Re: [patch] x86, tsc: fix SMI induced variation in quick_pit_calibrate()

From: Suresh Siddha <suresh.b.siddha@intel.com>
To: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Ingo Molnar <mingo@elte.hu>, Thomas Gleixner <tglx@linutronix.de>,
	"H. Peter Anvin" <hpa@zytor.com>,
	linux-kernel <linux-kernel@vger.kernel.org>,
	asit.k.mallick@intel.com
Subject: Re: [patch] x86, tsc: fix SMI induced variation in quick_pit_calibrate()
Date: Mon, 16 Jan 2012 16:41:09 -0800	[thread overview]
Message-ID: <1326760869.16150.16.camel@sbsiddha-desk.sc.intel.com> (raw)
In-Reply-To: <CA+55aFys9zddHAoKURGAeu0QYD_MC11uEyOQGU_4Ybsf200NkQ@mail.gmail.com>

On Mon, 2012-01-16 at 16:18 -0800, Linus Torvalds wrote:
> So I think your patch is a total hack that just compares the two last
> TSC deltas, but it's actually close to the "correct' thing in that it
> does start taking the time to see the last of the previous MSB into
> account.

yeah, it was a hack but your alternative sounds good.

> Now the delta is twice as big as before (it covers four PIT accesses,
> roughly 4us), so the comments might have to be updated to match, but
> the rest of the code should "just work" (except it might loop a bit
> longer, and maybe it gives closer to 250 ppm precision).
> 
> Does this fix it for you? I have NOT tested it in any way.
> 

also may be we should change the correction 

from: delta += (long)(d2 - d1)/2;
to:   delta += (long)(d2 - d1)/4;

I will give this patch a try.

thanks,
suresh