summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorMingzhe Zou <mingzhe.zou@easystack.cn>2023-06-15 20:12:23 +0800
committerJens Axboe <axboe@kernel.dk>2023-06-15 07:32:55 -0600
commitf0854489fc07d2456f7cc71a63f4faf9c716ffbe (patch)
tree3bd78d0064c665d9d3852ea7cce972bf314d8c56
parent80fca8a10b604afad6c14213fdfd816c4eda3ee4 (diff)
downloadlwn-f0854489fc07d2456f7cc71a63f4faf9c716ffbe.tar.gz
lwn-f0854489fc07d2456f7cc71a63f4faf9c716ffbe.zip
bcache: fixup btree_cache_wait list damage
We get a kernel crash about "list_add corruption. next->prev should be prev (ffff9c801bc01210), but was ffff9c77b688237c. (next=ffffae586d8afe68)." crash> struct list_head 0xffff9c801bc01210 struct list_head { next = 0xffffae586d8afe68, prev = 0xffffae586d8afe68 } crash> struct list_head 0xffff9c77b688237c struct list_head { next = 0x0, prev = 0x0 } crash> struct list_head 0xffffae586d8afe68 struct list_head struct: invalid kernel virtual address: ffffae586d8afe68 type: "gdb_readmem_callback" Cannot access memory at address 0xffffae586d8afe68 [230469.019492] Call Trace: [230469.032041] prepare_to_wait+0x8a/0xb0 [230469.044363] ? bch_btree_keys_free+0x6c/0xc0 [escache] [230469.056533] mca_cannibalize_lock+0x72/0x90 [escache] [230469.068788] mca_alloc+0x2ae/0x450 [escache] [230469.080790] bch_btree_node_get+0x136/0x2d0 [escache] [230469.092681] bch_btree_check_thread+0x1e1/0x260 [escache] [230469.104382] ? finish_wait+0x80/0x80 [230469.115884] ? bch_btree_check_recurse+0x1a0/0x1a0 [escache] [230469.127259] kthread+0x112/0x130 [230469.138448] ? kthread_flush_work_fn+0x10/0x10 [230469.149477] ret_from_fork+0x35/0x40 bch_btree_check_thread() and bch_dirty_init_thread() may call mca_cannibalize() to cannibalize other cached btree nodes. Only one thread can do it at a time, so the op of other threads will be added to the btree_cache_wait list. We must call finish_wait() to remove op from btree_cache_wait before free it's memory address. Otherwise, the list will be damaged. Also should call bch_cannibalize_unlock() to release the btree_cache_alloc_lock and wake_up other waiters. Fixes: 8e7102273f59 ("bcache: make bch_btree_check() to be multithreaded") Fixes: b144e45fc576 ("bcache: make bch_sectors_dirty_init() to be multithreaded") Cc: stable@vger.kernel.org Signed-off-by: Mingzhe Zou <mingzhe.zou@easystack.cn> Signed-off-by: Coly Li <colyli@suse.de> Link: https://lore.kernel.org/r/20230615121223.22502-7-colyli@suse.de Signed-off-by: Jens Axboe <axboe@kernel.dk>
-rw-r--r--drivers/md/bcache/btree.c11
-rw-r--r--drivers/md/bcache/btree.h1
-rw-r--r--drivers/md/bcache/writeback.c10
3 files changed, 21 insertions, 1 deletions
diff --git a/drivers/md/bcache/btree.c b/drivers/md/bcache/btree.c
index 0ddf91204782..68b9d7ca864e 100644
--- a/drivers/md/bcache/btree.c
+++ b/drivers/md/bcache/btree.c
@@ -885,7 +885,7 @@ static struct btree *mca_cannibalize(struct cache_set *c, struct btree_op *op,
* cannibalize_bucket() will take. This means every time we unlock the root of
* the btree, we need to release this lock if we have it held.
*/
-static void bch_cannibalize_unlock(struct cache_set *c)
+void bch_cannibalize_unlock(struct cache_set *c)
{
spin_lock(&c->btree_cannibalize_lock);
if (c->btree_cache_alloc_lock == current) {
@@ -1970,6 +1970,15 @@ static int bch_btree_check_thread(void *arg)
c->gc_stats.nodes++;
bch_btree_op_init(&op, 0);
ret = bcache_btree(check_recurse, p, c->root, &op);
+ /*
+ * The op may be added to cache_set's btree_cache_wait
+ * in mca_cannibalize(), must ensure it is removed from
+ * the list and release btree_cache_alloc_lock before
+ * free op memory.
+ * Otherwise, the btree_cache_wait will be damaged.
+ */
+ bch_cannibalize_unlock(c);
+ finish_wait(&c->btree_cache_wait, &(&op)->wait);
if (ret)
goto out;
}
diff --git a/drivers/md/bcache/btree.h b/drivers/md/bcache/btree.h
index 1b5fdbc0d83e..a2920bbfcad5 100644
--- a/drivers/md/bcache/btree.h
+++ b/drivers/md/bcache/btree.h
@@ -282,6 +282,7 @@ void bch_initial_gc_finish(struct cache_set *c);
void bch_moving_gc(struct cache_set *c);
int bch_btree_check(struct cache_set *c);
void bch_initial_mark_key(struct cache_set *c, int level, struct bkey *k);
+void bch_cannibalize_unlock(struct cache_set *c);
static inline void wake_up_gc(struct cache_set *c)
{
diff --git a/drivers/md/bcache/writeback.c b/drivers/md/bcache/writeback.c
index d4a5fc0650bb..24c049067f61 100644
--- a/drivers/md/bcache/writeback.c
+++ b/drivers/md/bcache/writeback.c
@@ -890,6 +890,16 @@ static int bch_root_node_dirty_init(struct cache_set *c,
if (ret < 0)
pr_warn("sectors dirty init failed, ret=%d!\n", ret);
+ /*
+ * The op may be added to cache_set's btree_cache_wait
+ * in mca_cannibalize(), must ensure it is removed from
+ * the list and release btree_cache_alloc_lock before
+ * free op memory.
+ * Otherwise, the btree_cache_wait will be damaged.
+ */
+ bch_cannibalize_unlock(c);
+ finish_wait(&c->btree_cache_wait, &(&op.op)->wait);
+
return ret;
}