Skip to content

K
Kernel

openeuler / Kernel
1 年多 前同步成功

通知 8
Star 0
Fork 0
K
Kernel
切换分支/标签
  1. 02 6月, 2018 7 次提交
  3. 17 5月, 2018 2 次提交
  5. 16 5月, 2018 1 次提交
  7. 10 5月, 2018 2 次提交
    • D
      iomap: Use FUA for pure data O_DSYNC DIO writes · 3460cac1
      由 Dave Chinner 提交于 
      If we are doing direct IO writes with datasync semantics, we often
have to flush metadata changes along with the data write. However,
if we are overwriting existing data, there are no metadata changes
that we need to flush. In this case, optimising the IO by using
FUA write makes sense.

We know from the IOMAP_F_DIRTY flag as to whether a specific inode
requires a metadata flush - this is currently used by DAX to ensure
extent modification as stable in page fault operations. For direct
IO writes, we can use it to determine if we need to flush metadata
or not once the data is on disk.

Hence if we have been returned a mapped extent that is not new and
the IO mapping is not dirty, then we can use a FUA write to provide
datasync semantics. This allows us to short-cut the
generic_write_sync() call in IO completion and hence avoid
unnecessary operations. This makes pure direct IO data write
behaviour identical to the way block devices use REQ_FUA to provide
datasync semantics.

On a FUA enabled device, a synchronous direct IO write workload
(sequential 4k overwrites in 32MB file) had the following results:

# xfs_io -fd -c "pwrite -V 1 -D 0 32m" /mnt/scratch/boo

kernel		time	write()s	write iops	Write b/w
------		----	--------	----------	---------
(no dsync)	 4s	2173/s		2173		8.5MB/s
vanilla		22s	 370/s		 750		1.4MB/s
patched		19s	 420/s		 420		1.6MB/s

The patched code clearly doesn't send cache flushes anymore, but
instead uses FUA (confirmed via blktrace), and performance improves
a bit as a result. However, the benefits will be higher on workloads
that mix O_DSYNC overwrites with other write IO as we won't be
flushing the entire device cache on every DSYNC overwrite IO
anymore.
Signed-Off-By: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
      3460cac1
    • D
      iomap: iomap_dio_rw() handles all sync writes · 4f8ff44b
      由 Dave Chinner 提交于 
      Currently iomap_dio_rw() only handles (data)sync write completions
for AIO. This means we can't optimised non-AIO IO to minimise device
flushes as we can't tell the caller whether a flush is required or
not.

To solve this problem and enable further optimisations, make
iomap_dio_rw responsible for data sync behaviour for all IO, not
just AIO.

In doing so, the sync operation is now accounted as part of the DIO
IO by inode_dio_end(), hence post-IO data stability updates will no
long race against operations that serialise via inode_dio_wait()
such as truncate or hole punch.
Signed-Off-By: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
      4f8ff44b
  9. 29 1月, 2018 1 次提交
  11. 09 1月, 2018 1 次提交
    • D
      iomap: report collisions between directio and buffered writes to userspace · 5a9d929d
      由 Darrick J. Wong 提交于 
      If two programs simultaneously try to write to the same part of a file
via direct IO and buffered IO, there's a chance that the post-diowrite
pagecache invalidation will fail on the dirty page.  When this happens,
the dio write succeeded, which means that the page cache is no longer
coherent with the disk!

Programs are not supposed to mix IO types and this is a clear case of
data corruption, so store an EIO which will be reflected to userspace
during the next fsync.  Replace the WARN_ON with a ratelimited pr_crit
so that the developers have /some/ kind of breadcrumb to track down the
offending program(s) and file(s) involved.
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NLiu Bo <bo.li.liu@oracle.com>
      5a9d929d
  13. 04 11月, 2017 1 次提交
  15. 17 10月, 2017 1 次提交
    • E
      fs: invalidate page cache after end_io() in dio completion · 5e25c269
      由 Eryu Guan 提交于 
      Commit 332391a9 ("fs: Fix page cache inconsistency when mixing
buffered and AIO DIO") moved page cache invalidation from
iomap_dio_rw() to iomap_dio_complete() for iomap based direct write
path, but before the dio->end_io() call, and it re-introdued the bug
fixed by commit c771c14b ("iomap: invalidate page caches should
be after iomap_dio_complete() in direct write").

I found this because fstests generic/418 started failing on XFS with
v4.14-rc3 kernel, which is the regression test for this specific
bug.

So similarly, fix it by moving dio->end_io() (which does the
unwritten extent conversion) before page cache invalidation, to make
sure next buffer read reads the final real allocations not unwritten
extents. I also add some comments about why should end_io() go first
in case we get it wrong again in the future.

Note that, there's no such problem in the non-iomap based direct
write path, because we didn't remove the page cache invalidation
after the ->direct_IO() in generic_file_direct_write() call, but I
decided to fix dio_complete() too so we don't leave a landmine
there, also be consistent with iomap_dio_complete().

Fixes: 332391a9 ("fs: Fix page cache inconsistency when mixing buffered and AIO DIO")
Signed-off-by: NEryu Guan <eguan@redhat.com>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NJan Kara <jack@suse.cz>
Reviewed-by: NLukas Czerner <lczerner@redhat.com>
      5e25c269
  17. 12 10月, 2017 1 次提交
    • A
      iomap_dio_actor(): fix iov_iter bugs · cfe057f7
      由 Al Viro 提交于 
      1) Ignoring return value from iov_iter_zero() is wrong
for iovec-backed case as well as for pipes - it can fail.

2) Failure to fault destination pages in 25Mb into a 50Mb iovec
should not act as if nothing in the area had been read, nevermind
that the first 25Mb might have *already* been read by that point.
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
      cfe057f7
  19. 02 10月, 2017 2 次提交
  21. 27 9月, 2017 1 次提交
    • C
      iomap_dio_rw: Allocate AIO completion queue before submitting dio · 546e7be8
      由 Chandan Rajendra 提交于 
      Executing xfs/104 test in a loop on Linux-v4.13 kernel on a ppc64
machine can cause the following NULL pointer dereference,

.queue_work_on+0x4c/0x80
.iomap_dio_bio_end_io+0xbc/0x1f0
.bio_endio+0x118/0x1f0
.blk_update_request+0xd0/0x470
.blk_mq_end_request+0x24/0xc0
.lo_complete_rq+0x40/0xe0
.__blk_mq_complete_request_remote+0x28/0x40
.flush_smp_call_function_queue+0xc4/0x1e0
.smp_ipi_demux_relaxed+0x8c/0x100
.icp_hv_ipi_action+0x54/0xa0
.__handle_irq_event_percpu+0x84/0x2c0
.handle_irq_event_percpu+0x28/0x80
.handle_percpu_irq+0x78/0xc0
.generic_handle_irq+0x40/0x70
.__do_irq+0x88/0x200
.call_do_irq+0x14/0x24
.do_IRQ+0x84/0x130

This occurs due to the following sequence of events,

1. Allocate dio for Direct I/O write.
2. Invoke iomap_apply() until iov_iter_count() bytes have been submitted.
   - Assume that we have submitted atleast one bio. Hence iomap_dio->ref value
     will be >= 2.
   - If during the second iteration, iomap_apply() ends up returning -ENOSPC, we would
     break out of the loop and since the 'ret' value is a negative number we
     end up not allocating memory for super_block->s_dio_done_wq.
3. Meanwhile, iomap_dio_bio_end_io() is invoked for bios that have been
   submitted and here the code ends up dereferencing the NULL pointer stored
   at super_block->s_dio_done_wq.

This commit fixes the bug by allocating memory for
super_block->s_dio_done_wq before iomap_apply() is invoked.
Reported-by: NEryu Guan <eguan@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Tested-by: NEryu Guan <eguan@redhat.com>
Signed-off-by: NChandan Rajendra <chandan@linux.vnet.ibm.com>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
      546e7be8
  23. 25 9月, 2017 1 次提交
    • L
      fs: Fix page cache inconsistency when mixing buffered and AIO DIO · 332391a9
      由 Lukas Czerner 提交于 
      Currently when mixing buffered reads and asynchronous direct writes it
is possible to end up with the situation where we have stale data in the
page cache while the new data is already written to disk. This is
permanent until the affected pages are flushed away. Despite the fact
that mixing buffered and direct IO is ill-advised it does pose a thread
for a data integrity, is unexpected and should be fixed.

Fix this by deferring completion of asynchronous direct writes to a
process context in the case that there are mapped pages to be found in
the inode. Later before the completion in dio_complete() invalidate
the pages in question. This ensures that after the completion the pages
in the written area are either unmapped, or populated with up-to-date
data. Also do the same for the iomap case which uses
iomap_dio_complete() instead.

This has a side effect of deferring the completion to a process context
for every AIO DIO that happens on inode that has pages mapped. However
since the consensus is that this is ill-advised practice the performance
implication should not be a problem.

This was based on proposal from Jeff Moyer, thanks!
Reviewed-by: NJan Kara <jack@suse.cz>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NJeff Moyer <jmoyer@redhat.com>
Signed-off-by: NLukas Czerner <lczerner@redhat.com>
Signed-off-by: NJens Axboe <axboe@kernel.dk>
      332391a9
  25. 02 9月, 2017 1 次提交
  27. 24 8月, 2017 1 次提交
    • C
      block: replace bi_bdev with a gendisk pointer and partitions index · 74d46992
      由 Christoph Hellwig 提交于 
      This way we don't need a block_device structure to submit I/O.  The
block_device has different life time rules from the gendisk and
request_queue and is usually only available when the block device node
is open.  Other callers need to explicitly create one (e.g. the lightnvm
passthrough code, or the new nvme multipathing code).

For the actual I/O path all that we need is the gendisk, which exists
once per block device.  But given that the block layer also does
partition remapping we additionally need a partition index, which is
used for said remapping in generic_make_request.

Note that all the block drivers generally want request_queue or
sometimes the gendisk, so this removes a layer of indirection all
over the stack.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJens Axboe <axboe@kernel.dk>
      74d46992
  29. 12 8月, 2017 1 次提交
  31. 14 7月, 2017 1 次提交
  33. 03 7月, 2017 1 次提交
  35. 28 6月, 2017 1 次提交
  37. 20 6月, 2017 1 次提交
  39. 09 6月, 2017 1 次提交
  41. 09 5月, 2017 1 次提交
  43. 04 5月, 2017 1 次提交
    • A
      fs: fix data invalidation in the cleancache during direct IO · 55635ba7
      由 Andrey Ryabinin 提交于 
      Patch series "Properly invalidate data in the cleancache", v2.

We've noticed that after direct IO write, buffered read sometimes gets
stale data which is coming from the cleancache.  The reason for this is
that some direct write hooks call call invalidate_inode_pages2[_range]()
conditionally iff mapping->nrpages is not zero, so we may not invalidate
data in the cleancache.

Another odd thing is that we check only for ->nrpages and don't check
for ->nrexceptional, but invalidate_inode_pages2[_range] also
invalidates exceptional entries as well.  So we invalidate exceptional
entries only if ->nrpages != 0? This doesn't feel right.

 - Patch 1 fixes direct IO writes by removing ->nrpages check.
 - Patch 2 fixes similar case in invalidate_bdev().
     Note: I only fixed conditional cleancache_invalidate_inode() here.
       Do we also need to add ->nrexceptional check in into invalidate_bdev()?

 - Patches 3-4: some optimizations.

This patch (of 4):

Some direct IO write fs hooks call invalidate_inode_pages2[_range]()
conditionally iff mapping->nrpages is not zero.  This can't be right,
because invalidate_inode_pages2[_range]() also invalidate data in the
cleancache via cleancache_invalidate_inode() call.  So if page cache is
empty but there is some data in the cleancache, buffered read after
direct IO write would get stale data from the cleancache.

Also it doesn't feel right to check only for ->nrpages because
invalidate_inode_pages2[_range] invalidates exceptional entries as well.

Fix this by calling invalidate_inode_pages2[_range]() regardless of
nrpages state.

Note: nfs,cifs,9p doesn't need similar fix because the never call
cleancache_get_page() (nor directly, nor via mpage_readpage[s]()), so
they are not affected by this bug.

Fixes: c515e1fd ("mm/fs: add hooks to support cleancache")
Link: http://lkml.kernel.org/r/20170424164135.22350-2-aryabinin@virtuozzo.comSigned-off-by: NAndrey Ryabinin <aryabinin@virtuozzo.com>
Reviewed-by: NJan Kara <jack@suse.cz>
Acked-by: NKonrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: Ross Zwisler <ross.zwisler@linux.intel.com>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Alexey Kuznetsov <kuznet@virtuozzo.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Nikolay Borisov <n.borisov.lkml@gmail.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      55635ba7
  45. 26 4月, 2017 2 次提交
    • D
      filesystem-dax: convert to dax_direct_access() · cccbce67
      由 Dan Williams 提交于 
      Now that a dax_device is plumbed through all dax-capable drivers we can
switch from block_device_operations to dax_operations for invoking
->direct_access.

This also lets us kill off some usages of struct blk_dax_ctl on the way
to its eventual removal.
Suggested-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NDan Williams <dan.j.williams@intel.com>
      cccbce67
    • C
      iomap_dio_rw: Prevent reading file data beyond iomap_dio->i_size · a008c31c
      由 Chandan Rajendra 提交于 
      On a ppc64 machine executing overlayfs/019 with xfs as the lower and
upper filesystem causes the following call trace,

WARNING: CPU: 2 PID: 8034 at /root/repos/linux/fs/iomap.c:765 .iomap_dio_actor+0xcc/0x420
Modules linked in:
CPU: 2 PID: 8034 Comm: fsstress Tainted: G             L  4.11.0-rc5-next-20170405 #100
task: c000000631314880 task.stack: c0000003915d4000
NIP: c00000000035a72c LR: c00000000035a6f4 CTR: c00000000035a660
REGS: c0000003915d7570 TRAP: 0700   Tainted: G             L   (4.11.0-rc5-next-20170405)
MSR: 800000000282b032 <SF,VEC,VSX,EE,FP,ME,IR,DR,RI>
  CR: 24004284  XER: 00000000
CFAR: c0000000006f7190 SOFTE: 1
GPR00: c00000000035a6f4 c0000003915d77f0 c0000000015a3f00 000000007c22f600
GPR04: 000000000022d000 0000000000002600 c0000003b2d56360 c0000003915d7960
GPR08: c0000003915d7cd0 0000000000000002 0000000000002600 c000000000521cc0
GPR12: 0000000024004284 c00000000fd80a00 000000004b04ae64 ffffffffffffffff
GPR16: 000000001000ca70 0000000000000000 c0000003b2d56380 c00000000153d2b8
GPR20: 0000000000000010 c0000003bc87bac8 0000000000223000 000000000022f5ff
GPR24: c0000003b2d56360 000000000000000c 0000000000002600 000000000022d000
GPR28: 0000000000000000 c0000003915d7960 c0000003b2d56360 00000000000001ff
NIP [c00000000035a72c] .iomap_dio_actor+0xcc/0x420
LR [c00000000035a6f4] .iomap_dio_actor+0x94/0x420
Call Trace:
[c0000003915d77f0] [c00000000035a6f4] .iomap_dio_actor+0x94/0x420 (unreliable)
[c0000003915d78f0] [c00000000035b9f4] .iomap_apply+0xf4/0x1f0
[c0000003915d79d0] [c00000000035c320] .iomap_dio_rw+0x230/0x420
[c0000003915d7ae0] [c000000000512a14] .xfs_file_dio_aio_read+0x84/0x160
[c0000003915d7b80] [c000000000512d24] .xfs_file_read_iter+0x104/0x130
[c0000003915d7c10] [c0000000002d6234] .__vfs_read+0x114/0x1a0
[c0000003915d7cf0] [c0000000002d7a8c] .vfs_read+0xac/0x1a0
[c0000003915d7d90] [c0000000002d96b8] .SyS_read+0x58/0x100
[c0000003915d7e30] [c00000000000b8e0] system_call+0x38/0xfc
Instruction dump:
78630020 7f831b78 7ffc07b4 7c7ce039 40820360 a13d0018 2f890003 419e0288
2f890004 419e00a0 2f890001 419e02a8 <0fe00000> 3b80fffb 38210100 7f83e378

The above problem can also be recreated on a regular xfs filesystem
using the command,

$ fsstress -d /mnt -l 1000 -n 1000 -p 1000

The reason for the call trace is,
1. When 'reserving' blocks for delayed allocation , XFS reserves more
   blocks (i.e. past file's current EOF) than required. This is done
   because XFS assumes that userspace might write more data and hence
   'reserving' more blocks might lead to the file's new data being
   stored contiguously on disk.
2. The in-memory 'struct xfs_bmbt_irec' mapping the file's last extent would
   then cover the prealloc-ed EOF blocks in addition to the regular blocks.
3. When flushing the dirty blocks to disk, we only flush data till the
   file's EOF. But before writing out the dirty data, we allocate blocks
   on the disk for holding the file's new data. This allocation includes
   the blocks that are part of the 'prealloc EOF blocks'.
4. Later, when the last reference to the inode is being closed, XFS frees the
   unused 'prealloc EOF blocks' in xfs_inactive().

In step 3 above, When allocating space on disk for the delayed allocation
range, the space allocator might sometimes allocate less blocks than
required. If such an allocation ends right at the current EOF of the
file, We will not be able to clear the "delayed allocation" flag for the
'prealloc EOF blocks', since we won't have dirty buffer heads associated
with that range of the file.

In such a situation if a Direct I/O read operation is performed on file
range [X, Y] (where X < EOF and Y > EOF), we flush dirty data in the
range [X, Y] and invalidate page cache for that range (Refer to
iomap_dio_rw()). Later for performing the Direct I/O read, XFS obtains
the extent items (which are still cached in memory) for the file
range. When doing so we are not supposed to get an extent item with
IOMAP_DELALLOC flag set, since the previous "flush" operation should
have converted any delayed allocation data in the range [X, Y]. Hence we
end up hitting a WARN_ON_ONCE(1) statement in iomap_dio_actor().

This commit fixes the bug by preventing the read operation from going
beyond iomap_dio->i_size.
Reported-by: NSanthosh G <santhog4@linux.vnet.ibm.com>
Signed-off-by: NChandan Rajendra <chandan@linux.vnet.ibm.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
      a008c31c
  47. 07 3月, 2017 1 次提交
    • E
      iomap: invalidate page caches should be after iomap_dio_complete() in direct write · c771c14b
      由 Eryu Guan 提交于 
      After XFS switching to iomap based DIO (commit acdda3aa ("xfs:
use iomap_dio_rw")), I started to notice dio29/dio30 tests failures
from LTP run on ppc64 hosts, and they can be reproduced on x86_64
hosts with 512B/1k block size XFS too.

dio29	diotest3 -b 65536 -n 100 -i 1000 -o 1024000
dio30	diotest6 -b 65536 -n 100 -i 1000 -o 1024000

The failure message is like:
bufcmp: offset 0: Expected: 0x62, got 0x0
diotest03    1  TPASS  :  Read with Direct IO, Write without
diotest03    2  TFAIL  :  diotest3.c:142: comparsion failed; child=98 offset=1425408
diotest03    3  TFAIL  :  diotest3.c:194: Write Direct-child 98 failed

Direct write wrote 0x62 but buffer read got zero. This is because,
when doing direct write to a hole or preallocated file, we
invalidate the page caches before converting the extent from
unwritten state to normal state, which is done by
iomap_dio_complete(), thus leave a window for other buffer reader to
cache the unwritten state extent.

Consider this case, with sub-page blocksize XFS, two processes are
direct writing to different blocksize-aligned regions (say 512B) of
the same preallocated file, and reading the region back via buffered
I/O to compare contents.

process A, region [0,512]		process B, region [512,1024]
xfs_file_write_iter
 xfs_file_aio_dio_write
  iomap_dio_rw
   iomap_apply
   invalidate_inode_pages2_range
   					xfs_file_write_iter
				 	xfs_file_aio_dio_write
					  iomap_dio_rw
					   iomap_apply
					   invalidate_inode_pages2_range
					   iomap_dio_complete
					xfs_file_read_iter
					 xfs_file_buffered_aio_read
					  generic_file_read_iter
					   do_generic_file_read
					    <readahead fills pagecache with 0>
   iomap_dio_complete
xfs_file_read_iter
 <read gets 0 from pagecache>

Process A first invalidates page caches, at this point the
underlying extent is still in unwritten state (iomap_dio_complete
not called yet), and process B finishs direct write and populates
page caches via readahead, which caches zeros in page for region A,
then process A reads zeros from page cache, instead of the actual
data.

Fix it by invalidating page caches after converting unwritten extent
to make sure we read content from disk after extent state changed,
as what we did before switching to iomap based dio.

Also introduce a new 'start' variable to save the original write
offset (iomap_dio_complete() updates iocb->ki_pos), and a 'err'
variable for invalidating caches result, cause we can't reuse 'ret'
anymore.
Signed-off-by: NEryu Guan <eguan@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
      c771c14b
  49. 02 3月, 2017 1 次提交
  51. 28 2月, 2017 1 次提交
  53. 25 2月, 2017 1 次提交
  55. 04 2月, 2017 1 次提交
    • M
      fs: break out of iomap_file_buffered_write on fatal signals · d1908f52
      由 Michal Hocko 提交于 
      Tetsuo has noticed that an OOM stress test which performs large write
requests can cause the full memory reserves depletion.  He has tracked
this down to the following path

	__alloc_pages_nodemask+0x436/0x4d0
	alloc_pages_current+0x97/0x1b0
	__page_cache_alloc+0x15d/0x1a0          mm/filemap.c:728
	pagecache_get_page+0x5a/0x2b0           mm/filemap.c:1331
	grab_cache_page_write_begin+0x23/0x40   mm/filemap.c:2773
	iomap_write_begin+0x50/0xd0             fs/iomap.c:118
	iomap_write_actor+0xb5/0x1a0            fs/iomap.c:190
	? iomap_write_end+0x80/0x80             fs/iomap.c:150
	iomap_apply+0xb3/0x130                  fs/iomap.c:79
	iomap_file_buffered_write+0x68/0xa0     fs/iomap.c:243
	? iomap_write_end+0x80/0x80
	xfs_file_buffered_aio_write+0x132/0x390 [xfs]
	? remove_wait_queue+0x59/0x60
	xfs_file_write_iter+0x90/0x130 [xfs]
	__vfs_write+0xe5/0x140
	vfs_write+0xc7/0x1f0
	? syscall_trace_enter+0x1d0/0x380
	SyS_write+0x58/0xc0
	do_syscall_64+0x6c/0x200
	entry_SYSCALL64_slow_path+0x25/0x25

the oom victim has access to all memory reserves to make a forward
progress to exit easier.  But iomap_file_buffered_write and other
callers of iomap_apply loop to complete the full request.  We need to
check for fatal signals and back off with a short write instead.

As the iomap_apply delegates all the work down to the actor we have to
hook into those.  All callers that work with the page cache are calling
iomap_write_begin so we will check for signals there.  dax_iomap_actor
has to handle the situation explicitly because it copies data to the
userspace directly.  Other callers like iomap_page_mkwrite work on a
single page or iomap_fiemap_actor do not allocate memory based on the
given len.

Fixes: 68a9f5e7 ("xfs: implement iomap based buffered write path")
Link: http://lkml.kernel.org/r/20170201092706.9966-2-mhocko@kernel.orgSigned-off-by: NMichal Hocko <mhocko@suse.com>
Reported-by: NTetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: <stable@vger.kernel.org>	[4.8+]
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
      d1908f52
  57. 31 1月, 2017 1 次提交
  59. 30 11月, 2016 1 次提交
    • C
      iomap: implement direct I/O · ff6a9292
      由 Christoph Hellwig 提交于 
      This adds a full fledget direct I/O implementation using the iomap
interface. Full fledged in this case means all features are supported:
AIO, vectored I/O, any iov_iter type including kernel pointers, bvecs
and pipes, support for hole filling and async apending writes.  It does
not mean supporting all the warts of the old generic code.  We expect
i_rwsem to be held over the duration of the call, and we expect to
maintain i_dio_count ourselves, and we pass on any kinds of mapping
to the file system for now.

The algorithm used is very simple: We use iomap_apply to iterate over
the range of the I/O, and then we use the new bio_iov_iter_get_pages
helper to lock down the user range for the size of the extent.
bio_iov_iter_get_pages can currently lock down twice as many pages as
the old direct I/O code did, which means that we will have a better
batch factor for everything but overwrites of badly fragmented files.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NKent Overstreet <kent.overstreet@gmail.com>
Tested-by: NJens Axboe <axboe@fb.com>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDave Chinner <david@fromorbit.com>
      ff6a9292