linux-bcache.git - Unnamed repository; edit this file 'description' to name the repository.

Age	Commit message (Collapse)	Author
2016-12-12	bcache: treebitvec thingtreebitvec	Kent Overstreet

2016-12-12	bcache: don't flush journal unnecessarily	Kent Overstreet

2016-12-12	bcache: short circuit bch_prio_write() and journal flush when possible	Kent Overstreet

2016-12-12	bcache: add field to superblock for maximum journal entry size	Kent Overstreet

2016-12-12	block: add sysfs entry for fua support	Kent Overstreet

2016-12-12	bcache: Refactor write path to not split bios unnecessarily	Kent Overstreet
	compression code is a good deal saner now, too
2016-12-11	lz4: return amount of data we were able to compress on failure	Kent Overstreet

2016-12-11	bcache: stash journal entry sequence number in struct journal_res	Kent Overstreet

2016-12-11	bcache: cancel journal write timer when doing journal write	Kent Overstreet

2016-12-11	bcache: fix journal bug with prio_buckets	Kent Overstreet
	this code is all really stupid, need to finish off the new transaction machinery and get prio/gen journal entry converted to that
2016-12-09	bcachefs: fix find_get_entry() usage	Kent Overstreet

2016-12-09	bcache: increase default journal write delay for spinning rust	Kent Overstreet

2016-12-08	bcache: factor out eytzinger.h	Kent Overstreet

2016-12-07	bcache: fix bad shifts in inorder_to_tree()	Kent Overstreet
	caught by ubsan
2016-12-07	bcache: fix hash usage for journal has_inode hash	Kent Overstreet

2016-12-07	bcache: fix uninitialized field	Kent Overstreet
	bug was only when replication was enabled
2016-12-06	bcache: rw aux search tree improvement	Kent Overstreet

2016-12-06	bcache: bkey_unpack_pos()	Kent Overstreet
	slightly more efficient when we don't have a dynamically generated bkey_unpack_key()
2016-12-06	bcache: add end_offset to struct bset_tree	Kent Overstreet
	duplicates bset->u64s - for better data locality
2016-12-06	bcache: drop a comparison in bset_search_tree()	Kent Overstreet

2016-12-06	bcache: squish struct btree some more	Kent Overstreet
	finally under half a kilobyte
2016-12-06	bcache: kill struct btree_keys	Kent Overstreet

2016-12-06	bcache: larger bfloats for base of aux search tree	Kent Overstreet

2016-12-06	bcache: kill bset_tree->prev_offset	Kent Overstreet

2016-12-06	bcache: squish struct btree some more	Kent Overstreet

2016-12-06	bcache: use a single buffer for btree node aux trees	Kent Overstreet

2016-12-06	bcache: uninline bkey_cmp_left_packed()	Kent Overstreet
	this is generally a slowpath
2016-12-06	bcache: more aux search tree microoptimizations	Kent Overstreet

2016-12-06	Revert "bcache: Optimization for aux search tree slowpath"	Kent Overstreet
	This reverts commit 27b3474087af56511602a52050a3f2ec0fc69325. Turned out to not really help, gonna go with a different approach.
2016-12-06	bcache: prefetch btree root in bch_btree_iter_init()	Kent Overstreet

2016-12-06	bcache: inline some fastpaths	Kent Overstreet

2016-12-06	bcache: btree_node_iter_init() optimizations	Kent Overstreet

2016-12-06	bcache: round packed bkey format fields up to nearest byte when possible	Kent Overstreet
	wasn't useful with the old bkey_unpack_key(), but with the new dynamic codegen bkey_unpack_key() we can take advantage of this to avoid shifts and generate smaller code
2016-12-06	bcache: jit bkey_unpack_key()	Kent Overstreet

2016-12-06	bcache: packed bkeys refactoring	Kent Overstreet

2016-12-06	bcache: asm bkey_cmp_packed()	Kent Overstreet
	shouldn't need asm for this, but gcc is generating retarded code... later, we should precompute bkey_format_key_bits() and stash that somewhere
2016-12-06	bcache: drop trace_bkey_pack_pos_fail	Kent Overstreet
	this isn't actually a slowpath - move the tracepoint to the actual slowpath
2016-12-06	bcache: fix another copygc deadlock	Kent Overstreet
	since btree node reserves got reworked, btree updates can end up blocking on copygc - which means we have to run index updates for copygc out of a separate workqueue. also fix another deadlock related to rwsems being unfair.
2016-12-06	bcache: change copygc assertion to just print a warning	Kent Overstreet
	also, fix the assertion...
2016-12-06	bcache: fix btree node write tracepoint	Kent Overstreet
	b->nsets was changed, and it wasn't updated
2016-12-05	bcachefs: fix a regression from extent whiteouts optimizations	Kent Overstreet
	checking bkey_whiteout(&s->insert->k->k) in extent_insert_committed() was incorrect, because if we were already finished, the insert key's size would be 0 and 0 size extents are always KEY_TYPE_DELETED (i.e. whiteouts).
2016-12-02	bcache: allow __bch_write() to fragment io more than twice without slowpath loop	Kent Overstreet

2016-11-28	bcache: Whiteout optimizations: extents	Kent Overstreet

2016-11-28	bcache: extent insert path refactoring	Kent Overstreet

2016-11-28	bcache: re-enable btree_node_write() idx_to_write	Kent Overstreet

2016-11-28	bcache: btree_node_verify() improvements	Kent Overstreet

2016-11-28	bcache: Move unwritten whiteouts to a separate area	Kent Overstreet

2016-11-23	bcache: Optimize btree node writes	Kent Overstreet

2016-11-23	bcache: fix a race with journal_seq_blacklist flushing	Kent Overstreet
	the code was checking for missing journal entries in journal replay - but by that point, journal_seq_blacklist entries could already have been flushed and removed, causing us to spuriously think we were missing journal entries
2016-11-23	bcache: New sorting infrastructure	Kent Overstreet
	The upcoming whiteout optimizations have to do new kinds of sorting, add some new infrastructure to abstract out the comparison function Later, we should use this for the sorts we do when reading in a btree node (bch_key_sort_fix_overlapping and bch_extent_sort_fix_overlapping).