The way we handle delalloc metadata reservations has gotten
progressively more complicated over the years. There is so much cruft
and weirdness around keeping the reserved count and outstanding counters
consistent and handling the error cases that it's impossible to
understand.
Fix this by making the delalloc block rsv per-inode. This way we can
calculate the actual size of the outstanding metadata reservations every
time we make a change, and then reserve the delta based on that amount.
This greatly simplifies the code everywhere, and makes the error
handling in btrfs_delalloc_reserve_metadata far less terrifying.
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Right now we do a lot of weird hoops around outstanding_extents in order
to keep the extent count consistent. This is because we logically
transfer the outstanding_extent count from the initial reservation
through the set_delalloc_bits. This makes it pretty difficult to get a
handle on how and when we need to mess with outstanding_extents.
Fix this by revamping the rules of how we deal with outstanding_extents.
Now instead everybody that is holding on to a delalloc extent is
required to increase the outstanding extents count for itself. This
means we'll have something like this
btrfs_delalloc_reserve_metadata - outstanding_extents = 1
btrfs_set_extent_delalloc - outstanding_extents = 2
btrfs_release_delalloc_extents - outstanding_extents = 1
for an initial file write. Now take the append write where we extend an
existing delalloc range but still under the maximum extent size
btrfs_delalloc_reserve_metadata - outstanding_extents = 2
btrfs_set_extent_delalloc
btrfs_set_bit_hook - outstanding_extents = 3
btrfs_merge_extent_hook - outstanding_extents = 2
btrfs_delalloc_release_extents - outstanding_extnets = 1
In order to make the ordered extent transition we of course must now
make ordered extents carry their own outstanding_extent reservation, so
for cow_file_range we end up with
btrfs_add_ordered_extent - outstanding_extents = 2
clear_extent_bit - outstanding_extents = 1
btrfs_remove_ordered_extent - outstanding_extents = 0
This makes all manipulations of outstanding_extents much more explicit.
Every successful call to btrfs_delalloc_reserve_metadata _must_ now be
combined with btrfs_release_delalloc_extents, even in the error case, as
that is the only function that actually modifies the
outstanding_extents counter.
The drawback to this is now we are much more likely to have transient
cases where outstanding_extents is much larger than it actually should
be. This could happen before as we manipulated the delalloc bits, but
now it happens basically at every write. This may put more pressure on
the ENOSPC flushing code, but I think making this code simpler is worth
the cost. I have another change coming to mitigate this side-effect
somewhat.
I also added trace points for the counter manipulation. These were used
by a bpf script I wrote to help track down leak issues.
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Preliminary support for setting compression level for zlib, the
following works:
$ mount -o compess=zlib # default
$ mount -o compess=zlib0 # same
$ mount -o compess=zlib9 # level 9, slower sync, less data
$ mount -o compess=zlib1 # level 1, faster sync, more data
$ mount -o remount,compress=zlib3 # level set by remount
The compress-force works the same as compress'. The level is visible in
the same format in /proc/mounts. Level set via file property does not
work yet.
Required patch: "btrfs: prepare for extensions in compression options"
Signed-off-by: David Sterba <dsterba@suse.com>
Currently btrfs' code uses a mix of opencoded sizes and defines from sizes.h.
Let's unifiy the code base to always use the symbolic constants. No functional
changes
Signed-off-by: Nikolay Borisov <nborisov@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
We were having corruption issues that were tied back to problems with
the extent tree. In order to track them down I built this tool to try
and find the culprit, which was pretty successful. If you compile with
this tool on it will live verify every ref update that the fs makes and
make sure it is consistent and valid. I've run this through with
xfstests and haven't gotten any false positives. Thanks,
Signed-off-by: Josef Bacik <jbacik@fb.com>
Reviewed-by: David Sterba <dsterba@suse.com>
[ update error messages, add fixup from Dan Carpenter to handle errors
of read_tree_block ]
Signed-off-by: David Sterba <dsterba@suse.com>
We need the actual root for the ref verifier tool to work, so change
these functions to pass the root around instead. This will be used in
a subsequent patch.
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: David Sterba <dsterba@suse.com>
This adds the infrastructure for turning ref verify on and off for a
mount, to be used by a later patch.
Signed-off-by: Josef Bacik <jbacik@fb.com>
Reviewed-by: David Sterba <dsterba@suse.com>
[ enhnance btrfs_print_mod_info to print if ref-verify is compiled in ]
Signed-off-by: David Sterba <dsterba@suse.com>
Now that we have the combo of flushing twice, which can make sure IO
have started since the second flush will wait for page lock which
won't be unlocked unless setting page writeback and queuing ordered
extents, we don't need %async_submit_draining, %async_delalloc_pages
and %nr_async_submits to tell whether the IO has actually started.
Moreover, all the flushers in use are followed by functions that wait
for ordered extents to complete, so %nr_async_submits, which tracks
whether bio's async submit has made progress, doesn't really make
sense.
However, %async_delalloc_pages is still required by shrink_delalloc()
as that function doesn't flush twice in the normal case (just issues a
writeback with WB_REASON_FS_FREE_SPACE).
Signed-off-by: Liu Bo <bo.li.liu@oracle.com>
Signed-off-by: David Sterba <dsterba@suse.com>
This was intended to congest higher layers to not send bios, but as
1) the congested bit has been taken by writeback
Async bios come from buffered writes and DIO writes.
For DIO writes, we want to submit them ASAP, while for buffered writes,
writeback uses balance_dirty_pages() to throttle how much dirty pages we
can have.
2) and no one is waiting for %nr_async_bios down to zero,
Historically, it was introduced along with changes which let
checksumming workload spread accross different cpus. And at that time,
pdflush was used instead of per-bdi flushing, perhaps pdflush did not
have the necessary information for writeback to do throttling.
We can safely remove them now.
Signed-off-by: Liu Bo <bo.li.liu@oracle.com>
[ additional explanation from mails, removed unused variable 'limit' ]
Signed-off-by: David Sterba <dsterba@suse.com>
btrfs_changed_cb_t represents the signature of the callback being passed
to btrfs_compare_trees. Currently there is only one such callback,
namely changed_cb in send.c. This function doesn't really uses the first
2 parameters, i.e. the roots. Since there are not other functions
implementing the btrfs_changed_cb_t let's remove the unused parameters
from the prototype and implementation.
Signed-off-by: Nikolay Borisov <nborisov@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Reviewed-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Pull btrfs fixes from David Sterba:
"Two more fixes for bugs introduced in 4.13.
The sector_t problem with 32bit architecture and !LBDAF config seems
serious but the number of affected deployments is hopefully low.
The clashing status bits could lead to a confusing in-memory state of
the whole-filesystem operations if used with the quota override sysfs
knob"
* 'for-4.14-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux:
Btrfs: fix overlap of fs_info::flags values
btrfs: avoid overflow when sector_t is 32 bit
Because the values of BTRFS_FS_EXCL_OP and BTRFS_FS_QUOTA_OVERRIDE overlap,
we should change the value.
First, BTRFS_FS_EXCL_OP was set to 14.
commit 171938e528 ("btrfs: track exclusive filesystem operation in flags")
Next, the value of BTRFS_FS_QUOTA_OVERRIDE was set to 14.
commit f29efe2921 ("btrfs: add quota override flag to enable quota override for CAP_SYS_RESOURCE")
As a result, the value 14 overlapped, by accident.
This problem is solved by defining the value of BTRFS_FS_EXCL_OP as 16,
the flags are internal.
Fixes: f29efe2921 ("btrfs: add quota override flag to enable quota override for CAP_SYS_RESOURCE")
CC: stable@vger.kernel.org # 4.13+
Signed-off-by: Tsutomu Itoh <t-itoh@jp.fujitsu.com>
Reviewed-by: David Sterba <dsterba@suse.com>
[ minimize the change, update only BTRFS_FS_EXCL_OP ]
Signed-off-by: David Sterba <dsterba@suse.com>
Pull btrfs fixes from David Sterba:
"We've collected a bunch of isolated fixes, for crashes, user-visible
behaviour or missing bits from other subsystem cleanups from the past.
The overall number is not small but I was not able to make it
significantly smaller. Most of the patches are supposed to go to
stable"
* 'for-4.14-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux:
btrfs: log csums for all modified extents
Btrfs: fix unexpected result when dio reading corrupted blocks
btrfs: Report error on removing qgroup if del_qgroup_item fails
Btrfs: skip checksum when reading compressed data if some IO have failed
Btrfs: fix kernel oops while reading compressed data
Btrfs: use btrfs_op instead of bio_op in __btrfs_map_block
Btrfs: do not backup tree roots when fsync
btrfs: remove BTRFS_FS_QUOTA_DISABLING flag
btrfs: propagate error to btrfs_cmp_data_prepare caller
btrfs: prevent to set invalid default subvolid
Btrfs: send: fix error number for unknown inode types
btrfs: fix NULL pointer dereference from free_reloc_roots()
btrfs: finish ordered extent cleaning if no progress is found
btrfs: clear ordered flag on cleaning up ordered extents
Btrfs: fix incorrect {node,sector}size endianness from BTRFS_IOC_FS_INFO
Btrfs: do not reset bio->bi_ops while writing bio
Btrfs: use the new helper wbc_to_write_flags
Currently, "btrfs quota enable" would fail after "btrfs quota disable" on
the first time with syslog output "qgroup_rescan_init failed with -22", but
it would succeed on the second time.
When "quota disable" is called, BTRFS_FS_QUOTA_DISABLING flag bit will be
set in fs_info->flags in btrfs_quota_disable(), but it will not be droppd
in btrfs_run_qgroups() (which is called in btrfs_commit_transaction())
because quota_root has already been freed. If "quota enable" is called
after that, both BTRFS_FS_QUOTA_DISABLING and BTRFS_FS_QUOTA_ENABLED flag
would be dropped in the btrfs_run_qgroups() since quota_root is not NULL.
This leads to the failure of "quota enable" on the first time.
BTRFS_FS_QUOTA_DISABLING flag is not used outside of "quota disable"
context and is equivalent to whether quota_root is NULL or not.
btrfs_run_qgroups() checks whether quota_root is NULL or not in the first
place.
So, let's remove BTRFS_FS_QUOTA_DISABLING flag.
Signed-off-by: Tomohiro Misono <misono.tomohiro@jp.fujitsu.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Pull zstd support from Chris Mason:
"Nick Terrell's patch series to add zstd support to the kernel has been
floating around for a while. After talking with Dave Sterba, Herbert
and Phillip, we decided to send the whole thing in as one pull
request.
zstd is a big win in speed over zlib and in compression ratio over
lzo, and the compression team here at FB has gotten great results
using it in production. Nick will continue to update the kernel side
with new improvements from the open source zstd userland code.
Nick has a number of benchmarks for the main zstd code in his lib/zstd
commit:
I ran the benchmarks on a Ubuntu 14.04 VM with 2 cores and 4 GiB
of RAM. The VM is running on a MacBook Pro with a 3.1 GHz Intel
Core i7 processor, 16 GB of RAM, and a SSD. I benchmarked using
`silesia.tar` [3], which is 211,988,480 B large. Run the following
commands for the benchmark:
sudo modprobe zstd_compress_test
sudo mknod zstd_compress_test c 245 0
sudo cp silesia.tar zstd_compress_test
The time is reported by the time of the userland `cp`.
The MB/s is computed with
1,536,217,008 B / time(buffer size, hash)
which includes the time to copy from userland.
The Adjusted MB/s is computed with
1,536,217,088 B / (time(buffer size, hash) - time(buffer size, none)).
The memory reported is the amount of memory the compressor
requests.
| Method | Size (B) | Time (s) | Ratio | MB/s | Adj MB/s | Mem (MB) |
|----------|----------|----------|-------|---------|----------|----------|
| none | 11988480 | 0.100 | 1 | 2119.88 | - | - |
| zstd -1 | 73645762 | 1.044 | 2.878 | 203.05 | 224.56 | 1.23 |
| zstd -3 | 66988878 | 1.761 | 3.165 | 120.38 | 127.63 | 2.47 |
| zstd -5 | 65001259 | 2.563 | 3.261 | 82.71 | 86.07 | 2.86 |
| zstd -10 | 60165346 | 13.242 | 3.523 | 16.01 | 16.13 | 13.22 |
| zstd -15 | 58009756 | 47.601 | 3.654 | 4.45 | 4.46 | 21.61 |
| zstd -19 | 54014593 | 102.835 | 3.925 | 2.06 | 2.06 | 60.15 |
| zlib -1 | 77260026 | 2.895 | 2.744 | 73.23 | 75.85 | 0.27 |
| zlib -3 | 72972206 | 4.116 | 2.905 | 51.50 | 52.79 | 0.27 |
| zlib -6 | 68190360 | 9.633 | 3.109 | 22.01 | 22.24 | 0.27 |
| zlib -9 | 67613382 | 22.554 | 3.135 | 9.40 | 9.44 | 0.27 |
I benchmarked zstd decompression using the same method on the same
machine. The benchmark file is located in the upstream zstd repo
under `contrib/linux-kernel/zstd_decompress_test.c` [4]. The
memory reported is the amount of memory required to decompress
data compressed with the given compression level. If you know the
maximum size of your input, you can reduce the memory usage of
decompression irrespective of the compression level.
| Method | Time (s) | MB/s | Adjusted MB/s | Memory (MB) |
|----------|----------|---------|---------------|-------------|
| none | 0.025 | 8479.54 | - | - |
| zstd -1 | 0.358 | 592.15 | 636.60 | 0.84 |
| zstd -3 | 0.396 | 535.32 | 571.40 | 1.46 |
| zstd -5 | 0.396 | 535.32 | 571.40 | 1.46 |
| zstd -10 | 0.374 | 566.81 | 607.42 | 2.51 |
| zstd -15 | 0.379 | 559.34 | 598.84 | 4.61 |
| zstd -19 | 0.412 | 514.54 | 547.77 | 8.80 |
| zlib -1 | 0.940 | 225.52 | 231.68 | 0.04 |
| zlib -3 | 0.883 | 240.08 | 247.07 | 0.04 |
| zlib -6 | 0.844 | 251.17 | 258.84 | 0.04 |
| zlib -9 | 0.837 | 253.27 | 287.64 | 0.04 |
I ran a long series of tests and benchmarks on the btrfs side and the
gains are very similar to the core benchmarks Nick ran"
* 'zstd-minimal' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs:
squashfs: Add zstd support
btrfs: Add zstd support
lib: Add zstd modules
lib: Add xxhash module
Now that btrfs_get_extent_inline_ref_type() can report if type is a
valid one and all callers can gracefully deal with that, we don't need
to crash here.
Signed-off-by: Liu Bo <bo.li.liu@oracle.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
An invalid value of extent inline ref type may be read from a
malicious image which may force btrfs to crash.
This adds a helper which does sanity check for the ref type, so we can
know if it's sane, return he type, otherwise return an error.
Signed-off-by: Liu Bo <bo.li.liu@oracle.com>
Reviewed-by: David Sterba <dsterba@suse.com>
[ minimal tweak const types, causing warnings due to other cleanup patches ]
Signed-off-by: David Sterba <dsterba@suse.com>
btrfs_make_block_group is always called with chunk_objectid set to
BTRFS_FIRST_CHUNK_TREE_OBJECTID. There's no reason why this behavior will
change anytime soon, so let's remove the argument and decrease the cognitive
load when reading the code path. No functional change
Signed-off-by: Nikolay Borisov <nborisov@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
This patch provides a band aid to improve the 'out of the box'
behaviour of btrfs for disks that are detected as being an ssd. In a
general purpose mixed workload scenario, the current ssd mode causes
overallocation of available raw disk space for data, while leaving
behind increasing amounts of unused fragmented free space. This
situation leads to early ENOSPC problems which are harming user
experience and adoption of btrfs as a general purpose filesystem.
This patch modifies the data extent allocation behaviour of the ssd mode
to make it behave identical to nossd mode. The metadata behaviour and
additional ssd_spread option stay untouched so far.
Recommendations for future development are to reconsider the current
oversimplified nossd / ssd distinction and the broken detection
mechanism based on the rotational attribute in sysfs and provide
experienced users with a more flexible way to choose allocator behaviour
for data and metadata, optimized for certain use cases, while keeping
sane 'out of the box' default settings. The internals of the current
btrfs code have more potential than what currently gets exposed to the
user to choose from.
The SSD story...
In the first year of btrfs development, around early 2008, btrfs
gained a mount option which enables specific functionality for
filesystems on solid state devices. The first occurance of this
functionality is in commit e18e4809, labeled "Add mount -o ssd, which
includes optimizations for seek free storage".
The effect on allocating free space for doing (data) writes is to
'cluster' writes together, writing them out in contiguous space, as
opposed to a 'tetris' way of putting all separate writes into any free
space fragment that fits (which is what the -o nossd behaviour does).
A somewhat simplified explanation of what happens is that, when for
example, the 'cluster' size is set to 2MiB, when we do some writes, the
data allocator will search for a free space block that is 2MiB big, and
put the writes in there. The ssd mode itself might allow a 2MiB cluster
to be composed of multiple free space extents with some existing data in
between, while the additional ssd_spread mount option kills off this
option and requires fully free space.
The idea behind this is (commit 536ac8ae): "The [...] clusters make it
more likely a given IO will completely overwrite the ssd block, so it
doesn't have to do an internal rwm cycle."; ssd block meaning nand erase
block. So, effectively this means applying a "locality based algorithm"
and trying to outsmart the actual ssd.
Since then, various changes have been made to the involved code, but the
basic idea is still present, and gets activated whenever the ssd mount
option is active. This also happens by default, when the rotational flag
as seen at /sys/block/<device>/queue/rotational is set to 0.
However, there's a number of problems with this approach.
First, what the optimization is trying to do is outsmart the ssd by
assuming there is a relation between the physical address space of the
block device as seen by btrfs and the actual physical storage of the
ssd, and then adjusting data placement. However, since the introduction
of the Flash Translation Layer (FTL) which is a part of the internal
controller of an ssd, these attempts are futile. The use of good quality
FTL in consumer ssd products might have been limited in 2008, but this
situation has changed drastically soon after that time. Today, even the
flash memory in your automatic cat feeding machine or your grandma's
wheelchair has a full featured one.
Second, the behaviour as described above results in the filesystem being
filled up with badly fragmented free space extents because of relatively
small pieces of space that are freed up by deletes, but not selected
again as part of a 'cluster'. Since the algorithm prefers allocating a
new chunk over going back to tetris mode, the end result is a filesystem
in which all raw space is allocated, but which is composed of
underutilized chunks with a 'shotgun blast' pattern of fragmented free
space. Usually, the next problematic thing that happens is the
filesystem wanting to allocate new space for metadata, which causes the
filesystem to fail in spectacular ways.
Third, the default mount options you get for an ssd ('ssd' mode enabled,
'discard' not enabled), in combination with spreading out writes over
the full address space and ignoring freed up space leads to worst case
behaviour in providing information to the ssd itself, since it will
never learn that all the free space left behind is actually free. There
are two ways to let an ssd know previously written data does not have to
be preserved, which are sending explicit signals using discard or
fstrim, or by simply overwriting the space with new data. The worst
case behaviour is the btrfs ssd_spread mount option in combination with
not having discard enabled. It has a side effect of minimizing the reuse
of free space previously written in.
Fourth, the rotational flag in /sys/ does not reliably indicate if the
device is a locally attached ssd. For example, iSCSI or NBD displays as
non-rotational, while a loop device on an ssd shows up as rotational.
The combination of the second and third problem effectively means that
despite all the good intentions, the btrfs ssd mode reliably causes the
ssd hardware and the filesystem structures and performance to be choked
to death. The clickbait version of the title of this story would have
been "Btrfs ssd optimizations considered harmful for ssds".
The current nossd 'tetris' mode (even still without discard) allows a
pattern of overwriting much more previously used space, causing many
more implicit discards to happen because of the overwrite information
the ssd gets. The actual location in the physical address space, as seen
from the point of view of btrfs is irrelevant, because the actual writes
to the low level flash are reordered anyway thanks to the FTL.
Changes made in the code
1. Make ssd mode data allocation identical to tetris mode, like nossd.
2. Adjust and clean up filesystem mount messages so that we can easily
identify if a kernel has this patch applied or not, when providing
support to end users. Also, make better use of the *_and_info helpers to
only trigger messages on actual state changes.
Backporting notes
Notes for whoever wants to backport this patch to their 4.9 LTS kernel:
* First apply commit 951e7966 "btrfs: drop the nossd flag when
remounting with -o ssd", or fixup the differences manually.
* The rest of the conflicts are because of the fs_info refactoring. So,
for example, instead of using fs_info, it's root->fs_info in
extent-tree.c
Signed-off-by: Hans van Kranenburg <hans.van.kranenburg@mendix.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Readdir does dir_emit while under the btree lock. dir_emit can trigger
the page fault which means we can deadlock. Fix this by allocating a
buffer on opening a directory and copying the readdir into this buffer
and doing dir_emit from outside of the tree lock.
Thread A
readdir <holding tree lock>
dir_emit
<page fault>
down_read(mmap_sem)
Thread B
mmap write
down_write(mmap_sem)
page_mkwrite
wait_ordered_extents
Process C
finish_ordered_extent
insert_reserved_file_extent
try to lock leaf <hang>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Reviewed-by: David Sterba <dsterba@suse.com>
[ copy the deadlock scenario to changelog ]
Signed-off-by: David Sterba <dsterba@suse.com>
btrfs_new_inode() is the only consumer move it to inode.c,
from ioctl.c.
Signed-off-by: Anand Jain <anand.jain@oracle.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
As we use per-chunk degradable check, the global
num_tolerated_disk_barrier_failures is of no use.
We can now remove it.
Signed-off-by: Qu Wenruo <quwenruo@cn.fujitsu.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
We have reader helpers for most of the on-disk structures that use
an extent_buffer and pointer as offset into the buffer that are
read-only. We should mark them as const and, in turn, allow consumers
of these interfaces to mark the buffers const as well.
No impact on code, but serves as documentation that a buffer is intended
not to be modified.
Signed-off-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
The sectorsize member of btrfs_block_group_cache is unused. So remove it, this
reduces the number of holes in the struct.
With patch:
/* size: 856, cachelines: 14, members: 40 */
/* sum members: 837, holes: 4, sum holes: 19 */
/* bit holes: 1, sum bit holes: 29 bits */
/* last cacheline: 24 bytes */
Without patch:
/* size: 864, cachelines: 14, members: 41 */
/* sum members: 841, holes: 5, sum holes: 23 */
/* bit holes: 1, sum bit holes: 29 bits */
/* last cacheline: 32 bytes */
Signed-off-by: Nikolay Borisov <nborisov@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Add zstd compression and decompression support to BtrFS. zstd at its
fastest level compresses almost as well as zlib, while offering much
faster compression and decompression, approaching lzo speeds.
I benchmarked btrfs with zstd compression against no compression, lzo
compression, and zlib compression. I benchmarked two scenarios. Copying
a set of files to btrfs, and then reading the files. Copying a tarball
to btrfs, extracting it to btrfs, and then reading the extracted files.
After every operation, I call `sync` and include the sync time.
Between every pair of operations I unmount and remount the filesystem
to avoid caching. The benchmark files can be found in the upstream
zstd source repository under
`contrib/linux-kernel/{btrfs-benchmark.sh,btrfs-extract-benchmark.sh}`
[1] [2].
I ran the benchmarks on a Ubuntu 14.04 VM with 2 cores and 4 GiB of RAM.
The VM is running on a MacBook Pro with a 3.1 GHz Intel Core i7 processor,
16 GB of RAM, and a SSD.
The first compression benchmark is copying 10 copies of the unzipped
Silesia corpus [3] into a BtrFS filesystem mounted with
`-o compress-force=Method`. The decompression benchmark times how long
it takes to `tar` all 10 copies into `/dev/null`. The compression ratio is
measured by comparing the output of `df` and `du`. See the benchmark file
[1] for details. I benchmarked multiple zstd compression levels, although
the patch uses zstd level 1.
| Method | Ratio | Compression MB/s | Decompression speed |
|---------|-------|------------------|---------------------|
| None | 0.99 | 504 | 686 |
| lzo | 1.66 | 398 | 442 |
| zlib | 2.58 | 65 | 241 |
| zstd 1 | 2.57 | 260 | 383 |
| zstd 3 | 2.71 | 174 | 408 |
| zstd 6 | 2.87 | 70 | 398 |
| zstd 9 | 2.92 | 43 | 406 |
| zstd 12 | 2.93 | 21 | 408 |
| zstd 15 | 3.01 | 11 | 354 |
The next benchmark first copies `linux-4.11.6.tar` [4] to btrfs. Then it
measures the compression ratio, extracts the tar, and deletes the tar.
Then it measures the compression ratio again, and `tar`s the extracted
files into `/dev/null`. See the benchmark file [2] for details.
| Method | Tar Ratio | Extract Ratio | Copy (s) | Extract (s)| Read (s) |
|--------|-----------|---------------|----------|------------|----------|
| None | 0.97 | 0.78 | 0.981 | 5.501 | 8.807 |
| lzo | 2.06 | 1.38 | 1.631 | 8.458 | 8.585 |
| zlib | 3.40 | 1.86 | 7.750 | 21.544 | 11.744 |
| zstd 1 | 3.57 | 1.85 | 2.579 | 11.479 | 9.389 |
[1] https://github.com/facebook/zstd/blob/dev/contrib/linux-kernel/btrfs-benchmark.sh
[2] https://github.com/facebook/zstd/blob/dev/contrib/linux-kernel/btrfs-extract-benchmark.sh
[3] http://sun.aei.polsl.pl/~sdeor/index.php?page=silesia
[4] https://cdn.kernel.org/pub/linux/kernel/v4.x/linux-4.11.6.tar.xz
zstd source repository: https://github.com/facebook/zstd
Signed-off-by: Nick Terrell <terrelln@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
Pull btrfs updates from David Sterba:
"The core updates improve error handling (mostly related to bios), with
the usual incremental work on the GFP_NOFS (mis)use removal,
refactoring or cleanups. Except the two top patches, all have been in
for-next for an extensive amount of time.
User visible changes:
- statx support
- quota override tunable
- improved compression thresholds
- obsoleted mount option alloc_start
Core updates:
- bio-related updates:
- faster bio cloning
- no allocation failures
- preallocated flush bios
- more kvzalloc use, memalloc_nofs protections, GFP_NOFS updates
- prep work for btree_inode removal
- dir-item validation
- qgoup fixes and updates
- cleanups:
- removed unused struct members, unused code, refactoring
- argument refactoring (fs_info/root, caller -> callee sink)
- SEARCH_TREE ioctl docs"
* 'for-4.13-part1' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux: (115 commits)
btrfs: Remove false alert when fiemap range is smaller than on-disk extent
btrfs: Don't clear SGID when inheriting ACLs
btrfs: fix integer overflow in calc_reclaim_items_nr
btrfs: scrub: fix target device intialization while setting up scrub context
btrfs: qgroup: Fix qgroup reserved space underflow by only freeing reserved ranges
btrfs: qgroup: Introduce extent changeset for qgroup reserve functions
btrfs: qgroup: Fix qgroup reserved space underflow caused by buffered write and quotas being enabled
btrfs: qgroup: Return actually freed bytes for qgroup release or free data
btrfs: qgroup: Cleanup btrfs_qgroup_prepare_account_extents function
btrfs: qgroup: Add quick exit for non-fs extents
Btrfs: rework delayed ref total_bytes_pinned accounting
Btrfs: return old and new total ref mods when adding delayed refs
Btrfs: always account pinned bytes when dropping a tree block ref
Btrfs: update total_bytes_pinned when pinning down extents
Btrfs: make BUG_ON() in add_pinned_bytes() an ASSERT()
Btrfs: make add_pinned_bytes() take an s64 num_bytes instead of u64
btrfs: fix validation of XATTR_ITEM dir items
btrfs: Verify dir_item in iterate_object_props
btrfs: Check name_len before in btrfs_del_root_ref
btrfs: Check name_len before reading btrfs_get_name
...
[BUG]
For the following case, btrfs can underflow qgroup reserved space
at an error path:
(Page size 4K, function name without "btrfs_" prefix)
Task A | Task B
----------------------------------------------------------------------
Buffered_write [0, 2K) |
|- check_data_free_space() |
| |- qgroup_reserve_data() |
| Range aligned to page |
| range [0, 4K) <<< |
| 4K bytes reserved <<< |
|- copy pages to page cache |
| Buffered_write [2K, 4K)
| |- check_data_free_space()
| | |- qgroup_reserved_data()
| | Range alinged to page
| | range [0, 4K)
| | Already reserved by A <<<
| | 0 bytes reserved <<<
| |- delalloc_reserve_metadata()
| | And it *FAILED* (Maybe EQUOTA)
| |- free_reserved_data_space()
|- qgroup_free_data()
Range aligned to page range
[0, 4K)
Freeing 4K
(Special thanks to Chandan for the detailed report and analyse)
[CAUSE]
Above Task B is freeing reserved data range [0, 4K) which is actually
reserved by Task A.
And at writeback time, page dirty by Task A will go through writeback
routine, which will free 4K reserved data space at file extent insert
time, causing the qgroup underflow.
[FIX]
For btrfs_qgroup_free_data(), add @reserved parameter to only free
data ranges reserved by previous btrfs_qgroup_reserve_data().
So in above case, Task B will try to free 0 byte, so no underflow.
Reported-by: Chandan Rajendra <chandan@linux.vnet.ibm.com>
Signed-off-by: Qu Wenruo <quwenruo@cn.fujitsu.com>
Reviewed-by: Chandan Rajendra <chandan@linux.vnet.ibm.com>
Tested-by: Chandan Rajendra <chandan@linux.vnet.ibm.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Introduce a new parameter, struct extent_changeset for
btrfs_qgroup_reserved_data() and its callers.
Such extent_changeset was used in btrfs_qgroup_reserve_data() to record
which range it reserved in current reserve, so it can free it in error
paths.
The reason we need to export it to callers is, at buffered write error
path, without knowing what exactly which range we reserved in current
allocation, we can free space which is not reserved by us.
This will lead to qgroup reserved space underflow.
Reviewed-by: Chandan Rajendra <chandan@linux.vnet.ibm.com>
Signed-off-by: Qu Wenruo <quwenruo@cn.fujitsu.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Introduce function btrfs_is_name_len_valid.
The function compares parameter @name_len with item boundary then
returns true if name_len is valid.
Signed-off-by: Su Yue <suy.fnst@cn.fujitsu.com>
Reviewed-by: David Sterba <dsterba@suse.com>
[ s/btrfs_leaf_data/BTRFS_LEAF_DATA_OFFSET/ ]
Signed-off-by: David Sterba <dsterba@suse.com>
We got an internal report about a file system not wanting to mount
following 99e3ecfcb9 ("Btrfs: add more validation checks for
superblock").
BTRFS error (device sdb1): super_total_bytes 1000203816960 mismatch with
fs_devices total_rw_bytes 1000203820544
Subtracting the numbers we get a difference of less than a 4kb. Upon
closer inspection it became apparent that mkfs actually rounds down the
size of the device to a multiple of sector size. However, the same
cannot be said for various functions which modify the total size and are
called from btrfs_balance as well as when adding a new device. So this
patch ensures that values being saved into on-disk data structures are
always rounded down to a multiple of sectorsize.
Signed-off-by: Nikolay Borisov <nborisov@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
The device->total_bytes member needs to always be rounded down to sectorsize
so that it corresponds to the value of super->total_bytes. However, there are
multiple places where the setter is fed a value which is not rounded which
can cause a fs to be unmountable due to the check introduced in
99e3ecfcb9 ("Btrfs: add more validation checks for superblock"). This patch
implements the getter/setter manually so that in a later patch I can add
necessary code to catch offenders.
Signed-off-by: Nikolay Borisov <nborisov@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
The mount option alloc_start was used in the past for debugging and
stressing the chunk allocator. Not meant to be used by users, so we're
not breaking anybody's setup.
There was some added complexity handling changes of the value and when
it was not same as default. Such code has likely been untested and I
think it's better to remove it.
This patch kills all use of alloc_start, and by doing that also fixes
a bug when alloc_size is set, potentially called from statfs:
in btrfs_calc_avail_data_space, traversing the list in RCU, the RCU
protection is temporarily dropped so btrfs_account_dev_extents_size can
be called and then RCU is locked again! Doing that inside
list_for_each_entry_rcu is just asking for trouble, but unlikely to be
observed in practice.
Signed-off-by: David Sterba <dsterba@suse.com>
We can keep the state among the other fs_info flags, there's no reason
why fs_frozen would need to be separate.
Reviewed-by: Nikolay Borisov <nborisov@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Observing the number of slab objects of btrfs_transaction, there's just
one active on an almost quiescent filesystem, and the number of objects
goes to about ten when sync is in progress. Then the nubmer goes down to
1. This matches the expectations of the transaction lifetime.
For such use the separate slab cache is not justified, as we do not
reuse objects frequently. For the shortlived transaction, the generic
slab (size 512) should be ok. We can optimistically expect that the 512
slabs are not all used (fragmentation) and there are free slots to take
when we do the allocation, compared to potentially allocating a whole new
page for the separate slab.
We'll lose the stats about the object use, which could be added later if
we really need them.
Signed-off-by: David Sterba <dsterba@suse.com>
__BTRFS_LAF_DATA_SIZE is used only by BTRFS_LEAF_DATA_SIZE. Make the
latter subsume the former.
Signed-off-by: Nikolay Borisov <nborisov@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Commit 5f39d397df ("Btrfs: Create extent_buffer interface
for large blocksizes") refactored btrfs_leaf_data function to take
extent_buffer rather than struct btrfs_leaf. However, as it turns out the
parameter being passed is never used. Furthermore this function no longer
returns the leaf data but rather the offset to it. So rename the function
to BTRFS_LEAF_DATA_OFFSET to make it consistent with other BTRFS_LEAF_*
helpers and turn it into a macro.
Signed-off-by: Nikolay Borisov <nborisov@suse.com>
[ removed () from the macro ]
Signed-off-by: David Sterba <dsterba@suse.com>
There are two places where we don't already know what kind of alloc
profile we need before calling btrfs_get_alloc_profile, but we need
access to a root everywhere we call it.
This patch adds helpers for btrfs_{data,metadata,system}_alloc_profile()
and relegates btrfs_system_alloc_profile to a static for use in those
two cases. The next patch will eliminate one of those.
Signed-off-by: Jeff Mahoney <jeffm@suse.com>
Reviewed-by: Liu Bo <bo.li.liu@oracle.com>
Signed-off-by: David Sterba <dsterba@suse.com>
For extent_io tree's we have carried the address_mapping of the inode
around in the io tree in order to pull the inode back out for calling
into various tree ops hooks. This works fine when everything that has
an extent_io_tree has an inode. But we are going to remove the
btree_inode, so we need to change this. Instead just have a generic
void * for private data that we can initialize with, and have all the
tree ops use that instead. This had a lot of cascading changes but
should be relatively straightforward.
Signed-off-by: Josef Bacik <jbacik@fb.com>
Reviewed-by: Chandan Rajendra <chandan@linux.vnet.ibm.com>
Reviewed-by: David Sterba <dsterba@suse.com>
[ minor reordering of the callback prototypes ]
Signed-off-by: David Sterba <dsterba@suse.com>
This patch introduces the quota override flag to btrfs_fs_info, and a
change to quota limit checking code to temporarily allow for quota to be
overridden for processes with CAP_SYS_RESOURCE.
It's useful for administrative programs, such as log rotation, that may
need to temporarily use more disk space in order to free up a greater
amount of overall disk space without yielding more disk space to the
rest of userland.
Eventually, we may want to add the idea of an operator-specific quota,
operator reserved space, or something else to allow for administrative
override, but this is perhaps the simplest solution.
Signed-off-by: Sargun Dhillon <sargun@sargun.me>
Reviewed-by: David Sterba <dsterba@suse.com>
[ minor changelog edits ]
Signed-off-by: David Sterba <dsterba@suse.com>
The ->free_chunk_space variable is used to track the unallocated space
and access to it is protected by a spinlock, which is not used for
anything else. Make the code a bit self-explanatory by switching the
variable to an atomic64_t type and kill the spinlock.
Signed-off-by: Nikolay Borisov <nborisov@suse.com>
[ not a performance critical code, use of atomic type is ok ]
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
-----BEGIN PGP SIGNATURE-----
iQEcBAABAgAGBQJZPdbLAAoJEHm+PkMAQRiGx4wH/1nCjfnl6fE8oJ24/1gEAOUh
biFdqJkYZmlLYHVtYfLm4Ueg4adJdg0wx6qM/4RaAzmQVvLfDV34bc1qBf1+P95G
kVF+osWyXrZo5cTwkwapHW/KNu4VJwAx2D1wrlxKDVG5AOrULH1pYOYGOpApEkZU
4N+q5+M0ce0GJpqtUZX+UnI33ygjdDbBxXoFKsr24B7eA0ouGbAJ7dC88WcaETL+
2/7tT01SvDMo0jBSV0WIqlgXwZ5gp3yPGnklC3F4159Yze6VFrzHMKS/UpPF8o8E
W9EbuzwxsKyXUifX2GY348L1f+47glen/1sedbuKnFhP6E9aqUQQJXvEO7ueQl4=
=m2Gx
-----END PGP SIGNATURE-----
Merge tag 'v4.12-rc5' into for-4.13/block
We've already got a few conflicts and upcoming work depends on some of the
changes that have gone into mainline as regression fixes for this series.
Pull in 4.12-rc5 to resolve these conflicts and make it easier on down stream
trees to continue working on 4.13 changes.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Pull btrfs fixes from Chris Mason:
"Some fixes that Dave Sterba collected.
We've been hitting an early enospc problem on production machines that
Omar tracked down to an old int->u64 mistake. I waited a bit on this
pull to make sure it was really the problem from production, but it's
on ~2100 hosts now and I think we're good.
Omar also noticed a commit in the queue would make new early ENOSPC
problems. I pulled that out for now, which is why the top three
commits are younger than the rest.
Otherwise these are all fixes, some explaining very old bugs that
we've been poking at for a while"
* 'for-linus-4.12' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs:
Btrfs: fix delalloc accounting leak caused by u32 overflow
Btrfs: clear EXTENT_DEFRAG bits in finish_ordered_io
btrfs: tree-log.c: Wrong printk information about namelen
btrfs: fix race with relocation recovery and fs_root setup
btrfs: fix memory leak in update_space_info failure path
btrfs: use correct types for page indices in btrfs_page_exists_in_range
btrfs: fix incorrect error return ret being passed to mapping_set_error
btrfs: Make flush bios explicitely sync
btrfs: fiemap: Cache and merge fiemap extent before submit it to user
btrfs_calc_trans_metadata_size() does an unsigned 32-bit multiplication,
which can overflow if num_items >= 4 GB / (nodesize * BTRFS_MAX_LEVEL * 2).
For a nodesize of 16kB, this overflow happens at 16k items. Usually,
num_items is a small constant passed to btrfs_start_transaction(), but
we also use btrfs_calc_trans_metadata_size() for metadata reservations
for extent items in btrfs_delalloc_{reserve,release}_metadata().
In drop_outstanding_extents(), num_items is calculated as
inode->reserved_extents - inode->outstanding_extents. The difference
between these two counters is usually small, but if many delalloc
extents are reserved and then the outstanding extents are merged in
btrfs_merge_extent_hook(), the difference can become large enough to
overflow in btrfs_calc_trans_metadata_size().
The overflow manifests itself as a leak of a multiple of 4 GB in
delalloc_block_rsv and the metadata bytes_may_use counter. This in turn
can cause early ENOSPC errors. Additionally, these WARN_ONs in
extent-tree.c will be hit when unmounting:
WARN_ON(fs_info->delalloc_block_rsv.size > 0);
WARN_ON(fs_info->delalloc_block_rsv.reserved > 0);
WARN_ON(space_info->bytes_pinned > 0 ||
space_info->bytes_reserved > 0 ||
space_info->bytes_may_use > 0);
Fix it by casting nodesize to a u64 so that
btrfs_calc_trans_metadata_size() does a full 64-bit multiplication.
While we're here, do the same in btrfs_calc_trunc_metadata_size(); this
can't overflow with any existing uses, but it's better to be safe here
than have another hard-to-debug problem later on.
Cc: stable@vger.kernel.org
Signed-off-by: Omar Sandoval <osandov@fb.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
Replace bi_error with a new bi_status to allow for a clear conversion.
Note that device mapper overloaded bi_error with a private value, which
we'll have to keep arround at least for now and thus propagate to a
proper blk_status_t value.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
Pull btrfs updates from Chris Mason:
"This has fixes and cleanups Dave Sterba collected for the merge
window.
The biggest functional fixes are between btrfs raid5/6 and scrub, and
raid5/6 and device replacement. Some of our pending qgroup fixes are
included as well while I bash on the rest in testing.
We also have the usual set of cleanups, including one that makes
__btrfs_map_block() much more maintainable, and conversions from
atomic_t to refcount_t"
* 'for-linus-4.12' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs: (71 commits)
btrfs: fix the gfp_mask for the reada_zones radix tree
Btrfs: fix reported number of inode blocks
Btrfs: send, fix file hole not being preserved due to inline extent
Btrfs: fix extent map leak during fallocate error path
Btrfs: fix incorrect space accounting after failure to insert inline extent
Btrfs: fix invalid attempt to free reserved space on failure to cow range
btrfs: Handle delalloc error correctly to avoid ordered extent hang
btrfs: Fix metadata underflow caused by btrfs_reloc_clone_csum error
btrfs: check if the device is flush capable
btrfs: delete unused member nobarriers
btrfs: scrub: Fix RAID56 recovery race condition
btrfs: scrub: Introduce full stripe lock for RAID56
btrfs: Use ktime_get_real_ts for root ctime
Btrfs: handle only applicable errors returned by btrfs_get_extent
btrfs: qgroup: Fix qgroup corruption caused by inode_cache mount option
btrfs: use q which is already obtained from bdev_get_queue
Btrfs: switch to div64_u64 if with a u64 divisor
Btrfs: update scrub_parity to use u64 stripe_len
Btrfs: enable repair during read for raid56 profile
btrfs: use clear_page where appropriate
...
