lk/linux-kernel-test
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

Btrfs: properly honor wbc->nr_to_write changes

Browse Source 
When btrfs fills a delayed allocation, it tries to increase
the wbc nr_to_write to cover a big part of allocation.  The
theory is that we're doing contiguous IO and writing a few
more blocks will save seeks overall at a very low cost.

The problem is that extent_write_cache_pages could ignore
the new higher nr_to_write if nr_to_write had already gone
down to zero.  We fix that by rechecking the nr_to_write
for every page that is processed in the pagevec.

This updates the math around bumping the nr_to_write value
to make sure we don't leave a tiny amount of IO hanging
around for the very end of a new extent.

Signed-off-by: Chris Mason <chris.mason@oracle.com>
This commit is contained in:
Chris Mason
2009-09-18 16:03:16 -04:00
parent 11833d66be
commit f85d7d6c8f
1 changed files with 27 additions and 11 deletions
Download Patch File Download Diff File Expand all files Collapse all files

38
fs/btrfs/extent_io.c
View File

@@ -2182,7 +2182,7 @@ static int __extent_writepage(struct page *page, struct writeback_control *wbc,
delalloc_end = 0; delalloc_end = 0;
page_started = 0; page_started = 0;
if (!epd->extent_locked) { if (!epd->extent_locked) {
u64 delalloc_to_write; u64 delalloc_to_write = 0;
/* /*
* make sure the wbc mapping index is at least updated * make sure the wbc mapping index is at least updated
* to this page. * to this page.
@@ -2202,16 +2202,24 @@ static int __extent_writepage(struct page *page, struct writeback_control *wbc,
tree->ops->fill_delalloc(inode, page, delalloc_start, tree->ops->fill_delalloc(inode, page, delalloc_start,
delalloc_end, &page_started, delalloc_end, &page_started,
&nr_written); &nr_written);
delalloc_to_write = (delalloc_end - /*
max_t(u64, page_offset(page), * delalloc_end is already one less than the total
delalloc_start) + 1) >> * length, so we don't subtract one from
PAGE_CACHE_SHIFT; * PAGE_CACHE_SIZE
if (wbc->nr_to_write < delalloc_to_write) { */
wbc->nr_to_write = min_t(long, 8192, delalloc_to_write += (delalloc_end - delalloc_start +
delalloc_to_write); PAGE_CACHE_SIZE) >>
} PAGE_CACHE_SHIFT;
delalloc_start = delalloc_end + 1; delalloc_start = delalloc_end + 1;
} }
if (wbc->nr_to_write < delalloc_to_write) {
int thresh = 8192;
if (delalloc_to_write < thresh * 2)
thresh = delalloc_to_write;
wbc->nr_to_write = min_t(u64, delalloc_to_write,
thresh);
}
/* did the fill delalloc function already unlock and start /* did the fill delalloc function already unlock and start
* the IO? * the IO?
@@ -2388,6 +2396,7 @@ static int extent_write_cache_pages(struct extent_io_tree *tree,
{ {
int ret = 0; int ret = 0;
int done = 0; int done = 0;
int nr_to_write_done = 0;
struct pagevec pvec; struct pagevec pvec;
int nr_pages; int nr_pages;
pgoff_t index; pgoff_t index;
@@ -2407,7 +2416,7 @@ static int extent_write_cache_pages(struct extent_io_tree *tree,
scanned = 1; scanned = 1;
} }
retry: retry:
while (!done && (index <= end) && while (!done && !nr_to_write_done && (index <= end) &&
(nr_pages = pagevec_lookup_tag(&pvec, mapping, &index, (nr_pages = pagevec_lookup_tag(&pvec, mapping, &index,
PAGECACHE_TAG_DIRTY, min(end - index, PAGECACHE_TAG_DIRTY, min(end - index,
(pgoff_t)PAGEVEC_SIZE-1) + 1))) { (pgoff_t)PAGEVEC_SIZE-1) + 1))) {
@@ -2458,8 +2467,15 @@ retry:
unlock_page(page); unlock_page(page);
ret = 0; ret = 0;
} }
if (ret || wbc->nr_to_write <= 0) if (ret)
done = 1; done = 1;
/*
* the filesystem may choose to bump up nr_to_write.
* We have to make sure to honor the new nr_to_write
* at any time
*/
nr_to_write_done = wbc->nr_to_write <= 0;
} }
pagevec_release(&pvec); pagevec_release(&pvec);
cond_resched(); cond_resched();