We go through various hoops to disallow signals for the IO threads, but
there's really no reason why we cannot just allow them. The IO threads
never return to userspace like a normal thread, and hence don't go through
normal signal processing. Instead, just check for a pending signal as part
of the work loop, and call get_signal() to handle it for us if anything
is pending.
With that, we can support receiving signals, including special ones like
SIGSTOP.
Acked-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
There were two problems (one of which could cause data corruption)
that were noticed with duplicate extents (ie reflink)
when debugging why various xfstests were being incorrectly skipped
(e.g. generic/138, generic/140, generic/142). First, we were not
updating the file size locally in the cache when extending a
file due to reflink (it would refresh after actimeo expires)
but xfstest was checking the size immediately which was still
0 so caused the test to be skipped. Second, we were setting
the target file size (which could shrink the file) in all cases
to the end of the reflinked range rather than only setting the
target file size when reflink would extend the file.
CC: <stable@vger.kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
Make SMB2 not print out an error when an oplock break is received for an
unknown handle, similar to SMB1. The debug message which is printed for
these unknown handles may also be misleading, so fix that too.
The SMB2 lease break path is not affected by this patch.
Without this, a program which writes to a file from one thread, and
opens, reads, and writes the same file from another thread triggers the
below errors several times a minute when run against a Samba server
configured with "smb2 leases = no".
CIFS: VFS: \\192.168.0.1 No task to wake, unknown frame received! NumMids 2
00000000: 424d53fe 00000040 00000000 00000012 .SMB@...........
00000010: 00000001 00000000 ffffffff ffffffff ................
00000020: 00000000 00000000 00000000 00000000 ................
00000030: 00000000 00000000 00000000 00000000 ................
Signed-off-by: Vincent Whitchurch <vincent.whitchurch@axis.com>
Reviewed-by: Tom Talpey <tom@talpey.com>
Reviewed-by: Paulo Alcantara (SUSE) <pc@cjr.nz>
Signed-off-by: Steve French <stfrench@microsoft.com>
RHBZ: 1933527
Under SMB1 + POSIX, if an inode is reused on a server after we have read and
cached a part of a file, when we then open the new file with the
re-cycled inode there is a chance that we may serve the old data out of cache
to the application.
This only happens for SMB1 (deprecated) and when posix are used.
The simplest solution to avoid this race is to force a revalidate
on smb1-posix open.
Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com>
Reviewed-by: Paulo Alcantara (SUSE) <pc@cjr.nz>
Signed-off-by: Steve French <stfrench@microsoft.com>
My recent fixes to cifsacl to maintain inherited ACEs had
regressed modefromsid when an older ACL already exists.
Found testing xfstest 495 with modefromsid mount option
Fixes: f506550889 ("cifs: Retain old ACEs when converting between mode bits and ACL")
Signed-off-by: Shyam Prasad N <sprasad@microsoft.com>
Reviewed-by: Paulo Alcantara (SUSE) <pc@cjr.nz>
Signed-off-by: Steve French <stfrench@microsoft.com>
Print the path name of the swapfile that failed to active to ease
debugging the problem and to avoid a scare if xfstests hits these
cases. Also reword one warning a bit, as the error is not about
a file being on multiple devices, but one that has at least an
extent outside the main device known to the VFS and swap code.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
In order to avoid race with f2fs_do_replace_block().
Fixes: f5a53edcf0 ("f2fs: support aligned pinned file")
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
If the alloc_type of the original curseg is LFS,
when we change_curseg and then do recover curseg,
the alloc_type becomes SSR.
Signed-off-by: Wang Xiaojun <wangxiaojun11@huawei.com>
Reviewed-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
With the default DPOLICY_BG discard thread is ioaware, which prevents
the discard thread from issuing the discard commands. On low RAM setups,
it is observed that these discard commands in the cache are consuming
high memory. This patch aims to relax the memory pressure on the system
due to f2fs pending discard cmds by changing the policy to DPOLICY_FORCE
based on the nm_i->ram_thresh configured.
Signed-off-by: Sahitya Tummala <stummala@codeaurora.org>
Reviewed-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
In CP disabling mode, there are two issues when using LFS or SSR | AT_SSR
mode to select victim:
1. LFS is set to find source section during GC, the victim should have
no checkpointed data, since after GC, section could not be set free for
reuse.
Previously, we only check valid chpt blocks in current segment rather
than section, fix it.
2. SSR | AT_SSR are set to find target segment for writes which can be
fully filled by checkpointed and newly written blocks, we should never
select such segment, otherwise it can cause panic or data corruption
during allocation, potential case is described as below:
a) target segment has 'n' (n < 512) ckpt valid blocks
b) GC migrates 'n' valid blocks to other segment (segment is still
in dirty list)
c) GC migrates '512 - n' blocks to target segment (segment has 'n'
cp_vblocks and '512 - n' vblocks)
d) If GC selects target segment via {AT,}SSR allocator, however there
is no free space in targe segment.
Fixes: 4354994f09 ("f2fs: checkpoint disabling")
Fixes: 093749e296 ("f2fs: support age threshold based garbage collection")
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
For AES256 encryption (GCM and CCM), we need to adjust the size of a few
fields to 32 bytes instead of 16 to accommodate the larger keys.
Also, the L value supplied to the key generator needs to be changed from
to 256 when these algorithms are used.
Keeping the ioctl struct for dumping keys of the same size for now.
Will send out a different patch for that one.
Signed-off-by: Shyam Prasad N <sprasad@microsoft.com>
Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com>
CC: <stable@vger.kernel.org> # v5.10+
Signed-off-by: Steve French <stfrench@microsoft.com>
In the case of expanding pinned file, map.m_lblk and map.m_len
will update in each round of section allocation, so in error
path, last i_size will be calculated with wrong m_lblk and m_len,
fix it.
Fixes: f5a53edcf0 ("f2fs: support aligned pinned file")
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
In error path of f2fs_remount(), it missed to restart/stop kernel thread
or enable/disable checkpoint, then mount option status may not be
consistent with real condition of filesystem, so let's reorder remount
flow a bit as below and do recovery correctly in error path:
1) handle gc thread
2) handle ckpt thread
3) handle flush thread
4) handle checkpoint disabling
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
butt3rflyh4ck <butterflyhuangxx@gmail.com> reported a bug found by
syzkaller fuzzer with custom modifications in 5.12.0-rc3+ [1]:
dump_stack+0xfa/0x151 lib/dump_stack.c:120
print_address_description.constprop.0.cold+0x82/0x32c mm/kasan/report.c:232
__kasan_report mm/kasan/report.c:399 [inline]
kasan_report.cold+0x7c/0xd8 mm/kasan/report.c:416
f2fs_test_bit fs/f2fs/f2fs.h:2572 [inline]
current_nat_addr fs/f2fs/node.h:213 [inline]
get_next_nat_page fs/f2fs/node.c:123 [inline]
__flush_nat_entry_set fs/f2fs/node.c:2888 [inline]
f2fs_flush_nat_entries+0x258e/0x2960 fs/f2fs/node.c:2991
f2fs_write_checkpoint+0x1372/0x6a70 fs/f2fs/checkpoint.c:1640
f2fs_issue_checkpoint+0x149/0x410 fs/f2fs/checkpoint.c:1807
f2fs_sync_fs+0x20f/0x420 fs/f2fs/super.c:1454
__sync_filesystem fs/sync.c:39 [inline]
sync_filesystem fs/sync.c:67 [inline]
sync_filesystem+0x1b5/0x260 fs/sync.c:48
generic_shutdown_super+0x70/0x370 fs/super.c:448
kill_block_super+0x97/0xf0 fs/super.c:1394
The root cause is, if nat entry in checkpoint journal area is corrupted,
e.g. nid of journalled nat entry exceeds max nid value, during checkpoint,
once it tries to flush nat journal to NAT area, get_next_nat_page() may
access out-of-bounds memory on nat_bitmap due to it uses wrong nid value
as bitmap offset.
[1] https://lore.kernel.org/lkml/CAFcO6XOMWdr8pObek6eN6-fs58KG9doRFadgJj-FnF-1x43s2g@mail.gmail.com/T/#u
Reported-and-tested-by: butt3rflyh4ck <butterflyhuangxx@gmail.com>
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
In readonly mountpoint, there should be no write IOs include checkpoint
IO, so that it's not needed to create kernel checkpoint thread.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
AT_SSR mode is introduced by age threshold based GC for better
hot/cold data seperation and avoiding free segment cost. However,
LFS write mode is preferred in the scenario of foreground or high
urgent GC, which should be finished ASAP. Let's only use AT_SSR
in background GC and not high urgent GC modes.
Signed-off-by: Weichao Guo <guoweichao@oppo.com>
Signed-off-by: Huang Jianan <huangjianan@oppo.com>
Reviewed-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
I've added new sysfs nodes to show runtime compression stat since mount.
compr_written_block - show the block count written after compression
compr_saved_block - show the saved block count with compression
compr_new_inode - show the count of inode newly enabled for compression
Signed-off-by: Daeho Jeong <daehojeong@google.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Networking has many sysctls that could fit in one u8.
This patch adds proc_dou8vec_minmax() for this purpose.
Note that the .extra1 and .extra2 fields are pointing
to integers, because it makes conversions easier.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Removal of kmem_zone_init wrappers accidentally changed a slab cache
name from "xfs_trans" to "xf_trans". Fix this so that userspace
consumers of /proc/slabinfo and /sys/kernel/slab can find it again.
Fixes: b1231760e4 ("xfs: Remove slab init wrappers")
Signed-off-by: Anthony Iliopoulos <ailiop@suse.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
per-AG resv failure after fixing up freespace is hard to test in an
effective way, so directly add an error injection path to observe
such error handling path works as expected.
Signed-off-by: Gao Xiang <hsiangkao@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
As the first step of shrinking, this attempts to enable shrinking
unused space in the last allocation group by fixing up freespace
btree, agi, agf and adjusting super block and use a helper
xfs_ag_shrink_space() to fixup the last AG.
This can be all done in one transaction for now, so I think no
additional protection is needed.
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Gao Xiang <hsiangkao@redhat.com>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
This patch introduces a helper to shrink unused space in the last AG
by fixing up the freespace btree.
Also make sure that the per-AG reservation works under the new AG
size. If such per-AG reservation or extent allocation fails, roll
the transaction so the new transaction could cancel without any side
effects.
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Gao Xiang <hsiangkao@redhat.com>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Move out related logic for initializing new added AGs to a new helper
in preparation for shrinking. No logic changes.
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Gao Xiang <hsiangkao@redhat.com>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
sb_fdblocks will be updated lazily if lazysbcount is enabled,
therefore when shrinking the filesystem sb_fdblocks could be
larger than sb_dblocks and xfs_validate_sb_write() would fail.
Even for growfs case, it'd be better to update lazy sb counters
immediately to reflect the real sb counters.
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Signed-off-by: Gao Xiang <hsiangkao@redhat.com>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
- 21.92% __xfs_trans_commit
- 21.62% xfs_log_commit_cil
- 11.69% xfs_trans_unreserve_and_mod_sb
- 11.58% __percpu_counter_compare
- 11.45% __percpu_counter_sum
- 10.29% _raw_spin_lock_irqsave
- 10.28% do_raw_spin_lock
__pv_queued_spin_lock_slowpath
We debated just getting rid of it last time this came up and
there was no real objection to removing it. Now it's the biggest
scalability limitation for debug kernels even on smallish machines,
so let's just get rid of it.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
On debug kernels, we call xfs_dir3_leaf_check_int() multiple times
on every directory modification. The robust hash ordering checks it
does on every entry in the leaf on every call results in a massive
CPU overhead which slows down debug kernels by a large amount.
We use xfs_dir3_leaf_check_int() for the verifiers as well, so we
can't just gut the function to reduce overhead. What we can do,
however, is reduce the work it does when it is called from the
debug interfaces, just leaving the high level checks in place and
leaving the robust validation to the verifiers. This means the debug
checks will catch gross errors, but subtle bugs might not be caught
until a verifier is run.
It is easy enough to restore the existing debug behaviour if the
developer needs it (just change a call parameter in the debug code),
but overwise the overhead makes testing large directory block sizes
on debug kernels very slow.
Profile at an unlink rate of ~80k file/s on a 64k block size
filesystem before the patch:
40.30% [kernel] [k] xfs_dir3_leaf_check_int
10.98% [kernel] [k] __xfs_dir3_data_check
8.10% [kernel] [k] xfs_verify_dir_ino
4.42% [kernel] [k] memcpy
2.22% [kernel] [k] xfs_dir2_data_get_ftype
1.52% [kernel] [k] do_raw_spin_lock
Profile after, at an unlink rate of ~125k files/s (+50% improvement)
has largely dropped the leaf verification debug overhead out of the
profile.
16.53% [kernel] [k] __xfs_dir3_data_check
12.53% [kernel] [k] xfs_verify_dir_ino
7.97% [kernel] [k] memcpy
3.36% [kernel] [k] xfs_dir2_data_get_ftype
2.86% [kernel] [k] __pv_queued_spin_lock_slowpath
Create shows a similar change in profile and a +25% improvement in
performance.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
We call xfs_dir_ino_validate() for every dir entry in a directory
when doing validity checking of the directory. It calls
xfs_verify_dir_ino() then emits a corruption report if bad or does
error injection if good. It is extremely costly:
43.27% [kernel] [k] xfs_dir3_leaf_check_int
10.28% [kernel] [k] __xfs_dir3_data_check
6.61% [kernel] [k] xfs_verify_dir_ino
4.16% [kernel] [k] xfs_errortag_test
4.00% [kernel] [k] memcpy
3.48% [kernel] [k] xfs_dir_ino_validate
7% of the cpu usage in this directory traversal workload is
xfs_dir_ino_validate() doing absolutely nothing.
We don't need error injection to simulate a bad inode numbers in the
directory structure because we can do that by fuzzing the structure
on disk.
And we don't need a corruption report, because the
__xfs_dir3_data_check() will emit one if the inode number is bad.
So just call xfs_verify_dir_ino() directly here, and get rid of all
this unnecessary overhead:
40.30% [kernel] [k] xfs_dir3_leaf_check_int
10.98% [kernel] [k] __xfs_dir3_data_check
8.10% [kernel] [k] xfs_verify_dir_ino
4.42% [kernel] [k] memcpy
2.22% [kernel] [k] xfs_dir2_data_get_ftype
1.52% [kernel] [k] do_raw_spin_lock
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
From a concurrent rm -rf workload:
41.04% [kernel] [k] xfs_dir3_leaf_check_int
9.85% [kernel] [k] __xfs_dir3_data_check
5.60% [kernel] [k] xfs_verify_ino
5.32% [kernel] [k] xfs_agino_range
4.21% [kernel] [k] memcpy
3.06% [kernel] [k] xfs_errortag_test
2.57% [kernel] [k] xfs_dir_ino_validate
1.66% [kernel] [k] xfs_dir2_data_get_ftype
1.17% [kernel] [k] do_raw_spin_lock
1.11% [kernel] [k] xfs_verify_dir_ino
0.84% [kernel] [k] __raw_callee_save___pv_queued_spin_unlock
0.83% [kernel] [k] xfs_buf_find
0.64% [kernel] [k] xfs_log_commit_cil
THere's an awful lot of overhead in just range checking inode
numbers in that, but each inode number check is not a lot of code.
The total is a bit over 14.5% of the CPU time is spent validating
inode numbers.
The problem is that they deeply nested global scope functions so the
overhead here is all in function call marshalling.
text data bss dec hex filename
2077 0 0 2077 81d fs/xfs/libxfs/xfs_types.o.orig
2197 0 0 2197 895 fs/xfs/libxfs/xfs_types.o
There's a small increase in binary size by inlining all the local
nested calls in the verifier functions, but the same workload now
profiles as:
40.69% [kernel] [k] xfs_dir3_leaf_check_int
10.52% [kernel] [k] __xfs_dir3_data_check
6.68% [kernel] [k] xfs_verify_dir_ino
4.22% [kernel] [k] xfs_errortag_test
4.15% [kernel] [k] memcpy
3.53% [kernel] [k] xfs_dir_ino_validate
1.87% [kernel] [k] xfs_dir2_data_get_ftype
1.37% [kernel] [k] do_raw_spin_lock
0.98% [kernel] [k] xfs_buf_find
0.94% [kernel] [k] __raw_callee_save___pv_queued_spin_unlock
0.73% [kernel] [k] xfs_log_commit_cil
Now we only spend just over 10% of the time validing inode numbers
for the same workload. Hence a few "inline" keyworks is good enough
to reduce the validation overhead by 30%...
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
We process the buf_log_item bitmap one set bit at a time with
xfs_next_bit() so we can detect if a region crosses a memcpy
discontinuity in the buffer data address. This has massive overhead
on large buffers (e.g. 64k directory blocks) because we do a lot of
unnecessary checks and xfs_buf_offset() calls.
For example, 16-way concurrent create workload on debug kernel
running CPU bound has this at the top of the profile at ~120k
create/s on 64kb directory block size:
20.66% [kernel] [k] xfs_dir3_leaf_check_int
7.10% [kernel] [k] memcpy
6.22% [kernel] [k] xfs_next_bit
3.55% [kernel] [k] xfs_buf_offset
3.53% [kernel] [k] xfs_buf_item_format
3.34% [kernel] [k] __pv_queued_spin_lock_slowpath
3.04% [kernel] [k] do_raw_spin_lock
2.84% [kernel] [k] xfs_buf_item_size_segment.isra.0
2.31% [kernel] [k] __raw_callee_save___pv_queued_spin_unlock
1.36% [kernel] [k] xfs_log_commit_cil
(debug checks hurt large blocks)
The only buffers with discontinuities in the data address are
unmapped buffers, and they are only used for inode cluster buffers
and only for logging unlinked pointers. IOWs, it is -rare- that we
even need to detect a discontinuity in the buffer item formatting
code.
Optimise all this by using xfs_contig_bits() to find the size of
the contiguous regions, then test for a discontiunity inside it. If
we find one, do the slow "bit at a time" method we do now. If we
don't, then just copy the entire contiguous range in one go.
Profile now looks like:
25.26% [kernel] [k] xfs_dir3_leaf_check_int
9.25% [kernel] [k] memcpy
5.01% [kernel] [k] __pv_queued_spin_lock_slowpath
2.84% [kernel] [k] do_raw_spin_lock
2.22% [kernel] [k] __raw_callee_save___pv_queued_spin_unlock
1.88% [kernel] [k] xfs_buf_find
1.53% [kernel] [k] memmove
1.47% [kernel] [k] xfs_log_commit_cil
....
0.34% [kernel] [k] xfs_buf_item_format
....
0.21% [kernel] [k] xfs_buf_offset
....
0.16% [kernel] [k] xfs_contig_bits
....
0.13% [kernel] [k] xfs_buf_item_size_segment.isra.0
So the bit scanning over for the dirty region tracking for the
buffer log items is basically gone. Debug overhead hurts even more
now...
Perf comparison
dir block creates unlink
size (kb) time rate time
Original 4 4m08s 220k 5m13s
Original 64 7m21s 115k 13m25s
Patched 4 3m59s 230k 5m03s
Patched 64 6m23s 143k 12m33s
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Otherwise it doesn't correctly calculate the number of vectors
in a logged buffer that has a contiguous map that gets split into
multiple regions because the range spans discontigous memory.
Probably never been hit in practice - we don't log contiguous ranges
on unmapped buffers (inode clusters).
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
When we modify btrees repeatedly, we regularly increase the size of
the logged region by a single chunk at a time (per transaction
commit). This results in the CIL formatting code having to
reallocate the log vector buffer every time the buffer dirty region
grows. Hence over a typical 4kB btree buffer, we might grow the log
vector 4096/128 = 32x over a short period where we repeatedly add
or remove records to/from the buffer over a series of running
transaction. This means we are doing 32 memory allocations and frees
over this time during a performance critical path in the journal.
The amount of space tracked in the CIL for the object is calculated
during the ->iop_format() call for the buffer log item, but the
buffer memory allocated for it is calculated by the ->iop_size()
call. The size callout determines the size of the buffer, the format
call determines the space used in the buffer.
Hence we can oversize the buffer space required in the size
calculation without impacting the amount of space used and accounted
to the CIL for the changes being logged. This allows us to reduce
the number of allocations by rounding up the buffer size to allow
for future growth. This can safe a substantial amount of CPU time in
this path:
- 46.52% 2.02% [kernel] [k] xfs_log_commit_cil
- 44.49% xfs_log_commit_cil
- 30.78% _raw_spin_lock
- 30.75% do_raw_spin_lock
30.27% __pv_queued_spin_lock_slowpath
(oh, ouch!)
....
- 1.05% kmem_alloc_large
- 1.02% kmem_alloc
0.94% __kmalloc
This overhead here us what this patch is aimed at. After:
- 0.76% kmem_alloc_large
- 0.75% kmem_alloc
0.70% __kmalloc
The size of 512 bytes is based on the bitmap chunk size being 128
bytes and that random directory entry updates almost never require
more than 3-4 128 byte regions to be logged in the directory block.
The other observation is for per-ag btrees. When we are inserting
into a new btree block, we'll pack it from the front. Hence the
first few records land in the first 128 bytes so we log only 128
bytes, the next 8-16 records land in the second region so now we log
256 bytes. And so on. If we are doing random updates, it will only
allocate every 4 random 128 byte regions that are dirtied instead of
every single one.
Any larger than 512 bytes and I noticed an increase in memory
footprint in my scalability workloads. Any less than this and I
didn't really see any significant benefit to CPU usage.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Chandan Babu R <chandanrlinux@gmail.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Gao Xiang <hsiangkao@redhat.com>
When we allocate a new inode, we often need to add an attribute to
the inode as part of the create. This can happen as a result of
needing to add default ACLs or security labels before the inode is
made visible to userspace.
This is highly inefficient right now. We do the create transaction
to allocate the inode, then we do an "add attr fork" transaction to
modify the just created empty inode to set the inode fork offset to
allow attributes to be stored, then we go and do the attribute
creation.
This means 3 transactions instead of 1 to allocate an inode, and
this greatly increases the load on the CIL commit code, resulting in
excessive contention on the CIL spin locks and performance
degradation:
18.99% [kernel] [k] __pv_queued_spin_lock_slowpath
3.57% [kernel] [k] do_raw_spin_lock
2.51% [kernel] [k] __raw_callee_save___pv_queued_spin_unlock
2.48% [kernel] [k] memcpy
2.34% [kernel] [k] xfs_log_commit_cil
The typical profile resulting from running fsmark on a selinux enabled
filesytem is adds this overhead to the create path:
- 15.30% xfs_init_security
- 15.23% security_inode_init_security
- 13.05% xfs_initxattrs
- 12.94% xfs_attr_set
- 6.75% xfs_bmap_add_attrfork
- 5.51% xfs_trans_commit
- 5.48% __xfs_trans_commit
- 5.35% xfs_log_commit_cil
- 3.86% _raw_spin_lock
- do_raw_spin_lock
__pv_queued_spin_lock_slowpath
- 0.70% xfs_trans_alloc
0.52% xfs_trans_reserve
- 5.41% xfs_attr_set_args
- 5.39% xfs_attr_set_shortform.constprop.0
- 4.46% xfs_trans_commit
- 4.46% __xfs_trans_commit
- 4.33% xfs_log_commit_cil
- 2.74% _raw_spin_lock
- do_raw_spin_lock
__pv_queued_spin_lock_slowpath
0.60% xfs_inode_item_format
0.90% xfs_attr_try_sf_addname
- 1.99% selinux_inode_init_security
- 1.02% security_sid_to_context_force
- 1.00% security_sid_to_context_core
- 0.92% sidtab_entry_to_string
- 0.90% sidtab_sid2str_get
0.59% sidtab_sid2str_put.part.0
- 0.82% selinux_determine_inode_label
- 0.77% security_transition_sid
0.70% security_compute_sid.part.0
And fsmark creation rate performance drops by ~25%. The key point to
note here is that half the additional overhead comes from adding the
attribute fork to the newly created inode. That's crazy, considering
we can do this same thing at inode create time with a couple of
lines of code and no extra overhead.
So, if we know we are going to add an attribute immediately after
creating the inode, let's just initialise the attribute fork inside
the create transaction and chop that whole chunk of code out of
the create fast path. This completely removes the performance
drop caused by enabling SELinux, and the profile looks like:
- 8.99% xfs_init_security
- 9.00% security_inode_init_security
- 6.43% xfs_initxattrs
- 6.37% xfs_attr_set
- 5.45% xfs_attr_set_args
- 5.42% xfs_attr_set_shortform.constprop.0
- 4.51% xfs_trans_commit
- 4.54% __xfs_trans_commit
- 4.59% xfs_log_commit_cil
- 2.67% _raw_spin_lock
- 3.28% do_raw_spin_lock
3.08% __pv_queued_spin_lock_slowpath
0.66% xfs_inode_item_format
- 0.90% xfs_attr_try_sf_addname
- 0.60% xfs_trans_alloc
- 2.35% selinux_inode_init_security
- 1.25% security_sid_to_context_force
- 1.21% security_sid_to_context_core
- 1.19% sidtab_entry_to_string
- 1.20% sidtab_sid2str_get
- 0.86% sidtab_sid2str_put.part.0
- 0.62% _raw_spin_lock_irqsave
- 0.77% do_raw_spin_lock
__pv_queued_spin_lock_slowpath
- 0.84% selinux_determine_inode_label
- 0.83% security_transition_sid
0.86% security_compute_sid.part.0
Which indicates the XFS overhead of creating the selinux xattr has
been halved. This doesn't fix the CIL lock contention problem, just
means it's not a limiting factor for this workload. Lock contention
in the security subsystems is going to be an issue soon, though...
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
[djwong: fix compilation error when CONFIG_SECURITY=n]
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Gao Xiang <hsiangkao@redhat.com>
Add the BUILD_BUG_ON to xfs_errortag_add() in order to make sure that
the length of xfs_errortag_random_default matches XFS_ERRTAG_MAX when
building.
Signed-off-by: Gao Xiang <hsiangkao@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Rename mp variable to parsisng_mp so it is easy to distinguish
between current mount point handle and handle for mount point
which mount options are being parsed.
Suggested-by: Eric Sandeen <sandeen@redhat.com>
Signed-off-by: Pavel Reichl <preichl@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Carlos Maiolino <cmaiolino@redhat.com>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Since we're about to start using the blockgc workqueue to dispose of
inactivated inodes, strip the "block" prefix from the name; now it's
merely the general garbage collection (gc) workqueue.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Files containing metadata (quota records, rt bitmap and summary info)
are fully managed by the filesystem, which means that all resource
cleanup must be explicit, not automatic. This means that they should
never be subjected automatic to post-eof truncation, nor should they be
freed automatically even if the link count drops to zero.
In other words, xfs_inactive() should leave these files alone. Add the
necessary predicate functions to make this happen. This adds a second
layer of prevention for the kinds of fs corruption that was fixed by
commit f4c32e87de. If we ever decide to support removing metadata
files, we should make all those metadata updates explicit.
Rearrange the order of #includes to fix compiler errors, since
xfs_mount.h is supposed to be included before xfs_inode.h
Followup-to: f4c32e87de ("xfs: fix realtime bitmap/summary file truncation when growing rt volume")
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Use the AG btree height limits that we precomputed into the xfs_mount to
validate the AG headers instead of using XFS_BTREE_MAXLEVELS.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Functions called by this function cannot fail, so get rid of the return
and error checking.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Since xchk_ag_read_headers initializes fields in struct xchk_ag, we
might as well set the AG number and save the callers the trouble.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
If scrub observes cross-referencing errors while scanning a data
structure, mark the data structure sick. There's /something/
inconsistent, even if we can't really tell what it is.
Fixes: 4860a05d24 ("xfs: scrub/repair should update filesystem metadata health")
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
If a scrubber cannot complete its check and signals an incomplete check,
we must bail out immediately without updating health status, trying a
repair, etc. because our scan is incomplete and we therefore do not know
much more.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
