RE: [PATCH] idle using PNI monitor/mwait

From: "Saxena, Sunil" <sunil.saxena@intel.com>
To: "Zwane Mwaikambo" <zwane@arm.linux.org.uk>,
	"Nakajima, Jun" <jun.nakajima@intel.com>
Cc: "Linus Torvalds" <torvalds@osdl.org>,
	<linux-kernel@vger.kernel.org>,
	"Mallick, Asit K" <asit.k.mallick@intel.com>,
	"Pallipadi, Venkatesh" <venkatesh.pallipadi@intel.com>
Subject: RE: [PATCH] idle using PNI monitor/mwait
Date: Wed, 9 Jul 2003 18:17:32 -0700	[thread overview]
Message-ID: <763EDB03559AA1428E4D41045B43F17B0304D8C4@fmsmsx406.fm.intel.com> (raw)

Thermal advantages may be there and like "pause" they would be
implementation specific.

Thanks
Sunil

-----Original Message-----
From: Zwane Mwaikambo [mailto:zwane@arm.linux.org.uk] 
Sent: Tuesday, July 08, 2003 11:42 PM
To: Nakajima, Jun
Cc: Linus Torvalds; linux-kernel@vger.kernel.org; Saxena, Sunil;
Mallick, Asit K; Pallipadi, Venkatesh
Subject: RE: [PATCH] idle using PNI monitor/mwait

On Tue, 8 Jul 2003, Nakajima, Jun wrote:

> That's right. If we have a lot of high-contention locks in the kernel,
> we need to fix the code first, to get benefits for the other
> architectures. 
> 
> "mwait" granularity (64-byte, for example) is given by the cpuid
> instruction, and we did not use it because 1) it's unlikely that the
> other fields of the task structure are modified when it's idle, 2) the
> processor needs to check the flag after mwait anyway, to avoid waking
up
> with a false signal caused by other break events (i.e. mwait is a
hint).

It could still be very handy for polling loops of the form;

while (!ready)
	__asm__ ("pause;");

Jun would there be any thermal advantages over using poll and pause ?

Thanks,
	Zwane
-- 
function.linuxpower.ca