linux-kernel - Re: [tip: x86/pasid] x86/asm: Carve out a generic movdir64b() helper for general usage

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] [day] [month] [year] [list]

Message-ID: <20201008064928.GR2628@hirez.programming.kicks-ass.net>
Date:   Thu, 8 Oct 2020 08:49:28 +0200
From:   Peter Zijlstra <peterz@...radead.org>
To:     Borislav Petkov <bp@...en8.de>
Cc:     linux-kernel@...r.kernel.org, linux-tip-commits@...r.kernel.org,
        Michael Matz <matz@...e.de>, Dave Jiang <dave.jiang@...el.com>,
        Borislav Petkov <bp@...e.de>, Tony Luck <tony.luck@...el.com>,
        x86 <x86@...nel.org>, hjl.tools@...il.com,
        linux-toolchains@...r.kernel.org
Subject: Re: [tip: x86/pasid] x86/asm: Carve out a generic movdir64b() helper
 for general usage

On Wed, Oct 07, 2020 at 11:13:27PM +0200, Borislav Petkov wrote:
> On Wed, Oct 07, 2020 at 07:08:35PM +0200, Peter Zijlstra wrote:
> > (%rdx), %rax, surely?
> 
> Right, later. Already tagged the branch so that Vinod can base stuff ontop.
> 
> > Also, that's a horrible convention, but I suppose (%rdx), (%rax) was
> > out?
> 
> See the end of this mail:
> 
> https://lkml.kernel.org/r/alpine.LSU.2.20.2009241356020.20802@wotan.suse.de

That, 100x that. Why wasn't it fixed then? How about we fix binutils to
accept the sane mnemonic as well?

> > Can we pretty please get a binutils version that knows about this
> > instruction, such that we know when we can get rid of the silly .byte
> > encoded mess?
> 
> It looks like support for this insn got introduced in this binutils commit:
> 
> c0a30a9f0ab4 ("Enable Intel MOVDIRI, MOVDIR64B instructions")
> 
> So I guess from 2.31 onwards:

Then we'll just keep the byte code around until we reach the min
binutils that's sane, but at least we can fix the comment to not be
insane.