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:   Wed, 22 Apr 2020 18:53:39 +0200
From:   Borislav Petkov <bp@alien8.de>
To:     Martin =?utf-8?B?TGnFoWth?= <mliska@suse.cz>
Cc:     Jakub Jelinek <jakub@redhat.com>,
        Peter Zijlstra <peterz@infradead.org>,
        Nick Desaulniers <ndesaulniers@google.com>,
        Michael Matz <matz@suse.de>,
        Sergei Trofimovich <slyfox@gentoo.org>,
        LKML <linux-kernel@vger.kernel.org>,
        Thomas Gleixner <tglx@linutronix.de>,
        Ingo Molnar <mingo@redhat.com>,
        "H. Peter Anvin" <hpa@zytor.com>,
        Andy Lutomirski <luto@kernel.org>,
        "maintainer:X86 ARCHITECTURE (32-BIT AND 64-BIT)" <x86@kernel.org>,
        clang-built-linux <clang-built-linux@googlegroups.com>
Subject: Re: [PATCH v2] x86: fix early boot crash on gcc-10
Message-ID: <20200422165339.GE26846@zn.tnic>
References: <CAKwvOdnFXPBJsAUD++HtYS5JiR2KmX73M5GAUe-tvX-JYV7DaA@mail.gmail.com>
 <CAKwvOdmNwNwa6rMC27-QZq8VDrYdTQeQqss-bAwF1EMmnAHxdw@mail.gmail.com>
 <20200417190607.GY2424@tucnak>
 <CAKwvOdkkbWgWmNthq5KijCdtatM9PEAaCknaq8US9w4qaDuwug@mail.gmail.com>
 <alpine.LSU.2.21.2004201401120.11688@wotan.suse.de>
 <20200422102309.GA26846@zn.tnic>
 <20200422114007.GC20730@hirez.programming.kicks-ass.net>
 <20200422134924.GB26846@zn.tnic>
 <20200422135531.GM2424@tucnak>
 <20a91f2e-0f25-8dba-e441-3233cc1ef398@suse.cz>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
In-Reply-To: <20a91f2e-0f25-8dba-e441-3233cc1ef398@suse.cz>
User-Agent: Mutt/1.10.1 (2018-07-13)
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Wed, Apr 22, 2020 at 04:16:53PM +0200, Martin Liška wrote:
> And as I talked to Boris, I would recommend to come up with a "configure" check
> that a compiler does not optimize the key code sequence:
> 
> $ cat asm-detect.c
> int foo(int a);
> int bar(int a)
> {
>   int r = foo(a);
>   asm ("");
>   return r;
> }
> 
> $ gcc -O2 -c asm-detect.c -S -o/dev/stdout | grep jmp
> [no output]

That is a good test to run at the beginning of the compilation I guess.

Without the asm("") it produces:

bar:
.LFB0:
	.cfi_startproc
	jmp	foo@PLT
	.cfi_endproc

I'd like for LLVM folks to confirm that this is a good test for LLVM too
Trying that here with clang gives:

bar:                                    # @bar
        .cfi_startproc
# %bb.0:
        jmp     foo                     # TAILCALL
.Lfunc_end0:

so this *looks* like it would work with LLVM too but I might be missing
something...

Thx.

-- 
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette