Re: LVM snapshot broke between 4.14 and 4.16

[Linus Torvalds]

On Fri, Aug 3, 2018 at 12:06 PM Mike Snitzer <snitzer@xxxxxxxxxx> wrote:
>
> How does that pass for a fix to this issue?
>
> That'll unilaterally mark all dm device's readonly.

Well, if it wasn't honored before anyway, then...

But yes, a much more targeted patch would be preferred.

And in fact, maybe the right thing to do is to not revert the original
commit entirely, but instead just weaken it a lot. Turn the "you did a
write request on a RO disk" into a WARN_ON_ONCE() instead of a hard
error.

Something like the attached patch.

WGH, do you build your own kernels? Does this attached (untested)
patch make things work for you? It should give a big warning, but let
the old behavior through..

                    Linus
 block/blk-core.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/block/blk-core.c b/block/blk-core.c
index f84a9b7b6f5a..95ca45bc4fc9 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -2155,11 +2155,12 @@ static inline bool bio_check_ro(struct bio *bio, struct hd_struct *part)
 	if (part->policy && op_is_write(bio_op(bio))) {
 		char b[BDEVNAME_SIZE];
 
-		printk(KERN_ERR
+		WARN_ONCE(
 		       "generic_make_request: Trying to write "
 			"to read-only block-device %s (partno %d)\n",
 			bio_devname(bio, b), part->partno);
-		return true;
+		/* Older lvm-tools actually trigger this */
+		return false;
 	}
 
 	return false;