From d8c951c313ed1d7144b55c0d56f7c53220044dda Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Mon, 13 Jan 2014 12:08:11 -0500 Subject: NFSv4.1: Don't trust attributes if a pNFS LAYOUTCOMMIT is outstanding If a LAYOUTCOMMIT is outstanding, then chances are that the metadata server may still be returning incorrect values for the change attribute, ctime, mtime and/or size. Just ignore those attributes for now, and wait for the LAYOUTCOMMIT rpc call to finish. Reported-by: shaobingqing Signed-off-by: Trond Myklebust --- fs/nfs/nfs4proc.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) (limited to 'fs/nfs/nfs4proc.c') diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 15052b81df42..f4908eb40a21 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -7780,10 +7780,7 @@ nfs4_layoutcommit_done(struct rpc_task *task, void *calldata) case -NFS4ERR_BADLAYOUT: /* no layout */ case -NFS4ERR_GRACE: /* loca_recalim always false */ task->tk_status = 0; - break; case 0: - nfs_post_op_update_inode_force_wcc(data->args.inode, - data->res.fattr); break; default: if (nfs4_async_handle_error(task, server, NULL) == -EAGAIN) { @@ -7798,6 +7795,8 @@ static void nfs4_layoutcommit_release(void *calldata) struct nfs4_layoutcommit_data *data = calldata; pnfs_cleanup_layoutcommit(data); + nfs_post_op_update_inode_force_wcc(data->args.inode, + data->res.fattr); put_rpccred(data->cred); kfree(data); } -- cgit v1.2.3 From 78b19bae0813bd6f921ca58490196abd101297bd Mon Sep 17 00:00:00 2001 From: Weston Andros Adamson Date: Mon, 13 Jan 2014 16:54:45 -0500 Subject: nfs4.1: properly handle ENOTSUP in SECINFO_NO_NAME Don't check for -NFS4ERR_NOTSUPP, it's already been mapped to -ENOTSUPP by nfs4_stat_to_errno. This allows the client to mount v4.1 servers that don't support SECINFO_NO_NAME by falling back to the "guess and check" method of nfs4_find_root_sec. Signed-off-by: Weston Andros Adamson Cc: stable@vger.kernel.org # 3.1+ Signed-off-by: Trond Myklebust --- fs/nfs/nfs4proc.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs/nfs/nfs4proc.c') diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index f4908eb40a21..a9eaaaa436cb 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -7919,7 +7919,7 @@ nfs41_proc_secinfo_no_name(struct nfs_server *server, struct nfs_fh *fhandle, switch (err) { case 0: case -NFS4ERR_WRONGSEC: - case -NFS4ERR_NOTSUPP: + case -ENOTSUPP: goto out; default: err = nfs4_handle_exception(server, err, &exception); @@ -7953,7 +7953,7 @@ nfs41_find_root_sec(struct nfs_server *server, struct nfs_fh *fhandle, * Fall back on "guess and check" method if * the server doesn't support SECINFO_NO_NAME */ - if (err == -NFS4ERR_WRONGSEC || err == -NFS4ERR_NOTSUPP) { + if (err == -NFS4ERR_WRONGSEC || err == -ENOTSUPP) { err = nfs4_find_root_sec(server, fhandle, info); goto out_freepage; } -- cgit v1.2.3 From ed7e5423014ad89720fcf315c0b73f2c5d0c7bd2 Mon Sep 17 00:00:00 2001 From: Boaz Harrosh Date: Wed, 22 Jan 2014 20:34:54 +0200 Subject: pnfs: Proper delay for NFS4ERR_RECALLCONFLICT in layout_get_done An NFS4ERR_RECALLCONFLICT is returned by server from a GET_LAYOUT only when a Server Sent a RECALL do to that GET_LAYOUT, or the RECALL and GET_LAYOUT crossed on the wire. In any way this means we want to wait at most until in-flight IO is finished and the RECALL can be satisfied. So a proper wait here is more like 1/10 of a second, not 15 seconds like we have now. In case of a server bug we delay exponentially longer on each retry. Current code totally craps out performance of very large files on most pnfs-objects layouts, because of how the map changes when the file has grown into the next raid group. [Stable: This will patch back to 3.9. If there are earlier still maintained trees, please tell me I'll send a patch] CC: Stable Tree Signed-off-by: Boaz Harrosh Signed-off-by: Trond Myklebust --- fs/nfs/nfs4proc.c | 34 ++++++++++++++++++++++++++++++---- 1 file changed, 30 insertions(+), 4 deletions(-) (limited to 'fs/nfs/nfs4proc.c') diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index a9eaaaa436cb..a1965329a12c 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -7409,9 +7409,9 @@ static void nfs4_layoutget_done(struct rpc_task *task, void *calldata) struct nfs_server *server = NFS_SERVER(inode); struct pnfs_layout_hdr *lo; struct nfs4_state *state = NULL; - unsigned long timeo, giveup; + unsigned long timeo, now, giveup; - dprintk("--> %s\n", __func__); + dprintk("--> %s tk_status => %d\n", __func__, -task->tk_status); if (!nfs41_sequence_done(task, &lgp->res.seq_res)) goto out; @@ -7419,12 +7419,38 @@ static void nfs4_layoutget_done(struct rpc_task *task, void *calldata) switch (task->tk_status) { case 0: goto out; + /* + * NFS4ERR_LAYOUTTRYLATER is a conflict with another client + * (or clients) writing to the same RAID stripe + */ case -NFS4ERR_LAYOUTTRYLATER: + /* + * NFS4ERR_RECALLCONFLICT is when conflict with self (must recall + * existing layout before getting a new one). + */ case -NFS4ERR_RECALLCONFLICT: timeo = rpc_get_timeout(task->tk_client); giveup = lgp->args.timestamp + timeo; - if (time_after(giveup, jiffies)) - task->tk_status = -NFS4ERR_DELAY; + now = jiffies; + if (time_after(giveup, now)) { + unsigned long delay; + + /* Delay for: + * - Not less then NFS4_POLL_RETRY_MIN. + * - One last time a jiffie before we give up + * - exponential backoff (time_now minus start_attempt) + */ + delay = max_t(unsigned long, NFS4_POLL_RETRY_MIN, + min((giveup - now - 1), + now - lgp->args.timestamp)); + + dprintk("%s: NFS4ERR_RECALLCONFLICT waiting %lu\n", + __func__, delay); + rpc_delay(task, delay); + task->tk_status = 0; + rpc_restart_call_prepare(task); + goto out; /* Do not call nfs4_async_handle_error() */ + } break; case -NFS4ERR_EXPIRED: case -NFS4ERR_BAD_STATEID: -- cgit v1.2.3 From 7dd7d95916fe7c8494aa8708204d5a2b8689d270 Mon Sep 17 00:00:00 2001 From: Malahal Naineni Date: Thu, 23 Jan 2014 08:54:55 -0600 Subject: nfs: handle servers that support only ALLOW ACE type. Currently we support ACLs if the NFS server file system supports both ALLOW and DENY ACE types. This patch makes the Linux client work with ACLs even if the server supports only 'ALLOW' ACE type. Signed-off-by: Malahal Naineni Signed-off-by: Trond Myklebust --- fs/nfs/nfs4proc.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) (limited to 'fs/nfs/nfs4proc.c') diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index a1965329a12c..ed10d0d4f860 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -2744,7 +2744,8 @@ static int _nfs4_server_capabilities(struct nfs_server *server, struct nfs_fh *f NFS_CAP_OWNER_GROUP|NFS_CAP_ATIME| NFS_CAP_CTIME|NFS_CAP_MTIME| NFS_CAP_SECURITY_LABEL); - if (res.attr_bitmask[0] & FATTR4_WORD0_ACL) + if (res.attr_bitmask[0] & FATTR4_WORD0_ACL && + res.acl_bitmask & ACL4_SUPPORT_ALLOW_ACL) server->caps |= NFS_CAP_ACLS; if (res.has_links != 0) server->caps |= NFS_CAP_HARDLINKS; @@ -4321,9 +4322,7 @@ static int nfs4_proc_renew(struct nfs_client *clp, struct rpc_cred *cred) static inline int nfs4_server_supports_acls(struct nfs_server *server) { - return (server->caps & NFS_CAP_ACLS) - && (server->acl_bitmask & ACL4_SUPPORT_ALLOW_ACL) - && (server->acl_bitmask & ACL4_SUPPORT_DENY_ACL); + return server->caps & NFS_CAP_ACLS; } /* Assuming that XATTR_SIZE_MAX is a multiple of PAGE_SIZE, and that -- cgit v1.2.3 From f9c96fcc501a43dbc292b17fc0ded4b54e63b79d Mon Sep 17 00:00:00 2001 From: Andy Adamson Date: Wed, 29 Jan 2014 11:34:38 -0500 Subject: NFSv4.1 free slot before resending I/O to MDS Fix a dynamic session slot leak where a slot is preallocated and I/O is resent through the MDS. Signed-off-by: Andy Adamson Signed-off-by: Trond Myklebust --- fs/nfs/nfs4proc.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'fs/nfs/nfs4proc.c') diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index ed10d0d4f860..ae00c3ed733f 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -592,7 +592,7 @@ out_unlock: nfs41_server_notify_highest_slotid_update(session->clp); } -static int nfs41_sequence_done(struct rpc_task *task, struct nfs4_sequence_res *res) +int nfs41_sequence_done(struct rpc_task *task, struct nfs4_sequence_res *res) { struct nfs4_session *session; struct nfs4_slot *slot; @@ -692,6 +692,7 @@ out_retry: rpc_delay(task, NFS4_POLL_RETRY_MAX); return 0; } +EXPORT_SYMBOL_GPL(nfs41_sequence_done); static int nfs4_sequence_done(struct rpc_task *task, struct nfs4_sequence_res *res) -- cgit v1.2.3 From cab92c19821a814ecf5a5279e2699bf28e66caee Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 29 Jan 2014 12:12:15 -0500 Subject: NFSv4: Fix a slot leak in nfs40_sequence_done The check for whether or not we sent an RPC call in nfs40_sequence_done is insufficient to decide whether or not we are holding a session slot, and thus should not be used to decide when to free that slot. This patch replaces the RPC_WAS_SENT() test with the correct test for whether or not slot == NULL. Cc: Chuck Lever Cc: stable@vger.kernel.org # 3.12+ Signed-off-by: Trond Myklebust --- fs/nfs/nfs4proc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs/nfs/nfs4proc.c') diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index ae00c3ed733f..493e9cce1f11 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -539,7 +539,7 @@ static int nfs40_sequence_done(struct rpc_task *task, struct nfs4_slot *slot = res->sr_slot; struct nfs4_slot_table *tbl; - if (!RPC_WAS_SENT(task)) + if (slot == NULL) goto out; tbl = slot->table; -- cgit v1.2.3 From a13ce7c629366880c1072d4d113d3dab6c53510d Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 29 Jan 2014 12:24:03 -0500 Subject: NFSv4.1: Clean up nfs41_sequence_done Move the test for res->sr_slot == NULL out of the nfs41_sequence_free_slot helper and into the main function for efficiency. Signed-off-by: Trond Myklebust --- fs/nfs/nfs4proc.c | 19 ++++++++----------- 1 file changed, 8 insertions(+), 11 deletions(-) (limited to 'fs/nfs/nfs4proc.c') diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 493e9cce1f11..42da6af77587 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -559,15 +559,10 @@ static void nfs41_sequence_free_slot(struct nfs4_sequence_res *res) { struct nfs4_session *session; struct nfs4_slot_table *tbl; + struct nfs4_slot *slot = res->sr_slot; bool send_new_highest_used_slotid = false; - if (!res->sr_slot) { - /* just wake up the next guy waiting since - * we may have not consumed a slot after all */ - dprintk("%s: No slot\n", __func__); - return; - } - tbl = res->sr_slot->table; + tbl = slot->table; session = tbl->session; spin_lock(&tbl->slot_tbl_lock); @@ -577,11 +572,11 @@ static void nfs41_sequence_free_slot(struct nfs4_sequence_res *res) if (tbl->highest_used_slotid > tbl->target_highest_slotid) send_new_highest_used_slotid = true; - if (nfs41_wake_and_assign_slot(tbl, res->sr_slot)) { + if (nfs41_wake_and_assign_slot(tbl, slot)) { send_new_highest_used_slotid = false; goto out_unlock; } - nfs4_free_slot(tbl, res->sr_slot); + nfs4_free_slot(tbl, slot); if (tbl->highest_used_slotid != NFS4_NO_SLOT) send_new_highest_used_slotid = false; @@ -595,16 +590,17 @@ out_unlock: int nfs41_sequence_done(struct rpc_task *task, struct nfs4_sequence_res *res) { struct nfs4_session *session; - struct nfs4_slot *slot; + struct nfs4_slot *slot = res->sr_slot; struct nfs_client *clp; bool interrupted = false; int ret = 1; + if (slot == NULL) + goto out_noaction; /* don't increment the sequence number if the task wasn't sent */ if (!RPC_WAS_SENT(task)) goto out; - slot = res->sr_slot; session = slot->table->session; if (slot->interrupted) { @@ -679,6 +675,7 @@ out: /* The session may be reset by one of the error handlers. */ dprintk("%s: Error %d free the slot \n", __func__, res->sr_status); nfs41_sequence_free_slot(res); +out_noaction: return ret; retry_nowait: if (rpc_restart_call_prepare(task)) { -- cgit v1.2.3