linux-kernel - Re: [PATCH] x86/ACPI/cstate: Allow ACPI C1 FFH MWAIT use on AMD systems

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: <20170523075911.GA25383@nazgul.tnic>
Date:   Tue, 23 May 2017 09:59:11 +0200
From:   Borislav Petkov <bp@...en8.de>
To:     "Ghannam, Yazen" <Yazen.Ghannam@....com>
Cc:     "linux-pm@...r.kernel.org" <linux-pm@...r.kernel.org>,
        "x86@...nel.org" <x86@...nel.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        "rjw@...ysocki.net" <rjw@...ysocki.net>,
        "len.brown@...el.com" <len.brown@...el.com>,
        "pavel@....cz" <pavel@....cz>
Subject: Re: [PATCH] x86/ACPI/cstate: Allow ACPI C1 FFH MWAIT use on AMD
 systems

On Mon, May 22, 2017 at 08:20:56PM +0000, Ghannam, Yazen wrote:
> On the other hand, MWAIT on AMD limits the hardware to using only certain,
> shallower Cstates. This is okay if we define individual states and use MWAIT
> for some of them. But it would consume more power if used always.

Let me see if I understand it correctly:

Even though we used to do HLT on previous families as idling with HLT
*is* the preferred method until now, with your change you're moving
*every* AMD machine out there to do MWAIT now.

Lemme look at the F15h BKDG:

"2.5.3.2

C-state Request Interface

C-states are dynamically requested by software and are exposed through
ACPI objects (see 2.5.3.6 [ACPI Processor C-state Objects]). C-states
can be requested on a per-core basis. Software requests a C-state change
in one of two ways, either by executing the HLT instruction or by
reading from an IO address specified by MSRC001_0073[CstateAddr] plus an
offset of 0 through 7 (see D18F4x11[C:8])."

So this doesn't say anything about using MWAIT.

What's up?

-- 
Regards/Gruss,
    Boris.

ECO tip #101: Trim your mails when you reply.
--