linux-kernel - Re: [PATCH 2/6] x86: bug: prevent gcc distortions

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <A0A24A0C-7ED5-40DA-A017-7FE292F57939@vmware.com>
Date:   Fri, 18 May 2018 14:36:21 +0000
From:   Nadav Amit <namit@...are.com>
To:     Borislav Petkov <bp@...en8.de>
CC:     Ingo Molnar <mingo@...nel.org>,
        Peter Zijlstra <peterz@...radead.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        "x86@...nel.org" <x86@...nel.org>,
        Thomas Gleixner <tglx@...utronix.de>,
        Ingo Molnar <mingo@...hat.com>,
        "H. Peter Anvin" <hpa@...or.com>,
        Josh Poimboeuf <jpoimboe@...hat.com>,
        Linus Torvalds <torvalds@...ux-foundation.org>,
        Andy Lutomirski <luto@...nel.org>,
        Peter Zijlstra <a.p.zijlstra@...llo.nl>,
        Denys Vlasenko <dvlasenk@...hat.com>
Subject: Re: [PATCH 2/6] x86: bug: prevent gcc distortions

Borislav Petkov <bp@...en8.de> wrote:

> On Fri, May 18, 2018 at 10:13:54AM +0200, Ingo Molnar wrote:
>> Yes, that's my main worry too about all these inlining changes:
>> the very, very marked reduction in the readability of assembly code.
> 
> Same reaction here: the small improvements this brings is simply not
> enough to sacrifice readability so much IMO.

I didn’t try too hard to find more affected (micro)benchmarks, but I am
pretty sure there are: pretty much all the paravirt ops are currently not
inlined, which can affect kernel operations that bang on the page-tables.

Notice that this is not an x86-specific issue. Other architectures, with
not-as-good OOOE for instance, might incur higher overheads. So I don’t
think that this issue should be ignored.