Age | Commit message (Collapse) | Author | |
---|---|---|---|
2016-11-02 | bcache: get rid of a memcmp() in bch_btree_node_get() | Kent Overstreet | |
2016-11-02 | bcache: don't touch journal flags in fast paths | Kent Overstreet | |
2016-11-02 | bcache: optimize bch_btree_node_iter_init() | Kent Overstreet | |
Avoid the memmove in bch_btree_node_iter_push() by just doing a sort at the end. | |||
2016-10-27 | bcache: delete some duplicated code | Kent Overstreet | |
bch_btree_node_fill() does the unlock | |||
2016-10-27 | bcache: rename some bset aux tree stuff | Kent Overstreet | |
2016-10-27 | bcachefs: be consistent with generic_write_sync() | Kent Overstreet | |
2016-10-25 | bcache: fix handling of ptr_offset in allocation code | Kent Overstreet | |
only affects replication | |||
2016-10-24 | bcache: fix for btree code getting wrong journal sequence number | Kent Overstreet | |
2016-10-21 | bcache: shrink struct btree a bit | Kent Overstreet | |
2016-10-20 | bcache: Make btree coalescing triggerable from sysfs | Kent Overstreet | |
2016-10-20 | bcache: improve btree node coalesce tracepoints | Kent Overstreet | |
2016-10-18 | bcache: Make debugfs interface a bit more informative | Kent Overstreet | |
2016-10-14 | bcache: workaround for buggy FUA | Kent Overstreet | |
2016-10-14 | bcache: fix a shutdown deadlock | Kent Overstreet | |
2016-10-10 | bcache: fix a race in the journalling code | Kent Overstreet | |
2016-10-10 | bcache: hack around a race with journal reclaim | Kent Overstreet | |
2016-10-10 | bcache: rip out a bunch more data move code | Kent Overstreet | |
switch to a much much simpler mechanism for perserving io ordering/locality | |||
2016-10-10 | bcache: rip out bch_queue_start()/stop() | Kent Overstreet | |
this code wasn't even doing anything, since when shutting down tiering/copygc we were stopping the kthreads before stopping the queues... | |||
2016-10-10 | bcache: rip out generational gc | Kent Overstreet | |
generational copygc has the side effect of potentially fragmenting data, we probably don't do it by default, if at all | |||
2016-10-10 | bcache: move ratelimit accounting to more sensible location | Kent Overstreet | |
2016-10-10 | bcache: create a sensible crc entry for ptrs lacking crcs | Kent Overstreet | |
2016-10-07 | bcache: More fsck improvements | Kent Overstreet | |
2016-10-07 | bcache: copygc fixes | Kent Overstreet | |
Copygc now passes really nasty torture tests without deadlocking | |||
2016-10-07 | bcache: another allocator fixup | Kent Overstreet | |
2016-10-07 | bcache: tweak a fault injection thing | Kent Overstreet | |
copygc doesn't retry stale reads, because if the read was stale that means the data was overwritten and doesn't have to be moved anymore - so injecting a fault there causes it to not move data it should have, and breaks tests. | |||
2016-10-07 | bcache: Simpler keylists | Kent Overstreet | |
With scan keylists gone, keylists don't have to hold more than a few keys anymore, and we can go back to a much simpler implementation. Yay for deleting code. This also means we can finally fix the definition of BKEY_EXTENT_VAL_U64s_MAX... | |||
2016-10-07 | bcache: scan keylists get to die in a fire | Kent Overstreet | |
2016-10-07 | bcache: add a journal error check to __bch_btree_reserve_get() | Kent Overstreet | |
2016-10-07 | bcache: better error messages | Kent Overstreet | |
2016-10-07 | bcache: fix an awesome deadlock | Kent Overstreet | |
2016-10-07 | bcache: fix a divide by zero | Kent Overstreet | |
2016-10-07 | bcache: fix tiering + compression | Kent Overstreet | |
2016-10-07 | bcache: fix an assertion in bch_extent_invalid() | Kent Overstreet | |
2016-10-07 | bcache: fix corner case in allocation code with replication | Kent Overstreet | |
2016-10-07 | bcache: add mount option to fix errors | Kent Overstreet | |
2016-10-07 | bcache: don't mark as clean when there were errors | Kent Overstreet | |
2016-10-07 | bcache: Make allocation code more intelligent | Kent Overstreet | |
- rework bch_bucket_alloc_set so that it doesn't fail unless it can't allocate from any devices - split out the policy crap - "foreground writes try to allocate from tier 0 first" - into a separate wrapper function - make btree allocations also just prefer tier 0, instead of _only_ allocating from tier 0 | |||
2016-10-07 | bcache: Track inodes in a journal write with a bloom filter | Kent Overstreet | |
This lets us make fsync work when the inode's been evicted - fixes xfstests 106. | |||
2016-10-07 | bcache: ensure that data writes aren't cached if journal write won't do flush | Kent Overstreet | |
2016-10-07 | bcache: improve alloc_debug | Kent Overstreet | |
2016-10-07 | bcache: more useful checksum errors | Kent Overstreet | |
2016-10-07 | bcache: git rid of a branch in bkey_pack_key() | Kent Overstreet | |
2016-10-07 | bcache: fix a lost wakeup | Kent Overstreet | |
the way bch_next_cache() was using buckets_free_cache() was racy | |||
2016-10-07 | bcachefs: fs-gc, fsck improvements | Kent Overstreet | |
Make some checks tighter | |||
2016-10-07 | bcache: fix bch_btree_iter_advance_pos() | Kent Overstreet | |
iter->k doesn't get updated when bch_btree_iter_peek() returns null | |||
2016-10-07 | bcachefs: update i_nlink synchronously | Kent Overstreet | |
Fixes a bug uncovered by the previous patch | |||
2016-10-07 | bcache: add flag to superblock indicating if we shutdown cleanly | Kent Overstreet | |
2016-10-07 | bcache: fix ordering of index updates after unclean shutdown | Kent Overstreet | |
the journal sequence number blacklist machinery doesn't work if we don't actually hit all the btree nodes before starting journal replay... | |||
2016-10-07 | bcache: fix an ordering issue in btree_interior_update code | Kent Overstreet | |
Also a bunch of refactoring | |||
2016-10-07 | bcache: add bch_journal_pin_add_if_older() | Kent Overstreet | |
also uninline the journal pin stuff, there's enough callers (or will be soon) that it shouldn't be inlined. |