The typos accumulate over time so once in a while time they get fixed in
a large patch.
Signed-off-by: Andrea Gelmini <andrea.gelmini@gelma.net>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Traditionally we've had voodoo in btrfs to account for the space that
delayed refs may take up by having a global_block_rsv. This works most
of the time, except when it doesn't. We've had issues reported and seen
in production where sometimes the global reserve is exhausted during
transaction commit before we can run all of our delayed refs, resulting
in an aborted transaction. Because of this voodoo we have equally
dubious flushing semantics around throttling delayed refs which we often
get wrong.
So instead give them their own block_rsv. This way we can always know
exactly how much outstanding space we need for delayed refs. This
allows us to make sure we are constantly filling that reservation up
with space, and allows us to put more precise pressure on the enospc
system. Instead of doing math to see if its a good time to throttle,
the normal enospc code will be invoked if we have a lot of delayed refs
pending, and they will be run via the normal flushing mechanism.
For now the delayed_refs_rsv will hold the reservations for the delayed
refs, the block group updates, and deleting csums. We could have a
separate rsv for the block group updates, but the csum deletion stuff is
still handled via the delayed_refs so that will stay there.
Historical background:
The global reserve has grown to cover everything we don't reserve space
explicitly for, and we've grown a lot of weird ad-hoc heuristics to know
if we're running short on space and when it's time to force a commit. A
failure rate of 20-40 file systems when we run hundreds of thousands of
them isn't super high, but cleaning up this code will make things less
ugly and more predictible.
Thus the delayed refs rsv. We always know how many delayed refs we have
outstanding, and although running them generates more we can use the
global reserve for that spill over, which fits better into it's desired
use than a full blown reservation. This first approach is to simply
take how many times we're reserving space for and multiply that by 2 in
order to save enough space for the delayed refs that could be generated.
This is a niave approach and will probably evolve, but for now it works.
Signed-off-by: Josef Bacik <jbacik@fb.com>
Reviewed-by: David Sterba <dsterba@suse.com> # high-level review
[ added background notes from the cover letter ]
Signed-off-by: David Sterba <dsterba@suse.com>
After the rw semaphore has been added, the custom blocking using
::blocking_readers and ::read_lock_wq is redundant.
The blocking logic in __btrfs_map_block is replaced by extending the
time the semaphore is held, that has the same blocking effect on writes
as the previous custom scheme that waited until ::blocking_readers was
zero.
Signed-off-by: David Sterba <dsterba@suse.com>
This is the first part of removing the custom locking and waiting scheme
used for device replace. It was probably copied from extent buffer
locking, but there's nothing that would require more than is provided by
the common locking primitives.
The rw spinlock protects waiting tasks counter in case of incompatible
locks and the waitqueue. Same as rw semaphore.
This patch only switches the locking primitive, for better
bisectability. There should be no functional change other than the
overhead of the locking and potential sleeping instead of spinning when
the lock is contended.
Signed-off-by: David Sterba <dsterba@suse.com>
Document why map_private_extent_buffer() cannot return '1' (i.e. the map
spans two pages) for the csum_tree_block() case.
The current algorithm for detecting a page boundary crossing in
map_private_extent_buffer() will return a '1' *IFF* the extent buffer's
offset in the page + the offset passed in by csum_tree_block() and the
minimal length passed in by csum_tree_block() - 1 are bigger than
PAGE_SIZE.
We always pass BTRFS_CSUM_SIZE (32) as offset and a minimal length of 32
and the current extent buffer allocator always guarantees page aligned
extends, so the above condition can't be true.
Reviewed-by: Nikolay Borisov <nborisov@suse.com>
Signed-off-by: Johannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: David Sterba <dsterba@suse.com>
For data inodes this hook does nothing but to return -EAGAIN which is
used to signal to the endio routines that this bio belongs to a data
inode. If this is the case the actual retrying is handled by
bio_readpage_error. Alternatively, if this bio belongs to the btree
inode then btree_io_failed_hook just does some cleanup and doesn't retry
anything.
This patch simplifies the code flow by eliminating
readpage_io_failed_hook and instead open-coding btree_io_failed_hook in
end_bio_extent_readpage. Also eliminate some needless checks since IO is
always performed on either data inode or btree inode, both of which are
guaranteed to have their extent_io_tree::ops set.
Signed-off-by: Nikolay Borisov <nborisov@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
There's one caller and its code is simple, we can open code it in
run_one_async_done. The errors are passed through bio.
Reviewed-by: Nikolay Borisov <nborisov@suse.com>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: David Sterba <dsterba@suse.com>
We can have a lot freed extents during the life span of transaction, so
the red black tree that keeps track of the ranges of each freed extent
(fs_info->freed_extents[]) can get quite big. When finishing a
transaction commit we find each range, process it (discard the extents,
unpin them) and then remove it from the red black tree.
We can use an extent state record as a cache when searching for a range,
so that when we clean the range we can use the cached extent state we
passed to the search function instead of iterating the red black tree
again. Doing things as fast as possible when finishing a transaction (in
state TRANS_STATE_UNBLOCKED) is convenient as it reduces the time we
block another task that wants to commit the next transaction.
So change clear_extent_dirty() to allow an optional extent state record to
be passed as an argument, which will be passed down to __clear_extent_bit.
Reviewed-by: Nikolay Borisov <nborisov@suse.com>
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Even though fsid change without rewrite is a very quick operation it's
still possible to experience a split-brain scenario if power loss occurs
at the most inconvenient time. This patch handles the case where power
failure occurs while the first transaction (the one setting
CHANGING_FSID_V2) flag is being persisted on disk. This can cause the
btrfs_fs_devices of this filesystem to be created by a device which:
a) has the CHANGING_FSID_V2 flag set but its fsid value is intact
b) or a device which doesn't have CHANGING_FSID_V2 flag set and its
fsid value is intact
This situation is trivially handled by the current find_fsid code since
in both cases the devices are going to be treated like ordinary devices.
Since btrfs is always mounted using the superblock of the latest
device (the one with highest generation number), meaning it will have
the CHANGING_FSID_V2 flag set, ensure it's being cleared on mount. On
the first transaction commit following mount all disks will have it
cleared.
Signed-off-by: Nikolay Borisov <nborisov@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Currently btrfs_fs_info structure contains a copy of the
fsid/metadata_uuid fields. Same values are also contained in the
btrfs_fs_devices structure which fs_info has a reference to. Let's
reduce duplication by removing the fields from fs_info and always refer
to the ones in fs_devices. No functional changes.
Signed-off-by: Nikolay Borisov <nborisov@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
This field is going to be used when the user wants to change the UUID
of the filesystem without having to rewrite all metadata blocks. This
field adds another level of indirection such that when the FSID is
changed what really happens is the current UUID (the one with which the
fs was created) is copied to the 'metadata_uuid' field in the superblock
as well as a new incompat flag is set METADATA_UUID. When the kernel
detects this flag is set it knows that the superblock in fact has 2
UUIDs:
1. Is the UUID which is user-visible, currently known as FSID.
2. Metadata UUID - this is the UUID which is stamped into all on-disk
datastructures belonging to this file system.
When the new incompat flag is present device scanning checks whether
both fsid/metadata_uuid of the scanned device match any of the
registered filesystems. When the flag is not set then both UUIDs are
equal and only the FSID is retained on disk, metadata_uuid is set only
in-memory during mount.
Additionally a new metadata_uuid field is also added to the fs_info
struct. It's initialised either with the FSID in case METADATA_UUID
incompat flag is not set or with the metdata_uuid of the superblock
otherwise.
This commit introduces the new fields as well as the new incompat flag
and switches all users of the fsid to the new logic.
Signed-off-by: Nikolay Borisov <nborisov@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
[ minor updates in comments ]
Signed-off-by: David Sterba <dsterba@suse.com>
A later patch will implement swap file support for Btrfs, but before we
do that, we need to make sure that the various Btrfs ioctls cannot
change a swap file.
When a swap file is active, we must make sure that the extents of the
file are not moved and that they don't become shared. That means that
the following are not safe:
- chattr +c (enable compression)
- reflink
- dedupe
- snapshot
- defrag
Don't allow those to happen on an active swap file.
Additionally, balance, resize, device remove, and device replace are
also unsafe if they affect an active swapfile. Add a red-black tree of
block groups and devices which contain an active swapfile. Relocation
checks each block group against this tree and skips it or errors out for
balance or resize, respectively. Device remove and device replace check
the tree for the device they will operate on.
Note that we don't have to worry about chattr -C (disable nocow), which
we ignore for non-empty files, because an active swapfile must be
non-empty and can't be truncated. We also don't have to worry about
autodefrag because it's only done on COW files. Truncate and fallocate
are already taken care of by the generic code. Device add doesn't do
relocation so it's not an issue, either.
Signed-off-by: Omar Sandoval <osandov@fb.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
When a metadata read is served the endio routine btree_readpage_end_io_hook
is called which eventually runs the tree-checker. If tree-checker fails
to validate the read eb then it sets EXTENT_BUFFER_CORRUPT flag. This
leads to btree_read_extent_buffer_pages wrongly assuming that all
available copies of this extent buffer are wrong and failing prematurely.
Fix this modify btree_read_extent_buffer_pages to read all copies of
the data.
This failure was exhibitted in xfstests btrfs/124 which would
spuriously fail its balance operations. The reason was that when balance
was run following re-introduction of the missing raid1 disk
__btrfs_map_block would map the read request to stripe 0, which
corresponded to devid 2 (the disk which is being removed in the test):
item 2 key (FIRST_CHUNK_TREE CHUNK_ITEM 3553624064) itemoff 15975 itemsize 112
length 1073741824 owner 2 stripe_len 65536 type DATA|RAID1
io_align 65536 io_width 65536 sector_size 4096
num_stripes 2 sub_stripes 1
stripe 0 devid 2 offset 2156920832
dev_uuid 8466c350-ed0c-4c3b-b17d-6379b445d5c8
stripe 1 devid 1 offset 3553624064
dev_uuid 1265d8db-5596-477e-af03-df08eb38d2ca
This caused read requests for a checksum item that to be routed to the
stale disk which triggered the aforementioned logic involving
EXTENT_BUFFER_CORRUPT flag. This then triggered cascading failures of
the balance operation.
Fixes: a826d6dcb3 ("Btrfs: check items for correctness as we search")
CC: stable@vger.kernel.org # 4.4+
Suggested-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: Nikolay Borisov <nborisov@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
There's a race between close_ctree() and cleaner_kthread().
close_ctree() sets btrfs_fs_closing(), and the cleaner stops when it
sees it set, but this is racy; the cleaner might have already checked
the bit and could be cleaning stuff. In particular, if it deletes unused
block groups, it will create delayed iputs for the free space cache
inodes. As of "btrfs: don't run delayed_iputs in commit", we're no
longer running delayed iputs after a commit. Therefore, if the cleaner
creates more delayed iputs after delayed iputs are run in
btrfs_commit_super(), we will leak inodes on unmount and get a busy
inode crash from the VFS.
Fix it by parking the cleaner before we actually close anything. Then,
any remaining delayed iputs will always be handled in
btrfs_commit_super(). This also ensures that the commit in close_ctree()
is really the last commit, so we can get rid of the commit in
cleaner_kthread().
The fstest/generic/475 followed by 476 can trigger a crash that
manifests as a slab corruption caused by accessing the freed kthread
structure by a wake up function. Sample trace:
[ 5657.077612] BUG: unable to handle kernel NULL pointer dereference at 00000000000000cc
[ 5657.079432] PGD 1c57a067 P4D 1c57a067 PUD da10067 PMD 0
[ 5657.080661] Oops: 0000 [#1] PREEMPT SMP
[ 5657.081592] CPU: 1 PID: 5157 Comm: fsstress Tainted: G W 4.19.0-rc8-default+ #323
[ 5657.083703] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.11.2-0-gf9626cc-prebuilt.qemu-project.org 04/01/2014
[ 5657.086577] RIP: 0010:shrink_page_list+0x2f9/0xe90
[ 5657.091937] RSP: 0018:ffffb5c745c8f728 EFLAGS: 00010287
[ 5657.092953] RAX: 0000000000000074 RBX: ffffb5c745c8f830 RCX: 0000000000000000
[ 5657.094590] RDX: 0000000000000000 RSI: 0000000000000001 RDI: ffff9a8747fdf3d0
[ 5657.095987] RBP: ffffb5c745c8f9e0 R08: 0000000000000000 R09: 0000000000000000
[ 5657.097159] R10: ffff9a8747fdf5e8 R11: 0000000000000000 R12: ffffb5c745c8f788
[ 5657.098513] R13: ffff9a877f6ff2c0 R14: ffff9a877f6ff2c8 R15: dead000000000200
[ 5657.099689] FS: 00007f948d853b80(0000) GS:ffff9a877d600000(0000) knlGS:0000000000000000
[ 5657.101032] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 5657.101953] CR2: 00000000000000cc CR3: 00000000684bd000 CR4: 00000000000006e0
[ 5657.103159] Call Trace:
[ 5657.103776] shrink_inactive_list+0x194/0x410
[ 5657.104671] shrink_node_memcg.constprop.84+0x39a/0x6a0
[ 5657.105750] shrink_node+0x62/0x1c0
[ 5657.106529] try_to_free_pages+0x1a4/0x500
[ 5657.107408] __alloc_pages_slowpath+0x2c9/0xb20
[ 5657.108418] __alloc_pages_nodemask+0x268/0x2b0
[ 5657.109348] kmalloc_large_node+0x37/0x90
[ 5657.110205] __kmalloc_node+0x236/0x310
[ 5657.111014] kvmalloc_node+0x3e/0x70
Fixes: 30928e9baa ("btrfs: don't run delayed_iputs in commit")
Signed-off-by: Omar Sandoval <osandov@fb.com>
Reviewed-by: David Sterba <dsterba@suse.com>
[ add trace ]
Signed-off-by: David Sterba <dsterba@suse.com>
I ran into an issue where there was some reference being held on an
inode that I couldn't track. This assert wasn't triggered, but it at
least rules out we're doing something stupid.
Reviewed-by: Omar Sandoval <osandov@fb.com>
Signed-off-by: Josef Bacik <josef@toxicpanda.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
The replace_wait and bio_counter were mistakenly added to fs_info in
commit c404e0dc2c ("Btrfs: fix use-after-free in the finishing
procedure of the device replace"), but they logically belong to
fs_info::dev_replace. Besides, bio_counter is a very generic name and is
confusing in bare fs_info context.
Reviewed-by: Anand Jain <anand.jain@oracle.com>
Signed-off-by: David Sterba <dsterba@suse.com>
This member seems to be copied from the extent_buffer locking scheme and
is at least used to assert that the read lock/unlock is properly nested.
In some way. While the _inc/_dec are called inside the read lock
section, the asserts are both inside and outside, so the ordering is not
guaranteed and we can see read/inc/dec ordered in any way
(theoretically).
A missing call of btrfs_dev_replace_clear_lock_blocking could cause
unexpected read_locks count, so this at least looks like a valid
assertion, but this will become unnecessary with later updates.
Reviewed-by: Anand Jain <anand.jain@oracle.com>
Signed-off-by: David Sterba <dsterba@suse.com>
rb_first_cached() trades an extra pointer "leftmost" for doing the same
job as rb_first() but in O(1).
Functions manipulating href->ref_tree need to get the first entry, this
converts href->ref_tree to use rb_first_cached().
For more details about the optimization see patch "Btrfs: delayed-refs:
use rb_first_cached for href_root".
Tested-by: Holger Hoffstätte <holger@applied-asynchrony.com>
Signed-off-by: Liu Bo <bo.liu@linux.alibaba.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
rb_first_cached() trades an extra pointer "leftmost" for doing the same
job as rb_first() but in O(1).
Functions manipulating href_root need to get the first entry, this
converts href_root to use rb_first_cached().
This patch is first in the sequenct of similar updates to other rbtrees
and this is analysis of the expected behaviour and improvements.
There's a common pattern:
while (node = rb_first) {
entry = rb_entry(node)
next = rb_next(node)
rb_erase(node)
cleanup(entry)
}
rb_first needs to traverse the tree up to logN depth, rb_erase can
completely reshuffle the tree. With the caching we'll skip the traversal
in rb_first. That's a cached memory access vs looped pointer
dereference trade-off that IMHO has a clear winner.
Measurements show there's not much difference in a sample tree with
10000 nodes: 4.5s / rb_first and 4.8s / rb_first_cached. Real effects of
caching and pointer chasing are unpredictable though.
Further optimzations can be done to avoid the expensive rb_erase step.
In some cases it's ok to process the nodes in any order, so the tree can
be traversed in post-order, not rebalancing the children nodes and just
calling free. Care must be taken regarding the next node.
Tested-by: Holger Hoffstätte <holger@applied-asynchrony.com>
Signed-off-by: Liu Bo <bo.liu@linux.alibaba.com>
Reviewed-by: David Sterba <dsterba@suse.com>
[ update changelog from mail discussions ]
Signed-off-by: David Sterba <dsterba@suse.com>
There are two members in struct btrfs_root which indicate root's
objectid: objectid and root_key.objectid.
They are both set to the same value in __setup_root():
static void __setup_root(struct btrfs_root *root,
struct btrfs_fs_info *fs_info,
u64 objectid)
{
...
root->objectid = objectid;
...
root->root_key.objectid = objecitd;
...
}
and not changed to other value after initialization.
grep in btrfs directory shows both are used in many places:
$ grep -rI "root->root_key.objectid" | wc -l
133
$ grep -rI "root->objectid" | wc -l
55
(4.17, inc. some noise)
It is confusing to have two similar variable names and it seems
that there is no rule about which should be used in a certain case.
Since ->root_key itself is needed for tree reloc tree, let's remove
'objecitd' member and unify code to use ->root_key.objectid in all places.
Signed-off-by: Misono Tomohiro <misono.tomohiro@jp.fujitsu.com>
Reviewed-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Commit e9894fd3e3 ("Btrfs: fix snapshot vs nocow writting") forced
nocow writes to fallback to COW, during writeback, when a snapshot is
created. This resulted in writes made before creating the snapshot to
unexpectedly fail with ENOSPC during writeback when success (0) was
returned to user space through the write system call.
The steps leading to this problem are:
1. When it's not possible to allocate data space for a write, the
buffered write path checks if a NOCOW write is possible. If it is,
it will not reserve space and success (0) is returned to user space.
2. Then when a snapshot is created, the root's will_be_snapshotted
atomic is incremented and writeback is triggered for all inode's that
belong to the root being snapshotted. Incrementing that atomic forces
all previous writes to fallback to COW during writeback (running
delalloc).
3. This results in the writeback for the inodes to fail and therefore
setting the ENOSPC error in their mappings, so that a subsequent
fsync on them will report the error to user space. So it's not a
completely silent data loss (since fsync will report ENOSPC) but it's
a very unexpected and undesirable behaviour, because if a clean
shutdown/unmount of the filesystem happens without previous calls to
fsync, it is expected to have the data present in the files after
mounting the filesystem again.
So fix this by adding a new atomic named snapshot_force_cow to the
root structure which prevents this behaviour and works the following way:
1. It is incremented when we start to create a snapshot after triggering
writeback and before waiting for writeback to finish.
2. This new atomic is now what is used by writeback (running delalloc)
to decide whether we need to fallback to COW or not. Because we
incremented this new atomic after triggering writeback in the
snapshot creation ioctl, we ensure that all buffered writes that
happened before snapshot creation will succeed and not fallback to
COW (which would make them fail with ENOSPC).
3. The existing atomic, will_be_snapshotted, is kept because it is used
to force new buffered writes, that start after we started
snapshotting, to reserve data space even when NOCOW is possible.
This makes these writes fail early with ENOSPC when there's no
available space to allocate, preventing the unexpected behaviour of
writeback later failing with ENOSPC due to a fallback to COW mode.
Fixes: e9894fd3e3 ("Btrfs: fix snapshot vs nocow writting")
Signed-off-by: Robbie Ko <robbieko@synology.com>
Reviewed-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
This patch will introduce chunk <-> dev extent mapping check, to protect
us against invalid dev extents or chunks.
Since chunk mapping is the fundamental infrastructure of btrfs, extra
check at mount time could prevent a lot of unexpected behavior (BUG_ON).
Reported-by: Xu Wen <wen.xu@gatech.edu>
Link: https://bugzilla.kernel.org/show_bug.cgi?id=200403
Link: https://bugzilla.kernel.org/show_bug.cgi?id=200407
Signed-off-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: Su Yue <suy.fnst@cn.fujitsu.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
The exported helper just calls the static one. There's no obvious reason
to have them separate eg. for performance reasons where the static one
could be better optimized in the same unit. There's a slight decrease in
code size and stack consumption.
Signed-off-by: David Sterba <dsterba@suse.com>
Lock owner and nesting level have been unused since day 1, probably
copy&pasted from the extent_buffer locking scheme without much thinking.
The locking of device replace is simpler and does not need any lock
nesting.
Signed-off-by: David Sterba <dsterba@suse.com>
Added in 58176a9604 ("Btrfs: Add per-root block accounting and sysfs
entries") in 2007, the roots had names exported in sysfs. The code
was commented out in 4df27c4d5c ("Btrfs: change how subvolumes
are organized") and cleaned by 182608c829 ("btrfs: remove old
unused commented out code").
Signed-off-by: David Sterba <dsterba@suse.com>
The data and metadata callback implementation both use the same
function. We can remove the call indirection and intermediate helper
completely.
Reviewed-by: Nikolay Borisov <nborisov@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
The data and metadata callback implementation both use the same
function. We can remove the call indirection completely.
Reviewed-by: Nikolay Borisov <nborisov@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
All implementations of the callback are trivial and do the same and
there's only one user. Merge everything together.
Reviewed-by: Nikolay Borisov <nborisov@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
The end_io callbacks passed to btrfs_wq_submit_bio
(btrfs_submit_bio_done and btree_submit_bio_done) are effectively the
same code, there's no point to do the indirection. Export
btrfs_submit_bio_done and call it directly.
Reviewed-by: Nikolay Borisov <nborisov@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
After splitting the start and end hooks in a758781d4b ("btrfs:
separate types for submit_bio_start and submit_bio_done"), some of
the function arguments were dropped but not removed from the structure.
Reviewed-by: Nikolay Borisov <nborisov@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Introduced by c6100a4b4e ("Btrfs: replace tree->mapping with
tree->private_data") to be used in run_one_async_done where it got
unused after 736cd52e0c ("Btrfs: remove nr_async_submits and
async_submit_draining").
Reviewed-by: Nikolay Borisov <nborisov@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
EXTENT_BUFFER_DUMMY is an awful name for this flag. Buffers which have
this flag set are not in any way dummy. Rather, they are private in the
sense that are not mapped and linked to the global buffer tree. This
flag has subtle implications to the way free_extent_buffer works for
example, as well as controls whether page->mapping->private_lock is held
during extent_buffer release. Pages for an unmapped buffer cannot be
under io, nor can they be written by a 3rd party so taking the lock is
unnecessary.
Signed-off-by: Nikolay Borisov <nborisov@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
[ EXTENT_BUFFER_UNMAPPED, update changelog ]
Signed-off-by: David Sterba <dsterba@suse.com>
We use customized, nodesize batch value to update dirty_metadata_bytes.
We should also use batch version of compare function or we will easily
goto fast path and get false result from percpu_counter_compare().
Fixes: e2d845211e ("Btrfs: use percpu counter for dirty metadata count")
CC: stable@vger.kernel.org # 4.4+
Signed-off-by: Ethan Lien <ethanlien@synology.com>
Reviewed-by: Nikolay Borisov <nborisov@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Functions that get btrfs inode can simply reach the fs_info by
dereferencing the root and this looks a bit more straightforward
compared to the btrfs_sb(...) indirection.
If the transaction handle is available and not NULL it's used instead.
Signed-off-by: David Sterba <dsterba@suse.com>
For easier debugging, print eb->start if level is invalid. Also make
clear if bytenr found is not expected.
Signed-off-by: Su Yue <suy.fnst@cn.fujitsu.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
The transaction times were changed to ktime_get_real_seconds to avoid
the y2038 overflow, but they still have a minor problem when they go
backwards or jump due to settimeofday() or leap seconds.
This changes the transaction handling to instead use ktime_get_seconds(),
which returns a CLOCK_MONOTONIC timestamp that has neither of those
problems.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Nikolay Borisov <nborisov@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
When a new extent buffer is allocated there are a few mandatory fields
which need to be set in order for the buffer to be sane: level,
generation, bytenr, backref_rev, owner and FSID/UUID. Currently this
is open coded in the callers of btrfs_alloc_tree_block, meaning it's
fairly high in the abstraction hierarchy of operations. This patch
solves this by simply moving this init code in btrfs_init_new_buffer,
since this is the function which initializes a newly allocated
extent buffer. No functional changes.
Signed-off-by: Nikolay Borisov <nborisov@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
The get_seconds() function is deprecated as it truncates the timestamp
to 32 bits. Change it to or ktime_get_real_seconds().
Signed-off-by: Allen Pais <allen.lkml@gmail.com>
Reviewed-by: David Sterba <dsterba@suse.com>
[ update changelog ]
Signed-off-by: David Sterba <dsterba@suse.com>
As verify_level_key() is checked after verify_parent_transid(), i.e.
if (verify_parent_transid())
ret = -EIO;
else if (verify_level_key())
ret = -EUCLEAN;
if parent_transid is 0, verify_parent_transid() skips verifying
parent_transid and considers eb as valid, and if verify_level_key()
reports something wrong, we're not going to know if it's caused by
corrupted metadata or non-checkecd eb (e.g. stale eb).
The stale eb can be from an outdated raid1 mirror after a degraded
mount, see eg "btrfs: fix reading stale metadata blocks after degraded
raid1 mounts" (02a3307aa9) for more details.
@parent_transid is able to tell whether the eb's generation has been
verified by the caller.
Signed-off-by: Liu Bo <bo.liu@linux.alibaba.com>
Reviewed-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Now that we don't keep long-standing reservations for orphan items,
root->orphan_block_rsv isn't used. We can git rid of it, along with:
- root->orphan_lock, which was used to protect root->orphan_block_rsv
- root->orphan_inodes, which was used as a refcount for root->orphan_block_rsv
- BTRFS_INODE_ORPHAN_META_RESERVED, which was used to track reservations
in root->orphan_block_rsv
- btrfs_orphan_commit_root(), which was the last user of any of these
and does nothing else
Reviewed-by: Nikolay Borisov <nborisov@suse.com>
Signed-off-by: Omar Sandoval <osandov@fb.com>
Signed-off-by: David Sterba <dsterba@suse.com>
There are already 2 reports about strangely corrupted super blocks,
where csum still matches but extra garbage gets slipped into super block.
The corruption would looks like:
------
superblock: bytenr=65536, device=/dev/sdc1
---------------------------------------------------------
csum_type 41700 (INVALID)
csum 0x3b252d3a [match]
bytenr 65536
flags 0x1
( WRITTEN )
magic _BHRfS_M [match]
...
incompat_flags 0x5b22400000000169
( MIXED_BACKREF |
COMPRESS_LZO |
BIG_METADATA |
EXTENDED_IREF |
SKINNY_METADATA |
unknown flag: 0x5b22400000000000 )
...
------
Or
------
superblock: bytenr=65536, device=/dev/mapper/x
---------------------------------------------------------
csum_type 35355 (INVALID)
csum_size 32
csum 0xf0dbeddd [match]
bytenr 65536
flags 0x1
( WRITTEN )
magic _BHRfS_M [match]
...
incompat_flags 0x176d200000000169
( MIXED_BACKREF |
COMPRESS_LZO |
BIG_METADATA |
EXTENDED_IREF |
SKINNY_METADATA |
unknown flag: 0x176d200000000000 )
------
Obviously, csum_type and incompat_flags get some garbage, but its csum
still matches, which means kernel calculates the csum based on corrupted
super block memory.
And after manually fixing these values, the filesystem is completely
healthy without any problem exposed by btrfs check.
Although the cause is still unknown, at least detect it and prevent further
corruption.
Both reports have same symptoms, there's an overwrite on offset 192 of
the superblock, by 4 bytes. The superblock structure is not allocated or
freed and stays in the memory for the whole filesystem lifetime, so it's
not a use-after-free kind of error on someone else's leaked page.
As a vague point for the problable cause is mentioning of other system
freezing related to graphic card drivers.
Reported-by: Ken Swenson <flat@imo.uto.moe>
Reported-by: Ben Parsons <9parsonsb@gmail.com>
Signed-off-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
[ add brief analysis of the reports ]
Signed-off-by: David Sterba <dsterba@suse.com>
Refactor btrfs_check_super_valid:
1) Rename it to btrfs_validate_mount_super()
Now it's more obvious when the function should be called.
2) Extract core check routine into validate_super()
Later write time check can reuse it, and if needed, we could also
use validate_super() to check each super block.
3) Add more comments about btrfs_validate_mount_super()
Mostly about what it doesn't check and when it should be called.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
[ rename to validate_super ]
Signed-off-by: David Sterba <dsterba@suse.com>
Move btrfs_check_super_valid() before its single caller to avoid forward
declaration.
Though such code motion is not recommended as it pollutes git history,
in this case the following patches would need to add new forward
declarations for static functions that we want to avoid.
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 member struct btrfs_raid_attr::bg_flag so that
btrfs_raid_array can maintain the bit map flag of the raid type, and
so we can drop btrfs_raid_group.
Signed-off-by: Anand Jain <anand.jain@oracle.com>
Reviewed-by: Nikolay Borisov <nborisov@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Currently fs_info::balance_running is 0 or 1 and does not use the
semantics of atomics. The pause and cancel check for 0, that can happen
only after __btrfs_balance exits for whatever reason.
Parallel calls to balance ioctl may enter btrfs_ioctl_balance multiple
times but will block on the balance_mutex that protects the
fs_info::flags bit.
Reviewed-by: Anand Jain <anand.jain@oracle.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Mutual exclusion of device add/rm and balance was done by the volume
mutex up to version 3.7. The commit 5ac00addc7 ("Btrfs: disallow
mutually exclusive admin operations from user mode") added a bit that
essentially tracked the same information.
The status bit has an advantage over a mutex that it can be set without
restrictions of function context, so it started to be used in the
mount-time resuming of balance or device replace.
But we don't really need to track the same information in two ways.
1) After the previous cleanups, the main ioctl handlers for
add/del/resize copy the EXCL_OP bit next to the volume mutex, here
it's clearly safe.
2) Resuming balance during mount or after rw remount will set only the
EXCL_OP bit and the volume_mutex is held in the kernel thread that
calls btrfs_balance.
3) Resuming device replace during mount or after rw remount is done
after balance and is excluded by the EXCL_OP bit. It does not take
the volume_mutex at all and completely relies on the EXCL_OP bit.
4) The resuming of balance and dev-replace cannot hapen at the same time
as the ioctls cannot be started in parallel. Nevertheless, a crafted
image could trigger that and a warning is printed.
5) Balance is normally excluded by EXCL_OP and also uses own mutex to
protect against concurrent access to its status data. There's some
trickery to maintain the right lock nesting in case we need to
reexamine the status in btrfs_ioctl_balance. The volume_mutex is
removed and the unlock/lock sequence is left in place as we might
expect other waiters to proceed.
6) Similar to 5, the unlock/lock sequence is kept in
btrfs_cancel_balance to allow waiters to continue.
Reviewed-by: Anand Jain <anand.jain@oracle.com>
Signed-off-by: David Sterba <dsterba@suse.com>
When a transaction is aborted btrfs_cleanup_transaction is called to
cleanup all the various in-flight bits and pieces which migth be
active. One of those is delalloc inodes - inodes which have dirty
pages which haven't been persisted yet. Currently the process of
freeing such delalloc inodes in exceptional circumstances such as
transaction abort boiled down to calling btrfs_invalidate_inodes whose
sole job is to invalidate the dentries for all inodes related to a
root. This is in fact wrong and insufficient since such delalloc inodes
will likely have pending pages or ordered-extents and will be linked to
the sb->s_inode_list. This means that unmounting a btrfs instance with
an aborted transaction could potentially lead inodes/their pages
visible to the system long after their superblock has been freed. This
in turn leads to a "use-after-free" situation once page shrink is
triggered. This situation could be simulated by running generic/019
which would cause such inodes to be left hanging, followed by
generic/176 which causes memory pressure and page eviction which lead
to touching the freed super block instance. This situation is
additionally detected by the unmount code of VFS with the following
message:
"VFS: Busy inodes after unmount of Self-destruct in 5 seconds. Have a nice day..."
Additionally btrfs hits WARN_ON(!RB_EMPTY_ROOT(&root->inode_tree));
in free_fs_root for the same reason.
This patch aims to rectify the sitaution by doing the following:
1. Change btrfs_destroy_delalloc_inodes so that it calls
invalidate_inode_pages2 for every inode on the delalloc list, this
ensures that all the pages of the inode are released. This function
boils down to calling btrfs_releasepage. During test I observed cases
where inodes on the delalloc list were having an i_count of 0, so this
necessitates using igrab to be sure we are working on a non-freed inode.
2. Since calling btrfs_releasepage might queue delayed iputs move the
call out to btrfs_cleanup_transaction in btrfs_error_commit_super before
calling run_delayed_iputs for the last time. This is necessary to ensure
that delayed iputs are run.
Note: this patch is tagged for 4.14 stable but the fix applies to older
versions too but needs to be backported manually due to conflicts.
CC: stable@vger.kernel.org # 4.14.x: 2b87733134: btrfs: Split btrfs_del_delalloc_inode into 2 functions
CC: stable@vger.kernel.org # 4.14.x
Signed-off-by: Nikolay Borisov <nborisov@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
[ add comment to igrab ]
Signed-off-by: David Sterba <dsterba@suse.com>
Unlike previous method that tries to commit transaction inside
qgroup_reserve(), this time we will try to commit transaction using
fs_info->transaction_kthread to avoid nested transaction and no need to
worry about locking context.
Since it's an asynchronous function call and we won't wait for
transaction commit, unlike previous method, we must call it before we
hit the qgroup limit.
So this patch will use the ratio and size of qgroup meta_pertrans
reservation as indicator to check if we should trigger a transaction
commit. (meta_prealloc won't be cleaned in transaction committ, it's
useless anyway)
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>