Add a helper to complete an ordered_extent without first doing a lookup.
The tracepoint cannot use the ordered_extent class as we also want to
print the range.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Factor out a helper to queue up an ordered_extent completion in a work
queue. This new helper will later be used complete an ordered_extent
without first doing a lookup.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Factor out a helper from btrfs_mark_ordered_io_finished that does the
actual per-ordered_extent work to check if we want to schedule an I/O
completion. This new helper will later be used complete an
ordered_extent without first doing a lookup.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Use the ordered_extent pointer in the btrfs_bio instead of looking it
up manually.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Add a pointer to the ordered_extent to the existing union in struct
btrfs_bio, so all code dealing with data write bios can just use a
pointer dereference to retrieve the ordered_extent instead of doing
multiple rbtree lookups per I/O.
The reference to this ordered_extent is dropped at end I/O time,
which implies that an extra one must be acquired when the bio is split.
This also requires moving the btrfs_extract_ordered_extent call into
btrfs_split_bio so that the invariant of always having a valid
ordered_extent reference for the btrfs_bio is kept.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
btrfs_dio_submit_io is the only place that uses btrfs_bio_end_io to end a
bio that hasn't been submitted using btrfs_submit_bio yet, and this
invariant will become a problem with upcoming changes to the btrfs bio
layer. Just open code the assignment of bi_status and the call to
btrfs_dio_end_io.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Add a helper to check for that a btrfs_bio has a valid inode, and that
it is a data inode to key off all the special handling for data path
checksumming. Note that this uses is_data_inode instead of REQ_META as
REQ_META is only set directly before submission in submit_one_bio and
we'll also want to use this helper for error handling where REQ_META
isn't set yet.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
All callers are gone now.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
btrfs_submit_compressed_write always operates on a single ordered_extent.
Make that explicit by using btrfs_alloc_ordered_extent in the callers
and passing the ordered_extent to btrfs_submit_compressed_write. This
will help with storing and ordered_extent pointer in the btrfs_bio in
subsequent patches.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Both callers of btrfs_reloc_clone_csums allocate the ordered_extent that
btrfs_reloc_clone_csums operates on. Switch them to use
btrfs_alloc_ordered_extent instead of btrfs_add_ordered_extent and
pass the ordered_extent to btrfs_reloc_clone_csums instead of doing an
extra lookup.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Refactor run_delalloc_nocow a little bit so that there is only a single
call to btrfs_add_ordered_extent instead of two.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Currently buffered writeback bios are allowed to span multiple
ordered_extents, although that basically never actually happens since
commit 4a445b7b61 ("btrfs: don't merge pages into bio if their page
offset is not contiguous").
Supporting bios than span ordered_extents complicates the file
checksumming code, and prevents us from adding an ordered_extent pointer
to the btrfs_bio structure. Use the existing code to limit a bio to
single ordered_extent for zoned device writes for all writes.
This allows to remove the REQ_BTRFS_ONE_ORDERED flags, and the
handling of multiple ordered_extents in btrfs_csum_one_bio.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
If a bio gets split, it needs to have a proper file_offset for checksum
validation and repair to work properly.
Based on feedback from Josef, commit 852eee62d3 ("btrfs: allow
btrfs_submit_bio to split bios") skipped this adjustment for ONE_ORDERED
bios. But if we actually ever need to split a ONE_ORDERED read bio, this
will lead to a wrong file offset in the repair code. Right now the only
user of the file_offset is logging of an error message so this is mostly
harmless, but the wrong offset might be more problematic for additional
users in the future.
Fixes: 852eee62d3 ("btrfs: allow btrfs_submit_bio to split bios")
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
The block group tree was not present among the lockdep classes. We could
get potentially lockdep warnings but so far none has been seen, also
because block-group-tree is a relatively new feature.
CC: stable@vger.kernel.org # 6.1+
Signed-off-by: David Sterba <dsterba@suse.com>
When extent_write_locked_range was originally added, it was only used
writing back compressed pages from an async helper thread. But it is
now also used for writing back pages on zoned devices, where it is
called directly from the ->writepage context. In this case we want to
be able to pass on the writeback_control instead of creating a new one,
and more importantly want to use all the normal cgroup interaction
instead of potentially deferring writeback to another helper.
Fixes: 898793d992 ("btrfs: zoned: write out partially allocated region")
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
__extent_writepage does a lot of things that make no sense for
extent_write_locked_range, given that extent_write_locked_range itself is
called from __extent_writepage either directly or through a workqueue,
and all this work has already been done in the first invocation and the
pages haven't been unlocked since. Call __extent_writepage_io directly
instead and open code the logic tracked in
btrfs_bio_ctrl::extent_locked.
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Move the nr_to_write accounting from __extent_writepage_io to
__extent_writepage_io as we'll grow another __extent_writepage_io that
doesn't want this accounting soon. Also drop the obsolete comment -
decrementing a counter in the on-stack writeback_control data structure
doesn't need the page lock.
Reviewed-by: Anand Jain <anand.jain@oracle.com>
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
__extent_writepage_io is never called for compressed or inline extents,
or holes. Remove the not quite working code for them and replace it with
asserts that these cases don't happen.
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Now that the btrfs writeback code has stopped using PageError, using
PAGE_SET_ERROR to just set the per-address_space error flag is confusing.
Open code the mapping_set_error calls in the callers and remove
the PAGE_SET_ERROR flag.
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
PageError is not used by the VFS/MM and deprecated because it uses up a
page bit and has no coherent rules. Instead read errors are usually
propagated by not setting or clearing the uptodate bit, and write errors
are propagated through the address_space. Btrfs now only sets the flag
and never clears it for data pages, so just remove all places setting it,
and the subpage error bit.
Note that the error propagation for superblock writes that work on the
block device mapping still uses PageError for now, but that will be
addressed in a separate series.
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
__extent_writepage currenly sets PageError whenever any error happens,
and the also checks for PageError to decide if to call error handling.
This leads to very unclear responsibility for cleaning up on errors.
In the VM and generic writeback helpers the basic idea is that once
I/O is fired off all error handling responsibility is delegated to the
end I/O handler. But if that end I/O handler sets the PageError bit,
and the submitter checks it, the bit could in some cases leak into the
submission context for fast enough I/O.
Fix this by simply not checking PageError and just using the local
ret variable to check for submission errors. This also fundamentally
solves the long problem documented in a comment in __extent_writepage
by never leaking the error bit into the submission context.
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
cow_file_range_async is only used for compressed writeback. Rename it
to run_delalloc_compressed, which also fits in with run_delalloc_nocow
and run_delalloc_zoned.
Reviewed-by: Anand Jain <anand.jain@oracle.com>
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
If cow_file_range_async fails to allocate the asynchronous writeback
context, it currently returns an error and entirely fails the writeback.
This is not a good idea as a writeback failure is a non-temporary error
condition that will make the file system unusable. Just fall back to
synchronous uncompressed writeback instead. This requires us to delay
setting the BTRFS_INODE_HAS_ASYNC_EXTENT flag until we've committed to
the async writeback.
The compression checks INODE_NOCOMPRESS and FORCE_COMPRESS are moved
from cow_file_range_async to the preceding checks btrfs_run_delalloc_range().
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
btrfs_verify_page is called from the readpage completion handler, which
is only used to read pages, or parts of pages that aren't uptodate yet.
The only case where PageError could be set on a page in btrfs is if we
had a previous writeback error, but in that case we won't called readpage
on it, as it has previously been marked uptodate.
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Also clear the uptodate bit to make sure the page isn't seen as uptodate
in the page cache if fsverity verification fails.
Fixes: 146054090b ("btrfs: initial fsverity support")
CC: stable@vger.kernel.org # 5.15+
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Split all the conditionals for the fsverity calls in end_page_read into
a btrfs_verify_page helper to keep the code readable and make additional
refactoring easier.
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
The range_end field in struct writeback_control is inclusive, just like
the end parameter passed to extent_write_locked_range. Not doing this
could cause extra writeout, which is harmless but suboptimal.
Fixes: 771ed689d2 ("Btrfs: Optimize compressed writeback and reads")
CC: stable@vger.kernel.org # 5.9+
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
There is a fairly unlikely race condition in tree mod log rewind that
can result in a kernel panic which has the following trace:
[530.569] BTRFS critical (device sda3): unable to find logical 0 length 4096
[530.585] BTRFS critical (device sda3): unable to find logical 0 length 4096
[530.602] BUG: kernel NULL pointer dereference, address: 0000000000000002
[530.618] #PF: supervisor read access in kernel mode
[530.629] #PF: error_code(0x0000) - not-present page
[530.641] PGD 0 P4D 0
[530.647] Oops: 0000 [#1] SMP
[530.654] CPU: 30 PID: 398973 Comm: below Kdump: loaded Tainted: G S O K 5.12.0-0_fbk13_clang_7455_gb24de3bdb045 #1
[530.680] Hardware name: Quanta Mono Lake-M.2 SATA 1HY9U9Z001G/Mono Lake-M.2 SATA, BIOS F20_3A15 08/16/2017
[530.703] RIP: 0010:__btrfs_map_block+0xaa/0xd00
[530.755] RSP: 0018:ffffc9002c2f7600 EFLAGS: 00010246
[530.767] RAX: ffffffffffffffea RBX: ffff888292e41000 RCX: f2702d8b8be15100
[530.784] RDX: ffff88885fda6fb8 RSI: ffff88885fd973c8 RDI: ffff88885fd973c8
[530.800] RBP: ffff888292e410d0 R08: ffffffff82fd7fd0 R09: 00000000fffeffff
[530.816] R10: ffffffff82e57fd0 R11: ffffffff82e57d70 R12: 0000000000000000
[530.832] R13: 0000000000001000 R14: 0000000000001000 R15: ffffc9002c2f76f0
[530.848] FS: 00007f38d64af000(0000) GS:ffff88885fd80000(0000) knlGS:0000000000000000
[530.866] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[530.880] CR2: 0000000000000002 CR3: 00000002b6770004 CR4: 00000000003706e0
[530.896] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[530.912] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[530.928] Call Trace:
[530.934] ? btrfs_printk+0x13b/0x18c
[530.943] ? btrfs_bio_counter_inc_blocked+0x3d/0x130
[530.955] btrfs_map_bio+0x75/0x330
[530.963] ? kmem_cache_alloc+0x12a/0x2d0
[530.973] ? btrfs_submit_metadata_bio+0x63/0x100
[530.984] btrfs_submit_metadata_bio+0xa4/0x100
[530.995] submit_extent_page+0x30f/0x360
[531.004] read_extent_buffer_pages+0x49e/0x6d0
[531.015] ? submit_extent_page+0x360/0x360
[531.025] btree_read_extent_buffer_pages+0x5f/0x150
[531.037] read_tree_block+0x37/0x60
[531.046] read_block_for_search+0x18b/0x410
[531.056] btrfs_search_old_slot+0x198/0x2f0
[531.066] resolve_indirect_ref+0xfe/0x6f0
[531.076] ? ulist_alloc+0x31/0x60
[531.084] ? kmem_cache_alloc_trace+0x12e/0x2b0
[531.095] find_parent_nodes+0x720/0x1830
[531.105] ? ulist_alloc+0x10/0x60
[531.113] iterate_extent_inodes+0xea/0x370
[531.123] ? btrfs_previous_extent_item+0x8f/0x110
[531.134] ? btrfs_search_path_in_tree+0x240/0x240
[531.146] iterate_inodes_from_logical+0x98/0xd0
[531.157] ? btrfs_search_path_in_tree+0x240/0x240
[531.168] btrfs_ioctl_logical_to_ino+0xd9/0x180
[531.179] btrfs_ioctl+0xe2/0x2eb0
This occurs when logical inode resolution takes a tree mod log sequence
number, and then while backref walking hits a rewind on a busy node
which has the following sequence of tree mod log operations (numbers
filled in from a specific example, but they are somewhat arbitrary)
REMOVE_WHILE_FREEING slot 532
REMOVE_WHILE_FREEING slot 531
REMOVE_WHILE_FREEING slot 530
...
REMOVE_WHILE_FREEING slot 0
REMOVE slot 455
REMOVE slot 454
REMOVE slot 453
...
REMOVE slot 0
ADD slot 455
ADD slot 454
ADD slot 453
...
ADD slot 0
MOVE src slot 0 -> dst slot 456 nritems 533
REMOVE slot 455
REMOVE slot 454
REMOVE slot 453
...
REMOVE slot 0
When this sequence gets applied via btrfs_tree_mod_log_rewind, it
allocates a fresh rewind eb, and first inserts the correct key info for
the 533 elements, then overwrites the first 456 of them, then decrements
the count by 456 via the add ops, then rewinds the move by doing a
memmove from 456:988->0:532. We have never written anything past 532, so
that memmove writes garbage into the 0:532 range. In practice, this
results in a lot of fully 0 keys. The rewind then puts valid keys into
slots 0:455 with the last removes, but 456:532 are still invalid.
When search_old_slot uses this eb, if it uses one of those invalid
slots, it can then read the extent buffer and issue a bio for offset 0
which ultimately panics looking up extent mappings.
This bad tree mod log sequence gets generated when the node balancing
code happens to do a balance_node_right followed by a push_node_left
while logging in the tree mod log. Illustrated for ebs L and R (left and
right):
L R
start:
[XXX|YYY|...] [ZZZ|...|...]
balance_node_right:
[XXX|YYY|...] [...|ZZZ|...] move Z to make room for Y
[XXX|...|...] [YYY|ZZZ|...] copy Y from L to R
push_node_left:
[XXX|YYY|...] [...|ZZZ|...] copy Y from R to L
[XXX|YYY|...] [ZZZ|...|...] move Z into emptied space (NOT LOGGED!)
This is because balance_node_right logs a move, but push_node_left
explicitly doesn't. That is because logging the move would remove the
overwritten src < dst range in the right eb, which was already logged
when we called btrfs_tree_mod_log_eb_copy. The correct sequence would
include a move from 456:988 to 0:532 after remove 0:455 and before
removing 0:532. Reversing that sequence would entail creating keys for
0:532, then moving those keys out to 456:988, then creating more keys
for 0:455.
i.e.,
REMOVE_WHILE_FREEING slot 532
REMOVE_WHILE_FREEING slot 531
REMOVE_WHILE_FREEING slot 530
...
REMOVE_WHILE_FREEING slot 0
MOVE src slot 456 -> dst slot 0 nritems 533
REMOVE slot 455
REMOVE slot 454
REMOVE slot 453
...
REMOVE slot 0
ADD slot 455
ADD slot 454
ADD slot 453
...
ADD slot 0
MOVE src slot 0 -> dst slot 456 nritems 533
REMOVE slot 455
REMOVE slot 454
REMOVE slot 453
...
REMOVE slot 0
Fix this to log the move but avoid the double remove by putting all the
logging logic in btrfs_tree_mod_log_eb_copy which has enough information
to detect these cases and properly log moves, removes, and adds. Leave
btrfs_tree_mod_log_insert_move to handle insert_ptr and delete_ptr's
tree mod logging.
(Un)fortunately, this is quite difficult to reproduce, and I was only
able to reproduce it by adding sleeps in btrfs_search_old_slot that
would encourage more log rewinding during ino_to_logical ioctls. I was
able to hit the warning in the previous patch in the series without the
fix quite quickly, but not after this patch.
CC: stable@vger.kernel.org # 5.15+
Reviewed-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Boris Burkov <boris@bur.io>
Signed-off-by: David Sterba <dsterba@suse.com>
The way that tree mod log tracks the ultimate length of the eb, the
variable 'n', eventually turns up the correct value, but at intermediate
steps during the rewind, n can be inaccurate as a representation of the
end of the eb. For example, it doesn't get updated on move rewinds, and
it does get updated for add/remove in the middle of the eb.
To detect cases with invalid moves, introduce a separate variable called
max_slot which tries to track the maximum valid slot in the rewind eb.
We can then warn if we do a move whose src range goes beyond the max
valid slot.
There is a commented caveat that it is possible to have this value be an
overestimate due to the challenge of properly handling 'add' operations
in the middle of the eb, but in practice it doesn't cause enough of a
problem to throw out the max idea in favor of tracking every valid slot.
CC: stable@vger.kernel.org # 5.15+
Reviewed-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Boris Burkov <boris@bur.io>
Signed-off-by: David Sterba <dsterba@suse.com>
The workspaces for compression are typically much larger than a page and
for high zstd levels in the range of megabytes. There's a fallback to
vmalloc but this can still fail (see the report).
Some of the workspaces are preallocated at module load time so we have a
safe fallback, otherwise when a new workspace is needed it's allocated
but if this fails then the process waits. Which means the warning is
only causing noise and we can use the GFP flag to disable it.
Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=217466
Signed-off-by: David Sterba <dsterba@suse.com>
need_full_stripe is just a somewhat complicated way to say
"op != BTRFS_MAP_READ". Just spell that explicit check out, which makes
a lot of the code currently using the helper easier to understand.
Reviewed-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
btrfs_map_sblock just hard codes three arguments and calls
btrfs_map_sblock. Remove it as it doesn't provide any real value, but
makes following the btrfs_map_block call chains harder.
Reviewed-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Now that the old btrfs_map_block is gone, drop the leading underscores
from __btrfs_map_block.
Reviewed-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
There are no users of btrfs_map_block left, so remove it.
Reviewed-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Pass a smap into __btrfs_map_block so that the usual case of a read that
doesn't require parity raid recovery doesn't need an extra memory
allocation for the btrfs_io_context.
Reviewed-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
BTRFS_MAP_DISCARD is never set, as REQ_OP_DISCARD is never passed to
btrfs_op() only only checked in two ASSERTS.
Remove it and let the catchall WARN_ON in btrfs_op() deal with accidental
REQ_OP_DISCARDs leaked into btrfs_op(). Last use was in a4012f06f1
("btrfs: split discard handling out of btrfs_map_block").
Reviewed-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
The implementation of XXHASH is now CPU only but still fast enough to be
considered for the synchronous checksumming, like non-generic crc32c.
A userspace benchmark comparing it to various implementations (patched
hash-speedtest from btrfs-progs):
Block size: 4096
Iterations: 1000000
Implementation: builtin
Units: CPU cycles
NULL-NOP: cycles: 73384294, cycles/i 73
NULL-MEMCPY: cycles: 228033868, cycles/i 228, 61664.320 MiB/s
CRC32C-ref: cycles: 24758559416, cycles/i 24758, 567.950 MiB/s
CRC32C-NI: cycles: 1194350470, cycles/i 1194, 11773.433 MiB/s
CRC32C-ADLERSW: cycles: 6150186216, cycles/i 6150, 2286.372 MiB/s
CRC32C-ADLERHW: cycles: 626979180, cycles/i 626, 22427.453 MiB/s
CRC32C-PCL: cycles: 466746732, cycles/i 466, 30126.699 MiB/s
XXHASH: cycles: 860656400, cycles/i 860, 16338.188 MiB/s
Comparing purely software implementation (ref), current outdated
accelerated using crc32q instruction (NI), optimized implementations by
M. Adler (https://stackoverflow.com/questions/17645167/implementing-sse-4-2s-crc32c-in-software/17646775#17646775)
and the best one that was taken from kernel using the PCLMULQDQ
instruction (PCL).
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: David Sterba <dsterba@suse.com>
split_extent_map splits off the first chunk of an extent map into a new
one. One of the two users is the zoned I/O completion code that wants to
rewrite the logical block start address right after this split. Pass in
the logical address to be set in the split off first extent_map as an
argument to avoid an extra extent tree lookup for this case.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: David Sterba <dsterba@suse.com>
The btrfs zoned completion code currently needs an ordered_extent and
extent_map per bio so that it can account for the non-predictable
write location from Zone Append. To archive that it currently splits
the ordered_extent and extent_map at I/O submission time, and then
records the actual physical address in the ->physical field of the
ordered_extent.
This patch instead switches to record the "original" physical address
that the btrfs allocator assigned in spare space in the btrfs_bio,
and then rewrites the logical address in the btrfs_ordered_sum
structure at I/O completion time. This allows the ordered extent
completion handler to simply walk the list of ordered csums and
split the ordered extent as needed. This removes an extra ordered
extent and extent_map lookup and manipulation during the I/O
submission path, and instead batches it in the I/O completion path
where we need to touch these anyway.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: David Sterba <dsterba@suse.com>
To delay splitting ordered_extents to I/O completion time we need to be
able to handle fully completed ordered extents in
btrfs_split_ordered_extent. Besides a bit of accounting this primarily
involved moving over the csums to the split bio for the range that it
covers, which is simple enough because we always have one
btrfs_ordered_sum per bio.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: David Sterba <dsterba@suse.com>
Currently there is a small race window in btrfs_split_ordered_extent,
where the reduced old extent can be looked up on the per-inode rbtree
or the per-root list while the newly split out one isn't visible yet.
Fix this by open coding btrfs_alloc_ordered_extent in
btrfs_split_ordered_extent, and holding the tree lock and
root->ordered_extent_lock over the entire tree and extent manipulation.
Note that this introduces new lock ordering because previously
ordered_extent_lock was never held over the tree lock.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Split two low-level helpers out of btrfs_alloc_ordered_extent to allocate
and insert the logic extent. The pure alloc helper will be used to
improve btrfs_split_ordered_extent.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Return the ordered_extent split from the passed in one. This will be
needed to be able to store an ordered_extent in the btrfs_bio.
Reviewed-by: Naohiro Aota <naohiro.aota@wdc.com>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
There is no good reason for doing one before the other in terms of
failure implications, but doing the extent_map split first will
simplify some upcoming refactoring.
Reviewed-by: Naohiro Aota <naohiro.aota@wdc.com>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
split_extent_map doesn't have anything to do with the other code in
inode.c, so move it to extent_map.c.
This also allows marking replace_extent_mapping static.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
btrfs_submit_dev_bio is also called for clone bios that aren't embedded
into a btrfs_bio structure, but previous commit "btrfs: optimize the
logical to physical mapping for zoned writes" added code to assign
btrfs_bio.orig_physical in it.
This is harmless right now as only the single data profile can be used
on zoned devices, but will blow up when the RAID stripe tree is added.
Move it out into the single I/O specific branch in the caller.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: David Sterba <dsterba@suse.com>
The current code to store the final logical to physical mapping for a
zone append write in the extent tree is rather inefficient. It first has
to split the ordered extent so that there is one ordered extent per bio,
so that it can look up the ordered extent on I/O completion in
btrfs_record_physical_zoned and store the physical LBA returned by the
block driver in the ordered extent.
btrfs_rewrite_logical_zoned then has to do a lookup in the chunk tree to
see what physical address the logical address for this bio / ordered
extent is mapped to, and then rewrite it in the extent tree.
To optimize this process, we can store the physical address assigned in
the chunk tree to the original logical address and a pointer to
btrfs_ordered_sum structure the in the btrfs_bio structure, and then use
this information to rewrite the logical address in the btrfs_ordered_sum
structure directly at I/O completion time in btrfs_record_physical_zoned.
btrfs_rewrite_logical_zoned then simply updates the logical address in
the extent tree and the ordered_extent itself.
The code in btrfs_rewrite_logical_zoned now runs for all data I/O
completions in zoned file systems, which is fine as there is no remapping
to do for non-append writes to conventional zones or for relocation, and
the overhead for quickly breaking out of the loop is very low.
Because zoned file systems now need the ordered_sums structure to
record the actual write location returned by zone append, allocate dummy
structures without the csum array for them when the I/O doesn't use
checksums, and free them when completing the ordered_extent.
Note that the btrfs_bio doesn't grow as the new field are places into
a union that is so far not used for data writes and has plenty of space
left in it.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
btrfs_ordered_sum::bytendr stores a logical address. Make that clear by
renaming it to ->logical.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
len can't ever be negative, so mark it as an u32 instead of int.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.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 a zoned append command fails there is no written address reported,
so don't try to record it.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Add an IS_ENABLED check for CONFIG_BLK_DEV_ZONED in addition to the
run-time check for the zone size. This will allow to make use of
compiler dead code elimination for code guarded by btrfs_is_zoned, and
for example provide just a dangling prototype for a function instead of
adding a stub.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
btrfs_destroy_delayed_refs() always returns 0 and its single caller does
not check its return value, as it also returns void, and so does the
callers' caller and so on. This is because we are in the transaction abort
path, where we have no way to deal with errors (we are in a critical
situation) and all cleanup of resources works in a best effort fashion.
So make btrfs_destroy_delayed_refs() return void.
Reviewed-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
We have a few static functions at disk-io.c for which we have a forward
declaration of their prototype, but it's not needed because all those
functions are defined before they are called, so remove them.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
At init_delayed_ref_head(), we are using two separate if statements to
check the delayed ref head action, and initializing 'must_insert_reserved'
to false twice, once when the variable is declared and once again in an
else branch.
Make this simpler and more straightforward by having a single switch
statement, also moving the comment about a drop action to the
corresponding switch case to make it more clear and eliminating the
duplicated initialization of 'must_insert_reserved' to false.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
There's no point in have several fields defined as 1 bit unsigned int in
struct btrfs_delayed_ref_head, we can instead use a bool type, it makes
the code a bit more readable and it doesn't change the structure size.
So switch them to proper booleans.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
The function btrfs_select_ref_head() iterates over the red black tree of
delayed reference heads, which is protected by the spinlock in the delayed
refs root. The function doesn't take the lock, it's taken by its single
caller, btrfs_obtain_ref_head(), because it needs to call that function
and btrfs_delayed_ref_lock() in the same critical section (delimited by
that spinlock). So assert at btrfs_select_ref_head() that we are holding
the expected lock.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
At insert_delayed_ref() there's no point of having a label and goto in the
case we were able to insert the delayed ref head. We can just add the code
under label to the if statement's body and return immediately, and also
there is no need to track the return value in a variable, we can just
return a literal true or false value directly. So do those changes.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
insert_delayed_ref() can only return 0 or 1, to indicate if the given
delayed reference was added to the head reference or if it was merged
into an existing delayed ref, respectively. So just make it return a
boolean instead.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
We are using an integer as a boolean to track the qgroup record insertion
status when adding a delayed reference head. Since all we need is a
boolean, switch the type from int to bool to make it more obvious.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
The 'in_tree' field is really not needed in struct btrfs_delayed_ref_node,
as we can check whether a reference is in the tree or not simply by
checking its red black tree node member with RB_EMPTY_NODE(), as when we
remove it from the tree we always call RB_CLEAR_NODE(). So remove that
field and use RB_EMPTY_NODE().
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
The 'is_head' field of struct btrfs_delayed_ref_node is no longer after
commit d278850eff ("btrfs: remove delayed_ref_node from ref_head"),
so remove it.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
In function end_bio_extent_readpage() we call
endio_readpage_release_extent() to unlock the extent io tree.
However we pass PageUptodate(page) as @uptodate parameter for it, while
for previous end_page_read() call, we use a dedicated @uptodate local
variable.
This is not a big deal, as even for subpage cases, either the bio only
covers part of the page, then the @uptodate is always false, and the
subpage ranges can still be merged.
But for the sake of consistency, always use @uptodate variable when
possible.
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Anand Jain <anand.jain@oracle.com>
Signed-off-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Currently alloc_extent_buffer() would make the extent buffer uptodate if
the corresponding pages are also uptodate.
But this check is only checking PageUptodate, which is fine for regular
cases, but not for subpage cases, as we can have multiple extent buffers
in the same page.
So here we go btrfs_page_test_uptodate() instead.
The old code doesn't cause any problem, but is not efficient, as it
would cause extra metadata read even if the range is already uptodate.
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Assertions reports are split into two parts, the exact file and location
of the condition and then the stack trace printed from
btrfs_assertfail(). This means all the stack traces report the same line
and this is what's typically reported by various tools, making it harder
to distinguish the reports.
[403.2467] assertion failed: refcount_read(&block_group->refs) == 1, in fs/btrfs/block-group.c:4259
[403.2479] ------------[ cut here ]------------
[403.2484] kernel BUG at fs/btrfs/messages.c:259!
[403.2488] invalid opcode: 0000 [#1] PREEMPT SMP KASAN
[403.2493] CPU: 2 PID: 23202 Comm: umount Not tainted 6.2.0-rc4-default+ #67
[403.2499] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.16.0-0-gd239552-rebuilt.opensuse.org 04/01/2014
[403.2509] RIP: 0010:btrfs_assertfail+0x19/0x1b [btrfs]
...
[403.2595] Call Trace:
[403.2598] <TASK>
[403.2601] btrfs_free_block_groups.cold+0x52/0xae [btrfs]
[403.2608] close_ctree+0x6c2/0x761 [btrfs]
[403.2613] ? __wait_for_common+0x2b8/0x360
[403.2618] ? btrfs_cleanup_one_transaction.cold+0x7a/0x7a [btrfs]
[403.2626] ? mark_held_locks+0x6b/0x90
[403.2630] ? lockdep_hardirqs_on_prepare+0x13d/0x200
[403.2636] ? __call_rcu_common.constprop.0+0x1ea/0x3d0
[403.2642] ? trace_hardirqs_on+0x2d/0x110
[403.2646] ? __call_rcu_common.constprop.0+0x1ea/0x3d0
[403.2652] generic_shutdown_super+0xb0/0x1c0
[403.2657] kill_anon_super+0x1e/0x40
[403.2662] btrfs_kill_super+0x25/0x30 [btrfs]
[403.2668] deactivate_locked_super+0x4c/0xc0
By making btrfs_assertfail a macro we'll get the same line number for
the BUG output:
[63.5736] assertion failed: 0, in fs/btrfs/super.c:1572
[63.5758] ------------[ cut here ]------------
[63.5782] kernel BUG at fs/btrfs/super.c:1572!
[63.5807] invalid opcode: 0000 [#2] PREEMPT SMP KASAN
[63.5831] CPU: 0 PID: 859 Comm: mount Tainted: G D 6.3.0-rc7-default+ #2062
[63.5868] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.14.0-0-g155821a-rebuilt.opensuse.org 04/01/2014
[63.5905] RIP: 0010:btrfs_mount+0x24/0x30 [btrfs]
[63.5964] RSP: 0018:ffff88800e69fcd8 EFLAGS: 00010246
[63.5982] RAX: 000000000000002d RBX: ffff888008fc1400 RCX: 0000000000000000
[63.6004] RDX: 0000000000000000 RSI: ffffffffb90fd868 RDI: ffffffffbcc3ff20
[63.6026] RBP: ffffffffc081b200 R08: 0000000000000001 R09: ffff88800e69fa27
[63.6046] R10: ffffed1001cd3f44 R11: 0000000000000001 R12: ffff888005a3c370
[63.6062] R13: ffffffffc058e830 R14: 0000000000000000 R15: 00000000ffffffff
[63.6081] FS: 00007f7b3561f800(0000) GS:ffff88806c600000(0000) knlGS:0000000000000000
[63.6105] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[63.6120] CR2: 00007fff83726e10 CR3: 0000000002a9e000 CR4: 00000000000006b0
[63.6137] Call Trace:
[63.6143] <TASK>
[63.6148] legacy_get_tree+0x80/0xd0
[63.6158] vfs_get_tree+0x43/0x120
[63.6166] do_new_mount+0x1f3/0x3d0
[63.6176] ? do_add_mount+0x140/0x140
[63.6187] ? cap_capable+0xa4/0xe0
[63.6197] path_mount+0x223/0xc10
This comes at a cost of bloating the final btrfs.ko module due all the
inlining, as long as assertions are compiled in. This is a must for
debugging builds but this is often enabled on release builds too.
Release build:
text data bss dec hex filename
1251676 20317 16088 1288081 13a791 pre/btrfs.ko
1260612 29473 16088 1306173 13ee3d post/btrfs.ko
DELTA: +8936
CC: Josh Poimboeuf <jpoimboe@kernel.org>
Signed-off-by: David Sterba <dsterba@suse.com>
There is a bug report that assert_eb_page_uptodate() gets triggered for
free space tree metadata.
Without proper dump for the subpage bitmaps it's much harder to debug.
Thus this patch would dump all the subpage bitmaps (split them into
their own bitmaps) for a easier debugging.
The output would look like this:
(Dumped after a tree block got read from disk)
page:000000006e34bf49 refcount:4 mapcount:0 mapping:0000000067661ac4 index:0x1d1 pfn:0x110e9
memcg:ffff0000d7d62000
aops:btree_aops [btrfs] ino:1
flags: 0x8000000000002002(referenced|private|zone=2)
page_type: 0xffffffff()
raw: 8000000000002002 0000000000000000 dead000000000122 ffff00000188bed0
raw: 00000000000001d1 ffff0000c7992700 00000004ffffffff ffff0000d7d62000
page dumped because: btrfs subpage dump
BTRFS warning (device dm-1): start=30490624 len=16384 page=30474240 bitmaps: uptodate=4-7 error= dirty= writeback= ordered= checked=
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
BACKGROUND
==========
When multiple work items are queued to a workqueue, their execution order
doesn't match the queueing order. They may get executed in any order and
simultaneously. When fully serialized execution - one by one in the queueing
order - is needed, an ordered workqueue should be used which can be created
with alloc_ordered_workqueue().
However, alloc_ordered_workqueue() was a later addition. Before it, an
ordered workqueue could be obtained by creating an UNBOUND workqueue with
@max_active==1. This originally was an implementation side-effect which was
broken by 4c16bd327c ("workqueue: restore WQ_UNBOUND/max_active==1 to be
ordered"). Because there were users that depended on the ordered execution,
5c0338c687 ("workqueue: restore WQ_UNBOUND/max_active==1 to be ordered")
made workqueue allocation path to implicitly promote UNBOUND workqueues w/
@max_active==1 to ordered workqueues.
While this has worked okay, overloading the UNBOUND allocation interface
this way creates other issues. It's difficult to tell whether a given
workqueue actually needs to be ordered and users that legitimately want a
min concurrency level wq unexpectedly gets an ordered one instead. With
planned UNBOUND workqueue updates to improve execution locality and more
prevalence of chiplet designs which can benefit from such improvements, this
isn't a state we wanna be in forever.
This patch series audits all call sites that create an UNBOUND workqueue w/
@max_active==1 and converts them to alloc_ordered_workqueue() as necessary.
BTRFS
=====
* fs_info->scrub_workers initialized in scrub_workers_get() was setting
@max_active to 1 when @is_dev_replace is set and it seems that the
workqueue actually needs to be ordered if @is_dev_replace. Update the code
so that alloc_ordered_workqueue() is used if @is_dev_replace.
* fs_info->discard_ctl.discard_workers initialized in
btrfs_init_workqueues() was directly using alloc_workqueue() w/
@max_active==1. Converted to alloc_ordered_workqueue().
* fs_info->fixup_workers and fs_info->qgroup_rescan_workers initialized in
btrfs_queue_work() use the btrfs's workqueue wrapper, btrfs_workqueue,
which are allocated with btrfs_alloc_workqueue().
btrfs_workqueue implements automatic @max_active adjustment which is
disabled when the specified max limit is below a certain threshold, so
calling btrfs_alloc_workqueue() with @limit_active==1 yields an ordered
workqueue whose @max_active won't be changed as the auto-tuning is
disabled.
This is rather brittle in that nothing clearly indicates that the two
workqueues should be ordered or btrfs_alloc_workqueue() must disable
auto-tuning when @limit_active==1.
This patch factors out the common btrfs_workqueue init code into
btrfs_init_workqueue() and add explicit btrfs_alloc_ordered_workqueue().
The two workqueues are converted to use the new ordered allocation
interface.
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: David Sterba <dsterba@suse.com>
Now that all extent state bit helpers effectively take the GFP_NOFS mask
(and GFP_NOWAIT is encoded in the bits) we can remove the parameter.
This reduces stack consumption in many functions and simplifies a lot of
code.
Net effect on module on a release build:
text data bss dec hex filename
1250432 20985 16088 1287505 13a551 pre/btrfs.ko
1247074 20985 16088 1284147 139833 post/btrfs.ko
DELTA: -3358
Signed-off-by: David Sterba <dsterba@suse.com>
The only flags we now pass to set_extent_bit/__clear_extent_bit are
GFP_NOFS and GFP_NOWAIT (a few functions handling mappings). This
requires an extra parameter to be passed everywhere but is almost always
the same.
Encode the GFP_NOWAIT as an artificial extent bit and extract the
real bits and gfp mask in the lowest level helpers. Now the passed
gfp mask is not actually used and can be removed.
Signed-off-by: David Sterba <dsterba@suse.com>
The __GFP_NOFAIL passed to set_extent_bit first appeared in 2010
(commit f0486c68e4 ("Btrfs: Introduce contexts for metadata
reservation")), without any explanation why it would be needed.
Meanwhile we've updated the semantics of set_extent_bit to handle failed
allocations and do unlock, sleep and retry if needed. The use of the
NOFAIL flag is also an outlier, we never want any of the set/clear
extent bit helpers to fail, they're used for many critical changes like
extent locking, besides the extent state bit changes.
Signed-off-by: David Sterba <dsterba@suse.com>
This helper calls set_extent_bit with two more parameters set to default
values, but otherwise it's purpose is not clear.
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
The helper only passes GFP_NOWAIT as gfp flags and is used two times.
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
The helper is used a few times, that it's setting the DIRTY extent bit
is still clear.
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
The helper is used only once.
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
The helper is used once in fs code and a few times in the self test
code.
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
The helper is used only once.
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
btrfs_grab_root already checks for a NULL root itself.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Use a switch statement instead of an endless chain of if statements
to make the code a little cleaner.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
btrfs_grab_root returns either the root or NULL, and the callers of
btrfs_get_global_root expect it to return the same. But all the more
recently added roots instead return an ERR_PTR, so fix this.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
There are three ways the fsid is validated in btrfs_validate_super():
- verify that super_copy::fsid is the same as fs_devices::fsid
- if the metadata_uuid flag is set, verify if super_copy::metadata_uuid
and fs_devices::metadata_uuid are the same.
- a few lines below, often missed out, verify if dev_item::fsid is the
same as fs_devices::metadata_uuid.
The function btrfs_validate_super() contains multiple if-statements with
memcmp() to check UUIDs. This patch consolidates them into a single
location.
Signed-off-by: Anand Jain <anand.jain@oracle.com>
Signed-off-by: David Sterba <dsterba@suse.com>
We often check if the metadata_uuid is not the same as fsid, and then we
check if the given fsid matches the metadata_uuid. This patch refactors
this logic into function match_fsid_changed and utilize it.
Signed-off-by: Anand Jain <anand.jain@oracle.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Refactor the functions find_fsid() and find_fsid_with_metadata_uuid(),
as they currently share a common set of code to compare the fsid and
metadata_uuid. Create a common helper function, match_fsid_fs_devices().
Signed-off-by: Anand Jain <anand.jain@oracle.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Simplify the return type of check_tree_block_fsid() from int (1 or 0) to
bool. Its only user is interested in knowing the success or failure.
Signed-off-by: Anand Jain <anand.jain@oracle.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Add comment about metadata_uuid in btrfs_fs_devices.
No functional change.
Signed-off-by: Anand Jain <anand.jain@oracle.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Simplify has_metadata_uuid checks - by localizing the has_metadata_uuid
checked within alloc_fs_devices()'s second argument, it improves the
code readability.
Signed-off-by: Anand Jain <anand.jain@oracle.com>
Signed-off-by: David Sterba <dsterba@suse.com>
We currently have redundant checks for the non-null value of fsid
simplify it.
And, no one is using alloc_fs_devices() with a NULL metadata_uuid
while fsid is not NULL, add an assert() to verify this condition.
Signed-off-by: Anand Jain <anand.jain@oracle.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Pack bool fsid_change and bool seeding with other bool declarations in the
struct btrfs_fs_devices, approximately 6 bytes is saved, depending on
the config.
before: 512 bytes
after: 496 bytes
Signed-off-by: Anand Jain <anand.jain@oracle.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Most of the code in write_one_subpage_eb and write_one_eb is shared,
so merge the two functions into one.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Instead of locking and unlocking every page or the extent, just add a
new EXTENT_BUFFER_READING bit that mirrors EXTENT_BUFFER_WRITEBACK
for synchronizing threads trying to read an extent_buffer and to wait
for I/O completion.
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
The only other place that locks extents on the btree inode is
read_extent_buffer_subpage while reading in the partial page for a
buffer. This means locking the extent in btrfs_buffer_uptodate does not
synchronize with anything on non-subpage file systems, and on subpage
file systems it only waits for a parallel read(-ahead) to finish,
which seems to be counter to what the callers actually expect.
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
The only place that reads in pages and thus marks them uptodate for
the btree inode is read_extent_buffer_pages. Which means that either
pages are already uptodate from an old buffer when creating a new
one in alloc_extent_buffer, or they will be updated by ca call
to read_extent_buffer_pages. This means the checks for uptodate
pages in read_extent_buffer_pages and read_extent_buffer_subpage are
superfluous and can be removed.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
PageError is only used to limit the uptodate check in
assert_eb_page_uptodate. But we have a much more useful flag indicating
the exact condition we are about with the EXTENT_BUFFER_WRITE_ERR flag,
so use that instead and help the kernel toward eventually removing
PageError.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
No need to track the number of pages under I/O now that each
extent_buffer is read and written using a single bio. For the
read side we need to grab an extra reference for the duration of
the I/O to prevent eviction, though.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
The checksumming of btree blocks always operates on the entire
extent_buffer, and because btree blocks are always allocated contiguously
on disk they are never split by btrfs_submit_bio.
Simplify the checksumming code by finding the extent_buffer in the
btrfs_bio private data instead of trying to search through the bio_vec.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Now that we always use a single bio to write an extent_buffer, the buffer
can be passed to the end_io handler as private data. This allows
to simplify the metadata write end I/O handler, and merge the subpage
end_io handler into the main one.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
The btrfs_bio_ctrl machinery is overkill for writing extent_buffers
as we always operate on PAGE_SIZE chunks (or one smaller one for the
subpage case) that are contiguous and are guaranteed to fit into a
single bio. Replace it with open coded btrfs_bio_alloc, __bio_add_page
and btrfs_submit_bio calls.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Locking the pages in lock_extent_buffer_for_io only for the non-subpage
case is very confusing. Move it to write_one_eb to mirror the subpage
case and simplify the code. Now lock_extent_buffer_for_io does not leave
all the pages locked and each is individually locked/unlocked in
write_one_eb.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Stop trying to cluster writes of multiple extent_buffers into a single
bio. There is no need for that as the blk_plug mechanism used all the
way up in writeback_inodes_wb gives us the same I/O pattern even with
multiple bios. Removing the clustering simplifies
lock_extent_buffer_for_io a lot and will also allow passing the eb
as private data to the end I/O handler.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
lock_extent_buffer_for_io never returns a negative error value, so switch
the return value to a simple bool.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
[ keep noinline_for_stack ]
Signed-off-by: David Sterba <dsterba@suse.com>
Only subpage metadata reads lock the extent. Don't try to unlock it and
waste cycles in the extent tree lookup for PAGE_SIZE or larger metadata.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: David Sterba <dsterba@suse.com>
Now that we always use a single bio to read an extent_buffer, the buffer
can be passed to the end_io handler as private data. This allows
implementing a much simplified dedicated end I/O handler for metadata
reads.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: David Sterba <dsterba@suse.com>
Given that read recovery for data I/O is handled in the storage layer,
the mirror_num argument to btrfs_submit_compressed_read is always 0,
so remove it.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
The btrfs_bio_ctrl machinery is overkill for reading extent_buffers
as we always operate on PAGE_SIZE chunks (or one smaller one for the
subpage case) that are contiguous and are guaranteed to fit into a
single bio. Replace it with open coded btrfs_bio_alloc, __bio_add_page
and btrfs_submit_bio calls in a helper function shared between
the subpage and node size >= PAGE_SIZE cases.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Currently read_extent_buffer_pages skips pages that are already uptodate
when reading in an extent_buffer. While this reduces the amount of data
read, it increases the number of I/O operations as we now need to do
multiple I/Os when reading an extent buffer with one or more uptodate
pages in the middle of it. On any modern storage device, be that hard
drives or SSDs this actually decreases I/O performance. Fortunately
this case is pretty rare as the pages are always initially read together
and then aged the same way. Besides simplifying the code a bit as-is
this will allow for major simplifications to the I/O completion handler
later on.
Note that the case where all pages are uptodate is still handled by an
optimized fast path that does not read any data from disk.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Qu Wenruo <wqu@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>
verify_parent_transid is only called by btrfs_buffer_uptodate, which
confusingly inverts the return value. Merge the two functions and
reflow the parent_transid so that error handling is in a branch.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Setting the buffer uptodate in a function that is named as a validation
helper is a it confusing. Move the call from validate_extent_buffer to
the one of its two callers that didn't already have a duplicate call
to set_extent_buffer_uptodate.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Call btrfs_page_clear_uptodate instead of ClearPageUptodate to properly
manage the uptodate bit for the subpage case.
Reported-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
extent_buffer_under_io is only used in extent_io.c, so mark it static.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Qu Wenruo <wqu@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>
There is an internal error report that scrub found an error in an orphan
inode's data.
However there are very limited ways to cleanup such orphan inodes:
- btrfs_start_pre_rw_mount()
This happens at either mount, or RO->RW switch.
This is not a viable solution for root fs which may not be unmounted
or RO mounted.
Furthermore this doesn't cover every subvolume, it only covers the
currently cached subvolumes.
- btrfs_lookup_dentry()
This happens when we first lookup the subvolume dentry.
But dentry can be cached thus it's not ensured to be triggered every
time.
- create_snapshot()
This only happens for the created snapshot, not the source one.
This means if we didn't trigger orphan items cleanup, there is really no
other way to manually trigger it. Add this step to the START_SYNC ioctl.
This is a slight change in the semantics of the ioctl but as sync can be
potentially slow and is usually paired with WAIT_SYNC ioctl.
The errors are not handled because the main point of the ioctl is the
async commit, orphan cleanup is a side effect.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
There's a comment at btrfs_init_new_buffer() that refers to a function
named btrfs_clean_tree_block(), however the function was renamed to
btrfs_clear_buffer_dirty() in commit 190a83391b ("btrfs: rename
btrfs_clean_tree_block to btrfs_clear_buffer_dirty"). So update the
comment to refer to the current name.
Reviewed-by: Anand Jain <anand.jain@oracle.com>
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
The for_rename argument of btrfs_record_unlink_dir() is defined as an
integer, but the argument is in fact used as a boolean. So change it to
a boolean to make its use more clear.
Reviewed-by: Anand Jain <anand.jain@oracle.com>
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
There's no point of having a label and goto at btrfs_record_unlink_dir()
because the function is trivial and can just return early if we are not
in a rename context. So remove the label and goto and instead return
early if we are not in a rename.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Update the comments at btrfs_record_unlink_dir() so that they mention
where new names are logged and where old names are removed. Also, while
at it make the width of the comments closer to 80 columns and capitalize
the sentences and finish them with punctuation.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
At btrfs_record_unlink_dir() we directly check the logged_trans field of
the given inodes to check if they were previously logged in the current
transaction, and if any of them were, then we can avoid setting the field
last_unlink_trans of the directory to the id of the current transaction if
we are in a rename path. Avoiding that can later prevent falling back to
a transaction commit if anyone attempts to log the directory.
However the logged_trans field, store in struct btrfs_inode, is transient,
not persisted in the inode item on its subvolume b+tree, so that means
that if an inode is evicted and then loaded again, its original value is
lost and it's reset to 0. So directly checking the logged_trans field can
lead to some false negative, and that only results in a performance impact
as mentioned before.
Instead of directly checking the logged_trans field of the inodes, use the
inode_logged() helper, which will check in the log tree if an inode was
logged before in case its logged_trans field has a value of 0. This way
we can avoid setting the directory inode's last_unlink_trans and cause
future logging attempts of it to fallback to transaction commits. The
following test script shows one example where this happens without this
patch:
$ cat test.sh
#!/bin/bash
DEV=/dev/nullb0
MNT=/mnt/nullb0
num_init_files=10000
num_new_files=10000
mkfs.btrfs -f $DEV
mount -o ssd $DEV $MNT
mkdir $MNT/testdir
for ((i = 1; i <= $num_init_files; i++)); do
echo -n > $MNT/testdir/file_$i
done
echo -n > $MNT/testdir/foo
sync
# Add some files so that there's more work in the transaction other
# than just renaming file foo.
for ((i = 1; i <= $num_new_files; i++)); do
echo -n > $MNT/testdir/new_file_$i
done
# Change the file, fsync it.
setfattr -n user.x1 -v 123 $MNT/testdir/foo
xfs_io -c "fsync" $MNT/testdir/foo
# Now triggger eviction of file foo but no eviction for our test
# directory, since it is being used by the process below. This will
# set logged_trans of the file's inode to 0 once it is loaded again.
(
cd $MNT/testdir
while true; do
:
done
) &
pid=$!
echo 2 > /proc/sys/vm/drop_caches
kill $pid
wait $pid
# Move foo out of our testdir. This will set last_unlink_trans
# of the directory inode to the current transaction, because
# logged_trans of both the directory and the file are set to 0.
mv $MNT/testdir/foo $MNT/foo
# Change file foo again and fsync it.
# This fsync will result in a transaction commit because the rename
# above has set last_unlink_trans of the parent directory to the id
# of the current transaction and because our inode for file foo has
# last_unlink_trans set to the current transaction, since it was
# evicted and reloaded and it was previously modified in the current
# transaction (the xattr addition).
xfs_io -c "pwrite 0 64K" $MNT/foo
start=$(date +%s%N)
xfs_io -c "fsync" $MNT/foo
end=$(date +%s%N)
dur=$(( (end - start) / 1000000 ))
echo "file fsync took: $dur milliseconds"
umount $MNT
Before this patch: fsync took 19 milliseconds
After this patch: fsync took 5 milliseconds
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
At need_log_inode() we directly check the ->logged_trans field of the
given inode to check if it was previously logged in the transaction, with
the goal of skipping logging the inode again when it's not necessary.
The ->logged_trans field in not persisted in the inode item or elsewhere,
it's only stored in memory (struct btrfs_inode), so it's transient and
lost once the inode is evicted and then loaded again. Once an inode is
loaded, we are conservative and set ->logged_trans to 0, which may mean
that either the inode was never logged in the current transaction or it
was logged but evicted before being loaded again.
Instead of checking the inode's ->logged_trans field directly, we can
use instead the helper inode_logged(), which will really check if the
inode was logged before in the current transaction in case we have a
->logged_trans field with a value of 0. This will prevent unnecessarily
logging an inode when it's not needed, and in some cases preventing a
transaction commit, in case the logging requires a fallback to a
transaction commit. The following test script shows a scenario where
due to eviction we fallback a transaction commit when trying to fsync
a file that was renamed:
$ cat test.sh
#!/bin/bash
DEV=/dev/nullb0
MNT=/mnt/nullb0
num_init_files=10000
num_new_files=10000
mkfs.btrfs -f $DEV
mount -o ssd $DEV $MNT
mkdir $MNT/testdir
for ((i = 1; i <= $num_init_files; i++)); do
echo -n > $MNT/testdir/file_$i
done
echo -n > $MNT/testdir/foo
sync
# Add some files so that there's more work in the transaction other
# than just renaming file foo.
for ((i = 1; i <= $num_new_files; i++)); do
echo -n > $MNT/testdir/new_file_$i
done
# Fsync the directory first.
xfs_io -c "fsync" $MNT/testdir
# Rename file foo.
mv $MNT/testdir/foo $MNT/testdir/bar
# Now trigger eviction of the test directory's inode.
# Once loaded again, it will have logged_trans set to 0 and
# last_unlink_trans set to the current transaction.
echo 2 > /proc/sys/vm/drop_caches
# Fsync file bar (ex-foo).
# Before the patch the fsync would result in a transaction commit
# because the inode for file bar has last_unlink_trans set to the
# current transaction, so it will attempt to log the parent directory
# as well, which will fallback to a full transaction commit because
# it also has its last_unlink_trans set to the current transaction,
# due to the inode eviction.
start=$(date +%s%N)
xfs_io -c "fsync" $MNT/testdir/bar
end=$(date +%s%N)
dur=$(( (end - start) / 1000000 ))
echo "file fsync took: $dur milliseconds"
umount $MNT
Before this patch: fsync took 22 milliseconds
After this patch: fsync took 8 milliseconds
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
These functions are defined in the scrub.c file, but last callers were
removed in e9255d6c40 ("btrfs: scrub: remove the old scrub recheck
code").
fs/btrfs/scrub.c:553:20: warning: unused function 'scrub_stripe_index_and_offset'.
fs/btrfs/scrub.c:543:19: warning: unused function 'scrub_nr_raid_mirrors'.
Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Link: https://bugzilla.openanolis.cn/show_bug.cgi?id=4937
Signed-off-by: Jiapeng Chong <jiapeng.chong@linux.alibaba.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
[BUG]
Smatch reports the following errors related to commit ("btrfs: output
affected files when relocation fails"):
fs/btrfs/inode.c:283 print_data_reloc_error()
error: uninitialized symbol 'ref_level'.
[CAUSE]
That part of code is mostly copied from scrub, but unfortunately scrub
code from the beginning is not doing the error handling properly.
The offending code looks like this:
do {
ret = tree_backref_for_extent();
btrfs_warn_rl();
} while (ret != 1);
There are several problems involved:
- No error handling
If that tree_backref_for_extent() failed, we would output the same
error again and again, never really exit as it requires ret == 1 to
exit.
- Always do one extra output
As tree_backref_for_extent() only return > 0 if there is no more
backref item.
This means after the last item we hit, we would output an invalid
error message for ret > 0 case.
[FIX]
Fix the old code by:
- Move @ref_root and @ref_level into the if branch
And do not initialize them, so we can catch such uninitialized values
just like what we do in the inode.c
- Explicitly check the return value of tree_backref_for_extent()
And handle ret < 0 and ret > 0 cases properly.
- No more do {} while () loop
Instead go while (true) {} loop since we will handle @ret manually.
Reported-by: Dan Carpenter <dan.carpenter@linaro.org>
Signed-off-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
When btrfs_redirty_list_add redirties a buffer, it also acquires
an extra reference that is released on transaction commit. But
this is not required as buffers that are dirty or under writeback
are never freed (look for calls to extent_buffer_under_io())).
Remove the extra reference and the infrastructure used to drop it
again.
History behind redirty logic:
In the first place, it used releasing_list to hold all the
to-be-released extent buffers, and decided which buffers to re-dirty at
the commit time. Then, in a later version, the behaviour got changed to
re-dirty a necessary buffer and add re-dirtied one to the list in
btrfs_free_tree_block(). In short, the list was there mostly for the
patch series' historical reason.
Reviewed-by: Naohiro Aota <naohiro.aota@wdc.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
[ add Naohiro's comment regarding history ]
Signed-off-by: David Sterba <dsterba@suse.com>
dirty_metadata_bytes is decremented in both places that clear the dirty
bit in a buffer, but only incremented in btrfs_mark_buffer_dirty, which
means that a buffer that is redirtied using btrfs_redirty_list_add won't
be added to dirty_metadata_bytes, but it will be subtracted when written
out, leading an inconsistency in the counter.
Move the dirty_metadata_bytes from btrfs_mark_buffer_dirty into
set_extent_buffer_dirty to also account for the redirty case, and remove
the now unused set_extent_buffer_dirty return value.
Fixes: d3575156f6 ("btrfs: zoned: redirty released extent buffers")
CC: stable@vger.kernel.org # 5.15+
Reviewed-by: Naohiro Aota <naohiro.aota@wdc.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: David Sterba <dsterba@suse.com>
Mark btrfs_run_discard_work static and move it above its callers.
Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
This exists internal to ctree.c, however btrfs check needs to use it for
some of its operations. I'd rather not duplicate that code inside of
btrfs check as this is low level and I want to keep this code in one
place, so rename the function to btrfs_del_ptr and export it so that it
can be used inside of btrfs-progs safely. Add a comment to make sure
this doesn't get removed by a future cleanup.
Signed-off-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: David Sterba <dsterba@suse.com>
This is needed in btrfs-progs for the tools that convert the checksum
types for file systems and a few other things. We don't have it in the
kernel as we just want to get the size for the super blocks type.
However I don't want to have to manually add this every time we sync
ctree.c into btrfs-progs, so add the helper in the kernel with a note so
it doesn't get removed by a later cleanup.
Signed-off-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: David Sterba <dsterba@suse.com>
We want to override this in btrfs-progs, so wrap this in the __KERNEL__
check so we can easily sync this to btrfs-progs and have our local
version of btrfs_no_printk do the work.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: David Sterba <dsterba@suse.com>
These are more related to the inode item flags on disk than the
in-memory btrfs_inode, move the helpers to inode-item.h.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: David Sterba <dsterba@suse.com>
This is more a buffer validation helper, move it into the tree-checker
files where it makes more sense.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: David Sterba <dsterba@suse.com>
This helper returns a btrfs_tree_block_status for the various errors,
and then btrfs_check_node() will return -EUCLEAN if it gets anything
other than BTRFS_TREE_BLOCK_CLEAN which will be used by the kernel. In
the future btrfs-progs will use this helper instead.
Signed-off-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Instead of blanket returning -EUCLEAN for all the failures in
btrfs_check_leaf, use btrfs_tree_block_status and return the appropriate
status for each failure. Rename the helper to __btrfs_check_leaf and
then make a wrapper of btrfs_check_leaf that will return -EUCLEAN to
non-clean error codes. This will allow us to have the
__btrfs_check_leaf variant in btrfs-progs while keeping the behavior in
the kernel consistent.
Signed-off-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: David Sterba <dsterba@suse.com>
We have a variety of item specific errors that can occur. For now
simply put these under the umbrella of BTRFS_TREE_BLOCK_INVALID_ITEM,
this can be fleshed out as we need in the future.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: David Sterba <dsterba@suse.com>
We use this in btrfs-progs to determine if we can fix different types of
corruptions. We don't care about this in the kernel, however it would
be good to share this code between the kernel and btrfs-progs, so add
the status definitions so we can start converting the tree-checker code
over to using these status flags instead of blanket returning -EUCLEAN.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: David Sterba <dsterba@suse.com>
We have two helpers for checking leaves, because we have an extra check
for debugging in btrfs_mark_buffer_dirty(), and at that stage we may
have item data that isn't consistent yet. However we can handle this
case internally in the helper, if BTRFS_HEADER_FLAG_WRITTEN is set we
know the buffer should be internally consistent, otherwise we need to
skip checking the item data.
Simplify this helper down a single helper and handle the item data
checking logic internally to the helper.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: David Sterba <dsterba@suse.com>
We just pass in btrfs_header_level(eb) for the level, and we're passing
in the eb already, so simply get the level from the eb inside of
btrfs_set_block_flags.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: David Sterba <dsterba@suse.com>
This is completely related to block rsv's, move it out of the free space
cache code and into block-rsv.c.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: David Sterba <dsterba@suse.com>
For P/Q stripe scrub, we have quite some duplicated read IO:
- Data stripes read for verification
This is triggered by the scrub_submit_initial_read() inside
scrub_raid56_parity_stripe().
- Data stripes read (again) for P/Q stripe verification
This is triggered by scrub_assemble_read_bios() from scrub_rbio().
Although we can have hit rbio cache and avoid unnecessary read, the
chance is very low, as scrub would easily flush the whole rbio cache.
This means, even we're just scrubbing a single P/Q stripe, we would read
the data stripes twice for the best case scenario. If we need to
recover some data stripes, it would cause more reads on the same data
stripes, again and again.
However before we call raid56_parity_submit_scrub_rbio() we already
have all data stripes repaired and their contents ready to use.
But RAID56 cache is unaware about the scrub cache, thus RAID56 layer
itself still needs to re-read the data stripes.
To avoid such cache miss, this patch would:
- Introduce a new helper, raid56_parity_cache_data_pages()
This function would grab the pages from an array, and copy the content
to the rbio, marking all the involved sectors uptodate.
The page copy is unavoidable because of the cache pages of rbio are all
self managed, thus can not utilize outside pages without screwing up
the lifespan.
- Use the repaired data stripes as cache inside
scrub_raid56_parity_stripe()
By this, we ensure all the data sectors of the scrub rbio are already
uptodate, and no need to read them again from disk.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Removing a free space entry from an in memory space cache requires having
the corresponding btrfs_free_space_ctl's 'tree_lock' held. We have several
code paths that remove an entry, so add assertions where appropriate to
verify we are holding the lock, as the lock is acquired by some other
function up in the call chain, which makes it easy to miss in the future.
Note: for this to work we need to lock the local btrfs_free_space_ctl at
load_free_space_cache(), which was not being done because it's local,
declared on the stack, so no other task has access to it.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
When linking a free space entry, at link_free_space(), the caller should
be holding the spinlock 'tree_lock' of the given btrfs_free_space_ctl
argument, which is necessary for manipulating the red black tree of free
space entries (done by tree_insert_offset(), which already asserts the
lock is held) and for manipulating the 'free_space', 'free_extents',
'discardable_extents' and 'discardable_bytes' counters of the given
struct btrfs_free_space_ctl.
So assert that the spinlock 'tree_lock' of the given btrfs_free_space_ctl
is held by the current task. We have multiple code paths that end up
calling link_free_space(), and all currently take the lock before calling
it.
Reviewed-by: Anand Jain <anand.jain@oracle.com>
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
When searching for a free space entry by offset, at tree_search_offset(),
we are supposed to have the btrfs_free_space_ctl's 'tree_lock' held, so
assert that. We have multiple callers of tree_search_offset(), and all
currently hold the necessary lock before calling it.
Reviewed-by: Anand Jain <anand.jain@oracle.com>
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
There are multiple code paths leading to tree_insert_offset(), and each
path takes the necessary locks before tree_insert_offset() is called,
since they do other things that require those locks to be held. This makes
it easy to miss the locking somewhere, so make tree_insert_offset() assert
that the required locks are being held by the calling task.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
For the in-memory component of space caching (free space cache and free
space tree), three of the arguments passed to tree_insert_offset() can
always be taken from the new free space entry that we are about to add.
So simplify tree_insert_offset() to take the new entry instead of the
'offset', 'node' and 'bitmap' arguments. This will also allow to make
further changes simpler.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
The are two computations of end offsets at do_trimming() that are not
necessary, as they were previously computed and stored in local const
variables. So just use the variables instead, to make the source code
shorter and easier to read.
Reviewed-by: Anand Jain <anand.jain@oracle.com>
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
At try_merge_free_space(), avoid calling twice rb_prev() to find the
previous node, as that requires looping through the red black tree, so
store the result of the rb_prev() call and then use it.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
At copy_free_space_cache(), we add a new entry to the block group's ctl
before we free the entry from the temporary ctl. Adding a new entry
requires the allocation of a new struct btrfs_free_space, so we can
avoid a temporary extra allocation by freeing the entry from the
temporary ctl before we add a new entry to the main ctl, which possibly
also reduces the chances for a memory allocation failure in case of very
high memory pressure. So just do that.
Reviewed-by: Anand Jain <anand.jain@oracle.com>
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
A small code simplification, move the default value of transid to its
initialization and remove the else-statement.
Signed-off-by: Tom Rix <trix@redhat.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
[PROBLEM]
When relocation fails (mostly due to checksum mismatch), we only got
very cryptic error messages like:
BTRFS info (device dm-4): relocating block group 13631488 flags data
BTRFS warning (device dm-4): csum failed root -9 ino 257 off 0 csum 0x373e1ae3 expected csum 0x98757625 mirror 1
BTRFS error (device dm-4): bdev /dev/mapper/test-scratch1 errs: wr 0, rd 0, flush 0, corrupt 1, gen 0
BTRFS info (device dm-4): balance: ended with status: -5
The end user has to decipher the above messages and use various tools to
locate the affected files and find a way to fix the problem (mostly
deleting the file). This is not an easy work even for experienced
developer, not to mention the end users.
[SCRUB IS DOING BETTER]
By contrast, scrub is providing much better error messages:
BTRFS error (device dm-4): unable to fixup (regular) error at logical 13631488 on dev /dev/mapper/test-scratch1 physical 13631488
BTRFS warning (device dm-4): checksum error at logical 13631488 on dev /dev/mapper/test-scratch1, physical 13631488, root 5, inode 257, offset 0, length 4096, links 1 (path: file)
BTRFS info (device dm-4): scrub: finished on devid 1 with status: 0
Which provides the affected files directly to the end user.
[IMPROVEMENT]
Instead of the generic data checksum error messages, which is not doing
a good job for data reloc inodes, this patch introduce a scrub like
backref walking based solution.
When a sector fails its checksum for data reloc inode, we go the
following workflow:
- Get the real logical bytenr
For data reloc inode, the file offset is the offset inside the block
group.
Thus the real logical bytenr is @file_off + @block_group->start.
- Do an extent type check
If it's tree blocks it's much easier to handle, just go through
all the tree block backref.
- Do a backref walk and inode path resolution for data extents
This is mostly the same as scrub.
But unfortunately we can not reuse the same function as the output
format is different.
Now the new output would be more user friendly:
BTRFS info (device dm-4): relocating block group 13631488 flags data
BTRFS warning (device dm-4): csum failed root -9 ino 257 off 0 logical 13631488 csum 0x373e1ae3 expected csum 0x98757625 mirror 1
BTRFS warning (device dm-4): checksum error at logical 13631488 mirror 1 root 5 inode 257 offset 0 length 4096 links 1 (path: file)
BTRFS error (device dm-4): bdev /dev/mapper/test-scratch1 errs: wr 0, rd 0, flush 0, corrupt 2, gen 0
BTRFS info (device dm-4): balance: ended with status: -5
Reviewed-by: Anand Jain <anand.jain@oracle.com>
Signed-off-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Now that btrfs_wq_submit_bio is never called for synchronous I/O,
the hipri_workers workqueue is not used anymore and can be removed.
Reviewed-by: Chris Mason <clm@fb.com>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
The writeback_control structure already passes down the information about
a writeback being synchronous from the core VM code, and thus information
is propagated into the bio REQ_SYNC flag through the wbc_to_write_flags
helper.
Use that information to decide if checksums calculation is offloaded to
a workqueue instead of btrfs_inode::sync_writers field that not only
bloats the inode but also has too wide scope, being inode wide instead
of limited to the actual writeback request.
The sync writes were set in:
- btrfs_do_write_iter - regular IO, sync status is set
- start_ordered_ops - ordered write start, writeback with WB_SYNC_ALL
mode
- btrfs_write_marked_extents - write marked extents, writeback with
WB_SYNC_ALL mode
Reviewed-by: Chris Mason <clm@fb.com>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
[ update changelog ]
Signed-off-by: David Sterba <dsterba@suse.com>
Most modern hardware supports very fast accelerated crc32c calculation.
If that is supported the CPU overhead of the checksum calculation is
very limited, and offloading the calculation to special worker threads
has a lot of overhead for no gain.
E.g. on an Intel Optane device is actually very much slows down even
1M buffered writes with fio:
Unpatched:
write: IOPS=3316, BW=3316MiB/s (3477MB/s)(200GiB/61757msec); 0 zone resets
With synchronous CRCs:
write: IOPS=4882, BW=4882MiB/s (5119MB/s)(200GiB/41948msec); 0 zone resets
With a lot of variation during the unpatched run going down as low as
1100MB/s, while the synchronous CRC version has about the same peak write
speed but much lower dips, and fewer kworkers churning around.
Both tests had fio saturated at 100% CPU.
(thanks to Jens Axboe via Chris Mason for the benchmarking)
Reviewed-by: Chris Mason <clm@fb.com>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Using SECTOR_SHIFT to convert LBA to physical address makes it more
readable.
Signed-off-by: Anand Jain <anand.jain@oracle.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Use SECTOR_SHIFT while converting a physical address to an LBA, makes
it more readable.
Signed-off-by: Anand Jain <anand.jain@oracle.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Improve the leaf dump behavior by:
- Always dump the leaf first, then the error message
- Output the slot number if possible
Especially in __btrfs_free_extent() the leaf dump of extent tree can
be pretty large.
With an extra slot number it's much easier to locate the problem.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Since print-tree infrastructure only prints the content of a tree block,
we can make them to accept const extent buffer pointer.
This removes a forced type convert in extent-tree, where we convert a
const extent buffer pointer to regular one, just to avoid compiler
warning.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
bitmap_test_range_all_{set,zero} defined in subpage.c are useful for other
components. Move them to misc.h and use them in zoned.c. Also, as
find_next{,_zero}_bit take/return "unsigned long" instead of "unsigned
int", convert the type to "unsigned long".
While at it, also rewrite the "if (...) return true; else return false;"
pattern and add const to the input bitmap.
Signed-off-by: Naohiro Aota <naohiro.aota@wdc.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
When checking siblings keys, before moving keys from one node/leaf to a
sibling node/leaf, it's very unexpected to have the last key of the left
sibling greater than or equals to the first key of the right sibling, as
that means we have a (serious) corruption that breaks the key ordering
properties of a b+tree. Since this is unexpected, surround the comparison
with the unlikely macro, which helps the compiler generate better code
for the most expected case (no existing b+tree corruption). This is also
what we do for other unexpected cases of invalid key ordering (like at
btrfs_set_item_key_safe()).
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>
The function btrfs_free_device() is never used outside of volumes.c, so
make it static and remove its prototype declaration at volumes.h.
Reviewed-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: Anand Jain <anand.jain@oracle.com>
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Recently a Meta-internal workload encountered subvolume creation taking
up to 2s each, significantly slower than directory creation. As they
were hoping to be able to use subvolumes instead of directories, and
were looking to create hundreds, this was a significant issue. After
Josef investigated, it turned out to be due to the transaction commit
currently performed at the end of subvolume creation.
This change improves the workload by not doing transaction commit for every
subvolume creation, and merely requiring a transaction commit on fsync.
In the worst case, of doing a subvolume create and fsync in a loop, this
should require an equal amount of time to the current scheme; and in the
best case, the internal workload creating hundreds of subvolumes before
fsyncing is greatly improved.
While it would be nice to be able to use the log tree and use the normal
fsync path, log tree replay can't deal with new subvolume inodes
presently.
It's possible that there's some reason that the transaction commit is
necessary for correctness during subvolume creation; however,
git logs indicate that the commit dates back to the beginning of
subvolume creation, and there are no notes on why it would be necessary.
Reviewed-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Reviewed-by: Neal Gompa <neal@gompa.dev>
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>
btrfs_prev_leaf() is not used outside ctree.c, so there's no need to
export it at ctree.h - just make it static at ctree.c and move its
definition above btrfs_search_slot_for_read(), since that function
calls it.
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
-----BEGIN PGP SIGNATURE-----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=NAyd
-----END PGP SIGNATURE-----
Merge tag 'for-6.4-rc6-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux
Pull btrfs fixes from David Sterba:
"Two fixes for NOCOW files, a regression fix in scrub and an assertion
fix:
- NOCOW fixes:
- keep length of iomap direct io request in case of a failure
- properly pass mode of extent reference checking, this can break
some cases for swapfile
- fix error value confusion when scrubbing a stripe
- convert assertion to a proper error handling when loading global
roots, reported by syzbot"
* tag 'for-6.4-rc6-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux:
btrfs: scrub: fix a return value overwrite in scrub_stripe()
btrfs: do not ASSERT() on duplicated global roots
btrfs: can_nocow_file_extent should pass down args->strict from callers
btrfs: fix iomap_begin length for nocow writes
[RETURN VALUE OVERWRITE]
Inside scrub_stripe(), we would submit all the remaining stripes after
iterating all extents.
But since flush_scrub_stripes() can return error, we need to avoid
overwriting the existing @ret if there is any error.
However the existing check is doing the wrong check:
ret2 = flush_scrub_stripes();
if (!ret2)
ret = ret2;
This would overwrite the existing @ret to 0 as long as the final flush
detects no critical errors.
[FIX]
We should check @ret other than @ret2 in that case.
Fixes: 8eb3dd17ea ("btrfs: dev-replace: error out if we have unrepaired metadata error during")
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
[BUG]
Syzbot reports a reproducible ASSERT() when using rescue=usebackuproot
mount option on a corrupted fs.
The full report can be found here:
https://syzkaller.appspot.com/bug?extid=c4614eae20a166c25bf0
BTRFS error (device loop0: state C): failed to load root csum
assertion failed: !tmp, in fs/btrfs/disk-io.c:1103
------------[ cut here ]------------
kernel BUG at fs/btrfs/ctree.h:3664!
invalid opcode: 0000 [#1] PREEMPT SMP KASAN
CPU: 1 PID: 3608 Comm: syz-executor356 Not tainted 6.0.0-rc7-syzkaller-00029-g3800a713b607 #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 08/26/2022
RIP: 0010:assertfail+0x1a/0x1c fs/btrfs/ctree.h:3663
RSP: 0018:ffffc90003aaf250 EFLAGS: 00010246
RAX: 0000000000000032 RBX: 0000000000000000 RCX: f21c13f886638400
RDX: 0000000000000000 RSI: 0000000080000000 RDI: 0000000000000000
RBP: ffff888021c640a0 R08: ffffffff816bd38d R09: ffffed10173667f1
R10: ffffed10173667f1 R11: 1ffff110173667f0 R12: dffffc0000000000
R13: ffff8880229c21f7 R14: ffff888021c64060 R15: ffff8880226c0000
FS: 0000555556a73300(0000) GS:ffff8880b9b00000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 000055a2637d7a00 CR3: 00000000709c4000 CR4: 00000000003506e0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
<TASK>
btrfs_global_root_insert+0x1a7/0x1b0 fs/btrfs/disk-io.c:1103
load_global_roots_objectid+0x482/0x8c0 fs/btrfs/disk-io.c:2467
load_global_roots fs/btrfs/disk-io.c:2501 [inline]
btrfs_read_roots fs/btrfs/disk-io.c:2528 [inline]
init_tree_roots+0xccb/0x203c fs/btrfs/disk-io.c:2939
open_ctree+0x1e53/0x33df fs/btrfs/disk-io.c:3574
btrfs_fill_super+0x1c6/0x2d0 fs/btrfs/super.c:1456
btrfs_mount_root+0x885/0x9a0 fs/btrfs/super.c:1824
legacy_get_tree+0xea/0x180 fs/fs_context.c:610
vfs_get_tree+0x88/0x270 fs/super.c:1530
fc_mount fs/namespace.c:1043 [inline]
vfs_kern_mount+0xc9/0x160 fs/namespace.c:1073
btrfs_mount+0x3d3/0xbb0 fs/btrfs/super.c:1884
[CAUSE]
Since the introduction of global roots, we handle
csum/extent/free-space-tree roots as global roots, even if no
extent-tree-v2 feature is enabled.
So for regular csum/extent/fst roots, we load them into
fs_info::global_root_tree rb tree.
And we should not expect any conflicts in that rb tree, thus we have an
ASSERT() inside btrfs_global_root_insert().
But rescue=usebackuproot can break the assumption, as we will try to
load those trees again and again as long as we have bad roots and have
backup roots slot remaining.
So in that case we can have conflicting roots in the rb tree, and
triggering the ASSERT() crash.
[FIX]
We can safely remove that ASSERT(), as the caller will properly put the
offending root.
To make further debugging easier, also add two explicit error messages:
- Error message for conflicting global roots
- Error message when using backup roots slot
Reported-by: syzbot+a694851c6ab28cbcfb9c@syzkaller.appspotmail.com
Fixes: abed4aaae4 ("btrfs: track the csum, extent, and free space trees in a rb tree")
CC: stable@vger.kernel.org # 6.1+
Signed-off-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Commit 619104ba45 ("btrfs: move common NOCOW checks against a file
extent into a helper") changed our call to btrfs_cross_ref_exist() to
always pass false for the 'strict' parameter. We're passing this down
through the stack so that we can do a full check for cross references
during swapfile activation.
With strict always false, this test fails:
btrfs subvol create swappy
chattr +C swappy
fallocate -l1G swappy/swapfile
chmod 600 swappy/swapfile
mkswap swappy/swapfile
btrfs subvol snap swappy swapsnap
btrfs subvol del -C swapsnap
btrfs fi sync /
sync;sync;sync
swapon swappy/swapfile
The fix is to just use args->strict, and everyone except swapfile
activation is passing false.
Fixes: 619104ba45 ("btrfs: move common NOCOW checks against a file extent into a helper")
CC: stable@vger.kernel.org # 6.1+
Reviewed-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
Signed-off-by: David Sterba <dsterba@suse.com>
can_nocow_extent can reduce the len passed in, which needs to be
propagated to btrfs_dio_iomap_begin so that iomap does not submit
more data then is mapped.
This problems exists since the btrfs_get_blocks_direct helper was added
in commit c5794e5178 ("btrfs: Factor out write portion of
btrfs_get_blocks_direct"), but the ordered_extent splitting added in
commit b73a6fd1b1 ("btrfs: split partial dio bios before submit")
added a WARN_ON that made a syzkaller test fail.
Reported-by: syzbot+ee90502d5c8fd1d0dd93@syzkaller.appspotmail.com
Fixes: c5794e5178 ("btrfs: Factor out write portion of btrfs_get_blocks_direct")
CC: stable@vger.kernel.org # 6.1+
Tested-by: syzbot+ee90502d5c8fd1d0dd93@syzkaller.appspotmail.com
Reviewed-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: David Sterba <dsterba@suse.com>
-----BEGIN PGP SIGNATURE-----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=ABpK
-----END PGP SIGNATURE-----
Merge tag 'for-6.4-rc6-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux
Pull btrfs fixes from David Sterba:
"A more fixes and regression fixes:
- in subpage mode, fix crash when repairing metadata at the end of
a stripe
- properly enable async discard when remounting from read-only to
read-write
- scrub regression fixes:
- respect read-only scrub when attempting to do a repair
- fix reporting of found errors, the stats don't get properly
accounted after a stripe repair"
* tag 'for-6.4-rc6-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux:
btrfs: scrub: also report errors hit during the initial read
btrfs: scrub: respect the read-only flag during repair
btrfs: properly enable async discard when switching from RO->RW
btrfs: subpage: fix a crash in metadata repair path
The only overlap between the block open flags mapped into the fmode_t and
other uses of fmode_t are FMODE_READ and FMODE_WRITE. Define a new
blk_mode_t instead for use in blkdev_get_by_{dev,path}, ->open and
->ioctl and stop abusing fmode_t.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Jack Wang <jinpu.wang@ionos.com> [rnbd]
Reviewed-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Christian Brauner <brauner@kernel.org>
Link: https://lore.kernel.org/r/20230608110258.189493-28-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Add a helper to return the open flags for blkdev_get_by* for passed in
super block flags instead of open coding the logic in many places.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Acked-by: Christian Brauner <brauner@kernel.org>
Link: https://lore.kernel.org/r/20230608110258.189493-17-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
The current interface for exclusive opens is rather confusing as it
requires both the FMODE_EXCL flag and a holder. Remove the need to pass
FMODE_EXCL and just key off the exclusive open off a non-NULL holder.
For blkdev_put this requires adding the holder argument, which provides
better debug checking that only the holder actually releases the hold,
but at the same time allows removing the now superfluous mode argument.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Acked-by: Christian Brauner <brauner@kernel.org>
Acked-by: David Sterba <dsterba@suse.com> [btrfs]
Acked-by: Jack Wang <jinpu.wang@ionos.com> [rnbd]
Link: https://lore.kernel.org/r/20230608110258.189493-16-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Passing a holder to blkdev_get_by_path when FMODE_EXCL isn't set doesn't
make sense, so pass NULL instead and remove the holder argument from the
call chains the only end up in non-FMODE_EXCL blkdev_get_by_path calls.
Exclusive mode for device scanning is not used since commit 50d281fc43
("btrfs: scan device in non-exclusive mode")".
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Acked-by: Christian Brauner <brauner@kernel.org>
Acked-by: David Sterba <dsterba@suse.com>
Link: https://lore.kernel.org/r/20230608110258.189493-15-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Patch series "cleanup the filemap / direct I/O interaction", v4.
This series cleans up some of the generic write helper calling conventions
and the page cache writeback / invalidation for direct I/O. This is a
spinoff from the no-bufferhead kernel project, for which we'll want to an
use iomap based buffered write path in the block layer.
This patch (of 12):
The last user of current->backing_dev_info disappeared in commit
b9b1335e64 ("remove bdi_congested() and wb_congested() and related
functions"). Remove the field and all assignments to it.
Link: https://lkml.kernel.org/r/20230601145904.1385409-1-hch@lst.de
Link: https://lkml.kernel.org/r/20230601145904.1385409-2-hch@lst.de
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Christian Brauner <brauner@kernel.org>
Reviewed-by: Damien Le Moal <dlemoal@kernel.org>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Acked-by: Theodore Ts'o <tytso@mit.edu>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Andreas Gruenbacher <agruenba@redhat.com>
Cc: Anna Schumaker <anna@kernel.org>
Cc: Chao Yu <chao@kernel.org>
Cc: Ilya Dryomov <idryomov@gmail.com>
Cc: Jaegeuk Kim <jaegeuk@kernel.org>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Miklos Szeredi <miklos@szeredi.hu>
Cc: Miklos Szeredi <mszeredi@redhat.com>
Cc: Trond Myklebust <trond.myklebust@hammerspace.com>
Cc: Xiubo Li <xiubli@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
[BUG]
After the recent scrub rework introduced in commit e02ee89baa ("btrfs:
scrub: switch scrub_simple_mirror() to scrub_stripe infrastructure"),
btrfs scrub no longer reports repaired errors any more:
# mkfs.btrfs -f $dev -d DUP
# mount $dev $mnt
# xfs_io -f -d -c "pwrite -b 64K -S 0xaa 0 64" $mnt/file
# umount $dev
# xfs_io -f -c "pwrite -S 0xff $phy1 64K" $dev # Corrupt the first mirror
# mount $dev $mnt
# btrfs scrub start -BR $mnt
scrub done for 725e7cb7-8a4a-4c77-9f2a-86943619e218
Scrub started: Tue Jun 6 14:56:50 2023
Status: finished
Duration: 0:00:00
data_extents_scrubbed: 2
tree_extents_scrubbed: 18
data_bytes_scrubbed: 131072
tree_bytes_scrubbed: 294912
read_errors: 0
csum_errors: 0 <<< No errors here
verify_errors: 0
[...]
uncorrectable_errors: 0
unverified_errors: 0
corrected_errors: 16 <<< Only corrected errors
last_physical: 2723151872
This can confuse btrfs-progs, as it relies on the csum_errors to
determine if there is anything wrong.
While on v6.3.x kernels, the report is different:
csum_errors: 16 <<<
verify_errors: 0
[...]
uncorrectable_errors: 0
unverified_errors: 0
corrected_errors: 16 <<<
[CAUSE]
In the reworked scrub, we update the scrub progress inside
scrub_stripe_report_errors(), using various bitmaps to update the
result.
For example for csum_errors, we use bitmap_weight() of
stripe->csum_error_bitmap.
Unfortunately at that stage, all error bitmaps (except
init_error_bitmap) are the result of the latest repair attempt, thus if
the stripe is fully repaired, those error bitmaps will all be empty,
resulting the above output mismatch.
To fix this, record the number of errors into stripe->init_nr_*_errors.
Since we don't really care about where those errors are, we only need to
record the number of errors.
Then in scrub_stripe_report_errors(), use those initial numbers to
update the progress other than using the latest error bitmaps.
Fixes: e02ee89baa ("btrfs: scrub: switch scrub_simple_mirror() to scrub_stripe infrastructure")
Signed-off-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
[BUG]
With recent scrub rework, the scrub operation no longer respects the
read-only flag passed by "-r" option of "btrfs scrub start" command.
# mkfs.btrfs -f -d raid1 $dev1 $dev2
# mount $dev1 $mnt
# xfs_io -f -d -c "pwrite -b 128K -S 0xaa 0 128k" $mnt/file
# sync
# xfs_io -c "pwrite -S 0xff $phy1 64k" $dev1
# xfs_io -c "pwrite -S 0xff $((phy2 + 65536)) 64k" $dev2
# mount $dev1 $mnt -o ro
# btrfs scrub start -BrRd $mnt
Scrub device $dev1 (id 1) done
Scrub started: Tue Jun 6 09:59:14 2023
Status: finished
Duration: 0:00:00
[...]
corrected_errors: 16 <<< Still has corrupted sectors
last_physical: 1372585984
Scrub device $dev2 (id 2) done
Scrub started: Tue Jun 6 09:59:14 2023
Status: finished
Duration: 0:00:00
[...]
corrected_errors: 16 <<< Still has corrupted sectors
last_physical: 1351614464
# btrfs scrub start -BrRd $mnt
Scrub device $dev1 (id 1) done
Scrub started: Tue Jun 6 10:00:17 2023
Status: finished
Duration: 0:00:00
[...]
corrected_errors: 0 <<< No more errors
last_physical: 1372585984
Scrub device $dev2 (id 2) done
[...]
corrected_errors: 0 <<< No more errors
last_physical: 1372585984
[CAUSE]
In the newly reworked scrub code, repair is always submitted no matter
if we're doing a read-only scrub.
[FIX]
Fix it by skipping the write submission if the scrub is a read-only one.
Unfortunately for the report part, even for a read-only scrub we will
still report it as corrected errors, as we know it's repairable, even we
won't really submit the write.
Fixes: e02ee89baa ("btrfs: scrub: switch scrub_simple_mirror() to scrub_stripe infrastructure")
Signed-off-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
The async discard uses the BTRFS_FS_DISCARD_RUNNING bit in the fs_info
to force discards off when the filesystem has aborted or we're generally
not able to run discards. This gets flipped on when we're mounted rw,
and also when we go from ro->rw.
Commit 63a7cb1307 ("btrfs: auto enable discard=async when possible")
enabled async discard by default, and this meant
"mount -o ro /dev/xxx /yyy" had async discards turned on.
Unfortunately, this meant our check in btrfs_remount_cleanup() would see
that discards are already on:
/* If we toggled discard async */
if (!btrfs_raw_test_opt(old_opts, DISCARD_ASYNC) &&
btrfs_test_opt(fs_info, DISCARD_ASYNC))
btrfs_discard_resume(fs_info);
So, we'd never call btrfs_discard_resume() when remounting the root
filesystem from ro->rw.
drgn shows this really nicely:
import os
import sys
from drgn.helpers.linux.fs import path_lookup
from drgn import NULL, Object, Type, cast
def btrfs_sb(sb):
return cast("struct btrfs_fs_info *", sb.s_fs_info)
if len(sys.argv) == 1:
path = "/"
else:
path = sys.argv[1]
fs_info = cast("struct btrfs_fs_info *", path_lookup(prog, path).mnt.mnt_sb.s_fs_info)
BTRFS_FS_DISCARD_RUNNING = 1 << prog['BTRFS_FS_DISCARD_RUNNING']
if fs_info.flags & BTRFS_FS_DISCARD_RUNNING:
print("discard running flag is on")
else:
print("discard running flag is off")
[root]# mount | grep nvme
/dev/nvme0n1p3 on / type btrfs
(rw,relatime,compress-force=zstd:3,ssd,discard=async,space_cache=v2,subvolid=5,subvol=/)
[root]# ./discard_running.drgn
discard running flag is off
[root]# mount -o remount,discard=sync /
[root]# mount -o remount,discard=async /
[root]# ./discard_running.drgn
discard running flag is on
The fix is to call btrfs_discard_resume() when we're going from ro->rw.
It already checks to make sure the async discard flag is on, so it'll do
the right thing.
Fixes: 63a7cb1307 ("btrfs: auto enable discard=async when possible")
CC: stable@vger.kernel.org # 6.3+
Reviewed-by: Boris Burkov <boris@bur.io>
Signed-off-by: Chris Mason <clm@fb.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
[BUG]
Test case btrfs/027 would crash with subpage (64K page size, 4K
sectorsize) with the following dying messages:
debug: map_length=16384 length=65536 type=metadata|raid6(0x104)
assertion failed: map_length >= length, in fs/btrfs/volumes.c:8093
------------[ cut here ]------------
kernel BUG at fs/btrfs/messages.c:259!
Hardware name: QEMU KVM Virtual Machine, BIOS 0.0.0 02/06/2015
Call trace:
btrfs_assertfail+0x28/0x2c [btrfs]
btrfs_map_repair_block+0x150/0x2b8 [btrfs]
btrfs_repair_io_failure+0xd4/0x31c [btrfs]
btrfs_read_extent_buffer+0x150/0x16c [btrfs]
read_tree_block+0x38/0xbc [btrfs]
read_tree_root_path+0xfc/0x1bc [btrfs]
btrfs_get_root_ref.part.0+0xd4/0x3a8 [btrfs]
open_ctree+0xa30/0x172c [btrfs]
btrfs_mount_root+0x3c4/0x4a4 [btrfs]
legacy_get_tree+0x30/0x60
vfs_get_tree+0x28/0xec
vfs_kern_mount.part.0+0x90/0xd4
vfs_kern_mount+0x14/0x28
btrfs_mount+0x114/0x418 [btrfs]
legacy_get_tree+0x30/0x60
vfs_get_tree+0x28/0xec
path_mount+0x3e0/0xb64
__arm64_sys_mount+0x200/0x2d8
invoke_syscall+0x48/0x114
el0_svc_common.constprop.0+0x60/0x11c
do_el0_svc+0x38/0x98
el0_svc+0x40/0xa8
el0t_64_sync_handler+0xf4/0x120
el0t_64_sync+0x190/0x194
Code: aa0403e2 b0fff060 91010000 959c2024 (d4210000)
[CAUSE]
In btrfs/027 we test RAID6 with missing devices, in this particular
case, we're repairing a metadata at the end of a data stripe.
But at btrfs_repair_io_failure(), we always pass a full PAGE for repair,
and for subpage case this can cross stripe boundary and lead to the
above BUG_ON().
This metadata repair code is always there, since the introduction of
subpage support, but this can trigger BUG_ON() after the bio split
ability at btrfs_map_bio().
[FIX]
Instead of passing the old PAGE_SIZE, we calculate the correct length
based on the eb size and page size for both regular and subpage cases.
CC: stable@vger.kernel.org # 6.3+
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Add a new blk_holder_ops structure, which is passed to blkdev_get_by_* and
installed in the block_device for exclusive claims. It will be used to
allow the block layer to call back into the user of the block device for
thing like notification of a removed device or a device resize.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Jan Kara <jack@suse.cz>
Acked-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Link: https://lore.kernel.org/r/20230601094459.1350643-10-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
-----BEGIN PGP SIGNATURE-----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=O3yC
-----END PGP SIGNATURE-----
Merge tag 'for-6.4-rc4-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux
Pull btrfs fix from David Sterba:
"One regression fix.
The rewrite of scrub code in 6.4 broke device replace in zoned mode,
some of the writes could happen out of order so this had to be
adjusted for all cases"
* tag 'for-6.4-rc4-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux:
btrfs: zoned: fix dev-replace after the scrub rework
[BUG]
After commit e02ee89baa ("btrfs: scrub: switch scrub_simple_mirror()
to scrub_stripe infrastructure"), scrub no longer works for zoned device
at all.
Even an empty zoned btrfs cannot be replaced:
# mkfs.btrfs -f /dev/nvme0n1
# mount /dev/nvme0n1 /mnt/btrfs
# btrfs replace start -Bf 1 /dev/nvme0n2 /mnt/btrfs
Resetting device zones /dev/nvme1n1 (160 zones) ...
ERROR: ioctl(DEV_REPLACE_START) failed on "/mnt/btrfs/": Input/output error
And we can hit kernel crash related to that:
BTRFS info (device nvme1n1): host-managed zoned block device /dev/nvme3n1, 160 zones of 134217728 bytes
BTRFS info (device nvme1n1): dev_replace from /dev/nvme2n1 (devid 2) to /dev/nvme3n1 started
nvme3n1: Zone Management Append(0x7d) @ LBA 65536, 4 blocks, Zone Is Full (sct 0x1 / sc 0xb9) DNR
I/O error, dev nvme3n1, sector 786432 op 0xd:(ZONE_APPEND) flags 0x4000 phys_seg 3 prio class 2
BTRFS error (device nvme1n1): bdev /dev/nvme3n1 errs: wr 1, rd 0, flush 0, corrupt 0, gen 0
BUG: kernel NULL pointer dereference, address: 00000000000000a8
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.16.0-0-gd239552ce722-prebuilt.qemu.org 04/01/2014
RIP: 0010:_raw_spin_lock_irqsave+0x1e/0x40
Call Trace:
<IRQ>
btrfs_lookup_ordered_extent+0x31/0x190
btrfs_record_physical_zoned+0x18/0x40
btrfs_simple_end_io+0xaf/0xc0
blk_update_request+0x153/0x4c0
blk_mq_end_request+0x15/0xd0
nvme_poll_cq+0x1d3/0x360
nvme_irq+0x39/0x80
__handle_irq_event_percpu+0x3b/0x190
handle_irq_event+0x2f/0x70
handle_edge_irq+0x7c/0x210
__common_interrupt+0x34/0xa0
common_interrupt+0x7d/0xa0
</IRQ>
<TASK>
asm_common_interrupt+0x22/0x40
[CAUSE]
Dev-replace reuses scrub code to iterate all extents and write the
existing content back to the new device.
And for zoned devices, we call fill_writer_pointer_gap() to make sure
all the writes into the zoned device is sequential, even if there may be
some gaps between the writes.
However we have several different bugs all related to zoned dev-replace:
- We are using ZONE_APPEND operation for metadata style write back
For zoned devices, btrfs has two ways to write data:
* ZONE_APPEND for data
This allows higher queue depth, but will not be able to know where
the write would land.
Thus needs to grab the real on-disk physical location in it's endio.
* WRITE for metadata
This requires single queue depth (new writes can only be submitted
after previous one finished), and all writes must be sequential.
For scrub, we go single queue depth, but still goes with ZONE_APPEND,
which requires btrfs_bio::inode being populated.
This is the cause of that crash.
- No correct tracing of write_pointer
After a write finished, we should forward sctx->write_pointer, or
fill_writer_pointer_gap() would not work properly and cause more
than necessary zero out, and fill the whole zone prematurely.
- Incorrect physical bytenr passed to fill_writer_pointer_gap()
In scrub_write_sectors(), one call site passes logical address, which
is completely wrong.
The other call site passes physical address of current sector, but
we should pass the physical address of the btrfs_bio we're submitting.
This is the cause of the -EIO errors.
[FIX]
- Do not use ZONE_APPEND for btrfs_submit_repair_write().
- Manually forward sctx->write_pointer after successful writeback
- Use the physical address of the to-be-submitted btrfs_bio for
fill_writer_pointer_gap()
Now zoned device replace would work as expected.
Reported-by: Christoph Hellwig <hch@lst.de>
Fixes: e02ee89baa ("btrfs: scrub: switch scrub_simple_mirror() to scrub_stripe infrastructure")
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
-----BEGIN PGP SIGNATURE-----
iQIzBAABCgAdFiEE8rQSAMVO+zA4DBdWxWXV+ddtWDsFAmR2TDwACgkQxWXV+ddt
WDsMvQ/+KgUXW+Liu5BaOyD5UzPL4BgHWiPTmJyRpsWTkGm8LE/yRCRoxqp1XbU+
nOjQpjkxI+ziRgKpDTAGFK/w51TV9ECM5wyZiXx93TO6iaTOuYCtSnSsWylzEC1H
q9I3znLJSWrnBPTktwTZ29rvKvXj1k3th8ypyI9ho7N+3H0Uzt2VIPxrH2oVXZNz
f2vkjSX9pKGN5zxM2ahd3Nde4Ma6yAlJLD+pnlYK20zH/30cAXdJsUCsUqQLXDL1
sUR++Br7qym3Wqn9Qa5R71IPJ1FieW2NaHgAz4dBBFfqe5PR7YCGL/Md6G+CFJ1E
qLLFOWpELpqkeQdvivBnMZWqgpw+54Pdfuqxg7VylEmUc1y6CK4ab5XctpXIf75h
6bK0RPZ7D9jZl6JukkWftoS4XnW2cseyEfHneDMZDty4v1bxwR6g7i4ZTym413Gx
Td1Z+G6BN5O5ih0Pc0CgSS3QnndWTUl3LAHiuxRErrK4dxpeuQlDTGWWY7YVyRPJ
O9yC24GbHyWYBYHtNACEn6/GlXQjtswhjlHxqONmQfnstZL7Fz8si9EQEOWwssJE
PIlb022a1mvR42yHr64TE0SzpDZbMY8mnULAsSrWgPXh3IAt1ztUuJajcFs84MZr
qWewi4F/3wDAB0m1lUbAOmeBbpAw5gSGHhwBrjdK3EWJr2kxQ50=
=viyP
-----END PGP SIGNATURE-----
Merge tag 'for-6.4-rc4-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux
Pull btrfs fixes from David Sterba:
"One bug fix and two build warning fixes:
- call proper end bio callback for metadata RAID0 in a rare case of
an unaligned block
- fix uninitialized variable (reported by gcc 10.2)
- fix warning about potential access beyond array bounds on mips64
with 64k pages (runtime check would not allow that)"
* tag 'for-6.4-rc4-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux:
btrfs: fix csum_tree_block page iteration to avoid tripping on -Werror=array-bounds
btrfs: fix an uninitialized variable warning in btrfs_log_inode
btrfs: call btrfs_orig_bbio_end_io in btrfs_end_bio_work
When compiling on a MIPS 64-bit machine we get these warnings:
In file included from ./arch/mips/include/asm/cacheflush.h:13,
from ./include/linux/cacheflush.h:5,
from ./include/linux/highmem.h:8,
from ./include/linux/bvec.h:10,
from ./include/linux/blk_types.h:10,
from ./include/linux/blkdev.h:9,
from fs/btrfs/disk-io.c:7:
fs/btrfs/disk-io.c: In function ‘csum_tree_block’:
fs/btrfs/disk-io.c💯34: error: array subscript 1 is above array bounds of ‘struct page *[1]’ [-Werror=array-bounds]
100 | kaddr = page_address(buf->pages[i]);
| ~~~~~~~~~~^~~
./include/linux/mm.h:2135:48: note: in definition of macro ‘page_address’
2135 | #define page_address(page) lowmem_page_address(page)
| ^~~~
cc1: all warnings being treated as errors
We can check if i overflows to solve the problem. However, this doesn't make
much sense, since i == 1 and num_pages == 1 doesn't execute the body of the loop.
In addition, i < num_pages can also ensure that buf->pages[i] will not cross
the boundary. Unfortunately, this doesn't help with the problem observed here:
gcc still complains.
To fix this add a compile-time condition for the extent buffer page
array size limit, which would eventually lead to eliminating the whole
for loop.
CC: stable@vger.kernel.org # 5.10+
Signed-off-by: pengfuyuan <pengfuyuan@kylinos.cn>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
This fixes the following warning reported by gcc 10.2.1 under x86_64:
../fs/btrfs/tree-log.c: In function ‘btrfs_log_inode’:
../fs/btrfs/tree-log.c:6211:9: error: ‘last_range_start’ may be used uninitialized in this function [-Werror=maybe-uninitialized]
6211 | ret = insert_dir_log_key(trans, log, path, key.objectid,
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
6212 | first_dir_index, last_dir_index);
| ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
../fs/btrfs/tree-log.c:6161:6: note: ‘last_range_start’ was declared here
6161 | u64 last_range_start;
| ^~~~~~~~~~~~~~~~
This might be a false positive fixed in later compiler versions but we
want to have it fixed.
Reported-by: k2ci <kernel-bot@kylinos.cn>
Reviewed-by: Anand Jain <anand.jain@oracle.com>
Signed-off-by: Shida Zhang <zhangshida@kylinos.cn>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
When I implemented the storage layer bio splitting, I was under the
assumption that we'll never split metadata bios. But Qu reminded me that
this can actually happen with very old file systems with unaligned
metadata chunks and RAID0.
I still haven't seen such a case in practice, but we better handled this
case, especially as it is fairly easily to do not calling the ->end_іo
method directly in btrfs_end_io_work, and using the proper
btrfs_orig_bbio_end_io helper instead.
In addition to the old file system with unaligned metadata chunks case
documented in the commit log, the combination of the new scrub code
with Johannes pending raid-stripe-tree also triggers this case. We
spent some time debugging it and found that this patch solves
the problem.
Fixes: 103c19723c ("btrfs: split the bio submission path into a separate file")
CC: stable@vger.kernel.org # 6.3+
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Tested-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: David Sterba <dsterba@suse.com>
-----BEGIN PGP SIGNATURE-----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=elHB
-----END PGP SIGNATURE-----
Merge tag 'for-6.4-rc3-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux
Pull btrfs fixes from David Sterba:
- handle memory allocation error in checksumming helper (reported by
syzbot)
- fix lockdep splat when aborting a transaction, add NOFS protection
around invalidate_inode_pages2 that could allocate with GFP_KERNEL
- reduce chances to hit an ENOSPC during scrub with RAID56 profiles
* tag 'for-6.4-rc3-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux:
btrfs: use nofs when cleaning up aborted transactions
btrfs: handle memory allocation failure in btrfs_csum_one_bio
btrfs: scrub: try harder to mark RAID56 block groups read-only
Our CI system caught a lockdep splat:
======================================================
WARNING: possible circular locking dependency detected
6.3.0-rc7+ #1167 Not tainted
------------------------------------------------------
kswapd0/46 is trying to acquire lock:
ffff8c6543abd650 (sb_internal#2){++++}-{0:0}, at: btrfs_commit_inode_delayed_inode+0x5f/0x120
but task is already holding lock:
ffffffffabe61b40 (fs_reclaim){+.+.}-{0:0}, at: balance_pgdat+0x4aa/0x7a0
which lock already depends on the new lock.
the existing dependency chain (in reverse order) is:
-> #1 (fs_reclaim){+.+.}-{0:0}:
fs_reclaim_acquire+0xa5/0xe0
kmem_cache_alloc+0x31/0x2c0
alloc_extent_state+0x1d/0xd0
__clear_extent_bit+0x2e0/0x4f0
try_release_extent_mapping+0x216/0x280
btrfs_release_folio+0x2e/0x90
invalidate_inode_pages2_range+0x397/0x470
btrfs_cleanup_dirty_bgs+0x9e/0x210
btrfs_cleanup_one_transaction+0x22/0x760
btrfs_commit_transaction+0x3b7/0x13a0
create_subvol+0x59b/0x970
btrfs_mksubvol+0x435/0x4f0
__btrfs_ioctl_snap_create+0x11e/0x1b0
btrfs_ioctl_snap_create_v2+0xbf/0x140
btrfs_ioctl+0xa45/0x28f0
__x64_sys_ioctl+0x88/0xc0
do_syscall_64+0x38/0x90
entry_SYSCALL_64_after_hwframe+0x72/0xdc
-> #0 (sb_internal#2){++++}-{0:0}:
__lock_acquire+0x1435/0x21a0
lock_acquire+0xc2/0x2b0
start_transaction+0x401/0x730
btrfs_commit_inode_delayed_inode+0x5f/0x120
btrfs_evict_inode+0x292/0x3d0
evict+0xcc/0x1d0
inode_lru_isolate+0x14d/0x1e0
__list_lru_walk_one+0xbe/0x1c0
list_lru_walk_one+0x58/0x80
prune_icache_sb+0x39/0x60
super_cache_scan+0x161/0x1f0
do_shrink_slab+0x163/0x340
shrink_slab+0x1d3/0x290
shrink_node+0x300/0x720
balance_pgdat+0x35c/0x7a0
kswapd+0x205/0x410
kthread+0xf0/0x120
ret_from_fork+0x29/0x50
other info that might help us debug this:
Possible unsafe locking scenario:
CPU0 CPU1
---- ----
lock(fs_reclaim);
lock(sb_internal#2);
lock(fs_reclaim);
lock(sb_internal#2);
*** DEADLOCK ***
3 locks held by kswapd0/46:
#0: ffffffffabe61b40 (fs_reclaim){+.+.}-{0:0}, at: balance_pgdat+0x4aa/0x7a0
#1: ffffffffabe50270 (shrinker_rwsem){++++}-{3:3}, at: shrink_slab+0x113/0x290
#2: ffff8c6543abd0e0 (&type->s_umount_key#44){++++}-{3:3}, at: super_cache_scan+0x38/0x1f0
stack backtrace:
CPU: 0 PID: 46 Comm: kswapd0 Not tainted 6.3.0-rc7+ #1167
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.13.0-2.fc32 04/01/2014
Call Trace:
<TASK>
dump_stack_lvl+0x58/0x90
check_noncircular+0xd6/0x100
? save_trace+0x3f/0x310
? add_lock_to_list+0x97/0x120
__lock_acquire+0x1435/0x21a0
lock_acquire+0xc2/0x2b0
? btrfs_commit_inode_delayed_inode+0x5f/0x120
start_transaction+0x401/0x730
? btrfs_commit_inode_delayed_inode+0x5f/0x120
btrfs_commit_inode_delayed_inode+0x5f/0x120
btrfs_evict_inode+0x292/0x3d0
? lock_release+0x134/0x270
? __pfx_wake_bit_function+0x10/0x10
evict+0xcc/0x1d0
inode_lru_isolate+0x14d/0x1e0
__list_lru_walk_one+0xbe/0x1c0
? __pfx_inode_lru_isolate+0x10/0x10
? __pfx_inode_lru_isolate+0x10/0x10
list_lru_walk_one+0x58/0x80
prune_icache_sb+0x39/0x60
super_cache_scan+0x161/0x1f0
do_shrink_slab+0x163/0x340
shrink_slab+0x1d3/0x290
shrink_node+0x300/0x720
balance_pgdat+0x35c/0x7a0
kswapd+0x205/0x410
? __pfx_autoremove_wake_function+0x10/0x10
? __pfx_kswapd+0x10/0x10
kthread+0xf0/0x120
? __pfx_kthread+0x10/0x10
ret_from_fork+0x29/0x50
</TASK>
This happens because when we abort the transaction in the transaction
commit path we call invalidate_inode_pages2_range on our block group
cache inodes (if we have space cache v1) and any delalloc inodes we may
have. The plain invalidate_inode_pages2_range() call passes through
GFP_KERNEL, which makes sense in most cases, but not here. Wrap these
two invalidate callees with memalloc_nofs_save/memalloc_nofs_restore to
make sure we don't end up with the fs reclaim dependency under the
transaction dependency.
CC: stable@vger.kernel.org # 4.14+
Signed-off-by: Josef Bacik <josef@toxicpanda.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Since f8a53bb58e ("btrfs: handle checksum generation in the storage
layer") the failures of btrfs_csum_one_bio() are handled via
bio_end_io().
This means, we can return BLK_STS_RESOURCE from btrfs_csum_one_bio() in
case the allocation of the ordered sums fails.
This also fixes a syzkaller report, where injecting a failure into the
kvzalloc() call results in a BUG_ON().
Reported-by: syzbot+d8941552e21eac774778@syzkaller.appspotmail.com
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Anand Jain <anand.jain@oracle.com>
Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Currently we allow a block group not to be marked read-only for scrub.
But for RAID56 block groups if we require the block group to be
read-only, then we're allowed to use cached content from scrub stripe to
reduce unnecessary RAID56 reads.
So this patch would:
- Make btrfs_inc_block_group_ro() try harder
During my tests, for cases like btrfs/061 and btrfs/064, we can hit
ENOSPC from btrfs_inc_block_group_ro() calls during scrub.
The reason is if we only have one single data chunk, and trying to
scrub it, we won't have any space left for any newer data writes.
But this check should be done by the caller, especially for scrub
cases we only temporarily mark the chunk read-only.
And newer data writes would always try to allocate a new data chunk
when needed.
- Return error for scrub if we failed to mark a RAID56 chunk read-only
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
-----BEGIN PGP SIGNATURE-----
iQIzBAABCgAdFiEE8rQSAMVO+zA4DBdWxWXV+ddtWDsFAmRebDIACgkQxWXV+ddt
WDu3vA//RNyRGjEz0HgfhTc1119DXJLwK6j544waYLrzRcMtBK4xKByiaFkAA4tL
PQidGX+nAQPm+pZl0jcK30cBMObik5GXJwoSOZGl7/ectx4O7aFfXqiSfwPTyqZU
3fTavoqoJxbxJCVbifcXOPNhsUxMlEGYJmA3CVRsllLviXY+3HMpX2ZpWZ7vch+N
MLENNBfUo1HVdWaxOYfQif/qT5iR9G7D8dBjX9DUK0kVwrbwBB0rolJy4fPrY6z5
gBLED9Ks3FBgyU3mYq4qrfPmbfF8mPiaU0+1j+B46vw3PdPtIwjIForR+91GsZ1v
iHojbykf6VWTQV+gO78mgv4O4vRtn3C+UJaGxLL86OMOaiQQHFYdSETn9arPmoho
p1wCBidI82tvfIOGYXgrTGorLN27hhyPJinHe/2Bqo+1wUL8/J8mwCWunIox7a8z
rxO5QhDIDFX7gamsvYjkW3tBkYuGiGvBjx+Ic2cBHTkVp9wSPL9PCvqNNru2qexA
t0BpAL9DxvN+T1xO1thC3qsm2Ogx0QEmgdDfRglbEVASnRZKZZsJEMO90FzFbkFg
vLbs0KnT7yS7mTwq4NklDrgHZ0eiiJLZVCb8bR8xkzVW+ADrUmZuDM8WOcCgJAUp
fUoMmFsJZi5zsdAOygDWr1bBHorLV5szrY0bSB5L2eHwJjYZ6KE=
=uWUN
-----END PGP SIGNATURE-----
Merge tag 'for-6.4-rc1-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux
Pull more btrfs fixes from David Sterba:
- fix incorrect number of bitmap entries for space cache if loading is
interrupted by some error
- fix backref walking, this breaks a mode of LOGICAL_INO_V2 ioctl that
is used in deduplication tools
- zoned mode fixes:
- properly finish zone reserved for relocation
- correctly calculate super block zone end on ZNS
- properly initialize new extent buffer for redirty
- make mount option clear_cache work with block-group-tree, to rebuild
free-space-tree instead of temporarily disabling it that would lead
to a forced read-only mount
- fix alignment check for offset when printing extent item
* tag 'for-6.4-rc1-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux:
btrfs: make clear_cache mount option to rebuild FST without disabling it
btrfs: zero the buffer before marking it dirty in btrfs_redirty_list_add
btrfs: zoned: fix full zone super block reading on ZNS
btrfs: zoned: zone finish data relocation BG with last IO
btrfs: fix backref walking not returning all inode refs
btrfs: fix space cache inconsistency after error loading it from disk
btrfs: print-tree: parent bytenr must be aligned to sector size
Previously clear_cache mount option would simply disable free-space-tree
feature temporarily then re-enable it to rebuild the whole free space
tree.
But this is problematic for block-group-tree feature, as we have an
artificial dependency on free-space-tree feature.
If we go the existing method, after clearing the free-space-tree
feature, we would flip the filesystem to read-only mode, as we detect a
super block write with block-group-tree but no free-space-tree feature.
This patch would change the behavior by properly rebuilding the free
space tree without disabling this feature, thus allowing clear_cache
mount option to work with block group tree.
Now we can mount a filesystem with block-group-tree feature and
clear_mount option:
$ mkfs.btrfs -O block-group-tree /dev/test/scratch1 -f
$ sudo mount /dev/test/scratch1 /mnt/btrfs -o clear_cache
$ sudo dmesg -t | head -n 5
BTRFS info (device dm-1): force clearing of disk cache
BTRFS info (device dm-1): using free space tree
BTRFS info (device dm-1): auto enabling async discard
BTRFS info (device dm-1): rebuilding free space tree
BTRFS info (device dm-1): checking UUID tree
CC: stable@vger.kernel.org # 6.1+
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
btrfs_redirty_list_add zeroes the buffer data and sets the
EXTENT_BUFFER_NO_CHECK to make sure writeback is fine with a bogus
header. But it does that after already marking the buffer dirty, which
means that writeback could already be looking at the buffer.
Switch the order of operations around so that the buffer is only marked
dirty when we're ready to write it.
Fixes: d3575156f6 ("btrfs: zoned: redirty released extent buffers")
CC: stable@vger.kernel.org # 5.15+
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
When both of the superblock zones are full, we need to check which
superblock is newer. The calculation of last superblock position is wrong
as it does not consider zone_capacity and uses the length.
Fixes: 9658b72ef3 ("btrfs: zoned: locate superblock position using zone capacity")
CC: stable@vger.kernel.org # 6.1+
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Naohiro Aota <naohiro.aota@wdc.com>
Signed-off-by: David Sterba <dsterba@suse.com>
For data block groups, we zone finish a zone (or, just deactivate it) when
seeing the last IO in btrfs_finish_ordered_io(). That is only called for
IOs using ZONE_APPEND, but we use a regular WRITE command for data
relocation IOs. Detect it and call btrfs_zone_finish_endio() properly.
Fixes: be1a1d7a5d ("btrfs: zoned: finish fully written block group")
CC: stable@vger.kernel.org # 6.1+
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Naohiro Aota <naohiro.aota@wdc.com>
Signed-off-by: David Sterba <dsterba@suse.com>
When using the logical to ino ioctl v2, if the flag to ignore offsets of
file extent items (BTRFS_LOGICAL_INO_ARGS_IGNORE_OFFSET) is given, the
backref walking code ends up not returning references for all file offsets
of an inode that point to the given logical bytenr. This happens since
kernel 6.2, commit 6ce6ba5344 ("btrfs: use a single argument for extent
offset in backref walking functions") because:
1) It mistakenly skipped the search for file extent items in a leaf that
point to the target extent if that flag is given. Instead it should
only skip the filtering done by check_extent_in_eb() - that is, it
should not avoid the calls to that function (or find_extent_in_eb(),
which uses it).
2) It was also not building a list of inode extent elements (struct
extent_inode_elem) if we have multiple inode references for an extent
when the ignore offset flag is given to the logical to ino ioctl - it
would leave a single element, only the last one that was found.
These stem from the confusing old interface for backref walking functions
where we had an extent item offset argument that was a pointer to a u64
and another boolean argument that indicated if the offset should be
ignored, but the pointer could be NULL. That NULL case is used by
relocation, qgroup extent accounting and fiemap, simply to avoid building
the inode extent list for each reference, as it's not necessary for those
use cases and therefore avoids memory allocations and some computations.
Fix this by adding a boolean argument to the backref walk context
structure to indicate that the inode extent list should not be built,
make relocation set that argument to true and fix the backref walking
logic to skip the calls to check_extent_in_eb() and find_extent_in_eb()
only if this new argument is true, instead of 'ignore_extent_item_pos'
being true.
A test case for fstests will be added soon, to provide cover not only
for these cases but to the logical to ino ioctl in general as well, as
currently we do not have a test case for it.
Reported-by: Vladimir Panteleev <git@vladimir.panteleev.md>
Link: https://lore.kernel.org/linux-btrfs/CAHhfkvwo=nmzrJSqZ2qMfF-rZB-ab6ahHnCD_sq9h4o8v+M7QQ@mail.gmail.com/
Fixes: 6ce6ba5344 ("btrfs: use a single argument for extent offset in backref walking functions")
CC: stable@vger.kernel.org # 6.2+
Tested-by: Vladimir Panteleev <git@vladimir.panteleev.md>
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
When loading a free space cache from disk, at __load_free_space_cache(),
if we fail to insert a bitmap entry, we still increment the number of
total bitmaps in the btrfs_free_space_ctl structure, which is incorrect
since we failed to add the bitmap entry. On error we then empty the
cache by calling __btrfs_remove_free_space_cache(), which will result
in getting the total bitmaps counter set to 1.
A failure to load a free space cache is not critical, so if a failure
happens we just rebuild the cache by scanning the extent tree, which
happens at block-group.c:caching_thread(). Yet the failure will result
in having the total bitmaps of the btrfs_free_space_ctl always bigger
by 1 then the number of bitmap entries we have. So fix this by having
the total bitmaps counter be incremented only if we successfully added
the bitmap entry.
Fixes: a67509c300 ("Btrfs: add a io_ctl struct and helpers for dealing with the space cache")
Reviewed-by: Anand Jain <anand.jain@oracle.com>
CC: stable@vger.kernel.org # 4.4+
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Check nodesize to sectorsize in alignment check in print_extent_item.
The comment states that and this is correct, similar check is done
elsewhere in the functions.
Found by Linux Verification Center (linuxtesting.org) with SVACE.
Fixes: ea57788eb7 ("btrfs: require only sector size alignment for parent eb bytenr")
CC: stable@vger.kernel.org # 4.14+
Reviewed-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: Anastasia Belova <abelova@astralinux.ru>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
-----BEGIN PGP SIGNATURE-----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=K25T
-----END PGP SIGNATURE-----
Merge tag 'for-6.4-rc1-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux
Pull btrfs fixes from David Sterba:
- fix backward leaf iteration which could possibly return the same key
- fix assertion when device add and balance race for exclusive
operation
- fix regression when freeing device, state tree would leak after
device replace
- fix attempt to clear space cache v1 when block-group-tree is enabled
- fix potential i_size corruption when encoded write races with send v2
and enabled no-holes (the race is hard to hit though, the window is a
few instructions wide)
- fix wrong bitmap API use when checking empty zones, parameters were
swapped but not causing a bug due to other code
- prevent potential qgroup leak if subvolume create does not commit
transaction (which is pending in the development queue)
- error handling and reporting:
- abort transaction when sibling keys check fails for leaves
- print extent buffers when sibling keys check fails
* tag 'for-6.4-rc1-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux:
btrfs: don't free qgroup space unless specified
btrfs: fix encoded write i_size corruption with no-holes
btrfs: zoned: fix wrong use of bitops API in btrfs_ensure_empty_zones
btrfs: properly reject clear_cache and v1 cache for block-group-tree
btrfs: print extent buffers when sibling keys check fails
btrfs: abort transaction when sibling keys check fails for leaves
btrfs: fix leak of source device allocation state after device replace
btrfs: fix assertion of exclop condition when starting balance
btrfs: fix btrfs_prev_leaf() to not return the same key twice
Boris noticed in his simple quotas testing that he was getting a leak
with Sweet Tea's change to subvol create that stopped doing a
transaction commit. This was just a side effect of that change.
In the delayed inode code we have an optimization that will free extra
reservations if we think we can pack a dir item into an already modified
leaf. Previously this wouldn't be triggered in the subvolume create
case because we'd commit the transaction, it was still possible but
much harder to trigger. It could actually be triggered if we did a
mkdir && subvol create with qgroups enabled.
This occurs because in btrfs_insert_delayed_dir_index(), which gets
called when we're adding the dir item, we do the following:
btrfs_block_rsv_release(fs_info, trans->block_rsv, bytes, NULL);
if we're able to skip reserving space.
The problem here is that trans->block_rsv points at the temporary block
rsv for the subvolume create, which has qgroup reservations in the block
rsv.
This is a problem because btrfs_block_rsv_release() will do the
following:
if (block_rsv->qgroup_rsv_reserved >= block_rsv->qgroup_rsv_size) {
qgroup_to_release = block_rsv->qgroup_rsv_reserved -
block_rsv->qgroup_rsv_size;
block_rsv->qgroup_rsv_reserved = block_rsv->qgroup_rsv_size;
}
The temporary block rsv just has ->qgroup_rsv_reserved set,
->qgroup_rsv_size == 0. The optimization in
btrfs_insert_delayed_dir_index() sets ->qgroup_rsv_reserved = 0. Then
later on when we call btrfs_subvolume_release_metadata() which has
btrfs_block_rsv_release(fs_info, rsv, (u64)-1, &qgroup_to_release);
btrfs_qgroup_convert_reserved_meta(root, qgroup_to_release);
qgroup_to_release is set to 0, and we do not convert the reserved
metadata space.
The problem here is that the block rsv code has been unconditionally
messing with ->qgroup_rsv_reserved, because the main place this is used
is delalloc, and any time we call btrfs_block_rsv_release() we do it
with qgroup_to_release set, and thus do the proper accounting.
The subvolume code is the only other code that uses the qgroup
reservation stuff, but it's intermingled with the above optimization,
and thus was getting its reservation freed out from underneath it and
thus leaking the reserved space.
The solution is to simply not mess with the qgroup reservations if we
don't have qgroup_to_release set. This works with the existing code as
anything that messes with the delalloc reservations always have
qgroup_to_release set. This fixes the leak that Boris was observing.
Reviewed-by: Qu Wenruo <wqu@suse.com>
CC: stable@vger.kernel.org # 5.4+
Signed-off-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: David Sterba <dsterba@suse.com>
We have observed a btrfs filesystem corruption on workloads using
no-holes and encoded writes via send stream v2. The symptom is that a
file appears to be truncated to the end of its last aligned extent, even
though the final unaligned extent and even the file extent and otherwise
correctly updated inode item have been written.
So if we were writing out a 1MiB+X file via 8 128K extents and one
extent of length X, i_size would be set to 1MiB, but the ninth extent,
nbyte, etc. would all appear correct otherwise.
The source of the race is a narrow (one line of code) window in which a
no-holes fs has read in an updated i_size, but has not yet set a shared
disk_i_size variable to write. Therefore, if two ordered extents run in
parallel (par for the course for receive workloads), the following
sequence can play out: (following "threads" a bit loosely, since there
are callbacks involved for endio but extra threads aren't needed to
cause the issue)
ENC-WR1 (second to last) ENC-WR2 (last)
------- -------
btrfs_do_encoded_write
set i_size = 1M
submit bio B1 ending at 1M
endio B1
btrfs_inode_safe_disk_i_size_write
local i_size = 1M
falls off a cliff for some reason
btrfs_do_encoded_write
set i_size = 1M+X
submit bio B2 ending at 1M+X
endio B2
btrfs_inode_safe_disk_i_size_write
local i_size = 1M+X
disk_i_size = 1M+X
disk_i_size = 1M
btrfs_delayed_update_inode
btrfs_delayed_update_inode
And the delayed inode ends up filled with nbytes=1M+X and isize=1M, and
writes respect i_size and present a corrupted file missing its last
extents.
Fix this by holding the inode lock in the no-holes case so that a thread
can't sneak in a write to disk_i_size that gets overwritten with an out
of date i_size.
Fixes: 41a2ee75aa ("btrfs: introduce per-inode file extent tree")
CC: stable@vger.kernel.org # 5.10+
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Boris Burkov <boris@bur.io>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
find_next_bit and find_next_zero_bit take @size as the second parameter and
@offset as the third parameter. They are specified opposite in
btrfs_ensure_empty_zones(). Thanks to the later loop, it never failed to
detect the empty zones. Fix them and (maybe) return the result a bit
faster.
Note: the naming is a bit confusing, size has two meanings here, bitmap
and our range size.
Fixes: 1cd6121f2a ("btrfs: zoned: implement zoned chunk allocator")
CC: stable@vger.kernel.org # 5.15+
Signed-off-by: Naohiro Aota <naohiro.aota@wdc.com>
Signed-off-by: David Sterba <dsterba@suse.com>
[BUG]
With block-group-tree feature enabled, mounting it with clear_cache
would cause the following transaction abort at mount or remount:
BTRFS info (device dm-4): force clearing of disk cache
BTRFS info (device dm-4): using free space tree
BTRFS info (device dm-4): auto enabling async discard
BTRFS info (device dm-4): clearing free space tree
BTRFS info (device dm-4): clearing compat-ro feature flag for FREE_SPACE_TREE (0x1)
BTRFS info (device dm-4): clearing compat-ro feature flag for FREE_SPACE_TREE_VALID (0x2)
BTRFS error (device dm-4): block-group-tree feature requires fres-space-tree and no-holes
BTRFS error (device dm-4): super block corruption detected before writing it to disk
BTRFS: error (device dm-4) in write_all_supers:4288: errno=-117 Filesystem corrupted (unexpected superblock corruption detected)
BTRFS warning (device dm-4: state E): Skipping commit of aborted transaction.
[CAUSE]
For block-group-tree feature, we have an artificial dependency on
free-space-tree.
This means if we detect block-group-tree without v2 cache, we consider
it a corruption and cause the problem.
For clear_cache mount option, it would temporary disable v2 cache, then
re-enable it.
But unfortunately for that temporary v2 cache disabled status, we refuse
to write a superblock with bg tree only flag, thus leads to the above
transaction abortion.
[FIX]
For now, just reject clear_cache and v1 cache mount option for block
group tree. So now we got a graceful rejection other than a transaction
abort:
BTRFS info (device dm-4): force clearing of disk cache
BTRFS error (device dm-4): cannot disable free space tree with block-group-tree feature
BTRFS error (device dm-4): open_ctree failed
CC: stable@vger.kernel.org # 6.1+
Signed-off-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
When trying to move keys from one node/leaf to another sibling node/leaf,
if the sibling keys check fails we just print an error message with the
last key of the left sibling and the first key of the right sibling.
However it's also useful to print all the keys of each sibling, as it
may provide some clues to what went wrong, which code path may be
inserting keys in an incorrect order. So just do that, print the siblings
with btrfs_print_tree(), as it works for both leaves and nodes.
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>
If the sibling keys check fails before we move keys from one sibling
leaf to another, we are not aborting the transaction - we leave that to
some higher level caller of btrfs_search_slot() (or anything else that
uses it to insert items into a b+tree).
This means that the transaction abort will provide a stack trace that
omits the b+tree modification call chain. So change this to immediately
abort the transaction and therefore get a more useful stack trace that
shows us the call chain in the bt+tree modification code.
It's also important to immediately abort the transaction just in case
some higher level caller is not doing it, as this indicates a very
serious corruption and we should stop the possibility of doing further
damage.
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 a device replace finishes, the source device is freed by calling
btrfs_free_device() at btrfs_rm_dev_replace_free_srcdev(), but the
allocation state, tracked in the device's alloc_state io tree, is never
freed.
This is a regression recently introduced by commit f0bb5474cf ("btrfs:
remove redundant release of btrfs_device::alloc_state"), which removed a
call to extent_io_tree_release() from btrfs_free_device(), with the
rationale that btrfs_close_one_device() already releases the allocation
state from a device and btrfs_close_one_device() is always called before
a device is freed with btrfs_free_device(). However that is not true for
the device replace case, as btrfs_free_device() is called without any
previous call to btrfs_close_one_device().
The issue is trivial to reproduce, for example, by running test btrfs/027
from fstests:
$ ./check btrfs/027
$ rmmod btrfs
$ dmesg
(...)
[84519.395485] BTRFS info (device sdc): dev_replace from <missing disk> (devid 2) to /dev/sdg started
[84519.466224] BTRFS info (device sdc): dev_replace from <missing disk> (devid 2) to /dev/sdg finished
[84519.552251] BTRFS info (device sdc): scrub: started on devid 1
[84519.552277] BTRFS info (device sdc): scrub: started on devid 2
[84519.552332] BTRFS info (device sdc): scrub: started on devid 3
[84519.552705] BTRFS info (device sdc): scrub: started on devid 4
[84519.604261] BTRFS info (device sdc): scrub: finished on devid 4 with status: 0
[84519.609374] BTRFS info (device sdc): scrub: finished on devid 3 with status: 0
[84519.610818] BTRFS info (device sdc): scrub: finished on devid 1 with status: 0
[84519.610927] BTRFS info (device sdc): scrub: finished on devid 2 with status: 0
[84559.503795] BTRFS: state leak: start 1048576 end 1351614463 state 1 in tree 1 refs 1
[84559.506764] BTRFS: state leak: start 1048576 end 1347420159 state 1 in tree 1 refs 1
[84559.510294] BTRFS: state leak: start 1048576 end 1351614463 state 1 in tree 1 refs 1
So fix this by adding back the call to extent_io_tree_release() at
btrfs_free_device().
Fixes: f0bb5474cf ("btrfs: remove redundant release of btrfs_device::alloc_state")
Reviewed-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
A call to btrfs_prev_leaf() may end up returning a path that points to the
same item (key) again. This happens if while btrfs_prev_leaf(), after we
release the path, a concurrent insertion happens, which moves items off
from a sibling into the front of the previous leaf, and an item with the
computed previous key does not exists.
For example, suppose we have the two following leaves:
Leaf A
-------------------------------------------------------------
| ... key (300 96 10) key (300 96 15) key (300 96 16) |
-------------------------------------------------------------
slot 20 slot 21 slot 22
Leaf B
-------------------------------------------------------------
| key (300 96 20) key (300 96 21) key (300 96 22) ... |
-------------------------------------------------------------
slot 0 slot 1 slot 2
If we call btrfs_prev_leaf(), from btrfs_previous_item() for example, with
a path pointing to leaf B and slot 0 and the following happens:
1) At btrfs_prev_leaf() we compute the previous key to search as:
(300 96 19), which is a key that does not exists in the tree;
2) Then we call btrfs_release_path() at btrfs_prev_leaf();
3) Some other task inserts a key at leaf A, that sorts before the key at
slot 20, for example it has an objectid of 299. In order to make room
for the new key, the key at slot 22 is moved to the front of leaf B.
This happens at push_leaf_right(), called from split_leaf().
After this leaf B now looks like:
--------------------------------------------------------------------------------
| key (300 96 16) key (300 96 20) key (300 96 21) key (300 96 22) ... |
--------------------------------------------------------------------------------
slot 0 slot 1 slot 2 slot 3
4) At btrfs_prev_leaf() we call btrfs_search_slot() for the computed
previous key: (300 96 19). Since the key does not exists,
btrfs_search_slot() returns 1 and with a path pointing to leaf B
and slot 1, the item with key (300 96 20);
5) This makes btrfs_prev_leaf() return a path that points to slot 1 of
leaf B, the same key as before it was called, since the key at slot 0
of leaf B (300 96 16) is less than the computed previous key, which is
(300 96 19);
6) As a consequence btrfs_previous_item() returns a path that points again
to the item with key (300 96 20).
For some users of btrfs_prev_leaf() or btrfs_previous_item() this may not
be functional a problem, despite not making sense to return a new path
pointing again to the same item/key. However for a caller such as
tree-log.c:log_dir_items(), this has a bad consequence, as it can result
in not logging some dir index deletions in case the directory is being
logged without holding the inode's VFS lock (logging triggered while
logging a child inode for example) - for the example scenario above, in
case the dir index keys 17, 18 and 19 were deleted in the current
transaction.
CC: stable@vger.kernel.org # 4.14+
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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=Jt7p
-----END PGP SIGNATURE-----
Merge tag 'for-6.4-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux
Pull btrfs updates from David Sterba:
"Mostly core changes and cleanups, some notable fixes and two
performance improvements in directory logging.
The IO path cleanups are removing or refactoring old code, scrub main
loop has been completely rewritten also refactoring old code.
There are some changes to non-btrfs code, mostly trivial, the cgroup
punt bio logic is only moved from generic code.
Performance improvements:
- improve logging changes in a directory during one transaction,
avoid iterating over items and reduce lock contention (fsync time
4x lower)
- when logging directory entries during one transaction, reduce
locking of subvolume trees by checking tree-log instead
(improvement in throughput and latency for concurrent access to a
subvolume)
Notable fixes:
- dev-replace:
- properly honor read mode when requested to avoid reading from
source device
- target device won't be used for eventual read repair, this is
unreliable for NODATASUM files
- when there are unpaired (and unrepairable) metadata during
replace, exit early with error and don't try to finish whole
operation
- scrub ioctl properly rejects unknown flags
- fix global block reserve calculations
- fix partial direct io write when there's a page fault in the
middle, iomap will try to continue with partial request but the
btrfs part did not match that, this can lead to zeros written
instead of data
Core changes:
- io path:
- continued cleanups and refactoring around bio handling
- extent io submit path simplifications and cleanups
- flush write path simplifications and cleanups
- rework logic of passing sync mode of bio, with further cleanups
- rewrite scrub code flow, restructure how the stripes are enumerated
and verified in a more unified way
- allow to set lower threshold for block group reclaim in debug mode
to aid zoned mode testing
- remove obsolete time-based delayed ref throttling logic when
truncating items
- DREW locks are not using percpu variables anymore
- more warning fixes (-Wmaybe-uninitialized)
- u64 division simplifications
- error handling improvements
Non-btrfs code changes:
- push cgroup punt bio logic to btrfs code (there was no other user
of that), the functionality can be now selected separately by
BLK_CGROUP_PUNT_BIO
- crc32c_impl removed after removing last uses in btrfs code
- add btrfs_assertfail() to objtool table"
* tag 'for-6.4-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux: (147 commits)
btrfs: mark btrfs_assertfail() __noreturn
btrfs: fix uninitialized variable warnings
btrfs: use log root when iterating over index keys when logging directory
btrfs: avoid iterating over all indexes when logging directory
btrfs: dev-replace: error out if we have unrepaired metadata error during
btrfs: remove pointless loop at btrfs_get_next_valid_item()
btrfs: scrub: reject unsupported scrub flags
btrfs: reinterpret async discard iops_limit=0 as no delay
btrfs: set default discard iops_limit to 1000
btrfs: remove unused raid56 functions which were dedicated for scrub
btrfs: scrub: remove scrub_bio structure
btrfs: scrub: remove scrub_block and scrub_sector structures
btrfs: scrub: remove the old scrub recheck code
btrfs: scrub: remove the old writeback infrastructure
btrfs: scrub: remove scrub_parity structure
btrfs: scrub: use scrub_stripe to implement RAID56 P/Q scrub
btrfs: scrub: switch scrub_simple_mirror() to scrub_stripe infrastructure
btrfs: scrub: introduce helper to queue a stripe for scrub
btrfs: scrub: introduce error reporting functionality for scrub_stripe
btrfs: scrub: introduce a writeback helper for scrub_stripe
...
-----BEGIN PGP SIGNATURE-----
iHUEABYKAB0WIQRAhzRXHqcMeLMyaSiRxhvAZXjcogUCZEEhwgAKCRCRxhvAZXjc
otwgAQDXHnKiPm/d76lITXbxdUNCtvZz+ig26EbOrD+vEszzIQEA81dru0QbCNCt
ctoZdcsmtKbt2VaYQF1CDOhlnNg5VQM=
=pER1
-----END PGP SIGNATURE-----
Merge tag 'v6.4/vfs.acl' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs
Pull acl updates from Christian Brauner:
"After finishing the introduction of the new posix acl api last cycle
the generic POSIX ACL xattr handlers are still around in the
filesystems xattr handlers for two reasons:
(1) Because a few filesystems rely on the ->list() method of the
generic POSIX ACL xattr handlers in their ->listxattr() inode
operation.
(2) POSIX ACLs are only available if IOP_XATTR is raised. The
IOP_XATTR flag is raised in inode_init_always() based on whether
the sb->s_xattr pointer is non-NULL. IOW, the registered xattr
handlers of the filesystem are used to raise IOP_XATTR. Removing
the generic POSIX ACL xattr handlers from all filesystems would
risk regressing filesystems that only implement POSIX ACL support
and no other xattrs (nfs3 comes to mind).
This contains the work to decouple POSIX ACLs from the IOP_XATTR flag
as they don't depend on xattr handlers anymore. So it's now possible
to remove the generic POSIX ACL xattr handlers from the sb->s_xattr
list of all filesystems. This is a crucial step as the generic POSIX
ACL xattr handlers aren't used for POSIX ACLs anymore and POSIX ACLs
don't depend on the xattr infrastructure anymore.
Adressing problem (1) will require more long-term work. It would be
best to get rid of the ->list() method of xattr handlers completely at
some point.
For erofs, ext{2,4}, f2fs, jffs2, ocfs2, and reiserfs the nop POSIX
ACL xattr handler is kept around so they can continue to use
array-based xattr handler indexing.
This update does simplify the ->listxattr() implementation of all
these filesystems however"
* tag 'v6.4/vfs.acl' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs:
acl: don't depend on IOP_XATTR
ovl: check for ->listxattr() support
reiserfs: rework priv inode handling
fs: rename generic posix acl handlers
reiserfs: rework ->listxattr() implementation
fs: simplify ->listxattr() implementation
fs: drop unused posix acl handlers
xattr: remove unused argument
xattr: add listxattr helper
xattr: simplify listxattr helpers
-----BEGIN PGP SIGNATURE-----
iQJEBAABCAAuFiEEwPw5LcreJtl1+l5K99NY+ylx4KYFAmRCvdsQHGF4Ym9lQGtl
cm5lbC5kawAKCRD301j7KXHgpg4oD/457EJ21Fm36NuyT/S0Cr8ok9Tdk7t9BeBh
V/9CYThoXr5aqAox0Vq23FF+Rhzm81GzwYERN4493LBblliNeNOo2IaXF9/7qrUW
11v9Bkug2J3k3hRGtEa6Zl0EpMu+FRLsNpchjFS2KPuOq+iMDxrvwuy50kidWg7n
r25e4UwpExVO9fIoUSmzgWVfRHOTuj9yiG/UsaH2+2BRXerIX0Q1tyElwmcGh25M
Ad2hN+yDnuIbNA5gNUpnzY32Dp0zjAsquc//QOvq9mltcNTElokB8idGliismvyd
8qF0lkwQwewOBT/sSD5EY3K0Qd8IJu425bvT/yPUDScHz1chxHUoxo5eisIr2M9l
5AL5KHAf7Zzs8ZuV+IYPzZ5qM6a/vF3mHUisKRNKYVhF46Nmd4cBratfXwWb1MxV
clQM2qr0TLOYli9mOeTXph3hg/rBVqKqf90boAZoN8b2tWBKlMykpqRadbepjrgx
bmBSwwAF99NxIHEjU3U5DMdUloCSiMZIfMfDxQrPNDrfWAW4xJs5Ym0VeOjEotTt
oFEs1fr6c3Mn7KEuPPfOtnDxvs51IP/B8+gDgMt/edf+wHiCU1Zm31u2gxt2dsKh
g73Y92i5SHjIf36H5szBTeioyMy1E1VA9HF14xWz2eKdQ+wxQ9VNWoctcJ85k3F4
6AZDYRIrWA==
=EaE9
-----END PGP SIGNATURE-----
Merge tag 'iter-ubuf.2-2023-04-21' of git://git.kernel.dk/linux
Pull ITER_UBUF updates from Jens Axboe:
"This turns singe vector imports into ITER_UBUF, rather than
ITER_IOVEC.
The former is more trivial to iterate and advance, and hence a bit
more efficient. From some very unscientific testing, ~60% of all iovec
imports are single vector"
* tag 'iter-ubuf.2-2023-04-21' of git://git.kernel.dk/linux:
iov_iter: Mark copy_compat_iovec_from_user() noinline
iov_iter: import single vector iovecs as ITER_UBUF
iov_iter: convert import_single_range() to ITER_UBUF
iov_iter: overlay struct iovec and ubuf/len
iov_iter: set nr_segs = 1 for ITER_UBUF
iov_iter: remove iov_iter_iovec()
iov_iter: add iter_iov_addr() and iter_iov_len() helpers
ALSA: pcm: check for user backed iterator, not specific iterator type
IB/qib: check for user backed iterator, not specific iterator type
IB/hfi1: check for user backed iterator, not specific iterator type
iov_iter: add iter_iovec() helper
block: ensure bio_alloc_map_data() deals with ITER_UBUF correctly
-----BEGIN PGP SIGNATURE-----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=mGL5
-----END PGP SIGNATURE-----
Merge tag 'for-6.3-rc7-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux
Pull btrfs fixes from David Sterba:
"Two patches fixing the problem with aync discard.
The default settings had a low IOPS limit and processing a large batch
to discard would take a long time. On laptops this can cause increased
power consumption due to disk activity.
As async discard has been on by default since 6.2 this likely affects
a lot of users.
Summary:
- increase the default IOPS limit 10x which reportedly helped
- setting the sysfs IOPS value to 0 now does not throttle anymore
allowing the discards to be processed at full speed. Previously
there was an arbitrary 6 hour target for processing the pending
batch"
* tag 'for-6.3-rc7-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux:
btrfs: reinterpret async discard iops_limit=0 as no delay
btrfs: set default discard iops_limit to 1000
Currently, a limit of 0 results in a hard coded metering over 6 hours.
Since the default is a set limit, I suspect no one truly depends on this
rather arbitrary setting. Repurpose it for an arguably more useful
"unlimited" mode, where the delay is 0.
Note that if block groups are too new, or go fully empty, there is still
a delay associated with those conditions. Those delays implement
heuristics for not trimming a region we are relatively likely to fully
overwrite soon.
CC: stable@vger.kernel.org # 6.2+
Reviewed-by: Neal Gompa <neal@gompa.dev>
Signed-off-by: Boris Burkov <boris@bur.io>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Previously, the default was a relatively conservative 10. This results
in a 100ms delay, so with ~300 discards in a commit, it takes the full
30s till the next commit to finish the discards. On a workstation, this
results in the disk never going idle, wasting power/battery, etc.
Set the default to 1000, which results in using the smallest possible
delay, currently, which is 1ms. This has shown to not pathologically
keep the disk busy by the original reporter.
Link: https://lore.kernel.org/linux-btrfs/Y%2F+n1wS%2F4XAH7X1p@nz/
Link: https://bugzilla.redhat.com/show_bug.cgi?id=2182228
CC: stable@vger.kernel.org # 6.2+
Reviewed-by: Neal Gompa <neal@gompa.dev
Signed-off-by: Boris Burkov <boris@bur.io>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
There are some warnings on older compilers (gcc 10, 7) or non-x86_64
architectures (aarch64). As btrfs wants to enable -Wmaybe-uninitialized
by default, fix the warnings even though it's not necessary on recent
compilers (gcc 12+).
../fs/btrfs/volumes.c: In function ‘btrfs_init_new_device’:
../fs/btrfs/volumes.c:2703:3: error: ‘seed_devices’ may be used uninitialized in this function [-Werror=maybe-uninitialized]
2703 | btrfs_setup_sprout(fs_info, seed_devices);
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
../fs/btrfs/send.c: In function ‘get_cur_inode_state’:
../include/linux/compiler.h:70:32: error: ‘right_gen’ may be used uninitialized in this function [-Werror=maybe-uninitialized]
70 | (__if_trace.miss_hit[1]++,1) : \
| ^
../fs/btrfs/send.c:1878:6: note: ‘right_gen’ was declared here
1878 | u64 right_gen;
| ^~~~~~~~~
Reported-by: k2ci <kernel-bot@kylinos.cn>
Signed-off-by: Genjian Zhang <zhanggenjian@kylinos.cn>
Reviewed-by: David Sterba <dsterba@suse.com>
[ update changelog ]
Signed-off-by: David Sterba <dsterba@suse.com>
When logging dir dentries of a directory, we iterate over the subvolume
tree to find dir index keys on leaves modified in the current transaction.
This however is heavy on locking, since btrfs_search_forward() may often
keep locks on extent buffers for quite a while when walking the tree to
find a suitable leaf modified in the current transaction and with a key
not smaller than then the provided minimum key. That means it will block
other tasks trying to access the subvolume tree, which may be common fs
operations like creating, renaming, linking, unlinking, reflinking files,
etc.
A better solution is to iterate the log tree, since it's much smaller than
a subvolume tree and just use plain btrfs_search_slot() (or the wrapper
btrfs_for_each_slot()) and only contains dir index keys added in the
current transaction.
The following bonnie++ test on a non-debug kernel (with Debian's default
kernel config) on a 20G null block device, was used to measure the impact:
$ cat test.sh
#!/bin/bash
DEV=/dev/nullb0
MNT=/mnt/nullb0
NR_DIRECTORIES=20
NR_FILES=20480 # must be a multiple of 1024
DATASET_SIZE=$(( (8 * 1024 * 1024 * 1024) / 1048576 )) # 8 GiB as megabytes
DIRECTORY_SIZE=$(( DATASET_SIZE / NR_FILES ))
NR_FILES=$(( NR_FILES / 1024 ))
umount $DEV &> /dev/null
mkfs.btrfs -f $DEV
mount $DEV $MNT
bonnie++ -u root -d $MNT \
-n $NR_FILES:$DIRECTORY_SIZE:$DIRECTORY_SIZE:$NR_DIRECTORIES \
-r 0 -s $DATASET_SIZE -b
umount $MNT
Before patchset:
Version 2.00a ------Sequential Output------ --Sequential Input- --Random-
-Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
Name:Size etc /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP
debian0 8G 376k 99 1.1g 98 939m 92 1527k 99 3.2g 99 9060 256
Latency 24920us 207us 680ms 5594us 171us 2891us
Version 2.00a ------Sequential Create------ --------Random Create--------
debian0 -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete--
files /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP
20/20 20480 96 +++++ +++ 20480 95 20480 99 +++++ +++ 20480 97
Latency 8708us 137us 5128us 6743us 60us 19712us
After patchset:
Version 2.00a ------Sequential Output------ --Sequential Input- --Random-
-Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
Name:Size etc /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP
debian0 8G 384k 99 1.2g 99 971m 91 1533k 99 3.3g 99 9180 309
Latency 24930us 125us 661ms 5587us 46us 2020us
Version 2.00a ------Sequential Create------ --------Random Create--------
debian0 -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete--
files /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP
20/20 20480 90 +++++ +++ 20480 99 20480 99 +++++ +++ 20480 97
Latency 7030us 61us 1246us 4942us 56us 16855us
The patchset consists of this patch plus a previous one that has the
following subject:
"btrfs: avoid iterating over all indexes when logging directory"
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
When logging a directory, after copying all directory index items from the
subvolume tree to the log tree, we iterate over the subvolume tree to find
all dir index items that are located in leaves COWed (or created) in the
current transaction. If we keep logging a directory several times during
the same transaction, we end up iterating over the same dir index items
everytime we log the directory, wasting time and adding extra lock
contention on the subvolume tree.
So just keep track of the last logged dir index offset in order to start
the search for that index (+1) the next time the directory is logged, as
dir index values (key offsets) come from a monotonically increasing
counter.
The following test measures the difference before and after this change:
$ cat test.sh
#!/bin/bash
DEV=/dev/nullb0
MNT=/mnt/nullb0
umount $DEV &> /dev/null
mkfs.btrfs -f $DEV
mount -o ssd $DEV $MNT
# Time values in milliseconds.
declare -a fsync_times
# Total number of files added to the test directory.
num_files=1000000
# Fsync directory after every N files are added.
fsync_period=100
mkdir $MNT/testdir
fsync_total_time=0
for ((i = 1; i <= $num_files; i++)); do
echo -n > $MNT/testdir/file_$i
if [ $((i % fsync_period)) -eq 0 ]; then
start=$(date +%s%N)
xfs_io -c "fsync" $MNT/testdir
end=$(date +%s%N)
fsync_total_time=$((fsync_total_time + (end - start)))
fsync_times[i]=$(( (end - start) / 1000000 ))
echo -n -e "Progress $i / $num_files\r"
fi
done
echo -e "\nHistogram of directory fsync duration in ms:\n"
printf '%s\n' "${fsync_times[@]}" | \
perl -MStatistics::Histogram -e '@d = <>; print get_histogram(\@d);'
fsync_total_time=$((fsync_total_time / 1000000))
echo -e "\nTotal time spent in fsync: $fsync_total_time ms\n"
echo
umount $MNT
The test was run on a non-debug kernel (Debian's default kernel config)
against a 15G null block device.
Result before this change:
Histogram of directory fsync duration in ms:
Count: 10000
Range: 3.000 - 362.000; Mean: 34.556; Median: 31.000; Stddev: 25.751
Percentiles: 90th: 71.000; 95th: 77.000; 99th: 81.000
3.000 - 5.278: 1423 #################################
5.278 - 8.854: 1173 ###########################
8.854 - 14.467: 591 ##############
14.467 - 23.277: 1025 #######################
23.277 - 37.105: 1422 #################################
37.105 - 58.809: 2036 ###############################################
58.809 - 92.876: 2316 #####################################################
92.876 - 146.346: 6 |
146.346 - 230.271: 6 |
230.271 - 362.000: 2 |
Total time spent in fsync: 350527 ms
Result after this change:
Histogram of directory fsync duration in ms:
Count: 10000
Range: 3.000 - 1088.000; Mean: 8.704; Median: 8.000; Stddev: 12.576
Percentiles: 90th: 12.000; 95th: 14.000; 99th: 17.000
3.000 - 6.007: 3222 #################################
6.007 - 11.276: 5197 #####################################################
11.276 - 20.506: 1551 ################
20.506 - 36.674: 24 |
36.674 - 201.552: 1 |
201.552 - 353.841: 4 |
353.841 - 1088.000: 1 |
Total time spent in fsync: 92114 ms
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
[BUG]
Even before the scrub rework, if we have some corrupted metadata failed
to be repaired during replace, we still continue replacing and let it
finish just as there is nothing wrong:
BTRFS info (device dm-4): dev_replace from /dev/mapper/test-scratch1 (devid 1) to /dev/mapper/test-scratch2 started
BTRFS warning (device dm-4): tree block 5578752 mirror 1 has bad csum, has 0x00000000 want 0xade80ca1
BTRFS warning (device dm-4): tree block 5578752 mirror 0 has bad csum, has 0x00000000 want 0xade80ca1
BTRFS warning (device dm-4): checksum error at logical 5578752 on dev /dev/mapper/test-scratch1, physical 5578752: metadata leaf (level 0) in tree 5
BTRFS warning (device dm-4): checksum error at logical 5578752 on dev /dev/mapper/test-scratch1, physical 5578752: metadata leaf (level 0) in tree 5
BTRFS error (device dm-4): bdev /dev/mapper/test-scratch1 errs: wr 0, rd 0, flush 0, corrupt 1, gen 0
BTRFS warning (device dm-4): tree block 5578752 mirror 1 has bad bytenr, has 0 want 5578752
BTRFS error (device dm-4): unable to fixup (regular) error at logical 5578752 on dev /dev/mapper/test-scratch1
BTRFS info (device dm-4): dev_replace from /dev/mapper/test-scratch1 (devid 1) to /dev/mapper/test-scratch2 finished
This can lead to unexpected problems for the resulting filesystem.
[CAUSE]
Btrfs reuses scrub code path for dev-replace to iterate all dev extents.
But unlike scrub, dev-replace doesn't really bother to check the scrub
progress, which records all the errors found during replace.
And even if we check the progress, we cannot really determine which
errors are minor, which are critical just by the plain numbers.
(remember we don't treat metadata/data checksum error differently).
This behavior is there from the very beginning.
[FIX]
Instead of continuing the replace, just error out if we hit an
unrepaired metadata sector.
Now the dev-replace would be rejected with -EIO, to let the user know.
Although it also means, the filesystem has some metadata error which
cannot be repaired, the user would be upset anyway.
The new dmesg would look like this:
BTRFS info (device dm-4): dev_replace from /dev/mapper/test-scratch1 (devid 1) to /dev/mapper/test-scratch2 started
BTRFS warning (device dm-4): tree block 5578752 mirror 1 has bad csum, has 0x00000000 want 0xade80ca1
BTRFS warning (device dm-4): tree block 5578752 mirror 1 has bad csum, has 0x00000000 want 0xade80ca1
BTRFS error (device dm-4): unable to fixup (regular) error at logical 5570560 on dev /dev/mapper/test-scratch1 physical 5570560
BTRFS warning (device dm-4): header error at logical 5570560 on dev /dev/mapper/test-scratch1, physical 5570560: metadata leaf (level 0) in tree 5
BTRFS warning (device dm-4): header error at logical 5570560 on dev /dev/mapper/test-scratch1, physical 5570560: metadata leaf (level 0) in tree 5
BTRFS error (device dm-4): stripe 5570560 has unrepaired metadata sector at 5578752
BTRFS error (device dm-4): btrfs_scrub_dev(/dev/mapper/test-scratch1, 1, /dev/mapper/test-scratch2) failed -5
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
It's pointless to have a while loop at btrfs_get_next_valid_item(), as if
the slot on the current leaf is beyond the last item, we call
btrfs_next_leaf(), which leaves us at a valid slot of the next leaf (or
a valid slot in the current leaf if after releasing the path an item gets
pushed from the next leaf to the current leaf).
So just call btrfs_next_leaf() if the current slot on the current leaf is
beyond the last item.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Since the introduction of scrub interface, the only flag that we support
is BTRFS_SCRUB_READONLY. Thus there is no sanity checks, if there are
some undefined flags passed in, we just ignore them.
This is problematic if we want to introduce new scrub flags, as we have
no way to determine if such flags are supported.
Address the problem by introducing a check for the flags, and if
unsupported flags are set, return -EOPNOTSUPP to inform the user space.
This check should be backported for all supported kernels before any new
scrub flags are introduced.
CC: stable@vger.kernel.org # 4.14+
Reviewed-by: Anand Jain <anand.jain@oracle.com>
Signed-off-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Currently, a limit of 0 results in a hard coded metering over 6 hours.
Since the default is a set limit, I suspect no one truly depends on this
rather arbitrary setting. Repurpose it for an arguably more useful
"unlimited" mode, where the delay is 0.
Note that if block groups are too new, or go fully empty, there is still
a delay associated with those conditions. Those delays implement
heuristics for not trimming a region we are relatively likely to fully
overwrite soon.
CC: stable@vger.kernel.org # 6.2+
Reviewed-by: Neal Gompa <neal@gompa.dev>
Signed-off-by: Boris Burkov <boris@bur.io>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Previously, the default was a relatively conservative 10. This results
in a 100ms delay, so with ~300 discards in a commit, it takes the full
30s till the next commit to finish the discards. On a workstation, this
results in the disk never going idle, wasting power/battery, etc.
Set the default to 1000, which results in using the smallest possible
delay, currently, which is 1ms. This has shown to not pathologically
keep the disk busy by the original reporter.
Link: https://lore.kernel.org/linux-btrfs/Y%2F+n1wS%2F4XAH7X1p@nz/
Link: https://bugzilla.redhat.com/show_bug.cgi?id=2182228
CC: stable@vger.kernel.org # 6.2+
Reviewed-by: Neal Gompa <neal@gompa.dev
Signed-off-by: Boris Burkov <boris@bur.io>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Since the scrub rework, the following RAID56 functions are no longer
called:
- raid56_add_scrub_pages()
- raid56_alloc_missing_rbio()
- raid56_submit_missing_rbio()
Those functions are all utilized by scrub to handle missing device cases
for RAID56.
However the new scrub code handle them in a completely different way:
- If it's data stripe, go recovery path through btrfs_submit_bio()
- If it's P/Q stripe, it would be handled through
raid56_parity_submit_scrub_rbio()
And that function would handle dev-replace and repair properly.
Thus we can safely remove those functions.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Since scrub path has been fully moved to scrub_stripe based facilities,
no more scrub_bio would be submitted.
Thus we can remove it completely, this involves:
- SCRUB_SECTORS_PER_BIO macro
- SCRUB_BIOS_PER_SCTX macro
- SCRUB_MAX_PAGES macro
- BTRFS_MAX_MIRRORS macro
- scrub_bio structure
- scrub_ctx::bios member
- scrub_ctx::curr member
- scrub_ctx::bios_in_flight member
- scrub_ctx::workers_pending member
- scrub_ctx::list_lock member
- scrub_ctx::list_wait member
- function scrub_bio_end_io_worker()
- function scrub_pending_bio_inc()
- function scrub_pending_bio_dec()
- function scrub_throttle()
- function scrub_submit()
- function scrub_find_csum()
- function drop_csum_range()
- Some unnecessary flush and scrub pauses
Signed-off-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Those two structures are used to represent a bunch of sectors for scrub,
but now they are fully replaced by scrub_stripe in one go, so we can
remove them. This involves:
- structure scrub_block
- structure scrub_sector
- structure scrub_page_private
- function attach_scrub_page_private()
- function detach_scrub_page_private()
Now we no longer need to use page::private to handle subpage.
- function alloc_scrub_block()
- function alloc_scrub_sector()
- function scrub_sector_get_page()
- function scrub_sector_get_page_offset()
- function scrub_sector_get_kaddr()
- function bio_add_scrub_sector()
- function scrub_checksum_data()
- function scrub_checksum_tree_block()
- function scrub_checksum_super()
- function scrub_check_fsid()
- function scrub_block_get()
- function scrub_block_put()
- function scrub_sector_get()
- function scrub_sector_put()
- function scrub_bio_end_io()
- function scrub_block_complete()
- function scrub_add_sector_to_rd_bio()
Signed-off-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
The old scrub code has different entrance to verify the content, and
since we have removed the writeback path, now we can start removing the
re-check part, including:
- scrub_recover structure
- scrub_sector::recover member
- function scrub_setup_recheck_block()
- function scrub_recheck_block()
- function scrub_recheck_block_checksum()
- function scrub_repair_block_group_good_copy()
- function scrub_repair_sector_from_good_copy()
- function scrub_is_page_on_raid56()
- function full_stripe_lock()
- function search_full_stripe_lock()
- function get_full_stripe_logical()
- function insert_full_stripe_lock()
- function lock_full_stripe()
- function unlock_full_stripe()
- btrfs_block_group::full_stripe_locks_root member
- btrfs_full_stripe_locks_tree structure
This infrastructure is to ensure RAID56 scrub is properly handling
recovery and P/Q scrub correctly.
This is no longer needed, before P/Q scrub we will wait for all
the involved data stripes to be scrubbed first, and RAID56 code has
internal lock to ensure no race in the same full stripe.
- function scrub_print_warning()
- function scrub_get_recover()
- function scrub_put_recover()
- function scrub_handle_errored_block()
- function scrub_setup_recheck_block()
- function scrub_bio_wait_endio()
- function scrub_submit_raid56_bio_wait()
- function scrub_recheck_block_on_raid56()
- function scrub_recheck_block()
- function scrub_recheck_block_checksum()
- function scrub_repair_block_from_good_copy()
- function scrub_repair_sector_from_good_copy()
And two more functions exported temporarily for later cleanup:
- alloc_scrub_sector()
- alloc_scrub_block()
Signed-off-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Since the whole scrub path has been switched to scrub_stripe based
solution, the old writeback path can be removed completely, which
involves:
- scrub_ctx::wr_curr_bio member
- scrub_ctx::flush_all_writes member
- function scrub_write_block_to_dev_replace()
- function scrub_write_sector_to_dev_replace()
- function scrub_add_sector_to_wr_bio()
- function scrub_wr_submit()
- function scrub_wr_bio_end_io()
- function scrub_wr_bio_end_io_worker()
And one more function needs to be exported temporarily:
- scrub_sector_get()
Signed-off-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
The structure scrub_parity is used to indicate that some extents are
scrubbed for the purpose of RAID56 P/Q scrubbing.
Since the whole RAID56 P/Q scrubbing path has been replaced with new
scrub_stripe infrastructure, and we no longer need to use scrub_parity
to modify the behavior of data stripes, we can remove it completely.
This removal involves:
- scrub_parity_workers
Now only one worker would be utilized, scrub_workers, to do the read
and repair.
All writeback would happen at the main scrub thread.
- scrub_block::sparity member
- scrub_parity structure
- function scrub_parity_get()
- function scrub_parity_put()
- function scrub_free_parity()
- function __scrub_mark_bitmap()
- function scrub_parity_mark_sectors_error()
- function scrub_parity_mark_sectors_data()
These helpers are no longer needed, scrub_stripe has its bitmaps and
we can use bitmap helpers to get the error/data status.
- scrub_parity_bio_endio()
- scrub_parity_check_and_repair()
- function scrub_sectors_for_parity()
- function scrub_extent_for_parity()
- function scrub_raid56_data_stripe_for_parity()
- function scrub_raid56_parity()
The new code would reuse the scrub read-repair and writeback path.
Just skip the dev-replace phase.
And scrub_stripe infrastructure allows us to submit and wait for those
data stripes before scrubbing P/Q, without extra infrastructure.
The following two functions are temporarily exported for later cleanup:
- scrub_find_csum()
- scrub_add_sector_to_rd_bio()
Signed-off-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Implement the only missing part for scrub: RAID56 P/Q stripe scrub.
The workflow is pretty straightforward for the new function,
scrub_raid56_parity_stripe():
- Go through the regular scrub path for each data stripe
- Wait for the verification and repair to finish
- Writeback the repaired sectors to data stripes
- Make sure all stripes are properly repaired
If we have sectors unrepaired, we cannot continue, or we could further
corrupt the P/Q stripe.
- Submit the rbio for P/Q stripe
The dev-replace would be handled inside
raid56_parity_submit_scrub_rbio() path.
- Wait for the above bio to finish
Although the old code is no longer used, we still keep the declaration,
as the cleanup can be several times larger than this patch itself.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Switch scrub_simple_mirror() to the new scrub_stripe infrastructure.
Since scrub_simple_mirror() is the core part of scrub (only RAID56
P/Q stripes don't utilize it), we can get rid of a big chunk of code,
mostly scrub_extent(), scrub_sectors() and directly called functions.
There is a functionality change:
- Scrub speed throttle now only affects read on the scrubbing device
Writes (for repair and replace), and reads from other mirrors won't
be limited by the set limits.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
The new helper, queue_scrub_stripe(), would try to queue a stripe for
scrub. If all stripes are already in use, we will submit all the
existing ones and wait for them to finish.
Currently we would queue up to 8 stripes, to enlarge the blocksize to
512KiB to improve the performance. Sectors repaired on zoned need to be
relocated instead of in-place fix.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
The new helper, scrub_stripe_report_errors(), will report the result of
the scrub to system log.
The main reporting is done by introducing a new helper,
scrub_print_common_warning(), which is mostly the same content from
scrub_print_wanring(), but without the need for a scrub_block.
Since we're reporting the errors, it's the perfect time to update the
scrub stats too.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Add a new helper, scrub_write_sectors(), to submit write bios for
specified sectors to the target disk.
There are several differences compared to read path:
- Utilize btrfs_submit_scrub_write()
Now we still rely on the @mirror_num based writeback, but the
requirement is also a little different than regular writeback or read,
thus we have to call btrfs_submit_scrub_write().
- We cannot write the full stripe back
We can only write the sectors we have. There will be two call sites
later, one for repaired sectors, one for all utilized sectors of
dev-replace.
Thus the callers should specify their own write_bitmap.
This function only submit the bios, will not wait for them unless for
zoned case.
Caller must explicitly wait for the IO to finish.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
The new helper, scrub_stripe_read_repair_worker(), would handle the
read-repair part:
- Wait for the previous submitted read IO to finish
- Verify the contents of the stripe
- Go through the remaining mirrors, using as large blocksize as possible
At this stage, we just read out all the failed sectors from each
mirror and re-verify.
If no more failed sector, we can exit.
- Go through all mirrors again, sector-by-sector
This time, we read sector by sector, this is to address cases where
one bad sector mismatches the drive's internal checksum, and cause the
whole read range to fail.
We put this recovery method as the last resort, as sector-by-sector
reading is slow, and reading from other mirrors may have already fixed
the errors.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
The new helper, scrub_verify_stripe(), shares the same main workflow of
the old scrub code.
The major differences are:
- How pages/page_offset is grabbed
Everything can be grabbed from scrub_stripe easily.
- When error report happens
Currently the helper only verifies the sectors, not really doing any
error reporting.
The error reporting would be done after we have done the repair.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
The new helper, scrub_verify_one_metadata(), is almost the same as
scrub_checksum_tree_block().
The difference is in how we grab the pages from other structures.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
The new helper will search the extent tree to find the first extent of a
logical range, then fill the sectors array by two loops:
- Loop 1 to fill common bits and metadata generation
- Loop 2 to fill csum data (only for data bgs)
This loop will use the new btrfs_lookup_csums_bitmap() to fill
the full csum buffer, and set scrub_sector_verification::csum.
With all the needed info filled by this function, later we only need to
submit and verify the stripe.
Here we temporarily export the helper to avoid warning on unused static
function.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
This patch introduces the following structures:
- scrub_sector_verification
Contains all the needed info to verify one sector (data or metadata).
- scrub_stripe
Contains all needed members (mostly bitmap based) to scrub one stripe
(with a length of BTRFS_STRIPE_LEN).
The basic idea is, we keep the existing per-device scrub behavior, but
merge all the scrub_bio/scrub_bio into one generic structure, and read
the full BTRFS_STRIPE_LEN stripe on the first try.
This means we will read some sectors which are not scrub target, but
that's fine. At dev-replace time we only writeback the utilized and good
sectors, and for read-repair we only writeback the repaired sectors.
With every read submitted in BTRFS_STRIPE_LEN, the need for complex bio
form shaping would be gone.
Although to get the same performance of the old scrub behavior, we would
need to submit the initial read for two stripes at once.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Both scrub and read-repair are utilizing a special repair writes that:
- Only writes back to a single device
Even for read-repair on RAID56, we only update the corrupted data
stripe itself, not triggering the full RMW path.
- Requires a valid @mirror_num
For RAID56 case, only @mirror_num == 1 is valid.
For non-RAID56 cases, we need @mirror_num to locate our stripe.
- No data csum generation needed
These two call sites still have some differences though:
- Read-repair goes plain bio
It doesn't need a full btrfs_bio, and goes submit_bio_wait().
- New scrub repair would go btrfs_bio
To simplify both read and write path.
So here this patch would:
- Introduce a common helper, btrfs_map_repair_block()
Due to the single device nature, we can use an on-stack
btrfs_io_stripe to pass device and its physical bytenr.
- Introduce a new interface, btrfs_submit_repair_bio(), for later scrub
code
This is for the incoming scrub code.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Currently we're doing a lot of work for btrfs_bio:
- Checksum verification for data read bios
- Bio splits if it crosses stripe boundary
- Read repair for data read bios
However for the incoming scrub patches, we don't want this extra
functionality at all, just plain logical + mirror -> physical mapping
ability.
Thus here we do the following changes:
- Introduce btrfs_bio::fs_info
This is for the new scrub specific btrfs_bio, which would not populate
btrfs_bio::inode.
Thus we need such new member to grab a fs_info
This new member will always be populated.
- Replace @inode argument with @fs_info for btrfs_bio_init() and its
caller
Since @inode is no longer a mandatory member, replace it with
@fs_info, and let involved users populate @inode.
- Skip checksum verification and generation if @bbio->inode is NULL
- Add extra ASSERT()s
To make sure:
* bbio->inode is properly set for involved read repair path
* if @file_offset is set, bbio->inode is also populated
- Grab @fs_info from @bbio directly
We can no longer go @bbio->inode->root->fs_info, as bbio->inode can be
NULL. This involves:
* btrfs_simple_end_io()
* should_async_write()
* btrfs_wq_submit_bio()
* btrfs_use_zone_append()
Signed-off-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
There is really no need to go through the super complex scrub_sectors()
to just handle super blocks. Introduce a dedicated function to handle
super block scrubbing.
This new function will introduce a behavior change, instead of using the
complex but concurrent scrub_bio system, here we just go submit-and-wait.
There is really not much sense to care the performance of super block
scrubbing. It only has 3 super blocks at most, and they are all
scattered around the devices already.
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Anand Jain <anand.jain@oracle.com>
Signed-off-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Commit 321f69f86a ("btrfs: reset device back to allocation state when
removing") included adding extent_io_tree_release(&device->alloc_state)
to btrfs_close_one_device(), which had already been called in
btrfs_free_device().
The alloc_state tree (IO_TREE_DEVICE_ALLOC_STATE), is created in
btrfs_alloc_device() and released in btrfs_close_one_device(). Therefore,
the additional call to extent_io_tree_release(&device->alloc_state) in
btrfs_free_device() is unnecessary and can be removed.
Signed-off-by: Anand Jain <anand.jain@oracle.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
During my recent search for the root cause of a reported bug, I realized
that it's a good idea to issue a warning for missed cleanup instead of
using debug-only assertions. Since most installations run with debug off,
missed cleanups and premature calls to close could go unnoticed. However,
these issues are serious enough to warrant reporting and fixing.
Signed-off-by: Anand Jain <anand.jain@oracle.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Btrfs can use various different checksumming algorithms, and prints
the one used for a given file system at mount time. Don't bother
printing the crc32c implementation at module load time, the information
is available in /sys/fs/btrfs/FSID/checksum.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
The tree-log code has three almost identical copies for the accounting on
an extent_buffer that doesn't need to be written any more. The only
difference is that walk_down_log_tree passed the bytenr used to find the
buffer instead of extent_buffer.start and calculates the length using the
nodesize, while the other two callers look at the extent_buffer.len
field that must always be equivalent to the nodesize.
Factor the code into a common helper.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Guard all the code to punt bios to a per-cgroup submission helper by a
new CONFIG_BLK_CGROUP_PUNT_BIO symbol that is selected by btrfs.
This way non-btrfs kernel builds don't need to have this code.
Reviewed-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
REQ_CGROUP_PUNT is a bit annoying as it is hard to follow and adds
a branch to the bio submission hot path. To fix this, export
blkcg_punt_bio_submit and let btrfs call it directly. Add a new
REQ_FS_PRIVATE flag for btrfs to indicate to it's own low-level
bio submission code that a punt to the cgroup submission helper
is required.
Reviewed-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
punt_to_cgroup is only used by extent_write_locked_range, but that
function also directly controls the bio flags for the actual submission.
Remove th punt_to_cgroup field, and just set REQ_CGROUP_PUNT directly
in extent_write_locked_range.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: David Sterba <dsterba@suse.com>
submit_one_async_extent needs to use submit_one_async_extent no matter
if the range it handles ends up beeing compressed or not as the deadlock
risk due to cgroup thottling is the same. Call kthread_associate_blkcg
earlier to cover submit_uncompressed_range case as well.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Let submit_one_async_extent, which is the only caller of
submit_uncompressed_range handle freeing of the async_extent in one
central place.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
btrfs_submit_compressed_write should not have to care if it is called
from a helper thread or not. Move the kthread_associate_blkcg handling
into submit_one_async_extent, as that is the one caller that needs it.
Also move the assignment of REQ_CGROUP_PUNT into cow_file_range_async,
as that is the routine that sets up the helper thread offload.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
When starting a transaction, we are assuming the number of bytes used for
each delayed ref update matches the number of bytes used for each item
update, that is the return value of:
btrfs_calc_insert_metadata_size(fs_info, num_items)
However that is not correct when we are using the free space tree, as we
need to multiply that value by 2, since delayed ref updates need to modify
the free space tree besides the extent tree.
So fix this by using btrfs_calc_delayed_ref_bytes() to get the correct
number of bytes used for delayed ref updates.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
When starting a transaction we are comparing the result of a call to
btrfs_block_rsv_full() with 0, but the function returns a boolean. While
in practice it is not incorrect, as 0 is equivalent to false, it makes it
a bit odd and less readable. So update the check to not compare against 0
and instead use the logical not (!) operator.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
If an application is doing direct io to a btrfs file and experiences a
page fault reading from the write buffer, iomap will issue a partial
bio, and allow the fs to keep going. However, there was a subtle bug in
this code path in the btrfs dio iomap implementation that led to the
partial write ending up as a gap in the file's extents and to be read
back as zeros.
The sequence of events in a partial write, lightly summarized and
trimmed down for brevity is as follows:
==== WRITING TASK ====
btrfs_direct_write
__iomap_dio_write
iomap_iter
btrfs_dio_iomap_begin # create full ordered extent
iomap_dio_bio_iter
bio_iov_iter_get_pages # page fault; partial read
submit_bio # partial bio
iomap_iter
btrfs_dio_iomap_end
btrfs_mark_ordered_io_finished # sets BTRFS_ORDERED_IOERR;
# submit to finish_ordered_fn wq
fault_in_iov_iter_readable # btrfs_direct_write detects partial write
__iomap_dio_write
iomap_iter
btrfs_dio_iomap_begin # create second partial ordered extent
iomap_dio_bio_iter
bio_iov_iter_get_pages # read all of remainder
submit_bio # partial bio with all of remainder
iomap_iter
btrfs_dio_iomap_end # nothing exciting to do with ordered io
==== DIO ENDIO ====
== FIRST PARTIAL BIO ==
btrfs_dio_end_io
btrfs_mark_ordered_io_finished # bytes_left > 0
# don't submit to finish_ordered_fn wq
== SECOND PARTIAL BIO ==
btrfs_dio_end_io
btrfs_mark_ordered_io_finished # bytes_left == 0
# submit to finish_ordered_fn wq
==== BTRFS FINISH ORDERED WQ ====
== FIRST PARTIAL BIO ==
btrfs_finish_ordered_io # called by dio_iomap_end_io, sees
# BTRFS_ORDERED_IOERR, just drops the
# ordered_extent
==SECOND PARTIAL BIO==
btrfs_finish_ordered_io # called by btrfs_dio_end_io, writes out file
# extents, csums, etc...
The essence of the problem is that while btrfs_direct_write and iomap
properly interact to submit all the correct bios, there is insufficient
logic in the btrfs dio functions (btrfs_dio_iomap_begin,
btrfs_dio_submit_io, btrfs_dio_end_io, and btrfs_dio_iomap_end) to
ensure that every bio is at least a part of a completed ordered_extent.
And it is completing an ordered_extent that results in crucial
functionality like writing out a file extent for the range.
More specifically, btrfs_dio_end_io treats the ordered extent as
unfinished but btrfs_dio_iomap_end sets BTRFS_ORDERED_IOERR on it.
Thus, the finish io work doesn't result in file extents, csums, etc.
In the aftermath, such a file behaves as though it has a hole in it,
instead of the purportedly written data.
We considered a few options for fixing the bug:
1. treat the partial bio as if we had truncated the file, which would
result in properly finishing it.
2. split the ordered extent when submitting a partial bio.
3. cache the ordered extent across calls to __iomap_dio_rw in
iter->private, so that we could reuse it and correctly apply
several bios to it.
I had trouble with 1, and it felt the most like a hack, so I tried 2
and 3. Since 3 has the benefit of also not creating an extra file
extent, and avoids an ordered extent lookup during bio submission, it
felt like the best option. However, that turned out to re-introduce a
deadlock which this code discarding the ordered_extent between faults
was meant to fix in the first place. (Link to an explanation of the
deadlock below.)
Therefore, go with fix 2, which requires a bit more setup work but fixes
the corruption without introducing the deadlock, which is fundamentally
caused by the ordered extent existing when we attempt to fault in a
range that overlaps with it.
Put succinctly, what this patch does is: when we submit a dio bio, check
if it is partial against the ordered extent stored in dio_data, and if it
is, extract the ordered_extent that matches the bio exactly out of the
larger ordered_extent. Keep the remaining ordered_extent around in dio_data
for cancellation in iomap_end.
Thanks to Josef, Christoph, and Filipe with their help figuring out the
bug and the fix.
Fixes: 51bd9563b6 ("btrfs: fix deadlock due to page faults during direct IO reads and writes")
Link: https://bugzilla.redhat.com/show_bug.cgi?id=2169947
Link: https://lore.kernel.org/linux-btrfs/aa1fb69e-b613-47aa-a99e-a0a2c9ed273f@app.fastmail.com/
Link: https://pastebin.com/3SDaH8C6
Link: https://lore.kernel.org/linux-btrfs/20230315195231.GW10580@twin.jikos.cz/T/#t
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Tested-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Boris Burkov <boris@bur.io>
[ hch: refactored the ordered_extent extraction ]
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: David Sterba <dsterba@suse.com>
NOCOW writes just overwrite an existing extent map, which thus should
not be split in btrfs_extract_ordered_extent. The NOCOW case can't
currently happen as btrfs_extract_ordered_extent is only used on zoned
devices that do not support NOCOW writes, but this will change soon.
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Tested-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Boris Burkov <boris@bur.io>
[ hch: split from a larger patch, wrote a commit log ]
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: David Sterba <dsterba@suse.com>
To prepare for a new caller that already has the ordered_extent
available, change btrfs_extract_ordered_extent to take an argument
for it. Add a wrapper for the bio case that still has to do the
lookup (for now).
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Tested-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Christoph Hellwig
David Sterba
Boris Burkov
Filipe Manana
Qu Wenruo
Tejun Heo
Anand Jain
Jiapeng Chong
Johannes Thumshirn
Josef Bacik
Tom Rix
Naohiro Aota
Sweet Tea Dorminy
Linus Torvalds
Chris Mason
pengfuyuan
Shida Zhang
David Howells
Naohiro Aota
Anastasia Belova
xiaoshoukui
Josh Poimboeuf
Genjian Zhang