diff options
author | Linus Torvalds <torvalds@linux-foundation.org> | 2022-08-03 15:16:49 -0700 |
---|---|---|
committer | Linus Torvalds <torvalds@linux-foundation.org> | 2022-08-03 15:16:49 -0700 |
commit | f18d73096c0eca1275f586cb984e6e28330447a0 (patch) | |
tree | d8c41b69e3f9373fcf0fbfad057d344db2dfd4a1 /fs/iomap/buffered-io.c | |
parent | 2e4f8c729db5f3c0b8ea8b1b99f1ae124152e8cc (diff) | |
parent | f8189d5d5fbf082786fb91c549f5127f23daec09 (diff) |
Merge tag 'iomap-5.20-merge-1' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux
Pull iomap updates from Darrick Wong:
"The most notable change in this first batch is that we no longer
schedule pages beyond i_size for writeback, preferring instead to let
truncate deal with those pages.
Next week, there may be a second pull request to remove
iomap_writepage from the other two filesystems (gfs2/zonefs) that use
iomap for buffered IO. This follows in the same vein as the recent
removal of writepage from XFS, since it hasn't been triggered in a few
years; it does nothing during direct reclaim; and as far as the people
who examined the patchset can tell, it's moving the codebase in the
right direction.
However, as it was a late addition to for-next, I'm holding off on
that section for another week of testing to see if anyone can come up
with a solid reason for holding off in the meantime.
Summary:
- Skip writeback for pages that are completely beyond EOF
- Minor code cleanups"
* tag 'iomap-5.20-merge-1' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux:
dax: set did_zero to true when zeroing successfully
iomap: set did_zero to true when zeroing successfully
iomap: skip pages past eof in iomap_do_writepage()
Diffstat (limited to 'fs/iomap/buffered-io.c')
-rw-r--r-- | fs/iomap/buffered-io.c | 15 |
1 files changed, 8 insertions, 7 deletions
diff --git a/fs/iomap/buffered-io.c b/fs/iomap/buffered-io.c index 6505d45a99e0..2b82c7f1de88 100644 --- a/fs/iomap/buffered-io.c +++ b/fs/iomap/buffered-io.c @@ -926,10 +926,10 @@ static loff_t iomap_zero_iter(struct iomap_iter *iter, bool *did_zero) pos += bytes; length -= bytes; written += bytes; - if (did_zero) - *did_zero = true; } while (length > 0); + if (did_zero) + *did_zero = true; return written; } @@ -1487,10 +1487,10 @@ iomap_do_writepage(struct page *page, struct writeback_control *wbc, void *data) pgoff_t end_index = isize >> PAGE_SHIFT; /* - * Skip the page if it's fully outside i_size, e.g. due to a - * truncate operation that's in progress. We must redirty the - * page so that reclaim stops reclaiming it. Otherwise - * iomap_release_folio() is called on it and gets confused. + * Skip the page if it's fully outside i_size, e.g. + * due to a truncate operation that's in progress. We've + * cleaned this page and truncate will finish things off for + * us. * * Note that the end_index is unsigned long. If the given * offset is greater than 16TB on a 32-bit system then if we @@ -1505,7 +1505,7 @@ iomap_do_writepage(struct page *page, struct writeback_control *wbc, void *data) */ if (folio->index > end_index || (folio->index == end_index && poff == 0)) - goto redirty; + goto unlock; /* * The page straddles i_size. It must be zeroed out on each @@ -1523,6 +1523,7 @@ iomap_do_writepage(struct page *page, struct writeback_control *wbc, void *data) redirty: folio_redirty_for_writepage(wbc, folio); +unlock: folio_unlock(folio); return 0; } |