Date: Tue, 11 Oct 2005 21:53:01 +0200
From: Ingo Molnar <mingo@elte.hu>
To: Lee Revell <rlrevell@joe-job.com>
Cc: linux-kernel <linux-kernel@vger.kernel.org>
Subject: Re: 2.6.13-rt12: irqs hard off for 657 usecs
Message-ID: <20051011195301.GA29042@elte.hu>
References: <1128724690.17981.57.camel@mindpipe> <20051011112043.GB15995@elte.hu> <1129048106.12702.21.camel@mindpipe> <1129056268.4718.1.camel@mindpipe>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <1129056268.4718.1.camel@mindpipe>
User-Agent: Mutt/1.4.2.1i
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 998
Lines: 23


* Lee Revell <rlrevell@joe-job.com> wrote:

> > The max latency is close to the period of the timer IRQ.  So either it's
> > an instrumentation bug or something really disables IRQs for up to 1/HZ
> > seconds.
> 
> I can no longer reproduce this with 2.6.14-rc4-rt1.  It seems to have 
> been a real IRQs disabled bug (not an instrumentation bug) as I was 
> also getting xruns with 2.6.13-rt12.

ok, please re-report if it occurs again. I havent fixed any preemption 
bug since -rt12, but maybe as a side-effect of other things? Some false 
positive latencies were fixed in the latency tracer, but since you also 
got xruns, they must have been real latencies. (albeit saving a long 
false-positive latency trace can cause a latency by itself!)

	Ingo
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/