Linux-NFS Archive on lore.kernel.org
 help / color / Atom feed
* [PATCH AUTOSEL 5.2 25/94] NFSv4: Fix return values for nfs4_file_open()
       [not found] <20190904155739.2816-1-sashal@kernel.org>
@ 2019-09-04 15:56 ` Sasha Levin
  2019-09-04 15:56 ` [PATCH AUTOSEL 5.2 26/94] NFSv4: Fix return value in nfs_finish_open() Sasha Levin
                   ` (10 subsequent siblings)
  11 siblings, 0 replies; 12+ messages in thread
From: Sasha Levin @ 2019-09-04 15:56 UTC (permalink / raw)
  To: linux-kernel, stable; +Cc: Trond Myklebust, Sasha Levin, linux-nfs

From: Trond Myklebust <trond.myklebust@hammerspace.com>

[ Upstream commit 90cf500e338ab3f3c0f126ba37e36fb6a9058441 ]

Currently, we are translating RPC level errors such as timeouts,
as well as interrupts etc into EOPENSTALE, which forces a single
replay of the open attempt. What we actually want to do is
force the replay only in the cases where the returned error
indicates that the file may have changed on the server.

So the fix is to spell out the exact set of errors where we want
to return EOPENSTALE.

Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 fs/nfs/nfs4file.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/fs/nfs/nfs4file.c b/fs/nfs/nfs4file.c
index 3a507c42c1cae..113be154a1ec6 100644
--- a/fs/nfs/nfs4file.c
+++ b/fs/nfs/nfs4file.c
@@ -73,13 +73,13 @@ nfs4_file_open(struct inode *inode, struct file *filp)
 	if (IS_ERR(inode)) {
 		err = PTR_ERR(inode);
 		switch (err) {
-		case -EPERM:
-		case -EACCES:
-		case -EDQUOT:
-		case -ENOSPC:
-		case -EROFS:
-			goto out_put_ctx;
 		default:
+			goto out_put_ctx;
+		case -ENOENT:
+		case -ESTALE:
+		case -EISDIR:
+		case -ENOTDIR:
+		case -ELOOP:
 			goto out_drop;
 		}
 	}
-- 
2.20.1


^ permalink raw reply	[flat|nested] 12+ messages in thread

* [PATCH AUTOSEL 5.2 26/94] NFSv4: Fix return value in nfs_finish_open()
       [not found] <20190904155739.2816-1-sashal@kernel.org>
  2019-09-04 15:56 ` [PATCH AUTOSEL 5.2 25/94] NFSv4: Fix return values for nfs4_file_open() Sasha Levin
@ 2019-09-04 15:56 ` Sasha Levin
  2019-09-04 15:56 ` [PATCH AUTOSEL 5.2 27/94] NFS: Fix initialisation of I/O result struct in nfs_pgio_rpcsetup Sasha Levin
                   ` (9 subsequent siblings)
  11 siblings, 0 replies; 12+ messages in thread
From: Sasha Levin @ 2019-09-04 15:56 UTC (permalink / raw)
  To: linux-kernel, stable; +Cc: Trond Myklebust, Sasha Levin, linux-nfs

From: Trond Myklebust <trond.myklebust@hammerspace.com>

[ Upstream commit 9821421a291b548ef4369c6998745baa36ddecd5 ]

If the file turns out to be of the wrong type after opening, we want
to revalidate the path and retry, so return EOPENSTALE rather than
ESTALE.

Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 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 9f44ddc34c7bf..3321cc7a7ead1 100644
--- a/fs/nfs/dir.c
+++ b/fs/nfs/dir.c
@@ -1483,7 +1483,7 @@ static int nfs_finish_open(struct nfs_open_context *ctx,
 	if (S_ISREG(file->f_path.dentry->d_inode->i_mode))
 		nfs_file_set_open_context(file, ctx);
 	else
-		err = -ESTALE;
+		err = -EOPENSTALE;
 out:
 	return err;
 }
-- 
2.20.1


^ permalink raw reply	[flat|nested] 12+ messages in thread

* [PATCH AUTOSEL 5.2 27/94] NFS: Fix initialisation of I/O result struct in nfs_pgio_rpcsetup
       [not found] <20190904155739.2816-1-sashal@kernel.org>
  2019-09-04 15:56 ` [PATCH AUTOSEL 5.2 25/94] NFSv4: Fix return values for nfs4_file_open() Sasha Levin
  2019-09-04 15:56 ` [PATCH AUTOSEL 5.2 26/94] NFSv4: Fix return value in nfs_finish_open() Sasha Levin
@ 2019-09-04 15:56 ` Sasha Levin
  2019-09-04 15:56 ` [PATCH AUTOSEL 5.2 28/94] NFS: On fatal writeback errors, we need to call nfs_inode_remove_request() Sasha Levin
                   ` (8 subsequent siblings)
  11 siblings, 0 replies; 12+ messages in thread
From: Sasha Levin @ 2019-09-04 15:56 UTC (permalink / raw)
  To: linux-kernel, stable; +Cc: Trond Myklebust, Sasha Levin, linux-nfs

From: Trond Myklebust <trond.myklebust@hammerspace.com>

[ Upstream commit 17d8c5d145000070c581f2a8aa01edc7998582ab ]

Initialise the result count to 0 rather than initialising it to the
argument count. The reason is that we want to ensure we record the
I/O stats correctly in the case where an error is returned (for
instance in the layoutstats).

Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 fs/nfs/pagelist.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/nfs/pagelist.c b/fs/nfs/pagelist.c
index 6ef5278326b67..3f6760b17d6a9 100644
--- a/fs/nfs/pagelist.c
+++ b/fs/nfs/pagelist.c
@@ -590,7 +590,7 @@ static void nfs_pgio_rpcsetup(struct nfs_pgio_header *hdr,
 	}
 
 	hdr->res.fattr   = &hdr->fattr;
-	hdr->res.count   = count;
+	hdr->res.count   = 0;
 	hdr->res.eof     = 0;
 	hdr->res.verf    = &hdr->verf;
 	nfs_fattr_init(&hdr->fattr);
-- 
2.20.1


^ permalink raw reply	[flat|nested] 12+ messages in thread

* [PATCH AUTOSEL 5.2 28/94] NFS: On fatal writeback errors, we need to call nfs_inode_remove_request()
       [not found] <20190904155739.2816-1-sashal@kernel.org>
                   ` (2 preceding siblings ...)
  2019-09-04 15:56 ` [PATCH AUTOSEL 5.2 27/94] NFS: Fix initialisation of I/O result struct in nfs_pgio_rpcsetup Sasha Levin
@ 2019-09-04 15:56 ` Sasha Levin
  2019-09-04 15:56 ` [PATCH AUTOSEL 5.2 45/94] SUNRPC: Handle EADDRINUSE and ENOBUFS correctly Sasha Levin
                   ` (7 subsequent siblings)
  11 siblings, 0 replies; 12+ messages in thread
From: Sasha Levin @ 2019-09-04 15:56 UTC (permalink / raw)
  To: linux-kernel, stable; +Cc: Trond Myklebust, Sasha Levin, linux-nfs

From: Trond Myklebust <trond.myklebust@hammerspace.com>

[ Upstream commit 06c9fdf3b9f1acc6e53753c99c54c39764cc979f ]

If the writeback error is fatal, we need to remove the tracking structures
(i.e. the nfs_page) from the inode.

Fixes: 6fbda89b257f ("NFS: Replace custom error reporting mechanism...")
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 fs/nfs/write.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index 059a7c38bc4fc..bf3a3f5e1884e 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -57,6 +57,7 @@ static const struct rpc_call_ops nfs_commit_ops;
 static const struct nfs_pgio_completion_ops nfs_async_write_completion_ops;
 static const struct nfs_commit_completion_ops nfs_commit_completion_ops;
 static const struct nfs_rw_ops nfs_rw_write_ops;
+static void nfs_inode_remove_request(struct nfs_page *req);
 static void nfs_clear_request_commit(struct nfs_page *req);
 static void nfs_init_cinfo_from_inode(struct nfs_commit_info *cinfo,
 				      struct inode *inode);
@@ -591,7 +592,9 @@ nfs_lock_and_join_requests(struct page *page)
 
 static void nfs_write_error(struct nfs_page *req, int error)
 {
+	nfs_set_pageerror(page_file_mapping(req->wb_page));
 	nfs_mapping_set_error(req->wb_page, error);
+	nfs_inode_remove_request(req);
 	nfs_end_page_writeback(req);
 	nfs_release_request(req);
 }
-- 
2.20.1


^ permalink raw reply	[flat|nested] 12+ messages in thread

* [PATCH AUTOSEL 5.2 45/94] SUNRPC: Handle EADDRINUSE and ENOBUFS correctly
       [not found] <20190904155739.2816-1-sashal@kernel.org>
                   ` (3 preceding siblings ...)
  2019-09-04 15:56 ` [PATCH AUTOSEL 5.2 28/94] NFS: On fatal writeback errors, we need to call nfs_inode_remove_request() Sasha Levin
@ 2019-09-04 15:56 ` Sasha Levin
  2019-09-04 15:56 ` [PATCH AUTOSEL 5.2 46/94] SUNRPC: Handle connection breakages correctly in call_status() Sasha Levin
                   ` (6 subsequent siblings)
  11 siblings, 0 replies; 12+ messages in thread
From: Sasha Levin @ 2019-09-04 15:56 UTC (permalink / raw)
  To: linux-kernel, stable; +Cc: Trond Myklebust, Sasha Levin, linux-nfs, netdev

From: Trond Myklebust <trond.myklebust@hammerspace.com>

[ Upstream commit 80f455da6cd0998a5be30a8af24ea2a22815c212 ]

If a connect or bind attempt returns EADDRINUSE, that means we want to
retry with a different port. It is not a fatal connection error.
Similarly, ENOBUFS is not fatal, but just indicates a memory allocation
issue. Retry after a short delay.

Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 net/sunrpc/clnt.c | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)

diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c
index 9e1743b364ec4..3f090a75f3721 100644
--- a/net/sunrpc/clnt.c
+++ b/net/sunrpc/clnt.c
@@ -1926,6 +1926,9 @@ call_bind_status(struct rpc_task *task)
 		task->tk_rebind_retry--;
 		rpc_delay(task, 3*HZ);
 		goto retry_timeout;
+	case -ENOBUFS:
+		rpc_delay(task, HZ >> 2);
+		goto retry_timeout;
 	case -EAGAIN:
 		goto retry_timeout;
 	case -ETIMEDOUT:
@@ -1949,7 +1952,6 @@ call_bind_status(struct rpc_task *task)
 	case -ENETDOWN:
 	case -EHOSTUNREACH:
 	case -ENETUNREACH:
-	case -ENOBUFS:
 	case -EPIPE:
 		dprintk("RPC: %5u remote rpcbind unreachable: %d\n",
 				task->tk_pid, task->tk_status);
@@ -2040,8 +2042,6 @@ call_connect_status(struct rpc_task *task)
 	case -ENETDOWN:
 	case -ENETUNREACH:
 	case -EHOSTUNREACH:
-	case -EADDRINUSE:
-	case -ENOBUFS:
 	case -EPIPE:
 		xprt_conditional_disconnect(task->tk_rqstp->rq_xprt,
 					    task->tk_rqstp->rq_connect_cookie);
@@ -2050,6 +2050,7 @@ call_connect_status(struct rpc_task *task)
 		/* retry with existing socket, after a delay */
 		rpc_delay(task, 3*HZ);
 		/* fall through */
+	case -EADDRINUSE:
 	case -ENOTCONN:
 	case -EAGAIN:
 	case -ETIMEDOUT:
@@ -2058,6 +2059,9 @@ call_connect_status(struct rpc_task *task)
 		clnt->cl_stats->netreconn++;
 		task->tk_action = call_transmit;
 		return;
+	case -ENOBUFS:
+		rpc_delay(task, HZ >> 2);
+		goto out_retry;
 	}
 	rpc_call_rpcerror(task, status);
 	return;
-- 
2.20.1


^ permalink raw reply	[flat|nested] 12+ messages in thread

* [PATCH AUTOSEL 5.2 46/94] SUNRPC: Handle connection breakages correctly in call_status()
       [not found] <20190904155739.2816-1-sashal@kernel.org>
                   ` (4 preceding siblings ...)
  2019-09-04 15:56 ` [PATCH AUTOSEL 5.2 45/94] SUNRPC: Handle EADDRINUSE and ENOBUFS correctly Sasha Levin
@ 2019-09-04 15:56 ` Sasha Levin
  2019-09-04 15:56 ` [PATCH AUTOSEL 5.2 47/94] pNFS/flexfiles: Don't time out requests on hard mounts Sasha Levin
                   ` (5 subsequent siblings)
  11 siblings, 0 replies; 12+ messages in thread
From: Sasha Levin @ 2019-09-04 15:56 UTC (permalink / raw)
  To: linux-kernel, stable; +Cc: Trond Myklebust, Sasha Levin, linux-nfs, netdev

From: Trond Myklebust <trond.myklebust@hammerspace.com>

[ Upstream commit c82e5472c9980e0e483f4b689044150eefaca408 ]

If the connection breaks while we're waiting for a reply from the
server, then we want to immediately try to reconnect.

Fixes: ec6017d90359 ("SUNRPC fix regression in umount of a secure mount")
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 net/sunrpc/clnt.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c
index 3f090a75f3721..49adae7cf4c2d 100644
--- a/net/sunrpc/clnt.c
+++ b/net/sunrpc/clnt.c
@@ -2292,7 +2292,7 @@ call_status(struct rpc_task *task)
 	case -ECONNABORTED:
 	case -ENOTCONN:
 		rpc_force_rebind(clnt);
-		/* fall through */
+		break;
 	case -EADDRINUSE:
 		rpc_delay(task, 3*HZ);
 		/* fall through */
-- 
2.20.1


^ permalink raw reply	[flat|nested] 12+ messages in thread

* [PATCH AUTOSEL 5.2 47/94] pNFS/flexfiles: Don't time out requests on hard mounts
       [not found] <20190904155739.2816-1-sashal@kernel.org>
                   ` (5 preceding siblings ...)
  2019-09-04 15:56 ` [PATCH AUTOSEL 5.2 46/94] SUNRPC: Handle connection breakages correctly in call_status() Sasha Levin
@ 2019-09-04 15:56 ` Sasha Levin
  2019-09-04 15:56 ` [PATCH AUTOSEL 5.2 48/94] NFS: Fix spurious EIO read errors Sasha Levin
                   ` (4 subsequent siblings)
  11 siblings, 0 replies; 12+ messages in thread
From: Sasha Levin @ 2019-09-04 15:56 UTC (permalink / raw)
  To: linux-kernel, stable; +Cc: Trond Myklebust, Sasha Levin, linux-nfs

From: Trond Myklebust <trond.myklebust@hammerspace.com>

[ Upstream commit 7af46292dadcf8870946916f79fdddf79bd7267f ]

If the mount is hard, we should ignore the 'io_maxretrans' module
parameter so that we always keep retrying.

Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 fs/nfs/flexfilelayout/flexfilelayout.c | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

diff --git a/fs/nfs/flexfilelayout/flexfilelayout.c b/fs/nfs/flexfilelayout/flexfilelayout.c
index bcff3bf5ae09e..e7d365ccc258b 100644
--- a/fs/nfs/flexfilelayout/flexfilelayout.c
+++ b/fs/nfs/flexfilelayout/flexfilelayout.c
@@ -8,6 +8,7 @@
  */
 
 #include <linux/nfs_fs.h>
+#include <linux/nfs_mount.h>
 #include <linux/nfs_page.h>
 #include <linux/module.h>
 #include <linux/sched/mm.h>
@@ -928,7 +929,9 @@ ff_layout_pg_init_read(struct nfs_pageio_descriptor *pgio,
 	pgm = &pgio->pg_mirrors[0];
 	pgm->pg_bsize = mirror->mirror_ds->ds_versions[0].rsize;
 
-	pgio->pg_maxretrans = io_maxretrans;
+	if (NFS_SERVER(pgio->pg_inode)->flags &
+			(NFS_MOUNT_SOFT|NFS_MOUNT_SOFTERR))
+		pgio->pg_maxretrans = io_maxretrans;
 	return;
 out_nolseg:
 	if (pgio->pg_error < 0)
@@ -936,6 +939,7 @@ ff_layout_pg_init_read(struct nfs_pageio_descriptor *pgio,
 out_mds:
 	pnfs_put_lseg(pgio->pg_lseg);
 	pgio->pg_lseg = NULL;
+	pgio->pg_maxretrans = 0;
 	nfs_pageio_reset_read_mds(pgio);
 }
 
@@ -996,12 +1000,15 @@ ff_layout_pg_init_write(struct nfs_pageio_descriptor *pgio,
 		pgm->pg_bsize = mirror->mirror_ds->ds_versions[0].wsize;
 	}
 
-	pgio->pg_maxretrans = io_maxretrans;
+	if (NFS_SERVER(pgio->pg_inode)->flags &
+			(NFS_MOUNT_SOFT|NFS_MOUNT_SOFTERR))
+		pgio->pg_maxretrans = io_maxretrans;
 	return;
 
 out_mds:
 	pnfs_put_lseg(pgio->pg_lseg);
 	pgio->pg_lseg = NULL;
+	pgio->pg_maxretrans = 0;
 	nfs_pageio_reset_write_mds(pgio);
 }
 
-- 
2.20.1


^ permalink raw reply	[flat|nested] 12+ messages in thread

* [PATCH AUTOSEL 5.2 48/94] NFS: Fix spurious EIO read errors
       [not found] <20190904155739.2816-1-sashal@kernel.org>
                   ` (6 preceding siblings ...)
  2019-09-04 15:56 ` [PATCH AUTOSEL 5.2 47/94] pNFS/flexfiles: Don't time out requests on hard mounts Sasha Levin
@ 2019-09-04 15:56 ` Sasha Levin
  2019-09-04 15:56 ` [PATCH AUTOSEL 5.2 49/94] NFS: Fix writepage(s) error handling to not report errors twice Sasha Levin
                   ` (3 subsequent siblings)
  11 siblings, 0 replies; 12+ messages in thread
From: Sasha Levin @ 2019-09-04 15:56 UTC (permalink / raw)
  To: linux-kernel, stable; +Cc: Trond Myklebust, Sasha Levin, linux-nfs

From: Trond Myklebust <trond.myklebust@hammerspace.com>

[ Upstream commit 8f54c7a4babf58bbaf849e126f7ae9664bdc9e04 ]

If the client attempts to read a page, but the read fails due to some
spurious error (e.g. an ACCESS error or a timeout, ...) then we need
to allow other processes to retry.
Also try to report errors correctly when doing a synchronous readpage.

Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 fs/nfs/internal.h | 10 ++++++++++
 fs/nfs/read.c     | 35 ++++++++++++++++++++++++++---------
 fs/nfs/write.c    | 12 ------------
 3 files changed, 36 insertions(+), 21 deletions(-)

diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h
index 81e2fdff227ed..9ab9427405f3f 100644
--- a/fs/nfs/internal.h
+++ b/fs/nfs/internal.h
@@ -773,3 +773,13 @@ static inline bool nfs_error_is_fatal(int err)
 	}
 }
 
+static inline bool nfs_error_is_fatal_on_server(int err)
+{
+	switch (err) {
+	case 0:
+	case -ERESTARTSYS:
+	case -EINTR:
+		return false;
+	}
+	return nfs_error_is_fatal(err);
+}
diff --git a/fs/nfs/read.c b/fs/nfs/read.c
index c19841c82b6a3..cfe0b586eadd4 100644
--- a/fs/nfs/read.c
+++ b/fs/nfs/read.c
@@ -91,19 +91,25 @@ void nfs_pageio_reset_read_mds(struct nfs_pageio_descriptor *pgio)
 }
 EXPORT_SYMBOL_GPL(nfs_pageio_reset_read_mds);
 
-static void nfs_readpage_release(struct nfs_page *req)
+static void nfs_readpage_release(struct nfs_page *req, int error)
 {
 	struct inode *inode = d_inode(nfs_req_openctx(req)->dentry);
+	struct page *page = req->wb_page;
 
 	dprintk("NFS: read done (%s/%llu %d@%lld)\n", inode->i_sb->s_id,
 		(unsigned long long)NFS_FILEID(inode), req->wb_bytes,
 		(long long)req_offset(req));
 
+	if (nfs_error_is_fatal_on_server(error) && error != -ETIMEDOUT)
+		SetPageError(page);
 	if (nfs_page_group_sync_on_bit(req, PG_UNLOCKPAGE)) {
-		if (PageUptodate(req->wb_page))
-			nfs_readpage_to_fscache(inode, req->wb_page, 0);
+		struct address_space *mapping = page_file_mapping(page);
 
-		unlock_page(req->wb_page);
+		if (PageUptodate(page))
+			nfs_readpage_to_fscache(inode, page, 0);
+		else if (!PageError(page) && !PagePrivate(page))
+			generic_error_remove_page(mapping, page);
+		unlock_page(page);
 	}
 	nfs_release_request(req);
 }
@@ -131,7 +137,7 @@ int nfs_readpage_async(struct nfs_open_context *ctx, struct inode *inode,
 			     &nfs_async_read_completion_ops);
 	if (!nfs_pageio_add_request(&pgio, new)) {
 		nfs_list_remove_request(new);
-		nfs_readpage_release(new);
+		nfs_readpage_release(new, pgio.pg_error);
 	}
 	nfs_pageio_complete(&pgio);
 
@@ -153,6 +159,7 @@ static void nfs_page_group_set_uptodate(struct nfs_page *req)
 static void nfs_read_completion(struct nfs_pgio_header *hdr)
 {
 	unsigned long bytes = 0;
+	int error;
 
 	if (test_bit(NFS_IOHDR_REDO, &hdr->flags))
 		goto out;
@@ -179,14 +186,19 @@ static void nfs_read_completion(struct nfs_pgio_header *hdr)
 				zero_user_segment(page, start, end);
 			}
 		}
+		error = 0;
 		bytes += req->wb_bytes;
 		if (test_bit(NFS_IOHDR_ERROR, &hdr->flags)) {
 			if (bytes <= hdr->good_bytes)
 				nfs_page_group_set_uptodate(req);
+			else {
+				error = hdr->error;
+				xchg(&nfs_req_openctx(req)->error, error);
+			}
 		} else
 			nfs_page_group_set_uptodate(req);
 		nfs_list_remove_request(req);
-		nfs_readpage_release(req);
+		nfs_readpage_release(req, error);
 	}
 out:
 	hdr->release(hdr);
@@ -213,7 +225,7 @@ nfs_async_read_error(struct list_head *head, int error)
 	while (!list_empty(head)) {
 		req = nfs_list_entry(head->next);
 		nfs_list_remove_request(req);
-		nfs_readpage_release(req);
+		nfs_readpage_release(req, error);
 	}
 }
 
@@ -337,8 +349,13 @@ int nfs_readpage(struct file *file, struct page *page)
 			goto out;
 	}
 
+	xchg(&ctx->error, 0);
 	error = nfs_readpage_async(ctx, inode, page);
-
+	if (!error) {
+		error = wait_on_page_locked_killable(page);
+		if (!PageUptodate(page) && !error)
+			error = xchg(&ctx->error, 0);
+	}
 out:
 	put_nfs_open_context(ctx);
 	return error;
@@ -372,8 +389,8 @@ readpage_async_filler(void *data, struct page *page)
 		zero_user_segment(page, len, PAGE_SIZE);
 	if (!nfs_pageio_add_request(desc->pgio, new)) {
 		nfs_list_remove_request(new);
-		nfs_readpage_release(new);
 		error = desc->pgio->pg_error;
+		nfs_readpage_release(new, error);
 		goto out;
 	}
 	return 0;
diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index bf3a3f5e1884e..f15dda5efb741 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -599,18 +599,6 @@ static void nfs_write_error(struct nfs_page *req, int error)
 	nfs_release_request(req);
 }
 
-static bool
-nfs_error_is_fatal_on_server(int err)
-{
-	switch (err) {
-	case 0:
-	case -ERESTARTSYS:
-	case -EINTR:
-		return false;
-	}
-	return nfs_error_is_fatal(err);
-}
-
 /*
  * Find an associated nfs write request, and prepare to flush it out
  * May return an error if the user signalled nfs_wait_on_request().
-- 
2.20.1


^ permalink raw reply	[flat|nested] 12+ messages in thread

* [PATCH AUTOSEL 5.2 49/94] NFS: Fix writepage(s) error handling to not report errors twice
       [not found] <20190904155739.2816-1-sashal@kernel.org>
                   ` (7 preceding siblings ...)
  2019-09-04 15:56 ` [PATCH AUTOSEL 5.2 48/94] NFS: Fix spurious EIO read errors Sasha Levin
@ 2019-09-04 15:56 ` Sasha Levin
  2019-09-04 15:56 ` [PATCH AUTOSEL 5.2 53/94] NFSv2: Fix eof handling Sasha Levin
                   ` (2 subsequent siblings)
  11 siblings, 0 replies; 12+ messages in thread
From: Sasha Levin @ 2019-09-04 15:56 UTC (permalink / raw)
  To: linux-kernel, stable; +Cc: Trond Myklebust, Sasha Levin, linux-nfs

From: Trond Myklebust <trond.myklebust@hammerspace.com>

[ Upstream commit 96c4145599b30c0eb6cbeaa24207802452dd1872 ]

If writepage()/writepages() saw an error, but handled it without
reporting it, we should not be re-reporting that error on exit.

Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 fs/nfs/write.c | 21 +++++++++++++--------
 1 file changed, 13 insertions(+), 8 deletions(-)

diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index f15dda5efb741..0d6d7beb85053 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -621,12 +621,12 @@ static int nfs_page_async_flush(struct nfs_pageio_descriptor *pgio,
 	WARN_ON_ONCE(test_bit(PG_CLEAN, &req->wb_flags));
 
 	/* If there is a fatal error that covers this write, just exit */
-	ret = 0;
 	mapping = page_file_mapping(page);
-	if (test_bit(AS_ENOSPC, &mapping->flags) ||
-	    test_bit(AS_EIO, &mapping->flags))
+	ret = pgio->pg_error;
+	if (nfs_error_is_fatal_on_server(ret))
 		goto out_launder;
 
+	ret = 0;
 	if (!nfs_pageio_add_request(pgio, req)) {
 		ret = pgio->pg_error;
 		/*
@@ -638,6 +638,7 @@ static int nfs_page_async_flush(struct nfs_pageio_descriptor *pgio,
 		} else
 			ret = -EAGAIN;
 		nfs_redirty_request(req);
+		pgio->pg_error = 0;
 	} else
 		nfs_add_stats(page_file_mapping(page)->host,
 				NFSIOS_WRITEPAGES, 1);
@@ -657,7 +658,7 @@ static int nfs_do_writepage(struct page *page, struct writeback_control *wbc,
 	ret = nfs_page_async_flush(pgio, page);
 	if (ret == -EAGAIN) {
 		redirty_page_for_writepage(wbc, page);
-		ret = 0;
+		ret = AOP_WRITEPAGE_ACTIVATE;
 	}
 	return ret;
 }
@@ -676,10 +677,11 @@ static int nfs_writepage_locked(struct page *page,
 	nfs_pageio_init_write(&pgio, inode, 0,
 				false, &nfs_async_write_completion_ops);
 	err = nfs_do_writepage(page, wbc, &pgio);
+	pgio.pg_error = 0;
 	nfs_pageio_complete(&pgio);
 	if (err < 0)
 		return err;
-	if (pgio.pg_error < 0)
+	if (nfs_error_is_fatal(pgio.pg_error))
 		return pgio.pg_error;
 	return 0;
 }
@@ -689,7 +691,8 @@ int nfs_writepage(struct page *page, struct writeback_control *wbc)
 	int ret;
 
 	ret = nfs_writepage_locked(page, wbc);
-	unlock_page(page);
+	if (ret != AOP_WRITEPAGE_ACTIVATE)
+		unlock_page(page);
 	return ret;
 }
 
@@ -698,7 +701,8 @@ static int nfs_writepages_callback(struct page *page, struct writeback_control *
 	int ret;
 
 	ret = nfs_do_writepage(page, wbc, data);
-	unlock_page(page);
+	if (ret != AOP_WRITEPAGE_ACTIVATE)
+		unlock_page(page);
 	return ret;
 }
 
@@ -725,6 +729,7 @@ int nfs_writepages(struct address_space *mapping, struct writeback_control *wbc)
 				&nfs_async_write_completion_ops);
 	pgio.pg_io_completion = ioc;
 	err = write_cache_pages(mapping, wbc, nfs_writepages_callback, &pgio);
+	pgio.pg_error = 0;
 	nfs_pageio_complete(&pgio);
 	nfs_io_completion_put(ioc);
 
@@ -733,7 +738,7 @@ int nfs_writepages(struct address_space *mapping, struct writeback_control *wbc)
 	if (err < 0)
 		goto out_err;
 	err = pgio.pg_error;
-	if (err < 0)
+	if (nfs_error_is_fatal(err))
 		goto out_err;
 	return 0;
 out_err:
-- 
2.20.1


^ permalink raw reply	[flat|nested] 12+ messages in thread

* [PATCH AUTOSEL 5.2 53/94] NFSv2: Fix eof handling
       [not found] <20190904155739.2816-1-sashal@kernel.org>
                   ` (8 preceding siblings ...)
  2019-09-04 15:56 ` [PATCH AUTOSEL 5.2 49/94] NFS: Fix writepage(s) error handling to not report errors twice Sasha Levin
@ 2019-09-04 15:56 ` Sasha Levin
  2019-09-04 15:56 ` [PATCH AUTOSEL 5.2 54/94] NFSv2: Fix write regression Sasha Levin
  2019-09-04 15:57 ` [PATCH AUTOSEL 5.2 55/94] NFS: remove set but not used variable 'mapping' Sasha Levin
  11 siblings, 0 replies; 12+ messages in thread
From: Sasha Levin @ 2019-09-04 15:56 UTC (permalink / raw)
  To: linux-kernel, stable; +Cc: Trond Myklebust, Sasha Levin, linux-nfs

From: Trond Myklebust <trond.myklebust@hammerspace.com>

[ Upstream commit 71affe9be45a5c60b9772e1b2701710712637274 ]

If we received a reply from the server with a zero length read and
no error, then that implies we are at eof.

Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 fs/nfs/proc.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/fs/nfs/proc.c b/fs/nfs/proc.c
index 5552fa8b6e128..ec79d2214a78c 100644
--- a/fs/nfs/proc.c
+++ b/fs/nfs/proc.c
@@ -594,7 +594,8 @@ static int nfs_read_done(struct rpc_task *task, struct nfs_pgio_header *hdr)
 		/* Emulate the eof flag, which isn't normally needed in NFSv2
 		 * as it is guaranteed to always return the file attributes
 		 */
-		if (hdr->args.offset + hdr->res.count >= hdr->res.fattr->size)
+		if ((hdr->res.count == 0 && hdr->args.count > 0) ||
+		    hdr->args.offset + hdr->res.count >= hdr->res.fattr->size)
 			hdr->res.eof = 1;
 	}
 	return 0;
-- 
2.20.1


^ permalink raw reply	[flat|nested] 12+ messages in thread

* [PATCH AUTOSEL 5.2 54/94] NFSv2: Fix write regression
       [not found] <20190904155739.2816-1-sashal@kernel.org>
                   ` (9 preceding siblings ...)
  2019-09-04 15:56 ` [PATCH AUTOSEL 5.2 53/94] NFSv2: Fix eof handling Sasha Levin
@ 2019-09-04 15:56 ` Sasha Levin
  2019-09-04 15:57 ` [PATCH AUTOSEL 5.2 55/94] NFS: remove set but not used variable 'mapping' Sasha Levin
  11 siblings, 0 replies; 12+ messages in thread
From: Sasha Levin @ 2019-09-04 15:56 UTC (permalink / raw)
  To: linux-kernel, stable
  Cc: Trond Myklebust, Jan Stancek, Naresh Kamboju, Sasha Levin, linux-nfs

From: Trond Myklebust <trond.myklebust@hammerspace.com>

[ Upstream commit d33d4beb522987d1c305c12500796f9be3687dee ]

Ensure we update the write result count on success, since the
RPC call itself does not do so.

Reported-by: Jan Stancek <jstancek@redhat.com>
Reported-by: Naresh Kamboju <naresh.kamboju@linaro.org>
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Tested-by: Jan Stancek <jstancek@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 fs/nfs/proc.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/fs/nfs/proc.c b/fs/nfs/proc.c
index ec79d2214a78c..0f7288b94633b 100644
--- a/fs/nfs/proc.c
+++ b/fs/nfs/proc.c
@@ -616,8 +616,10 @@ static int nfs_proc_pgio_rpc_prepare(struct rpc_task *task,
 
 static int nfs_write_done(struct rpc_task *task, struct nfs_pgio_header *hdr)
 {
-	if (task->tk_status >= 0)
+	if (task->tk_status >= 0) {
+		hdr->res.count = hdr->args.count;
 		nfs_writeback_update_inode(hdr);
+	}
 	return 0;
 }
 
-- 
2.20.1


^ permalink raw reply	[flat|nested] 12+ messages in thread

* [PATCH AUTOSEL 5.2 55/94] NFS: remove set but not used variable 'mapping'
       [not found] <20190904155739.2816-1-sashal@kernel.org>
                   ` (10 preceding siblings ...)
  2019-09-04 15:56 ` [PATCH AUTOSEL 5.2 54/94] NFSv2: Fix write regression Sasha Levin
@ 2019-09-04 15:57 ` Sasha Levin
  11 siblings, 0 replies; 12+ messages in thread
From: Sasha Levin @ 2019-09-04 15:57 UTC (permalink / raw)
  To: linux-kernel, stable
  Cc: YueHaibing, Hulk Robot, Trond Myklebust, Sasha Levin, linux-nfs

From: YueHaibing <yuehaibing@huawei.com>

[ Upstream commit 99300a85260c2b7febd57082a617d1062532067e ]

Fixes gcc '-Wunused-but-set-variable' warning:

fs/nfs/write.c: In function nfs_page_async_flush:
fs/nfs/write.c:609:24: warning: variable mapping set but not used [-Wunused-but-set-variable]

It is not use since commit aefb623c422e ("NFS: Fix
writepage(s) error handling to not report errors twice")

Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 fs/nfs/write.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index 0d6d7beb85053..ee6932c9819e0 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -606,7 +606,6 @@ static void nfs_write_error(struct nfs_page *req, int error)
 static int nfs_page_async_flush(struct nfs_pageio_descriptor *pgio,
 				struct page *page)
 {
-	struct address_space *mapping;
 	struct nfs_page *req;
 	int ret = 0;
 
@@ -621,7 +620,6 @@ static int nfs_page_async_flush(struct nfs_pageio_descriptor *pgio,
 	WARN_ON_ONCE(test_bit(PG_CLEAN, &req->wb_flags));
 
 	/* If there is a fatal error that covers this write, just exit */
-	mapping = page_file_mapping(page);
 	ret = pgio->pg_error;
 	if (nfs_error_is_fatal_on_server(ret))
 		goto out_launder;
-- 
2.20.1


^ permalink raw reply	[flat|nested] 12+ messages in thread

end of thread, back to index

Thread overview: 12+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
     [not found] <20190904155739.2816-1-sashal@kernel.org>
2019-09-04 15:56 ` [PATCH AUTOSEL 5.2 25/94] NFSv4: Fix return values for nfs4_file_open() Sasha Levin
2019-09-04 15:56 ` [PATCH AUTOSEL 5.2 26/94] NFSv4: Fix return value in nfs_finish_open() Sasha Levin
2019-09-04 15:56 ` [PATCH AUTOSEL 5.2 27/94] NFS: Fix initialisation of I/O result struct in nfs_pgio_rpcsetup Sasha Levin
2019-09-04 15:56 ` [PATCH AUTOSEL 5.2 28/94] NFS: On fatal writeback errors, we need to call nfs_inode_remove_request() Sasha Levin
2019-09-04 15:56 ` [PATCH AUTOSEL 5.2 45/94] SUNRPC: Handle EADDRINUSE and ENOBUFS correctly Sasha Levin
2019-09-04 15:56 ` [PATCH AUTOSEL 5.2 46/94] SUNRPC: Handle connection breakages correctly in call_status() Sasha Levin
2019-09-04 15:56 ` [PATCH AUTOSEL 5.2 47/94] pNFS/flexfiles: Don't time out requests on hard mounts Sasha Levin
2019-09-04 15:56 ` [PATCH AUTOSEL 5.2 48/94] NFS: Fix spurious EIO read errors Sasha Levin
2019-09-04 15:56 ` [PATCH AUTOSEL 5.2 49/94] NFS: Fix writepage(s) error handling to not report errors twice Sasha Levin
2019-09-04 15:56 ` [PATCH AUTOSEL 5.2 53/94] NFSv2: Fix eof handling Sasha Levin
2019-09-04 15:56 ` [PATCH AUTOSEL 5.2 54/94] NFSv2: Fix write regression Sasha Levin
2019-09-04 15:57 ` [PATCH AUTOSEL 5.2 55/94] NFS: remove set but not used variable 'mapping' Sasha Levin

Linux-NFS Archive on lore.kernel.org

Archives are clonable:
	git clone --mirror https://lore.kernel.org/linux-nfs/0 linux-nfs/git/0.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 linux-nfs linux-nfs/ https://lore.kernel.org/linux-nfs \
		linux-nfs@vger.kernel.org
	public-inbox-index linux-nfs

Example config snippet for mirrors

Newsgroup available over NNTP:
	nntp://nntp.lore.kernel.org/org.kernel.vger.linux-nfs


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git