Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Sat, 1 May 2021 23:03:37 +0200
From:   Oleksandr Natalenko <oleksandr@natalenko.name>
To:     Linus Torvalds <torvalds@linux-foundation.org>
Cc:     Vineet Gupta <Vineet.Gupta1@synopsys.com>,
        Arnd Bergmann <arnd@kernel.org>,
        Peter Zijlstra <peterz@infradead.org>,
        Masahiro Yamada <masahiroy@kernel.org>,
        Jann Horn <jannh@google.com>,
        lkml <linux-kernel@vger.kernel.org>,
        arcml <linux-snps-arc@lists.infradead.org>,
        "linux-arch@vger.kernel.org" <linux-arch@vger.kernel.org>
Subject: Re: Heads up: gcc miscompiling initramfs zlib decompression code at
 -O3
Message-ID: <20210501210337.66qxgomc643lbdyi@spock.localdomain>
References: <75d07691-1e4f-741f-9852-38c0b4f520bc@synopsys.com>
 <CAHk-=wjJEgjCYzHZFPxTs01p7FMEHKKqXyqwRVBk6KnvHB1qVA@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <CAHk-=wjJEgjCYzHZFPxTs01p7FMEHKKqXyqwRVBk6KnvHB1qVA@mail.gmail.com>
Precedence: bulk

Hello.

On Fri, Apr 30, 2021 at 03:06:30PM -0700, Linus Torvalds wrote:
> On Fri, Apr 30, 2021 at 1:46 PM Vineet Gupta <Vineet.Gupta1@synopsys.com> wrote:
> >
> > I've hit a mainline gcc 10.2 (also gcc 9.3) bug which triggers at -O3
> > causing wrong codegen.
> 
> I'd be more than happy to just disable CC_OPTIMIZE_FOR_PERFORMANCE_O3 entirely.

FWIW, we used to find real bugs using -O3 in the past [1].

> The advantages are very questionable - with a lot of the optimizations
> at O3 being about loops, something which the kernel to a close
> approximation doesn't have.
> 
> Most kernel loops are "count on one hand" iterations, and loop
> optimizations generally just make things worse.
> 
> And we've had problems with -O3 before, because not only are the
> optimizations a bit esoteric, they are often relatively untested. If
> you look around at various projects (outside the kernel), -O2 is
> generally the "default".
> 
> And that's entirely ignoring the gcc history - where -O3 has often
> been very buggy indeed. It's gotten much better, but I just don't see
> the upside of using -O3.
> 
> In fact, it looks like we already have that
> 
>         depends on ARC
> 
> for -O3, exactly because nobody really wants to use this.
> 
> So this bug seems to be entirely ARC-specific, in that only ARC can
> use -O3 for the kernel already.
> 
>              Linus

[1] https://lore.kernel.org/lkml/673b885183fb64f1cbb3ed2387524077@natalenko.name/

-- 
  Oleksandr Natalenko (post-factum)