Re: LVM snapshot broke between 4.14 and 4.16

From: Mike Snitzer <snitzer@redhat.com>
To: Linus Torvalds <torvalds@linux-foundation.org>
Cc: wgh@torlan.ru, Theodore Ts'o <tytso@mit.edu>,
	Jens Axboe <axboe@kernel.dk>, Sagi Grimberg <sagi@grimberg.me>,
	Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
	linux-block <linux-block@vger.kernel.org>,
	dm-devel@redhat.com, Ilya Dryomov <idryomov@gmail.com>,
	Zdenek Kabelac <zkabelac@redhat.com>
Subject: Re: LVM snapshot broke between 4.14 and 4.16
Date: Sat, 4 Aug 2018 14:19:51 -0400	[thread overview]
Message-ID: <20180804181950.GA10514@redhat.com> (raw)
In-Reply-To: <CA+55aFza3_+poqmw34ev+a_jZ1Tz0rxjTu7YfnU4HX-SU52abQ@mail.gmail.com>

On Sat, Aug 04 2018 at  1:04pm -0400,
Linus Torvalds <torvalds@linux-foundation.org> wrote:

> On Sat, Aug 4, 2018 at 3:03 AM WGH <wgh@torlan.ru> wrote:
> >
> > >
> > The patch works for me.
> >
> > However, there's no text messsage in the kernel log, just a traceback. I
> > think that's because WARN_ONCE is supposed to take condition as a first
> > argument.
> 
> Duh.
> 
> It needs to be WARN_ONCE(1, ...);
> 
> I obviously didn't test that patch, but I _did_ compile it. I wonder
> why I didn't get a compiler warning for it...
> 
> [ Goes off and looks ]
> 
> Oh, because the "bio_devname(bio, b)" argument ended up being
> interpreted as the format string, and since it was a dynamic string
> the compiler felt it was all fine. Just bad luck.
> 
> Anyway, just out of curiosity, what was the traceback?
> 
> I'm not entirely happy with that patch either (even after the obvious
> fix to add the "1" argument), but it does seem like the minimal
> temporary workaround for now.

I agree.

Please feel free to add my:

Acked-by: Mike Snitzer <snitzer@redhat.com>