The allocation stack switch at xfs_bmapi_allocate() has served it's
purpose, but is no longer a sufficient solution to the stack usage
problem we have in the XFS allocation path.

Whilst the kernel stack size is now 16k, that is not a valid reason
for undoing all our "keep stack usage down" modifications. What it
does allow us to do is have the freedom to refine and perfect the
modifications knowing that if we get it wrong it won't blow up in
our faces - we have a safety net now.

This is important because we still have the issue of older kernels
having smaller stacks and that they are still supported and are
demonstrating a wide range of different stack overflows.  Red Hat
has several open bugs for allocation based stack overflows from
directory modifications and direct IO block allocation and these
problems still need to be solved. If we can solve them upstream,
then distro's won't need to bake their own unique solutions.

To that end, I've observed that every allocation based stack
overflow report has had a specific characteristic - it has happened
during or directly after a bmap btree block split. That event
requires a new block to be allocated to the tree, and so we
effectively stack one allocation stack on top of another, and that's
when we get into trouble.

A further observation is that bmap btree block splits are much rarer
than writeback allocation - over a range of different workloads I've
observed the ratio of bmap btree inserts to splits ranges from 100:1
(xfstests run) to 10000:1 (local VM image server with sparse files
that range in the hundreds of thousands to millions of extents).
Either way, bmap btree split events are much, much rarer than
allocation events.

Finally, we have to move the kswapd state to the allocation workqueue
work when allocation is done on behalf of kswapd. This is proving to
cause significant perturbation in performance under memory pressure
and appears to be generating allocation deadlock warnings under some
workloads, so avoiding the use of a workqueue for the majority of
kswapd writeback allocation will minimise the impact of such
behaviour.

Hence it makes sense to move the stack switch to xfs_btree_split()
and only do it for bmap btree splits. Stack switches during
allocation will be much rarer, so there won't be significant
performacne overhead caused by switching stacks. The worse case
stack from all allocation paths will be split, not just writeback.
And the majority of memory allocations will be done in the correct
context (e.g. kswapd) without causing additional latency, and so we
simplify the memory reclaim interactions between processes,
workqueues and kswapd.

The worst stack I've been able to generate with this patch in place
is 5600 bytes deep. It's very revealing because we exit XFS at:

37)     1768      64   kmem_cache_alloc+0x13b/0x170

about 1800 bytes of stack consumed, and the remaining 3800 bytes
(and 36 functions) is memory reclaim, swap and the IO stack. And
this occurs in the inode allocation from an open(O_CREAT) syscall,
not writeback.

The amount of stack being used is much less than I've previously be
able to generate - fs_mark testing has been able to generate stack
usage of around 7k without too much trouble; with this patch it's
only just getting to 5.5k. This is primarily because the metadata
allocation paths (e.g. directory blocks) are no longer causing
double splits on the same stack, and hence now stack tracing is
showing swapping being the worst stack consumer rather than XFS.

Performance of fs_mark inode create workloads is unchanged.
Performance of fs_mark async fsync workloads is consistently good
with context switches reduced by around 150,000/s (30%).
Performance of dbench, streaming IO and postmark is unchanged.
Allocation deadlock warnings have not been seen on the workloads
that generated them since adding this patch.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NBrian Foster <bfoster@redhat.com>
Signed-off-by: NDave Chinner <david@fromorbit.com>

This reverts commit 1f6d6482.

This commit resulted in regressions in performance in low
memory situations where kswapd was doing writeback of delayed
allocation blocks. It resulted in significant parallelism of the
kswapd work and with the special kswapd flags meant that hundreds of
active allocation could dip into kswapd specific memory reserves and
avoid being throttled. This cause a large amount of performance
variation, as well as random OOM-killer invocations that didn't
previously exist.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NBrian Foster <bfoster@redhat.com>
Signed-off-by: NDave Chinner <david@fromorbit.com>

Upon memory pressure, kswapd calls xfs_vm_writepage() from
shrink_page_list(). This can result in delayed allocation occurring
and that gets deferred to the the allocation workqueue.

The allocation then runs outside kswapd context, which means if it
needs memory (and it does to demand page metadata from disk) it can
block in shrink_inactive_list() waiting for IO congestion. These
blocking waits are normally avoiding in kswapd context, so under
memory pressure writeback from kswapd can be arbitrarily delayed by
memory reclaim.

To avoid this, pass the kswapd context to the allocation being done
by the workqueue, so that memory reclaim understands correctly that
the work is being done for kswapd and therefore it is not blocked
and does not delay memory reclaim.

To avoid issues with int->char conversion of flag fields (as noticed
in v1 of this patch) convert the flag fields in the struct
xfs_bmalloca to bool types. pahole indicates these variables are
still single byte variables, so no extra space is consumed by this
change.

cc: <stable@vger.kernel.org>
Reported-by: NTetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NDave Chinner <david@fromorbit.com>

This patch implements fallocate's FALLOC_FL_COLLAPSE_RANGE for XFS.

The semantics of this flag are following:
1) It collapses the range lying between offset and length by removing any data
   blocks which are present in this range and than updates all the logical
   offsets of extents beyond "offset + len" to nullify the hole created by
   removing blocks. In short, it does not leave a hole.
2) It should be used exclusively. No other fallocate flag in combination.
3) Offset and length supplied to fallocate should be fs block size aligned
   in case of xfs and ext4.
4) Collaspe range does not work beyond i_size.
Signed-off-by: NNamjae Jeon <namjae.jeon@samsung.com>
Signed-off-by: NAshish Sangwan <a.sangwan@samsung.com>
Reviewed-by: NDave Chinner <dchinner@redhat.com>
Signed-off-by: NDave Chinner <david@fromorbit.com>

Now that only one caller of xfs_change_file_space is left it can be merged
into said caller.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDave Chinner <dchinner@redhat.com>
Signed-off-by: NBen Myers <bpm@sgi.com>

Call xfs_alloc_file_space or xfs_free_file_space directly from
xfs_file_fallocate instead of going through xfs_change_file_space.

This simplified the code by removing the unessecary marshalling of the
arguments into an xfs_flock64_t structure and allows removing checks that
are already done in the VFS code.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDave Chinner <dchinner@redhat.com>
Signed-off-by: NBen Myers <bpm@sgi.com>

So we don't need xfs_dfrag.h in userspace anymore, move the extent
swap ioctl structure definition to xfs_fs.h where most of the other
ioctl structure definitions are.

Now that we don't need separate files for extent swapping, separate
the basic file descriptor checking code to xfs_ioctl.c, and the code
that does the extent swap operation to xfs_bmap_util.c.  This
cleanly separates the user interface code from the physical
mechanism used to do the extent swap.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NMark Tinguely <tinguely@sgi.com>
Signed-off-by: NBen Myers <bpm@sgi.com>

Now we have xfs_inode.c for holding kernel-only XFS inode
operations, move all the inode operations from xfs_vnodeops.c to
this new file as it holds another set of kernel-only inode
operations. The name of this file traces back to the days of Irix
and it's vnodes which we don't have anymore.

Essentially this move consolidates the inode locking functions
and a bunch of XFS inode operations into the one file. Eventually
the high level functions will be merged into the VFS interface
functions in xfs_iops.c.

This leaves only internal preallocation, EOF block manipulation and
hole punching functions in vnodeops.c. Move these to xfs_bmap_util.c
where we are already consolidating various in-kernel physical extent
manipulation and querying functions.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NMark Tinguely <tinguely@sgi.com>
Signed-off-by: NBen Myers <bpm@sgi.com>

There is a bunch of code in xfs_bmap.c that is kernel specific and
not shared with userspace. To minimise the difference between the
kernel and userspace code, shift this unshared code to
xfs_bmap_util.c, and the declarations to xfs_bmap_util.h.

The biggest issue here is xfs_bmap_finish() - userspace has it's own
definition of this function, and so we need to move it out of
xfs_bmap.[ch]. This means several other files need to include
xfs_bmap_util.h as well.

It also introduces and interesting dance for the stack switching
code in xfs_bmapi_allocate(). The stack switching/workqueue code is
actually moved to xfs_bmap_util.c, so that userspace can simply use
a #define in a header file to connect the dots without needing to
know about the stack switch code at all.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NMark Tinguely <tinguely@sgi.com>
Signed-off-by: NBen Myers <bpm@sgi.com>