On Mon, Feb 26, 2024 at 10:58:29AM +0100, Christoph Hellwig wrote: > On Mon, Feb 26, 2024 at 12:38:28AM +0800, Zorro Lang wrote: > > This patch makes sense to me, but I think this patch might cause > > failures on old kernel which doesn't has kernel commit a951104333bd, right? > > > > Is there a simple way to check if dm-error supports zoned block device? If > > there's not, I think we have to let it fails on old kernel, to make it run > > on newer kernel. > > I don't know any way. But as this just affects zoned devices on btrfs > for now I guess the usual parties can just make sure it works by > backporting the right patches. The previous don't work wasn't even > a not supported, but a case of dm-error being buggy for zoned devices. > > In other words the original exclusion in xfstests should have never > gone in and whoever put it there should have fixed dm-error instead. OK, let's have this patch, and see the response from others :) Reviewed-by: Zorro Lang <zlang@xxxxxxxxxx> >
