linux-stable.git/fs/fs-writeback.c, branch linux-2.6.32.y

writeback: disable periodic old data writeback for !dirty_writeback_centisecs

2010-07-05T18:10:45+00:00

commit 69b62d01ec44fe0d505d89917392347732135a4d upstream.

Prior to 2.6.32, setting /proc/sys/vm/dirty_writeback_centisecs disabled
periodic dirty writeback from kupdate. This got broken and now causes
excessive sys CPU usage if set to zero, as we'll keep beating on
schedule().

Reported-by: Justin Maggard 
Signed-off-by: Jens Axboe 
Signed-off-by: Greg Kroah-Hartman

fs-writeback: Add helper function to start writeback if idle

2010-04-26T14:41:36+00:00

commit 17bd55d037a02b04d9119511cfd1a4b985d20f63 upstream.

ext4, at least, would like to start pushing on writeback if it starts
to get close to ENOSPC when reserving worst-case blocks for delalloc
writes.  Writing out delalloc data will convert those worst-case
predictions into usually smaller actual usage, freeing up space
before we hit ENOSPC based on this speculation.

Thanks to Jens for the suggestion for the helper function,
& the naming help.

I've made the helper return status on whether writeback was
started even though I don't plan to use it in the ext4 patch;
it seems like it would be potentially useful to test this
in some cases.

Signed-off-by: Eric Sandeen 
Acked-by: Jan Kara 
Cc: maximilian attems 
Signed-off-by: Greg Kroah-Hartman

writeback: pass in super_block to bdi_start_writeback()

2009-09-25T22:10:40+00:00

Sometimes we only want to write pages from a specific super_block,
so allow that to be passed in.

This fixes a problem with commit 56a131dcf7ed36c3c6e36bea448b674ea85ed5bb
causing writeback on all super_blocks on a bdi, where we only really
want to sync a specific sb from writeback_inodes_sb().

Signed-off-by: Jens Axboe

writeback: writeback_inodes_sb() should use bdi_start_writeback()

2009-09-25T16:08:26+00:00

Pointless to iterate other devices looking for a super, when
we have a bdi mapping.

Signed-off-by: Jens Axboe

writeback: don't delay inodes redirtied by a fast dirtier

2009-09-25T16:08:26+00:00

Debug traces show that in per-bdi writeback, the inode under writeback
almost always get redirtied by a busy dirtier.  We used to call
redirty_tail() in this case, which could delay inode for up to 30s.

This is unacceptable because it now happens so frequently for plain cp/dd,
that the accumulated delays could make writeback of big files very slow.

So let's distinguish between data redirty and metadata only redirty.
The first one is caused by a busy dirtier, while the latter one could
happen in XFS, NFS, etc. when they are doing delalloc or updating isize.

The inode being busy dirtied will now be requeued for next io, while
the inode being redirtied by fs will continue to be delayed to avoid
repeated IO.

CC: Jan Kara 
CC: Theodore Ts'o 
CC: Dave Chinner 
CC: Chris Mason 
CC: Christoph Hellwig 
Signed-off-by: Wu Fengguang 
Signed-off-by: Jens Axboe

writeback: make the super_block pinning more efficient

2009-09-25T16:08:26+00:00

Currently we pin the inode->i_sb for every single inode. This
increases cache traffic on sb->s_umount sem. Lets instead
cache the inode sb pin state and keep the super_block pinned
for as long as keep writing out inodes from the same
super_block.

Signed-off-by: Jens Axboe

writeback: don't resort for a single super_block in move_expired_inodes()

2009-09-25T16:08:26+00:00

If we only moved inodes from a single super_block to the temporary
list, there's no point in doing a resort for multiple super_blocks.

Signed-off-by: Jens Axboe

writeback: move inodes from one super_block together

2009-09-25T16:08:25+00:00

__mark_inode_dirty adds inode to wb dirty list in random order. If a disk has
several partitions, writeback might keep spindle moving between partitions.
To reduce the move, better write big chunk of one partition and then move to
another. Inodes from one fs usually are in one partion, so idealy move indoes
from one fs together should reduce spindle move. This patch tries to address
this. Before per-bdi writeback is added, the behavior is write indoes
from one fs first and then another, so the patch restores previous behavior.
The loop in the patch is a bit ugly, should we add a dirty list for each
superblock in bdi_writeback?

Test in a two partition disk with attached fio script shows about 3% ~ 6%
improvement.

Signed-off-by: Shaohua Li 
Reviewed-by: Wu Fengguang 
Signed-off-by: Jens Axboe

writeback: get rid to incorrect references to pdflush in comments

2009-09-25T16:08:25+00:00

Signed-off-by: Jens Axboe

writeback: improve readability of the wb_writeback() continue/break logic

2009-09-25T16:08:25+00:00

And throw some comments in there, too.

Reviewed-by: Wu Fengguang 
Signed-off-by: Jens Axboe