From 5794351146199b9ac67a5ab1beab82be8bfd7b5d Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Thu, 25 Apr 2013 13:58:35 -0700 Subject: bcache: Refactor btree io The most significant change is that btree reads are now done synchronously, instead of asynchronously and doing the post read stuff from a workqueue. This was originally done because we can't block on IO under generic_make_request(). But - we already have a mechanism to punt cache lookups to workqueue if needed, so if we just use that we don't have to deal with the complexity of doing things asynchronously. The main benefit is this makes the locking situation saner; we can hold our write lock on the btree node until we're finished reading it, and we don't need that btree_node_read_done() flag anymore. Also, for writes, btree_write() was broken out into btree_node_write() and btree_leaf_dirty() - the old code with the boolean argument was dumb and confusing. The prio_blocked mechanism was improved a bit too, now the only counter is in struct btree_write, we don't mess with transfering a count from struct btree anymore. This required changing garbage collection to block prios at the start and unblock when it finishes, which is cleaner than what it was doing anyways (the old code had mostly the same effect, but was doing it in a convoluted way) And the btree iter btree_node_read_done() uses was converted to a real mempool. Signed-off-by: Kent Overstreet --- drivers/md/bcache/debug.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'drivers/md/bcache/debug.c') diff --git a/drivers/md/bcache/debug.c b/drivers/md/bcache/debug.c index 89fd5204924e..ae6096c6845d 100644 --- a/drivers/md/bcache/debug.c +++ b/drivers/md/bcache/debug.c @@ -144,7 +144,7 @@ void bch_btree_verify(struct btree *b, struct bset *new) v->written = 0; v->level = b->level; - bch_btree_read(v); + bch_btree_node_read(v); closure_wait_event(&v->io.wait, &cl, atomic_read(&b->io.cl.remaining) == -1); @@ -512,7 +512,7 @@ static ssize_t btree_fuzz(struct kobject *k, struct kobj_attribute *a, bch_btree_sort(b); fill->written = 0; - bch_btree_read_done(&fill->io.cl); + bch_btree_node_read_done(fill); if (b->sets[0].data->keys != fill->sets[0].data->keys || memcmp(b->sets[0].data->start, -- cgit v1.2.3 From 85b1492ee113486d871de7676a61f506a43ca475 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 14 May 2013 20:33:16 -0700 Subject: bcache: Rip out pkey()/pbtree() Old gcc doesnt like the struct hack, and it is kind of ugly. So finish off the work to convert pr_debug() statements to tracepoints, and delete pkey()/pbtree(). Signed-off-by: Kent Overstreet --- drivers/md/bcache/debug.c | 40 +++++++++++++++++++++++++--------------- 1 file changed, 25 insertions(+), 15 deletions(-) (limited to 'drivers/md/bcache/debug.c') diff --git a/drivers/md/bcache/debug.c b/drivers/md/bcache/debug.c index ae6096c6845d..82e3a07771ec 100644 --- a/drivers/md/bcache/debug.c +++ b/drivers/md/bcache/debug.c @@ -47,11 +47,10 @@ const char *bch_ptr_status(struct cache_set *c, const struct bkey *k) return ""; } -struct keyprint_hack bch_pkey(const struct bkey *k) +int bch_bkey_to_text(char *buf, size_t size, const struct bkey *k) { unsigned i = 0; - struct keyprint_hack r; - char *out = r.s, *end = r.s + KEYHACK_SIZE; + char *out = buf, *end = buf + size; #define p(...) (out += scnprintf(out, end - out, __VA_ARGS__)) @@ -75,16 +74,14 @@ struct keyprint_hack bch_pkey(const struct bkey *k) if (KEY_CSUM(k)) p(" cs%llu %llx", KEY_CSUM(k), k->ptr[1]); #undef p - return r; + return out - buf; } -struct keyprint_hack bch_pbtree(const struct btree *b) +int bch_btree_to_text(char *buf, size_t size, const struct btree *b) { - struct keyprint_hack r; - - snprintf(r.s, 40, "%zu level %i/%i", PTR_BUCKET_NR(b->c, &b->key, 0), - b->level, b->c->root ? b->c->root->level : -1); - return r; + return scnprintf(buf, size, "%zu level %i/%i", + PTR_BUCKET_NR(b->c, &b->key, 0), + b->level, b->c->root ? b->c->root->level : -1); } #if defined(CONFIG_BCACHE_DEBUG) || defined(CONFIG_BCACHE_EDEBUG) @@ -100,10 +97,12 @@ static void dump_bset(struct btree *b, struct bset *i) { struct bkey *k; unsigned j; + char buf[80]; for (k = i->start; k < end(i); k = bkey_next(k)) { + bch_bkey_to_text(buf, sizeof(buf), k); printk(KERN_ERR "block %zu key %zi/%u: %s", index(i, b), - (uint64_t *) k - i->d, i->keys, pkey(k)); + (uint64_t *) k - i->d, i->keys, buf); for (j = 0; j < KEY_PTRS(k); j++) { size_t n = PTR_BUCKET_NR(b->c, k, j); @@ -252,6 +251,7 @@ static void vdump_bucket_and_panic(struct btree *b, const char *fmt, va_list args) { unsigned i; + char buf[80]; console_lock(); @@ -262,7 +262,8 @@ static void vdump_bucket_and_panic(struct btree *b, const char *fmt, console_unlock(); - panic("at %s\n", pbtree(b)); + bch_btree_to_text(buf, sizeof(buf), b); + panic("at %s\n", buf); } void bch_check_key_order_msg(struct btree *b, struct bset *i, @@ -337,6 +338,7 @@ static ssize_t bch_dump_read(struct file *file, char __user *buf, { struct dump_iterator *i = file->private_data; ssize_t ret = 0; + char kbuf[80]; while (size) { struct keybuf_key *w; @@ -359,7 +361,8 @@ static ssize_t bch_dump_read(struct file *file, char __user *buf, if (!w) break; - i->bytes = snprintf(i->buf, PAGE_SIZE, "%s\n", pkey(&w->key)); + bch_bkey_to_text(kbuf, sizeof(kbuf), &w->key); + i->bytes = snprintf(i->buf, PAGE_SIZE, "%s\n", kbuf); bch_keybuf_del(&i->keys, w); } @@ -526,10 +529,17 @@ static ssize_t btree_fuzz(struct kobject *k, struct kobj_attribute *a, k < end(i); k = bkey_next(k), l = bkey_next(l)) if (bkey_cmp(k, l) || - KEY_SIZE(k) != KEY_SIZE(l)) + KEY_SIZE(k) != KEY_SIZE(l)) { + char buf1[80]; + char buf2[80]; + + bch_bkey_to_text(buf1, sizeof(buf1), k); + bch_bkey_to_text(buf2, sizeof(buf2), l); + pr_err("key %zi differs: %s != %s", (uint64_t *) k - i->d, - pkey(k), pkey(l)); + buf1, buf2); + } for (j = 0; j < 3; j++) { pr_err("**** Set %i ****", j); -- cgit v1.2.3 From 72c270612bd33192fa836ad0f2939af1ca218292 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Wed, 5 Jun 2013 06:24:39 -0700 Subject: bcache: Write out full stripes Now that we're tracking dirty data per stripe, we can add two optimizations for raid5/6: * If a stripe is already dirty, force writes to that stripe to writeback mode - to help build up full stripes of dirty data * When flushing dirty data, preferentially write out full stripes first if there are any. Signed-off-by: Kent Overstreet --- drivers/md/bcache/debug.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'drivers/md/bcache/debug.c') diff --git a/drivers/md/bcache/debug.c b/drivers/md/bcache/debug.c index 82e3a07771ec..1c8fd319846e 100644 --- a/drivers/md/bcache/debug.c +++ b/drivers/md/bcache/debug.c @@ -357,7 +357,7 @@ static ssize_t bch_dump_read(struct file *file, char __user *buf, if (i->bytes) break; - w = bch_keybuf_next_rescan(i->c, &i->keys, &MAX_KEY); + w = bch_keybuf_next_rescan(i->c, &i->keys, &MAX_KEY, dump_pred); if (!w) break; @@ -380,7 +380,7 @@ static int bch_dump_open(struct inode *inode, struct file *file) file->private_data = i; i->c = c; - bch_keybuf_init(&i->keys, dump_pred); + bch_keybuf_init(&i->keys); i->keys.last_scanned = KEY(0, 0, 0); return 0; -- cgit v1.2.3 From f3059a54610f6c516c0942d58b9435921768ce2d Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Wed, 15 May 2013 17:13:45 -0700 Subject: bcache: Delete fuzz tester This code has rotted and it hasn't been used in ages anyways. Signed-off-by: Kent Overstreet --- drivers/md/bcache/debug.c | 148 ---------------------------------------------- 1 file changed, 148 deletions(-) (limited to 'drivers/md/bcache/debug.c') diff --git a/drivers/md/bcache/debug.c b/drivers/md/bcache/debug.c index 1c8fd319846e..ba2ceee0fdbb 100644 --- a/drivers/md/bcache/debug.c +++ b/drivers/md/bcache/debug.c @@ -412,149 +412,6 @@ void bch_debug_init_cache_set(struct cache_set *c) #endif -/* Fuzz tester has rotted: */ -#if 0 - -static ssize_t btree_fuzz(struct kobject *k, struct kobj_attribute *a, - const char *buffer, size_t size) -{ - void dump(struct btree *b) - { - struct bset *i; - - for (i = b->sets[0].data; - index(i, b) < btree_blocks(b) && - i->seq == b->sets[0].data->seq; - i = ((void *) i) + set_blocks(i, b->c) * block_bytes(b->c)) - dump_bset(b, i); - } - - struct cache_sb *sb; - struct cache_set *c; - struct btree *all[3], *b, *fill, *orig; - int j; - - struct btree_op op; - bch_btree_op_init_stack(&op); - - sb = kzalloc(sizeof(struct cache_sb), GFP_KERNEL); - if (!sb) - return -ENOMEM; - - sb->bucket_size = 128; - sb->block_size = 4; - - c = bch_cache_set_alloc(sb); - if (!c) - return -ENOMEM; - - for (j = 0; j < 3; j++) { - BUG_ON(list_empty(&c->btree_cache)); - all[j] = list_first_entry(&c->btree_cache, struct btree, list); - list_del_init(&all[j]->list); - - all[j]->key = KEY(0, 0, c->sb.bucket_size); - bkey_copy_key(&all[j]->key, &MAX_KEY); - } - - b = all[0]; - fill = all[1]; - orig = all[2]; - - while (1) { - for (j = 0; j < 3; j++) - all[j]->written = all[j]->nsets = 0; - - bch_bset_init_next(b); - - while (1) { - struct bset *i = write_block(b); - struct bkey *k = op.keys.top; - unsigned rand; - - bkey_init(k); - rand = get_random_int(); - - op.type = rand & 1 - ? BTREE_INSERT - : BTREE_REPLACE; - rand >>= 1; - - SET_KEY_SIZE(k, bucket_remainder(c, rand)); - rand >>= c->bucket_bits; - rand &= 1024 * 512 - 1; - rand += c->sb.bucket_size; - SET_KEY_OFFSET(k, rand); -#if 0 - SET_KEY_PTRS(k, 1); -#endif - bch_keylist_push(&op.keys); - bch_btree_insert_keys(b, &op); - - if (should_split(b) || - set_blocks(i, b->c) != - __set_blocks(i, i->keys + 15, b->c)) { - i->csum = csum_set(i); - - memcpy(write_block(fill), - i, set_bytes(i)); - - b->written += set_blocks(i, b->c); - fill->written = b->written; - if (b->written == btree_blocks(b)) - break; - - bch_btree_sort_lazy(b); - bch_bset_init_next(b); - } - } - - memcpy(orig->sets[0].data, - fill->sets[0].data, - btree_bytes(c)); - - bch_btree_sort(b); - fill->written = 0; - bch_btree_node_read_done(fill); - - if (b->sets[0].data->keys != fill->sets[0].data->keys || - memcmp(b->sets[0].data->start, - fill->sets[0].data->start, - b->sets[0].data->keys * sizeof(uint64_t))) { - struct bset *i = b->sets[0].data; - struct bkey *k, *l; - - for (k = i->start, - l = fill->sets[0].data->start; - k < end(i); - k = bkey_next(k), l = bkey_next(l)) - if (bkey_cmp(k, l) || - KEY_SIZE(k) != KEY_SIZE(l)) { - char buf1[80]; - char buf2[80]; - - bch_bkey_to_text(buf1, sizeof(buf1), k); - bch_bkey_to_text(buf2, sizeof(buf2), l); - - pr_err("key %zi differs: %s != %s", - (uint64_t *) k - i->d, - buf1, buf2); - } - - for (j = 0; j < 3; j++) { - pr_err("**** Set %i ****", j); - dump(all[j]); - } - panic("\n"); - } - - pr_info("fuzz complete: %i keys", b->sets[0].data->keys); - } -} - -kobj_attribute_write(fuzz, btree_fuzz); -#endif - void bch_debug_exit(void) { if (!IS_ERR_OR_NULL(debug)) @@ -564,11 +421,6 @@ void bch_debug_exit(void) int __init bch_debug_init(struct kobject *kobj) { int ret = 0; -#if 0 - ret = sysfs_create_file(kobj, &ksysfs_fuzz.attr); - if (ret) - return ret; -#endif debug = debugfs_create_dir("bcache", NULL); return ret; -- cgit v1.2.3 From 8e51e414a3c6d92ef2cc41720c67342a8e2c0bf7 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Thu, 6 Jun 2013 18:15:57 -0700 Subject: bcache: Use standard utility code Some of bcache's utility code has made it into the rest of the kernel, so drop the bcache versions. Bcache used to have a workaround for allocating from a bio set under generic_make_request() (if you allocated more than once, the bios you already allocated would get stuck on current->bio_list when you submitted, and you'd risk deadlock) - bcache would mask out __GFP_WAIT when allocating bios under generic_make_request() so that allocation could fail and it could retry from workqueue. But bio_alloc_bioset() has a workaround now, so we can drop this hack and the associated error handling. Signed-off-by: Kent Overstreet --- drivers/md/bcache/debug.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers/md/bcache/debug.c') diff --git a/drivers/md/bcache/debug.c b/drivers/md/bcache/debug.c index ba2ceee0fdbb..88e6411eab4f 100644 --- a/drivers/md/bcache/debug.c +++ b/drivers/md/bcache/debug.c @@ -199,7 +199,7 @@ void bch_data_verify(struct search *s) if (!check) return; - if (bch_bio_alloc_pages(check, GFP_NOIO)) + if (bio_alloc_pages(check, GFP_NOIO)) goto out_put; check->bi_rw = READ_SYNC; -- cgit v1.2.3