_CLASS(nfs4_sparse_event, TP_ARGS(inode, args, error)) DEFINE_NFS4_SPARSE_EVENT(nfs4_fallocate); DEFINE_NFS4_SPARSE_EVENT(nfs4_deallocate); + +TRACE_EVENT(nfs4_copy, + TP_PROTO( + const struct inode *src_inode, + const struct inode *dst_inode, + const struct nfs42_copy_args *args, + const struct nfs42_copy_res *res, + const struct nl4_server *nss, + int error + ), + + TP_ARGS(src_inode, dst_inode, args, res, nss, error), + + TP_STRUCT__entry( + __field(unsigned long, error) + __field(u32, src_fhandle) + __field(u32, src_fileid) + __field(u32, dst_fhandle) + __field(u32, dst_fileid) + __field(dev_t, src_dev) + __field(dev_t, dst_dev) + __field(int, src_stateid_seq) + __field(u32, src_stateid_hash) + __field(int, dst_stateid_seq) + __field(u32, dst_stateid_hash) + __field(loff_t, src_offset) + __field(loff_t, dst_offset) + __field(bool, sync) + __field(loff_t, len) + __field(int, res_stateid_seq) + __field(u32, res_stateid_hash) + __field(loff_t, res_count) + __field(bool, res_sync) + __field(bool, res_cons) + __field(bool, intra) + ), + + TP_fast_assign( + const struct nfs_inode *src_nfsi = NFS_I(src_inode); + const struct nfs_inode *dst_nfsi = NFS_I(dst_inode); + + __entry->src_fileid = src_nfsi->fileid; + __entry->src_dev = src_inode->i_sb->s_dev; + __entry->src_fhandle = nfs_fhandle_hash(args->src_fh); + __entry->src_offset = args->src_pos; + __entry->dst_fileid = dst_nfsi->fileid; + __entry->dst_dev = dst_inode->i_sb->s_dev; + __entry->dst_fhandle = nfs_fhandle_hash(args->dst_fh); + __entry->dst_offset = args->dst_pos; + __entry->len = args->count; + __entry->sync = args->sync; + __entry->src_stateid_seq = + be32_to_cpu(args->src_stateid.seqid); + __entry->src_stateid_hash = + nfs_stateid_hash(&args->src_stateid); + __entry->dst_stateid_seq = + be32_to_cpu(args->dst_stateid.seqid); + __entry->dst_stateid_hash = + nfs_stateid_hash(&args->dst_stateid); + __entry->intra = nss ? 0 : 1; + if (error) { + __entry->error = -error; + __entry->res_stateid_seq = 0; + __entry->res_stateid_hash = 0; + __entry->res_count = 0; + __entry->res_sync = 0; + __entry->res_cons = 0; + } else { + __entry->error = 0; + __entry->res_stateid_seq = + be32_to_cpu(res->write_res.stateid.seqid); + __entry->res_stateid_hash = + nfs_stateid_hash(&res->write_res.stateid); + __entry->res_count = res->write_res.count; + __entry->res_sync = res->synchronous; + __entry->res_cons = res->consecutive; + } + ), + + TP_printk( + "error=%ld (%s) intra=%d src_fileid=%02x:%02x:%llu " + "src_fhandle=0x%08x dst_fileid=%02x:%02x:%llu " + "dst_fhandle=0x%08x src_stateid=%d:0x%08x " + "dst_stateid=%d:0x%08x src_offset=%llu dst_offset=%llu " + "len=%llu sync=%d cb_stateid=%d:0x%08x res_sync=%d " + "res_cons=%d res_count=%llu", + -__entry->error, + show_nfs4_status(__entry->error), + __entry->intra, + MAJOR(__entry->src_dev), MINOR(__entry->src_dev), + (unsigned long long)__entry->src_fileid, + __entry->src_fhandle, + MAJOR(__entry->dst_dev), MINOR(__entry->dst_dev), + (unsigned long long)__entry->dst_fileid, + __entry->dst_fhandle, + __entry->src_stateid_seq, __entry->src_stateid_hash, + __entry->dst_stateid_seq, __entry->dst_stateid_hash, + __entry->src_offset, + __entry->dst_offset, + __entry->len, + __entry->sync, + __entry->res_stateid_seq, __entry->res_stateid_hash, + __entry->res_sync, + __entry->res_cons, + __entry->res_count + ) +); #endif /* CONFIG_NFS_V4_2 */ #endif /* CONFIG_NFS_V4_1 */ From 2a65ca8b5850914e2ca67a0b8b8140bf7297353a Mon Sep 17 00:00:00 2001 From: Olga Kornievskaia Date: Thu, 4 Nov 2021 10:57:11 -0400 Subject: [PATCH 57/84] NFSv4.2 add tracepoint to CLONE Add a tracepoint to the CLONE operation. Signed-off-by: Olga Kornievskaia Signed-off-by: Trond Myklebust --- fs/nfs/nfs42proc.c | 1 + fs/nfs/nfs4trace.h | 73 ++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 74 insertions(+) diff --git a/fs/nfs/nfs42proc.c b/fs/nfs/nfs42proc.c index a072cdaf7bdc..d3d9ea71702f 100644 --- a/fs/nfs/nfs42proc.c +++ b/fs/nfs/nfs42proc.c @@ -1077,6 +1077,7 @@ static int _nfs42_proc_clone(struct rpc_message *msg, struct file *src_f, status = nfs4_call_sync(server->client, server, msg, &args.seq_args, &res.seq_res, 0); + trace_nfs4_clone(src_inode, dst_inode, &args, status); if (status == 0) { nfs42_copy_dest_done(dst_inode, dst_offset, count); status = nfs_post_op_update_inode(dst_inode, res.dst_fattr); diff --git a/fs/nfs/nfs4trace.h b/fs/nfs/nfs4trace.h index bdccaec504d8..af7e59aa9265 100644 --- a/fs/nfs/nfs4trace.h +++ b/fs/nfs/nfs4trace.h @@ -2290,6 +2290,79 @@ TRACE_EVENT(nfs4_copy, __entry->res_count ) ); + +TRACE_EVENT(nfs4_clone, + TP_PROTO( + const struct inode *src_inode, + const struct inode *dst_inode, + const struct nfs42_clone_args *args, + int error + ), + + TP_ARGS(src_inode, dst_inode, args, error), + + TP_STRUCT__entry( + __field(unsigned long, error) + __field(u32, src_fhandle) + __field(u32, src_fileid) + __field(u32, dst_fhandle) + __field(u32, dst_fileid) + __field(dev_t, src_dev) + __field(dev_t, dst_dev) + __field(loff_t, src_offset) + __field(loff_t, dst_offset) + __field(int, src_stateid_seq) + __field(u32, src_stateid_hash) + __field(int, dst_stateid_seq) + __field(u32, dst_stateid_hash) + __field(loff_t, len) + ), + + TP_fast_assign( + const struct nfs_inode *src_nfsi = NFS_I(src_inode); + const struct nfs_inode *dst_nfsi = NFS_I(dst_inode); + + __entry->src_fileid = src_nfsi->fileid; + __entry->src_dev = src_inode->i_sb->s_dev; + __entry->src_fhandle = nfs_fhandle_hash(args->src_fh); + __entry->src_offset = args->src_offset; + __entry->dst_fileid = dst_nfsi->fileid; + __entry->dst_dev = dst_inode->i_sb->s_dev; + __entry->dst_fhandle = nfs_fhandle_hash(args->dst_fh); + __entry->dst_offset = args->dst_offset; + __entry->len = args->count; + __entry->error = error < 0 ? -error : 0; + __entry->src_stateid_seq = + be32_to_cpu(args->src_stateid.seqid); + __entry->src_stateid_hash = + nfs_stateid_hash(&args->src_stateid); + __entry->dst_stateid_seq = + be32_to_cpu(args->dst_stateid.seqid); + __entry->dst_stateid_hash = + nfs_stateid_hash(&args->dst_stateid); + ), + + TP_printk( + "error=%ld (%s) src_fileid=%02x:%02x:%llu " + "src_fhandle=0x%08x dst_fileid=%02x:%02x:%llu " + "dst_fhandle=0x%08x src_stateid=%d:0x%08x " + "dst_stateid=%d:0x%08x src_offset=%llu " + "dst_offset=%llu len=%llu", + -__entry->error, + show_nfs4_status(__entry->error), + MAJOR(__entry->src_dev), MINOR(__entry->src_dev), + (unsigned long long)__entry->src_fileid, + __entry->src_fhandle, + MAJOR(__entry->dst_dev), MINOR(__entry->dst_dev), + (unsigned long long)__entry->dst_fileid, + __entry->dst_fhandle, + __entry->src_stateid_seq, __entry->src_stateid_hash, + __entry->dst_stateid_seq, __entry->dst_stateid_hash, + __entry->src_offset, + __entry->dst_offset, + __entry->len + ) +); #endif /* CONFIG_NFS_V4_2 */ #endif /* CONFIG_NFS_V4_1 */ From 8db744ce45ee31d26bd2776070f31121b009f27e Mon Sep 17 00:00:00 2001 From: Olga Kornievskaia Date: Thu, 4 Nov 2021 10:57:12 -0400 Subject: [PATCH 58/84] NFSv4.2 add tracepoint to CB_OFFLOAD Add a tracepoint to the CB_OFFLOAD operation. Signed-off-by: Olga Kornievskaia Signed-off-by: Trond Myklebust --- fs/nfs/callback_proc.c | 3 +++ fs/nfs/nfs4trace.h | 44 ++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 47 insertions(+) diff --git a/fs/nfs/callback_proc.c b/fs/nfs/callback_proc.c index ed9d580826f5..09c5b1cb3e07 100644 --- a/fs/nfs/callback_proc.c +++ b/fs/nfs/callback_proc.c @@ -739,6 +739,9 @@ out: kfree(copy); spin_unlock(&cps->clp->cl_lock); + trace_nfs4_cb_offload(&args->coa_fh, &args->coa_stateid, + args->wr_count, args->error, + args->wr_writeverf.committed); return 0; } #endif /* CONFIG_NFS_V4_2 */ diff --git a/fs/nfs/nfs4trace.h b/fs/nfs/nfs4trace.h index af7e59aa9265..f971e38a6e3a 100644 --- a/fs/nfs/nfs4trace.h +++ b/fs/nfs/nfs4trace.h @@ -200,6 +200,50 @@ TRACE_EVENT(nfs4_cb_seqid_err, ) ); +TRACE_EVENT(nfs4_cb_offload, + TP_PROTO( + const struct nfs_fh *cb_fh, + const nfs4_stateid *cb_stateid, + uint64_t cb_count, + int cb_error, + int cb_how_stable + ), + + TP_ARGS(cb_fh, cb_stateid, cb_count, cb_error, + cb_how_stable), + + TP_STRUCT__entry( + __field(unsigned long, error) + __field(u32, fhandle) + __field(loff_t, cb_count) + __field(int, cb_how) + __field(int, cb_stateid_seq) + __field(u32, cb_stateid_hash) + ), + + TP_fast_assign( + __entry->error = cb_error < 0 ? -cb_error : 0; + __entry->fhandle = nfs_fhandle_hash(cb_fh); + __entry->cb_stateid_seq = + be32_to_cpu(cb_stateid->seqid); + __entry->cb_stateid_hash = + nfs_stateid_hash(cb_stateid); + __entry->cb_count = cb_count; + __entry->cb_how = cb_how_stable; + ), + + TP_printk( + "error=%ld (%s) fhandle=0x%08x cb_stateid=%d:0x%08x " + "cb_count=%llu cb_how=%s", + -__entry->error, + show_nfs4_status(__entry->error), + __entry->fhandle, + __entry->cb_stateid_seq, __entry->cb_stateid_hash, + __entry->cb_count, + show_nfs_stable_how(__entry->cb_how) + ) +); + #endif /* CONFIG_NFS_V4_1 */ TRACE_EVENT(nfs4_setup_sequence, From 488b170c7d78bae2ea493a14d71ac6566f4a0afa Mon Sep 17 00:00:00 2001 From: Olga Kornievskaia Date: Thu, 4 Nov 2021 10:57:13 -0400 Subject: [PATCH 59/84] NFSv4.2 add tracepoint to COPY_NOTIFY Add a tracepoint to COPY_NOTIFY operation. Signed-off-by: Olga Kornievskaia Signed-off-by: Trond Myklebust --- fs/nfs/nfs42proc.c | 1 + fs/nfs/nfs4trace.h | 58 +++++++++++++++++++++++++++++++++++++++++++++- 2 files changed, 58 insertions(+), 1 deletion(-) diff --git a/fs/nfs/nfs42proc.c b/fs/nfs/nfs42proc.c index d3d9ea71702f..7c7399b10050 100644 --- a/fs/nfs/nfs42proc.c +++ b/fs/nfs/nfs42proc.c @@ -603,6 +603,7 @@ static int _nfs42_proc_copy_notify(struct file *src, struct file *dst, status = nfs4_call_sync(src_server->client, src_server, &msg, &args->cna_seq_args, &res->cnr_seq_res, 0); + trace_nfs4_copy_notify(file_inode(src), args, res, status); if (status == -ENOTSUPP) src_server->caps &= ~NFS_CAP_COPY_NOTIFY; diff --git a/fs/nfs/nfs4trace.h b/fs/nfs/nfs4trace.h index f971e38a6e3a..f337a25c67b3 100644 --- a/fs/nfs/nfs4trace.h +++ b/fs/nfs/nfs4trace.h @@ -243,7 +243,6 @@ TRACE_EVENT(nfs4_cb_offload, show_nfs_stable_how(__entry->cb_how) ) ); - #endif /* CONFIG_NFS_V4_1 */ TRACE_EVENT(nfs4_setup_sequence, @@ -2407,6 +2406,63 @@ TRACE_EVENT(nfs4_clone, __entry->len ) ); + +TRACE_EVENT(nfs4_copy_notify, + TP_PROTO( + const struct inode *inode, + const struct nfs42_copy_notify_args *args, + const struct nfs42_copy_notify_res *res, + int error + ), + + TP_ARGS(inode, args, res, error), + + TP_STRUCT__entry( + __field(unsigned long, error) + __field(u32, fhandle) + __field(u32, fileid) + __field(dev_t, dev) + __field(int, stateid_seq) + __field(u32, stateid_hash) + __field(int, res_stateid_seq) + __field(u32, res_stateid_hash) + ), + + TP_fast_assign( + const struct nfs_inode *nfsi = NFS_I(inode); + + __entry->fileid = nfsi->fileid; + __entry->dev = inode->i_sb->s_dev; + __entry->fhandle = nfs_fhandle_hash(args->cna_src_fh); + __entry->stateid_seq = + be32_to_cpu(args->cna_src_stateid.seqid); + __entry->stateid_hash = + nfs_stateid_hash(&args->cna_src_stateid); + if (error) { + __entry->error = -error; + __entry->res_stateid_seq = 0; + __entry->res_stateid_hash = 0; + } else { + __entry->error = 0; + __entry->res_stateid_seq = + be32_to_cpu(res->cnr_stateid.seqid); + __entry->res_stateid_hash = + nfs_stateid_hash(&res->cnr_stateid); + } + ), + + TP_printk( + "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " + "stateid=%d:0x%08x res_stateid=%d:0x%08x", + -__entry->error, + show_nfs4_status(__entry->error), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, + __entry->stateid_seq, __entry->stateid_hash, + __entry->res_stateid_seq, __entry->res_stateid_hash + ) +); #endif /* CONFIG_NFS_V4_2 */ #endif /* CONFIG_NFS_V4_1 */ From 127becabad7bef13b2e2ab795953cf8e9cce8aab Mon Sep 17 00:00:00 2001 From: Olga Kornievskaia Date: Thu, 4 Nov 2021 10:57:14 -0400 Subject: [PATCH 60/84] NFSv4.2 add tracepoint to OFFLOAD_CANCEL Add tracepoint to OFFLOAD_CANCEL operation. Signed-off-by: Olga Kornievskaia Signed-off-by: Trond Myklebust --- fs/nfs/nfs42proc.c | 1 + fs/nfs/nfs4trace.h | 33 +++++++++++++++++++++++++++++++++ 2 files changed, 34 insertions(+) diff --git a/fs/nfs/nfs42proc.c b/fs/nfs/nfs42proc.c index 7c7399b10050..08355b66e7cb 100644 --- a/fs/nfs/nfs42proc.c +++ b/fs/nfs/nfs42proc.c @@ -509,6 +509,7 @@ static void nfs42_offload_cancel_done(struct rpc_task *task, void *calldata) { struct nfs42_offloadcancel_data *data = calldata; + trace_nfs4_offload_cancel(&data->args, task->tk_status); nfs41_sequence_done(task, &data->res.osr_seq_res); if (task->tk_status && nfs4_async_handle_error(task, data->seq_server, NULL, diff --git a/fs/nfs/nfs4trace.h b/fs/nfs/nfs4trace.h index f337a25c67b3..6ee6ad3674a2 100644 --- a/fs/nfs/nfs4trace.h +++ b/fs/nfs/nfs4trace.h @@ -2463,6 +2463,39 @@ TRACE_EVENT(nfs4_copy_notify, __entry->res_stateid_seq, __entry->res_stateid_hash ) ); + +TRACE_EVENT(nfs4_offload_cancel, + TP_PROTO( + const struct nfs42_offload_status_args *args, + int error + ), + + TP_ARGS(args, error), + + TP_STRUCT__entry( + __field(unsigned long, error) + __field(u32, fhandle) + __field(int, stateid_seq) + __field(u32, stateid_hash) + ), + + TP_fast_assign( + __entry->fhandle = nfs_fhandle_hash(args->osa_src_fh); + __entry->error = error < 0 ? -error : 0; + __entry->stateid_seq = + be32_to_cpu(args->osa_stateid.seqid); + __entry->stateid_hash = + nfs_stateid_hash(&args->osa_stateid); + ), + + TP_printk( + "error=%ld (%s) fhandle=0x%08x stateid=%d:0x%08x", + -__entry->error, + show_nfs4_status(__entry->error), + __entry->fhandle, + __entry->stateid_seq, __entry->stateid_hash + ) +); #endif /* CONFIG_NFS_V4_2 */ #endif /* CONFIG_NFS_V4_1 */ From d896ba8300ebd09e5fc9c43051d9eb36b65e996e Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Fri, 29 Oct 2021 12:26:17 -0400 Subject: [PATCH 61/84] SUNRPC: Fix races when closing the socket Ensure that we bump the xprt->connect_cookie when we set the XPRT_CLOSE_WAIT flag so that another call to xprt_conditional_disconnect() won't race with the reconnection. Signed-off-by: Trond Myklebust --- net/sunrpc/xprt.c | 2 ++ net/sunrpc/xprtsock.c | 1 + 2 files changed, 3 insertions(+) diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c index 48560188e84d..691fe5a682b6 100644 --- a/net/sunrpc/xprt.c +++ b/net/sunrpc/xprt.c @@ -735,6 +735,8 @@ static void xprt_autoclose(struct work_struct *work) unsigned int pflags = memalloc_nofs_save(); trace_xprt_disconnect_auto(xprt); + xprt->connect_cookie++; + smp_mb__before_atomic(); clear_bit(XPRT_CLOSE_WAIT, &xprt->state); xprt->ops->close(xprt); xprt_release_write(xprt, NULL); diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index aa293e4a77fa..7fb302e202bc 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -1134,6 +1134,7 @@ static void xs_run_error_worker(struct sock_xprt *transport, unsigned int nr) static void xs_sock_reset_connection_flags(struct rpc_xprt *xprt) { + xprt->connect_cookie++; smp_mb__before_atomic(); clear_bit(XPRT_CLOSE_WAIT, &xprt->state); clear_bit(XPRT_CLOSING, &xprt->state); From 3be232f11a3cc9b0ef0795e39fa11bdb8e422a06 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Tue, 26 Oct 2021 18:01:07 -0400 Subject: [PATCH 62/84] SUNRPC: Prevent immediate close+reconnect If we have already set up the socket and are waiting for it to connect, then don't immediately close and retry. Signed-off-by: Trond Myklebust --- net/sunrpc/xprt.c | 3 ++- net/sunrpc/xprtsock.c | 2 +- 2 files changed, 3 insertions(+), 2 deletions(-) diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c index 691fe5a682b6..a02de2bddb28 100644 --- a/net/sunrpc/xprt.c +++ b/net/sunrpc/xprt.c @@ -767,7 +767,8 @@ EXPORT_SYMBOL_GPL(xprt_disconnect_done); */ static void xprt_schedule_autoclose_locked(struct rpc_xprt *xprt) { - set_bit(XPRT_CLOSE_WAIT, &xprt->state); + if (test_and_set_bit(XPRT_CLOSE_WAIT, &xprt->state)) + return; if (test_and_set_bit(XPRT_LOCKED, &xprt->state) == 0) queue_work(xprtiod_workqueue, &xprt->task_cleanup); else if (xprt->snd_task && !test_bit(XPRT_SND_IS_COOKIE, &xprt->state)) diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index 7fb302e202bc..ae48c9c84ee1 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -2314,7 +2314,7 @@ static void xs_connect(struct rpc_xprt *xprt, struct rpc_task *task) WARN_ON_ONCE(!xprt_lock_connect(xprt, task, transport)); - if (transport->sock != NULL) { + if (transport->sock != NULL && !xprt_connecting(xprt)) { dprintk("RPC: xs_connect delayed xprt %p for %lu " "seconds\n", xprt, xprt->reestablish_timeout / HZ); From 4798f8058d6b6bcfaedf5560fc28a83f404cbb57 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Fri, 5 Nov 2021 12:35:26 -0400 Subject: [PATCH 63/84] NFS: Don't trace an uninitialised value If fhandle is NULL or fattr is NULL, then 'error' is uninitialised. Reported-by: Nathan Chancellor Signed-off-by: Trond Myklebust Reviewed-by: Nathan Chancellor --- fs/nfs/dir.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index 210c5945ac2b..36cb1012c7e1 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -1814,7 +1814,7 @@ no_entry: } nfs_set_verifier(dentry, dir_verifier); out_label: - trace_nfs_lookup_exit(dir, dentry, flags, error); + trace_nfs_lookup_exit(dir, dentry, flags, PTR_ERR_OR_ZERO(res)); nfs4_label_free(label); out: nfs_free_fattr(fattr); From 6659db4c59842343da46b97017574130f95143a9 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Thu, 4 Nov 2021 17:18:01 -0400 Subject: [PATCH 64/84] NFSv4: Ensure decode_compound_hdr() sanity checks the tag The server is supposed to return the same tag that the client sends in the outgoing RPC call, but we should still sanity check the length just in case. Reported-by: Signed-off-by: Trond Myklebust --- fs/nfs/nfs4xdr.c | 23 +++++++++++++---------- 1 file changed, 13 insertions(+), 10 deletions(-) diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c index a8cff19c6f00..f206d41d6bee 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -3168,20 +3168,23 @@ static int decode_opaque_inline(struct xdr_stream *xdr, unsigned int *len, char static int decode_compound_hdr(struct xdr_stream *xdr, struct compound_hdr *hdr) { - __be32 *p; + ssize_t ret; + void *ptr; + u32 tmp; - p = xdr_inline_decode(xdr, 8); - if (unlikely(!p)) + if (xdr_stream_decode_u32(xdr, &tmp) < 0) return -EIO; - hdr->status = be32_to_cpup(p++); - hdr->taglen = be32_to_cpup(p); + hdr->status = tmp; - p = xdr_inline_decode(xdr, hdr->taglen + 4); - if (unlikely(!p)) + ret = xdr_stream_decode_opaque_inline(xdr, &ptr, NFS4_OPAQUE_LIMIT); + if (ret < 0) return -EIO; - hdr->tag = (char *)p; - p += XDR_QUADLEN(hdr->taglen); - hdr->nops = be32_to_cpup(p); + hdr->taglen = ret; + hdr->tag = ptr; + + if (xdr_stream_decode_u32(xdr, &tmp) < 0) + return -EIO; + hdr->nops = tmp; if (unlikely(hdr->nops < 1)) return nfs4_stat_to_errno(hdr->status); return 0; From f114759c322edb4f293941a62748e31858852cd1 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Thu, 4 Nov 2021 17:33:36 -0400 Subject: [PATCH 65/84] NFSv4: Fix potential Oops in decode_op_map() The return value of xdr_inline_decode() is not being checked, leading to a potential Oops. Just replace the open coded array decode with the generic XDR version. Reported-by: Signed-off-by: Trond Myklebust --- fs/nfs/nfs4xdr.c | 15 ++------------- 1 file changed, 2 insertions(+), 13 deletions(-) diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c index f206d41d6bee..63a12181e6c7 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -5575,20 +5575,9 @@ static int decode_secinfo_no_name(struct xdr_stream *xdr, struct nfs4_secinfo_re static int decode_op_map(struct xdr_stream *xdr, struct nfs4_op_map *op_map) { - __be32 *p; - uint32_t bitmap_words; - unsigned int i; - - p = xdr_inline_decode(xdr, 4); - if (!p) + if (xdr_stream_decode_uint32_array(xdr, op_map->u.words, + ARRAY_SIZE(op_map->u.words)) < 0) return -EIO; - bitmap_words = be32_to_cpup(p++); - if (bitmap_words > NFS4_OP_MAP_NUM_WORDS) - return -EIO; - p = xdr_inline_decode(xdr, 4 * bitmap_words); - for (i = 0; i < bitmap_words; i++) - op_map->u.words[i] = be32_to_cpup(p++); - return 0; } From e48c81bbc188964ac3932539e53287491f2c1d87 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Fri, 5 Nov 2021 13:40:11 -0400 Subject: [PATCH 66/84] NFSv4: Remove unnecessary 'minor version' check It is completely redundant to the server capability check. Signed-off-by: Trond Myklebust --- fs/nfs/inode.c | 8 ++------ 1 file changed, 2 insertions(+), 6 deletions(-) diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index b81b2d2f47ad..b4cb75c45f90 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -378,14 +378,10 @@ void nfs_setsecurity(struct inode *inode, struct nfs_fattr *fattr, struct nfs4_label *nfs4_label_alloc(struct nfs_server *server, gfp_t flags) { - struct nfs4_label *label = NULL; - int minor_version = server->nfs_client->cl_minorversion; - - if (minor_version < 2) - return label; + struct nfs4_label *label; if (!(server->caps & NFS_CAP_SECURITY_LABEL)) - return label; + return NULL; label = kzalloc(sizeof(struct nfs4_label), flags); if (label == NULL) From 156cd28562a4e8ca454d11b234d9f634a45d6390 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Fri, 5 Nov 2021 14:23:30 -0400 Subject: [PATCH 67/84] NFS: Don't allocate nfs_fattr on the stack in __nfs42_ssc_open() The preferred behaviour is always to allocate struct nfs_fattr from the slab. Signed-off-by: Trond Myklebust --- fs/nfs/nfs4file.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) diff --git a/fs/nfs/nfs4file.c b/fs/nfs/nfs4file.c index c91565227ea2..f9f50fe1f3a4 100644 --- a/fs/nfs/nfs4file.c +++ b/fs/nfs/nfs4file.c @@ -317,7 +317,7 @@ static int read_name_gen = 1; static struct file *__nfs42_ssc_open(struct vfsmount *ss_mnt, struct nfs_fh *src_fh, nfs4_stateid *stateid) { - struct nfs_fattr fattr; + struct nfs_fattr *fattr = nfs_alloc_fattr(); struct file *filep, *res; struct nfs_server *server; struct inode *r_ino = NULL; @@ -328,9 +328,10 @@ static struct file *__nfs42_ssc_open(struct vfsmount *ss_mnt, server = NFS_SERVER(ss_mnt->mnt_root->d_inode); - nfs_fattr_init(&fattr); + if (!fattr) + return ERR_PTR(-ENOMEM); - status = nfs4_proc_getattr(server, src_fh, &fattr, NULL, NULL); + status = nfs4_proc_getattr(server, src_fh, fattr, NULL, NULL); if (status < 0) { res = ERR_PTR(status); goto out; @@ -343,7 +344,7 @@ static struct file *__nfs42_ssc_open(struct vfsmount *ss_mnt, goto out; snprintf(read_name, len, SSC_READ_NAME_BODY, read_name_gen++); - r_ino = nfs_fhget(ss_mnt->mnt_root->d_inode->i_sb, src_fh, &fattr, + r_ino = nfs_fhget(ss_mnt->mnt_root->d_inode->i_sb, src_fh, fattr, NULL); if (IS_ERR(r_ino)) { res = ERR_CAST(r_ino); @@ -388,6 +389,7 @@ static struct file *__nfs42_ssc_open(struct vfsmount *ss_mnt, out_free_name: kfree(read_name); out: + nfs_free_fattr(fattr); return res; out_stateowner: nfs4_put_state_owner(sp); From aa97a3ef15c303697ca58340938a5c385be00cf7 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Fri, 5 Nov 2021 14:32:28 -0400 Subject: [PATCH 68/84] NFSv4.2: alloc_file_pseudo() takes an open flag, not an f_mode Signed-off-by: Trond Myklebust --- fs/nfs/nfs4file.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/fs/nfs/nfs4file.c b/fs/nfs/nfs4file.c index f9f50fe1f3a4..92a1b992a141 100644 --- a/fs/nfs/nfs4file.c +++ b/fs/nfs/nfs4file.c @@ -351,13 +351,12 @@ static struct file *__nfs42_ssc_open(struct vfsmount *ss_mnt, goto out_free_name; } - filep = alloc_file_pseudo(r_ino, ss_mnt, read_name, FMODE_READ, + filep = alloc_file_pseudo(r_ino, ss_mnt, read_name, O_RDONLY, r_ino->i_fop); if (IS_ERR(filep)) { res = ERR_CAST(filep); goto out_free_name; } - filep->f_mode |= FMODE_READ; ctx = alloc_nfs_open_context(filep->f_path.dentry, filep->f_mode, filep); From d4a95a7e5a4d3b68b26f70668cf77324a11b5718 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Thu, 4 Nov 2021 18:03:26 -0400 Subject: [PATCH 69/84] NFS: Always initialise fattr->label in nfs_fattr_alloc() We're about to add a check in nfs_free_fattr() for whether or not the label is non-zero. Signed-off-by: Trond Myklebust --- fs/nfs/inode.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index b4cb75c45f90..0b5b1e44b2c4 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -1598,8 +1598,10 @@ struct nfs_fattr *nfs_alloc_fattr(void) struct nfs_fattr *fattr; fattr = kmalloc(sizeof(*fattr), GFP_NOFS); - if (fattr != NULL) + if (fattr != NULL) { nfs_fattr_init(fattr); + fattr->label = NULL; + } return fattr; } EXPORT_SYMBOL_GPL(nfs_alloc_fattr); From d755ad8dc752d44545613ea04d660aed674e540d Mon Sep 17 00:00:00 2001 From: Anna Schumaker Date: Fri, 22 Oct 2021 13:11:00 -0400 Subject: [PATCH 70/84] NFS: Create a new nfs_alloc_fattr_with_label() function For creating fattrs with the label field already allocated for us. I also update nfs_free_fattr() to free the label in the end. Signed-off-by: Anna Schumaker Signed-off-by: Trond Myklebust --- fs/nfs/getroot.c | 17 ++++++----------- fs/nfs/inode.c | 17 +++++++++++++++++ fs/nfs/internal.h | 9 --------- include/linux/nfs_fs.h | 13 +++++++++++++ 4 files changed, 36 insertions(+), 20 deletions(-) diff --git a/fs/nfs/getroot.c b/fs/nfs/getroot.c index 59355c106ece..7604cb6a0ac2 100644 --- a/fs/nfs/getroot.c +++ b/fs/nfs/getroot.c @@ -80,18 +80,15 @@ int nfs_get_root(struct super_block *s, struct fs_context *fc) goto out; /* get the actual root for this mount */ - fsinfo.fattr = nfs_alloc_fattr(); + fsinfo.fattr = nfs_alloc_fattr_with_label(server); if (fsinfo.fattr == NULL) goto out_name; - fsinfo.fattr->label = nfs4_label_alloc(server, GFP_KERNEL); - if (IS_ERR(fsinfo.fattr->label)) - goto out_fattr; error = server->nfs_client->rpc_ops->getroot(server, ctx->mntfh, &fsinfo); if (error < 0) { dprintk("nfs_get_root: getattr error = %d\n", -error); nfs_errorf(fc, "NFS: Couldn't getattr on root"); - goto out_label; + goto out_fattr; } inode = nfs_fhget(s, ctx->mntfh, fsinfo.fattr, NULL); @@ -99,12 +96,12 @@ int nfs_get_root(struct super_block *s, struct fs_context *fc) dprintk("nfs_get_root: get root inode failed\n"); error = PTR_ERR(inode); nfs_errorf(fc, "NFS: Couldn't get root inode"); - goto out_label; + goto out_fattr; } error = nfs_superblock_set_dummy_root(s, inode); if (error != 0) - goto out_label; + goto out_fattr; /* root dentries normally start off anonymous and get spliced in later * if the dentry tree reaches them; however if the dentry already @@ -115,7 +112,7 @@ int nfs_get_root(struct super_block *s, struct fs_context *fc) dprintk("nfs_get_root: get root dentry failed\n"); error = PTR_ERR(root); nfs_errorf(fc, "NFS: Couldn't get root dentry"); - goto out_label; + goto out_fattr; } security_d_instantiate(root, inode); @@ -154,8 +151,6 @@ int nfs_get_root(struct super_block *s, struct fs_context *fc) nfs_setsecurity(inode, fsinfo.fattr, fsinfo.fattr->label); error = 0; -out_label: - nfs4_label_free(fsinfo.fattr->label); out_fattr: nfs_free_fattr(fsinfo.fattr); out_name: @@ -165,5 +160,5 @@ out: error_splat_root: dput(fc->root); fc->root = NULL; - goto out_label; + goto out_fattr; } diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index 0b5b1e44b2c4..84c7efa2ea87 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -1606,6 +1606,23 @@ struct nfs_fattr *nfs_alloc_fattr(void) } EXPORT_SYMBOL_GPL(nfs_alloc_fattr); +struct nfs_fattr *nfs_alloc_fattr_with_label(struct nfs_server *server) +{ + struct nfs_fattr *fattr = nfs_alloc_fattr(); + + if (!fattr) + return NULL; + + fattr->label = nfs4_label_alloc(server, GFP_NOFS); + if (IS_ERR(fattr->label)) { + kfree(fattr); + return NULL; + } + + return fattr; +} +EXPORT_SYMBOL_GPL(nfs_alloc_fattr_with_label); + struct nfs_fh *nfs_alloc_fhandle(void) { struct nfs_fh *fh; diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h index 123078c76495..12f6acb483bb 100644 --- a/fs/nfs/internal.h +++ b/fs/nfs/internal.h @@ -342,14 +342,6 @@ nfs4_label_copy(struct nfs4_label *dst, struct nfs4_label *src) return dst; } -static inline void nfs4_label_free(struct nfs4_label *label) -{ - if (label) { - kfree(label->label); - kfree(label); - } - return; -} static inline void nfs_zap_label_cache_locked(struct nfs_inode *nfsi) { @@ -358,7 +350,6 @@ static inline void nfs_zap_label_cache_locked(struct nfs_inode *nfsi) } #else static inline struct nfs4_label *nfs4_label_alloc(struct nfs_server *server, gfp_t flags) { return NULL; } -static inline void nfs4_label_free(void *label) {} static inline void nfs_zap_label_cache_locked(struct nfs_inode *nfsi) { } diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index 739ca1ef934f..88c3aed8ad39 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -426,9 +426,22 @@ extern void nfs_fattr_set_barrier(struct nfs_fattr *fattr); extern unsigned long nfs_inc_attr_generation_counter(void); extern struct nfs_fattr *nfs_alloc_fattr(void); +extern struct nfs_fattr *nfs_alloc_fattr_with_label(struct nfs_server *server); + +static inline void nfs4_label_free(struct nfs4_label *label) +{ +#ifdef CONFIG_NFS_V4_SECURITY_LABEL + if (label) { + kfree(label->label); + kfree(label); + } +#endif +} static inline void nfs_free_fattr(const struct nfs_fattr *fattr) { + if (fattr) + nfs4_label_free(fattr->label); kfree(fattr); } From b1db9a401d464d526d5941f0544e7c9ea37fa731 Mon Sep 17 00:00:00 2001 From: Anna Schumaker Date: Fri, 22 Oct 2021 13:11:01 -0400 Subject: [PATCH 71/84] NFS: Remove the nfs4_label from the nfs_entry struct And instead allocate the fattr using nfs_alloc_fattr_with_label() Signed-off-by: Anna Schumaker Signed-off-by: Trond Myklebust --- fs/nfs/dir.c | 21 +++++++-------------- fs/nfs/nfs4xdr.c | 2 +- include/linux/nfs_xdr.h | 1 - 3 files changed, 8 insertions(+), 16 deletions(-) diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index 36cb1012c7e1..92530c3c1694 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -682,7 +682,8 @@ again: nfs_set_verifier(dentry, dir_verifier); status = nfs_refresh_inode(d_inode(dentry), entry->fattr); if (!status) - nfs_setsecurity(d_inode(dentry), entry->fattr, entry->label); + nfs_setsecurity(d_inode(dentry), entry->fattr, + entry->fattr->label); goto out; } else { d_invalidate(dentry); @@ -696,7 +697,7 @@ again: goto out; } - inode = nfs_fhget(dentry->d_sb, entry->fh, entry->fattr, entry->label); + inode = nfs_fhget(dentry->d_sb, entry->fh, entry->fattr, entry->fattr->label); alias = d_splice_alias(inode, dentry); d_lookup_done(dentry); if (alias) { @@ -732,8 +733,8 @@ static int nfs_readdir_page_filler(struct nfs_readdir_descriptor *desc, xdr_set_scratch_page(&stream, scratch); do { - if (entry->label) - entry->label->len = NFS4_MAXLABELLEN; + if (entry->fattr->label) + entry->fattr->label->len = NFS4_MAXLABELLEN; status = xdr_decode(desc, entry, &stream); if (status != 0) @@ -838,21 +839,15 @@ static int nfs_readdir_xdr_to_array(struct nfs_readdir_descriptor *desc, return -ENOMEM; entry->cookie = nfs_readdir_page_last_cookie(page); entry->fh = nfs_alloc_fhandle(); - entry->fattr = nfs_alloc_fattr(); + entry->fattr = nfs_alloc_fattr_with_label(NFS_SERVER(inode)); entry->server = NFS_SERVER(inode); if (entry->fh == NULL || entry->fattr == NULL) goto out; - entry->label = nfs4_label_alloc(NFS_SERVER(inode), GFP_NOWAIT); - if (IS_ERR(entry->label)) { - status = PTR_ERR(entry->label); - goto out; - } - array_size = (dtsize + PAGE_SIZE - 1) >> PAGE_SHIFT; pages = nfs_readdir_alloc_pages(array_size); if (!pages) - goto out_release_label; + goto out; do { unsigned int pglen; @@ -875,8 +870,6 @@ static int nfs_readdir_xdr_to_array(struct nfs_readdir_descriptor *desc, } while (!status && nfs_readdir_page_needs_filling(page)); nfs_readdir_free_pages(pages, array_size); -out_release_label: - nfs4_label_free(entry->label); out: nfs_free_fattr(entry->fattr); nfs_free_fhandle(entry->fh); diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c index 63a12181e6c7..fba89f82e7b7 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -7467,7 +7467,7 @@ int nfs4_decode_dirent(struct xdr_stream *xdr, struct nfs_entry *entry, return -EAGAIN; if (decode_getfattr_attrs(xdr, bitmap, entry->fattr, entry->fh, - NULL, entry->label, entry->server) < 0) + NULL, entry->fattr->label, entry->server) < 0) return -EAGAIN; if (entry->fattr->valid & NFS_ATTR_FATTR_MOUNTED_ON_FILEID) entry->ino = entry->fattr->mounted_on_fileid; diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index e9698b6278a5..9960f6628066 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -753,7 +753,6 @@ struct nfs_entry { int eof; struct nfs_fh * fh; struct nfs_fattr * fattr; - struct nfs4_label *label; unsigned char d_type; struct nfs_server * server; }; From 68be1742c22983558f0f148a4467eb9127d56b86 Mon Sep 17 00:00:00 2001 From: Anna Schumaker Date: Fri, 22 Oct 2021 13:11:02 -0400 Subject: [PATCH 72/84] NFS: Remove the nfs4_label from the nfs4_create_res struct Instead, use the label embedded in the attached fattr. Signed-off-by: Anna Schumaker Signed-off-by: Trond Myklebust --- fs/nfs/nfs4proc.c | 12 +++++------- fs/nfs/nfs4xdr.c | 2 +- include/linux/nfs_xdr.h | 1 - 3 files changed, 6 insertions(+), 9 deletions(-) diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 127388fabda8..2e07550dd0d4 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -4850,7 +4850,6 @@ struct nfs4_createdata { struct nfs4_create_res res; struct nfs_fh fh; struct nfs_fattr fattr; - struct nfs4_label *label; }; static struct nfs4_createdata *nfs4_alloc_createdata(struct inode *dir, @@ -4862,8 +4861,8 @@ static struct nfs4_createdata *nfs4_alloc_createdata(struct inode *dir, if (data != NULL) { struct nfs_server *server = NFS_SERVER(dir); - data->label = nfs4_label_alloc(server, GFP_KERNEL); - if (IS_ERR(data->label)) + data->fattr.label = nfs4_label_alloc(server, GFP_KERNEL); + if (IS_ERR(data->fattr.label)) goto out_free; data->msg.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_CREATE]; @@ -4874,12 +4873,11 @@ static struct nfs4_createdata *nfs4_alloc_createdata(struct inode *dir, data->arg.name = name; data->arg.attrs = sattr; data->arg.ftype = ftype; - data->arg.bitmask = nfs4_bitmask(server, data->label); + data->arg.bitmask = nfs4_bitmask(server, data->fattr.label); data->arg.umask = current_umask(); data->res.server = server; data->res.fh = &data->fh; data->res.fattr = &data->fattr; - data->res.label = data->label; nfs_fattr_init(data->res.fattr); } return data; @@ -4901,14 +4899,14 @@ static int nfs4_do_create(struct inode *dir, struct dentry *dentry, struct nfs4_ data->res.fattr->time_start, NFS_INO_INVALID_DATA); spin_unlock(&dir->i_lock); - status = nfs_instantiate(dentry, data->res.fh, data->res.fattr, data->res.label); + status = nfs_instantiate(dentry, data->res.fh, data->res.fattr, data->res.fattr->label); } return status; } static void nfs4_free_createdata(struct nfs4_createdata *data) { - nfs4_label_free(data->label); + nfs4_label_free(data->fattr.label); kfree(data); } diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c index fba89f82e7b7..38c74833f263 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -6353,7 +6353,7 @@ static int nfs4_xdr_dec_create(struct rpc_rqst *rqstp, struct xdr_stream *xdr, status = decode_getfh(xdr, res->fh); if (status) goto out; - decode_getfattr_label(xdr, res->fattr, res->label, res->server); + decode_getfattr_label(xdr, res->fattr, res->fattr->label, res->server); out: return status; } diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index 9960f6628066..5aba81b74c98 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -1040,7 +1040,6 @@ struct nfs4_create_res { const struct nfs_server * server; struct nfs_fh * fh; struct nfs_fattr * fattr; - struct nfs4_label *label; struct nfs4_change_info dir_cinfo; }; From aa7ca3b2de190675543d84adaa1ff74e7867c76f Mon Sep 17 00:00:00 2001 From: Anna Schumaker Date: Fri, 22 Oct 2021 13:11:03 -0400 Subject: [PATCH 73/84] NFS: Remove the nfs4_label from the nfs4_link_res struct Again, use the fattr's label field instead. Signed-off-by: Anna Schumaker Signed-off-by: Trond Myklebust --- fs/nfs/nfs4proc.c | 16 +++------------- fs/nfs/nfs4xdr.c | 2 +- include/linux/nfs_xdr.h | 1 - 3 files changed, 4 insertions(+), 15 deletions(-) diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 2e07550dd0d4..bde5b5723046 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -4790,7 +4790,6 @@ static int _nfs4_proc_link(struct inode *inode, struct inode *dir, const struct }; struct nfs4_link_res res = { .server = server, - .label = NULL, }; struct rpc_message msg = { .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_LINK], @@ -4799,18 +4798,12 @@ static int _nfs4_proc_link(struct inode *inode, struct inode *dir, const struct }; int status = -ENOMEM; - res.fattr = nfs_alloc_fattr(); + res.fattr = nfs_alloc_fattr_with_label(server); if (res.fattr == NULL) goto out; - res.label = nfs4_label_alloc(server, GFP_KERNEL); - if (IS_ERR(res.label)) { - status = PTR_ERR(res.label); - goto out; - } - nfs4_inode_make_writeable(inode); - nfs4_bitmap_copy_adjust(bitmask, nfs4_bitmask(server, res.label), inode, + nfs4_bitmap_copy_adjust(bitmask, nfs4_bitmask(server, res.fattr->label), inode, NFS_INO_INVALID_CHANGE); status = nfs4_call_sync(server->client, server, &msg, &arg.seq_args, &res.seq_res, 1); if (!status) { @@ -4819,12 +4812,9 @@ static int _nfs4_proc_link(struct inode *inode, struct inode *dir, const struct nfs4_inc_nlink(inode); status = nfs_post_op_update_inode(inode, res.fattr); if (!status) - nfs_setsecurity(inode, res.fattr, res.label); + nfs_setsecurity(inode, res.fattr, res.fattr->label); } - - nfs4_label_free(res.label); - out: nfs_free_fattr(res.fattr); return status; diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c index 38c74833f263..4c9d66fac3fd 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -6323,7 +6323,7 @@ static int nfs4_xdr_dec_link(struct rpc_rqst *rqstp, struct xdr_stream *xdr, status = decode_restorefh(xdr); if (status) goto out; - decode_getfattr_label(xdr, res->fattr, res->label, res->server); + decode_getfattr_label(xdr, res->fattr, res->fattr->label, res->server); out: return status; } diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index 5aba81b74c98..d55bf3fd5167 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -1079,7 +1079,6 @@ struct nfs4_link_res { struct nfs4_sequence_res seq_res; const struct nfs_server * server; struct nfs_fattr * fattr; - struct nfs4_label *label; struct nfs4_change_info cinfo; struct nfs_fattr * dir_attr; }; From 9558a007dbc383d48e7f5a123d0b5ff656c71068 Mon Sep 17 00:00:00 2001 From: Anna Schumaker Date: Fri, 22 Oct 2021 13:11:04 -0400 Subject: [PATCH 74/84] NFS: Remove the label from the nfs4_lookup_res struct And usethe fattr's label field instead. I also adjust function calls to remove labels along the way. Signed-off-by: Anna Schumaker Signed-off-by: Trond Myklebust --- fs/nfs/dir.c | 34 ++++++++++++---------------------- fs/nfs/namespace.c | 3 +-- fs/nfs/nfs3proc.c | 3 +-- fs/nfs/nfs4proc.c | 16 +++++++--------- fs/nfs/nfs4xdr.c | 4 ++-- fs/nfs/proc.c | 3 +-- include/linux/nfs_xdr.h | 4 +--- 7 files changed, 25 insertions(+), 42 deletions(-) diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index 92530c3c1694..8a327971d485 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -1494,19 +1494,17 @@ nfs_lookup_revalidate_dentry(struct inode *dir, struct dentry *dentry, { struct nfs_fh *fhandle; struct nfs_fattr *fattr; - struct nfs4_label *label; unsigned long dir_verifier; int ret; ret = -ENOMEM; fhandle = nfs_alloc_fhandle(); - fattr = nfs_alloc_fattr(); - label = nfs4_label_alloc(NFS_SERVER(inode), GFP_KERNEL); - if (fhandle == NULL || fattr == NULL || IS_ERR(label)) + fattr = nfs_alloc_fattr_with_label(NFS_SERVER(inode)); + if (fhandle == NULL || fattr == NULL) goto out; dir_verifier = nfs_save_change_attribute(dir); - ret = NFS_PROTO(dir)->lookup(dir, dentry, fhandle, fattr, label); + ret = NFS_PROTO(dir)->lookup(dir, dentry, fhandle, fattr); if (ret < 0) { switch (ret) { case -ESTALE: @@ -1525,7 +1523,7 @@ nfs_lookup_revalidate_dentry(struct inode *dir, struct dentry *dentry, if (nfs_refresh_inode(inode, fattr) < 0) goto out; - nfs_setsecurity(inode, fattr, label); + nfs_setsecurity(inode, fattr, fattr->label); nfs_set_verifier(dentry, dir_verifier); /* set a readdirplus hint that we had a cache miss */ @@ -1534,7 +1532,6 @@ nfs_lookup_revalidate_dentry(struct inode *dir, struct dentry *dentry, out: nfs_free_fattr(fattr); nfs_free_fhandle(fhandle); - nfs4_label_free(label); /* * If the lookup failed despite the dentry change attribute being @@ -1754,7 +1751,6 @@ struct dentry *nfs_lookup(struct inode *dir, struct dentry * dentry, unsigned in struct inode *inode = NULL; struct nfs_fh *fhandle = NULL; struct nfs_fattr *fattr = NULL; - struct nfs4_label *label = NULL; unsigned long dir_verifier; int error; @@ -1773,27 +1769,23 @@ struct dentry *nfs_lookup(struct inode *dir, struct dentry * dentry, unsigned in res = ERR_PTR(-ENOMEM); fhandle = nfs_alloc_fhandle(); - fattr = nfs_alloc_fattr(); + fattr = nfs_alloc_fattr_with_label(NFS_SERVER(dir)); if (fhandle == NULL || fattr == NULL) goto out; - label = nfs4_label_alloc(NFS_SERVER(dir), GFP_NOWAIT); - if (IS_ERR(label)) - goto out; - dir_verifier = nfs_save_change_attribute(dir); trace_nfs_lookup_enter(dir, dentry, flags); - error = NFS_PROTO(dir)->lookup(dir, dentry, fhandle, fattr, label); + error = NFS_PROTO(dir)->lookup(dir, dentry, fhandle, fattr); if (error == -ENOENT) goto no_entry; if (error < 0) { res = ERR_PTR(error); - goto out_label; + goto out; } - inode = nfs_fhget(dentry->d_sb, fhandle, fattr, label); + inode = nfs_fhget(dentry->d_sb, fhandle, fattr, fattr->label); res = ERR_CAST(inode); if (IS_ERR(res)) - goto out_label; + goto out; /* Notify readdir to use READDIRPLUS */ nfs_force_use_readdirplus(dir); @@ -1802,14 +1794,12 @@ no_entry: res = d_splice_alias(inode, dentry); if (res != NULL) { if (IS_ERR(res)) - goto out_label; + goto out; dentry = res; } nfs_set_verifier(dentry, dir_verifier); -out_label: - trace_nfs_lookup_exit(dir, dentry, flags, PTR_ERR_OR_ZERO(res)); - nfs4_label_free(label); out: + trace_nfs_lookup_exit(dir, dentry, flags, PTR_ERR_OR_ZERO(res)); nfs_free_fattr(fattr); nfs_free_fhandle(fhandle); return res; @@ -2058,7 +2048,7 @@ nfs_add_or_obtain(struct dentry *dentry, struct nfs_fh *fhandle, d_drop(dentry); if (fhandle->size == 0) { - error = NFS_PROTO(dir)->lookup(dir, dentry, fhandle, fattr, NULL); + error = NFS_PROTO(dir)->lookup(dir, dentry, fhandle, fattr); if (error) goto out_error; } diff --git a/fs/nfs/namespace.c b/fs/nfs/namespace.c index bc0c698f3350..3295af4110f1 100644 --- a/fs/nfs/namespace.c +++ b/fs/nfs/namespace.c @@ -308,8 +308,7 @@ int nfs_submount(struct fs_context *fc, struct nfs_server *server) /* Look it up again to get its attributes */ err = server->nfs_client->rpc_ops->lookup(d_inode(parent), dentry, - ctx->mntfh, ctx->clone_data.fattr, - NULL); + ctx->mntfh, ctx->clone_data.fattr); dput(parent); if (err != 0) return err; diff --git a/fs/nfs/nfs3proc.c b/fs/nfs/nfs3proc.c index f7524310ddf4..717eb651f0fd 100644 --- a/fs/nfs/nfs3proc.c +++ b/fs/nfs/nfs3proc.c @@ -193,8 +193,7 @@ __nfs3_proc_lookup(struct inode *dir, const char *name, size_t len, static int nfs3_proc_lookup(struct inode *dir, struct dentry *dentry, - struct nfs_fh *fhandle, struct nfs_fattr *fattr, - struct nfs4_label *label) + struct nfs_fh *fhandle, struct nfs_fattr *fattr) { unsigned short task_flags = 0; diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index bde5b5723046..7af73fd34b22 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -4294,7 +4294,7 @@ nfs4_proc_setattr(struct dentry *dentry, struct nfs_fattr *fattr, static int _nfs4_proc_lookup(struct rpc_clnt *clnt, struct inode *dir, struct dentry *dentry, struct nfs_fh *fhandle, - struct nfs_fattr *fattr, struct nfs4_label *label) + struct nfs_fattr *fattr) { struct nfs_server *server = NFS_SERVER(dir); int status; @@ -4306,7 +4306,6 @@ static int _nfs4_proc_lookup(struct rpc_clnt *clnt, struct inode *dir, struct nfs4_lookup_res res = { .server = server, .fattr = fattr, - .label = label, .fh = fhandle, }; struct rpc_message msg = { @@ -4323,7 +4322,7 @@ static int _nfs4_proc_lookup(struct rpc_clnt *clnt, struct inode *dir, if (nfs_lookup_is_soft_revalidate(dentry)) task_flags |= RPC_TASK_TIMEOUT; - args.bitmask = nfs4_bitmask(server, label); + args.bitmask = nfs4_bitmask(server, fattr->label); nfs_fattr_init(fattr); @@ -4345,7 +4344,7 @@ static void nfs_fixup_secinfo_attributes(struct nfs_fattr *fattr) static int nfs4_proc_lookup_common(struct rpc_clnt **clnt, struct inode *dir, struct dentry *dentry, struct nfs_fh *fhandle, - struct nfs_fattr *fattr, struct nfs4_label *label) + struct nfs_fattr *fattr) { struct nfs4_exception exception = { .interruptible = true, @@ -4354,7 +4353,7 @@ static int nfs4_proc_lookup_common(struct rpc_clnt **clnt, struct inode *dir, const struct qstr *name = &dentry->d_name; int err; do { - err = _nfs4_proc_lookup(client, dir, dentry, fhandle, fattr, label); + err = _nfs4_proc_lookup(client, dir, dentry, fhandle, fattr); trace_nfs4_lookup(dir, name, err); switch (err) { case -NFS4ERR_BADNAME: @@ -4390,13 +4389,12 @@ out: } static int nfs4_proc_lookup(struct inode *dir, struct dentry *dentry, - struct nfs_fh *fhandle, struct nfs_fattr *fattr, - struct nfs4_label *label) + struct nfs_fh *fhandle, struct nfs_fattr *fattr) { int status; struct rpc_clnt *client = NFS_CLIENT(dir); - status = nfs4_proc_lookup_common(&client, dir, dentry, fhandle, fattr, label); + status = nfs4_proc_lookup_common(&client, dir, dentry, fhandle, fattr); if (client != NFS_CLIENT(dir)) { rpc_shutdown_client(client); nfs_fixup_secinfo_attributes(fattr); @@ -4411,7 +4409,7 @@ nfs4_proc_lookup_mountpoint(struct inode *dir, struct dentry *dentry, struct rpc_clnt *client = NFS_CLIENT(dir); int status; - status = nfs4_proc_lookup_common(&client, dir, dentry, fhandle, fattr, NULL); + status = nfs4_proc_lookup_common(&client, dir, dentry, fhandle, fattr); if (status < 0) return ERR_PTR(status); return (client == NFS_CLIENT(dir)) ? rpc_clone_client(client) : client; diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c index 4c9d66fac3fd..960e2b2a7a58 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -6171,7 +6171,7 @@ static int nfs4_xdr_dec_lookup(struct rpc_rqst *rqstp, struct xdr_stream *xdr, status = decode_getfh(xdr, res->fh); if (status) goto out; - status = decode_getfattr_label(xdr, res->fattr, res->label, res->server); + status = decode_getfattr_label(xdr, res->fattr, res->fattr->label, res->server); out: return status; } @@ -6229,7 +6229,7 @@ static int nfs4_xdr_dec_lookup_root(struct rpc_rqst *rqstp, status = decode_getfh(xdr, res->fh); if (status == 0) status = decode_getfattr_label(xdr, res->fattr, - res->label, res->server); + res->fattr->label, res->server); out: return status; } diff --git a/fs/nfs/proc.c b/fs/nfs/proc.c index ecc4e717808c..98a8901ede2e 100644 --- a/fs/nfs/proc.c +++ b/fs/nfs/proc.c @@ -154,8 +154,7 @@ nfs_proc_setattr(struct dentry *dentry, struct nfs_fattr *fattr, static int nfs_proc_lookup(struct inode *dir, struct dentry *dentry, - struct nfs_fh *fhandle, struct nfs_fattr *fattr, - struct nfs4_label *label) + struct nfs_fh *fhandle, struct nfs_fattr *fattr) { struct nfs_diropargs arg = { .fh = NFS_FH(dir), diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index d55bf3fd5167..95219d5a8668 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -1095,7 +1095,6 @@ struct nfs4_lookup_res { const struct nfs_server * server; struct nfs_fattr * fattr; struct nfs_fh * fh; - struct nfs4_label *label; }; struct nfs4_lookupp_arg { @@ -1740,8 +1739,7 @@ struct nfs_rpc_ops { int (*setattr) (struct dentry *, struct nfs_fattr *, struct iattr *); int (*lookup) (struct inode *, struct dentry *, - struct nfs_fh *, struct nfs_fattr *, - struct nfs4_label *); + struct nfs_fh *, struct nfs_fattr *); int (*lookupp) (struct inode *, struct nfs_fh *, struct nfs_fattr *, struct nfs4_label *); int (*access) (struct inode *, struct nfs_access_entry *); From ba4bc8dc4d937df2b407393435a302550be0ad82 Mon Sep 17 00:00:00 2001 From: Anna Schumaker Date: Fri, 22 Oct 2021 13:11:05 -0400 Subject: [PATCH 75/84] NFS: Remove the nfs4_label from the nfs4_lookupp_res struct Signed-off-by: Anna Schumaker Signed-off-by: Trond Myklebust --- fs/nfs/export.c | 30 +++++++++--------------------- fs/nfs/nfs3proc.c | 2 +- fs/nfs/nfs4proc.c | 10 ++++------ fs/nfs/nfs4xdr.c | 2 +- include/linux/nfs_xdr.h | 3 +-- 5 files changed, 16 insertions(+), 31 deletions(-) diff --git a/fs/nfs/export.c b/fs/nfs/export.c index d772c20bbfd1..895b404888dd 100644 --- a/fs/nfs/export.c +++ b/fs/nfs/export.c @@ -131,7 +131,6 @@ nfs_get_parent(struct dentry *dentry) struct super_block *sb = inode->i_sb; struct nfs_server *server = NFS_SB(sb); struct nfs_fattr *fattr = NULL; - struct nfs4_label *label = NULL; struct dentry *parent; struct nfs_rpc_ops const *ops = server->nfs_client->rpc_ops; struct nfs_fh fh; @@ -139,31 +138,20 @@ nfs_get_parent(struct dentry *dentry) if (!ops->lookupp) return ERR_PTR(-EACCES); - fattr = nfs_alloc_fattr(); - if (fattr == NULL) { - parent = ERR_PTR(-ENOMEM); + fattr = nfs_alloc_fattr_with_label(server); + if (fattr == NULL) + return ERR_PTR(-ENOMEM); + + ret = ops->lookupp(inode, &fh, fattr); + if (ret) { + parent = ERR_PTR(ret); goto out; } - label = nfs4_label_alloc(server, GFP_KERNEL); - if (IS_ERR(label)) { - parent = ERR_CAST(label); - goto out_free_fattr; - } - - ret = ops->lookupp(inode, &fh, fattr, label); - if (ret) { - parent = ERR_PTR(ret); - goto out_free_label; - } - - pinode = nfs_fhget(sb, &fh, fattr, label); + pinode = nfs_fhget(sb, &fh, fattr, fattr->label); parent = d_obtain_alias(pinode); -out_free_label: - nfs4_label_free(label); -out_free_fattr: - nfs_free_fattr(fattr); out: + nfs_free_fattr(fattr); return parent; } diff --git a/fs/nfs/nfs3proc.c b/fs/nfs/nfs3proc.c index 717eb651f0fd..516f3340b226 100644 --- a/fs/nfs/nfs3proc.c +++ b/fs/nfs/nfs3proc.c @@ -208,7 +208,7 @@ nfs3_proc_lookup(struct inode *dir, struct dentry *dentry, } static int nfs3_proc_lookupp(struct inode *inode, struct nfs_fh *fhandle, - struct nfs_fattr *fattr, struct nfs4_label *label) + struct nfs_fattr *fattr) { const char dotdot[] = ".."; const size_t len = strlen(dotdot); diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 7af73fd34b22..caa5a1467f94 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -4416,8 +4416,7 @@ nfs4_proc_lookup_mountpoint(struct inode *dir, struct dentry *dentry, } static int _nfs4_proc_lookupp(struct inode *inode, - struct nfs_fh *fhandle, struct nfs_fattr *fattr, - struct nfs4_label *label) + struct nfs_fh *fhandle, struct nfs_fattr *fattr) { struct rpc_clnt *clnt = NFS_CLIENT(inode); struct nfs_server *server = NFS_SERVER(inode); @@ -4429,7 +4428,6 @@ static int _nfs4_proc_lookupp(struct inode *inode, struct nfs4_lookupp_res res = { .server = server, .fattr = fattr, - .label = label, .fh = fhandle, }; struct rpc_message msg = { @@ -4442,7 +4440,7 @@ static int _nfs4_proc_lookupp(struct inode *inode, if (NFS_SERVER(inode)->flags & NFS_MOUNT_SOFTREVAL) task_flags |= RPC_TASK_TIMEOUT; - args.bitmask = nfs4_bitmask(server, label); + args.bitmask = nfs4_bitmask(server, fattr->label); nfs_fattr_init(fattr); @@ -4454,14 +4452,14 @@ static int _nfs4_proc_lookupp(struct inode *inode, } static int nfs4_proc_lookupp(struct inode *inode, struct nfs_fh *fhandle, - struct nfs_fattr *fattr, struct nfs4_label *label) + struct nfs_fattr *fattr) { struct nfs4_exception exception = { .interruptible = true, }; int err; do { - err = _nfs4_proc_lookupp(inode, fhandle, fattr, label); + err = _nfs4_proc_lookupp(inode, fhandle, fattr); trace_nfs4_lookupp(inode, err); err = nfs4_handle_exception(NFS_SERVER(inode), err, &exception); diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c index 960e2b2a7a58..0044747f9314 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -6201,7 +6201,7 @@ static int nfs4_xdr_dec_lookupp(struct rpc_rqst *rqstp, struct xdr_stream *xdr, status = decode_getfh(xdr, res->fh); if (status) goto out; - status = decode_getfattr_label(xdr, res->fattr, res->label, res->server); + status = decode_getfattr_label(xdr, res->fattr, res->fattr->label, res->server); out: return status; } diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index 95219d5a8668..f0a685d9b8bd 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -1108,7 +1108,6 @@ struct nfs4_lookupp_res { const struct nfs_server *server; struct nfs_fattr *fattr; struct nfs_fh *fh; - struct nfs4_label *label; }; struct nfs4_lookup_root_arg { @@ -1741,7 +1740,7 @@ struct nfs_rpc_ops { int (*lookup) (struct inode *, struct dentry *, struct nfs_fh *, struct nfs_fattr *); int (*lookupp) (struct inode *, struct nfs_fh *, - struct nfs_fattr *, struct nfs4_label *); + struct nfs_fattr *); int (*access) (struct inode *, struct nfs_access_entry *); int (*readlink)(struct inode *, struct page *, unsigned int, unsigned int); From 76baa2b29c7161bc65a3051d311297b7d7fc827a Mon Sep 17 00:00:00 2001 From: Anna Schumaker Date: Fri, 22 Oct 2021 13:11:06 -0400 Subject: [PATCH 76/84] NFS: Remove the f_label from the nfs4_opendata and nfs_openres Signed-off-by: Anna Schumaker Signed-off-by: Trond Myklebust --- fs/nfs/nfs4_fs.h | 1 - fs/nfs/nfs4proc.c | 35 +++++++++++------------------------ fs/nfs/nfs4xdr.c | 2 +- include/linux/nfs_xdr.h | 1 - 4 files changed, 12 insertions(+), 27 deletions(-) diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h index ba78df4b13d9..b621e29e6187 100644 --- a/fs/nfs/nfs4_fs.h +++ b/fs/nfs/nfs4_fs.h @@ -234,7 +234,6 @@ struct nfs4_opendata { struct nfs4_string group_name; struct nfs4_label *a_label; struct nfs_fattr f_attr; - struct nfs4_label *f_label; struct dentry *dir; struct dentry *dentry; struct nfs4_state_owner *owner; diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index caa5a1467f94..cb0613e0ef8f 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -1329,7 +1329,6 @@ nfs4_map_atomic_open_claim(struct nfs_server *server, static void nfs4_init_opendata_res(struct nfs4_opendata *p) { p->o_res.f_attr = &p->f_attr; - p->o_res.f_label = p->f_label; p->o_res.seqid = p->o_arg.seqid; p->c_res.seqid = p->c_arg.seqid; p->o_res.server = p->o_arg.server; @@ -1355,8 +1354,8 @@ static struct nfs4_opendata *nfs4_opendata_alloc(struct dentry *dentry, if (p == NULL) goto err; - p->f_label = nfs4_label_alloc(server, gfp_mask); - if (IS_ERR(p->f_label)) + p->f_attr.label = nfs4_label_alloc(server, gfp_mask); + if (IS_ERR(p->f_attr.label)) goto err_free_p; p->a_label = nfs4_label_alloc(server, gfp_mask); @@ -1434,7 +1433,7 @@ static struct nfs4_opendata *nfs4_opendata_alloc(struct dentry *dentry, err_free_label: nfs4_label_free(p->a_label); err_free_f: - nfs4_label_free(p->f_label); + nfs4_label_free(p->f_attr.label); err_free_p: kfree(p); err: @@ -1456,7 +1455,7 @@ static void nfs4_opendata_free(struct kref *kref) nfs4_put_state_owner(p->owner); nfs4_label_free(p->a_label); - nfs4_label_free(p->f_label); + nfs4_label_free(p->f_attr.label); dput(p->dir); dput(p->dentry); @@ -2009,7 +2008,7 @@ nfs4_opendata_get_inode(struct nfs4_opendata *data) if (!(data->f_attr.valid & NFS_ATTR_FATTR)) return ERR_PTR(-EAGAIN); inode = nfs_fhget(data->dir->d_sb, &data->o_res.fh, - &data->f_attr, data->f_label); + &data->f_attr, data->f_attr.label); break; default: inode = d_inode(data->dentry); @@ -2709,7 +2708,7 @@ static int _nfs4_proc_open(struct nfs4_opendata *data, if (!(o_res->f_attr->valid & NFS_ATTR_FATTR)) { nfs4_sequence_free_slot(&o_res->seq_res); nfs4_proc_getattr(server, &o_res->fh, o_res->f_attr, - o_res->f_label, NULL); + o_res->f_attr->label, NULL); } return 0; } @@ -3125,7 +3124,6 @@ static int _nfs4_do_open(struct inode *dir, enum open_claim_type4 claim = NFS4_OPEN_CLAIM_NULL; struct iattr *sattr = c->sattr; struct nfs4_label *label = c->label; - struct nfs4_label *olabel = NULL; int status; /* Protect against reboot recovery conflicts */ @@ -3148,19 +3146,11 @@ static int _nfs4_do_open(struct inode *dir, if (opendata == NULL) goto err_put_state_owner; - if (label) { - olabel = nfs4_label_alloc(server, GFP_KERNEL); - if (IS_ERR(olabel)) { - status = PTR_ERR(olabel); - goto err_opendata_put; - } - } - if (server->attr_bitmask[2] & FATTR4_WORD2_MDSTHRESHOLD) { if (!opendata->f_attr.mdsthreshold) { opendata->f_attr.mdsthreshold = pnfs_mdsthreshold_alloc(); if (!opendata->f_attr.mdsthreshold) - goto err_free_label; + goto err_opendata_put; } opendata->o_arg.open_bitmap = &nfs4_pnfs_open_bitmap[0]; } @@ -3169,7 +3159,7 @@ static int _nfs4_do_open(struct inode *dir, status = _nfs4_open_and_get_state(opendata, flags, ctx); if (status != 0) - goto err_free_label; + goto err_opendata_put; state = ctx->state; if ((opendata->o_arg.open_flags & (O_CREAT|O_EXCL)) == (O_CREAT|O_EXCL) && @@ -3186,11 +3176,12 @@ static int _nfs4_do_open(struct inode *dir, nfs_fattr_init(opendata->o_res.f_attr); status = nfs4_do_setattr(state->inode, cred, opendata->o_res.f_attr, sattr, - ctx, label, olabel); + ctx, label, opendata->o_res.f_attr->label); if (status == 0) { nfs_setattr_update_inode(state->inode, sattr, opendata->o_res.f_attr); - nfs_setsecurity(state->inode, opendata->o_res.f_attr, olabel); + nfs_setsecurity(state->inode, opendata->o_res.f_attr, + opendata->o_res.f_attr->label); } sattr->ia_valid = ia_old; } @@ -3203,13 +3194,9 @@ static int _nfs4_do_open(struct inode *dir, opendata->f_attr.mdsthreshold = NULL; } - nfs4_label_free(olabel); - nfs4_opendata_put(opendata); nfs4_put_state_owner(sp); return 0; -err_free_label: - nfs4_label_free(olabel); err_opendata_put: nfs4_opendata_put(opendata); err_put_state_owner: diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c index 0044747f9314..09bd1d121318 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -6524,7 +6524,7 @@ static int nfs4_xdr_dec_open(struct rpc_rqst *rqstp, struct xdr_stream *xdr, goto out; if (res->access_request) decode_access(xdr, &res->access_supported, &res->access_result); - decode_getfattr_label(xdr, res->f_attr, res->f_label, res->server); + decode_getfattr_label(xdr, res->f_attr, res->f_attr->label, res->server); if (res->lg_res) decode_layoutget(xdr, rqstp, res->lg_res); out: diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index f0a685d9b8bd..cb28e01ea41e 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -488,7 +488,6 @@ struct nfs_openres { struct nfs4_change_info cinfo; __u32 rflags; struct nfs_fattr * f_attr; - struct nfs4_label *f_label; struct nfs_seqid * seqid; const struct nfs_server *server; fmode_t delegation_type; From 2ef61e0eaa333e4e9c348c41a4b7abfb34b8736d Mon Sep 17 00:00:00 2001 From: Anna Schumaker Date: Fri, 22 Oct 2021 13:11:07 -0400 Subject: [PATCH 77/84] NFS: Remove the nfs4_label from the nfs4_getattr_res Signed-off-by: Anna Schumaker Signed-off-by: Trond Myklebust --- fs/nfs/client.c | 2 +- fs/nfs/dir.c | 2 +- fs/nfs/export.c | 18 ++++-------------- fs/nfs/inode.c | 20 +++++--------------- fs/nfs/nfs3proc.c | 3 +-- fs/nfs/nfs4_fs.h | 3 +-- fs/nfs/nfs4file.c | 2 +- fs/nfs/nfs4proc.c | 25 +++++++++++-------------- fs/nfs/nfs4xdr.c | 2 +- fs/nfs/proc.c | 3 +-- include/linux/nfs_xdr.h | 4 +--- 11 files changed, 28 insertions(+), 56 deletions(-) diff --git a/fs/nfs/client.c b/fs/nfs/client.c index 960b9d87648e..1e4dc1ab9312 100644 --- a/fs/nfs/client.c +++ b/fs/nfs/client.c @@ -1048,7 +1048,7 @@ struct nfs_server *nfs_create_server(struct fs_context *fc) if (!(fattr->valid & NFS_ATTR_FATTR)) { error = ctx->nfs_mod->rpc_ops->getattr(server, ctx->mntfh, - fattr, NULL, NULL); + fattr, NULL); if (error < 0) { dprintk("nfs_create_server: getattr error = %d\n", -error); goto error; diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index 8a327971d485..aa95a898ad0f 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -2056,7 +2056,7 @@ nfs_add_or_obtain(struct dentry *dentry, struct nfs_fh *fhandle, if (!(fattr->valid & NFS_ATTR_FATTR)) { struct nfs_server *server = NFS_SB(dentry->d_sb); error = server->nfs_client->rpc_ops->getattr(server, fhandle, - fattr, NULL, NULL); + fattr, NULL); if (error < 0) goto out_error; } diff --git a/fs/nfs/export.c b/fs/nfs/export.c index 895b404888dd..a0462f7e7e35 100644 --- a/fs/nfs/export.c +++ b/fs/nfs/export.c @@ -64,7 +64,6 @@ static struct dentry * nfs_fh_to_dentry(struct super_block *sb, struct fid *fid, int fh_len, int fh_type) { - struct nfs4_label *label = NULL; struct nfs_fattr *fattr = NULL; struct nfs_fh *server_fh = nfs_exp_embedfh(fid->raw); size_t fh_size = offsetof(struct nfs_fh, data) + server_fh->size; @@ -79,7 +78,7 @@ nfs_fh_to_dentry(struct super_block *sb, struct fid *fid, if (fh_len < len || fh_type != len) return NULL; - fattr = nfs_alloc_fattr(); + fattr = nfs_alloc_fattr_with_label(NFS_SB(sb)); if (fattr == NULL) { dentry = ERR_PTR(-ENOMEM); goto out; @@ -95,28 +94,19 @@ nfs_fh_to_dentry(struct super_block *sb, struct fid *fid, if (inode) goto out_found; - label = nfs4_label_alloc(NFS_SB(sb), GFP_KERNEL); - if (IS_ERR(label)) { - dentry = ERR_CAST(label); - goto out_free_fattr; - } - rpc_ops = NFS_SB(sb)->nfs_client->rpc_ops; - ret = rpc_ops->getattr(NFS_SB(sb), server_fh, fattr, label, NULL); + ret = rpc_ops->getattr(NFS_SB(sb), server_fh, fattr, NULL); if (ret) { dprintk("%s: getattr failed %d\n", __func__, ret); trace_nfs_fh_to_dentry(sb, server_fh, fattr->fileid, ret); dentry = ERR_PTR(ret); - goto out_free_label; + goto out_free_fattr; } - inode = nfs_fhget(sb, server_fh, fattr, label); + inode = nfs_fhget(sb, server_fh, fattr, fattr->label); out_found: dentry = d_obtain_alias(inode); - -out_free_label: - nfs4_label_free(label); out_free_fattr: nfs_free_fattr(fattr); out: diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index 84c7efa2ea87..7d9dca781956 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -1194,7 +1194,6 @@ int __nfs_revalidate_inode(struct nfs_server *server, struct inode *inode) { int status = -ESTALE; - struct nfs4_label *label = NULL; struct nfs_fattr *fattr = NULL; struct nfs_inode *nfsi = NFS_I(inode); @@ -1216,20 +1215,13 @@ __nfs_revalidate_inode(struct nfs_server *server, struct inode *inode) } status = -ENOMEM; - fattr = nfs_alloc_fattr(); + fattr = nfs_alloc_fattr_with_label(NFS_SERVER(inode)); if (fattr == NULL) goto out; nfs_inc_stats(inode, NFSIOS_INODEREVALIDATE); - label = nfs4_label_alloc(NFS_SERVER(inode), GFP_KERNEL); - if (IS_ERR(label)) { - status = PTR_ERR(label); - goto out; - } - - status = NFS_PROTO(inode)->getattr(server, NFS_FH(inode), fattr, - label, inode); + status = NFS_PROTO(inode)->getattr(server, NFS_FH(inode), fattr, inode); if (status != 0) { dfprintk(PAGECACHE, "nfs_revalidate_inode: (%s/%Lu) getattr failed, error=%d\n", inode->i_sb->s_id, @@ -1246,7 +1238,7 @@ __nfs_revalidate_inode(struct nfs_server *server, struct inode *inode) else nfs_zap_caches(inode); } - goto err_out; + goto out; } status = nfs_refresh_inode(inode, fattr); @@ -1254,20 +1246,18 @@ __nfs_revalidate_inode(struct nfs_server *server, struct inode *inode) dfprintk(PAGECACHE, "nfs_revalidate_inode: (%s/%Lu) refresh failed, error=%d\n", inode->i_sb->s_id, (unsigned long long)NFS_FILEID(inode), status); - goto err_out; + goto out; } if (nfsi->cache_validity & NFS_INO_INVALID_ACL) nfs_zap_acl_cache(inode); - nfs_setsecurity(inode, fattr, label); + nfs_setsecurity(inode, fattr, fattr->label); dfprintk(PAGECACHE, "NFS: (%s/%Lu) revalidation complete\n", inode->i_sb->s_id, (unsigned long long)NFS_FILEID(inode)); -err_out: - nfs4_label_free(label); out: nfs_free_fattr(fattr); trace_nfs_revalidate_inode_exit(inode, status); diff --git a/fs/nfs/nfs3proc.c b/fs/nfs/nfs3proc.c index 516f3340b226..7bae21a2ba05 100644 --- a/fs/nfs/nfs3proc.c +++ b/fs/nfs/nfs3proc.c @@ -100,8 +100,7 @@ nfs3_proc_get_root(struct nfs_server *server, struct nfs_fh *fhandle, */ static int nfs3_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle, - struct nfs_fattr *fattr, struct nfs4_label *label, - struct inode *inode) + struct nfs_fattr *fattr, struct inode *inode) { struct rpc_message msg = { .rpc_proc = &nfs3_procedures[NFS3PROC_GETATTR], diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h index b621e29e6187..ed5eaca6801e 100644 --- a/fs/nfs/nfs4_fs.h +++ b/fs/nfs/nfs4_fs.h @@ -316,8 +316,7 @@ extern int nfs4_set_rw_stateid(nfs4_stateid *stateid, const struct nfs_lock_context *l_ctx, fmode_t fmode); extern int nfs4_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle, - struct nfs_fattr *fattr, struct nfs4_label *label, - struct inode *inode); + struct nfs_fattr *fattr, struct inode *inode); extern int update_open_stateid(struct nfs4_state *state, const nfs4_stateid *open_stateid, const nfs4_stateid *deleg_stateid, diff --git a/fs/nfs/nfs4file.c b/fs/nfs/nfs4file.c index 92a1b992a141..e2451f66024c 100644 --- a/fs/nfs/nfs4file.c +++ b/fs/nfs/nfs4file.c @@ -331,7 +331,7 @@ static struct file *__nfs42_ssc_open(struct vfsmount *ss_mnt, if (!fattr) return ERR_PTR(-ENOMEM); - status = nfs4_proc_getattr(server, src_fh, fattr, NULL, NULL); + status = nfs4_proc_getattr(server, src_fh, fattr, NULL); if (status < 0) { res = ERR_PTR(status); goto out; diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index cb0613e0ef8f..f0262397faec 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -93,7 +93,8 @@ struct nfs4_opendata; static int _nfs4_recover_proc_open(struct nfs4_opendata *data); static int nfs4_do_fsinfo(struct nfs_server *, struct nfs_fh *, struct nfs_fsinfo *); static void nfs_fixup_referral_attributes(struct nfs_fattr *fattr); -static int _nfs4_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle, struct nfs_fattr *fattr, struct nfs4_label *label, struct inode *inode); +static int _nfs4_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle, + struct nfs_fattr *fattr, struct inode *inode); static int nfs4_do_setattr(struct inode *inode, const struct cred *cred, struct nfs_fattr *fattr, struct iattr *sattr, struct nfs_open_context *ctx, struct nfs4_label *ilabel, @@ -2707,8 +2708,7 @@ static int _nfs4_proc_open(struct nfs4_opendata *data, } if (!(o_res->f_attr->valid & NFS_ATTR_FATTR)) { nfs4_sequence_free_slot(&o_res->seq_res); - nfs4_proc_getattr(server, &o_res->fh, o_res->f_attr, - o_res->f_attr->label, NULL); + nfs4_proc_getattr(server, &o_res->fh, o_res->f_attr, NULL); } return 0; } @@ -4090,7 +4090,6 @@ static int nfs4_proc_get_root(struct nfs_server *server, struct nfs_fh *mntfh, { int error; struct nfs_fattr *fattr = info->fattr; - struct nfs4_label *label = fattr->label; error = nfs4_server_capabilities(server, mntfh); if (error < 0) { @@ -4098,7 +4097,7 @@ static int nfs4_proc_get_root(struct nfs_server *server, struct nfs_fh *mntfh, return error; } - error = nfs4_proc_getattr(server, mntfh, fattr, label, NULL); + error = nfs4_proc_getattr(server, mntfh, fattr, NULL); if (error < 0) { dprintk("nfs4_get_root: getattr error = %d\n", -error); goto out; @@ -4161,8 +4160,7 @@ out: } static int _nfs4_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle, - struct nfs_fattr *fattr, struct nfs4_label *label, - struct inode *inode) + struct nfs_fattr *fattr, struct inode *inode) { __u32 bitmask[NFS4_BITMASK_SZ]; struct nfs4_getattr_arg args = { @@ -4171,7 +4169,6 @@ static int _nfs4_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle, }; struct nfs4_getattr_res res = { .fattr = fattr, - .label = label, .server = server, }; struct rpc_message msg = { @@ -4188,7 +4185,7 @@ static int _nfs4_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle, if (inode && (server->flags & NFS_MOUNT_SOFTREVAL)) task_flags |= RPC_TASK_TIMEOUT; - nfs4_bitmap_copy_adjust(bitmask, nfs4_bitmask(server, label), inode, 0); + nfs4_bitmap_copy_adjust(bitmask, nfs4_bitmask(server, fattr->label), inode, 0); nfs_fattr_init(fattr); nfs4_init_sequence(&args.seq_args, &res.seq_res, 0, 0); return nfs4_do_call_sync(server->client, server, &msg, @@ -4196,15 +4193,14 @@ static int _nfs4_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle, } int nfs4_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle, - struct nfs_fattr *fattr, struct nfs4_label *label, - struct inode *inode) + struct nfs_fattr *fattr, struct inode *inode) { struct nfs4_exception exception = { .interruptible = true, }; int err; do { - err = _nfs4_proc_getattr(server, fhandle, fattr, label, inode); + err = _nfs4_proc_getattr(server, fhandle, fattr, inode); trace_nfs4_getattr(server, fhandle, fattr, err); err = nfs4_handle_exception(server, err, &exception); @@ -5972,17 +5968,18 @@ static int _nfs4_get_security_label(struct inode *inode, void *buf, size_t buflen) { struct nfs_server *server = NFS_SERVER(inode); - struct nfs_fattr fattr; struct nfs4_label label = {0, 0, buflen, buf}; u32 bitmask[3] = { 0, 0, FATTR4_WORD2_SECURITY_LABEL }; + struct nfs_fattr fattr = { + .label = &label, + }; struct nfs4_getattr_arg arg = { .fh = NFS_FH(inode), .bitmask = bitmask, }; struct nfs4_getattr_res res = { .fattr = &fattr, - .label = &label, .server = server, }; struct rpc_message msg = { diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c index 09bd1d121318..d1b61b76bc82 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -6386,7 +6386,7 @@ static int nfs4_xdr_dec_getattr(struct rpc_rqst *rqstp, struct xdr_stream *xdr, status = decode_putfh(xdr); if (status) goto out; - status = decode_getfattr_label(xdr, res->fattr, res->label, res->server); + status = decode_getfattr_label(xdr, res->fattr, res->fattr->label, res->server); out: return status; } diff --git a/fs/nfs/proc.c b/fs/nfs/proc.c index 98a8901ede2e..baee21c2c091 100644 --- a/fs/nfs/proc.c +++ b/fs/nfs/proc.c @@ -100,8 +100,7 @@ nfs_proc_get_root(struct nfs_server *server, struct nfs_fh *fhandle, */ static int nfs_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle, - struct nfs_fattr *fattr, struct nfs4_label *label, - struct inode *inode) + struct nfs_fattr *fattr, struct inode *inode) { struct rpc_message msg = { .rpc_proc = &nfs_procedures[NFSPROC_GETATTR], diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index cb28e01ea41e..817f1bf5f187 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -1063,7 +1063,6 @@ struct nfs4_getattr_res { struct nfs4_sequence_res seq_res; const struct nfs_server * server; struct nfs_fattr * fattr; - struct nfs4_label *label; }; struct nfs4_link_arg { @@ -1732,8 +1731,7 @@ struct nfs_rpc_ops { int (*submount) (struct fs_context *, struct nfs_server *); int (*try_get_tree) (struct fs_context *); int (*getattr) (struct nfs_server *, struct nfs_fh *, - struct nfs_fattr *, struct nfs4_label *, - struct inode *); + struct nfs_fattr *, struct inode *); int (*setattr) (struct dentry *, struct nfs_fattr *, struct iattr *); int (*lookup) (struct inode *, struct dentry *, From 1b00ad657997c8984a9e627a3bd37ea14f20beb2 Mon Sep 17 00:00:00 2001 From: Anna Schumaker Date: Fri, 22 Oct 2021 13:11:08 -0400 Subject: [PATCH 78/84] NFS: Remove the nfs4_label from the nfs_setattrres Signed-off-by: Anna Schumaker Signed-off-by: Trond Myklebust --- fs/nfs/inode.c | 2 +- fs/nfs/nfs4proc.c | 56 ++++++++++++----------------------------- fs/nfs/nfs4xdr.c | 2 +- include/linux/nfs_xdr.h | 1 - 4 files changed, 18 insertions(+), 43 deletions(-) diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index 7d9dca781956..df109287f2e0 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -650,7 +650,7 @@ nfs_setattr(struct user_namespace *mnt_userns, struct dentry *dentry, if (S_ISREG(inode->i_mode)) nfs_sync_inode(inode); - fattr = nfs_alloc_fattr(); + fattr = nfs_alloc_fattr_with_label(NFS_SERVER(inode)); if (fattr == NULL) { error = -ENOMEM; goto out; diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index f0262397faec..a0762ecc0c73 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -97,8 +97,7 @@ static int _nfs4_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle, struct nfs_fattr *fattr, struct inode *inode); static int nfs4_do_setattr(struct inode *inode, const struct cred *cred, struct nfs_fattr *fattr, struct iattr *sattr, - struct nfs_open_context *ctx, struct nfs4_label *ilabel, - struct nfs4_label *olabel); + struct nfs_open_context *ctx, struct nfs4_label *ilabel); #ifdef CONFIG_NFS_V4_1 static struct rpc_task *_nfs41_proc_sequence(struct nfs_client *clp, const struct cred *cred, @@ -3176,7 +3175,7 @@ static int _nfs4_do_open(struct inode *dir, nfs_fattr_init(opendata->o_res.f_attr); status = nfs4_do_setattr(state->inode, cred, opendata->o_res.f_attr, sattr, - ctx, label, opendata->o_res.f_attr->label); + ctx, label); if (status == 0) { nfs_setattr_update_inode(state->inode, sattr, opendata->o_res.f_attr); @@ -3341,8 +3340,7 @@ zero_stateid: static int nfs4_do_setattr(struct inode *inode, const struct cred *cred, struct nfs_fattr *fattr, struct iattr *sattr, - struct nfs_open_context *ctx, struct nfs4_label *ilabel, - struct nfs4_label *olabel) + struct nfs_open_context *ctx, struct nfs4_label *ilabel) { struct nfs_server *server = NFS_SERVER(inode); __u32 bitmask[NFS4_BITMASK_SZ]; @@ -3356,7 +3354,6 @@ static int nfs4_do_setattr(struct inode *inode, const struct cred *cred, }; struct nfs_setattrres res = { .fattr = fattr, - .label = olabel, .server = server, }; struct nfs4_exception exception = { @@ -3373,7 +3370,7 @@ static int nfs4_do_setattr(struct inode *inode, const struct cred *cred, adjust_flags |= NFS_INO_INVALID_OTHER; do { - nfs4_bitmap_copy_adjust(bitmask, nfs4_bitmask(server, olabel), + nfs4_bitmap_copy_adjust(bitmask, nfs4_bitmask(server, fattr->label), inode, adjust_flags); err = _nfs4_do_setattr(inode, &arg, &res, cred, ctx); @@ -4232,7 +4229,6 @@ nfs4_proc_setattr(struct dentry *dentry, struct nfs_fattr *fattr, struct inode *inode = d_inode(dentry); const struct cred *cred = NULL; struct nfs_open_context *ctx = NULL; - struct nfs4_label *label = NULL; int status; if (pnfs_ld_layoutret_on_setattr(inode) && @@ -4258,20 +4254,15 @@ nfs4_proc_setattr(struct dentry *dentry, struct nfs_fattr *fattr, cred = ctx->cred; } - label = nfs4_label_alloc(NFS_SERVER(inode), GFP_KERNEL); - if (IS_ERR(label)) - return PTR_ERR(label); - /* Return any delegations if we're going to change ACLs */ if ((sattr->ia_valid & (ATTR_MODE|ATTR_UID|ATTR_GID)) != 0) nfs4_inode_make_writeable(inode); - status = nfs4_do_setattr(inode, cred, fattr, sattr, ctx, NULL, label); + status = nfs4_do_setattr(inode, cred, fattr, sattr, ctx, NULL); if (status == 0) { nfs_setattr_update_inode(inode, sattr, fattr); - nfs_setsecurity(inode, fattr, label); + nfs_setsecurity(inode, fattr, fattr->label); } - nfs4_label_free(label); return status; } @@ -6021,8 +6012,7 @@ static int nfs4_get_security_label(struct inode *inode, void *buf, static int _nfs4_do_set_security_label(struct inode *inode, struct nfs4_label *ilabel, - struct nfs_fattr *fattr, - struct nfs4_label *olabel) + struct nfs_fattr *fattr) { struct iattr sattr = {0}; @@ -6037,7 +6027,6 @@ static int _nfs4_do_set_security_label(struct inode *inode, }; struct nfs_setattrres res = { .fattr = fattr, - .label = olabel, .server = server, }; struct rpc_message msg = { @@ -6058,15 +6047,13 @@ static int _nfs4_do_set_security_label(struct inode *inode, static int nfs4_do_set_security_label(struct inode *inode, struct nfs4_label *ilabel, - struct nfs_fattr *fattr, - struct nfs4_label *olabel) + struct nfs_fattr *fattr) { struct nfs4_exception exception = { }; int err; do { - err = _nfs4_do_set_security_label(inode, ilabel, - fattr, olabel); + err = _nfs4_do_set_security_label(inode, ilabel, fattr); trace_nfs4_set_security_label(inode, err); err = nfs4_handle_exception(NFS_SERVER(inode), err, &exception); @@ -6077,32 +6064,21 @@ static int nfs4_do_set_security_label(struct inode *inode, static int nfs4_set_security_label(struct inode *inode, const void *buf, size_t buflen) { - struct nfs4_label ilabel, *olabel = NULL; - struct nfs_fattr fattr; + struct nfs4_label ilabel = {0, 0, buflen, (char *)buf }; + struct nfs_fattr *fattr; int status; if (!nfs_server_capable(inode, NFS_CAP_SECURITY_LABEL)) return -EOPNOTSUPP; - nfs_fattr_init(&fattr); + fattr = nfs_alloc_fattr_with_label(NFS_SERVER(inode)); + if (fattr == NULL) + return -ENOMEM; - ilabel.pi = 0; - ilabel.lfs = 0; - ilabel.label = (char *)buf; - ilabel.len = buflen; - - olabel = nfs4_label_alloc(NFS_SERVER(inode), GFP_KERNEL); - if (IS_ERR(olabel)) { - status = -PTR_ERR(olabel); - goto out; - } - - status = nfs4_do_set_security_label(inode, &ilabel, &fattr, olabel); + status = nfs4_do_set_security_label(inode, &ilabel, fattr); if (status == 0) - nfs_setsecurity(inode, &fattr, olabel); + nfs_setsecurity(inode, fattr, fattr->label); - nfs4_label_free(olabel); -out: return status; } #endif /* CONFIG_NFS_V4_SECURITY_LABEL */ diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c index d1b61b76bc82..e3df7ada5988 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -6608,7 +6608,7 @@ static int nfs4_xdr_dec_setattr(struct rpc_rqst *rqstp, status = decode_setattr(xdr); if (status) goto out; - decode_getfattr_label(xdr, res->fattr, res->label, res->server); + decode_getfattr_label(xdr, res->fattr, res->fattr->label, res->server); out: return status; } diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index 817f1bf5f187..967a0098f0a9 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -832,7 +832,6 @@ struct nfs_getaclres { struct nfs_setattrres { struct nfs4_sequence_res seq_res; struct nfs_fattr * fattr; - struct nfs4_label *label; const struct nfs_server * server; }; From d91bfc46426d3d772fc0d9d165e3435fd0f0a79e Mon Sep 17 00:00:00 2001 From: Anna Schumaker Date: Fri, 22 Oct 2021 13:11:09 -0400 Subject: [PATCH 79/84] NFS: Remove the nfs4_label argument from nfs_instantiate() Pull the label from the fattr instead. Signed-off-by: Anna Schumaker Signed-off-by: Trond Myklebust --- fs/nfs/dir.c | 5 ++--- fs/nfs/nfs4proc.c | 2 +- fs/nfs/proc.c | 8 ++++---- include/linux/nfs_fs.h | 2 +- 4 files changed, 8 insertions(+), 9 deletions(-) diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index aa95a898ad0f..48ea69af9446 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -2075,12 +2075,11 @@ EXPORT_SYMBOL_GPL(nfs_add_or_obtain); * Code common to create, mkdir, and mknod. */ int nfs_instantiate(struct dentry *dentry, struct nfs_fh *fhandle, - struct nfs_fattr *fattr, - struct nfs4_label *label) + struct nfs_fattr *fattr) { struct dentry *d; - d = nfs_add_or_obtain(dentry, fhandle, fattr, label); + d = nfs_add_or_obtain(dentry, fhandle, fattr, fattr->label); if (IS_ERR(d)) return PTR_ERR(d); diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index a0762ecc0c73..4d09c81502cf 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -4859,7 +4859,7 @@ static int nfs4_do_create(struct inode *dir, struct dentry *dentry, struct nfs4_ data->res.fattr->time_start, NFS_INO_INVALID_DATA); spin_unlock(&dir->i_lock); - status = nfs_instantiate(dentry, data->res.fh, data->res.fattr, data->res.fattr->label); + status = nfs_instantiate(dentry, data->res.fh, data->res.fattr); } return status; } diff --git a/fs/nfs/proc.c b/fs/nfs/proc.c index baee21c2c091..73dcaa99fa9b 100644 --- a/fs/nfs/proc.c +++ b/fs/nfs/proc.c @@ -255,7 +255,7 @@ nfs_proc_create(struct inode *dir, struct dentry *dentry, struct iattr *sattr, status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0); nfs_mark_for_revalidate(dir); if (status == 0) - status = nfs_instantiate(dentry, data->res.fh, data->res.fattr, NULL); + status = nfs_instantiate(dentry, data->res.fh, data->res.fattr); nfs_free_createdata(data); out: dprintk("NFS reply create: %d\n", status); @@ -302,7 +302,7 @@ nfs_proc_mknod(struct inode *dir, struct dentry *dentry, struct iattr *sattr, status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0); } if (status == 0) - status = nfs_instantiate(dentry, data->res.fh, data->res.fattr, NULL); + status = nfs_instantiate(dentry, data->res.fh, data->res.fattr); nfs_free_createdata(data); out: dprintk("NFS reply mknod: %d\n", status); @@ -434,7 +434,7 @@ nfs_proc_symlink(struct inode *dir, struct dentry *dentry, struct page *page, * should fill in the data with a LOOKUP call on the wire. */ if (status == 0) - status = nfs_instantiate(dentry, fh, fattr, NULL); + status = nfs_instantiate(dentry, fh, fattr); out_free: nfs_free_fattr(fattr); @@ -463,7 +463,7 @@ nfs_proc_mkdir(struct inode *dir, struct dentry *dentry, struct iattr *sattr) status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0); nfs_mark_for_revalidate(dir); if (status == 0) - status = nfs_instantiate(dentry, data->res.fh, data->res.fattr, NULL); + status = nfs_instantiate(dentry, data->res.fh, data->res.fattr); nfs_free_createdata(data); out: dprintk("NFS reply mkdir: %d\n", status); diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index 88c3aed8ad39..a8a9b71aeea6 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -532,7 +532,7 @@ extern struct dentry *nfs_add_or_obtain(struct dentry *dentry, struct nfs_fh *fh, struct nfs_fattr *fattr, struct nfs4_label *label); extern int nfs_instantiate(struct dentry *dentry, struct nfs_fh *fh, - struct nfs_fattr *fattr, struct nfs4_label *label); + struct nfs_fattr *fattr); extern int nfs_may_open(struct inode *inode, const struct cred *cred, int openflags); extern void nfs_access_zap_cache(struct inode *inode); extern int nfs_access_get_cached(struct inode *inode, const struct cred *cred, struct nfs_access_entry *res, From cc6f32989c3202349b90edde0c4702b098410fe8 Mon Sep 17 00:00:00 2001 From: Anna Schumaker Date: Fri, 22 Oct 2021 13:11:10 -0400 Subject: [PATCH 80/84] NFS: Remove the nfs4_label argument from nfs_add_or_obtain() Signed-off-by: Anna Schumaker Signed-off-by: Trond Myklebust --- fs/nfs/dir.c | 7 +++---- fs/nfs/nfs3proc.c | 2 +- include/linux/nfs_fs.h | 3 +-- 3 files changed, 5 insertions(+), 7 deletions(-) diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index 48ea69af9446..1c74f9d2f3a1 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -2036,8 +2036,7 @@ static int nfs4_lookup_revalidate(struct dentry *dentry, unsigned int flags) struct dentry * nfs_add_or_obtain(struct dentry *dentry, struct nfs_fh *fhandle, - struct nfs_fattr *fattr, - struct nfs4_label *label) + struct nfs_fattr *fattr) { struct dentry *parent = dget_parent(dentry); struct inode *dir = d_inode(parent); @@ -2060,7 +2059,7 @@ nfs_add_or_obtain(struct dentry *dentry, struct nfs_fh *fhandle, if (error < 0) goto out_error; } - inode = nfs_fhget(dentry->d_sb, fhandle, fattr, label); + inode = nfs_fhget(dentry->d_sb, fhandle, fattr, fattr->label); d = d_splice_alias(inode, dentry); out: dput(parent); @@ -2079,7 +2078,7 @@ int nfs_instantiate(struct dentry *dentry, struct nfs_fh *fhandle, { struct dentry *d; - d = nfs_add_or_obtain(dentry, fhandle, fattr, fattr->label); + d = nfs_add_or_obtain(dentry, fhandle, fattr); if (IS_ERR(d)) return PTR_ERR(d); diff --git a/fs/nfs/nfs3proc.c b/fs/nfs/nfs3proc.c index 7bae21a2ba05..7100514d306b 100644 --- a/fs/nfs/nfs3proc.c +++ b/fs/nfs/nfs3proc.c @@ -321,7 +321,7 @@ nfs3_do_create(struct inode *dir, struct dentry *dentry, struct nfs3_createdata if (status != 0) return ERR_PTR(status); - return nfs_add_or_obtain(dentry, data->res.fh, data->res.fattr, NULL); + return nfs_add_or_obtain(dentry, data->res.fh, data->res.fattr); } static void nfs3_free_createdata(struct nfs3_createdata *data) diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index a8a9b71aeea6..6eda001b306b 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -529,8 +529,7 @@ extern void nfs_set_verifier(struct dentry * dentry, unsigned long verf); extern void nfs_clear_verifier_delegated(struct inode *inode); #endif /* IS_ENABLED(CONFIG_NFS_V4) */ extern struct dentry *nfs_add_or_obtain(struct dentry *dentry, - struct nfs_fh *fh, struct nfs_fattr *fattr, - struct nfs4_label *label); + struct nfs_fh *fh, struct nfs_fattr *fattr); extern int nfs_instantiate(struct dentry *dentry, struct nfs_fh *fh, struct nfs_fattr *fattr); extern int nfs_may_open(struct inode *inode, const struct cred *cred, int openflags); From cf7ab00aabbf9c8f1ec72edff15849ddc23aa6a7 Mon Sep 17 00:00:00 2001 From: Anna Schumaker Date: Fri, 22 Oct 2021 13:11:11 -0400 Subject: [PATCH 81/84] NFS: Remove the nfs4_label argument from nfs_fhget() Signed-off-by: Anna Schumaker Signed-off-by: Trond Myklebust --- fs/nfs/dir.c | 6 +++--- fs/nfs/export.c | 4 ++-- fs/nfs/getroot.c | 2 +- fs/nfs/inode.c | 4 ++-- fs/nfs/nfs4file.c | 3 +-- fs/nfs/nfs4proc.c | 2 +- include/linux/nfs_fs.h | 2 +- 7 files changed, 11 insertions(+), 12 deletions(-) diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index 1c74f9d2f3a1..bd89f39e8ba9 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -697,7 +697,7 @@ again: goto out; } - inode = nfs_fhget(dentry->d_sb, entry->fh, entry->fattr, entry->fattr->label); + inode = nfs_fhget(dentry->d_sb, entry->fh, entry->fattr); alias = d_splice_alias(inode, dentry); d_lookup_done(dentry); if (alias) { @@ -1782,7 +1782,7 @@ struct dentry *nfs_lookup(struct inode *dir, struct dentry * dentry, unsigned in res = ERR_PTR(error); goto out; } - inode = nfs_fhget(dentry->d_sb, fhandle, fattr, fattr->label); + inode = nfs_fhget(dentry->d_sb, fhandle, fattr); res = ERR_CAST(inode); if (IS_ERR(res)) goto out; @@ -2059,7 +2059,7 @@ nfs_add_or_obtain(struct dentry *dentry, struct nfs_fh *fhandle, if (error < 0) goto out_error; } - inode = nfs_fhget(dentry->d_sb, fhandle, fattr, fattr->label); + inode = nfs_fhget(dentry->d_sb, fhandle, fattr); d = d_splice_alias(inode, dentry); out: dput(parent); diff --git a/fs/nfs/export.c b/fs/nfs/export.c index a0462f7e7e35..171c424cb6d5 100644 --- a/fs/nfs/export.c +++ b/fs/nfs/export.c @@ -103,7 +103,7 @@ nfs_fh_to_dentry(struct super_block *sb, struct fid *fid, goto out_free_fattr; } - inode = nfs_fhget(sb, server_fh, fattr, fattr->label); + inode = nfs_fhget(sb, server_fh, fattr); out_found: dentry = d_obtain_alias(inode); @@ -138,7 +138,7 @@ nfs_get_parent(struct dentry *dentry) goto out; } - pinode = nfs_fhget(sb, &fh, fattr, fattr->label); + pinode = nfs_fhget(sb, &fh, fattr); parent = d_obtain_alias(pinode); out: nfs_free_fattr(fattr); diff --git a/fs/nfs/getroot.c b/fs/nfs/getroot.c index 7604cb6a0ac2..0aedee201166 100644 --- a/fs/nfs/getroot.c +++ b/fs/nfs/getroot.c @@ -91,7 +91,7 @@ int nfs_get_root(struct super_block *s, struct fs_context *fc) goto out_fattr; } - inode = nfs_fhget(s, ctx->mntfh, fsinfo.fattr, NULL); + inode = nfs_fhget(s, ctx->mntfh, fsinfo.fattr); if (IS_ERR(inode)) { dprintk("nfs_get_root: get root inode failed\n"); error = PTR_ERR(inode); diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index df109287f2e0..be28f0251dee 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -448,7 +448,7 @@ static void nfs_inode_init_dir(struct nfs_inode *nfsi) * instead of inode number. */ struct inode * -nfs_fhget(struct super_block *sb, struct nfs_fh *fh, struct nfs_fattr *fattr, struct nfs4_label *label) +nfs_fhget(struct super_block *sb, struct nfs_fh *fh, struct nfs_fattr *fattr) { struct nfs_find_desc desc = { .fh = fh, @@ -581,7 +581,7 @@ nfs_fhget(struct super_block *sb, struct nfs_fh *fh, struct nfs_fattr *fattr, st fattr->size != 0) nfs_set_cache_invalid(inode, NFS_INO_INVALID_BLOCKS); - nfs_setsecurity(inode, fattr, label); + nfs_setsecurity(inode, fattr, fattr->label); nfsi->attrtimeo = NFS_MINATTRTIMEO(inode); nfsi->attrtimeo_timestamp = now; diff --git a/fs/nfs/nfs4file.c b/fs/nfs/nfs4file.c index e2451f66024c..e79ae4cbc395 100644 --- a/fs/nfs/nfs4file.c +++ b/fs/nfs/nfs4file.c @@ -344,8 +344,7 @@ static struct file *__nfs42_ssc_open(struct vfsmount *ss_mnt, goto out; snprintf(read_name, len, SSC_READ_NAME_BODY, read_name_gen++); - r_ino = nfs_fhget(ss_mnt->mnt_root->d_inode->i_sb, src_fh, fattr, - NULL); + r_ino = nfs_fhget(ss_mnt->mnt_root->d_inode->i_sb, src_fh, fattr); if (IS_ERR(r_ino)) { res = ERR_CAST(r_ino); goto out_free_name; diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 4d09c81502cf..b25adb2250ef 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -2008,7 +2008,7 @@ nfs4_opendata_get_inode(struct nfs4_opendata *data) if (!(data->f_attr.valid & NFS_ATTR_FATTR)) return ERR_PTR(-EAGAIN); inode = nfs_fhget(data->dir->d_sb, &data->o_res.fh, - &data->f_attr, data->f_attr.label); + &data->f_attr); break; default: inode = d_inode(data->dentry); diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index 6eda001b306b..c36c6a559fc9 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -388,7 +388,7 @@ extern void nfs_zap_caches(struct inode *); extern void nfs_set_inode_stale(struct inode *inode); extern void nfs_invalidate_atime(struct inode *); extern struct inode *nfs_fhget(struct super_block *, struct nfs_fh *, - struct nfs_fattr *, struct nfs4_label *); + struct nfs_fattr *); struct inode *nfs_ilookup(struct super_block *sb, struct nfs_fattr *, struct nfs_fh *); extern int nfs_refresh_inode(struct inode *, struct nfs_fattr *); extern int nfs_post_op_update_inode(struct inode *inode, struct nfs_fattr *fattr); From dd225cb3b02b827271a2284f89102fc81efcbf6f Mon Sep 17 00:00:00 2001 From: Anna Schumaker Date: Fri, 22 Oct 2021 13:11:12 -0400 Subject: [PATCH 82/84] NFS: Remove the nfs4_label argument from nfs_setsecurity Signed-off-by: Anna Schumaker Signed-off-by: Trond Myklebust --- fs/nfs/dir.c | 5 ++--- fs/nfs/getroot.c | 2 +- fs/nfs/inode.c | 20 +++++++++----------- fs/nfs/nfs4proc.c | 9 ++++----- include/linux/nfs_fs.h | 3 +-- 5 files changed, 17 insertions(+), 22 deletions(-) diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index bd89f39e8ba9..731d31015b6a 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -682,8 +682,7 @@ again: nfs_set_verifier(dentry, dir_verifier); status = nfs_refresh_inode(d_inode(dentry), entry->fattr); if (!status) - nfs_setsecurity(d_inode(dentry), entry->fattr, - entry->fattr->label); + nfs_setsecurity(d_inode(dentry), entry->fattr); goto out; } else { d_invalidate(dentry); @@ -1523,7 +1522,7 @@ nfs_lookup_revalidate_dentry(struct inode *dir, struct dentry *dentry, if (nfs_refresh_inode(inode, fattr) < 0) goto out; - nfs_setsecurity(inode, fattr, fattr->label); + nfs_setsecurity(inode, fattr); nfs_set_verifier(dentry, dir_verifier); /* set a readdirplus hint that we had a cache miss */ diff --git a/fs/nfs/getroot.c b/fs/nfs/getroot.c index 0aedee201166..11ff2b2e060f 100644 --- a/fs/nfs/getroot.c +++ b/fs/nfs/getroot.c @@ -148,7 +148,7 @@ int nfs_get_root(struct super_block *s, struct fs_context *fc) !(kflags_out & SECURITY_LSM_NATIVE_LABELS)) server->caps &= ~NFS_CAP_SECURITY_LABEL; - nfs_setsecurity(inode, fsinfo.fattr, fsinfo.fattr->label); + nfs_setsecurity(inode, fsinfo.fattr); error = 0; out_fattr: diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index be28f0251dee..dd53704c3f40 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -355,23 +355,22 @@ static void nfs_clear_label_invalid(struct inode *inode) spin_unlock(&inode->i_lock); } -void nfs_setsecurity(struct inode *inode, struct nfs_fattr *fattr, - struct nfs4_label *label) +void nfs_setsecurity(struct inode *inode, struct nfs_fattr *fattr) { int error; - if (label == NULL) + if (fattr->label == NULL) return; if ((fattr->valid & NFS_ATTR_FATTR_V4_SECURITY_LABEL) && inode->i_security) { - error = security_inode_notifysecctx(inode, label->label, - label->len); + error = security_inode_notifysecctx(inode, fattr->label->label, + fattr->label->len); if (error) printk(KERN_ERR "%s() %s %d " "security_inode_notifysecctx() %d\n", __func__, - (char *)label->label, - label->len, error); + (char *)fattr->label->label, + fattr->label->len, error); nfs_clear_label_invalid(inode); } } @@ -398,8 +397,7 @@ struct nfs4_label *nfs4_label_alloc(struct nfs_server *server, gfp_t flags) } EXPORT_SYMBOL_GPL(nfs4_label_alloc); #else -void nfs_setsecurity(struct inode *inode, struct nfs_fattr *fattr, - struct nfs4_label *label) +void nfs_setsecurity(struct inode *inode, struct nfs_fattr *fattr) { } #endif @@ -581,7 +579,7 @@ nfs_fhget(struct super_block *sb, struct nfs_fh *fh, struct nfs_fattr *fattr) fattr->size != 0) nfs_set_cache_invalid(inode, NFS_INO_INVALID_BLOCKS); - nfs_setsecurity(inode, fattr, fattr->label); + nfs_setsecurity(inode, fattr); nfsi->attrtimeo = NFS_MINATTRTIMEO(inode); nfsi->attrtimeo_timestamp = now; @@ -1252,7 +1250,7 @@ __nfs_revalidate_inode(struct nfs_server *server, struct inode *inode) if (nfsi->cache_validity & NFS_INO_INVALID_ACL) nfs_zap_acl_cache(inode); - nfs_setsecurity(inode, fattr, fattr->label); + nfs_setsecurity(inode, fattr); dfprintk(PAGECACHE, "NFS: (%s/%Lu) revalidation complete\n", inode->i_sb->s_id, diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index b25adb2250ef..535436dbdc9a 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -3179,8 +3179,7 @@ static int _nfs4_do_open(struct inode *dir, if (status == 0) { nfs_setattr_update_inode(state->inode, sattr, opendata->o_res.f_attr); - nfs_setsecurity(state->inode, opendata->o_res.f_attr, - opendata->o_res.f_attr->label); + nfs_setsecurity(state->inode, opendata->o_res.f_attr); } sattr->ia_valid = ia_old; } @@ -4261,7 +4260,7 @@ nfs4_proc_setattr(struct dentry *dentry, struct nfs_fattr *fattr, status = nfs4_do_setattr(inode, cred, fattr, sattr, ctx, NULL); if (status == 0) { nfs_setattr_update_inode(inode, sattr, fattr); - nfs_setsecurity(inode, fattr, fattr->label); + nfs_setsecurity(inode, fattr); } return status; } @@ -4782,7 +4781,7 @@ static int _nfs4_proc_link(struct inode *inode, struct inode *dir, const struct nfs4_inc_nlink(inode); status = nfs_post_op_update_inode(inode, res.fattr); if (!status) - nfs_setsecurity(inode, res.fattr, res.fattr->label); + nfs_setsecurity(inode, res.fattr); } out: @@ -6077,7 +6076,7 @@ nfs4_set_security_label(struct inode *inode, const void *buf, size_t buflen) status = nfs4_do_set_security_label(inode, &ilabel, fattr); if (status == 0) - nfs_setsecurity(inode, fattr, fattr->label); + nfs_setsecurity(inode, fattr); return status; } diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index c36c6a559fc9..05f249f20f55 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -409,8 +409,7 @@ extern int nfs_revalidate_mapping(struct inode *inode, struct address_space *map extern int nfs_revalidate_mapping_rcu(struct inode *inode); extern int nfs_setattr(struct user_namespace *, struct dentry *, struct iattr *); extern void nfs_setattr_update_inode(struct inode *inode, struct iattr *attr, struct nfs_fattr *); -extern void nfs_setsecurity(struct inode *inode, struct nfs_fattr *fattr, - struct nfs4_label *label); +extern void nfs_setsecurity(struct inode *inode, struct nfs_fattr *fattr); extern struct nfs_open_context *get_nfs_open_context(struct nfs_open_context *ctx); extern void put_nfs_open_context(struct nfs_open_context *ctx); extern struct nfs_open_context *nfs_find_open_context(struct inode *inode, const struct cred *cred, fmode_t mode); From 1e2f67da89310c3b879c4e18d0d1d57e9d901745 Mon Sep 17 00:00:00 2001 From: Anna Schumaker Date: Fri, 22 Oct 2021 13:11:13 -0400 Subject: [PATCH 83/84] NFS: Remove the nfs4_label argument from decode_getattr_*() functions Wa can check if the fattr has an allocated label when needed Signed-off-by: Anna Schumaker Signed-off-by: Trond Myklebust --- fs/nfs/nfs4xdr.c | 43 +++++++++++++++++-------------------------- 1 file changed, 17 insertions(+), 26 deletions(-) diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c index e3df7ada5988..69862bf6db00 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -4585,8 +4585,7 @@ static int decode_attr_mdsthreshold(struct xdr_stream *xdr, static int decode_getfattr_attrs(struct xdr_stream *xdr, uint32_t *bitmap, struct nfs_fattr *fattr, struct nfs_fh *fh, - struct nfs4_fs_locations *fs_loc, struct nfs4_label *label, - const struct nfs_server *server) + struct nfs4_fs_locations *fs_loc, const struct nfs_server *server) { int status; umode_t fmode = 0; @@ -4701,8 +4700,8 @@ static int decode_getfattr_attrs(struct xdr_stream *xdr, uint32_t *bitmap, if (status < 0) goto xdr_error; - if (label) { - status = decode_attr_security_label(xdr, bitmap, label); + if (fattr->label) { + status = decode_attr_security_label(xdr, bitmap, fattr->label); if (status < 0) goto xdr_error; fattr->valid |= status; @@ -4715,7 +4714,7 @@ xdr_error: static int decode_getfattr_generic(struct xdr_stream *xdr, struct nfs_fattr *fattr, struct nfs_fh *fh, struct nfs4_fs_locations *fs_loc, - struct nfs4_label *label, const struct nfs_server *server) + const struct nfs_server *server) { unsigned int savep; uint32_t attrlen, @@ -4734,8 +4733,7 @@ static int decode_getfattr_generic(struct xdr_stream *xdr, struct nfs_fattr *fat if (status < 0) goto xdr_error; - status = decode_getfattr_attrs(xdr, bitmap, fattr, fh, fs_loc, - label, server); + status = decode_getfattr_attrs(xdr, bitmap, fattr, fh, fs_loc, server); if (status < 0) goto xdr_error; @@ -4745,16 +4743,10 @@ xdr_error: return status; } -static int decode_getfattr_label(struct xdr_stream *xdr, struct nfs_fattr *fattr, - struct nfs4_label *label, const struct nfs_server *server) -{ - return decode_getfattr_generic(xdr, fattr, NULL, NULL, label, server); -} - static int decode_getfattr(struct xdr_stream *xdr, struct nfs_fattr *fattr, const struct nfs_server *server) { - return decode_getfattr_generic(xdr, fattr, NULL, NULL, NULL, server); + return decode_getfattr_generic(xdr, fattr, NULL, NULL, server); } /* @@ -6171,7 +6163,7 @@ static int nfs4_xdr_dec_lookup(struct rpc_rqst *rqstp, struct xdr_stream *xdr, status = decode_getfh(xdr, res->fh); if (status) goto out; - status = decode_getfattr_label(xdr, res->fattr, res->fattr->label, res->server); + status = decode_getfattr(xdr, res->fattr, res->server); out: return status; } @@ -6201,7 +6193,7 @@ static int nfs4_xdr_dec_lookupp(struct rpc_rqst *rqstp, struct xdr_stream *xdr, status = decode_getfh(xdr, res->fh); if (status) goto out; - status = decode_getfattr_label(xdr, res->fattr, res->fattr->label, res->server); + status = decode_getfattr(xdr, res->fattr, res->server); out: return status; } @@ -6228,8 +6220,7 @@ static int nfs4_xdr_dec_lookup_root(struct rpc_rqst *rqstp, goto out; status = decode_getfh(xdr, res->fh); if (status == 0) - status = decode_getfattr_label(xdr, res->fattr, - res->fattr->label, res->server); + status = decode_getfattr(xdr, res->fattr, res->server); out: return status; } @@ -6323,7 +6314,7 @@ static int nfs4_xdr_dec_link(struct rpc_rqst *rqstp, struct xdr_stream *xdr, status = decode_restorefh(xdr); if (status) goto out; - decode_getfattr_label(xdr, res->fattr, res->fattr->label, res->server); + decode_getfattr(xdr, res->fattr, res->server); out: return status; } @@ -6353,7 +6344,7 @@ static int nfs4_xdr_dec_create(struct rpc_rqst *rqstp, struct xdr_stream *xdr, status = decode_getfh(xdr, res->fh); if (status) goto out; - decode_getfattr_label(xdr, res->fattr, res->fattr->label, res->server); + decode_getfattr(xdr, res->fattr, res->server); out: return status; } @@ -6386,7 +6377,7 @@ static int nfs4_xdr_dec_getattr(struct rpc_rqst *rqstp, struct xdr_stream *xdr, status = decode_putfh(xdr); if (status) goto out; - status = decode_getfattr_label(xdr, res->fattr, res->fattr->label, res->server); + status = decode_getfattr(xdr, res->fattr, res->server); out: return status; } @@ -6524,7 +6515,7 @@ static int nfs4_xdr_dec_open(struct rpc_rqst *rqstp, struct xdr_stream *xdr, goto out; if (res->access_request) decode_access(xdr, &res->access_supported, &res->access_result); - decode_getfattr_label(xdr, res->f_attr, res->f_attr->label, res->server); + decode_getfattr(xdr, res->f_attr, res->server); if (res->lg_res) decode_layoutget(xdr, rqstp, res->lg_res); out: @@ -6608,7 +6599,7 @@ static int nfs4_xdr_dec_setattr(struct rpc_rqst *rqstp, status = decode_setattr(xdr); if (status) goto out; - decode_getfattr_label(xdr, res->fattr, res->fattr->label, res->server); + decode_getfattr(xdr, res->fattr, res->server); out: return status; } @@ -7023,7 +7014,7 @@ static int nfs4_xdr_dec_fs_locations(struct rpc_rqst *req, status = decode_getfattr_generic(xdr, &res->fs_locations->fattr, NULL, res->fs_locations, - NULL, res->fs_locations->server); + res->fs_locations->server); if (status) goto out; if (res->renew) @@ -7036,7 +7027,7 @@ static int nfs4_xdr_dec_fs_locations(struct rpc_rqst *req, status = decode_getfattr_generic(xdr, &res->fs_locations->fattr, NULL, res->fs_locations, - NULL, res->fs_locations->server); + res->fs_locations->server); } out: return status; @@ -7467,7 +7458,7 @@ int nfs4_decode_dirent(struct xdr_stream *xdr, struct nfs_entry *entry, return -EAGAIN; if (decode_getfattr_attrs(xdr, bitmap, entry->fattr, entry->fh, - NULL, entry->fattr->label, entry->server) < 0) + NULL, entry->server) < 0) return -EAGAIN; if (entry->fattr->valid & NFS_ATTR_FATTR_MOUNTED_ON_FILEID) entry->ino = entry->fattr->mounted_on_fileid; From f96f8cc4a63dd645e07ea9712be4e0a76ea4ec1f Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Sun, 7 Nov 2021 09:15:44 -0500 Subject: [PATCH 84/84] NFSv4: Sanity check the parameters in nfs41_update_target_slotid() Ensure that the values supplied by the server do not exceed the size of the largest allowed slot table. Reported-by: Signed-off-by: Trond Myklebust --- fs/nfs/nfs4session.c | 12 ++++++++---- fs/nfs/nfs4session.h | 1 + 2 files changed, 9 insertions(+), 4 deletions(-) diff --git a/fs/nfs/nfs4session.c b/fs/nfs/nfs4session.c index 4145a0138907..5db460476bf2 100644 --- a/fs/nfs/nfs4session.c +++ b/fs/nfs/nfs4session.c @@ -511,12 +511,16 @@ void nfs41_update_target_slotid(struct nfs4_slot_table *tbl, struct nfs4_slot *slot, struct nfs4_sequence_res *res) { + u32 target_highest_slotid = min(res->sr_target_highest_slotid, + NFS4_MAX_SLOTID); + u32 highest_slotid = min(res->sr_highest_slotid, NFS4_MAX_SLOTID); + spin_lock(&tbl->slot_tbl_lock); - if (!nfs41_is_outlier_target_slotid(tbl, res->sr_target_highest_slotid)) - nfs41_set_target_slotid_locked(tbl, res->sr_target_highest_slotid); + if (!nfs41_is_outlier_target_slotid(tbl, target_highest_slotid)) + nfs41_set_target_slotid_locked(tbl, target_highest_slotid); if (tbl->generation == slot->generation) - nfs41_set_server_slotid_locked(tbl, res->sr_highest_slotid); - nfs41_set_max_slotid_locked(tbl, res->sr_target_highest_slotid); + nfs41_set_server_slotid_locked(tbl, highest_slotid); + nfs41_set_max_slotid_locked(tbl, target_highest_slotid); spin_unlock(&tbl->slot_tbl_lock); } diff --git a/fs/nfs/nfs4session.h b/fs/nfs/nfs4session.h index 3de425f59b3a..351616c61df5 100644 --- a/fs/nfs/nfs4session.h +++ b/fs/nfs/nfs4session.h @@ -12,6 +12,7 @@ #define NFS4_DEF_SLOT_TABLE_SIZE (64U) #define NFS4_DEF_CB_SLOT_TABLE_SIZE (16U) #define NFS4_MAX_SLOT_TABLE (1024U) +#define NFS4_MAX_SLOTID (NFS4_MAX_SLOT_TABLE - 1U) #define NFS4_NO_SLOT ((u32)-1) #if IS_ENABLED(CONFIG_NFS_V4)