Linux NFS development
 help / color / mirror / Atom feed
* [PATCH 1/2] NFS: Use of mapping_set_error() results in spurious errors
@ 2022-02-15 23:05 trondmy
  2022-02-15 23:05 ` [PATCH 2/2] NFS: Do not report writeback errors in nfs_getattr() trondmy
  0 siblings, 1 reply; 2+ messages in thread
From: trondmy @ 2022-02-15 23:05 UTC (permalink / raw)
  To: linux-nfs

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

The use of mapping_set_error() in conjunction with calls to
filemap_check_errors() is problematic because every error gets reported
as either an EIO or an ENOSPC by filemap_check_errors() in functions
such as filemap_write_and_wait() or filemap_write_and_wait_range().
In almost all cases, we prefer to use the more nuanced wb errors.

Fixes: b8946d7bfb94 ("NFS: Revalidate the file mapping on all fatal writeback errors")
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
---
 fs/nfs/write.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index f88b0eb9b18e..74d258781205 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -316,7 +316,10 @@ static void nfs_mapping_set_error(struct page *page, int error)
 	struct address_space *mapping = page_file_mapping(page);
 
 	SetPageError(page);
-	mapping_set_error(mapping, error);
+	filemap_set_wb_err(mapping, error);
+	if (mapping->host)
+		errseq_set(&mapping->host->i_sb->s_wb_err,
+			   error == -ENOSPC ? -ENOSPC : -EIO);
 	nfs_set_pageerror(mapping);
 }
 
-- 
2.35.1


^ permalink raw reply related	[flat|nested] 2+ messages in thread

* [PATCH 2/2] NFS: Do not report writeback errors in nfs_getattr()
  2022-02-15 23:05 [PATCH 1/2] NFS: Use of mapping_set_error() results in spurious errors trondmy
@ 2022-02-15 23:05 ` trondmy
  0 siblings, 0 replies; 2+ messages in thread
From: trondmy @ 2022-02-15 23:05 UTC (permalink / raw)
  To: linux-nfs

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

The result of the writeback, whether it is an ENOSPC or an EIO, or
anything else, does not inhibit the NFS client from reporting the
correct file timestamps.

Fixes: 79566ef018f5 ("NFS: Getattr doesn't require data sync semantics")
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
---
 fs/nfs/inode.c | 9 +++------
 1 file changed, 3 insertions(+), 6 deletions(-)

diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index 90432fc389a0..f9fc506ebb29 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -850,12 +850,9 @@ int nfs_getattr(struct user_namespace *mnt_userns, const struct path *path,
 	}
 
 	/* Flush out writes to the server in order to update c/mtime.  */
-	if ((request_mask & (STATX_CTIME|STATX_MTIME)) &&
-			S_ISREG(inode->i_mode)) {
-		err = filemap_write_and_wait(inode->i_mapping);
-		if (err)
-			goto out;
-	}
+	if ((request_mask & (STATX_CTIME | STATX_MTIME)) &&
+	    S_ISREG(inode->i_mode))
+		filemap_write_and_wait(inode->i_mapping);
 
 	/*
 	 * We may force a getattr if the user cares about atime.
-- 
2.35.1


^ permalink raw reply related	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2022-02-15 23:11 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2022-02-15 23:05 [PATCH 1/2] NFS: Use of mapping_set_error() results in spurious errors trondmy
2022-02-15 23:05 ` [PATCH 2/2] NFS: Do not report writeback errors in nfs_getattr() trondmy

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox