84056 Commits

Author SHA1 Message Date
Kent Overstreet
e812cf38c5 bcachefs: Check for key size > offset
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22 17:08:23 -04:00
Kent Overstreet
91052b9de8 bcachefs: Refactor trans_(get|update)_key
these are still pretty ugly...

Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22 17:08:23 -04:00
Kent Overstreet
88767d65d8 bcachefs: Update path now handles triggers that generate more triggers
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22 17:08:23 -04:00
Kent Overstreet
44da9767bb bcachefs: add inode_generation_to_text method
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22 17:08:23 -04:00
Kent Overstreet
2ded276b7e bcachefs: Fix array overrun with unknown btree roots
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22 17:08:23 -04:00
Kent Overstreet
f707e3d8f4 bcachefs: fix kasan splat
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22 17:08:23 -04:00
Kent Overstreet
09bf409b46 bcachefs: Delete a spurious assertion
bch_write_op->written used to be a u16, but it's not so the assertion
isn't needed anymore - and 5.1 can send larger bios.

Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22 17:08:23 -04:00
Kent Overstreet
2a488aaac1 bcachefs: fix __bch2_xattr_bcachefs_get()
We were returning -ERANGE when the size of the buffer passed in was
exactly the size of the xattr val

Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22 17:08:23 -04:00
Kent Overstreet
e0dfc08bc2 bcachefs: use memalloc_nofs_save() for vmalloc allocation
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22 17:08:23 -04:00
Kent Overstreet
3e66981690 bcachefs: Fix promoting to cache devices (durability = 0)
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22 17:08:23 -04:00
Kent Overstreet
66b095b008 bcachefs: Don't allow bkey vals that are too big in extents btree
Make sure we don't overflow BKEY_PADDED keys

Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22 17:08:23 -04:00
Kent Overstreet
9146b8ee03 bcachefs: Reduce BKEY_PADDED usage
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22 17:08:23 -04:00
Kent Overstreet
436c656d46 bcachefs: bkey_merge() now takes bkey_s
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22 17:08:23 -04:00
Kent Overstreet
1ae9733456 bcachefs: Don't overflow stack in bch2_extent_merge_inline()
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22 17:08:23 -04:00
Kent Overstreet
5884fddfe7 bcachefs: Fix starting copygc when already started
We can sometimes call bch2_dev_read_write() when the device is already
RW (in error paths).

Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22 17:08:22 -04:00
Kent Overstreet
37dd783474 bcachefs: Fix an error path in bch2_btree_iter_traverse()
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22 17:08:22 -04:00
Kent Overstreet
6e738539cd bcachefs: Improve key marking interface
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22 17:08:22 -04:00
Kent Overstreet
572ad769f5 bcachefs: Fix cached sectors not being updated on invalidate
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22 17:08:22 -04:00
Kent Overstreet
4ee202e2b7 bcachefs: better BTREE_INSERT_NO_CLEAR_REPLICAS
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22 17:08:22 -04:00
Kent Overstreet
cdeeb75ea9 bcachefs: fix a mount error path
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22 17:08:22 -04:00
Kent Overstreet
a62c78a794 bcachefs: fix bch2_extent_merge()
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22 17:08:22 -04:00
Kent Overstreet
3838be7841 bcachefs: Don't use a fixed size buffer for fs_usage_deltas
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22 17:08:22 -04:00
Kent Overstreet
61011ea237 bcachefs: Rip out old hacky transaction restart tracing
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22 17:08:22 -04:00
Kent Overstreet
20bceecb31 bcachefs: More work to avoid transaction restarts
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22 17:08:22 -04:00
Kent Overstreet
7d82586660 bcachefs: Avoid spurious transaction restarts
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22 17:08:22 -04:00
Kent Overstreet
0e6dd8fba0 bcachefs: Ensure bch2_btree_iter_next() always advances
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22 17:08:22 -04:00
Kent Overstreet
87c3beb4a5 bcachefs: Fix a bug with spinning on the journal
Transactional triggers meant that when we failed to get a journal
reservation, then bailed out into the error path to block on a journal
reservation, the second blocking call into the journal code was asking
for less space, which is not what we want.

Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22 17:08:22 -04:00
Kent Overstreet
58fbf80834 bcachefs: Delete duplicate code
Also rename for consistency

Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22 17:08:22 -04:00
Kent Overstreet
6fb076e60d bcachefs: Fix spurious inconsistency in recovery
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22 17:08:22 -04:00
Kent Overstreet
ed8413fdab bcachefs: improved btree locking tracepoints
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22 17:08:22 -04:00
Kent Overstreet
17758a6c23 bcachefs: bch2_btree_delete_at_range()
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22 17:08:22 -04:00
Kent Overstreet
42c7d748e4 bcachefs: stripe creation fixes
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22 17:08:22 -04:00
Kent Overstreet
6009b4e508 bcachefs: Merge extents with checksums
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22 17:08:22 -04:00
Kent Overstreet
60755344c6 bcachefs: kill BTREE_ITER_NOUNLOCK
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22 17:08:22 -04:00
Kent Overstreet
b03b81dfd2 bcachefs: Don't pass around may_drop_locks
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22 17:08:22 -04:00
Kent Overstreet
b7607ce98f bcachefs: Kill remaining bch2_btree_iter_unlock() uses
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22 17:08:21 -04:00
Kent Overstreet
7cfac5f506 bcachefs: Fix for the stripes mark path and gc
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22 17:08:21 -04:00
Kent Overstreet
ae0ff7b8b0 bcachefs: Redo replicas gc mechanism
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22 17:08:21 -04:00
Kent Overstreet
9516950c06 bcachefs: Fix return code from bch2_fs_start()
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22 17:08:21 -04:00
Kent Overstreet
44e63bcaaa bcachefs: Don't run fsck by default at mount time
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22 17:08:21 -04:00
Kent Overstreet
619f5bee86 bcachefs: some improvements to startup messages and options
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22 17:08:21 -04:00
Kent Overstreet
460651ee86 bcachefs: Various improvements to bch2_alloc_write()
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22 17:08:21 -04:00
Kent Overstreet
932aa83745 bcachefs: bch2_trans_mark_update()
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22 17:08:21 -04:00
Kent Overstreet
c43a6ef9a0 bcachefs: btree_bkey_cached_common
This is prep work for the btree key cache: btree iterators will point to
either struct btree, or a new struct bkey_cached.

Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22 17:08:21 -04:00
Kent Overstreet
5e82a9a1f4 bcachefs: Write out fs usage consistently
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22 17:08:21 -04:00
Kent Overstreet
fca1223ccf bcachefs: Avoid write lock on mark_lock
mark_lock is a frequently taken lock, and there's also potential for
deadlocks since currently bch2_clear_page_bits which is called from
memory reclaim has to take it to drop disk reservations.

The disk reservation get path takes it when it recalculates the number
of sectors known to be available, but it's not really needed for
consistency.  We just want to make sure we only have one thread updating
the sectors_available count, which we can do with a dedicated mutex.

Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22 17:08:21 -04:00
Kent Overstreet
33eb63e575 bcachefs: Fix a bug with multiple iterators being traversed
If upgrade fails on one iterator, but it was copied from another
iterator and will be freed before transaction restart, then the original
iterator will get traversed first, so we need to make required btree
nodes on the original iterator will be traversed too.

Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22 17:08:21 -04:00
Kent Overstreet
1fe93f88c3 bcachefs: fix bch2_rbio_narrow_crcs()
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22 17:08:21 -04:00
Kent Overstreet
1f7d45beb7 bcachefs: Fix journal shutdown path
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22 17:08:21 -04:00
Kent Overstreet
75812e70d9 bcachefs: Fix fsync error reporting
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22 17:08:21 -04:00