What's in the 'bugfixes' branch of the NFS client git tree

PATCH In review Trond Myklebust 2 participants 10 messages Jan 25, 2010 1 message not in the archive

Subsystems (2)

NFS client NFS server

lore

Share & tools

lore

Trond Myklebust

author

2010-01-25 22:04 UTC #0 lore raw

cover What's in the 'bugfixes' branch of the NFS client git tree

I've committed the following patches to the 'bugfixes' branch for merging into mainline before the release of 2.6.33. Please review

Cheers
Trond

---

David Howells (1):
NFS: Avoid warnings when CONFIG_NFS_V4=n

H Hartley Sweeten (1):
NFS: Make nfs_commitdata_release static

Trond Myklebust (6):
NFS: Ensure that we handle NFS4ERR_STALE_STATEID correctly NFSv4.1: Don't call nfs4_schedule_state_recovery() unnecessarily
NFSv4: Don't allow posix locking against servers that don't support it
NFSv4: Ensure that the NFSv4 locking can recover from stateid errors
NFS: Try to commit unstable writes in nfs_release_page()
NFS: Fix a reference leak in nfs_wb_cancel_page()

 fs/nfs/file.c      |    2 +
 fs/nfs/nfs4_fs.h   |    2 +
 fs/nfs/nfs4proc.c  |   78 +++++++++++++++++++++++++++++++++++++++-------------
 fs/nfs/nfs4state.c |    2 +
 fs/nfs/sysctl.c    |    2 +
 fs/nfs/write.c     |    3 +-
 6 files changed, 67 insertions(+), 22 deletions(-)

A message that is not in the archive (sent off-list or not yet received): 20100125215734.14716.19614.stgit-bi+AKbBUZKY6gyzm1THtWbp2dZbC/[email protected]

Trond Myklebust

author

2010-01-25 22:04 UTC #2 lore raw

PATCH 1/8 NFS: Fix a reference leak in nfs_wb_cancel_page()

Signed-off-by Trond Myklebust <[email protected]>

Cc [email protected]

---

 fs/nfs/write.c |    1 +
 1 files changed, 1 insertions(+), 0 deletions(-)

fs/nfs/write.c +1 -0

index d171696..dac8d76 100644
@@ -1541,6 +1541,7 @@ int nfs_wb_page_cancel(struct inode *inode, struct page *page)
 			break;
 		}
 		ret = nfs_wait_on_request(req);
+		nfs_release_request(req);
 		if (ret < 0)
 			goto out;
 	}

Trond Myklebust

author

2010-01-25 22:04 UTC #3 lore raw

PATCH 2/8 NFS: Try to commit unstable writes in nfs_release_page()

In reply to a message not in the archive

If someone calls nfs_release_page(), we presumably already know that the page is clean, however it may be holding an unstable write.

Signed-off-by Trond Myklebust <[email protected]>

Cc [email protected]

---

 fs/nfs/file.c |    2 ++
 1 files changed, 2 insertions(+), 0 deletions(-)

fs/nfs/file.c +2 -0

index 6b89132..63f2071 100644
@@ -486,6 +486,8 @@ static int nfs_release_page(struct page *page, gfp_t gfp)
 {
 	dfprintk(PAGECACHE, "NFS: release_page(%p)\n", page);
 
+	if (gfp & __GFP_WAIT)
+		nfs_wb_page(page->mapping->host, page);
 	/* If PagePrivate() is set, then the page is not freeable */
 	if (PagePrivate(page))
 		return 0;

Trond Myklebust

author

2010-01-25 22:04 UTC #4 lore raw

PATCH 3/8 NFS: Make nfs_commitdata_release static

In reply to a message not in the archive

From: H Hartley Sweeten <[email protected]>

The symbol nfs_commitdata_release is only used locally
in this file. Make it static to prevent the following sparse warning:

warning: symbol 'nfs_commitdata_release' was not declared. Should it be static?

Signed-off-by H Hartley Sweeten <[email protected]>

Cc Trond Myklebust <[email protected]>

Signed-off-by Trond Myklebust <[email protected]>

---

 fs/nfs/write.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

fs/nfs/write.c +1 -1

index dac8d76..7b54b8b 100644
@@ -1233,7 +1233,7 @@ int nfs_writeback_done(struct rpc_task *task, struct nfs_write_data *data)
 
 
 #if defined(CONFIG_NFS_V3) || defined(CONFIG_NFS_V4)
-void nfs_commitdata_release(void *data)
+static void nfs_commitdata_release(void *data)
 {
 	struct nfs_write_data *wdata = data;

Trond Myklebust

author

2010-01-25 22:04 UTC #5 lore raw

PATCH 4/8 NFS: Avoid warnings when CONFIG_NFS_V4=n

In reply to a message not in the archive

From: David Howells <[email protected]>

Avoid the following warnings when CONFIG_NFS_V4=n:

fs/nfs/sysctl.c:19: warning: unused variable nfs_set_port_max' fs/nfs/sysctl.c:18: warning: unused variable nfs_set_port_min'

by making those variables contingent on NFSv4 being configured.

Signed-off-by David Howells <[email protected]>

Signed-off-by Trond Myklebust <[email protected]>

---

 fs/nfs/sysctl.c |    2 ++
 1 files changed, 2 insertions(+), 0 deletions(-)

fs/nfs/sysctl.c +2 -0

index 70e1fbb..ad4d2e7 100644
@@ -15,8 +15,10 @@
 
 #include "callback.h"
 
+#ifdef CONFIG_NFS_V4
 static const int nfs_set_port_min = 0;
 static const int nfs_set_port_max = 65535;
+#endif
 static struct ctl_table_header *nfs_callback_sysctl_table;
 
 static ctl_table nfs_cb_sysctls[] = {

Trond Myklebust

author

2010-01-25 22:04 UTC #6 lore raw

PATCH 5/8 NFSv4: Ensure that the NFSv4 locking can recover from stateid errors

In reply to a message not in the archive

In most cases, we just want to mark the lock_stateid sequence id as being uninitialised.

Signed-off-by Trond Myklebust <[email protected]>

Cc [email protected]

---

 fs/nfs/nfs4proc.c |   19 +++++++++++++++++++
 1 files changed, 19 insertions(+), 0 deletions(-)

fs/nfs/nfs4proc.c +19 -0

index 198d51d..0b68238 100644
@@ -4088,6 +4088,22 @@ static const struct rpc_call_ops nfs4_recover_lock_ops = {
 	.rpc_release = nfs4_lock_release,
 };
 
+static void nfs4_handle_setlk_error(struct nfs_server *server, struct nfs4_lock_state *lsp, int new_lock_owner, int error)
+{
+	struct nfs_client *clp = server->nfs_client;
+	struct nfs4_state *state = lsp->ls_state;
+
+	switch (error) {
+	case -NFS4ERR_ADMIN_REVOKED:
+	case -NFS4ERR_BAD_STATEID:
+	case -NFS4ERR_EXPIRED:
+		if (new_lock_owner != 0 ||
+		   (lsp->ls_flags & NFS_LOCK_INITIALIZED) != 0)
+			nfs4_state_mark_reclaim_nograce(clp, state);
+		lsp->ls_seqid.flags &= ~NFS_SEQID_CONFIRMED;
+	};
+}
+
 static int _nfs4_do_setlk(struct nfs4_state *state, int cmd, struct file_lock *fl, int recovery_type)
 {
 	struct nfs4_lockdata *data;

@@ -4126,6 +4142,9 @@ static int _nfs4_do_setlk(struct nfs4_state *state, int cmd, struct file_lock *f
 	ret = nfs4_wait_for_completion_rpc_task(task);
 	if (ret == 0) {
 		ret = data->rpc_status;
+		if (ret)
+			nfs4_handle_setlk_error(data->server, data->lsp,
+					data->arg.new_lock_owner, ret);
 	} else
 		data->cancelled = 1;
 	rpc_put_task(task);

Trond Myklebust

author

2010-01-25 22:04 UTC #7 lore raw

PATCH 6/8 NFSv4: Don't allow posix locking against servers that don't support it

In reply to a message not in the archive

Signed-off-by Trond Myklebust <[email protected]>

Cc [email protected]

---

 fs/nfs/nfs4_fs.h  |    1 +
 fs/nfs/nfs4proc.c |    7 ++++++-
 2 files changed, 7 insertions(+), 1 deletions(-)

fs/nfs/nfs4_fs.h +1 -0

index 865265b..ea2f41b 100644
@@ -146,6 +146,7 @@ enum {
 	NFS_O_RDWR_STATE,		/* OPEN stateid has read/write state */
 	NFS_STATE_RECLAIM_REBOOT,	/* OPEN stateid server rebooted */
 	NFS_STATE_RECLAIM_NOGRACE,	/* OPEN stateid needs to recover state */
+	NFS_STATE_POSIX_LOCKS,		/* Posix locks are supported */
 };
 
 struct nfs4_state {

fs/nfs/nfs4proc.c +6 -1

index 0b68238..be044b5 100644
@@ -1658,6 +1658,8 @@ static int _nfs4_do_open(struct inode *dir, struct path *path, fmode_t fmode, in
 	status = PTR_ERR(state);
 	if (IS_ERR(state))
 		goto err_opendata_put;
+	if ((opendata->o_res.rflags & NFS4_OPEN_RESULT_LOCKTYPE_POSIX) != 0)
+		set_bit(NFS_STATE_POSIX_LOCKS, &state->flags);
 	nfs4_opendata_put(opendata);
 	nfs4_put_state_owner(sp);
 	*res = state;

@@ -4200,8 +4202,11 @@ static int _nfs4_proc_setlk(struct nfs4_state *state, int cmd, struct file_lock
 {
 	struct nfs_inode *nfsi = NFS_I(state->inode);
 	unsigned char fl_flags = request->fl_flags;
-	int status;
+	int status = -ENOLCK;
 
+	if ((fl_flags & FL_POSIX) &&
+			!test_bit(NFS_STATE_POSIX_LOCKS, &state->flags))
+		goto out;
 	/* Is this a delegated open? */
 	status = nfs4_set_lock_state(state, request);
 	if (status != 0)

Trond Myklebust

author

2010-01-25 22:04 UTC #8 lore raw

PATCH 7/8 NFSv4.1: Don't call nfs4_schedule_state_recovery() unnecessarily

In reply to a message not in the archive

Currently, nfs4_handle_exception() will call it twice if called with an error of -NFS4ERR_STALE_CLIENTID, -NFS4ERR_STALE_STATEID or -NFS4ERR_EXPIRED.

Signed-off-by Trond Myklebust <[email protected]>

---

 fs/nfs/nfs4proc.c |    8 ++------
 1 files changed, 2 insertions(+), 6 deletions(-)

fs/nfs/nfs4proc.c +2 -6

index be044b5..afbfe67 100644
@@ -256,12 +256,8 @@ static int nfs4_handle_exception(const struct nfs_server *server, int errorcode,
 			ret = nfs4_wait_clnt_recover(clp);
 			if (ret == 0)
 				exception->retry = 1;
-#if !defined(CONFIG_NFS_V4_1)
 			break;
-#else /* !defined(CONFIG_NFS_V4_1) */
-			if (!nfs4_has_session(server->nfs_client))
-				break;
-			/* FALLTHROUGH */
+#if defined(CONFIG_NFS_V4_1)
 		case -NFS4ERR_BADSESSION:
 		case -NFS4ERR_BADSLOT:
 		case -NFS4ERR_BAD_HIGH_SLOT:

@@ -274,7 +270,7 @@ static int nfs4_handle_exception(const struct nfs_server *server, int errorcode,
 			nfs4_schedule_state_recovery(clp);
 			exception->retry = 1;
 			break;
-#endif /* !defined(CONFIG_NFS_V4_1) */
+#endif /* defined(CONFIG_NFS_V4_1) */
 		case -NFS4ERR_FILE_OPEN:
 			if (exception->timeout > HZ) {
 				/* We have retried a decent amount, time to

Trond Myklebust

author

2010-01-25 22:04 UTC #9 lore raw

PATCH 8/8 NFS: Ensure that we handle NFS4ERR_STALE_STATEID correctly

In reply to a message not in the archive

Even if the server is crazy, we should be able to mark the stateid as being bad, to ensure it gets recovered.

Signed-off-by Trond Myklebust <[email protected]>

---

 fs/nfs/nfs4_fs.h   |    1 +
 fs/nfs/nfs4proc.c  |   44 +++++++++++++++++++++++++++++++-------------
 fs/nfs/nfs4state.c |    2 +-
 3 files changed, 33 insertions(+), 14 deletions(-)

fs/nfs/nfs4_fs.h +1 -0

index ea2f41b..0c6fda3 100644
@@ -278,6 +278,7 @@ extern void nfs4_state_set_mode_locked(struct nfs4_state *, fmode_t);
 extern void nfs4_schedule_state_recovery(struct nfs_client *);
 extern void nfs4_schedule_state_manager(struct nfs_client *);
 extern int nfs4_state_mark_reclaim_nograce(struct nfs_client *clp, struct nfs4_state *state);
+extern int nfs4_state_mark_reclaim_reboot(struct nfs_client *clp, struct nfs4_state *state);
 extern void nfs41_handle_sequence_flag_errors(struct nfs_client *clp, u32 flags);
 extern void nfs4_put_lock_state(struct nfs4_lock_state *lsp);
 extern int nfs4_set_lock_state(struct nfs4_state *state, struct file_lock *fl);

fs/nfs/nfs4proc.c +31 ~~-13~~

78 lines not on this page · raw message

fs/nfs/nfs4state.c +1 -1

index 6d263ed..c1e2733 100644
@@ -901,7 +901,7 @@ void nfs4_schedule_state_recovery(struct nfs_client *clp)
 	nfs4_schedule_state_manager(clp);
 }
 
-static int nfs4_state_mark_reclaim_reboot(struct nfs_client *clp, struct nfs4_state *state)
+int nfs4_state_mark_reclaim_reboot(struct nfs_client *clp, struct nfs4_state *state)
 {
 
 	set_bit(NFS_STATE_RECLAIM_REBOOT, &state->flags);

Chuck Lever

Reviewed-by on the v1 series

2010-01-25 22:40 UTC #10 lore raw

In reply to Trond Myklebust

Quoted 2 lines · Trond Myklebust

I've committed the following patches to the 'bugfixes' branch for merging into mainline before the release of 2.6.33. Please review

Reviewed-by Chuck Lever <[email protected]>

I checked for mechanical problems. The semantics of NFSv4 locking is beyond me, but fortunately the changes looked mostly obvious.

Quoted 31 lines

Cheers
Trond
---
David Howells (1):
NFS: Avoid warnings when CONFIG_NFS_V4=n
H Hartley Sweeten (1):
NFS: Make nfs_commitdata_release static
Trond Myklebust (6):
NFS: Ensure that we handle NFS4ERR_STALE_STATEID correctly NFSv4.1: Don't call nfs4_schedule_state_recovery() unnecessarily
NFSv4: Don't allow posix locking against servers that don't
support it
NFSv4: Ensure that the NFSv4 locking can recover from stateid
errors
NFS: Try to commit unstable writes in nfs_release_page()
NFS: Fix a reference leak in nfs_wb_cancel_page()
fs/nfs/file.c      |    2 +
fs/nfs/nfs4_fs.h   |    2 +
fs/nfs/nfs4proc.c  |   78 ++++++++++++++++++++++++++++++++++++++ 
+-------------
fs/nfs/nfs4state.c |    2 +
fs/nfs/sysctl.c    |    2 +
fs/nfs/write.c     |    3 +-
6 files changed, 67 insertions(+), 22 deletions(-)

Quoted 4 lines

--
To unsubscribe from this list: send the line "unsubscribe linux-nfs" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

--
Chuck Lever
chuck[dot]lever[at]oracle[dot]com