We have two variants of lock/unlock extent, one set that takes a cached
state, another that does not. This is slightly annoying, and generally
speaking there are only a few places where we don't have a cached state.
Simplify this by making lock_extent/unlock_extent the only variant and
make it take a cached state, then convert all the callers appropriately.
Signed-off-by: Josef Bacik <josef@toxicpanda.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
This is only used in the case that we are clearing EXTENT_LOCKED, so
infer this value from the bits passed in instead of taking it as an
argument.
Signed-off-by: Josef Bacik <josef@toxicpanda.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
The current fiemap implementation does not scale very well with the number
of extents a file has. This is both because the main algorithm to find out
the extents has a high algorithmic complexity and because for each extent
we have to check if it's shared. This second part, checking if an extent
is shared, is significantly improved by the two previous patches in this
patchset, while the first part is improved by this specific patch. Every
now and then we get reports from users mentioning fiemap is too slow or
even unusable for files with a very large number of extents, such as the
two recent reports referred to by the Link tags at the bottom of this
change log.
To understand why the part of finding which extents a file has is very
inefficient, consider the example of doing a full ranged fiemap against
a file that has over 100K extents (normal for example for a file with
more than 10G of data and using compression, which limits the extent size
to 128K). When we enter fiemap at extent_fiemap(), the following happens:
1) Before entering the main loop, we call get_extent_skip_holes() to get
the first extent map. This leads us to btrfs_get_extent_fiemap(), which
in turn calls btrfs_get_extent(), to find the first extent map that
covers the file range [0, LLONG_MAX).
btrfs_get_extent() will first search the inode's extent map tree, to
see if we have an extent map there that covers the range. If it does
not find one, then it will search the inode's subvolume b+tree for a
fitting file extent item. After finding the file extent item, it will
allocate an extent map, fill it in with information extracted from the
file extent item, and add it to the inode's extent map tree (which
requires a search for insertion in the tree).
2) Then we enter the main loop at extent_fiemap(), emit the details of
the extent, and call again get_extent_skip_holes(), with a start
offset matching the end of the extent map we previously processed.
We end up at btrfs_get_extent() again, will search the extent map tree
and then search the subvolume b+tree for a file extent item if we could
not find an extent map in the extent tree. We allocate an extent map,
fill it in with the details in the file extent item, and then insert
it into the extent map tree (yet another search in this tree).
3) The second step is repeated over and over, until we have processed the
whole file range. Each iteration ends at btrfs_get_extent(), which
does a red black tree search on the extent map tree, then searches the
subvolume b+tree, allocates an extent map and then does another search
in the extent map tree in order to insert the extent map.
In the best scenario we have all the extent maps already in the extent
tree, and so for each extent we do a single search on a red black tree,
so we have a complexity of O(n log n).
In the worst scenario we don't have any extent map already loaded in
the extent map tree, or have very few already there. In this case the
complexity is much higher since we do:
- A red black tree search on the extent map tree, which has O(log n)
complexity, initially very fast since the tree is empty or very
small, but as we end up allocating extent maps and adding them to
the tree when we don't find them there, each subsequent search on
the tree gets slower, since it's getting bigger and bigger after
each iteration.
- A search on the subvolume b+tree, also O(log n) complexity, but it
has items for all inodes in the subvolume, not just items for our
inode. Plus on a filesystem with concurrent operations on other
inodes, we can block doing the search due to lock contention on
b+tree nodes/leaves.
- Allocate an extent map - this can block, and can also fail if we
are under serious memory pressure.
- Do another search on the extent maps red black tree, with the goal
of inserting the extent map we just allocated. Again, after every
iteration this tree is getting bigger by 1 element, so after many
iterations the searches are slower and slower.
- We will not need the allocated extent map anymore, so it's pointless
to add it to the extent map tree. It's just wasting time and memory.
In short we end up searching the extent map tree multiple times, on a
tree that is growing bigger and bigger after each iteration. And
besides that we visit the same leaf of the subvolume b+tree many times,
since a leaf with the default size of 16K can easily have more than 200
file extent items.
This is very inefficient overall. This patch changes the algorithm to
instead iterate over the subvolume b+tree, visiting each leaf only once,
and only searching in the extent map tree for file ranges that have holes
or prealloc extents, in order to figure out if we have delalloc there.
It will never allocate an extent map and add it to the extent map tree.
This is very similar to what was previously done for the lseek's hole and
data seeking features.
Also, the current implementation relying on extent maps for figuring out
which extents we have is not correct. This is because extent maps can be
merged even if they represent different extents - we do this to minimize
memory utilization and keep extent map trees smaller. For example if we
have two extents that are contiguous on disk, once we load the two extent
maps, they get merged into a single one - however if only one of the
extents is shared, we end up reporting both as shared or both as not
shared, which is incorrect.
This reproducer triggers that bug:
$ cat fiemap-bug.sh
#!/bin/bash
DEV=/dev/sdj
MNT=/mnt/sdj
mkfs.btrfs -f $DEV
mount $DEV $MNT
# Create a file with two 256K extents.
# Since there is no other write activity, they will be contiguous,
# and their extent maps merged, despite having two distinct extents.
xfs_io -f -c "pwrite -S 0xab 0 256K" \
-c "fsync" \
-c "pwrite -S 0xcd 256K 256K" \
-c "fsync" \
$MNT/foo
# Now clone only the second extent into another file.
xfs_io -f -c "reflink $MNT/foo 256K 0 256K" $MNT/bar
# Filefrag will report a single 512K extent, and say it's not shared.
echo
filefrag -v $MNT/foo
umount $MNT
Running the reproducer:
$ ./fiemap-bug.sh
wrote 262144/262144 bytes at offset 0
256 KiB, 64 ops; 0.0038 sec (65.479 MiB/sec and 16762.7030 ops/sec)
wrote 262144/262144 bytes at offset 262144
256 KiB, 64 ops; 0.0040 sec (61.125 MiB/sec and 15647.9218 ops/sec)
linked 262144/262144 bytes at offset 0
256 KiB, 1 ops; 0.0002 sec (1.034 GiB/sec and 4237.2881 ops/sec)
Filesystem type is: 9123683e
File size of /mnt/sdj/foo is 524288 (128 blocks of 4096 bytes)
ext: logical_offset: physical_offset: length: expected: flags:
0: 0.. 127: 3328.. 3455: 128: last,eof
/mnt/sdj/foo: 1 extent found
We end up reporting that we have a single 512K that is not shared, however
we have two 256K extents, and the second one is shared. Changing the
reproducer to clone instead the first extent into file 'bar', makes us
report a single 512K extent that is shared, which is algo incorrect since
we have two 256K extents and only the first one is shared.
This patch is part of a larger patchset that is comprised of the following
patches:
btrfs: allow hole and data seeking to be interruptible
btrfs: make hole and data seeking a lot more efficient
btrfs: remove check for impossible block start for an extent map at fiemap
btrfs: remove zero length check when entering fiemap
btrfs: properly flush delalloc when entering fiemap
btrfs: allow fiemap to be interruptible
btrfs: rename btrfs_check_shared() to a more descriptive name
btrfs: speedup checking for extent sharedness during fiemap
btrfs: skip unnecessary extent buffer sharedness checks during fiemap
btrfs: make fiemap more efficient and accurate reporting extent sharedness
The patchset was tested on a machine running a non-debug kernel (Debian's
default config) and compared the tests below on a branch without the
patchset versus the same branch with the whole patchset applied.
The following test for a large compressed file without holes:
$ cat fiemap-perf-test.sh
#!/bin/bash
DEV=/dev/sdi
MNT=/mnt/sdi
mkfs.btrfs -f $DEV
mount -o compress=lzo $DEV $MNT
# 40G gives 327680 128K file extents (due to compression).
xfs_io -f -c "pwrite -S 0xab -b 1M 0 20G" $MNT/foobar
umount $MNT
mount -o compress=lzo $DEV $MNT
start=$(date +%s%N)
filefrag $MNT/foobar
end=$(date +%s%N)
dur=$(( (end - start) / 1000000 ))
echo "fiemap took $dur milliseconds (metadata not cached)"
start=$(date +%s%N)
filefrag $MNT/foobar
end=$(date +%s%N)
dur=$(( (end - start) / 1000000 ))
echo "fiemap took $dur milliseconds (metadata cached)"
umount $MNT
Before patchset:
$ ./fiemap-perf-test.sh
(...)
/mnt/sdi/foobar: 327680 extents found
fiemap took 3597 milliseconds (metadata not cached)
/mnt/sdi/foobar: 327680 extents found
fiemap took 2107 milliseconds (metadata cached)
After patchset:
$ ./fiemap-perf-test.sh
(...)
/mnt/sdi/foobar: 327680 extents found
fiemap took 1214 milliseconds (metadata not cached)
/mnt/sdi/foobar: 327680 extents found
fiemap took 684 milliseconds (metadata cached)
That's a speedup of about 3x for both cases (no metadata cached and all
metadata cached).
The test provided by Pavel (first Link tag at the bottom), which uses
files with a large number of holes, was also used to measure the gains,
and it consists on a small C program and a shell script to invoke it.
The C program is the following:
$ cat pavels-test.c
#include <stdio.h>
#include <unistd.h>
#include <stdlib.h>
#include <fcntl.h>
#include <sys/stat.h>
#include <sys/time.h>
#include <sys/ioctl.h>
#include <linux/fs.h>
#include <linux/fiemap.h>
#define FILE_INTERVAL (1<<13) /* 8Kb */
long long interval(struct timeval t1, struct timeval t2)
{
long long val = 0;
val += (t2.tv_usec - t1.tv_usec);
val += (t2.tv_sec - t1.tv_sec) * 1000 * 1000;
return val;
}
int main(int argc, char **argv)
{
struct fiemap fiemap = {};
struct timeval t1, t2;
char data = 'a';
struct stat st;
int fd, off, file_size = FILE_INTERVAL;
if (argc != 3 && argc != 2) {
printf("usage: %s <path> [size]\n", argv[0]);
return 1;
}
if (argc == 3)
file_size = atoi(argv[2]);
if (file_size < FILE_INTERVAL)
file_size = FILE_INTERVAL;
file_size -= file_size % FILE_INTERVAL;
fd = open(argv[1], O_RDWR | O_CREAT | O_TRUNC, 0644);
if (fd < 0) {
perror("open");
return 1;
}
for (off = 0; off < file_size; off += FILE_INTERVAL) {
if (pwrite(fd, &data, 1, off) != 1) {
perror("pwrite");
close(fd);
return 1;
}
}
if (ftruncate(fd, file_size)) {
perror("ftruncate");
close(fd);
return 1;
}
if (fstat(fd, &st) < 0) {
perror("fstat");
close(fd);
return 1;
}
printf("size: %ld\n", st.st_size);
printf("actual size: %ld\n", st.st_blocks * 512);
fiemap.fm_length = FIEMAP_MAX_OFFSET;
gettimeofday(&t1, NULL);
if (ioctl(fd, FS_IOC_FIEMAP, &fiemap) < 0) {
perror("fiemap");
close(fd);
return 1;
}
gettimeofday(&t2, NULL);
printf("fiemap: fm_mapped_extents = %d\n",
fiemap.fm_mapped_extents);
printf("time = %lld us\n", interval(t1, t2));
close(fd);
return 0;
}
$ gcc -o pavels_test pavels_test.c
And the wrapper shell script:
$ cat fiemap-pavels-test.sh
#!/bin/bash
DEV=/dev/sdi
MNT=/mnt/sdi
mkfs.btrfs -f -O no-holes $DEV
mount $DEV $MNT
echo
echo "*********** 256M ***********"
echo
./pavels-test $MNT/testfile $((1 << 28))
echo
./pavels-test $MNT/testfile $((1 << 28))
echo
echo "*********** 512M ***********"
echo
./pavels-test $MNT/testfile $((1 << 29))
echo
./pavels-test $MNT/testfile $((1 << 29))
echo
echo "*********** 1G ***********"
echo
./pavels-test $MNT/testfile $((1 << 30))
echo
./pavels-test $MNT/testfile $((1 << 30))
umount $MNT
Running his reproducer before applying the patchset:
*********** 256M ***********
size: 268435456
actual size: 134217728
fiemap: fm_mapped_extents = 32768
time = 4003133 us
size: 268435456
actual size: 134217728
fiemap: fm_mapped_extents = 32768
time = 4895330 us
*********** 512M ***********
size: 536870912
actual size: 268435456
fiemap: fm_mapped_extents = 65536
time = 30123675 us
size: 536870912
actual size: 268435456
fiemap: fm_mapped_extents = 65536
time = 33450934 us
*********** 1G ***********
size: 1073741824
actual size: 536870912
fiemap: fm_mapped_extents = 131072
time = 224924074 us
size: 1073741824
actual size: 536870912
fiemap: fm_mapped_extents = 131072
time = 217239242 us
Running it after applying the patchset:
*********** 256M ***********
size: 268435456
actual size: 134217728
fiemap: fm_mapped_extents = 32768
time = 29475 us
size: 268435456
actual size: 134217728
fiemap: fm_mapped_extents = 32768
time = 29307 us
*********** 512M ***********
size: 536870912
actual size: 268435456
fiemap: fm_mapped_extents = 65536
time = 58996 us
size: 536870912
actual size: 268435456
fiemap: fm_mapped_extents = 65536
time = 59115 us
*********** 1G ***********
size: 1073741824
actual size: 536870912
fiemap: fm_mapped_extents = 116251
time = 124141 us
size: 1073741824
actual size: 536870912
fiemap: fm_mapped_extents = 131072
time = 119387 us
The speedup is massive, both on the first fiemap call and on the second
one as well, as his test creates files with many holes and small extents
(every extent follows a hole and precedes another hole).
For the 256M file we go from 4 seconds down to 29 milliseconds in the
first run, and then from 4.9 seconds down to 29 milliseconds again in the
second run, a speedup of 138x and 169x, respectively.
For the 512M file we go from 30.1 seconds down to 59 milliseconds in the
first run, and then from 33.5 seconds down to 59 milliseconds again in the
second run, a speedup of 510x and 568x, respectively.
For the 1G file, we go from 225 seconds down to 124 milliseconds in the
first run, and then from 217 seconds down to 119 milliseconds in the
second run, a speedup of 1815x and 1824x, respectively.
Reported-by: Pavel Tikhomirov <ptikhomirov@virtuozzo.com>
Link: https://lore.kernel.org/linux-btrfs/21dd32c6-f1f9-f44a-466a-e18fdc6788a7@virtuozzo.com/
Reported-by: Dominique MARTINET <dominique.martinet@atmark-techno.com>
Link: https://lore.kernel.org/linux-btrfs/Ysace25wh5BbLd5f@atmark-techno.com/
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
The current implementation of hole and data seeking for llseek does not
scale well in regards to the number of extents and the distance between
the start offset and the next hole or extent. This is due to a very high
algorithmic complexity. Often we also get reports of btrfs' hole and data
seeking (llseek) being too slow, such as at 2017's LSFMM (see the Link
tag at the bottom).
In order to better understand it, lets consider the case where the start
offset is 0, we are seeking for a hole and the file size is 16G. Between
file offset 0 and the first hole in the file there are 100K extents - this
is common for large files, specially if we have compression enabled, since
the maximum extent size is limited to 128K. The steps take by the main
loop of the current algorithm are the following:
1) We start by calling btrfs_get_extent_fiemap(), for file offset 0, which
calls btrfs_get_extent(). This will first lookup for an extent map in
the inode's extent map tree (a red black tree). If the extent map is
not loaded in memory, then it will do a lookup for the corresponding
file extent item in the subvolume's b+tree, create an extent map based
on the contents of the file extent item and then add the extent map to
the extent map tree of the inode;
2) The second iteration calls btrfs_get_extent_fiemap() again, this time
with a start offset matching the end offset of the previous extent.
Again, btrfs_get_extent() will first search the extent map tree, and
if it doesn't find an extent map there, it will again search in the
b+tree of the subvolume for a matching file extent item, build an
extent map based on the file extent item, and add the extent map to
to the extent map tree of the inode;
3) This repeats over and over until we find the first hole (when seeking
for holes) or until we find the first extent (when seeking for data).
If there no extent maps loaded in memory for each iteration, then on
each iteration we do 1 extent map tree search, 1 b+tree search, plus
1 more extent map tree traversal to insert an extent map - plus we
allocate memory for the extent map.
On each iteration we are growing the size of the extent map tree,
making each future search slower, and also visiting the same b+tree
leaves over and over again - taking into account with the default leaf
size of 16K we can fit more than 200 file extent items in a leaf - so
we can visit the same b+tree leaf 200+ times, on each visit walking
down a path from the root to the leaf.
So it's easy to see that what we have now doesn't scale well. Also, it
loads an extent map for every file extent item into memory, which is not
efficient - we should add extents maps only when doing IO (writing or
reading file data).
This change implements a new algorithm which scales much better, and
works like this:
1) We iterate over the subvolume's b+tree, visiting each leaf that has
file extent items once and only once;
2) For any file extent items found, that don't represent holes or prealloc
extents, it will not search the extent map tree - there's no need at
all for that - an extent map is just an in-memory representation of a
file extent item;
3) When a hole is found, or a prealloc extent, it will check if there's
delalloc for its range. For this it will search for EXTENT_DELALLOC
bits in the inode's io tree and check the extent map tree - this is
for accounting for unflushed delalloc and for flushed delalloc (the
period between running delalloc and ordered extent completion),
respectively. This is similar to what the current implementation does
when it finds a hole or prealloc extent, but without creating extent
maps and adding them to the extent map tree in case they are not
loaded in memory;
4) It never allocates extent maps, or adds extent maps to the inode's
extent map tree. This not only saves memory and time (from the tree
insertions and allocations), but also eliminates the possibility of
-ENOMEM due to allocating too many extent maps.
Part of this new code will also be used later for fiemap (which also
suffers similar scalability problems).
The following test example can be used to quickly measure the efficiency
before and after this patch:
$ cat test-seek-hole.sh
#!/bin/bash
DEV=/dev/sdi
MNT=/mnt/sdi
mkfs.btrfs -f $DEV
mount -o compress=lzo $DEV $MNT
# 16G file -> 131073 compressed extents.
xfs_io -f -c "pwrite -S 0xab -b 1M 0 16G" $MNT/foobar
# Leave a 1M hole at file offset 15G.
xfs_io -c "fpunch 15G 1M" $MNT/foobar
# Unmount and mount again, so that we can test when there's no
# metadata cached in memory.
umount $MNT
mount -o compress=lzo $DEV $MNT
# Test seeking for hole from offset 0 (hole is at offset 15G).
start=$(date +%s%N)
xfs_io -c "seek -h 0" $MNT/foobar
end=$(date +%s%N)
dur=$(( (end - start) / 1000000 ))
echo "Took $dur milliseconds to seek first hole (metadata not cached)"
echo
start=$(date +%s%N)
xfs_io -c "seek -h 0" $MNT/foobar
end=$(date +%s%N)
dur=$(( (end - start) / 1000000 ))
echo "Took $dur milliseconds to seek first hole (metadata cached)"
echo
umount $MNT
Before this change:
$ ./test-seek-hole.sh
(...)
Whence Result
HOLE 16106127360
Took 176 milliseconds to seek first hole (metadata not cached)
Whence Result
HOLE 16106127360
Took 17 milliseconds to seek first hole (metadata cached)
After this change:
$ ./test-seek-hole.sh
(...)
Whence Result
HOLE 16106127360
Took 43 milliseconds to seek first hole (metadata not cached)
Whence Result
HOLE 16106127360
Took 13 milliseconds to seek first hole (metadata cached)
That's about 4x faster when no metadata is cached and about 30% faster
when all metadata is cached.
In practice the differences may often be significantly higher, either due
to a higher number of extents in a file or because the subvolume's b+tree
is much bigger than in this example, where we only have one file.
Link: https://lwn.net/Articles/718805/
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Doing hole or data seeking on a file with a very large number of extents
can take a long time, and we have reports of it being too slow (such as
at LSFMM from 2017, see the Link below). So make it interruptible.
Link: https://lwn.net/Articles/718805/
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Reviewed-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
When logging an inode, if we detect the inode has a reference that
conflicts with some other inode that got renamed, we log that other inode
while holding the log mutex of the current inode. We then find out if
there are other inodes that conflict with the first conflicting inode,
and log them while under the log mutex of the original inode. This is
fine because the recursion can only happen once.
For the upcoming work where we directly log delayed items without flushing
them first to the subvolume tree, this recursion adds a lot of complexity
and it's hard to keep lockdep happy about it.
So collect a list of conflicting inodes and then log the inodes after
unlocking the log mutex of the inode we started with.
Also limit the maximum number of conflict inodes we log to 10, to avoid
spending too much time logging (and maybe allocating too many list
elements too), as typically we don't have more than 1 or 2 conflicting
inodes - if we go over the limit, simply fallback to a transaction commit.
It is possible to have a very long list of conflicting inodes to be
intentionally created by a user if he/she creates a very long succession
of renames like this:
(...)
rename E to F
rename D to E
rename C to D
rename B to C
rename A to B
touch A (create a new file named A)
fsync A
If that happened for a sequence of hundreds or thousands of renames, it
could massively slow down the logging and cause other secondary effects
like for example blocking other fsync operations and transaction commits
for a very long time (assuming it wouldn't run into -ENOSPC or -ENOMEM
first). However such cases are very uncommon to happen in practice,
nevertheless it's better to be prepared for them and avoid chaos.
Such long sequence of conflicting inodes could be created before this
change.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
btrfs_insert_file_extent() is only ever used to insert holes, so rename
it and remove the redundant parameters.
Reviewed-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: Omar Sandoval <osandov@osandov.com>
Signed-off-by: Sweet Tea Dorminy <sweettea-kernel@dorminy.me>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
With CONFIG_READ_ONLY_THP_FOR_FS, the Linux kernel supports using THPs for
read-only mmapped files, such as shared libraries. However, the kernel
makes no attempt to actually align those mappings on 2MB boundaries,
which makes it impossible to use those THPs most of the time. This issue
applies to general file mapping THP as well as existing setups using
CONFIG_READ_ONLY_THP_FOR_FS. This is easily fixed by using
thp_get_unmapped_area for the unmapped_area function in btrfs, which
is what ext2, ext4, fuse, and xfs all use.
Initially btrfs had been left out in commit 8c07fc452ac0 ("btrfs: fix
alignment of VMA for memory mapped files on THP") as btrfs does not support
DAX. However, commit 1854bc6e24 ("mm/readahead: Align file mappings
for non-DAX") removed the DAX requirement. We should now be able to call
thp_get_unmapped_area() for btrfs.
The problem can be seen in /proc/PID/smaps where THPeligible is set to 0
on mappings to eligible shared object files as shown below.
Before this patch:
7fc6a7e18000-7fc6a80cc000 r-xp 00000000 00:1e 199856
/usr/lib64/libcrypto.so.1.1.1k
Size: 2768 kB
THPeligible: 0
VmFlags: rd ex mr mw me
With this patch the library is mapped at a 2MB aligned address:
fbdfe200000-7fbdfe4b4000 r-xp 00000000 00:1e 199856
/usr/lib64/libcrypto.so.1.1.1k
Size: 2768 kB
THPeligible: 1
VmFlags: rd ex mr mw me
This fixes the alignment of VMAs for any mmap of a file that has the
rd and ex permissions and size >= 2MB. The VMA alignment and
THPeligible field for anonymous memory is handled separately and
is thus not effected by this change.
CC: stable@vger.kernel.org # 5.18+
Signed-off-by: Alexander Zhu <alexlzhu@fb.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
-----BEGIN PGP SIGNATURE-----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=p8YV
-----END PGP SIGNATURE-----
Merge tag 'for-6.0-rc3-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux
Pull btrfs fixes from David Sterba:
"Fixes:
- check that subvolume is writable when changing xattrs from security
namespace
- fix memory leak in device lookup helper
- update generation of hole file extent item when merging holes
- fix space cache corruption and potential double allocations; this
is a rare bug but can be serious once it happens, stable backports
and analysis tool will be provided
- fix error handling when deleting root references
- fix crash due to assert when attempting to cancel suspended device
replace, add message what to do if mount fails due to missing
replace item
Regressions:
- don't merge pages into bio if their page offset is not contiguous
- don't allow large NOWAIT direct reads, this could lead to short
reads eg. in io_uring"
* tag 'for-6.0-rc3-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux:
btrfs: add info when mount fails due to stale replace target
btrfs: replace: drop assert for suspended replace
btrfs: fix silent failure when deleting root reference
btrfs: fix space cache corruption and potential double allocations
btrfs: don't allow large NOWAIT direct reads
btrfs: don't merge pages into bio if their page offset is not contiguous
btrfs: update generation of hole file extent item when merging holes
btrfs: fix possible memory leak in btrfs_get_dev_args_from_path()
btrfs: check if root is readonly while setting security xattr
When punching a hole into a file range that is adjacent with a hole and we
are not using the no-holes feature, we expand the range of the adjacent
file extent item that represents a hole, to save metadata space.
However we don't update the generation of hole file extent item, which
means a full fsync will not log that file extent item if the fsync happens
in a later transaction (since commit 7f30c07288 ("btrfs: stop copying
old file extents when doing a full fsync")).
For example, if we do this:
$ mkfs.btrfs -f -O ^no-holes /dev/sdb
$ mount /dev/sdb /mnt
$ xfs_io -f -c "pwrite -S 0xab 2M 2M" /mnt/foobar
$ sync
We end up with 2 file extent items in our file:
1) One that represents the hole for the file range [0, 2M), with a
generation of 7;
2) Another one that represents an extent covering the range [2M, 4M).
After that if we do the following:
$ xfs_io -c "fpunch 2M 2M" /mnt/foobar
We end up with a single file extent item in the file, which represents a
hole for the range [0, 4M) and with a generation of 7 - because we end
dropping the data extent for range [2M, 4M) and then update the file
extent item that represented the hole at [0, 2M), by increasing
length from 2M to 4M.
Then doing a full fsync and power failing:
$ xfs_io -c "fsync" /mnt/foobar
<power failure>
will result in the full fsync not logging the file extent item that
represents the hole for the range [0, 4M), because its generation is 7,
which is lower than the generation of the current transaction (8).
As a consequence, after mounting again the filesystem (after log replay),
the region [2M, 4M) does not have a hole, it still points to the
previous data extent.
So fix this by always updating the generation of existing file extent
items representing holes when we merge/expand them. This solves the
problem and it's the same approach as when we merge prealloc extents that
got written (at btrfs_mark_extent_written()). Setting the generation to
the current transaction's generation is also what we do when merging
the new hole extent map with the previous one or the next one.
A test case for fstests, covering both cases of hole file extent item
merging (to the left and to the right), will be sent soon.
Fixes: 7f30c07288 ("btrfs: stop copying old file extents when doing a full fsync")
CC: stable@vger.kernel.org # 5.18+
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
-----BEGIN PGP SIGNATURE-----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=sS1q
-----END PGP SIGNATURE-----
Merge tag 'for-5.20-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux
Pull btrfs updates from David Sterba:
"This brings some long awaited changes, the send protocol bump,
otherwise lots of small improvements and fixes. The main core part is
reworking bio handling, cleaning up the submission and endio and
improving error handling.
There are some changes outside of btrfs adding helpers or updating
API, listed at the end of the changelog.
Features:
- sysfs:
- export chunk size, in debug mode add tunable for setting its size
- show zoned among features (was only in debug mode)
- show commit stats (number, last/max/total duration)
- send protocol updated to 2
- new commands:
- ability write larger data chunks than 64K
- send raw compressed extents (uses the encoded data ioctls),
ie. no decompression on send side, no compression needed on
receive side if supported
- send 'otime' (inode creation time) among other timestamps
- send file attributes (a.k.a file flags and xflags)
- this is first version bump, backward compatibility on send and
receive side is provided
- there are still some known and wanted commands that will be
implemented in the near future, another version bump will be
needed, however we want to minimize that to avoid causing
usability issues
- print checksum type and implementation at mount time
- don't print some messages at mount (mentioned as people asked about
it), we want to print messages namely for new features so let's
make some space for that
- big metadata - this has been supported for a long time and is
not a feature that's worth mentioning
- skinny metadata - same reason, set by default by mkfs
Performance improvements:
- reduced amount of reserved metadata for delayed items
- when inserted items can be batched into one leaf
- when deleting batched directory index items
- when deleting delayed items used for deletion
- overall improved count of files/sec, decreased subvolume lock
contention
- metadata item access bounds checker micro-optimized, with a few
percent of improved runtime for metadata-heavy operations
- increase direct io limit for read to 256 sectors, improved
throughput by 3x on sample workload
Notable fixes:
- raid56
- reduce parity writes, skip sectors of stripe when there are no
data updates
- restore reading from on-disk data instead of using stripe cache,
this reduces chances to damage correct data due to RMW cycle
- refuse to replay log with unknown incompat read-only feature bit
set
- zoned
- fix page locking when COW fails in the middle of allocation
- improved tracking of active zones, ZNS drives may limit the
number and there are ENOSPC errors due to that limit and not
actual lack of space
- adjust maximum extent size for zone append so it does not cause
late ENOSPC due to underreservation
- mirror reading error messages show the mirror number
- don't fallback to buffered IO for NOWAIT direct IO writes, we don't
have the NOWAIT semantics for buffered io yet
- send, fix sending link commands for existing file paths when there
are deleted and created hardlinks for same files
- repair all mirrors for profiles with more than 1 copy (raid1c34)
- fix repair of compressed extents, unify where error detection and
repair happen
Core changes:
- bio completion cleanups
- don't double defer compression bios
- simplify endio workqueues
- add more data to btrfs_bio to avoid allocation for read requests
- rework bio error handling so it's same what block layer does,
the submission works and errors are consumed in endio
- when asynchronous bio offload fails fall back to synchronous
checksum calculation to avoid errors under writeback or memory
pressure
- new trace points
- raid56 events
- ordered extent operations
- super block log_root_transid deprecated (never used)
- mixed_backref and big_metadata sysfs feature files removed, they've
been default for sufficiently long time, there are no known users
and mixed_backref could be confused with mixed_groups
Non-btrfs changes, API updates:
- minor highmem API update to cover const arguments
- switch all kmap/kmap_atomic to kmap_local
- remove redundant flush_dcache_page()
- address_space_operations::writepage callback removed
- add bdev_max_segments() helper"
* tag 'for-5.20-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux: (163 commits)
btrfs: don't call btrfs_page_set_checked in finish_compressed_bio_read
btrfs: fix repair of compressed extents
btrfs: remove the start argument to check_data_csum and export
btrfs: pass a btrfs_bio to btrfs_repair_one_sector
btrfs: simplify the pending I/O counting in struct compressed_bio
btrfs: repair all known bad mirrors
btrfs: merge btrfs_dev_stat_print_on_error with its only caller
btrfs: join running log transaction when logging new name
btrfs: simplify error handling in btrfs_lookup_dentry
btrfs: send: always use the rbtree based inode ref management infrastructure
btrfs: send: fix sending link commands for existing file paths
btrfs: send: introduce recorded_ref_alloc and recorded_ref_free
btrfs: zoned: wait until zone is finished when allocation didn't progress
btrfs: zoned: write out partially allocated region
btrfs: zoned: activate necessary block group
btrfs: zoned: activate metadata block group on flush_space
btrfs: zoned: disable metadata overcommit for zoned
btrfs: zoned: introduce space_info->active_total_bytes
btrfs: zoned: finish least available block group on data bg allocation
btrfs: let can_allocate_chunk return error
...
One of the goals is to reduce the overhead of using ->read_iter()
and ->write_iter() instead of ->read()/->write(); new_sync_{read,write}()
has a surprising amount of overhead, in particular inside iocb_flags().
That's why the beginning of the series is in this pile; it's not directly
iov_iter-related, but it's a part of the same work...
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
-----BEGIN PGP SIGNATURE-----
iHUEABYIAB0WIQQqUNBr3gm4hGXdBJlZ7Krx/gZQ6wUCYurGOQAKCRBZ7Krx/gZQ
6ysyAP91lvBfMRepcxpd9kvtuzWkU8A3rfSziZZteEHANB9Q7QEAiPn2a2OjWkcZ
uAyUWfCkHCNx+dSMkEvUgR5okQ0exAM=
=9UCV
-----END PGP SIGNATURE-----
Merge tag 'pull-work.iov_iter-base' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs
Pull vfs iov_iter updates from Al Viro:
"Part 1 - isolated cleanups and optimizations.
One of the goals is to reduce the overhead of using ->read_iter() and
->write_iter() instead of ->read()/->write().
new_sync_{read,write}() has a surprising amount of overhead, in
particular inside iocb_flags(). That's the explanation for the
beginning of the series is in this pile; it's not directly
iov_iter-related, but it's a part of the same work..."
* tag 'pull-work.iov_iter-base' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
first_iovec_segment(): just return address
iov_iter: massage calling conventions for first_{iovec,bvec}_segment()
iov_iter: first_{iovec,bvec}_segment() - simplify a bit
iov_iter: lift dealing with maxpages out of first_{iovec,bvec}_segment()
iov_iter_get_pages{,_alloc}(): cap the maxsize with MAX_RW_COUNT
iov_iter_bvec_advance(): don't bother with bvec_iter
copy_page_{to,from}_iter(): switch iovec variants to generic
keep iocb_flags() result cached in struct file
iocb: delay evaluation of IS_SYNC(...) until we want to check IOCB_DSYNC
struct file: use anonymous union member for rcuhead and llist
btrfs: use IOMAP_DIO_NOSYNC
teach iomap_dio_rw() to suppress dsync
No need of likely/unlikely on calls of check_copy_size()
Currently, for a direct IO write, if we need to fallback to buffered IO,
either to satisfy the whole write operation or just a part of it, we do
it in the current context even if it's a NOWAIT context. This is not ideal
because we currently don't have support for NOWAIT semantics in the
buffered IO path (we can block for several reasons), so we should instead
return -EAGAIN to the caller, so that it knows it should retry (the whole
operation or what's left of it) in a context where blocking is acceptable.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Use simple bool type for the block reserve failfast status, there's
short to save space as there used to be int but there's no reason for
that.
Reviewed-by: Anand Jain <anand.jain@oracle.com>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: David Sterba <dsterba@suse.com>
The chained assignments may be convenient to write, but make readability
a bit worse as it's too easy to overlook that there are several values
set on the same line while this is rather an exception. Making it
consistent everywhere avoids surprises.
The pattern where inode times are initialized reuses the first value and
the order is mtime, ctime. In other blocks the assignments are expanded
so the order of variables is similar to the neighboring code.
Signed-off-by: David Sterba <dsterba@suse.com>
Currently we will return 1 or -EAGAIN if we decide we need to commit
the transaction rather than sync the log. In practice this doesn't
really matter, we interpret any !0 and !BTRFS_NO_LOG_SYNC as needing to
commit the transaction. However this makes it hard to figure out what
the correct thing to do is.
Fix this up by defining BTRFS_LOG_FORCE_COMMIT and using this in all the
places where we want to force the transaction to be committed.
CC: stable@vger.kernel.org # 5.15+
Reviewed-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Josef Bacik <josef@toxicpanda.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
-----BEGIN PGP SIGNATURE-----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=rUrf
-----END PGP SIGNATURE-----
Merge tag 'for-5.19-rc3-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux
Pull btrfs fixes from David Sterba:
- zoned relocation fixes:
- fix critical section end for extent writeback, this could lead
to out of order write
- prevent writing to previous data relocation block group if space
gets low
- reflink fixes:
- fix race between reflinking and ordered extent completion
- proper error handling when block reserve migration fails
- add missing inode iversion/mtime/ctime updates on each iteration
when replacing extents
- fix deadlock when running fsync/fiemap/commit at the same time
- fix false-positive KCSAN report regarding pid tracking for read locks
and data race
- minor documentation update and link to new site
* tag 'for-5.19-rc3-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux:
Documentation: update btrfs list of features and link to readthedocs.io
btrfs: fix deadlock with fsync+fiemap+transaction commit
btrfs: don't set lock_owner when locking extent buffer for reading
btrfs: zoned: fix critical section of relocation inode writeback
btrfs: zoned: prevent allocation from previous data relocation BG
btrfs: do not BUG_ON() on failure to migrate space when replacing extents
btrfs: add missing inode updates on each iteration when replacing extents
btrfs: fix race between reflinking and ordered extent completion
We are hitting the following deadlock in production occasionally
Task 1 Task 2 Task 3 Task 4 Task 5
fsync(A)
start trans
start commit
falloc(A)
lock 5m-10m
start trans
wait for commit
fiemap(A)
lock 0-10m
wait for 5m-10m
(have 0-5m locked)
have btrfs_need_log_full_commit
!full_sync
wait_ordered_extents
finish_ordered_io(A)
lock 0-5m
DEADLOCK
We have an existing dependency of file extent lock -> transaction.
However in fsync if we tried to do the fast logging, but then had to
fall back to committing the transaction, we will be forced to call
btrfs_wait_ordered_range() to make sure all of our extents are updated.
This creates a dependency of transaction -> file extent lock, because
btrfs_finish_ordered_io() will need to take the file extent lock in
order to run the ordered extents.
Fix this by stopping the transaction if we have to do the full commit
and we attempted to do the fast logging. Then attach to the transaction
and commit it if we need to.
CC: stable@vger.kernel.org # 5.15+
Reviewed-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: David Sterba <dsterba@suse.com>
At btrfs_replace_file_extents(), if we fail to migrate reserved metadata
space from the transaction block reserve into the local block reserve,
we trigger a BUG_ON(). This is because it should not be possible to have
a failure here, as we reserved more space when we started the transaction
than the space we want to migrate. However having a BUG_ON() is way too
drastic, we can perfectly handle the failure and return the error to the
caller. So just do that instead, and add a WARN_ON() to make it easier
to notice the failure if it ever happens (which is particularly useful
for fstests, and the warning will trigger a failure of a test case).
Reviewed-by: Boris Burkov <boris@bur.io>
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
When replacing file extents, called during fallocate, hole punching,
clone and deduplication, we may not be able to replace/drop all the
target file extent items with a single transaction handle. We may get
-ENOSPC while doing it, in which case we release the transaction handle,
balance the dirty pages of the btree inode, flush delayed items and get
a new transaction handle to operate on what's left of the target range.
By dropping and replacing file extent items we have effectively modified
the inode, so we should bump its iversion and update its mtime/ctime
before we update the inode item. This is because if the transaction
we used for partially modifying the inode gets committed by someone after
we release it and before we finish the rest of the range, a power failure
happens, then after mounting the filesystem our inode has an outdated
iversion and mtime/ctime, corresponding to the values it had before we
changed it.
So add the missing iversion and mtime/ctime updates.
Reviewed-by: Boris Burkov <boris@bur.io>
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
New helper to be used instead of direct checks for IOCB_DSYNC:
iocb_is_dsync(iocb). Checks converted, which allows to avoid
the IS_SYNC(iocb->ki_filp->f_mapping->host) part (4 cache lines)
from iocb_flags() - it's checked in iocb_is_dsync() instead
Reviewed-by: Christian Brauner (Microsoft) <brauner@kernel.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
... instead of messing with iocb flags
Suggested-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Christian Brauner (Microsoft) <brauner@kernel.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
- Appoint myself page cache maintainer
- Fix how scsicam uses the page cache
- Use the memalloc_nofs_save() API to replace AOP_FLAG_NOFS
- Remove the AOP flags entirely
- Remove pagecache_write_begin() and pagecache_write_end()
- Documentation updates
- Convert several address_space operations to use folios:
- is_dirty_writeback
- readpage becomes read_folio
- releasepage becomes release_folio
- freepage becomes free_folio
- Change filler_t to require a struct file pointer be the first argument
like ->read_folio
-----BEGIN PGP SIGNATURE-----
iQEzBAABCgAdFiEEejHryeLBw/spnjHrDpNsjXcpgj4FAmKNMDUACgkQDpNsjXcp
gj4/mwf/bpHhXH4ZoNIvtUpTF6rZbqeffmc0VrbxCZDZ6igRnRPglxZ9H9v6L53O
7B0FBQIfxgNKHZpdqGdOkv8cjg/GMe/HJUbEy5wOakYPo4L9fZpHbDZ9HM2Eankj
xBqLIBgBJ7doKr+Y62DAN19TVD8jfRfVtli5mqXJoNKf65J7BkxljoTH1L3EXD9d
nhLAgyQjR67JQrT/39KMW+17GqLhGefLQ4YnAMONtB6TVwX/lZmigKpzVaCi4r26
bnk5vaR/3PdjtNxIoYvxdc71y2Eg05n2jEq9Wcy1AaDv/5vbyZUlZ2aBSaIVbtKX
WfrhN9O3L0bU5qS7p9PoyfLc9wpq8A==
=djLv
-----END PGP SIGNATURE-----
Merge tag 'folio-5.19' of git://git.infradead.org/users/willy/pagecache
Pull page cache updates from Matthew Wilcox:
- Appoint myself page cache maintainer
- Fix how scsicam uses the page cache
- Use the memalloc_nofs_save() API to replace AOP_FLAG_NOFS
- Remove the AOP flags entirely
- Remove pagecache_write_begin() and pagecache_write_end()
- Documentation updates
- Convert several address_space operations to use folios:
- is_dirty_writeback
- readpage becomes read_folio
- releasepage becomes release_folio
- freepage becomes free_folio
- Change filler_t to require a struct file pointer be the first
argument like ->read_folio
* tag 'folio-5.19' of git://git.infradead.org/users/willy/pagecache: (107 commits)
nilfs2: Fix some kernel-doc comments
Appoint myself page cache maintainer
fs: Remove aops->freepage
secretmem: Convert to free_folio
nfs: Convert to free_folio
orangefs: Convert to free_folio
fs: Add free_folio address space operation
fs: Convert drop_buffers() to use a folio
fs: Change try_to_free_buffers() to take a folio
jbd2: Convert release_buffer_page() to use a folio
jbd2: Convert jbd2_journal_try_to_free_buffers to take a folio
reiserfs: Convert release_buffer_page() to use a folio
fs: Remove last vestiges of releasepage
ubifs: Convert to release_folio
reiserfs: Convert to release_folio
orangefs: Convert to release_folio
ocfs2: Convert to release_folio
nilfs2: Remove comment about releasepage
nfs: Convert to release_folio
jfs: Convert to release_folio
...
Add a wrapper around iomap_dio_rw that keeps the direct I/O internals
isolated in inode.c.
Reviewed-by: Nikolay Borisov <nborisov@suse.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
When doing a NOWAIT direct IO write, if we can NOCOW then it means we can
proceed with the non-blocking, NOWAIT path. However reserving the metadata
space and qgroup meta space can often result in blocking - flushing
delalloc, wait for ordered extents to complete, trigger transaction
commits, etc, going against the semantics of a NOWAIT write.
So make the NOWAIT write path to try to reserve all the metadata it needs
without resulting in a blocking behaviour - if we get -ENOSPC or -EDQUOT
then return -EAGAIN to make the caller fallback to a blocking direct IO
write.
This is part of a patchset comprised of the following patches:
btrfs: avoid blocking on page locks with nowait dio on compressed range
btrfs: avoid blocking nowait dio when locking file range
btrfs: avoid double nocow check when doing nowait dio writes
btrfs: stop allocating a path when checking if cross reference exists
btrfs: free path at can_nocow_extent() before checking for checksum items
btrfs: release path earlier at can_nocow_extent()
btrfs: avoid blocking when allocating context for nowait dio read/write
btrfs: avoid blocking on space revervation when doing nowait dio writes
The following test was run before and after applying this patchset:
$ cat io-uring-nodatacow-test.sh
#!/bin/bash
DEV=/dev/sdc
MNT=/mnt/sdc
MOUNT_OPTIONS="-o ssd -o nodatacow"
MKFS_OPTIONS="-R free-space-tree -O no-holes"
NUM_JOBS=4
FILE_SIZE=8G
RUN_TIME=300
cat <<EOF > /tmp/fio-job.ini
[io_uring_rw]
rw=randrw
fsync=0
fallocate=posix
group_reporting=1
direct=1
ioengine=io_uring
iodepth=64
bssplit=4k/20:8k/20:16k/20:32k/10:64k/10:128k/5:256k/5:512k/5:1m/5
filesize=$FILE_SIZE
runtime=$RUN_TIME
time_based
filename=foobar
directory=$MNT
numjobs=$NUM_JOBS
thread
EOF
echo performance | \
tee /sys/devices/system/cpu/cpu*/cpufreq/scaling_governor
umount $MNT &> /dev/null
mkfs.btrfs -f $MKFS_OPTIONS $DEV &> /dev/null
mount $MOUNT_OPTIONS $DEV $MNT
fio /tmp/fio-job.ini
umount $MNT
The test was run a 12 cores box with 64G of ram, using a non-debug kernel
config (Debian's default config) and a spinning disk.
Result before the patchset:
READ: bw=407MiB/s (427MB/s), 407MiB/s-407MiB/s (427MB/s-427MB/s), io=119GiB (128GB), run=300175-300175msec
WRITE: bw=407MiB/s (427MB/s), 407MiB/s-407MiB/s (427MB/s-427MB/s), io=119GiB (128GB), run=300175-300175msec
Result after the patchset:
READ: bw=436MiB/s (457MB/s), 436MiB/s-436MiB/s (457MB/s-457MB/s), io=128GiB (137GB), run=300044-300044msec
WRITE: bw=435MiB/s (456MB/s), 435MiB/s-435MiB/s (456MB/s-456MB/s), io=128GiB (137GB), run=300044-300044msec
That's about +7.2% throughput for reads and +6.9% for writes.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
When doing a NOWAIT direct IO write we are checking twice if we can COW
into the target file range using can_nocow_extent() - once at the very
beginning of the write path, at btrfs_write_check() via
check_nocow_nolock(), and later again at btrfs_get_blocks_direct_write().
The can_nocow_extent() function does a lot of expensive things - searching
for the file extent item in the inode's subvolume tree, searching for the
extent item in the extent tree, checking delayed references, etc, so it
isn't a very cheap call.
We can remove the first check at btrfs_write_check(), and add there a
quick check to verify if the inode has the NODATACOW or PREALLOC flags,
and quickly bail out if it doesn't have neither of those flags, as that
means we have to COW and therefore can't comply with the NOWAIT semantics.
After this we do only one call to can_nocow_extent(), while we are at
btrfs_get_blocks_direct_write(), where we have already locked the file
range and we did a try lock on the range before, at
btrfs_dio_iomap_begin() (since the previous patch in the series).
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
We have four different scenarios where we don't expect to find ordered
extents after locking a file range:
1) During plain fallocate;
2) During hole punching;
3) During zero range;
4) During reflinks (both cloning and deduplication).
This is because in all these cases we follow the pattern:
1) Lock the inode's VFS lock in exclusive mode;
2) Lock the inode's i_mmap_lock in exclusive node, to serialize with
mmap writes;
3) Flush delalloc in a file range and wait for all ordered extents
to complete - both done through btrfs_wait_ordered_range();
4) Lock the file range in the inode's io_tree.
So add a helper that asserts that we don't have ordered extents for a
given range. Make the four scenarios listed above use this helper after
locking the respective file range.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
For hole punching and zero range we have this loop that checks if we have
ordered extents after locking the file range, and if so unlock the range,
wait for ordered extents, and retry until we don't find more ordered
extents.
This logic was needed in the past because:
1) Direct IO writes within the i_size boundary did not take the inode's
VFS lock. This was because that lock used to be a mutex, then some
years ago it was switched to a rw semaphore (commit 9902af79c0
("parallel lookups: actual switch to rwsem")), and then btrfs was
changed to take the VFS inode's lock in shared mode for writes that
don't cross the i_size boundary (commit e9adabb971 ("btrfs: use
shared lock for direct writes within EOF"));
2) We could race with memory mapped writes, because memory mapped writes
don't acquire the inode's VFS lock. We don't have that race anymore,
as we have a rw semaphore to synchronize memory mapped writes with
fallocate (and reflinking too). That change happened with commit
8d9b4a162a ("btrfs: exclude mmap from happening during all
fallocate operations").
So stop looking for ordered extents after locking the file range when
doing hole punching and zero range operations.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
When doing hole punching we are flushing delalloc and waiting for ordered
extents to complete before locking the inode (VFS lock and the btrfs
specific i_mmap_lock). This is fine because even if a write happens after
we call btrfs_wait_ordered_range() and before we lock the inode (call
btrfs_inode_lock()), we will notice the write at
btrfs_punch_hole_lock_range() and flush delalloc and wait for its ordered
extent.
We can however make this simpler by locking first the inode an then call
btrfs_wait_ordered_range(), which will allow us to remove the ordered
extent lookup logic from btrfs_punch_hole_lock_range() in the next patch.
It also makes the behaviour the same as plain fallocate, hole punching
and reflinks.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
For fallocate() we have this loop that checks if we have ordered extents
after locking the file range, and if so unlock the range, wait for ordered
extents, and retry until we don't find more ordered extents.
This logic was needed in the past because:
1) Direct IO writes within the i_size boundary did not take the inode's
VFS lock. This was because that lock used to be a mutex, then some
years ago it was switched to a rw semaphore (commit 9902af79c0
("parallel lookups: actual switch to rwsem")), and then btrfs was
changed to take the VFS inode's lock in shared mode for writes that
don't cross the i_size boundary (commit e9adabb971 ("btrfs: use
shared lock for direct writes within EOF"));
2) We could race with memory mapped writes, because memory mapped writes
don't acquire the inode's VFS lock. We don't have that race anymore,
as we have a rw semaphore to synchronize memory mapped writes with
fallocate (and reflinking too). That change happened with commit
8d9b4a162a ("btrfs: exclude mmap from happening during all
fallocate operations").
So stop looking for ordered extents after locking the file range when
doing a plain fallocate.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
When starting a fallocate zero range operation, before getting the first
extent map for the range, we make a call to inode_dio_wait().
This logic was needed in the past because direct IO writes within the
i_size boundary did not take the inode's VFS lock. This was because that
lock used to be a mutex, then some years ago it was switched to a rw
semaphore (by commit 9902af79c0 ("parallel lookups: actual switch to
rwsem")), and then btrfs was changed to take the VFS inode's lock in
shared mode for writes that don't cross the i_size boundary (done in
commit e9adabb971 ("btrfs: use shared lock for direct writes within
EOF")). The lockless direct IO writes could result in a race with the
zero range operation, resulting in the later getting a stale extent
map for the range.
So remove this no longer needed call to inode_dio_wait(), as fallocate
takes the inode's VFS lock in exclusive mode and direct IO writes within
i_size take that same lock in shared mode.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
During a plain fallocate, we always start by reserving an amount of data
space that matches the length of the range passed to fallocate. When we
already have extents allocated in that range, we may end up trying to
reserve a lot more data space then we need, which can result in several
undesired behaviours:
1) We fail with -ENOSPC. For example the passed range has a length
of 1G, but there's only one hole with a size of 1M in that range;
2) We temporarily reserve excessive data space that could be used by
other operations happening concurrently;
3) By reserving much more data space then we need, we can end up
doing expensive things like triggering dellaloc for other inodes,
waiting for the ordered extents to complete, trigger transaction
commits, allocate new block groups, etc.
Example:
$ cat test.sh
#!/bin/bash
DEV=/dev/sdj
MNT=/mnt/sdj
mkfs.btrfs -f -b 1g $DEV
mount $DEV $MNT
# Create a file with a size of 600M and two holes, one at [200M, 201M[
# and another at [401M, 402M[
xfs_io -f -c "pwrite -S 0xab 0 200M" \
-c "pwrite -S 0xcd 201M 200M" \
-c "pwrite -S 0xef 402M 198M" \
$MNT/foobar
# Now call fallocate against the whole file range, see if it fails
# with -ENOSPC or not - it shouldn't since we only need to allocate
# 2M of data space.
xfs_io -c "falloc 0 600M" $MNT/foobar
umount $MNT
$ ./test.sh
(...)
wrote 209715200/209715200 bytes at offset 0
200 MiB, 51200 ops; 0.8063 sec (248.026 MiB/sec and 63494.5831 ops/sec)
wrote 209715200/209715200 bytes at offset 210763776
200 MiB, 51200 ops; 0.8053 sec (248.329 MiB/sec and 63572.3172 ops/sec)
wrote 207618048/207618048 bytes at offset 421527552
198 MiB, 50688 ops; 0.7925 sec (249.830 MiB/sec and 63956.5548 ops/sec)
fallocate: No space left on device
$
So fix this by not allocating an amount of data space that matches the
length of the range passed to fallocate. Instead allocate an amount of
data space that corresponds to the sum of the sizes of each hole found
in the range. This reservation now happens after we have locked the file
range, which is safe since we know at this point there's no delalloc
in the range because we've taken the inode's VFS lock in exclusive mode,
we have taken the inode's i_mmap_lock in exclusive mode, we have flushed
delalloc and waited for all ordered extents in the range to complete.
This type of failure actually seems to happen in practice with systemd,
and we had at least one report about this in a very long thread which
is referenced by the Link tag below.
Link: https://lore.kernel.org/linux-btrfs/bdJVxLiFr_PyQSXRUbZJfFW_jAjsGgoMetqPHJMbg-hdy54Xt_ZHhRetmnJ6cJ99eBlcX76wy-AvWwV715c3YndkxneSlod11P1hlaADx0s=@protonmail.com/
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
I've only converted the outer layers of the btrfs release_folio paths
to use folios; the use of folios should be pushed further down into
btrfs from here.
Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Reviewed-by: Jeff Layton <jlayton@kernel.org>
With all implementations of aops->readpage converted to aops->read_folio,
we can stop checking whether it's set and remove the member from aops.
Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
This is a "weak" conversion which converts straight back to using pages.
A full conversion should be performed at some point, hopefully by
someone familiar with the filesystem.
Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Change all the callers of ->readpage to call ->read_folio in preference,
if it exists. This is a transitional duplication, and will be removed
by the end of the series.
Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Since the initial introduction of (posix) fallocate back at the turn of
the century, it has been possible to use this syscall to change the
user-visible contents of files. This can happen by extending the file
size during a preallocation, or through any of the newer modes (punch,
zero range). Because the call can be used to change file contents, we
should treat it like we do any other modification to a file -- update
the mtime, and drop set[ug]id privileges/capabilities.
The VFS function file_modified() does all this for us if pass it a
locked inode, so let's make fallocate drop permissions correctly.
Reviewed-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: David Sterba <dsterba@suse.com>
When an inode has a last_reflink_trans matching the current transaction,
we have to take special care when logging its checksums in order to
avoid getting checksum items with overlapping ranges in a log tree,
which could result in missing checksums after log replay (more on that
in the changelogs of commit 40e046acbd ("Btrfs: fix missing data
checksums after replaying a log tree") and commit e289f03ea7 ("btrfs:
fix corrupt log due to concurrent fsync of inodes with shared extents")).
We also need to make sure a full fsync will copy all old file extent
items it finds in modified leaves, because they might have been copied
from some other inode.
However once we fsync an inode, we don't need to keep paying the price of
that extra special care in future fsyncs done in the same transaction,
unless the inode is used for another reflink operation or the full sync
flag is set on it (truncate, failure to allocate extent maps for holes,
and other exceptional and infrequent cases).
So after we fsync an inode reset its last_unlink_trans to zero. In case
another reflink happens, we continue to update the last_reflink_trans of
the inode, just as before. Also set last_reflink_trans to the generation
of the last transaction that modified the inode whenever we need to set
the full sync flag on the inode, just like when we need to load an inode
from disk after eviction.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
The implementation resembles direct I/O: we have to flush any ordered
extents, invalidate the page cache, and do the io tree/delalloc/extent
map/ordered extent dance. From there, we can reuse the compression code
with a minor modification to distinguish the write from writeback. This
also creates inline extents when possible.
Signed-off-by: Omar Sandoval <osandov@fb.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Currently, we always reserve the same extent size in the file and extent
size on disk for delalloc because the former is the worst case for the
latter. For BTRFS_IOC_ENCODED_WRITE writes, we know the exact size of
the extent on disk, which may be less than or greater than (for
bookends) the size in the file. Add a disk_num_bytes parameter to
btrfs_delalloc_reserve_metadata() so that we can reserve the correct
amount of csum bytes. No functional change.
Reviewed-by: Nikolay Borisov <nborisov@suse.com>
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Omar Sandoval <osandov@fb.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
At btrfs_drop_extents(), we try to replace a range of file extent items
with a new file extent in a single btree search, to avoid the need to do
a search for deletion, followed by a path release and followed by yet
another search for insertion.
When I originally added that optimization, in commit 1acae57b16
("Btrfs: faster file extent item replace operations"), I left a constraint
to do the fast replace only if we visited a single leaf. That was because
in the most common case we find all file extent items that need to be
deleted (or trimmed) in a single leaf, however it can work for other
common cases like when we need to delete a few file extent items located
at the end of a leaf and a few more located at the beginning of the next
leaf. The key for the new file extent item is greater than the key of
any deleted or trimmed file extent item from previous leaves, so we are
fine to use the last leaf that we found as long as we are holding a
write lock on it - even if the new key ends up at slot 0, as if that's
the case, the btree search has obtained a write lock on any upper nodes
that need to have a key pointer updated.
So removed the constraint that limits the optimization to the case where
we visited only a single leaf.
This change if part of a patchset that is comprised of the following
patches:
1/6 btrfs: remove unnecessary leaf free space checks when pushing items
2/6 btrfs: avoid unnecessary COW of leaves when deleting items from a leaf
3/6 btrfs: avoid unnecessary computation when deleting items from a leaf
4/6 btrfs: remove constraint on number of visited leaves when replacing extents
5/6 btrfs: remove useless path release in the fast fsync path
6/6 btrfs: prepare extents to be logged before locking a log tree path
The last patch in the series has some performance test result in its
changelog.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
There is a big gap between inode_should_defrag() and autodefrag extent
size threshold. For inode_should_defrag() it has a flexible
@small_write value. For compressed extent is 16K, and for non-compressed
extent it's 64K.
However for autodefrag extent size threshold, it's always fixed to the
default value (256K).
This means, the following write sequence will trigger autodefrag to
defrag ranges which didn't trigger autodefrag:
pwrite 0 8k
sync
pwrite 8k 128K
sync
The latter 128K write will also be considered as a defrag target (if
other conditions are met). While only that 8K write is really
triggering autodefrag.
Such behavior can cause extra IO for autodefrag.
Close the gap, by copying the @small_write value into inode_defrag, so
that later autodefrag can use the same @small_write value which
triggered autodefrag.
With the existing transid value, this allows autodefrag really to scan
the ranges which triggered autodefrag.
Although this behavior change is mostly reducing the extent_thresh value
for autodefrag, I believe in the future we should allow users to specify
the autodefrag extent threshold through mount options, but that's an
other problem to consider in the future.
CC: stable@vger.kernel.org # 5.16+
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Although we have btrfs_requeue_inode_defrag(), for autodefrag we are
still just exhausting all inode_defrag items in the tree.
This means, it doesn't make much difference to requeue an inode_defrag,
other than scan the inode from the beginning till its end.
Change the behaviour to always scan from offset 0 of an inode, and till
the end.
By this we get the following benefit:
- Straight-forward code
- No more re-queue related check
- Fewer members in inode_defrag
We still keep the same btrfs_get_fs_root() and btrfs_iget() check for
each loop, and added extra should_auto_defrag() check per-loop.
Note: the patch needs to be backported and is intentionally written
to minimize the diff size, code will be cleaned up later.
CC: stable@vger.kernel.org # 5.16
Signed-off-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
If we do a direct IO read or write when the buffer given by the user is
memory mapped to the file range we are going to do IO, we end up ending
in a deadlock. This is triggered by the new test case generic/647 from
fstests.
For a direct IO read we get a trace like this:
[967.872718] INFO: task mmap-rw-fault:12176 blocked for more than 120 seconds.
[967.874161] Not tainted 5.14.0-rc7-btrfs-next-95 #1
[967.874909] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
[967.875983] task:mmap-rw-fault state:D stack: 0 pid:12176 ppid: 11884 flags:0x00000000
[967.875992] Call Trace:
[967.875999] __schedule+0x3ca/0xe10
[967.876015] schedule+0x43/0xe0
[967.876020] wait_extent_bit.constprop.0+0x1eb/0x260 [btrfs]
[967.876109] ? do_wait_intr_irq+0xb0/0xb0
[967.876118] lock_extent_bits+0x37/0x90 [btrfs]
[967.876150] btrfs_lock_and_flush_ordered_range+0xa9/0x120 [btrfs]
[967.876184] ? extent_readahead+0xa7/0x530 [btrfs]
[967.876214] extent_readahead+0x32d/0x530 [btrfs]
[967.876253] ? lru_cache_add+0x104/0x220
[967.876255] ? kvm_sched_clock_read+0x14/0x40
[967.876258] ? sched_clock_cpu+0xd/0x110
[967.876263] ? lock_release+0x155/0x4a0
[967.876271] read_pages+0x86/0x270
[967.876274] ? lru_cache_add+0x125/0x220
[967.876281] page_cache_ra_unbounded+0x1a3/0x220
[967.876291] filemap_fault+0x626/0xa20
[967.876303] __do_fault+0x36/0xf0
[967.876308] __handle_mm_fault+0x83f/0x15f0
[967.876322] handle_mm_fault+0x9e/0x260
[967.876327] __get_user_pages+0x204/0x620
[967.876332] ? get_user_pages_unlocked+0x69/0x340
[967.876340] get_user_pages_unlocked+0xd3/0x340
[967.876349] internal_get_user_pages_fast+0xbca/0xdc0
[967.876366] iov_iter_get_pages+0x8d/0x3a0
[967.876374] bio_iov_iter_get_pages+0x82/0x4a0
[967.876379] ? lock_release+0x155/0x4a0
[967.876387] iomap_dio_bio_actor+0x232/0x410
[967.876396] iomap_apply+0x12a/0x4a0
[967.876398] ? iomap_dio_rw+0x30/0x30
[967.876414] __iomap_dio_rw+0x29f/0x5e0
[967.876415] ? iomap_dio_rw+0x30/0x30
[967.876420] ? lock_acquired+0xf3/0x420
[967.876429] iomap_dio_rw+0xa/0x30
[967.876431] btrfs_file_read_iter+0x10b/0x140 [btrfs]
[967.876460] new_sync_read+0x118/0x1a0
[967.876472] vfs_read+0x128/0x1b0
[967.876477] __x64_sys_pread64+0x90/0xc0
[967.876483] do_syscall_64+0x3b/0xc0
[967.876487] entry_SYSCALL_64_after_hwframe+0x44/0xae
[967.876490] RIP: 0033:0x7fb6f2c038d6
[967.876493] RSP: 002b:00007fffddf586b8 EFLAGS: 00000246 ORIG_RAX: 0000000000000011
[967.876496] RAX: ffffffffffffffda RBX: 0000000000001000 RCX: 00007fb6f2c038d6
[967.876498] RDX: 0000000000001000 RSI: 00007fb6f2c17000 RDI: 0000000000000003
[967.876499] RBP: 0000000000001000 R08: 0000000000000003 R09: 0000000000000000
[967.876501] R10: 0000000000001000 R11: 0000000000000246 R12: 0000000000000003
[967.876502] R13: 0000000000000000 R14: 00007fb6f2c17000 R15: 0000000000000000
This happens because at btrfs_dio_iomap_begin() we lock the extent range
and return with it locked - we only unlock in the endio callback, at
end_bio_extent_readpage() -> endio_readpage_release_extent(). Then after
iomap called the btrfs_dio_iomap_begin() callback, it triggers the page
faults that resulting in reading the pages, through the readahead callback
btrfs_readahead(), and through there we end to attempt to lock again the
same extent range (or a subrange of what we locked before), resulting in
the deadlock.
For a direct IO write, the scenario is a bit different, and it results in
trace like this:
[1132.442520] run fstests generic/647 at 2021-08-31 18:53:35
[1330.349355] INFO: task mmap-rw-fault:184017 blocked for more than 120 seconds.
[1330.350540] Not tainted 5.14.0-rc7-btrfs-next-95 #1
[1330.351158] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
[1330.351900] task:mmap-rw-fault state:D stack: 0 pid:184017 ppid:183725 flags:0x00000000
[1330.351906] Call Trace:
[1330.351913] __schedule+0x3ca/0xe10
[1330.351930] schedule+0x43/0xe0
[1330.351935] btrfs_start_ordered_extent+0x108/0x1c0 [btrfs]
[1330.352020] ? do_wait_intr_irq+0xb0/0xb0
[1330.352028] btrfs_lock_and_flush_ordered_range+0x8c/0x120 [btrfs]
[1330.352064] ? extent_readahead+0xa7/0x530 [btrfs]
[1330.352094] extent_readahead+0x32d/0x530 [btrfs]
[1330.352133] ? lru_cache_add+0x104/0x220
[1330.352135] ? kvm_sched_clock_read+0x14/0x40
[1330.352138] ? sched_clock_cpu+0xd/0x110
[1330.352143] ? lock_release+0x155/0x4a0
[1330.352151] read_pages+0x86/0x270
[1330.352155] ? lru_cache_add+0x125/0x220
[1330.352162] page_cache_ra_unbounded+0x1a3/0x220
[1330.352172] filemap_fault+0x626/0xa20
[1330.352176] ? filemap_map_pages+0x18b/0x660
[1330.352184] __do_fault+0x36/0xf0
[1330.352189] __handle_mm_fault+0x1253/0x15f0
[1330.352203] handle_mm_fault+0x9e/0x260
[1330.352208] __get_user_pages+0x204/0x620
[1330.352212] ? get_user_pages_unlocked+0x69/0x340
[1330.352220] get_user_pages_unlocked+0xd3/0x340
[1330.352229] internal_get_user_pages_fast+0xbca/0xdc0
[1330.352246] iov_iter_get_pages+0x8d/0x3a0
[1330.352254] bio_iov_iter_get_pages+0x82/0x4a0
[1330.352259] ? lock_release+0x155/0x4a0
[1330.352266] iomap_dio_bio_actor+0x232/0x410
[1330.352275] iomap_apply+0x12a/0x4a0
[1330.352278] ? iomap_dio_rw+0x30/0x30
[1330.352292] __iomap_dio_rw+0x29f/0x5e0
[1330.352294] ? iomap_dio_rw+0x30/0x30
[1330.352306] btrfs_file_write_iter+0x238/0x480 [btrfs]
[1330.352339] new_sync_write+0x11f/0x1b0
[1330.352344] ? NF_HOOK_LIST.constprop.0.cold+0x31/0x3e
[1330.352354] vfs_write+0x292/0x3c0
[1330.352359] __x64_sys_pwrite64+0x90/0xc0
[1330.352365] do_syscall_64+0x3b/0xc0
[1330.352369] entry_SYSCALL_64_after_hwframe+0x44/0xae
[1330.352372] RIP: 0033:0x7f4b0a580986
[1330.352379] RSP: 002b:00007ffd34d75418 EFLAGS: 00000246 ORIG_RAX: 0000000000000012
[1330.352382] RAX: ffffffffffffffda RBX: 0000000000001000 RCX: 00007f4b0a580986
[1330.352383] RDX: 0000000000001000 RSI: 00007f4b0a3a4000 RDI: 0000000000000003
[1330.352385] RBP: 00007f4b0a3a4000 R08: 0000000000000003 R09: 0000000000000000
[1330.352386] R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000003
[1330.352387] R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000000
Unlike for reads, at btrfs_dio_iomap_begin() we return with the extent
range unlocked, but later when the page faults are triggered and we try
to read the extents, we end up btrfs_lock_and_flush_ordered_range() where
we find the ordered extent for our write, created by the iomap callback
btrfs_dio_iomap_begin(), and we wait for it to complete, which makes us
deadlock since we can't complete the ordered extent without reading the
pages (the iomap code only submits the bio after the pages are faulted
in).
Fix this by setting the nofault attribute of the given iov_iter and retry
the direct IO read/write if we get an -EFAULT error returned from iomap.
For reads, also disable page faults completely, this is because when we
read from a hole or a prealloc extent, we can still trigger page faults
due to the call to iov_iter_zero() done by iomap - at the moment, it is
oblivious to the value of the ->nofault attribute of an iov_iter.
We also need to keep track of the number of bytes written or read, and
pass it to iomap_dio_rw(), as well as use the new flag IOMAP_DIO_PARTIAL.
This depends on the iov_iter and iomap changes introduced in commit
c03098d4b9 ("Merge tag 'gfs2-v5.15-rc5-mmap-fault' of
git://git.kernel.org/pub/scm/linux/kernel/git/gfs2/linux-gfs2").
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Functions gfs2_file_read_iter and gfs2_file_write_iter are both
accessing the user buffer to write to or read from while holding the
inode glock. In the most basic scenario, that buffer will not be
resident and it will be mapped to the same file. Accessing the buffer
will trigger a page fault, and gfs2 will deadlock trying to take the
same inode glock again while trying to handle that fault.
Fix that and similar, more complex scenarios by disabling page faults
while accessing user buffers. To make this work, introduce a small
amount of new infrastructure and fix some bugs that didn't trigger so
far, with page faults enabled.
-----BEGIN PGP SIGNATURE-----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=WeNd
-----END PGP SIGNATURE-----
Merge tag 'gfs2-v5.15-rc5-mmap-fault' of git://git.kernel.org/pub/scm/linux/kernel/git/gfs2/linux-gfs2
Pull gfs2 mmap + page fault deadlocks fixes from Andreas Gruenbacher:
"Functions gfs2_file_read_iter and gfs2_file_write_iter are both
accessing the user buffer to write to or read from while holding the
inode glock.
In the most basic deadlock scenario, that buffer will not be resident
and it will be mapped to the same file. Accessing the buffer will
trigger a page fault, and gfs2 will deadlock trying to take the same
inode glock again while trying to handle that fault.
Fix that and similar, more complex scenarios by disabling page faults
while accessing user buffers. To make this work, introduce a small
amount of new infrastructure and fix some bugs that didn't trigger so
far, with page faults enabled"
* tag 'gfs2-v5.15-rc5-mmap-fault' of git://git.kernel.org/pub/scm/linux/kernel/git/gfs2/linux-gfs2:
gfs2: Fix mmap + page fault deadlocks for direct I/O
iov_iter: Introduce nofault flag to disable page faults
gup: Introduce FOLL_NOFAULT flag to disable page faults
iomap: Add done_before argument to iomap_dio_rw
iomap: Support partial direct I/O on user copy failures
iomap: Fix iomap_dio_rw return value for user copies
gfs2: Fix mmap + page fault deadlocks for buffered I/O
gfs2: Eliminate ip->i_gh
gfs2: Move the inode glock locking to gfs2_file_buffered_write
gfs2: Introduce flag for glock holder auto-demotion
gfs2: Clean up function may_grant
gfs2: Add wrapper for iomap_file_buffered_write
iov_iter: Introduce fault_in_iov_iter_writeable
iov_iter: Turn iov_iter_fault_in_readable into fault_in_iov_iter_readable
gup: Turn fault_in_pages_{readable,writeable} into fault_in_{readable,writeable}
powerpc/kvm: Fix kvm_use_magic_page
iov_iter: Fix iov_iter_get_pages{,_alloc} page fault return value
In order to make 'real_root' used only in ref-verify it's required to
have the necessary context to perform the same checks that this member
is used for. So add 'mod_root' which will contain the root on behalf of
which a delayed ref was created and a 'skip_group' parameter which
will contain callsite-specific override of skip_qgroup.
Signed-off-by: Nikolay Borisov <nborisov@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
We have a few flags that are inconsistently used to describe the fs in
different states of failure. As of 5963ffcaf3 ("btrfs: always abort
the transaction if we abort a trans handle") we will always set
BTRFS_FS_STATE_ERROR if we abort, so we don't have to check both ABORTED
and ERROR to see if things have gone wrong. Add a helper to check
BTRFS_FS_STATE_ERROR and then convert all checkers of FS_STATE_ERROR to
use the helper.
The TRANS_ABORTED bit check was added in af72273381 ("Btrfs: clean up
resources during umount after trans is aborted") but is not actually
specific.
Reviewed-by: Anand Jain <anand.jain@oracle.com>
Reviewed-by: Nikolay Borisov <nborisov@suse.com>
Signed-off-by: Josef Bacik <josef@toxicpanda.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Although in btrfs we have very limited usage of PageChecked flag, it's
still some page flag not yet subpage compatible.
Fix it by introducing btrfs_subpage::checked_offset to do the convert.
For most call sites, especially for free-space cache, COW fixup and
btrfs_invalidatepage(), they all work in full page mode anyway.
For other call sites, they work as subpage compatible mode.
Some call sites need extra modification:
- btrfs_drop_pages()
Needs extra parameter to get the real range we need to clear checked
flag.
Also since btrfs_drop_pages() will accept pages beyond the dirtied
range, update btrfs_subpage_clamp_range() to handle such case
by setting @len to 0 if the page is beyond target range.
- btrfs_invalidatepage()
We need to call subpage helper before calling __btrfs_releasepage(),
or it will trigger ASSERT() as page->private will be cleared.
- btrfs_verify_data_csum()
In theory we don't need the io_bio->csum check anymore, but it's
won't hurt. Just change the comment.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Since setup_items_for_insert() is not used anymore outside of ctree.c,
make it static and remove its prototype from ctree.h. This also requires
to move the definition of setup_item_for_insert() from ctree.h to ctree.c
and move down btrfs_duplicate_item() so that it's defined after
setup_items_for_insert().
Further, since setup_item_for_insert() is used outside ctree.c, rename it
to btrfs_setup_item_for_insert().
This patch is part of a small patchset that is comprised of the following
patches:
btrfs: loop only once over data sizes array when inserting an item batch
btrfs: unexport setup_items_for_insert()
btrfs: use single bulk copy operations when logging directories
This is patch 2/3 and performance results, and the specific tests, are
included in the changelog of patch 3/3.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
When inserting a batch of items into a btree, we end up looping over the
data sizes array 3 times:
1) Once in the caller of btrfs_insert_empty_items(), when it populates the
array with the data sizes for each item;
2) Once at btrfs_insert_empty_items() to sum the elements of the data
sizes array and compute the total data size;
3) And then once again at setup_items_for_insert(), where we do exactly
the same as what we do at btrfs_insert_empty_items(), to compute the
total data size.
That is not bad for small arrays, but when the arrays have hundreds of
elements, the time spent on looping is not negligible. For example when
doing batch inserts of delayed items for dir index items or when logging
a directory, it's common to have 200 to 260 dir index items in a single
batch when using a leaf size of 16K and using file names between 8 and 12
characters. For a 64K leaf size, multiply that by 4. Taking into account
that during directory logging or when flushing delayed dir index items we
can have many of those large batches, the time spent on the looping adds
up quickly.
It's also more important to avoid it at setup_items_for_insert(), since
we are holding a write lock on a leaf and, in some cases, on upper nodes
of the btree, which causes us to block other tasks that want to access
the leaf and nodes for longer than necessary.
So change the code so that setup_items_for_insert() and
btrfs_insert_empty_items() no longer compute the total data size, and
instead rely on the caller to supply it. This makes us loop over the
array only once, where we can both populate the data size array and
compute the total data size, taking advantage of spatial and temporal
locality. To make this more manageable, use a structure to contain
all the relevant details for a batch of items (keys array, data sizes
array, total data size, number of items), and use it as an argument
for btrfs_insert_empty_items() and setup_items_for_insert().
This patch is part of a small patchset that is comprised of the following
patches:
btrfs: loop only once over data sizes array when inserting an item batch
btrfs: unexport setup_items_for_insert()
btrfs: use single bulk copy operations when logging directories
This is patch 1/3 and performance results, and the specific tests, are
included in the changelog of patch 3/3.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
