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;
MIME-Version: 1.0
References: <20220316183833.1563139-1-arnd@kernel.org> <3b9cedd6-a084-5d87-2fdd-0da10146827b@arm.com>
In-Reply-To: <3b9cedd6-a084-5d87-2fdd-0da10146827b@arm.com>
From:   Nick Desaulniers <ndesaulniers@google.com>
Date:   Fri, 18 Mar 2022 12:52:18 -0700
Message-ID: <CAKwvOdnut8db1Z1FdZkkSXCvmggrzfKv9Y9GO3uz4ND=K5sB3g@mail.gmail.com>
Subject: Re: [PATCH] arm64: fix clang warning about TRAMP_VALIAS
To:     Robin Murphy <robin.murphy@arm.com>
Cc:     Arnd Bergmann <arnd@kernel.org>,
        Catalin Marinas <catalin.marinas@arm.com>,
        Will Deacon <will@kernel.org>,
        Nathan Chancellor <nathan@kernel.org>,
        James Morse <james.morse@arm.com>,
        Arnd Bergmann <arnd@arndb.de>,
        "Russell King (Oracle)" <rmk+kernel@armlinux.org.uk>,
        linux-arm-kernel@lists.infradead.org, linux-kernel@vger.kernel.org,
        llvm@lists.linux.dev
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Fri, Mar 18, 2022 at 12:11 PM Robin Murphy <robin.murphy@arm.com> wrote:
>
> On 2022-03-16 18:38, Arnd Bergmann wrote:
> > From: Arnd Bergmann <arnd@arndb.de>
> >
> > The newly introduced TRAMP_VALIAS definition causes a build warning
> > with clang-14:
> >
> > arch/arm64/include/asm/vectors.h:66:31: error: arithmetic on a null pointer treated as a cast from integer to pointer is a GNU extension [-Werror,-Wnull-pointer-arithmetic]
>
> Tangentially, all the other GNU extensions we depend on throughout Linux
> are also GNU extensions. Do we know what's special about this one?

My first thought is that -Wnull-pointer-arithmetic should be in the
-Wgnu flag group, such that -Wno-gnu (as set by Kbuild) implies
-Wno-null-pointer-arithmetic.

One issue seems to be that -Wnull-pointer-arithmetic seems to control
at least 2 different checks in clang.

https://godbolt.org/z/hdYYejj3d

So it's likely these two checks should have 2 distinct flags, such
that the warning we're observing can be grouped under -Wgnu.

Filed: https://github.com/llvm/llvm-project/issues/54444

>
> Robin.
>
> >                  return (char *)TRAMP_VALIAS + SZ_2K * slot;
> >
> > Change the addition to something clang does not complain about.
> >
> > Fixes: bd09128d16fa ("arm64: Add percpu vectors for EL1")
> > Signed-off-by: Arnd Bergmann <arnd@arndb.de>
> > ---
> >   arch/arm64/include/asm/vectors.h | 4 ++--
> >   1 file changed, 2 insertions(+), 2 deletions(-)
> >
> > I see this warning on 5.17-rc8, but did not test it on linux-next,
> > which may already have a fix.
> >
> > diff --git a/arch/arm64/include/asm/vectors.h b/arch/arm64/include/asm/vectors.h
> > index f64613a96d53..bc9a2145f419 100644
> > --- a/arch/arm64/include/asm/vectors.h
> > +++ b/arch/arm64/include/asm/vectors.h
> > @@ -56,14 +56,14 @@ enum arm64_bp_harden_el1_vectors {
> >   DECLARE_PER_CPU_READ_MOSTLY(const char *, this_cpu_vector);
> >
> >   #ifndef CONFIG_UNMAP_KERNEL_AT_EL0
> > -#define TRAMP_VALIAS 0
> > +#define TRAMP_VALIAS 0ul
> >   #endif
> >
> >   static inline const char *
> >   arm64_get_bp_hardening_vector(enum arm64_bp_harden_el1_vectors slot)
> >   {
> >       if (arm64_kernel_unmapped_at_el0())
> > -             return (char *)TRAMP_VALIAS + SZ_2K * slot;
> > +             return (char *)(TRAMP_VALIAS + SZ_2K * slot);
> >
> >       WARN_ON_ONCE(slot == EL1_VECTOR_KPTI);
> >


-- 
Thanks,
~Nick Desaulniers