[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20090402113234.GE10828@elte.hu>
Date: Thu, 2 Apr 2009 13:32:34 +0200
From: Ingo Molnar <mingo@...e.hu>
To: Peter Zijlstra <a.p.zijlstra@...llo.nl>
Cc: Paul Mackerras <paulus@...ba.org>,
Corey Ashford <cjashfor@...ux.vnet.ibm.com>,
linux-kernel@...r.kernel.org
Subject: Re: [PATCH 3/6] perf_counter: per event wakeups
* Peter Zijlstra <a.p.zijlstra@...llo.nl> wrote:
> By request, provide a way to request a wakeup every 'n' events instead
> of every page of output.
>
> Signed-off-by: Peter Zijlstra <a.p.zijlstra@...llo.nl>
> ---
> include/linux/perf_counter.h | 3 ++-
> kernel/perf_counter.c | 10 +++++++++-
> 2 files changed, 11 insertions(+), 2 deletions(-)
>
> Index: linux-2.6/include/linux/perf_counter.h
> ===================================================================
> --- linux-2.6.orig/include/linux/perf_counter.h
> +++ linux-2.6/include/linux/perf_counter.h
> @@ -146,7 +146,7 @@ struct perf_counter_hw_event {
> __reserved_1 : 52;
>
> __u32 extra_config_len;
> - __u32 __reserved_4;
> + __u32 wakeup_events; /* wakeup every n events */
[ Sidenote: i think we need more reserved bits :-) Expand the ABI
while it's still easy ;-) ]
I have a design observation here:
i think 'wakeup' is a user-space notification type. Another type of
notication can be 'signal', or 'none'.
Perhaps your current scheme makes the most sense: to treat them as
separate attributes, and allow signal delivery and wakeups at once.
(if user-space so wishes)
Agreed?
Ingo
--
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