summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2024-01-01bcachefs: growable btree_pathsKent Overstreet
XXX: we're allocating memory with btree locks held - bad We need to plumb through an error path so we can do allocate_dropping_locks() - but we're merging this now because it fixes a transaction path overflow caused by indirect extent fragmentation, and the resize path is rare. Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-01-01bcachefs: Fix interior update path btree_path usesKent Overstreet
Since the btree_paths array is now about to become growable, we have to be careful not to refer to paths by pointer across contexts where they may be reallocated. This fixes the remaining btree_interior_update() paths - split and merge. Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-01-01bcachefs: trans->nr_pathsKent Overstreet
Start to plumb through dynamically growable btree_paths; this patch replaces most BTREE_ITER_MAX references with trans->nr_paths. Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-01-01bcachefs: trans->updates will also be resizableKent Overstreet
the reflink triggers are also bumping up against the maximum number of paths in a transaction - and generating proportional numbers of updates. Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-01-01bcachefs: optimize __bch2_trans_get(), kill DEBUG_TRANSACTIONSKent Overstreet
- Some tweaks to greatly reduce locking overhead for the list of btree transactions, so that it can always be enabled: leave btree_trans objects on the list when they're on the percpu single item freelist, and only check for duplicates in the same process when CONFIG_BCACHEFS_DEBUG is enabled - don't zero out the full btree_trans() unless we allocated it from the mempool Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-01-01bcachefs: rcu protect trans->pathsKent Overstreet
Upcoming patches are going to be changing trans->paths to a reallocatable buffer. We need to guard against use after free when it's used by other threads; this introduces RCU protection to those paths and changes them to check for trans->paths == NULL Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-01-01bcachefs: Clean up btree_transKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-01-01bcachefs: kill btree_path.idxKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-01-01bcachefs: get_unlocked_mut_path() -> btree_path_idx_tKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-01-01bcachefs: bch2_btree_iter_peek_prev() no longer uses path->idxKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-01-01bcachefs: bch2_path_get() no longer uses path->idxKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-01-01bcachefs: trans_for_each_path_with_node() no longer uses path->idxKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-01-01bcachefs: trans_for_each_path() no longer uses path->idxKent Overstreet
path->idx is now a code smell: we should be using path_idx_t, since it's stable across btree path reallocation. This is also a bit faster, using the same loop counter vs. fetching path->idx from each path we iterate over. Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-01-01bcachefs: kill trans_for_each_path_from()Kent Overstreet
dead code Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-01-01bcachefs: bch2_btree_path_to_text() -> btree_path_idx_tKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-01-01bcachefs: struct trans_for_each_path_inorder_iterKent Overstreet
reducing our usage of path->idx Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-01-01bcachefs: btree_insert_entry -> btree_path_idx_tKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-01-01bcachefs: btree_iter -> btree_path_idx_tKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-01-01bcachefs: btree_path_alloc() -> btree_path_idx_tKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-01-01bcachefs: bch2_btree_path_traverse() -> btree_path_idx_tKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-01-01bcachefs: bch2_btree_path_make_mut() -> btree_path_idx_tKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-01-01bcachefs: bch2_btree_path_set_pos() -> btree_path_idx_tKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-01-01bcachefs; bch2_path_put() -> btree_path_idx_tKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-01-01bcachefs: bch2_path_get() -> btree_path_idx_tKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-01-01bcachefs: minor bch2_btree_path_set_pos() optimizationKent Overstreet
bpos_eq() is cheaper than bpos_cmp() Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-01-01bcachefs: Kill GFP_NOFAIL usage in readahead pathKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-01-01bcachefs: Convert split_devs() to darrayKent Overstreet
Bit of cleanup & modernization: also moving this code to util.c, it'll be used by userspace as well. Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-01-01bcachefs: skip journal more often in key cache reclaimKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-01-01bcachefs: for_each_keylist_key() declares loop iterKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-01-01bcachefs: bkey_for_each_ptr() now declares loop iterKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-01-01bcachefs: kill __bch2_btree_iter_peek_upto_and_restart()Kent Overstreet
dead code Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-01-01bcachefs: fsck -> bch2_trans_run()Kent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-01-01bcachefs: vstruct_for_each() now declares loop iterKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-01-01bcachefs: for_each_member_device_rcu() now declares loop iterKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-01-01bcachefs: for_each_member_device() now declares loop iterKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-01-01bcachefs: for_each_btree_key() now declares loop iterKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-01-01bcachefs: kill for_each_btree_key_norestart()Kent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-01-01bcachefs: kill for_each_btree_key_old_upto()Kent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-01-01bcachefs: for_each_btree_key_upto() -> for_each_btree_key_old_upto()Kent Overstreet
And for_each_btree_key2_upto -> for_each_btree_key_upto Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-01-01bcachefs: bch2_dirent_lookup() -> lockrestart_do()Kent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-01-01bcachefs: bch2_trans_srcu_lock() should be staticKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-01-01bcachefs: use track_event_change() for allocator blocked statsKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-01-01bcachefs: fix warning about uninitialized time_statsKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-01-01bcachefs: add more verbose loggingKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-01-01bcachefs: better error message in btree_node_write_work()Kent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-01-01bcachefs: simplify bch_devs_listKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-01-01bcachefs: darray_for_each() now declares loop iterKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-01-01bcachefs: trans_for_each_update() now declares loop iterKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-01-01bcachefs: Improve the nopromote tracepointKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-01-01bcachefs: Use GFP_KERNEL for promote allocationsKent Overstreet
We already have btree locks dropped here - no need for GFP_NOFS. Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
generated by cgit v1.2.3 (git 2.39.1) at 2025-07-29 16:05:48 +0000