summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2023-10-22bcachefs: Always increment bucket gen on bucket reuseKent Overstreet
Not doing so confuses copygc Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22bcachefs: Kill old allocator startup codeKent Overstreet
It's not needed anymore since we can now write to buckets before updating the alloc btree. Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22bcachefs: Improve assorted error messagesKent Overstreet
This also consolidates the various checks in bch2_mark_pointer() and bch2_trans_mark_pointer(). Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22bcachefs: Fix a deadlock in bch2_btree_node_get_sibling()Kent Overstreet
There was a bad interaction with bch2_btree_iter_set_pos_same_leaf(), which can leave a btree node locked that is just outside iter->pos, breaking the lock ordering checks in __bch2_btree_node_lock(). Ideally we should get rid of this corner case, but for now fix it locally with verbose comments. Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22bcachefs: Add debug code to print btree transactionsKent Overstreet
Intented to help debug deadlocks, since we can't use lockdep to check btree node lock ordering. Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22bcachefs: Set filesystem features earlier in fs init pathKent Overstreet
Before we were setting features after allocating btree nodes, which meant we were using the old btree pointer format. Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22bcachefs: Add an option to disable reflink supportKent Overstreet
Reflink might be buggy, so we're adding an option so users can help bisect what's going on. Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22bcachefs: Fixes for going ROKent Overstreet
Now that interior btree updates are fully transactional, we don't need to write out alloc info in a loop. However, interior btree updates do put more things in the journal, so we still need a loop in the RO sequence. Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22bcachefs: Don't require alloc btree to be updated before buckets are usedKent Overstreet
This is to break a circular dependency in the shutdown path. Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22bcachefs: fsck_error_lock requires GFP_NOFSKent Overstreet
this fixes a lockdep splat Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22bcachefs: Interior btree updates are now fully transactionalKent Overstreet
We now update the alloc info (bucket sector counts) atomically with journalling the update to the interior btree nodes, and we also set new btree roots atomically with the journalled part of the btree update. Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22bcachefs: Factor out bch2_fs_btree_interior_update_init()Kent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22bcachefs: Add a mechanism for passing extra journal entries to ↵Kent Overstreet
bch2_trans_commit() Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22bcachefs: Fix reading of alloc info after unclean shutdownKent Overstreet
When updates to interior nodes started being journalled, that meant that after an unclean shutdown, until journal replay is done we can't walk the btree without overlaying the updates from the journal. The initial btree gc was changed to walk the btree overlaying keys from the journal - but bch2_alloc_read() and bch2_stripes_read() were missed. Major whoops... Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22bcachefs: fix memalloc_nofs_restore() usageKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22bcachefs: Better error messages on bucket sector count overflowsKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22bcachefs: Be more rigorous about marking the filesystem cleanKent Overstreet
Previously, there was at least one error path where we could mark the filesystem clean when we hadn't sucessfully written out alloc info. Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22bcachefs: Handle printing of null bkeysKent Overstreet
This fixes a null ptr deref. Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22bcachefs: Add vmalloc fallback for decompress workspaceKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22bcachefs: Print out d_type in dirent_to_text()Kent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22bcachefs: fix stack corruptionYuxuan Shui
When a bkey_on_stack is passed to bch_read_indirect_extent, there is no guarantee that it will be big enough to hold the bkey. And bch_read_indirect_extent is not aware of bkey_on_stack to call realloc on it. This cause a stack corruption. This commit makes bch_read_indirect_extent aware of bkey_on_stack so it can call realloc when appropriate. Tested-by: Yuxuan Shui <yshuiv7@gmail.com> Signed-off-by: Yuxuan Shui <yshuiv7@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22bcachefs: Wrap vmap() in memalloc_nofs_save()/restore()Kent Overstreet
vmalloc() and vmap() don't take GFP_NOFS - this should be pushed further up the IO path, but for now just doing the simple fix. Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22bcachefs: Fix another iterator counting bugKent Overstreet
We were marking the end of where we could insert incorrectly for indirect extents. Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22bcachefs: Fix setquotaKent Overstreet
We were returning -EINTR because we were failing to retry the btree transaction. Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22bcachefs: Fix a workqueue deadlockKent Overstreet
writes running out of a workqueue (via dio path) could block and prevent other writes from calling bch2_write_index() and completing. Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22bcachefs: Validate that we read the correct btree nodeKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22bcachefs: Fixes for startup on very full filesystemsKent Overstreet
- Always pass BTREE_INSERT_USE_RESERVE when writing alloc btree keys - Don't strand buckest on the copygc freelist until after recovery is done and we're starting copygc. Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22bcachefs: Fix initialization of bounce mempoolsKent Overstreet
When they were converted to kvpmalloc pools they weren't converted to pass the actual size of the allocation. Oops. Also, validate the real length in the zstd decompression path. Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22bcachefs: Some compression improvementsKent Overstreet
In __bio_map_or_bounce(), the check for if the bio is physically contiguous is improved; it's now more readable and handles multi page but contiguous bios. Also when decompressing, we were doing a redundant memcpy in the case where we were able to use vmap to map a bio contigiously. Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22bcachefs: Fix two more deadlocksKent Overstreet
Deadlock on shutdown: btree_update_nodes_written() unblocks btree nodes from being written; after doing so, it has to check if they were marked as needing to be written and if so kick off those writes - if that doesn't happen, we'll never release journal pins and shutdown will get stuck when flushing the journal. There was an error path where this didn't happen, because in the error path we don't actually want those btree nodes write to happen; however, we still have to kick off the write path so the journal pins get released. The btree write path checks if we're in a journal error state and doesn't do the actual write if we are. Also - there was another deadlock because btree_update_nodes_written() was taking the btree update off of the unwritten_list too soon - before getting a journal reservation, which could fail and have to be retried. Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22bcachefs: Fix another deadlock in btree_update_nodes_written()Kent Overstreet
We also can't be blocking on btree node write locks while holding btree_interior_update_lock. Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22bcachefs: Add some printks for error pathsKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22bcachefs: Don't issue writes that are more than 1 MBKent Overstreet
the bcachefs io path in io.c can't bounce writes larger than that. Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22bcachefs: More fixes for counting extent update iteratorsKent Overstreet
This is unfortunately really fragile - hopefully we'll be able to think of a new approach at some point. Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22bcachefs: Fix a deadlockKent Overstreet
btree_node_lock_increment() was incorrectly skipping over the current iter when checking if we should increment a node we already have locked. Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22bcachefs: Handle -EINTR bch2_migrate_index_update()Kent Overstreet
peek_slot() shouldn't return -EINTR when there's only a single live iterator, but that's tricky to guarantee - we seem to be returning -EINTR when we shouldn't, but it's easy enough to handle in the caller. Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22bcachefs: Fix for the bkey compat pathKent Overstreet
In the write path, we were calling bch2_bkey_ops.compat() in the wrong place. Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22bcachefs: Add a few tracepointsKent Overstreet
Transaction restart tracing should probably be overhaulled at some point. Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22bcachefs: Slightly reduce btree split thresholdKent Overstreet
2/3rds performs a lot better than 3/4ths on the tested workloda, leading to significanly fewer btree node compactions. Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22bcachefs: Improve lockdep annotation in journalling codeKent Overstreet
bch2_journal_res_get() in nonblocking mode is equivalent to a trylock. Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22bcachefs: Fix a locking bug in bch2_journal_pin_copy()Kent Overstreet
There was a race where the src pin would be flushed - releasing the last pin on that sequence number - before adding the new journal pin. Oops. Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22bcachefs: Fix another deadlock in the btree interior update pathKent Overstreet
Can't take read locks on btree nodes while holding btree_interior_update_lock. Also, fix a bug where we were leaking journal prereservations. Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22bcachefs: Fix a locking bug in bch2_btree_ptr_debugcheck()Kent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22bcachefs: Account for ioclock slop when throttling rebalance threadKent Overstreet
This should fix an issue where the rebalance thread was spinning Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22bcachefs: Fix a deadlock on starting an interior btree updateKent Overstreet
Not legal to block on a journal prereservation with btree locks held. Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22bcachefs: Fix a debug mode assertionKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22bcachefs: Fix a debug assertionKent Overstreet
This assertion was passing the wrong btree node type when inserting into interior nodes. Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22bcachefs: Fix another error path locking bugKent Overstreet
btree_update_nodes_written() was leaking a btree node lock on failure to get a journal reservation. Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22bcachefs: Fix a null ptr deref during journal replayKent Overstreet
We were calling bch2_extent_can_insert() incorrectly; it should only be called when the extents-to-keys pass is running because that's when we could be splitting a compressed extent. Calling bch2_extent_can_insert() without passing in a disk reservation was causing a null ptr deref. Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2023-10-22bcachefs: Add another mssing bch2_trans_iter_put() callKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
generated by cgit (git 2.34.1) at 2025-04-25 05:09:57 +0000