summaryrefslogtreecommitdiff
path: root/drivers/md
diff options
context:
space:
mode:
authorColy Li <colyli@suse.de>2020-07-25 20:00:33 +0800
committerJens Axboe <axboe@kernel.dk>2020-07-25 07:38:20 -0600
commitbf6af17065079d29b9bd4e59de27cc2965e6fabf (patch)
tree2f7e5eb7382d26b28b6a2ae1d2fbea2584eb555a /drivers/md
parentc954ac8d6610255b2fb11a30fb9ebb206f3da48d (diff)
bcache: handle cache set verify_ondisk properly for bucket size > 8MB
In bch_btree_cache_alloc() when CONFIG_BCACHE_DEBUG is configured, allocate memory for c->verify_ondisk may fail if the bucket size > 8MB, which will require __get_free_pages() to allocate continuous pages with order > 11 (the default MAX_ORDER of Linux buddy allocator). Such over size allocation will fail, and cause 2 problems, - When CONFIG_BCACHE_DEBUG is configured, bch_btree_verify() does not work, because c->verify_ondisk is NULL and bch_btree_verify() returns immediately. - bch_btree_cache_alloc() will fail due to c->verify_ondisk allocation failed, then the whole cache device registration fails. And because of this failure, the first problem of bch_btree_verify() has no chance to be triggered. This patch fixes the above problem by two means, 1) If pages allocation of c->verify_ondisk fails, set it to NULL and returns bch_btree_cache_alloc() with -ENOMEM. 2) When calling __get_free_pages() to allocate c->verify_ondisk pages, use ilog2(meta_bucket_pages(&c->sb)) to make sure ilog2() will always generate a pages order <= MAX_ORDER (or CONFIG_FORCE_MAX_ZONEORDER). Then the buddy system won't directly reject the allocation request. Signed-off-by: Coly Li <colyli@suse.de> Reviewed-by: Hannes Reinecke <hare@suse.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
Diffstat (limited to 'drivers/md')
-rw-r--r--drivers/md/bcache/btree.c12
1 files changed, 10 insertions, 2 deletions
diff --git a/drivers/md/bcache/btree.c b/drivers/md/bcache/btree.c
index 1e3bd5ea5486..3d8bd0692af3 100644
--- a/drivers/md/bcache/btree.c
+++ b/drivers/md/bcache/btree.c
@@ -738,7 +738,7 @@ void bch_btree_cache_free(struct cache_set *c)
if (c->verify_data)
list_move(&c->verify_data->list, &c->btree_cache);
- free_pages((unsigned long) c->verify_ondisk, ilog2(bucket_pages(c)));
+ free_pages((unsigned long) c->verify_ondisk, ilog2(meta_bucket_pages(&c->sb)));
#endif
list_splice(&c->btree_cache_freeable,
@@ -785,7 +785,15 @@ int bch_btree_cache_alloc(struct cache_set *c)
mutex_init(&c->verify_lock);
c->verify_ondisk = (void *)
- __get_free_pages(GFP_KERNEL|__GFP_COMP, ilog2(bucket_pages(c)));
+ __get_free_pages(GFP_KERNEL|__GFP_COMP, ilog2(meta_bucket_pages(&c->sb)));
+ if (!c->verify_ondisk) {
+ /*
+ * Don't worry about the mca_rereserve buckets
+ * allocated in previous for-loop, they will be
+ * handled properly in bch_cache_set_unregister().
+ */
+ return -ENOMEM;
+ }
c->verify_data = mca_bucket_alloc(c, &ZERO_KEY, GFP_KERNEL);