The previous ceph-client merge resulted in ceph not even building,
because there was a merge conflict that wasn't visible as an actual data
conflict: commit 7221fe4c2e ("ceph: add acl for cephfs") added support
for POSIX ACL's into Ceph, but unluckily we also had the VFS tree change
a lot of the POSIX ACL helper functions to be much more helpful to
filesystems (see for example commits 2aeccbe957 "fs: add generic
xattr_acl handlers", 5bf3258fd2 "fs: make posix_acl_chmod more useful"
and 37bc15392a "fs: make posix_acl_create more useful")
The reason this conflict wasn't obvious was many-fold: because it was a
semantic conflict rather than a data conflict, it wasn't visible in the
git merge as a conflict. And because the VFS tree hadn't been in
linux-next, people hadn't become aware of it that way. And because I
was at jury duty this morning, I was using my laptop and as a result not
doing constant "allmodconfig" builds.
Anyway, this fixes the build and generally removes a fair chunk of the
Ceph POSIX ACL support code, since the improved helpers seem to match
really well for Ceph too. But I don't actually have any way to *test*
the end result, and I was really hoping for some ACK's for this. Oh,
well.
Not compiling certainly doesn't make things easier to test, so I'm
committing this without the acks after having waited for four hours...
Plus it's what I would have done for the merge had I noticed the
semantic conflict..
Reported-by: Dave Jones <davej@redhat.com>
Cc: Sage Weil <sage@inktank.com>
Cc: Guangliang Zhao <lucienchao@gmail.com>
Cc: Li Wang <li.wang@ubuntykylin.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The buffer size argument passed to snprintf must account for the
trailing null byte added by snprintf, and it returns a value >= then
sizeof(buffer) when the string can't fit in the buffer.
Since our buffer has a size of 64 characters, and the maximum orphan
name we can generate is 63 characters wide, we must pass 64 as the
buffer size to snprintf, and not 63.
Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
Reviewed-by: David Sterba <dsterba@suse.cz>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
When defragging a very large file, the cluster variable can wrap its 32-bit
signed int type and become negative, which eventually gets passed to
btrfs_force_ra() as a very large unsigned long value. On 32-bit platforms,
this eventually results in an Oops from the SLAB allocator.
Change the cluster and max_cluster signed int variables to unsigned long to
match the readahead functions. This also allows the min() comparison in
btrfs_defrag_file() to work as intended.
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
The COMPRESS_LZOv2 incompat featue is currently not implemented, the bit
is only reserved, no point to list it in sysfs.
Signed-off-by: David Sterba <dsterba@suse.cz>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
The owner and capability checks in IOC_SUBVOL_SETFLAGS and
SET_RECEIVED_SUBVOL should be called before any other checks are done.
Also unify the error code to EPERM.
Signed-off-by: David Sterba <dsterba@suse.cz>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
Currently, any user can snapshot any subvolume if the path is accessible and
thus indirectly create and keep files he does not own under his direcotries.
This is not possible with traditional directories.
In security context, a user can snapshot root filesystem and pin any
potentially buggy binaries, even if the updates are applied.
All the snapshots are visible to the administrator, so it's possible to
verify if there are suspicious snapshots.
Another more practical problem is that any user can pin the space used
by eg. root and cause ENOSPC.
Original report:
https://bugs.launchpad.net/ubuntu/+source/apparmor/+bug/484786
CC: stable@vger.kernel.org
Signed-off-by: David Sterba <dsterba@suse.cz>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
We allocate the free space from the former block group, not the current
one, so should use the former one to output the trace information.
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
used_block_group is just used for the space cluster which doesn't
belong to the current block group, the other place needn't use it.
Or the logic of code seems unclear.
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
It is better that the position of the lock is close to the data which is
protected by it, because they may be in the same cache line, we will load
less cache lines when we access them. So we rearrange the members' position
of btrfs_space_info structure to make the lock be closer to the its data.
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Reviewed-by: David Sterba <dsterba@suse.cz>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
To search tree root without transaction protection, we should neither search commit
root nor skip locking here, fix it.
Signed-off-by: Wang Shilong <wangsl.fnst@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
The performance of fsync dropped down suddenly sometimes, the main reason
of this problem was that we might only flush part dirty pages in a ordered
extent, then got that ordered extent, wait for the csum calcucation. But if
no task flushed the left part, we would wait until the flusher flushed them,
sometimes we need wait for several seconds, it made the performance drop
down suddenly. (On my box, it drop down from 56MB/s to 4-10MB/s)
This patch improves the above problem by flushing left dirty pages aggressively.
Test Environment:
CPU: 2CPU * 2Cores
Memory: 4GB
Partition: 20GB(HDD)
Test Command:
# sysbench --num-threads=8 --test=fileio --file-num=1 \
> --file-total-size=8G --file-block-size=32768 \
> --file-io-mode=sync --file-fsync-freq=100 \
> --file-fsync-end=no --max-requests=10000 \
> --file-test-mode=rndwr run
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
Steps to reproduce:
# mkfs.btrfs -f /dev/sda8
# mount /dev/sda8 /mnt -o flushoncommit
# dd if=/dev/zero of=/mnt/data bs=4k count=102400 &
# mount /dev/sda8 /mnt -o remount, ro
When remounting RW to RO, the logic is to firstly set flag
to RO and then commit transaction, however with option
flushoncommit enabled,we will do RO check within committing
transaction, so we get a transaction abortion here.
Actually,here check is wrong, we should check if FS_STATE_ERROR
is set, fix it.
Reported-by: Qu Wenruo <quwenruo@cn.fujitsu.com>
Suggested-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Wang Shilong <wangsl.fnst@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
When we are looking for file extent items that intersect the cloning
range, for each one that falls completely outside the range, don't
release the path and do another full tree search - just move on
to the next slot and copy the file extent item into our buffer only
if the item intersects the cloning range.
Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
When transaction is aborted, we fail to commit transaction, instead we do
cleanup work. After that when we umount btrfs, we get to free fs roots' log
trees respectively, but that happens after we unpin extents, so those extents
pinned by freeing log trees will remain in memory and lead to the leak.
Signed-off-by: Liu Bo <bo.li.liu@oracle.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
Since remount will pending the new mount options to the original mount
options, which will make btrfs_parse_options check the old options then
new options, causing some stupid output like "enabling XXX" following by
"disable XXX".
This patch will add extra check before every btrfs_info to skip the
output from old options checking.
Signed-off-by: Qu Wenruo <quwenruo@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
Add noinode_cache mount option for btrfs.
Since inode map cache involves all the btrfs_find_free_ino/return_ino
things and if just trigger the mount_opt,
an inode number get from inode map cache will not returned to inode map
cache.
To keep the find and return inode both in the same behavior,
a new bit in mount_opt, CHANGE_INODE_CACHE, is introduced for this idea.
CHANGE_INODE_CACHE is set/cleared in remounting, and the original
INODE_MAP_CACHE is set/cleared according to CHANGE_INODE_CACHE after a
success transaction.
Since find/return inode is all done between btrfs_start_transaction and
btrfs_commit_transaction, this will keep consistent behavior.
Also noinode_cache mount option will not stop the caching_kthread.
Cc: David Sterba <dsterba@suse.cz>
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Qu Wenruo <quwenruo@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
There is a bug that using btrfs_previous_item() to search metadata extent item.
This is because in btrfs_previous_item(), we need type match, however, since
skinny metada was introduced by josef, we may mix this two types. So just
use btrfs_previous_item() is not working right.
To keep btrfs_previous_item() like normal tree search, i introduce another
function btrfs_previous_extent_item().
Signed-off-by: Wang Shilong <wangsl.fnst@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
Check if we support skinny metadata firstly and fix to use
right type to search.
Signed-off-by: Wang Shilong <wangsl.fnst@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
It is possible for the send feature to send clone operations that
request a cloning range (offset + length) that is not aligned with
the block size. This makes the btrfs receive command send issue a
clone ioctl call that will fail, as the ioctl will return an -EINVAL
error because of the unaligned range.
Fix this by not sending clone operations for non block aligned ranges,
and instead send regular write operation for these (less common) cases.
The following xfstest reproduces this issue, which fails on the second
btrfs receive command without this change:
seq=`basename $0`
seqres=$RESULT_DIR/$seq
echo "QA output created by $seq"
tmp=`mktemp -d`
status=1 # failure is the default!
trap "_cleanup; exit \$status" 0 1 2 3 15
_cleanup()
{
rm -fr $tmp
}
# get standard environment, filters and checks
. ./common/rc
. ./common/filter
# real QA test starts here
_supported_fs btrfs
_supported_os Linux
_require_scratch
_need_to_be_root
rm -f $seqres.full
_scratch_mkfs >/dev/null 2>&1
_scratch_mount
$XFS_IO_PROG -f -c "truncate 819200" $SCRATCH_MNT/foo | _filter_xfs_io
$BTRFS_UTIL_PROG filesystem sync $SCRATCH_MNT | _filter_scratch
$XFS_IO_PROG -c "falloc -k 819200 667648" $SCRATCH_MNT/foo | _filter_xfs_io
$BTRFS_UTIL_PROG filesystem sync $SCRATCH_MNT | _filter_scratch
$XFS_IO_PROG -f -c "pwrite 1482752 2978" $SCRATCH_MNT/foo | _filter_xfs_io
$BTRFS_UTIL_PROG filesystem sync $SCRATCH_MNT | _filter_scratch
$BTRFS_UTIL_PROG subvol snapshot -r $SCRATCH_MNT $SCRATCH_MNT/mysnap1 | \
_filter_scratch
$XFS_IO_PROG -f -c "truncate 883305" $SCRATCH_MNT/foo | _filter_xfs_io
$BTRFS_UTIL_PROG filesystem sync $SCRATCH_MNT | _filter_scratch
$BTRFS_UTIL_PROG subvol snapshot -r $SCRATCH_MNT $SCRATCH_MNT/mysnap2 | \
_filter_scratch
$BTRFS_UTIL_PROG send $SCRATCH_MNT/mysnap1 -f $tmp/1.snap 2>&1 | _filter_scratch
$BTRFS_UTIL_PROG send -p $SCRATCH_MNT/mysnap1 $SCRATCH_MNT/mysnap2 \
-f $tmp/2.snap 2>&1 | _filter_scratch
md5sum $SCRATCH_MNT/foo | _filter_scratch
md5sum $SCRATCH_MNT/mysnap1/foo | _filter_scratch
md5sum $SCRATCH_MNT/mysnap2/foo | _filter_scratch
_scratch_unmount
_check_btrfs_filesystem $SCRATCH_DEV
_scratch_mkfs >/dev/null 2>&1
_scratch_mount
$BTRFS_UTIL_PROG receive $SCRATCH_MNT -f $tmp/1.snap
md5sum $SCRATCH_MNT/mysnap1/foo | _filter_scratch
$BTRFS_UTIL_PROG receive $SCRATCH_MNT -f $tmp/2.snap
md5sum $SCRATCH_MNT/mysnap2/foo | _filter_scratch
_scratch_unmount
_check_btrfs_filesystem $SCRATCH_DEV
status=0
exit
The tests expected output is:
QA output created by 025
FSSync 'SCRATCH_MNT'
FSSync 'SCRATCH_MNT'
wrote 2978/2978 bytes at offset 1482752
XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
FSSync 'SCRATCH_MNT'
Create a readonly snapshot of 'SCRATCH_MNT' in 'SCRATCH_MNT/mysnap1'
FSSync 'SCRATCH_MNT'
Create a readonly snapshot of 'SCRATCH_MNT' in 'SCRATCH_MNT/mysnap2'
At subvol SCRATCH_MNT/mysnap1
At subvol SCRATCH_MNT/mysnap2
129b8eaee8d3c2bcad49bec596591cb3 SCRATCH_MNT/foo
42b6369eae2a8725c1aacc0440e597aa SCRATCH_MNT/mysnap1/foo
129b8eaee8d3c2bcad49bec596591cb3 SCRATCH_MNT/mysnap2/foo
At subvol mysnap1
42b6369eae2a8725c1aacc0440e597aa SCRATCH_MNT/mysnap1/foo
At snapshot mysnap2
129b8eaee8d3c2bcad49bec596591cb3 SCRATCH_MNT/mysnap2/foo
Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
After the change titled "Btrfs: add support for inode properties", if
btrfs was built-in the kernel (i.e. not as a module), it would cause a
kernel panic, as reported recently by Fengguang:
[ 2.024722] BUG: unable to handle kernel NULL pointer dereference at (null)
[ 2.027814] IP: [<ffffffff81501594>] crc32c+0xc/0x6b
[ 2.028684] PGD 0
[ 2.028684] Oops: 0000 [#1] SMP
[ 2.028684] Modules linked in:
[ 2.028684] CPU: 0 PID: 1 Comm: swapper/0 Not tainted 3.13.0-rc7-04795-ga7b57c2 #1
[ 2.028684] Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011
[ 2.028684] task: ffff88000edba100 ti: ffff88000edd6000 task.ti: ffff88000edd6000
[ 2.028684] RIP: 0010:[<ffffffff81501594>] [<ffffffff81501594>] crc32c+0xc/0x6b
[ 2.028684] RSP: 0000:ffff88000edd7e58 EFLAGS: 00010246
[ 2.028684] RAX: 0000000000000000 RBX: ffffffff82295550 RCX: 0000000000000000
[ 2.028684] RDX: 0000000000000011 RSI: ffffffff81efe393 RDI: 00000000fffffffe
[ 2.028684] RBP: ffff88000edd7e60 R08: 0000000000000003 R09: 0000000000015d20
[ 2.028684] R10: ffffffff81ef225e R11: ffffffff811b0222 R12: ffffffffffffffff
[ 2.028684] R13: 0000000000000239 R14: 0000000000000000 R15: 0000000000000000
[ 2.028684] FS: 0000000000000000(0000) GS:ffff88000fa00000(0000) knlGS:0000000000000000
[ 2.028684] CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b
[ 2.028684] CR2: 0000000000000000 CR3: 000000000220c000 CR4: 00000000000006f0
[ 2.028684] Stack:
[ 2.028684] ffffffff82295550 ffff88000edd7e80 ffffffff8238af62 ffffffff8238ac05
[ 2.028684] 0000000000000000 ffff88000edd7e98 ffffffff8238ac0f ffffffff8238ac05
[ 2.028684] ffff88000edd7f08 ffffffff810002ba ffff88000edd7f00 ffffffff810e2404
[ 2.028684] Call Trace:
[ 2.028684] [<ffffffff8238af62>] btrfs_props_init+0x4f/0x96
[ 2.028684] [<ffffffff8238ac05>] ? ftrace_define_fields_btrfs_space_reservation+0x145/0x145
[ 2.028684] [<ffffffff8238ac0f>] init_btrfs_fs+0xa/0xf0
[ 2.028684] [<ffffffff8238ac05>] ? ftrace_define_fields_btrfs_space_reservation+0x145/0x145
[ 2.028684] [<ffffffff810002ba>] do_one_initcall+0xa4/0x13a
[ 2.028684] [<ffffffff810e2404>] ? parse_args+0x25f/0x33d
[ 2.028684] [<ffffffff8234cf75>] kernel_init_freeable+0x1aa/0x230
[ 2.028684] [<ffffffff8234c785>] ? do_early_param+0x88/0x88
[ 2.028684] [<ffffffff819f61b5>] ? rest_init+0x89/0x89
[ 2.028684] [<ffffffff819f61c3>] kernel_init+0xe/0x109
The issue here is that the initialization function of btrfs (super.c:init_btrfs_fs)
started using crc32c (from lib/libcrc32c.c). But when it needs to call crc32c (as
part of the properties initialization routine), the libcrc32c is not yet initialized,
so crc32c derreferenced a NULL pointer (lib/libcrc32c.c:tfm), causing the kernel
panic on boot.
The approach to fix this is to use crypto component directly to use its crc32c (which
is basically what lib/libcrc32c.c is, a wrapper around crypto). This is what ext4 is
doing as well, it uses crypto directly to get crc32c functionality.
Verified this works both when btrfs is built-in and when it's loadable kernel module.
Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
In the clone ioctl, when the source and target inodes are different,
we can acquire their mutexes in 2 possible different orders. After
we're done cloning, we were releasing the mutexes always in the same
order - the most correct way of doing it is to release them by the
reverse order they were acquired.
Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
Here we are not going to free memory, no need to remove every node
one by one, just init root node here is ok.
Cc: Liu Bo <bo.li.liu@oracle.com>
Signed-off-by: Wang Shilong <wangsl.fnst@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
We don't have to keep subvolume's block_rsv during transaction commit,
and within transaction commit, we may also need the free space reclaimed
from this block_rsv to process delayed refs.
Signed-off-by: Liu Bo <bo.li.liu@oracle.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
When we ran the 274th case of xfstests with nodatacow mount option,
We met the following warning message:
WARNING: CPU: 1 PID: 14185 at fs/btrfs/extent-tree.c:3734 btrfs_free_reserved_data_space+0xa6/0xd0
It is caused by the race between the write back and nocow buffered
write:
Task1 Task2
__btrfs_buffered_write()
skip data reservation
reserve the metadata space
copy the data
dirty the pages
unlock the pages
write back the pages
release the data space
becasue there is no
noreserve flag
set the noreserve flag
This patch fixes this problem by unlocking the pages after
the noreserve flag is set.
Reported-by: Tsutomu Itoh <t-itoh@jp.fujitsu.com>
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
The backref walking code will search down to the key it is looking for and then
proceed to walk _all_ of the extents on the file until it hits the end. This is
suboptimal with large files, we only need to look for as many extents as we have
references for that inode. I have a testcase that creates a randomly written 4
gig file and before this patch it took 6min 30sec to do the initial send, with
this patch it takes 2min 30sec to do the intial send. Thanks,
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
Could have sworn I fixed this before but apparently not. This makes us pass
btrfs/022 with skinny metadata enabled. Thanks,
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
I don't think this is an issue and I've not seen it in practice but
extent_from_logical will fail to find a skinny extent because it uses
btrfs_previous_item and gives it the normal extent item type. This is just not
a place to use btrfs_previous_item since we care about either normal extents or
skinny extents, so open code btrfs_previous_item to properly check. This would
only affect metadata and the only place this is used for metadata is scrub and
I'm pretty sure it's just for printing stuff out, not actually doing any work so
hopefully it was never a problem other than a cosmetic one. Thanks,
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
On one of our gluster clusters we noticed some pretty big lag spikes. This
turned out to be because our transaction commit was taking like 3 minutes to
complete. This is because we have like 30 gigs of metadata, so our global
reserve would end up being the max which is like 512 mb. So our throttling code
would allow a ridiculous amount of delayed refs to build up and then they'd all
get run at transaction commit time, and for a cold mounted file system that
could take up to 3 minutes to run. So fix the throttling to be based on both
the size of the global reserve and how long it takes us to run delayed refs.
This patch tracks the time it takes to run delayed refs and then only allows 1
seconds worth of outstanding delayed refs at a time. This way it will auto-tune
itself from cold cache up to when everything is in memory and it no longer has
to go to disk. This makes our transaction commits take much less time to run.
Thanks,
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
Currently we have two rb-trees, one for delayed ref heads and one for all of the
delayed refs, including the delayed ref heads. When we process the delayed refs
we have to hold onto the delayed ref lock for all of the selecting and merging
and such, which results in quite a bit of lock contention. This was solved by
having a waitqueue and only one flusher at a time, however this hurts if we get
a lot of delayed refs queued up.
So instead just have an rb tree for the delayed ref heads, and then attach the
delayed ref updates to an rb tree that is per delayed ref head. Then we only
need to take the delayed ref lock when adding new delayed refs and when
selecting a delayed ref head to process, all the rest of the time we deal with a
per delayed ref head lock which will be much less contentious.
The locking rules for this get a little more complicated since we have to lock
up to 3 things to properly process delayed refs, but I will address that problem
later. For now this passes all of xfstests and my overnight stress tests.
Thanks,
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
Looking into some performance related issues with large amounts of metadata
revealed that we can have some pretty huge swings in fsync() performance. If we
have a lot of delayed refs backed up (as you will tend to do with lots of
metadata) fsync() will wander off and try to run some of those delayed refs
which can result in reading from disk and such. Since the actual act of fsync()
doesn't create any delayed refs there is no need to make it throttle on delayed
ref stuff, that will be handled by other people. With this patch we get much
smoother fsync performance with large amounts of metadata. Thanks,
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
This change adds infrastructure to allow for generic properties for
inodes. Properties are name/value pairs that can be associated with
inodes for different purposes. They are stored as xattrs with the
prefix "btrfs."
Properties can be inherited - this means when a directory inode has
inheritable properties set, these are added to new inodes created
under that directory. Further, subvolumes can also have properties
associated with them, and they can be inherited from their parent
subvolume. Naturally, directory properties have priority over subvolume
properties (in practice a subvolume property is just a regular
property associated with the root inode, objectid 256, of the
subvolume's fs tree).
This change also adds one specific property implementation, named
"compression", whose values can be "lzo" or "zlib" and it's an
inheritable property.
The corresponding changes to btrfs-progs were also implemented.
A patch with xfstests for this feature will follow once there's
agreement on this change/feature.
Further, the script at the bottom of this commit message was used to
do some benchmarks to measure any performance penalties of this feature.
Basically the tests correspond to:
Test 1 - create a filesystem and mount it with compress-force=lzo,
then sequentially create N files of 64Kb each, measure how long it took
to create the files, unmount the filesystem, mount the filesystem and
perform an 'ls -lha' against the test directory holding the N files, and
report the time the command took.
Test 2 - create a filesystem and don't use any compression option when
mounting it - instead set the compression property of the subvolume's
root to 'lzo'. Then create N files of 64Kb, and report the time it took.
The unmount the filesystem, mount it again and perform an 'ls -lha' like
in the former test. This means every single file ends up with a property
(xattr) associated to it.
Test 3 - same as test 2, but uses 4 properties - 3 are duplicates of the
compression property, have no real effect other than adding more work
when inheriting properties and taking more btree leaf space.
Test 4 - same as test 3 but with 10 properties per file.
Results (in seconds, and averages of 5 runs each), for different N
numbers of files follow.
* Without properties (test 1)
file creation time ls -lha time
10 000 files 3.49 0.76
100 000 files 47.19 8.37
1 000 000 files 518.51 107.06
* With 1 property (compression property set to lzo - test 2)
file creation time ls -lha time
10 000 files 3.63 0.93
100 000 files 48.56 9.74
1 000 000 files 537.72 125.11
* With 4 properties (test 3)
file creation time ls -lha time
10 000 files 3.94 1.20
100 000 files 52.14 11.48
1 000 000 files 572.70 142.13
* With 10 properties (test 4)
file creation time ls -lha time
10 000 files 4.61 1.35
100 000 files 58.86 13.83
1 000 000 files 656.01 177.61
The increased latencies with properties are essencialy because of:
*) When creating an inode, we now synchronously write 1 more item
(an xattr item) for each property inherited from the parent dir
(or subvolume). This could be done in an asynchronous way such
as we do for dir intex items (delayed-inode.c), which could help
reduce the file creation latency;
*) With properties, we now have larger fs trees. For this particular
test each xattr item uses 75 bytes of leaf space in the fs tree.
This could be less by using a new item for xattr items, instead of
the current btrfs_dir_item, since we could cut the 'location' and
'type' fields (saving 18 bytes) and maybe 'transid' too (saving a
total of 26 bytes per xattr item) from the btrfs_dir_item type.
Also tried batching the xattr insertions (ignoring proper hash
collision handling, since it didn't exist) when creating files that
inherit properties from their parent inode/subvolume, but the end
results were (surprisingly) essentially the same.
Test script:
$ cat test.pl
#!/usr/bin/perl -w
use strict;
use Time::HiRes qw(time);
use constant NUM_FILES => 10_000;
use constant FILE_SIZES => (64 * 1024);
use constant DEV => '/dev/sdb4';
use constant MNT_POINT => '/home/fdmanana/btrfs-tests/dev';
use constant TEST_DIR => (MNT_POINT . '/testdir');
system("mkfs.btrfs", "-l", "16384", "-f", DEV) == 0 or die "mkfs.btrfs failed!";
# following line for testing without properties
#system("mount", "-o", "compress-force=lzo", DEV, MNT_POINT) == 0 or die "mount failed!";
# following 2 lines for testing with properties
system("mount", DEV, MNT_POINT) == 0 or die "mount failed!";
system("btrfs", "prop", "set", MNT_POINT, "compression", "lzo") == 0 or die "set prop failed!";
system("mkdir", TEST_DIR) == 0 or die "mkdir failed!";
my ($t1, $t2);
$t1 = time();
for (my $i = 1; $i <= NUM_FILES; $i++) {
my $p = TEST_DIR . '/file_' . $i;
open(my $f, '>', $p) or die "Error opening file!";
$f->autoflush(1);
for (my $j = 0; $j < FILE_SIZES; $j += 4096) {
print $f ('A' x 4096) or die "Error writing to file!";
}
close($f);
}
$t2 = time();
print "Time to create " . NUM_FILES . ": " . ($t2 - $t1) . " seconds.\n";
system("umount", DEV) == 0 or die "umount failed!";
system("mount", DEV, MNT_POINT) == 0 or die "mount failed!";
$t1 = time();
system("bash -c 'ls -lha " . TEST_DIR . " > /dev/null'") == 0 or die "ls failed!";
$t2 = time();
print "Time to ls -lha all files: " . ($t2 - $t1) . " seconds.\n";
system("umount", DEV) == 0 or die "umount failed!";
Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
When writing to a file we drop existing file extent items that cover the
write range and then add a new file extent item that represents that write
range.
Before this change we were doing a tree lookup to remove the file extent
items, and then after we did another tree lookup to insert the new file
extent item.
Most of the time all the file extent items we need to drop are located
within a single leaf - this is the leaf where our new file extent item ends
up at. Therefore, in this common case just combine these 2 operations into
a single one.
By avoiding the second btree navigation for insertion of the new file extent
item, we reduce btree node/leaf lock acquisitions/releases, btree block/leaf
COW operations, CPU time on btree node/leaf key binary searches, etc.
Besides for file writes, this is an operation that happens for file fsync's
as well. However log btrees are much less likely to big as big as regular
fs btrees, therefore the impact of this change is smaller.
The following benchmark was performed against an SSD drive and a
HDD drive, both for random and sequential writes:
sysbench --test=fileio --file-num=4096 --file-total-size=8G \
--file-test-mode=[rndwr|seqwr] --num-threads=512 \
--file-block-size=8192 \ --max-requests=1000000 \
--file-fsync-freq=0 --file-io-mode=sync [prepare|run]
All results below are averages of 10 runs of the respective test.
** SSD sequential writes
Before this change: 225.88 Mb/sec
After this change: 277.26 Mb/sec
** SSD random writes
Before this change: 49.91 Mb/sec
After this change: 56.39 Mb/sec
** HDD sequential writes
Before this change: 68.53 Mb/sec
After this change: 69.87 Mb/sec
** HDD random writes
Before this change: 13.04 Mb/sec
After this change: 14.39 Mb/sec
Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
We may return early in btrfs_drop_snapshot(), we shouldn't
call btrfs_std_err() for this case, fix it.
Cc: stable@vger.kernel.org
Signed-off-by: Wang Shilong <wangsl.fnst@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
We will finish orphan cleanups during snapshot, so we don't
have to commit transaction here.
Signed-off-by: Wang Shilong <wangsl.fnst@cn.fujitsu.com>
Reviewed-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
We should gurantee that parent and clone roots can not be destroyed
during send, for this we have two ideas.
1.by holding @subvol_sem, this might be a nightmare, because it will
block all subvolumes deletion for a long time.
2.Miao pointed out we can reuse @send_in_progress, that mean we will
skip snapshot deletion if root sending is in progress.
Here we adopt the second approach since it won't block other subvolumes
deletion for a long time.
Besides in btrfs_clean_one_deleted_snapshot(), we only check first root
, if this root is involved in send, we return directly rather than
continue to check.There are several reasons about it:
1.this case happen seldomly.
2.after sending,cleaner thread can continue to drop that root.
3.make code simple
Cc: David Sterba <dsterba@suse.cz>
Signed-off-by: Wang Shilong <wangsl.fnst@cn.fujitsu.com>
Reviewed-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
Steps to reproduce:
# mkfs.btrfs -f /dev/sda8
# mount /dev/sda8 /mnt
# btrfs sub snapshot -r /mnt /mnt/snap1
# btrfs sub snapshot -r /mnt /mnt/snap2
# btrfs send /mnt/snap1 -p /mnt/snap2 -f /mnt/1
# dmesg
The problem is that we will sort clone roots(include @send_root), it
might push @send_root before thus @send_root's @send_in_progress will
be decreased twice.
Cc: David Sterba <dsterba@suse.cz>
Signed-off-by: Wang Shilong <wangsl.fnst@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
Add treelog mount option to enable tree log with
remount option.
Signed-off-by: Qu Wenruo <quwenruo@cn.fujitsu.com>
Reviewed-by: David Sterba <dsterba@suse.cz>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
Add datasum mount option to enable checksum with
remount option.
Signed-off-by: Qu Wenruo <quwenruo@cn.fujitsu.com>
Reviewed-by: David Sterba <dsterba@suse.cz>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
Add datacow mount option to enable copy-on-write with
remount option.
Signed-off-by: Qu Wenruo <quwenruo@cn.fujitsu.com>
Reviewed-by: David Sterba <dsterba@suse.cz>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
Add acl mount option to enable acl with remount option.
Signed-off-by: Qu Wenruo <quwenruo@cn.fujitsu.com>
Reviewed-by: David Sterba <dsterba@suse.cz>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
Add noflushoncommit mount option to disable flush on commit with
remount option.
Signed-off-by: Qu Wenruo <quwenruo@cn.fujitsu.com>
Reviewed-by: David Sterba <dsterba@suse.cz>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
Add noenospc_debug mount option to disable ENOSPC debug with
remount option.
Signed-off-by: Qu Wenruo <quwenruo@cn.fujitsu.com>
Reviewed-by: David Sterba <dsterba@suse.cz>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
Add nodiscard mount option to disable discard with remount option.
Signed-off-by: Qu Wenruo <quwenruo@cn.fujitsu.com>
Reviewed-by: David Sterba <dsterba@suse.cz>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
Btrfs has autodefrag mount option but no pairing noautodefrag option,
which makes it impossible to disable autodefrag without umount.
Signed-off-by: Qu Wenruo <quwenruo@cn.fujitsu.com>
Reviewed-by: David Sterba <dsterba@suse.cz>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
Btrfs can be remounted without barrier, but there is no "barrier" option
so nobody can remount btrfs back with barrier on. Only umount and
mount again can re-enable barrier.(Quite awkward)
Also the mount options in the document is also changed slightly for the
further pairing options changes.
Reported-by: Daniel Blueman <daniel@quora.org>
Signed-off-by: Qu Wenruo <quwenruo@cn.fujitsu.com>
Signed-off-by: Mike Fleetwood <mike.fleetwood@googlemail.com>
Reviewed-by: David Sterba <dsterba@suse.cz>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
We only intent to fua the first superblock in every device from
comments, fix it.
Signed-off-by: Wang Shilong <wangsl.fnst@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
@full is not protected within global_rsv.lock, so we may think global_rsv
is already full but in fact it's not, so we miss the opportunity to return
free space to global_rsv directly when we release other block_rsvs.
Signed-off-by: Liu Bo <bo.li.liu@oracle.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
During balance test, we hit an oops:
[ 2013.841551] kernel BUG at fs/btrfs/relocation.c:1174!
The problem is that if we fail to relocate tree blocks, we should
update backref cache, otherwise, some pending nodes are not updated
while snapshot check @cache->last_trans is within one transaction
and won't update it and then oops happen.
Signed-off-by: Wang Shilong <wangsl.fnst@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
The following warning message was outputed when running the 274th case
of xfstests with nodatacow option:
BUG: Bad page state in process kswapd0 pfn:1c66f
page:ffffea0000636848 count:0 mapcount:0 mapping:(null) index:0x78000
page flags: 0x1000000000100a(error|uptodate|private_2)
It is because the check of nocow range was wrong, we should compare the
start and end position of the extent with the write position to verify
if the write position was in the extent, but the current code just used
the start postion to do the check, so we got the wrong extent and told
the caller that it was a nocow write. And then when we write back the
dirty pages, we found we should cow the extent, but at that time, there
was no space in the fs, we had to the error flag for the page. When
someone reclaimed that page, the above warning outputed. Fix it.
Reported-by: Tsutomu Itoh <t-itoh@jp.fujitsu.com>
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
Previously, we will free reloc root memory and then force filesystem
to be readonly. The problem is that there may be another thread commiting
transaction which will try to access freed reloc root during merging reloc
roots process.
To keep consistency snapshots shared space, we should allow snapshot
finished if possible, so here we don't free reloc root memory.
signed-off-by: Wang Shilong <wangsl.fnst@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
@nr is no longer used, remove it from select_reloc_root()
Signed-off-by: Wang Shilong <wangsl.fnst@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
If we do a btree search with the goal of updating an existing item
without changing its size (ins_len == 0 and cow == 1), then we never
need to hold locks on upper level nodes (even when slot == 0) after we
COW their child nodes/leaves, as we won't have node splits or merges
in this scenario (that is, no key additions, removals or shifts on any
nodes or leaves).
Therefore release the locks immediately after COWing the child nodes/leaves
while navigating the btree, even if their parent slot is 0, instead of
returning a path to the caller with those nodes locked, which would get
released only when the caller releases or frees the path (or if it calls
btrfs_unlock_up_safe).
This is a common scenario, for example when updating inode items in fs
trees and block group items in the extent tree.
The following benchmarks were performed on a quad core machine with 32Gb
of ram, using a leaf/node size of 4Kb (to generate deeper fs trees more
quickly).
sysbench --test=fileio --file-num=131072 --file-total-size=8G \
--file-test-mode=seqwr --num-threads=512 --file-block-size=8192 \
--max-requests=100000 --file-io-mode=sync [prepare|run]
Before this change: 49.85Mb/s (average of 5 runs)
After this change: 50.38Mb/s (average of 5 runs)
Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
The local variable 'new_size' comes from userspace. If a large number
was passed, there would be an integer overflow in the following line:
new_size = old_size + new_size;
Signed-off-by: Wenliang Fan <fanwlexca@gmail.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
We can starve out the transaction commit with a bunch of caching threads all
running at the same time. This is because we will only drop the
extent_commit_sem if we need_resched(), which isn't likely to happen since we
will be reading a lot from the disk so have already schedule()'ed plenty. Alex
observed that he could starve out a transaction commit for up to a minute with
32 caching threads all running at once. This will allow us to drop the
extent_commit_sem to allow the transaction commit to swap the commit_root out
and then all the cachers will start back up. Here is an explanation provided by
Igno
So, just to fill in what happens in this loop:
mutex_unlock(&caching_ctl->mutex);
cond_resched();
goto again;
where 'again:' takes caching_ctl->mutex and fs_info->extent_commit_sem
again:
again:
mutex_lock(&caching_ctl->mutex);
/* need to make sure the commit_root doesn't disappear */
down_read(&fs_info->extent_commit_sem);
So, if I'm reading the code correct, there can be a fair amount of
concurrency here: there may be multiple 'caching kthreads' per filesystem
active, while there's one fs_info->extent_commit_sem per filesystem
AFAICS.
So, what happens if there are a lot of CPUs all busy holding the
->extent_commit_sem rwsem read-locked and a writer arrives? They'd all
rush to try to release the fs_info->extent_commit_sem, and they'd block in
the down_read() because there's a writer waiting.
So there's a guarantee of forward progress. This should answer akpm's
concern I think.
Thanks,
Acked-by: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Signed-off-by: Chris Mason <clm@fb.com>
The inode reference item is close to inode item, so we insert it simultaneously
with the inode item insertion when we create a file/directory.. In fact, we also
can handle the inode reference deletion by the same way. So we made this patch to
introduce the delayed inode reference deletion for the single link inode(At most
case, the file doesn't has hard link, so we don't take the hard link into account).
This function is based on the delayed inode mechanism. After applying this patch,
we can reduce the time of the file/directory deletion by ~10%.
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Chris Mason <clm@fb.com>
Two reasons:
- btrfs_end_transaction_dmeta() is the same as btrfs_end_transaction_throttle()
so it is unnecessary.
- All the delayed items should be dealt in the current transaction, so the
workers should not commit the transaction, instead, deal with the delayed
items as many as possible.
So we can remove btrfs_end_transaction_dmeta()
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Chris Mason <clm@fb.com>
- move the condition check for wait into a function
- use wait_event_interruptible instead of prepare-schedule-finish process
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Chris Mason <clm@fb.com>
If the number of the delayed items is greater than the upper limit, we will
try to flush all the delayed items. After that, it is unnecessary to run
them again because they are being dealt with by the wokers or the number of
them is less than the lower limit.
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Chris Mason <clm@fb.com>
Before applying the patch
commit de3cb945db
title: Btrfs: improve the delayed inode throttling
We need requeue the async work after the current work was done, it
introduced a deadlock problem. So we wrote the code that this patch
removes to avoid the above problem. But after applying the above
patch, the deadlock problem didn't exist. So we should remove that
fix code.
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Chris Mason <clm@fb.com>
Convert all applicable cases of printk and pr_* to the btrfs_* macros.
Fix all uses of the BTRFS prefix.
Signed-off-by: Frank Holton <fholton@gmail.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
While running the test btrfs/004 from xfstests in a loop, it failed
about 1 time out of 20 runs in my desktop. The failure happened in
the backref walking part of the test, and the test's error message was
like this:
btrfs/004 93s ... [failed, exit status 1] - output mismatch (see /home/fdmanana/git/hub/xfstests_2/results//btrfs/004.out.bad)
--- tests/btrfs/004.out 2013-11-26 18:25:29.263333714 +0000
+++ /home/fdmanana/git/hub/xfstests_2/results//btrfs/004.out.bad 2013-12-10 15:25:10.327518516 +0000
@@ -1,3 +1,8 @@
QA output created by 004
*** test backref walking
-*** done
+unexpected output from
+ /home/fdmanana/git/hub/btrfs-progs/btrfs inspect-internal logical-resolve -P 141512704 /home/fdmanana/btrfs-tests/scratch_1
+expected inum: 405, expected address: 454656, file: /home/fdmanana/btrfs-tests/scratch_1/snap1/p0/d6/d3d/d156/fce, got:
+
...
(Run 'diff -u tests/btrfs/004.out /home/fdmanana/git/hub/xfstests_2/results//btrfs/004.out.bad' to see the entire diff)
Ran: btrfs/004
Failures: btrfs/004
Failed 1 of 1 tests
But immediately after the test finished, the btrfs inspect-internal command
returned the expected output:
$ btrfs inspect-internal logical-resolve -P 141512704 /home/fdmanana/btrfs-tests/scratch_1
inode 405 offset 454656 root 258
inode 405 offset 454656 root 5
It turned out this was because the btrfs_search_old_slot() calls performed
during backref walking (backref.c:__resolve_indirect_ref) were not finding
anything. The reason for this turned out to be that the tree mod logging
code was not logging some node multi-step operations atomically, therefore
btrfs_search_old_slot() callers iterated often over an incomplete tree that
wasn't fully consistent with any tree state from the past. Besides missing
items, this often (but not always) resulted in -EIO errors during old slot
searches, reported in dmesg like this:
[ 4299.933936] ------------[ cut here ]------------
[ 4299.933949] WARNING: CPU: 0 PID: 23190 at fs/btrfs/ctree.c:1343 btrfs_search_old_slot+0x57b/0xab0 [btrfs]()
[ 4299.933950] Modules linked in: btrfs raid6_pq xor pci_stub vboxpci(O) vboxnetadp(O) vboxnetflt(O) vboxdrv(O) bnep rfcomm bluetooth parport_pc ppdev binfmt_misc joydev snd_hda_codec_h
[ 4299.933977] CPU: 0 PID: 23190 Comm: btrfs Tainted: G W O 3.12.0-fdm-btrfs-next-16+ #70
[ 4299.933978] Hardware name: To Be Filled By O.E.M. To Be Filled By O.E.M./Z77 Pro4, BIOS P1.50 09/04/2012
[ 4299.933979] 000000000000053f ffff8806f3fd98f8 ffffffff8176d284 0000000000000007
[ 4299.933982] 0000000000000000 ffff8806f3fd9938 ffffffff8104a81c ffff880659c64b70
[ 4299.933984] ffff880659c643d0 ffff8806599233d8 ffff880701e2e938 0000160000000000
[ 4299.933987] Call Trace:
[ 4299.933991] [<ffffffff8176d284>] dump_stack+0x55/0x76
[ 4299.933994] [<ffffffff8104a81c>] warn_slowpath_common+0x8c/0xc0
[ 4299.933997] [<ffffffff8104a86a>] warn_slowpath_null+0x1a/0x20
[ 4299.934003] [<ffffffffa065d3bb>] btrfs_search_old_slot+0x57b/0xab0 [btrfs]
[ 4299.934005] [<ffffffff81775f3b>] ? _raw_read_unlock+0x2b/0x50
[ 4299.934010] [<ffffffffa0655001>] ? __tree_mod_log_search+0x81/0xc0 [btrfs]
[ 4299.934019] [<ffffffffa06dd9b0>] __resolve_indirect_refs+0x130/0x5f0 [btrfs]
[ 4299.934027] [<ffffffffa06a21f1>] ? free_extent_buffer+0x61/0xc0 [btrfs]
[ 4299.934034] [<ffffffffa06de39c>] find_parent_nodes+0x1fc/0xe40 [btrfs]
[ 4299.934042] [<ffffffffa06b13e0>] ? defrag_lookup_extent+0xe0/0xe0 [btrfs]
[ 4299.934048] [<ffffffffa06b13e0>] ? defrag_lookup_extent+0xe0/0xe0 [btrfs]
[ 4299.934056] [<ffffffffa06df980>] iterate_extent_inodes+0xe0/0x250 [btrfs]
[ 4299.934058] [<ffffffff817762db>] ? _raw_spin_unlock+0x2b/0x50
[ 4299.934065] [<ffffffffa06dfb82>] iterate_inodes_from_logical+0x92/0xb0 [btrfs]
[ 4299.934071] [<ffffffffa06b13e0>] ? defrag_lookup_extent+0xe0/0xe0 [btrfs]
[ 4299.934078] [<ffffffffa06b7015>] btrfs_ioctl+0xf65/0x1f60 [btrfs]
[ 4299.934080] [<ffffffff811658b8>] ? handle_mm_fault+0x278/0xb00
[ 4299.934083] [<ffffffff81075563>] ? up_read+0x23/0x40
[ 4299.934085] [<ffffffff8177a41c>] ? __do_page_fault+0x20c/0x5a0
[ 4299.934088] [<ffffffff811b2946>] do_vfs_ioctl+0x96/0x570
[ 4299.934090] [<ffffffff81776e23>] ? error_sti+0x5/0x6
[ 4299.934093] [<ffffffff810b71e8>] ? trace_hardirqs_off_caller+0x28/0xd0
[ 4299.934096] [<ffffffff81776a09>] ? retint_swapgs+0xe/0x13
[ 4299.934098] [<ffffffff811b2eb1>] SyS_ioctl+0x91/0xb0
[ 4299.934100] [<ffffffff813eecde>] ? trace_hardirqs_on_thunk+0x3a/0x3f
[ 4299.934102] [<ffffffff8177ef12>] system_call_fastpath+0x16/0x1b
[ 4299.934102] [<ffffffff8177ef12>] system_call_fastpath+0x16/0x1b
[ 4299.934104] ---[ end trace 48f0cfc902491414 ]---
[ 4299.934378] btrfs bad fsid on block 0
These tree mod log operations that must be performed atomically, tree_mod_log_free_eb,
tree_mod_log_eb_copy, tree_mod_log_insert_root and tree_mod_log_insert_move, used to
be performed atomically before the following commit:
c8cc634165
(Btrfs: stop using GFP_ATOMIC for the tree mod log allocations)
That change removed the atomicity of such operations. This patch restores the
atomicity while still not doing the GFP_ATOMIC allocations of tree_mod_elem
structures, so it has to do the allocations using GFP_NOFS before acquiring
the mod log lock.
This issue has been experienced by several users recently, such as for example:
http://www.spinics.net/lists/linux-btrfs/msg28574.html
After running the btrfs/004 test for 679 consecutive iterations with this
patch applied, I didn't ran into the issue anymore.
Cc: stable@vger.kernel.org
Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
Warn if the balance goes below zero, which appears to be unlikely
though. Otherwise cleans up the code a bit.
Signed-off-by: David Sterba <dsterba@suse.cz>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
Since daivd did the work that force us to use readonly snapshot,
we can safely remove transaction protection from btrfs send.
Signed-off-by: Wang Shilong <wangsl.fnst@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
We met the following oops when doing space balance:
kobject (ffff88081b590278): tried to init an initialized object, something is seriously wrong.
...
Call Trace:
[<ffffffff81937262>] dump_stack+0x49/0x5f
[<ffffffff8137d259>] kobject_init+0x89/0xa0
[<ffffffff8137d36a>] kobject_init_and_add+0x2a/0x70
[<ffffffffa009bd79>] ? clear_extent_bit+0x199/0x470 [btrfs]
[<ffffffffa005e82c>] __link_block_group+0xfc/0x120 [btrfs]
[<ffffffffa006b9db>] btrfs_make_block_group+0x24b/0x370 [btrfs]
[<ffffffffa00a899b>] __btrfs_alloc_chunk+0x54b/0x7e0 [btrfs]
[<ffffffffa00a8c6f>] btrfs_alloc_chunk+0x3f/0x50 [btrfs]
[<ffffffffa0060123>] do_chunk_alloc+0x363/0x440 [btrfs]
[<ffffffffa00633d4>] btrfs_check_data_free_space+0x104/0x310 [btrfs]
[<ffffffffa0069f4d>] btrfs_write_dirty_block_groups+0x48d/0x600 [btrfs]
[<ffffffffa007aad4>] commit_cowonly_roots+0x184/0x250 [btrfs]
...
Steps to reproduce:
# mkfs.btrfs -f <dev>
# mount -o nospace_cache <dev> <mnt>
# btrfs balance start <mnt>
# dd if=/dev/zero of=<mnt>/tmpfile bs=1M count=1
The reason of this problem is that we initialized the raid kobject when we added
a block group into a empty raid list. As we know, when we mounted a btrfs filesystem,
the raid list was empty, we would initialize the raid kobject when we added the first
block group. But if there was not data stored in the block group, the block group
would be freed when doing balance, and the raid list would be empty. And then if we
allocated a new block group and added it into the raid list, we would initialize
the raid kobject again, the oops happened.
Fix this problem by initializing the raid kobject just when mounting the fs.
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Reported-by: Wang Shilong <wangsl.fnst@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
All the subvolues that are involved in send must be read-only during the
whole operation. The ioctl SUBVOL_SETFLAGS could be used to change the
status to read-write and the result of send stream is undefined if the
data change unexpectedly.
Fix that by adding a refcount for all involved roots and verify that
there's no send in progress during SUBVOL_SETFLAGS ioctl call that does
read-only -> read-write transition.
We need refcounts because there are no restrictions on number of send
parallel operations currently run on a single subvolume, be it source,
parent or one of the multiple clone sources.
Kernel is silent when the RO checks fail and returns EPERM. The same set
of checks is done already in userspace before send starts.
Signed-off-by: David Sterba <dsterba@suse.cz>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
Unused since ed2590953b
"Btrfs: stop using vfs_read in send".
Signed-off-by: David Sterba <dsterba@suse.cz>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
Remove ifdefed code:
- tlv_put for 8, 16 and 32, add a generic tempalte if needed in future
- tlv_put_timespec - the btrfs_timespec fields are used
- fs_path_remove obsoleted long ago
Signed-off-by: David Sterba <dsterba@suse.cz>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
While running btrfs/004 from xfstests, after 503 iterations, dmesg reported
a deadlock between tasks iterating inode refs and tasks running delayed inodes
(during a transaction commit).
It turns out that iterating inode refs implies doing one tree search and
release all nodes in the path except the leaf node, and then passing that
leaf node to btrfs_ref_to_path(), which in turn does another tree search
without releasing the lock on the leaf node it received as parameter.
This is a problem when other task wants to write to the btree as well and
ends up updating the leaf that is read locked - the writer task locks the
parent of the leaf and then blocks waiting for the leaf's lock to be
released - at the same time, the task executing btrfs_ref_to_path()
does a second tree search, without releasing the lock on the first leaf,
and wants to access a leaf (the same or another one) that is a child of
the same parent, resulting in a deadlock.
The trace reported by lockdep follows.
[84314.936373] INFO: task fsstress:11930 blocked for more than 120 seconds.
[84314.936381] Tainted: G W O 3.12.0-fdm-btrfs-next-16+ #70
[84314.936383] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
[84314.936386] fsstress D ffff8806e1bf8000 0 11930 11926 0x00000000
[84314.936393] ffff8804d6d89b78 0000000000000046 ffff8804d6d89b18 ffffffff810bd8bd
[84314.936399] ffff8806e1bf8000 ffff8804d6d89fd8 ffff8804d6d89fd8 ffff8804d6d89fd8
[84314.936405] ffff880806308000 ffff8806e1bf8000 ffff8804d6d89c08 ffff8804deb8f190
[84314.936410] Call Trace:
[84314.936421] [<ffffffff810bd8bd>] ? trace_hardirqs_on+0xd/0x10
[84314.936428] [<ffffffff81774269>] schedule+0x29/0x70
[84314.936451] [<ffffffffa0715bf5>] btrfs_tree_lock+0x75/0x270 [btrfs]
[84314.936457] [<ffffffff810715c0>] ? __init_waitqueue_head+0x60/0x60
[84314.936470] [<ffffffffa06ba231>] btrfs_search_slot+0x7f1/0x930 [btrfs]
[84314.936489] [<ffffffffa0731c2a>] ? __btrfs_run_delayed_items+0x13a/0x1e0 [btrfs]
[84314.936504] [<ffffffffa06d2e1f>] btrfs_lookup_inode+0x2f/0xa0 [btrfs]
[84314.936510] [<ffffffff810bd6ef>] ? trace_hardirqs_on_caller+0x1f/0x1e0
[84314.936528] [<ffffffffa073173c>] __btrfs_update_delayed_inode+0x4c/0x1d0 [btrfs]
[84314.936543] [<ffffffffa0731c2a>] ? __btrfs_run_delayed_items+0x13a/0x1e0 [btrfs]
[84314.936558] [<ffffffffa0731c2a>] ? __btrfs_run_delayed_items+0x13a/0x1e0 [btrfs]
[84314.936573] [<ffffffffa0731c82>] __btrfs_run_delayed_items+0x192/0x1e0 [btrfs]
[84314.936589] [<ffffffffa0731d03>] btrfs_run_delayed_items+0x13/0x20 [btrfs]
[84314.936604] [<ffffffffa06dbcd4>] btrfs_flush_all_pending_stuffs+0x24/0x80 [btrfs]
[84314.936620] [<ffffffffa06ddc13>] btrfs_commit_transaction+0x223/0xa20 [btrfs]
[84314.936630] [<ffffffffa06ae5ae>] btrfs_sync_fs+0x6e/0x110 [btrfs]
[84314.936635] [<ffffffff811d0b50>] ? __sync_filesystem+0x60/0x60
[84314.936639] [<ffffffff811d0b50>] ? __sync_filesystem+0x60/0x60
[84314.936643] [<ffffffff811d0b70>] sync_fs_one_sb+0x20/0x30
[84314.936648] [<ffffffff811a3541>] iterate_supers+0xf1/0x100
[84314.936652] [<ffffffff811d0c45>] sys_sync+0x55/0x90
[84314.936658] [<ffffffff8177ef12>] system_call_fastpath+0x16/0x1b
[84314.936660] INFO: lockdep is turned off.
[84314.936663] INFO: task btrfs:11955 blocked for more than 120 seconds.
[84314.936666] Tainted: G W O 3.12.0-fdm-btrfs-next-16+ #70
[84314.936668] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
[84314.936670] btrfs D ffff880541729a88 0 11955 11608 0x00000000
[84314.936674] ffff880541729a38 0000000000000046 ffff8805417299d8 ffffffff810bd8bd
[84314.936680] ffff88075430c8a0 ffff880541729fd8 ffff880541729fd8 ffff880541729fd8
[84314.936685] ffffffff81c104e0 ffff88075430c8a0 ffff8804de8b00b8 ffff8804de8b0000
[84314.936690] Call Trace:
[84314.936695] [<ffffffff810bd8bd>] ? trace_hardirqs_on+0xd/0x10
[84314.936700] [<ffffffff81774269>] schedule+0x29/0x70
[84314.936717] [<ffffffffa0715815>] btrfs_tree_read_lock+0xd5/0x140 [btrfs]
[84314.936721] [<ffffffff810715c0>] ? __init_waitqueue_head+0x60/0x60
[84314.936733] [<ffffffffa06ba201>] btrfs_search_slot+0x7c1/0x930 [btrfs]
[84314.936746] [<ffffffffa06bd505>] btrfs_find_item+0x55/0x160 [btrfs]
[84314.936763] [<ffffffffa06ff689>] ? free_extent_buffer+0x49/0xc0 [btrfs]
[84314.936780] [<ffffffffa073c9ca>] btrfs_ref_to_path+0xba/0x1e0 [btrfs]
[84314.936797] [<ffffffffa06f9719>] ? release_extent_buffer+0xb9/0xe0 [btrfs]
[84314.936813] [<ffffffffa06ff689>] ? free_extent_buffer+0x49/0xc0 [btrfs]
[84314.936830] [<ffffffffa073cb50>] inode_to_path+0x60/0xd0 [btrfs]
[84314.936846] [<ffffffffa073d365>] paths_from_inode+0x115/0x3c0 [btrfs]
[84314.936851] [<ffffffff8118dd44>] ? kmem_cache_alloc_trace+0x114/0x200
[84314.936868] [<ffffffffa0714494>] btrfs_ioctl+0xf14/0x2030 [btrfs]
[84314.936873] [<ffffffff817762db>] ? _raw_spin_unlock+0x2b/0x50
[84314.936877] [<ffffffff8116598f>] ? handle_mm_fault+0x34f/0xb00
[84314.936882] [<ffffffff81075563>] ? up_read+0x23/0x40
[84314.936886] [<ffffffff8177a41c>] ? __do_page_fault+0x20c/0x5a0
[84314.936892] [<ffffffff811b2946>] do_vfs_ioctl+0x96/0x570
[84314.936896] [<ffffffff81776e23>] ? error_sti+0x5/0x6
[84314.936901] [<ffffffff810b71e8>] ? trace_hardirqs_off_caller+0x28/0xd0
[84314.936906] [<ffffffff81776a09>] ? retint_swapgs+0xe/0x13
[84314.936910] [<ffffffff811b2eb1>] SyS_ioctl+0x91/0xb0
[84314.936915] [<ffffffff813eecde>] ? trace_hardirqs_on_thunk+0x3a/0x3f
[84314.936920] [<ffffffff8177ef12>] system_call_fastpath+0x16/0x1b
[84314.936922] INFO: lockdep is turned off.
[84434.866873] INFO: task btrfs-transacti:11921 blocked for more than 120 seconds.
[84434.866881] Tainted: G W O 3.12.0-fdm-btrfs-next-16+ #70
[84434.866883] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
[84434.866886] btrfs-transacti D ffff880755b6a478 0 11921 2 0x00000000
[84434.866893] ffff8800735b9ce8 0000000000000046 ffff8800735b9c88 ffffffff810bd8bd
[84434.866899] ffff8805a1b848a0 ffff8800735b9fd8 ffff8800735b9fd8 ffff8800735b9fd8
[84434.866904] ffffffff81c104e0 ffff8805a1b848a0 ffff880755b6a478 ffff8804cece78f0
[84434.866910] Call Trace:
[84434.866920] [<ffffffff810bd8bd>] ? trace_hardirqs_on+0xd/0x10
[84434.866927] [<ffffffff81774269>] schedule+0x29/0x70
[84434.866948] [<ffffffffa06dd2ef>] wait_current_trans.isra.33+0xbf/0x120 [btrfs]
[84434.866954] [<ffffffff810715c0>] ? __init_waitqueue_head+0x60/0x60
[84434.866970] [<ffffffffa06dec18>] start_transaction+0x388/0x5a0 [btrfs]
[84434.866985] [<ffffffffa06db9b5>] ? transaction_kthread+0xb5/0x280 [btrfs]
[84434.866999] [<ffffffffa06dee97>] btrfs_attach_transaction+0x17/0x20 [btrfs]
[84434.867012] [<ffffffffa06dba9e>] transaction_kthread+0x19e/0x280 [btrfs]
[84434.867026] [<ffffffffa06db900>] ? open_ctree+0x2260/0x2260 [btrfs]
[84434.867030] [<ffffffff81070dad>] kthread+0xed/0x100
[84434.867035] [<ffffffff81070cc0>] ? flush_kthread_worker+0x190/0x190
[84434.867040] [<ffffffff8177ee6c>] ret_from_fork+0x7c/0xb0
[84434.867044] [<ffffffff81070cc0>] ? flush_kthread_worker+0x190/0x190
Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
Chris introduced hleper function read_csums() and this function
has been removed, but we forgot to remove its corresponding comments.
Signed-off-by: Wang Shilong <wangsl.fnst@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
It's not used anywhere, so just drop it.
Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
fs/btrfs/file.c: In function ‘prepare_pages.isra.18’:
fs/btrfs/file.c:1265:6: warning: ‘err’ may be used uninitialized in this function [-Wuninitialized]
Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
We have commited transaction before, remove redundant filemap writting and
waiting here, it can speed up balance relocation process.
Signed-off-by: Wang Shilong <wangsl.fnst@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
Clean up btrfs_lookup_dentry() to never return NULL, but PTR_ERR(-ENOENT)
instead. This keeps the return value convention consistent.
Callers who use btrfs_lookup_dentry() require a trivial update.
create_snapshot() in particular looks like it can also lose a BUG_ON(!inode)
which is not really needed - there seems less harm in returning ENOENT to
userspace at that point in the stack than there is to crash the machine.
Signed-off-by: Tsutomu Itoh <t-itoh@jp.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
In ctree.c:tree_mod_log_set_node_key() we were calling
__tree_mod_log_insert_key() even when the modification doesn't need
to be logged. This would allocate a tree_mod_elem structure, fill it
and pass it to __tree_mod_log_insert(), which would just acquire
the tree mod log write lock and then free the tree_mod_elem structure
and return (that is, a no-op).
Therefore call tree_mod_log_insert() instead of __tree_mod_log_insert()
which just returns immediately if the modification doesn't need to be
logged (without allocating the structure, fill it, acquire write lock,
free structure).
Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
I need to create a fake tree to test qgroups and I don't want to have to setup a
fake btree_inode. The fact is we only use the radix tree for the fs_info, so
everybody else who allocates an extent_io_tree is just wasting the space anyway.
This patch moves the radix tree and its lock into btrfs_fs_info so there is less
stuff I have to fake to do qgroup sanity tests. Thanks,
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
For creating a dummy in-memory btree I need to be able to use the radix tree to
keep track of the buffers like normal extent buffers. With dummy buffers we
skip the radix tree step, and we still want to do that for the tree mod log
dummy buffers but for my test buffers we need to be able to remove them from the
radix tree like normal. This will give me a way to do that. Thanks,
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
I need to add infrastructure to allocate dummy extent buffers for running sanity
tests, and to do this I need to not have to worry about having an
address_mapping for an io_tree, so just fix up the places where we assume that
all io_tree's have a non-NULL ->mapping. Thanks,
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
Currently when finding the leaf to insert a key into a btree, if the
leaf doesn't have enough space to store the item we attempt to move
off some items from our leaf to its right neighbor leaf, and if this
fails to create enough free space in our leaf, we try to move off more
items to the left neighbor leaf as well.
When trying to move off items to the right neighbor leaf, if it has
enough room to store the new key but not not enough room to move off
at least one item from our target leaf, __push_leaf_right returns 1 and
we have to attempt to move items to the left neighbor (push_leaf_left
function) without touching the right neighbor leaf.
For the case where the right leaf has enough room to store at least 1
item from our leaf, we end up modifying (and dirtying) both our leaf
and the right leaf. This is non-optimal for the case where the new key
is greater than any key in our target leaf because it can be inserted at
slot 0 of the right neighbor leaf and we don't need to touch our leaf
at all nor to attempt to move off items to the left neighbor leaf.
Therefore this change just selects the right neighbor leaf as our new
target leaf if it has enough room for the new key without modifying our
initial target leaf - we do this only if the new key is higher than any
key in the initial target leaf.
While running the following test, push_leaf_right was called by split_leaf
4802 times. Out of those 4802 calls, for 2571 calls (53.5%) we hit this
special case (right leaf has enough room and new key is higher than any key
in the initial target leaf).
Test:
sysbench --test=fileio --file-num=512 --file-total-size=5G \
--file-test-mode=[seqwr|rndwr] --num-threads=512 --file-block-size=8192 \
--max-requests=100000 --file-io-mode=sync [prepare|run]
Results:
sequential writes
Throughput before this change: 65.71Mb/sec (average of 10 runs)
Throughput after this change: 66.58Mb/sec (average of 10 runs)
random writes
Throughput before this change: 10.75Mb/sec (average of 10 runs)
Throughput after this change: 11.56Mb/sec (average of 10 runs)
Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
Reviewed-by: Liu Bo <bo.li.liu@oracle.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
Just wrap same code into one function scrub_blocked_if_needed().
This make a change that we will move waiting (@workers_pending = 0)
before we can wake up commiting transaction(atomic_inc(@scrub_paused)),
we must take carefully to not deadlock here.
Thread 1 Thread 2
|->btrfs_commit_transaction()
|->set trans type(COMMIT_DOING)
|->btrfs_scrub_paused()(blocked)
|->join_transaction(blocked)
Move btrfs_scrub_paused() before setting trans type which means we can
still join a transaction when commiting_transaction is blocked.
Signed-off-by: Wang Shilong <wangsl.fnst@cn.fujitsu.com>
Suggested-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
We came a race condition when scrubbing superblocks, the story is:
In commiting transaction, we will update @last_trans_commited after
writting superblocks, if scrubber start after writting superblocks
and before updating @last_trans_commited, generation mismatch happens!
We fix this by checking @scrub_pause_req, and we won't start a srubber
until commiting transaction is finished.(after btrfs_scrub_continue()
finished.)
Reported-by: Sebastian Ochmann <ochmann@informatik.uni-bonn.de>
Signed-off-by: Wang Shilong <wangsl.fnst@cn.fujitsu.com>
Reviewed-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
fs/btrfs/send.c:2190:9: warning: incorrect type in argument 3 (different base types)
fs/btrfs/send.c:2190:9: expected unsigned long long [unsigned] [usertype] value
fs/btrfs/send.c:2190:9: got restricted __le64 [usertype] ctransid
fs/btrfs/send.c:2195:17: warning: incorrect type in argument 3 (different base types)
fs/btrfs/send.c:2195:17: expected unsigned long long [unsigned] [usertype] value
fs/btrfs/send.c:2195:17: got restricted __le64 [usertype] ctransid
fs/btrfs/send.c:3716:9: warning: incorrect type in argument 3 (different base types)
fs/btrfs/send.c:3716:9: expected unsigned long long [unsigned] [usertype] value
fs/btrfs/send.c:3716:9: got restricted __le64 [usertype] ctransid
Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
When merging an extent_map with its right neighbor, increment
its block_len with the neighbor's block_len.
Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
[commit 8185554d: fix incorrect inode acl reset] introduced a dead
code by adding a condition which can never be true to an else
branch. The condition can never be true because it is already
checked by a previous if statement which causes function to return.
Signed-off-by: Michal Nazarewicz <mina86@mina86.com>
Reviewed-By: Filipe David Borba Manana <fdmanana@gmail.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
We were accounting for sizeof(struct btrfs_item) twice, once
in the data_size variable and another time in the if statement
below.
Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
Currently we do 2 traversals of an inode's extent_io_tree
before inserting an extent state structure: 1 to see if a
matching extent state already exists and 1 to do the insertion
if the fist traversal didn't found such extent state.
This change just combines those tree traversals into a single one.
While running sysbench tests (random writes) I captured the number
of elements in extent_io_tree trees for a while (into a procfs file
backed by a seq_list from seq_file module) and got this histogram:
Count: 9310
Range: 51.000 - 21386.000; Mean: 11785.243; Median: 18743.500; Stddev: 8923.688
Percentiles: 90th: 20985.000; 95th: 21155.000; 99th: 21369.000
51.000 - 93.933: 693 ########
93.933 - 172.314: 938 ##########
172.314 - 315.408: 856 #########
315.408 - 576.646: 95 #
576.646 - 6415.830: 888 ##########
6415.830 - 11713.809: 1024 ###########
11713.809 - 21386.000: 4816 #####################################################
So traversing such trees can take some significant time that can
easily be avoided.
Ran the following sysbench tests, 5 times each, for sequential and
random writes, and got the following results:
sysbench --test=fileio --file-num=1 --file-total-size=2G \
--file-test-mode=seqwr --num-threads=16 --file-block-size=65536 \
--max-requests=0 --max-time=60 --file-io-mode=sync
sysbench --test=fileio --file-num=1 --file-total-size=2G \
--file-test-mode=rndwr --num-threads=16 --file-block-size=65536 \
--max-requests=0 --max-time=60 --file-io-mode=sync
Before this change:
sequential writes: 69.28Mb/sec (average of 5 runs)
random writes: 4.14Mb/sec (average of 5 runs)
After this change:
sequential writes: 69.91Mb/sec (average of 5 runs)
random writes: 5.69Mb/sec (average of 5 runs)
Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
When we didn't find a matching extent state, we inserted a new one
but didn't cache it in the **cached_state parameter, which makes a
subsequent call do a tree lookup to get it.
Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
Before this change, adding an extent map to the extent map tree of an
inode required 2 tree nevigations:
1) doing a tree navigation to search for an existing extent map starting
at the same offset or an extent map that overlaps the extent map we
want to insert;
2) Another tree navigation to add the extent map to the tree (if the
former tree search didn't found anything).
This change just merges these 2 steps into a single one.
While running first few btrfs xfstests I had noticed these trees easily
had a few hundred elements, and then with the following sysbench test it
reached over 1100 elements very often.
Test:
sysbench --test=fileio --file-num=32 --file-total-size=10G \
--file-test-mode=seqwr --num-threads=512 --file-block-size=8192 \
--max-requests=1000000 --file-io-mode=sync [prepare|run]
(fs created with mkfs.btrfs -l 4096 -f /dev/sdb3 before each sysbench
prepare phase)
Before this patch:
run 1 - 41.894Mb/sec
run 2 - 40.527Mb/sec
run 3 - 40.922Mb/sec
run 4 - 49.433Mb/sec
run 5 - 40.959Mb/sec
average - 42.75Mb/sec
After this patch:
run 1 - 48.036Mb/sec
run 2 - 50.21Mb/sec
run 3 - 50.929Mb/sec
run 4 - 46.881Mb/sec
run 5 - 53.192Mb/sec
average - 49.85Mb/sec
Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
When attempting to move items from our target leaf to its neighbor
leaves (right and left), we only need to free data_size - free_space
bytes from our leaf in order to add the new item (which has size of
data_size bytes). Therefore attempt to move items to the right and
left leaves if they have at least data_size - free_space bytes free,
instead of data_size bytes free.
After 5 runs of the following test, I got a smaller number of btree
node splits overall:
sysbench --test=fileio --file-num=512 --file-total-size=5G \
--file-test-mode=seqwr --num-threads=512 \
--file-block-size=8192 --max-requests=100000 --file-io-mode=sync
Before this change:
* 6171 splits (average of 5 test runs)
* 61.508Mb/sec of throughput (average of 5 test runs)
After this change:
* 6036 splits (average of 5 test runs)
* 63.533Mb/sec of throughput (average of 5 test runs)
An ideal test would not just have multiple threads/processes writing
to a file (insertion of file extent items) but also do other operations
that result in insertion of items with varied sizes, like file/directory
creations, creation of links, symlinks, xattrs, etc.
Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
After an ordered extent completes, don't blindly reset the
inode's ordered tree last accessed ordered extent pointer.
While running the xfstests I noticed that about 29% of the
time the ordered extent to which tree->last pointed was not
the same as our just completed ordered extent. After that I
ran the following sysbench test (after a prepare phase) and
noticed that about 68% of the time tree->last pointed to
a different ordered extent too.
sysbench --test=fileio --file-num=32 --file-total-size=4G \
--file-test-mode=rndwr --num-threads=512 \
--file-block-size=32768 --max-time=60 --max-requests=0 run
Therefore reset tree->last on ordered extent removal only if
it pointed to the ordered extent we're removing from the tree.
Results from 4 runs of the following test before and after
applying this patch:
$ sysbench --test=fileio --file-num=32 --file-total-size=4G \
--file-test-mode=seqwr --num-threads=512 \
--file-block-size=32768 --max-time=60 --file-io-mode=sync prepare
$ sysbench --test=fileio --file-num=32 --file-total-size=4G \
--file-test-mode=seqwr --num-threads=512 \
--file-block-size=32768 --max-time=60 --file-io-mode=sync run
Before this path:
run 1 - 64.049Mb/sec
run 2 - 63.455Mb/sec
run 3 - 64.656Mb/sec
run 4 - 63.833Mb/sec
After this patch:
run 1 - 66.149Mb/sec
run 2 - 68.459Mb/sec
run 3 - 66.338Mb/sec
run 4 - 66.176Mb/sec
With random writes (--file-test-mode=rndwr) I had huge fluctuations
on the results (+- 35% easily).
Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
Filipe noticed that we were leaking the features attribute group
after umount. His fix of just calling sysfs_remove_group() wasn't enough
since that removes just the supported features and not the unsupported
features.
This patch changes the unknown feature handling to add them individually
so we can skip the kmalloc and uses the same iteration to tear them down
later.
We also fix the error handling during mount so that we catch the
failing creation of the per-super kobject, and handle proper teardown
of a half-setup sysfs context.
Tested properly with kmemleak enabled this time.
Reported-by: Filipe David Borba Manana <fdmanana@gmail.com>
Signed-off-by: Jeff Mahoney <jeffm@suse.com>
Tested-by: Filipe David Borba Manana <fdmanana@gmail.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
This patch fixes the following warnings:
fs/btrfs/extent-tree.c:6201:12: sparse: symbol 'get_raid_name' was not declared. Should it be static?
fs/btrfs/extent-tree.c:8430:9: error: format not a string literal and no format arguments [-Werror=format-security] get_raid_name(index));
Signed-off-by: Jeff Mahoney <jeffm@suse.com>
Reviewed-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
The inode eviction can be very slow, because during eviction we
tell the VFS to truncate all of the inode's pages. This results
in calls to btrfs_invalidatepage() which in turn does calls to
lock_extent_bits() and clear_extent_bit(). These calls result in
too many merges and splits of extent_state structures, which
consume a lot of time and cpu when the inode has many pages. In
some scenarios I have experienced umount times higher than 15
minutes, even when there's no pending IO (after a btrfs fs sync).
A quick way to reproduce this issue:
$ mkfs.btrfs -f /dev/sdb3
$ mount /dev/sdb3 /mnt/btrfs
$ cd /mnt/btrfs
$ sysbench --test=fileio --file-num=128 --file-total-size=16G \
--file-test-mode=seqwr --num-threads=128 \
--file-block-size=16384 --max-time=60 --max-requests=0 run
$ time btrfs fi sync .
FSSync '.'
real 0m25.457s
user 0m0.000s
sys 0m0.092s
$ cd ..
$ time umount /mnt/btrfs
real 1m38.234s
user 0m0.000s
sys 1m25.760s
The same test on ext4 runs much faster:
$ mkfs.ext4 /dev/sdb3
$ mount /dev/sdb3 /mnt/ext4
$ cd /mnt/ext4
$ sysbench --test=fileio --file-num=128 --file-total-size=16G \
--file-test-mode=seqwr --num-threads=128 \
--file-block-size=16384 --max-time=60 --max-requests=0 run
$ sync
$ cd ..
$ time umount /mnt/ext4
real 0m3.626s
user 0m0.004s
sys 0m3.012s
After this patch, the unmount (inode evictions) is much faster:
$ mkfs.btrfs -f /dev/sdb3
$ mount /dev/sdb3 /mnt/btrfs
$ cd /mnt/btrfs
$ sysbench --test=fileio --file-num=128 --file-total-size=16G \
--file-test-mode=seqwr --num-threads=128 \
--file-block-size=16384 --max-time=60 --max-requests=0 run
$ time btrfs fi sync .
FSSync '.'
real 0m26.774s
user 0m0.000s
sys 0m0.084s
$ cd ..
$ time umount /mnt/btrfs
real 0m1.811s
user 0m0.000s
sys 0m1.564s
Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
We hit a forever loop when doing balance relocation,the reason
is that we firstly reserve 4M(node size is 16k).and within transaction
we will try to add extra reservation for snapshot roots,this will
return -EAGAIN if there has been a thread flushing space to reserve
space.We will do this again and again with filesystem becoming nearly
full.
If the above '-EAGAIN' case happens, we try to refill reservation more
outsize of transaction, and this will return eariler in enospc case,however,
this dosen't really hurt because it makes no sense doing balance relocation
with the filesystem nearly full.
Miao Xie helped a lot to track this issue, thanks.
Signed-off-by: Wang Shilong <wangsl.fnst@cn.fujitsu.com>
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
If the ordered extent's last byte was 1 less than our region's
start byte, we would unnecessarily wait for the completion of
that ordered extent, because it doesn't intersect our target
range.
Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
