On Mon, Jul 13, 2020 at 11:18:36AM +0100, Christoph Hellwig wrote:

On Fri, Jul 10, 2020 at 10:34:19PM -0700, syzbot wrote:

quoted

Hello,

syzbot found the following crash on:

This is not a crash, but a WARN_ONCE.  A pre-existing one that just
slightly changed the printed message recently.

It doesn't really matter.  WARN is for indicating kernel bugs only.
A user-triggable WARN is a bug.  Either the bug that makes the WARN
reachable needs to be fixed, or if the WARN is legitimately user-reachable
it needs to be removed or replaced with a proper ratelimited log message.

This one looks legitimately user-reachable, so we could do:

diff --git a/block/blk-core.c b/block/blk-core.c
index d9d632639bd1..354c51ad5c81 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c

@@ -854,8 +854,8 @@ static inline bool bio_check_ro(struct bio *bio, struct hd_struct *part)
 		if (op_is_flush(bio->bi_opf) && !bio_sectors(bio))
 			return false;
 
-		WARN_ONCE(1,
-		       "Trying to write to read-only block-device %s (partno %d)\n",
+		pr_warn_ratelimited(
+		       "block: trying to write to read-only block-device %s (partno %d)\n",
 			bio_devname(bio, b), part->partno);
 		/* Older lvm-tools actually trigger this */
 		return false;

We could also show current->comm and current->pid if they would be useful here.

And yes, this is preexisting which is why syzbot has reported this before
(https://syzkaller.appspot.com/bug?id=79eda145ab047a0dc7d03ca5fcb1cf12206eb481).
Just no one has bothered to fix it yet.

- Eric