| Age | Commit message (Collapse) | Author |
|---|
|
Add a WARN_ON_ONCE if the file isn't actually open for write. This
matches the check done in vfs_write, but actually warn warns as a
kernel user calling write on a file not opened for writing is a pretty
obvious programming error.
Signed-off-by: Christoph Hellwig <hch@lst.de>
|
|
This is a very special interface that skips sb_writes protection, and not
used by modules anymore.
Signed-off-by: Christoph Hellwig <hch@lst.de>
|
|
While pipes don't really need sb_writers projection, __kernel_write is an
interface better kept private, and the additional rw_verify_area does not
hurt here.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Ian Kent <raven@themaw.net>
|
|
__kernel_write doesn't take a sb_writers references, which we need here.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Howells <dhowells@redhat.com>
|
|
Added a new gc_urgent mode, GC_URGENT_LOW, in which mode
F2FS will lower the bar of checking idle in order to
process outstanding discard commands and GC a little bit
aggressively.
Signed-off-by: Daeho Jeong <daehojeong@google.com>
Reviewed-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
|
|
If two readahead threads having same offset enter in readpages, every read
IOs are split and issued to the disk which giving lower bandwidth.
This patch tries to avoid redundant readahead calls.
Fixes one build error reported by Randy.
Fix build error when F2FS_FS_COMPRESSION is not set/enabled.
This label is needed in either case.
../fs/f2fs/data.c: In function ‘f2fs_mpage_readpages’:
../fs/f2fs/data.c:2327:5: error: label ‘next_page’ used but not defined
goto next_page;
Reviewed-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
|
|
If f2fs_grab_cache_page() fails, it needs to return -ENOMEM.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
|
|
The parameter op_flag is not used in f2fs_get_read_data_page(),
but it is used in f2fs_grab_read_bio(). Obviously, op_flag is
not passed to f2fs_grab_read_bio() successfully. We need to add
parameter in f2fs_submit_page_read() to pass it.
The case:
- gc_data_segment
- f2fs_get_read_data_page(.., op_flag = REQ_RAHEAD,..)
- f2fs_submit_page_read
- f2fs_grab_read_bio(.., op_flag = 0, ..)
Signed-off-by: Jia Yang <jiayang5@huawei.com>
Reviewed-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
|
|
to two independent functions:
- f2fs_allocate_new_segment() for specified type segment allocation
- f2fs_allocate_new_segments() for all data type segments allocation
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
|
|
if get_node_path() return -E2BIG and trace of
f2fs_truncate_inode_blocks_enter/exit enabled
then the matching-pair of trace_exit will lost
in log.
Signed-off-by: Yubo Feng <fengyubo3@huawei.com>
Reviewed-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
|
|
This patch is to fix a crash:
#3 [ffffb6580689f898] oops_end at ffffffffa2835bc2
#4 [ffffb6580689f8b8] no_context at ffffffffa28766e7
#5 [ffffb6580689f920] async_page_fault at ffffffffa320135e
[exception RIP: f2fs_is_compressed_page+34]
RIP: ffffffffa2ba83a2 RSP: ffffb6580689f9d8 RFLAGS: 00010213
RAX: 0000000000000001 RBX: fffffc0f50b34bc0 RCX: 0000000000002122
RDX: 0000000000002123 RSI: 0000000000000c00 RDI: fffffc0f50b34bc0
RBP: ffff97e815a40178 R8: 0000000000000000 R9: ffff97e83ffc9000
R10: 0000000000032300 R11: 0000000000032380 R12: ffffb6580689fa38
R13: fffffc0f50b34bc0 R14: ffff97e825cbd000 R15: 0000000000000c00
ORIG_RAX: ffffffffffffffff CS: 0010 SS: 0018
#6 [ffffb6580689f9d8] __is_cp_guaranteed at ffffffffa2b7ea98
#7 [ffffb6580689f9f0] f2fs_submit_page_write at ffffffffa2b81a69
#8 [ffffb6580689fa30] f2fs_do_write_meta_page at ffffffffa2b99777
#9 [ffffb6580689fae0] __f2fs_write_meta_page at ffffffffa2b75f1a
#10 [ffffb6580689fb18] f2fs_sync_meta_pages at ffffffffa2b77466
#11 [ffffb6580689fc98] do_checkpoint at ffffffffa2b78e46
#12 [ffffb6580689fd88] f2fs_write_checkpoint at ffffffffa2b79c29
#13 [ffffb6580689fdd0] f2fs_sync_fs at ffffffffa2b69d95
#14 [ffffb6580689fe20] sync_filesystem at ffffffffa2ad2574
#15 [ffffb6580689fe30] generic_shutdown_super at ffffffffa2a9b582
#16 [ffffb6580689fe48] kill_block_super at ffffffffa2a9b6d1
#17 [ffffb6580689fe60] kill_f2fs_super at ffffffffa2b6abe1
#18 [ffffb6580689fea0] deactivate_locked_super at ffffffffa2a9afb6
#19 [ffffb6580689feb8] cleanup_mnt at ffffffffa2abcad4
#20 [ffffb6580689fee0] task_work_run at ffffffffa28bca28
#21 [ffffb6580689ff00] exit_to_usermode_loop at ffffffffa28050b7
#22 [ffffb6580689ff38] do_syscall_64 at ffffffffa280560e
#23 [ffffb6580689ff50] entry_SYSCALL_64_after_hwframe at ffffffffa320008c
This occurred when umount f2fs if enable F2FS_FS_COMPRESSION
with F2FS_IO_TRACE. Fixes it by adding IS_IO_TRACED_PAGE to check
validity of pid for page_private.
Signed-off-by: Yu Changchun <yuchangchun1@huawei.com>
Reviewed-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
|
|
In the f2fs_unlink we do not add trace end for some
error paths, just add.
Signed-off-by: Lihong Kou <koulihong@huawei.com>
Reviewed-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
|
|
In f2fs_write_raw_pages(), we need to check page dirty status before
writeback, because there could be a racer (e.g. reclaimer) helps
writebacking the dirty page.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
|
|
The parameter compr is unused in the f2fs_cluster_blocks function
so we no longer need to pass it as a parameter.
Signed-off-by: Wang Xiaojun <wangxiaojun11@huawei.com>
Reviewed-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
|
|
to show f2fs_fiemap()'s result as below:
f2fs_fiemap: dev = (251,0), ino = 7, lblock:0, pblock:1625292800, len:2097152, flags:0, ret:0
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
|
|
to show f2fs_bmap()'s result as below:
f2fs_bmap: dev = (251,0), ino = 7, lblock:0, pblock:396800
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
|
|
If compression is disable, we should return zero rather than -EOPNOTSUPP
to indicate f2fs_bmap() is not supported.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
|
|
In current version, @state will only be FREE_NID. This parameter
has no real effect so remove it to keep clean.
Signed-off-by: Liu Song <liu.song11@zte.com.cn>
Reviewed-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
|
|
stakable/stackable
Signed-off-by: Liu Song <fishland@aliyun.com>
Reviewed-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
|
|
Filesystem including f2fs should support stable page for special
device like software raid, however there is one missing path that
page could be updated while it is writeback state as below, fix
this.
- gc_node_segment
- f2fs_move_node_page
- __write_node_page
- set_page_writeback
- do_read_inode
- f2fs_init_extent_tree
- __f2fs_init_extent_tree
i_ext->len = 0;
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
|
|
- Add to account and show per-log dirty_seg, full_seg and valid_blocks
in debugfs.
- reformat printed info.
TYPE segno secno zoneno dirty_seg full_seg valid_blk
- COLD data: 1523 1523 1523 1 0 399
- WARM data: 769 769 769 20 255 133098
- HOT data: 767 767 767 9 0 167
- Dir dnode: 22 22 22 3 0 70
- File dnode: 722 722 722 14 10 6505
- Indir nodes: 2 2 2 1 0 3
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
|
|
When f2fs_ioc_gc_range performs multiple segments gc ops, the return
value of f2fs_ioc_gc_range is determined by the last segment gc ops.
If its ops failed, the f2fs_ioc_gc_range will be considered to be failed
despite some of previous segments gc ops succeeded. Therefore, so we
fix: Redefine the return value of getting victim ops and add exception
handle for f2fs_gc. In particular, 1).if target has no valid block, it
will go on. 2).if target sectoion has valid block(s), but it is current
section, we will reminder the caller.
Signed-off-by: Qilong Zhang <zhangqilong3@huawei.com>
Reviewed-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
|
|
Use validation of @fio to inidcate whether caller want to serialize IOs
in io.io_list or not, then @add_list will be redundant, remove it.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
|
|
- to avoid race between checkpoint and quota file writeback, it
just needs to hold read lock of node_write in writeback path.
- node_write lock has covered all LFS data write paths, it's not
necessary, we only need to hold node_write lock at write path of
quota file.
This refactors commit ca7f76e68074 ("f2fs: fix wrong discard space").
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
|
|
to avoid polluting global symbol namespace.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
|
|
The caller of cifs_posix_lock_set will do retry(like
fcntl_setlk64->do_lock_file_wait) if we will wait for any file_lock.
So the retry in cifs_poxis_lock_set seems duplicated, remove it to
make a cleanup.
Signed-off-by: yangerkun <yangerkun@huawei.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
Reviewed-by: NeilBrown <neilb@suse.de>
|
|
read permission, not just read attributes permission, is required
on the directory.
See MS-SMB2 (protocol specification) section 3.3.5.19.
Signed-off-by: Steve French <stfrench@microsoft.com>
CC: Stable <stable@vger.kernel.org> # v5.6+
Reviewed-by: Pavel Shilovsky <pshilov@microsoft.com>
|
|
So far, gfs2 has taken the inode glocks inside the ->readpage and
->readahead address space operations. Since commit d4388340ae0b ("fs:
convert mpage_readpages to mpage_readahead"), gfs2_readahead is passed
the pages to read ahead locked. With that, the current holder of the
inode glock may be trying to lock one of those pages while
gfs2_readahead is trying to take the inode glock, resulting in a
deadlock.
Fix that by moving the lock taking to the higher-level ->read_iter file
and ->fault vm operations. This also gets rid of an ugly lock inversion
workaround in gfs2_readpage.
The cache consistency model of filesystems like gfs2 is such that if
data is found in the page cache, the data is up to date and can be used
without taking any filesystem locks. If a page is not cached,
filesystem locks must be taken before populating the page cache.
To avoid taking the inode glock when the data is already cached,
gfs2_file_read_iter first tries to read the data with the IOCB_NOIO flag
set. If that fails, the inode glock is taken and the operation is
retried with the IOCB_NOIO flag cleared.
Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux
Pull btrfs fixes from David Sterba:
- regression fix of a leak in global block reserve accounting
- fix a (hard to hit) race of readahead vs releasepage that could lead
to crash
- convert all remaining uses of comment fall through annotations to the
pseudo keyword
- fix crash when mounting a fuzzed image with -o recovery
* tag 'for-5.8-rc4-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux:
btrfs: reset tree root pointer after error in init_tree_roots
btrfs: fix reclaim_size counter leak after stealing from global reserve
btrfs: fix fatal extent_buffer readahead vs releasepage race
btrfs: convert comments to fallthrough annotations
|
|
First of all don't spin in io_ring_ctx_wait_and_kill() on iopoll.
Requests won't complete faster because of that, but only lengthen
io_uring_release().
The same goes for offloaded cleanup in io_ring_exit_work() -- it
already has waiting loop, don't do blocking active spinning.
For that, pass min=0 into io_iopoll_[try_]reap_events(), so it won't
actively spin. Leave the function if io_do_iopoll() there can't
complete a request to sleep in io_ring_exit_work().
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
Nobody checks io_iopoll_check()'s output parameter @nr_events.
Remove the parameter and declare it further down the stack.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
io_iopoll_reap_events() doesn't care about returned valued of
io_iopoll_getevents() and does the same checks for list emptiness
and need_resched(). Just use io_do_iopoll().
io_sq_thread() doesn't check return value as well. It also passes min=0,
so there never be the second iteration inside io_poll_getevents().
Inline it there too.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
Make sure the rtbitmap is large enough to store the entire bitmap.
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Allison Collins <allison.henderson@oracle.com>
|
|
Ensure that the realtime bitmap file is backed entirely by written
extents. No holes, no unwritten blocks, etc.
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Allison Collins <allison.henderson@oracle.com>
|
|
This debug code is called on every xfs_iflush() call, which then
checks every inode in the buffer for non-zero unlinked list field.
Hence it checks every inode in the cluster buffer every time a
single inode on that cluster it flushed. This is resulting in:
- 38.91% 5.33% [kernel] [k] xfs_iflush
- 17.70% xfs_iflush
- 9.93% xfs_inobp_check
4.36% xfs_buf_offset
10% of the CPU time spent flushing inodes is repeatedly checking
unlinked fields in the buffer. We don't need to do this.
The other place we call xfs_inobp_check() is
xfs_iunlink_update_dinode(), and this is after we've done this
assert for the agino we are about to write into that inode:
ASSERT(xfs_verify_agino_or_null(mp, agno, next_agino));
which means we've already checked that the agino we are about to
write is not 0 on debug kernels. The inode buffer verifiers do
everything else we need, so let's just remove this debug code.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
|
|
xfs_iflush_done() does 3 distinct operations to the inodes attached
to the buffer. Separate these operations out into functions so that
it is easier to modify these operations independently in future.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
|
|
Now that we have all the dirty inodes attached to the cluster
buffer, we don't actually have to do radix tree lookups to find
them. Sure, the radix tree is efficient, but walking a linked list
of just the dirty inodes attached to the buffer is much better.
We are also no longer dependent on having a locked inode passed into
the function to determine where to start the lookup. This means we
can drop it from the function call and treat all inodes the same.
We also make xfs_iflush_cluster skip inodes marked with
XFS_IRECLAIM. This we avoid races with inodes that reclaim is
actively referencing or are being re-initialised by inode lookup. If
they are actually dirty, they'll get written by a future cluster
flush....
We also add a shutdown check after obtaining the flush lock so that
we catch inodes that are dirty in memory and may have inconsistent
state due to the shutdown in progress. We abort these inodes
directly and so they remove themselves directly from the buffer list
and the AIL rather than having to wait for the buffer to be failed
and callbacks run to be processed correctly.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
|
|
with xfs_iflush() gone, we can rename xfs_iflush_int() back to
xfs_iflush(). Also move it up above xfs_iflush_cluster() so we don't
need the forward definition any more.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
|
|
Now we have a cached buffer on inode log items, we don't need
to do buffer lookups when flushing inodes anymore - all we need
to do is lock the buffer and we are ready to go.
This largely gets rid of the need for xfs_iflush(), which is
essentially just a mechanism to look up the buffer and flush the
inode to it. Instead, we can just call xfs_iflush_cluster() with a
few modifications to ensure it also flushes the inode we already
hold locked.
This allows the AIL inode item pushing to be almost entirely
non-blocking in XFS - we won't block unless memory allocation
for the cluster inode lookup blocks or the block device queues are
full.
Writeback during inode reclaim becomes a little more complex because
we now have to lock the buffer ourselves, but otherwise this change
is largely a functional no-op that removes a whole lot of code.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
|
|
Rather than attach inodes to the cluster buffer just when we are
doing IO, attach the inodes to the cluster buffer when they are
dirtied. The means the buffer always carries a list of dirty inodes
that reference it, and we can use that list to make more fundamental
changes to inode writeback that aren't otherwise possible.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
|
|
Once we have inodes pinning the cluster buffer and attached whenever
they are dirty, we no longer have a guarantee that the items are
flush locked when we lock the cluster buffer. Hence we cannot just
walk the buffer log item list and modify the attached inodes.
If the inode is not flush locked, we have to ILOCK it first and then
flush lock it to do all the prerequisite checks needed to avoid
races with other code. This is already handled by
xfs_ifree_get_one_inode(), so rework the inode iteration loop and
function to update all inodes in cache whether they are attached to
the buffer or not.
Note: we also remove the copying of the log item lsn to the
ili_flush_lsn as xfs_iflush_done() now uses the XFS_ISTALE flag to
trigger aborts and so flush lsn matching is not needed in IO
completion for processing freed inodes.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
|
|
Inode reclaim is quite different now to the way described in various
comments, so update all the comments explaining what it does and how
it works.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
|
|
Clean up xfs_reclaim_inodes() callers. Most callers want blocking
behaviour, so just make the existing SYNC_WAIT behaviour the
default.
For the xfs_reclaim_worker(), just call xfs_reclaim_inodes_ag()
directly because we just want optimistic clean inode reclaim to be
done in the background.
For xfs_quiesce_attr() we can just remove the inode reclaim calls as
they are a historic relic that was required to flush dirty inodes
that contained unlogged changes. We now log all changes to the
inodes, so the sync AIL push from xfs_log_quiesce() called by
xfs_quiesce_attr() will do all the required inode writeback for
freeze.
Seeing as we now want to loop until all reclaimable inodes have been
reclaimed, make xfs_reclaim_inodes() loop on the XFS_ICI_RECLAIM_TAG
tag rather than having xfs_reclaim_inodes_ag() tell it that inodes
were skipped. This is much more reliable and will always loop until
all reclaimable inodes are reclaimed.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
|
|
All background reclaim is SYNC_TRYLOCK already, and even blocking
reclaim (SYNC_WAIT) can use trylock mechanisms as
xfs_reclaim_inodes_ag() will keep cycling until there are no more
reclaimable inodes. Hence we can kill SYNC_TRYLOCK from inode
reclaim and make everything unconditionally non-blocking.
We remove all the optimistic "avoid blocking on locks" checks done
in xfs_reclaim_inode_grab() as nothing blocks on locks anymore.
Further, checking XFS_IFLOCK optimistically can result in detecting
inodes in the process of being cleaned (i.e. between being removed
from the AIL and having the flush lock dropped), so for
xfs_reclaim_inodes() to reliably reclaim all inodes we need to drop
these checks anyway.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
|
|
When we attempt to reclaim an inode, the first thing we do is take
the inode lock. This is blocking right now, so if the inode being
accessed by something else (e.g. being flushed to the cluster
buffer) we will block here.
Change this to a trylock so that we do not block inode reclaim
unnecessarily here.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
|
|
Inode reclaim will still throttle direct reclaim on the per-ag
reclaim locks. This is no longer necessary as reclaim can run
non-blocking now. Hence we can remove these locks so that we don't
arbitrarily block reclaimers just because there are more direct
reclaimers than there are AGs.
This can result in multiple reclaimers working on the same range of
an AG, but this doesn't cause any apparent issues. Optimising the
spread of concurrent reclaimers for best efficiency can be done in a
future patchset.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
|
|
We no longer need to issue IO from shrinker based inode reclaim to
prevent spurious OOM killer invocation. This leaves only the global
filesystem management operations such as unmount needing to
writeback dirty inodes and reclaim them.
Instead of using the reclaim pass to write dirty inodes before
reclaiming them, use the AIL to push all the dirty inodes before we
try to reclaim them. This allows us to remove all the conditional
SYNC_WAIT locking and the writeback code from xfs_reclaim_inode()
and greatly simplify the checks we need to do to reclaim an inode.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
|
|
Now that dirty inode writeback doesn't cause read-modify-write
cycles on the inode cluster buffer under memory pressure, the need
to throttle memory reclaim to the rate at which we can clean dirty
inodes goes away. That is due to the fact that we no longer thrash
inode cluster buffers under memory pressure to clean dirty inodes.
This means inode writeback no longer stalls on memory allocation
or read IO, and hence can be done asynchronously without generating
memory pressure. As a result, blocking inode writeback in reclaim is
no longer necessary to prevent reclaim priority windup as cleaning
dirty inodes is no longer dependent on having memory reserves
available for the filesystem to make progress reclaiming inodes.
Hence we can convert inode reclaim to be non-blocking for shrinker
callouts, both for direct reclaim and kswapd.
On a vanilla kernel, running a 16-way fsmark create workload on a
4 node/16p/16GB RAM machine, I can reliably pin 14.75GB of RAM via
userspace mlock(). The OOM killer gets invoked at 15GB of
pinned RAM.
Without the inode cluster pinning, this non-blocking reclaim patch
triggers premature OOM killer invocation with the same memory
pinning, sometimes with as much as 45% of RAM being free. It's
trivially easy to trigger the OOM killer when reclaim does not
block.
With pinning inode clusters in RAM and then adding this patch, I can
reliably pin 14.5GB of RAM and still have the fsmark workload run to
completion. The OOM killer gets invoked 14.75GB of pinned RAM, which
is only a small amount of memory less than the vanilla kernel. It is
much more reliable than just with async reclaim alone.
simoops shows that allocation stalls go away when async reclaim is
used. Vanilla kernel:
Run time: 1924 seconds
Read latency (p50: 3,305,472) (p95: 3,723,264) (p99: 4,001,792)
Write latency (p50: 184,064) (p95: 553,984) (p99: 807,936)
Allocation latency (p50: 2,641,920) (p95: 3,911,680) (p99: 4,464,640)
work rate = 13.45/sec (avg 13.44/sec) (p50: 13.46) (p95: 13.58) (p99: 13.70)
alloc stall rate = 3.80/sec (avg: 2.59) (p50: 2.54) (p95: 2.96) (p99: 3.02)
With inode cluster pinning and async reclaim:
Run time: 1924 seconds
Read latency (p50: 3,305,472) (p95: 3,715,072) (p99: 3,977,216)
Write latency (p50: 187,648) (p95: 553,984) (p99: 789,504)
Allocation latency (p50: 2,748,416) (p95: 3,919,872) (p99: 4,448,256)
work rate = 13.28/sec (avg 13.32/sec) (p50: 13.26) (p95: 13.34) (p99: 13.34)
alloc stall rate = 0.02/sec (avg: 0.02) (p50: 0.01) (p95: 0.03) (p99: 0.03)
Latencies don't really change much, nor does the work rate. However,
allocation almost never stalls with these changes, whilst the
vanilla kernel is sometimes reporting 20 stalls/s over a 60s sample
period. This difference is due to inode reclaim being largely
non-blocking now.
IOWs, once we have pinned inode cluster buffers, we can make inode
reclaim non-blocking without a major risk of premature and/or
spurious OOM killer invocation, and without any changes to memory
reclaim infrastructure.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
|
|
When we dirty an inode, we are going to have to write it disk at
some point in the near future. This requires the inode cluster
backing buffer to be present in memory. Unfortunately, under severe
memory pressure we can reclaim the inode backing buffer while the
inode is dirty in memory, resulting in stalling the AIL pushing
because it has to do a read-modify-write cycle on the cluster
buffer.
When we have no memory available, the read of the cluster buffer
blocks the AIL pushing process, and this causes all sorts of issues
for memory reclaim as it requires inode writeback to make forwards
progress. Allocating a cluster buffer causes more memory pressure,
and results in more cluster buffers to be reclaimed, resulting in
more RMW cycles to be done in the AIL context and everything then
backs up on AIL progress. Only the synchronous inode cluster
writeback in the the inode reclaim code provides some level of
forwards progress guarantees that prevent OOM-killer rampages in
this situation.
Fix this by pinning the inode backing buffer to the inode log item
when the inode is first dirtied (i.e. in xfs_trans_log_inode()).
This may mean the first modification of an inode that has been held
in cache for a long time may block on a cluster buffer read, but
we can do that in transaction context and block safely until the
buffer has been allocated and read.
Once we have the cluster buffer, the inode log item takes a
reference to it, pinning it in memory, and attaches it to the log
item for future reference. This means we can always grab the cluster
buffer from the inode log item when we need it.
When the inode is finally cleaned and removed from the AIL, we can
drop the reference the inode log item holds on the cluster buffer.
Once all inodes on the cluster buffer are clean, the cluster buffer
will be unpinned and it will be available for memory reclaim to
reclaim again.
This avoids the issues with needing to do RMW cycles in the AIL
pushing context, and hence allows complete non-blocking inode
flushing to be performed by the AIL pushing context.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
|
|
xfs_ail_delete_one() is called directly from dquot and inode IO
completion, as well as from the generic xfs_trans_ail_delete()
function. Inodes are about to have their own failure handling, and
dquots will in future, too. Pull the clearing of the LI_FAILED flag
up into the callers so we can customise the code appropriately.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
|
