Make sure that not only sync_filesystem but all callers of writeback_inodes_sb
have the superblock protected against remount.  As-is this disables all
functionality for these callers, but the next patch relies on this locking to
fix writeback_inodes_sb for sync_filesystem.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

If we want to rely on s_umount in the caller we need to wait for completion
of the I/O submission before returning to the caller.  Refactor
bdi_sync_writeback into a bdi_queue_work_onstack helper and use it for this
case.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

The code dealing with bdi_work->state and completion of a bdi_work is a
major mess currently.  This patch makes sure we directly use one set of
flags to deal with it, and use it consistently, which means:

 - always notify about completion from the rcu callback.  We only ever
   wait for it from on-stack callers, so this simplification does not
   even cause a theoretical slowdown currently.  It also makes sure we
   don't miss out on the notification if we ever add other callers to
   wait for it.
 - make earlier completion notification depending on the on-stack
   allocation, not the sync mode.  If we introduce new callers that
   want to do WB_SYNC_NONE writeback from on-stack callers this will
   be nessecary.

Also rename bdi_wait_on_work_clear to bdi_wait_on_work_done and inline
a few small functions into their only caller to make the code
understandable.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

This reverts commit e913fc82.

We are investigating a hang associated with the WB_SYNC_NONE changes,
so revert them for now.

Conflicts:

	fs/fs-writeback.c
	mm/page-writeback.c
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

This reverts commit 7c8a3554.

We are investigating a hang associated with the WB_SYNC_NONE changes,
so revert them for now.
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

When wb_writeback() hasn't written anything it will re-acquire the inode
lock before calling inode_wait_for_writeback.

This change tests the sync bit first so that is doesn't need to drop &
re-acquire the lock if the inode became available while wb_writeback() was
waiting to get the lock.
Signed-off-by: NRichard Kennedy <richard@rsk.demon.co.uk>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: Jens Axboe <jens.axboe@oracle.com>
Cc: Wu Fengguang <fengguang.wu@intel.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

This fixes sparse noise:
  error: dubious one-bit signed bitfield
Signed-off-by: NH Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

Calling schedule without setting the task state to non-running will
return immediately, so ensure that we set it properly and check our
sleep conditions after doing so.

This is a fixup for commit 69b62d01.
Signed-off-by: NJens Axboe <jens.axboe@oracle.com>

Even if the writeout itself isn't a data integrity operation, we need
to ensure that the caller doesn't drop the sb umount sem before we
have actually done the writeback.

This is a fixup for commit e913fc82.
Signed-off-by: NJens Axboe <jens.axboe@oracle.com>

Filesystems with delalloc support may dirty inode during writepages.
As result inode will have dirty metadata flags even after write_inode.
In fact we have two dedicated functions for proper data and metadata
writeback. It is reasonable to separate flags updates in two stages.

https://bugzilla.kernel.org/show_bug.cgi?id=15906Signed-off-by: NDmitry Monakhov <dmonakhov@openvz.org>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJens Axboe <jens.axboe@oracle.com>

When umount calls sync_filesystem(), we first do a WB_SYNC_NONE
writeback to kick off writeback of pending dirty inodes, then follow
that up with a WB_SYNC_ALL to wait for it. Since umount already holds
the sb s_umount mutex, WB_SYNC_NONE ends up doing nothing and all
writeback happens as WB_SYNC_ALL. This can greatly slow down umount,
since WB_SYNC_ALL writeback is a data integrity operation and thus
a bigger hammer than simple WB_SYNC_NONE. For barrier aware file systems
it's a lot slower.
Signed-off-by: NJens Axboe <jens.axboe@oracle.com>

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().

Cc: stable@kernel.org
Reported-by: NJustin Maggard <jmaggard10@gmail.com>
Signed-off-by: NJens Axboe <jens.axboe@oracle.com>

include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h

percpu.h is included by sched.h and module.h and thus ends up being
included when building most .c files.  percpu.h includes slab.h which
in turn includes gfp.h making everything defined by the two files
universally available and complicating inclusion dependencies.

percpu.h -> slab.h dependency is about to be removed.  Prepare for
this change by updating users of gfp and slab facilities include those
headers directly instead of assuming availability.  As this conversion
needs to touch large number of source files, the following script is
used as the basis of conversion.

  http://userweb.kernel.org/~tj/misc/slabh-sweep.py

The script does the followings.

* Scan files for gfp and slab usages and update includes such that
  only the necessary includes are there.  ie. if only gfp is used,
  gfp.h, if slab is used, slab.h.

* When the script inserts a new include, it looks at the include
  blocks and try to put the new include such that its order conforms
  to its surrounding.  It's put in the include block which contains
  core kernel includes, in the same order that the rest are ordered -
  alphabetical, Christmas tree, rev-Xmas-tree or at the end if there
  doesn't seem to be any matching order.

* If the script can't find a place to put a new include (mostly
  because the file doesn't have fitting include block), it prints out
  an error message indicating which .h file needs to be added to the
  file.

The conversion was done in the following steps.

1. The initial automatic conversion of all .c files updated slightly
   over 4000 files, deleting around 700 includes and adding ~480 gfp.h
   and ~3000 slab.h inclusions.  The script emitted errors for ~400
   files.

2. Each error was manually checked.  Some didn't need the inclusion,
   some needed manual addition while adding it to implementation .h or
   embedding .c file was more appropriate for others.  This step added
   inclusions to around 150 files.

3. The script was run again and the output was compared to the edits
   from #2 to make sure no file was left behind.

4. Several build tests were done and a couple of problems were fixed.
   e.g. lib/decompress_*.c used malloc/free() wrappers around slab
   APIs requiring slab.h to be added manually.

5. The script was run on all .h files but without automatically
   editing them as sprinkling gfp.h and slab.h inclusions around .h
   files could easily lead to inclusion dependency hell.  Most gfp.h
   inclusion directives were ignored as stuff from gfp.h was usually
   wildly available and often used in preprocessor macros.  Each
   slab.h inclusion directive was examined and added manually as
   necessary.

6. percpu.h was updated not to include slab.h.

7. Build test were done on the following configurations and failures
   were fixed.  CONFIG_GCOV_KERNEL was turned off for all tests (as my
   distributed build env didn't work with gcov compiles) and a few
   more options had to be turned off depending on archs to make things
   build (like ipr on powerpc/64 which failed due to missing writeq).

   * x86 and x86_64 UP and SMP allmodconfig and a custom test config.
   * powerpc and powerpc64 SMP allmodconfig
   * sparc and sparc64 SMP allmodconfig
   * ia64 SMP allmodconfig
   * s390 SMP allmodconfig
   * alpha SMP allmodconfig
   * um on x86_64 SMP allmodconfig

8. percpu.h modifications were reverted so that it could be applied as
   a separate patch and serve as bisection point.

Given the fact that I had only a couple of failures from tests on step
6, I'm fairly confident about the coverage of this conversion patch.
If there is a breakage, it's likely to be something in one of the arch
headers which should be easily discoverable easily on most builds of
the specific arch.
Signed-off-by: NTejun Heo <tj@kernel.org>
Guess-its-ok-by: NChristoph Lameter <cl@linux-foundation.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>

Do not pin/unpin superblock for every inode in writeback_inodes_wb(), pin
it for the whole group of inodes which belong to the same superblock and
call writeback_sb_inodes() handler for them.
Signed-off-by: NEdward Shishkin <edward.shishkin@gmail.com>
Cc: Jens Axboe <jens.axboe@oracle.com>
Cc: Wu Fengguang <fengguang.wu@intel.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NJens Axboe <jens.axboe@oracle.com>

This gives the filesystem more information about the writeback that
is happening.  Trond requested this for the NFS unstable write handling,
and other filesystems might benefit from this too by beeing able to
distinguish between the different callers in more detail.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

Similar to the fsync issue fixed a while ago in commit
2daea67e we need to write for data to
actually hit the disk before writing out the metadata to guarantee
data integrity for filesystems that modify the inode in the data I/O
completion path.  Currently XFS and NFS handle this manually, and AFS
has a write_inode method that does nothing but waiting for data, while
others are possibly missing out on this.

Fortunately this change has a lot less impact than the fsync change
as none of the write_inode methods starts data writeout of any form
by itself.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

Fix the following htmldocs warning:

  Warning(fs/fs-writeback.c:255): No description found for parameter 'sb'
Signed-off-by: NJaswinder Singh Rajput <jaswinderrajput@gmail.com>
Signed-off-by: NRandy Dunlap <randy.dunlap@oracle.com>
Acked-by: NWu Fengguang <fengguang.wu@intel.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Jan Kara <jack@suse.cz>
Cc: Jens Axboe <jens.axboe@oracle.com>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

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: NEric Sandeen <sandeen@redhat.com>
Acked-by: NJan Kara <jack@suse.cz>

- no one is calling wb_writeback and write_cache_pages with
  wbc.nonblocking=1 any more
- lumpy pageout will want to do nonblocking writeback without the
  congestion wait

So remove the congestion checks as suggested by Chris.
Signed-off-by: NWu Fengguang <fengguang.wu@intel.com>
Cc: Chris Mason <chris.mason@oracle.com>
Cc: Jens Axboe <jens.axboe@oracle.com>
Cc: Trond Myklebust <Trond.Myklebust@netapp.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Dave Chinner <david@fromorbit.com>
Cc: Evgeniy Polyakov <zbr@ioremap.net>
Cc: Alex Elder <aelder@sgi.com>
Signed-off-by: NJens Axboe <jens.axboe@oracle.com>

It will lower the flush priority for NFS, and maybe more in future.
Signed-off-by: NWu Fengguang <fengguang.wu@intel.com>
Cc: Trond Myklebust <Trond.Myklebust@netapp.com>
Cc: Jens Axboe <jens.axboe@oracle.com>
Cc: Christoph Hellwig <hch@infradead.org>
Signed-off-by: NJens Axboe <jens.axboe@oracle.com>

This is dead code because no bdi flush thread will be started for
!bdi_cap_writeback_dirty bdi.
Signed-off-by: NWu Fengguang <fengguang.wu@intel.com>
Cc: Jens Axboe <jens.axboe@oracle.com>
Cc: Trond Myklebust <Trond.Myklebust@netapp.com>
Cc: Christoph Hellwig <hch@infradead.org>
Signed-off-by: NJens Axboe <jens.axboe@oracle.com>

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 56a131dc
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: NJens Axboe <jens.axboe@oracle.com>

Pointless to iterate other devices looking for a super, when
we have a bdi mapping.
Signed-off-by: NJens Axboe <jens.axboe@oracle.com>

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 <jack@suse.cz>
CC: Theodore Ts'o <tytso@mit.edu>
CC: Dave Chinner <david@fromorbit.com>
CC: Chris Mason <chris.mason@oracle.com>
CC: Christoph Hellwig <hch@infradead.org>
Signed-off-by: NWu Fengguang <fengguang.wu@intel.com>
Signed-off-by: NJens Axboe <jens.axboe@oracle.com>

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: NJens Axboe <jens.axboe@oracle.com>

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: NJens Axboe <jens.axboe@oracle.com>

__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: NShaohua Li <shaohua.li@intel.com>
Reviewed-by: NWu Fengguang <fengguang.wu@intel.com>
Signed-off-by: NJens Axboe <jens.axboe@oracle.com>

Signed-off-by: NJens Axboe <jens.axboe@oracle.com>

And throw some comments in there, too.
Reviewed-by: NWu Fengguang <fengguang.wu@intel.com>
Signed-off-by: NJens Axboe <jens.axboe@oracle.com>

Make the if-else straight in writeback_single_inode().
No behavior change.

Cc: Jan Kara <jack@suse.cz>
Cc: Michael Rubin <mrubin@google.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: NFengguang Wu <wfg@mail.ustc.edu.cn>
Signed-off-by: NJens Axboe <jens.axboe@oracle.com>

Fix the kupdate case, which disregards wbc.more_io and stop writeback
prematurely even when there are more inodes to be synced.

wbc.more_io should always be respected.

Also remove the pages_skipped check. It will set when some page(s) of some
inode(s) cannot be written for now. Such inodes will be delayed for a while.
This variable has nothing to do with whether there are other writeable inodes.

CC: Jan Kara <jack@suse.cz>
CC: Dave Chinner <david@fromorbit.com>
CC: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: NWu Fengguang <fengguang.wu@intel.com>
Signed-off-by: NJens Axboe <jens.axboe@oracle.com>

Treat bdi_start_writeback(0) as a special request to do background write,
and stop such work when we are below the background dirty threshold.

Also simplify the (nr_pages <= 0) checks. Since we already pass in
nr_pages=LONG_MAX for WB_SYNC_ALL and background writes, we don't
need to worry about it being decreased to zero.
Reported-by: NRichard Kennedy <richard@rsk.demon.co.uk>
CC: Jan Kara <jack@suse.cz>
Acked-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: NWu Fengguang <fengguang.wu@intel.com>
Signed-off-by: NJens Axboe <jens.axboe@oracle.com>

If all inodes are under writeback (e.g. in case when there's only one inode
with dirty pages), wb_writeback() with WB_SYNC_NONE work basically degrades
to busylooping until I_SYNC flags of the inode is cleared. Fix the problem by
waiting on I_SYNC flags of an inode on b_more_io list in case we failed to
write anything.
Tested-by: NWu Fengguang <fengguang.wu@intel.com>
Signed-off-by: NJan Kara <jack@suse.cz>
Signed-off-by: NJens Axboe <jens.axboe@oracle.com>

wb_clear_pending AFAIKS should not be called after the item has been
put on the list, except by the worker threads. It could lead to the
situation where the refcount is decremented below 0 and cause lots of
problems.

Presumably the !wb_has_dirty_io case is not a common one, so it can
be discovered when the thread wakes up to check?

Also add a comment in bdi_work_clear.
Signed-off-by: NNick Piggin <npiggin@suse.de>
Signed-off-by: NJens Axboe <jens.axboe@oracle.com>

By the time bdi_work_on_stack gets evaluated again in bdi_work_free, it
can already have been deallocated and used for something else in the
!on stack case, giving a false positive in this test and causing
corruption.
Signed-off-by: NNick Piggin <npiggin@suse.de>
Signed-off-by: NJens Axboe <jens.axboe@oracle.com>

If you're going to do an atomic RMW on each list entry, there's not much
point in all the RCU complexities of the list walking. This is only going
to help the multi-thread case I guess, but it doesn't hurt to do now.
Signed-off-by: NNick Piggin <npiggin@suse.de>
Signed-off-by: NJens Axboe <jens.axboe@oracle.com>

list_add_tail_rcu contains required barriers.
Signed-off-by: NNick Piggin <npiggin@suse.de>
Signed-off-by: NJens Axboe <jens.axboe@oracle.com>

Gets rid of a manual set_current_state().
Signed-off-by: NJens Axboe <jens.axboe@oracle.com>

And document its retriever, get_next_work_item().
Acked-by: NJan Kara <jack@suse.cz>
Signed-off-by: NJens Axboe <jens.axboe@oracle.com>

bdi_start_writeback() is currently split into two paths, one for
WB_SYNC_NONE and one for WB_SYNC_ALL. Add bdi_sync_writeback()
for WB_SYNC_ALL writeback and let bdi_start_writeback() handle
only WB_SYNC_NONE.

Push down the writeback_control allocation and only accept the
parameters that make sense for each function. This cleans up
the API considerably.
Signed-off-by: NJens Axboe <jens.axboe@oracle.com>