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:   Thu, 13 May 2021 12:07:17 +0100
From:   Daniel Thompson <daniel.thompson@linaro.org>
To:     Ard Biesheuvel <ardb@kernel.org>
Cc:     Arnd Bergmann <arnd@kernel.org>,
        Russell King <linux@armlinux.org.uk>,
        Arnd Bergmann <arnd@arndb.de>,
        "# 3.4.x" <stable@vger.kernel.org>,
        Marek Vasut <marek.vasut@gmail.com>,
        Nathan Chancellor <nathan@kernel.org>,
        Nick Desaulniers <ndesaulniers@google.com>,
        Linus Walleij <linus.walleij@linaro.org>,
        Nicolas Pitre <nico@fluxnic.net>,
        Geert Uytterhoeven <geert+renesas@glider.be>,
        Mike Rapoport <rppt@kernel.org>,
        Linux ARM <linux-arm-kernel@lists.infradead.org>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        clang-built-linux <clang-built-linux@googlegroups.com>
Subject: Re: [PATCH] ARM: fix gcc-10 thumb2-kernel regression
Message-ID: <20210513110717.s2gr4l5upqzjkb5a@maple.lan>
References: <20210512081211.200025-1-arnd@kernel.org>
 <CAMj1kXECGjpxx5ouWuvnKUigzMGu=GcE8_ab2rrxt98yU1jUnw@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <CAMj1kXECGjpxx5ouWuvnKUigzMGu=GcE8_ab2rrxt98yU1jUnw@mail.gmail.com>
Precedence: bulk

On Wed, May 12, 2021 at 02:38:36PM +0200, Ard Biesheuvel wrote:
> On Wed, 12 May 2021 at 10:13, Arnd Bergmann <arnd@kernel.org> wrote:
> >
> > From: Arnd Bergmann <arnd@arndb.de>
> >
> > When building the kernel wtih gcc-10 or higher using the
> > CONFIG_CC_OPTIMIZE_FOR_PERFORMANCE=y flag, the compiler picks a slightly
> > different set of registers for the inline assembly in cpu_init() that
> > subsequently results in a corrupt kernel stack as well as remaining in
> > FIQ mode. If a banked register is used for the last argument, the wrong
> > version of that register gets loaded into CPSR_c.  When building in Arm
> > mode, the arguments are passed as immediate values and the bug cannot
> > happen.
> >
> > This got introduced when Daniel reworked the FIQ handling and was
> > technically always broken, but happened to work with both clang and gcc
> > before gcc-10 as long as they picked one of the lower registers.
> > This is probably an indication that still very few people build the
> > kernel in Thumb2 mode.
> >
> > Marek pointed out the problem on IRC, Arnd narrowed it down to this
> > inline assembly and Russell pinpointed the exact bug.
> >
> > Change the constraints to force the final mode switch to use a non-banked
> > register for the argument to ensure that the correct constant gets loaded.
> > Another alternative would be to always use registers for the constant
> > arguments to avoid the #ifdef that has now become more complex.
> >
> > Cc: <stable@vger.kernel.org> # v3.18+
> > Cc: Daniel Thompson <daniel.thompson@linaro.org>
> > Reported-by: Marek Vasut <marek.vasut@gmail.com>
> > Fixes: c0e7f7ee717e ("ARM: 8150/3: fiq: Replace default FIQ handler")
> > Signed-off-by: Arnd Bergmann <arnd@arndb.de>
> 
> Nice bug!

Indeed. Many thanks for those involved with the find and fix!


Daniel.