Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Sender: Tejun Heo <htejun@gmail.com>
Date:   Tue, 1 Nov 2022 06:46:35 -1000
From:   Tejun Heo <tj@kernel.org>
To:     Jiri Slaby <jirislaby@kernel.org>
Cc:     Christoph Hellwig <hch@infradead.org>,
        linux-kernel@vger.kernel.org, Martin Liska <mliska@suse.cz>,
        Josef Bacik <josef@toxicpanda.com>,
        Jens Axboe <axboe@kernel.dk>, cgroups@vger.kernel.org,
        linux-block@vger.kernel.org
Subject: Re: [PATCH] block/blk-iocost (gcc13): cast enum members to int in
 prints
Message-ID: <Y2FNa4bGhJoevRKT@slm.duckdns.org>
References: <20221031114520.10518-1-jirislaby@kernel.org>
 <Y1++fLJXkeZgtXR2@infradead.org>
 <Y2AMcSPAJpj6obSA@slm.duckdns.org>
 <d833ad15-f458-d43d-cab7-de62ff54a939@kernel.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <d833ad15-f458-d43d-cab7-de62ff54a939@kernel.org>
Precedence: bulk

Hello,

On Tue, Nov 01, 2022 at 06:46:56AM +0100, Jiri Slaby wrote:
> Yes. The real problem is that using anything else then an INT_MIN <= x <=
> INT_MAX _constant_ in an enum is undefined in ANSI C < 2x (in particular, 1
> << x is undefined too). gcc manual defines unsigned int on the top of that
> as defined too (so this holds for our -std=g*).
> 
> > I suppose the most reasonable thing to do here is just splitting them into
> > separate enum definitions. Does anyone know how this behavior change came to
> > be?
> 
> C2x which introduces un/signed long enums. See the bug I linked in the
> commit log:
> https://gcc.gnu.org/bugzilla/show_bug.cgi?id=36113

I see. So, it was an extension but the new standard is defined differently
and we're gonna end up with that behavior.

> The change is also turned on in < C2x on purpose. AIUI, unless there is too
> much breakage. So we'd need to sort it out in (rather distant) future anyway
> (when we come up to -std=g2x).

The part that the new behavior applying to <C2x feels like an odd decision.
I'm having a hard time seeing the upsides in doing so but maybe that's just
me not knowing the area well enough.

> > Do we know whether clang is gonna be changed the same way?
> 
> In C2x, Likely. In < C2x, dunno what'd be the default.

It looks like we can do one of the following two:

* If gcc actually changes the behavior for <c2x, split the enums according
  to their sizes. This feels rather silly but I can't think of a better way
  to cater to divergent compiler behaviors.

* If gcc doesn't change the behavior for <c2x, there's nothing to do for the
  time being. Later when we switch to -std=g2x, we can just change the
  format strings to use the now larger types.

Does the above make sense?

Thanks.

-- 
tejun