| Message ID | 9cdd449fc1d63cf2dba17cfa2fa7fb29b8f96a46.1700506526.git.ritesh.list@gmail.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | ext2: Use iomap in buffered-io for regular files and enable large folio support | expand |
On Tue, Nov 21, 2023 at 12:35:19AM +0530, Ritesh Harjani (IBM) wrote: > Commit "filemap: update ki_pos in generic_perform_write", made updating > of ki_pos into common code in generic_perform_write() function. > This also causes generic/091 to fail. > > Fixes: 182c25e9c157 ("filemap: update ki_pos in generic_perform_write") Looks like this really was an in-flight collision with: fb5de4358e1a ("ext2: Move direct-io to use iomap"). How did we manage to not notice the failure for months, though? Otherwise looks good: Reviewed-by: Christoph Hellwig <hch@lst.de>
Christoph Hellwig <hch@infradead.org> writes: > On Tue, Nov 21, 2023 at 12:35:19AM +0530, Ritesh Harjani (IBM) wrote: >> Commit "filemap: update ki_pos in generic_perform_write", made updating >> of ki_pos into common code in generic_perform_write() function. >> This also causes generic/091 to fail. >> >> Fixes: 182c25e9c157 ("filemap: update ki_pos in generic_perform_write") > > Looks like this really was an in-flight collision with: > fb5de4358e1a ("ext2: Move direct-io to use iomap"). How did we manage > to not notice the failure for months, though? Yes, it was due to in-flight collision. I found it during this conversion and also noticed that generic/091 fails on upstream kernel. > > Otherwise looks good: > > Reviewed-by: Christoph Hellwig <hch@lst.de> Thanks! -ritesh
On Tue, Nov 21, 2023 at 11:06:53AM +0530, Ritesh Harjani wrote: > Christoph Hellwig <hch@infradead.org> writes: > > > On Tue, Nov 21, 2023 at 12:35:19AM +0530, Ritesh Harjani (IBM) wrote: > >> Commit "filemap: update ki_pos in generic_perform_write", made updating > >> of ki_pos into common code in generic_perform_write() function. > >> This also causes generic/091 to fail. > >> > >> Fixes: 182c25e9c157 ("filemap: update ki_pos in generic_perform_write") > > > > Looks like this really was an in-flight collision with: > > fb5de4358e1a ("ext2: Move direct-io to use iomap"). How did we manage > > to not notice the failure for months, though? > > Yes, it was due to in-flight collision. I found it during this conversion and > also noticed that generic/091 fails on upstream kernel. Maybe ammend the commit message with that information and then it off for inclusion into 6.7-rc and -stable ASAP?
diff --git a/fs/ext2/file.c b/fs/ext2/file.c index 1039e5bf90af..4ddc36f4dbd4 100644 --- a/fs/ext2/file.c +++ b/fs/ext2/file.c @@ -258,7 +258,6 @@ static ssize_t ext2_dio_write_iter(struct kiocb *iocb, struct iov_iter *from) goto out_unlock; } - iocb->ki_pos += status; ret += status; endbyte = pos + status - 1; ret2 = filemap_write_and_wait_range(inode->i_mapping, pos,
Commit "filemap: update ki_pos in generic_perform_write", made updating of ki_pos into common code in generic_perform_write() function. This also causes generic/091 to fail. Fixes: 182c25e9c157 ("filemap: update ki_pos in generic_perform_write") Signed-off-by: Ritesh Harjani (IBM) <ritesh.list@gmail.com> --- fs/ext2/file.c | 1 - 1 file changed, 1 deletion(-)