Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Thu, 30 Jan 2020 18:39:10 +0100
From:   Borislav Petkov <bp@suse.de>
To:     Linus Torvalds <torvalds@linux-foundation.org>
Cc:     "Luck, Tony" <tony.luck@intel.com>, Ingo Molnar <mingo@kernel.org>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        Thomas Gleixner <tglx@linutronix.de>,
        Borislav Petkov <bp@alien8.de>,
        Peter Zijlstra <a.p.zijlstra@chello.nl>,
        Andrew Morton <akpm@linux-foundation.org>
Subject: Re: [GIT PULL] x86/asm changes for v5.6
Message-ID: <20200130173910.GK6684@zn.tnic>
References: <20200128165906.GA67781@gmail.com>
 <CAHk-=wgm+2ac4nnprPST6CnehHXScth=A7-ayrNyhydNC+xG-g@mail.gmail.com>
 <CAHk-=wi=otQxzhLAofWEvULLMk2X3G3zcWfUWz7e1CFz+xYs2Q@mail.gmail.com>
 <20200129132618.GA30979@zn.tnic>
 <20200129170725.GA21265@agluck-desk2.amr.corp.intel.com>
 <CAHk-=wgns2Tvph77XZWN=r_qAtUwxrTzDXNffi8nGKz1mLZNHw@mail.gmail.com>
 <20200129183404.GB30979@zn.tnic>
 <CAHk-=wh62anGKKEeey8ubD+-+3qSv059z7zSWZ4J=CoaOo4j_A@mail.gmail.com>
 <20200130085134.GB6684@zn.tnic>
 <CAHk-=wje_k92K6j0-=HH4F5Jmr8Fv7vB-ANObqbQeGS_RsikWA@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
In-Reply-To: <CAHk-=wje_k92K6j0-=HH4F5Jmr8Fv7vB-ANObqbQeGS_RsikWA@mail.gmail.com>
User-Agent: Mutt/1.10.1 (2018-07-13)
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Thu, Jan 30, 2020 at 07:27:28AM -0800, Linus Torvalds wrote:
> Fair enough. Let's not complicate things just to avoid a few nops.

Yeah, judging by the frequency this keeps popping up, we might end up
doing proper insn parsing for the alternatives soon. :)

> That does make me wonder about RIP-relative addressing in alternatives
> too. Particularly anything where we let gcc pick addressing modes. I
> guess we don't have any, but maybe this branch issue and possible RIP
> addressing is something that objtool could be taught to warn about?

Yeah, makes sense. It would help if one slaps a relative JMP as *not*
the first insn in an alternatives replacement and the build to warn that
it can't work. Lemme go stare at objtool.

-- 
Regards/Gruss,
    Boris.

SUSE Software Solutions Germany GmbH, GF: Felix Imendörffer, HRB 36809, AG Nürnberg