iv/linux
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
1ff7849f3b
linux/fs/bcachefs/btree_key_cache.h

48 lines
1.6 KiB
C
Raw Normal View History

bcachefs: Btree key cache This introduces a new kind of btree iterator, cached iterators, which point to keys cached in a hash table. The cache also acts as a write cache - in the update path, we journal the update but defer updating the btree until the cached entry is flushed by journal reclaim. Cache coherency is for now up to the users to handle, which isn't ideal but should be good enough for now. These new iterators will be used for updating inodes and alloc info (the alloc and stripes btrees). Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2019-03-08 03:46:10 +03:00
#ifndef _BCACHEFS_BTREE_KEY_CACHE_H
#define _BCACHEFS_BTREE_KEY_CACHE_H
bcachefs: Ensure journal reclaim runs when btree key cache is too dirty Ensuring the key cache isn't too dirty is critical for ensuring that the shrinker can reclaim memory. Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2020-11-20 03:54:40 +03:00
static inline size_t bch2_nr_btree_keys_need_flush(struct bch_fs *c)
{
bcachefs: btree key cache locking improvements The btree key cache mutex was becoming a significant bottleneck - it was mainly used to protect the lists of dirty, clean and freed cached keys. This patch eliminates the dirty and clean lists - instead, when we need to scan for keys to drop from the cache we iterate over the rhashtable, and thus we're able to remove most uses of that lock. Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2021-03-25 06:37:33 +03:00
size_t nr_dirty = atomic_long_read(&c->btree_key_cache.nr_dirty);
size_t nr_keys = atomic_long_read(&c->btree_key_cache.nr_keys);
bcachefs: Fix btree key cache dirty checks Had a type that meant we were triggering journal reclaim _much_ more aggressively than needed. Also, fix a potential integer overflow. Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2020-12-06 05:03:57 +03:00
size_t max_dirty = 1024 + nr_keys / 2;
bcachefs: Ensure journal reclaim runs when btree key cache is too dirty Ensuring the key cache isn't too dirty is critical for ensuring that the shrinker can reclaim memory. Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2020-11-20 03:54:40 +03:00
return max_t(ssize_t, 0, nr_dirty - max_dirty);
}
bcachefs: Throttle updates when btree key cache is too dirty This is needed to ensure we don't deadlock because journal reclaim and thus memory reclaim isn't making forward progress. Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2020-11-20 05:40:03 +03:00
static inline bool bch2_btree_key_cache_must_wait(struct bch_fs *c)
{
bcachefs: btree key cache locking improvements The btree key cache mutex was becoming a significant bottleneck - it was mainly used to protect the lists of dirty, clean and freed cached keys. This patch eliminates the dirty and clean lists - instead, when we need to scan for keys to drop from the cache we iterate over the rhashtable, and thus we're able to remove most uses of that lock. Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2021-03-25 06:37:33 +03:00
size_t nr_dirty = atomic_long_read(&c->btree_key_cache.nr_dirty);
size_t nr_keys = atomic_long_read(&c->btree_key_cache.nr_keys);
bcachefs: Throttle updates when btree key cache is too dirty This is needed to ensure we don't deadlock because journal reclaim and thus memory reclaim isn't making forward progress. Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2020-11-20 05:40:03 +03:00
size_t max_dirty = 4096 + (nr_keys * 3) / 4;
bcachefs: Simplify journal replay With BTREE_ITER_WITH_JOURNAL, there's no longer any restrictions on the order we have to replay keys from the journal in, and we can also start up journal reclaim right away - and delete a bunch of code. Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
2021-12-28 07:10:06 +03:00
return nr_dirty > max_dirty;
bcachefs: Throttle updates when btree key cache is too dirty This is needed to ensure we don't deadlock because journal reclaim and thus memory reclaim isn't making forward progress. Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2020-11-20 05:40:03 +03:00
}
bcachefs: Don't flush btree writes more aggressively because of btree key cache We need to flush the btree key cache when it's too dirty, because otherwise the shrinker won't be able to reclaim memory - this is done by journal reclaim. But journal reclaim also kicks btree node writes: this meant that btree node writes were getting kicked much too often just because we needed to flush btree key cache keys. This patch splits journal pins into two different lists, and teaches journal reclaim to not flush btree node writes when it only needs to flush key cache keys. Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2021-04-01 04:44:55 +03:00
int bch2_btree_key_cache_journal_flush(struct journal *,
struct journal_entry_pin *, u64);
bcachefs: Use cached iterators for inode updates This switches inode updates to use cached btree iterators - which should be a nice performance boost, since lock contention on the inodes btree can be a bottleneck on multithreaded workloads. Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2019-09-23 02:10:21 +03:00
struct bkey_cached *
bch2_btree_key_cache_find(struct bch_fs *, enum btree_id, struct bpos);
bcachefs: btree_path This splits btree_iter into two components: btree_iter is now the externally visible componont, and it points to a btree_path which is now reference counted. This means we no longer have to clone iterators up front if they might be mutated - btree_path can be shared by multiple iterators, and cloned if an iterator would mutate a shared btree_path. This will help us use iterators more efficiently, as well as slimming down the main long lived state in btree_trans, and significantly cleans up the logic for iterator lifetimes. Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2021-08-30 22:18:31 +03:00
int bch2_btree_path_traverse_cached(struct btree_trans *, struct btree_path *,
unsigned);
bcachefs: Btree key cache This introduces a new kind of btree iterator, cached iterators, which point to keys cached in a hash table. The cache also acts as a write cache - in the update path, we journal the update but defer updating the btree until the cached entry is flushed by journal reclaim. Cache coherency is for now up to the users to handle, which isn't ideal but should be good enough for now. These new iterators will be used for updating inodes and alloc info (the alloc and stripes btrees). Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2019-03-08 03:46:10 +03:00
bool bch2_btree_insert_key_cached(struct btree_trans *,
bcachefs: btree_path This splits btree_iter into two components: btree_iter is now the externally visible componont, and it points to a btree_path which is now reference counted. This means we no longer have to clone iterators up front if they might be mutated - btree_path can be shared by multiple iterators, and cloned if an iterator would mutate a shared btree_path. This will help us use iterators more efficiently, as well as slimming down the main long lived state in btree_trans, and significantly cleans up the logic for iterator lifetimes. Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2021-08-30 22:18:31 +03:00
struct btree_path *, struct bkey_i *);
bcachefs: Btree key cache This introduces a new kind of btree iterator, cached iterators, which point to keys cached in a hash table. The cache also acts as a write cache - in the update path, we journal the update but defer updating the btree until the cached entry is flushed by journal reclaim. Cache coherency is for now up to the users to handle, which isn't ideal but should be good enough for now. These new iterators will be used for updating inodes and alloc info (the alloc and stripes btrees). Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2019-03-08 03:46:10 +03:00
int bch2_btree_key_cache_flush(struct btree_trans *,
enum btree_id, struct bpos);
bcachefs: Btree key cache coherency - Updates to non key cache iterators will now be transparently redirected to the key cache for cached btrees. - Except when creating new keys: then the update goes to underlying btree For for iterating over a cached btree to work, we need to ensure that if a key exists in the key cache, it also exists in the btree - otherwise the iterator code will skip past it and not check the key cache. Otherwise, for consistency, all updates should go to the same place - the key cache. Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2022-01-12 09:14:47 +03:00
void bch2_btree_key_cache_drop(struct btree_trans *,
struct btree_path *);
bcachefs: Btree key cache This introduces a new kind of btree iterator, cached iterators, which point to keys cached in a hash table. The cache also acts as a write cache - in the update path, we journal the update but defer updating the btree until the cached entry is flushed by journal reclaim. Cache coherency is for now up to the users to handle, which isn't ideal but should be good enough for now. These new iterators will be used for updating inodes and alloc info (the alloc and stripes btrees). Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2019-03-08 03:46:10 +03:00
void bch2_fs_btree_key_cache_exit(struct btree_key_cache *);
void bch2_fs_btree_key_cache_init_early(struct btree_key_cache *);
int bch2_fs_btree_key_cache_init(struct btree_key_cache *);
bcachefs: Fix lock ordering with new btree cache code The code that checks lock ordering was recently changed to go off of the pos of the btree node, rather than the iterator, but the btree cache code didn't update to handle iterators that point to cached bkeys. Oops Also, update various debug code. Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2020-06-16 02:53:46 +03:00
void bch2_btree_key_cache_to_text(struct printbuf *, struct btree_key_cache *);
bcachefs: Add a kmem_cache for btree_key_cache objects We allocate a lot of these, and we're seeing sporading OOMs - this will help with tracking those down. Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2020-11-18 22:09:33 +03:00
void bch2_btree_key_cache_exit(void);
int __init bch2_btree_key_cache_init(void);
bcachefs: Btree key cache This introduces a new kind of btree iterator, cached iterators, which point to keys cached in a hash table. The cache also acts as a write cache - in the update path, we journal the update but defer updating the btree until the cached entry is flushed by journal reclaim. Cache coherency is for now up to the users to handle, which isn't ideal but should be good enough for now. These new iterators will be used for updating inodes and alloc info (the alloc and stripes btrees). Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2019-03-08 03:46:10 +03:00
#endif /* _BCACHEFS_BTREE_KEY_CACHE_H */
Reference in New Issue Copy Permalink