block: do not leak memory in bio_copy_user_iov()
authorJérôme Glisse <jglisse@redhat.com>
Wed, 10 Apr 2019 20:27:51 +0000 (16:27 -0400)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Wed, 17 Apr 2019 06:36:46 +0000 (08:36 +0200)
commit a3761c3c91209b58b6f33bf69dd8bb8ec0c9d925 upstream.

When bio_add_pc_page() fails in bio_copy_user_iov() we should free
the page we just allocated otherwise we are leaking it.

Cc: linux-block@vger.kernel.org
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: stable@vger.kernel.org
Reviewed-by: Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com>
Signed-off-by: Jérôme Glisse <jglisse@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
block/bio.c

index 68972e3..4c18a68 100644 (file)
@@ -1214,8 +1214,11 @@ struct bio *bio_copy_user_iov(struct request_queue *q,
                        }
                }
 
-               if (bio_add_pc_page(q, bio, page, bytes, offset) < bytes)
+               if (bio_add_pc_page(q, bio, page, bytes, offset) < bytes) {
+                       if (!map_data)
+                               __free_page(page);
                        break;
+               }
 
                len -= bytes;
                offset = 0;