NotificationsYou must be signed in to change notification settings
Fork1
Star1

Commit9b0d3ac

committed

Btrfs: don't return congestion in write_cache_pages as often

On fast devices that go from congested to uncongested very quickly, pdflushis waiting too often in congestion_wait, and the FS is backing off toeasily in write_cache_pages.For now, fix this on the btrfs side by only checking congestion aftersome bios have already gone down. Longer term a real fix is neededfor pdflush, but that is a larger project.Signed-off-by: Chris Mason <chris.mason@oracle.com>

1 parent7b78c17 commit9b0d3acCopy full SHA for 9b0d3ac

File tree

1 file changed

-5

lines changed

fs/btrfs
- extent_io.c

1 file changed

-5

lines changed

`‎fs/btrfs/extent_io.c‎`

Lines changed: 0 additions & 5 deletions

Original file line number	Diff line number	Diff line change
`@@ -2377,11 +2377,6 @@ static int extent_write_cache_pages(struct extent_io_tree *tree,`
`2377`	`2377`	`intscanned=0;`
`2378`	`2378`	`intrange_whole=0;`
`2379`	`2379`
`2380`		`-if (wbc->nonblocking&&bdi_write_congested(bdi)) {`
`2381`		`-wbc->encountered_congestion=1;`
`2382`		`-return0;`
`2383`		`-}`
`2384`		`-`
`2385`	`2380`	`pagevec_init(&pvec,0);`
`2386`	`2381`	`if (wbc->range_cyclic) {`
`2387`	`2382`	`index=mapping->writeback_index;/* Start from prev offset */`

0 commit comments

Comments

(0)

Movatterモバイル変換

Navigation Menu

Search code, repositories, users, issues, pull requests...

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Commit9b0d3ac

File tree

1 file changed

1 file changed

`‎fs/btrfs/extent_io.c‎`

0 commit comments