btrfs: fix incorrect error return ret being passed to mapping_set_error

The setting of return code ret should be based on the error code
passed into function end_extent_writepage and not on ret. Thanks
to Liu Bo for spotting this mistake in the original fix I submitted.

Detected by CoverityScan, CID#1414312 ("Logically dead code")

Fixes: 5dca6eea91 ("Btrfs: mark mapping with error flag to report errors to userspace")
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Reviewed-by: Liu Bo <bo.li.liu@oracle.com>
Signed-off-by: David Sterba <dsterba@suse.com>
This commit is contained in:
Colin Ian King 2017-05-09 18:14:01 +01:00 committed by David Sterba
parent 8d91012528
commit bff5baf8aa
1 changed files with 1 additions and 1 deletions

View File

@ -2458,7 +2458,7 @@ void end_extent_writepage(struct page *page, int err, u64 start, u64 end)
if (!uptodate) { if (!uptodate) {
ClearPageUptodate(page); ClearPageUptodate(page);
SetPageError(page); SetPageError(page);
ret = ret < 0 ? ret : -EIO; ret = err < 0 ? err : -EIO;
mapping_set_error(page->mapping, ret); mapping_set_error(page->mapping, ret);
} }
} }