[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20090626072638.GA21610@linux-sh.org>
Date: Fri, 26 Jun 2009 16:26:39 +0900
From: Paul Mundt <lethal@...ux-sh.org>
To: Ingo Molnar <mingo@...e.hu>
Cc: Peter Zijlstra <peterz@...radead.org>,
Martin Schwidefsky <schwidefsky@...ibm.com>,
linux-kernel <linux-kernel@...r.kernel.org>
Subject: Re: register_timer_hook use in arch/sh/oprofile
On Fri, Jun 26, 2009 at 09:23:50AM +0200, Ingo Molnar wrote:
>
> * Peter Zijlstra <peterz@...radead.org> wrote:
>
> > On Wed, 2009-06-24 at 15:24 +0200, Ingo Molnar wrote:
> > >
> > > [ And long-term analysis ('perf stat' type of runs) dont need IRQs
> > > anyway - perfcounters reads outs the counts and summarizes them
> > > across the measured workload. ]
> >
> > If the counter width is less than 64 bits we do need to have some
> > interrupt to read them from before they cycle so we can accumulate
> > the deltas into a proper u64.
>
> Yeah - but even that can be driven from some housekeeping hrtimer.
>
> > But for proper 64 bit hardware counters there is indeed no need
> > for that.
>
> Indeed - although they are rare.
>
We have both full 64-bit counters as well as 48-bit, although they do
both at least set an overflow bit that can be looked at, even if there is
no exception associated with it. Optionally they can be split up in to
multiple 32-bit counters and half of them handed over to the bus
controller.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists