[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20161018130818.GA25251@gmail.com>
Date: Tue, 18 Oct 2016 15:08:18 +0200
From: Ingo Molnar <mingo@...nel.org>
To: "Luc, Piotr" <Piotr.Luc@...el.com>
Cc: "x86@...nel.org" <x86@...nel.org>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
"peterz@...radead.org" <peterz@...radead.org>,
"torvalds@...ux-foundation.org" <torvalds@...ux-foundation.org>,
"tglx@...utronix.de" <tglx@...utronix.de>,
"jpoimboe@...hat.com" <jpoimboe@...hat.com>,
"hpa@...or.com" <hpa@...or.com>,
"brgerst@...il.com" <brgerst@...il.com>,
"mingo@...hat.com" <mingo@...hat.com>,
"luto@...nel.org" <luto@...nel.org>, "bp@...en8.de" <bp@...en8.de>,
"Hansen, Dave" <dave.hansen@...el.com>,
"dvlasenk@...hat.com" <dvlasenk@...hat.com>
Subject: Re: [v2] x86/cpufeature: Add AVX512_4VNNIW and AVX512_4FMAPS
features.
* Luc, Piotr <Piotr.Luc@...el.com> wrote:
> On Mon, 2016-10-17 at 17:03 +0200, Piotr Luc wrote:
> > v2: Initialize new bits in the scattered group. Add
>
> The commit message is obviously broken. Sorry for that.
> I will resend with fixed message.
In the v3 patchlog please also please describe to what extent new instructions are
enabled by the patch when run on real hardware (or on a simulator).
I.e. can user-space run those new instructions, while it couldn't before - or is
the patch purely for /proc/cpuinfo enumeration?
I.e. a comprehensive before/after comparison.
Thanks,
Ingo
Powered by blists - more mailing lists