MIME-Version: 1.0
In-Reply-To: <20161203150258.vwr5zzco7ctgc4pe@pd.tnic>
References: <0a21157c2233ba7d0781bbf07866b3f2d7e7c25d.1480638597.git.luto@kernel.org>
 <CA+55aFw3P++jGbRhkrM3MyqwZZ3mxmxQTC1BfYS2rhw6P3LKNA@mail.gmail.com>
 <CALCETrXCLLW9pd=h7dZ0u2oBSiRBJOBkxvEq2iz=Z8aTVtrZ=w@mail.gmail.com>
 <20161202180343.gehqor7lgtmzwqq3@pd.tnic> <CA+55aFyHptQ3=gjXA_LGGkvOgsRB9FBXvCT56fG4UqcFV4wFsA@mail.gmail.com>
 <20161202185008.tdziqrzi4a3axord@pd.tnic> <CA+55aFxuhzE0woFyRjZ8=Ji1EPR1+MohrbDm=2AiQH50dsptjg@mail.gmail.com>
 <20161202192050.l5l3rcwems6hptub@pd.tnic> <CA+55aFydg4RDv8a-Wpf=TyzPN59kONcUdCkk1KcSNMW_2Uix_g@mail.gmail.com>
 <20161202192844.elq2py3cxizjs2bx@pd.tnic> <20161203150258.vwr5zzco7ctgc4pe@pd.tnic>
From: Andy Lutomirski <luto@amacapital.net>
Date: Sat, 3 Dec 2016 09:05:39 -0800
Message-ID: <CALCETrXzdAUVV76KDJUKyk-3CMFoUteMYCHTjB+mXZeFzRoGog@mail.gmail.com>
Subject: Re: [PATCH] x86/alternatives: Do not use sync_core() to serialize I$
To: Borislav Petkov <bp@alien8.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>,
        Andy Lutomirski <luto@kernel.org>, Peter Anvin <hpa@zytor.com>,
        "the arch/x86 maintainers" <x86@kernel.org>,
        One Thousand Gnomes <gnomes@lxorguk.ukuu.org.uk>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        Brian Gerst <brgerst@gmail.com>,
        Matthew Whitehead <tedheadster@gmail.com>,
        Henrique de Moraes Holschuh <hmh@hmh.eng.br>,
        Peter Zijlstra <peterz@infradead.org>,
        Andrew Cooper <andrew.cooper3@citrix.com>
Content-Type: text/plain; charset=UTF-8
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1279
Lines: 33

On Sat, Dec 3, 2016 at 7:02 AM, Borislav Petkov <bp@alien8.de> wrote:
> On Fri, Dec 02, 2016 at 08:28:44PM +0100, Borislav Petkov wrote:
>> Ah, it is called only from apply_alternatives() but sure, it is safer
>> this way. Lemme do that and run it through the boxes to see whether
>> anything catches fire.
>
> Looks good, ran it on a bunch of machines, two of them huge AMD and
> Intel, 8-socket monsters.
>
> Here is a version with a commit message:
>
> ---
> From: Borislav Petkov <bp@suse.de>
> Date: Fri, 2 Dec 2016 23:18:51 +0100
> Subject: [PATCH] x86/alternatives: Do not use sync_core() to serialize I$
>
> We use sync_core() in the alternatives code to stop speculative
> execution of prefetched instructions because we are potentially changing
> them and don't want to execute stale bytes.
>
> What it does on most machines is call CPUID which is a serializing
> instruction. And that's expensive.
>
> However, the instruction cache is serialized when we're on the local CPU
> and are changing the data through the same virtual address. So then, we
> don't need the serializing CPUID but a simple control flow change. Last
> being accomplished with a CALL/RET which the noinline causes.
>
> So let's try it.

I like it.

Reviewed-by: Andy Lutomirski <luto@kernel.org>