MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Message-ID: <18890.59107.116394.771297@drongo.ozlabs.ibm.com>
Date: Thu, 26 Mar 2009 13:22:27 +1100
From: Paul Mackerras <paulus@samba.org>
To: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Ingo Molnar <mingo@elte.hu>, linux-kernel@vger.kernel.org,
       Mike Galbraith <efault@gmx.de>, Arjan van de Ven <arjan@infradead.org>,
       Wu Fengguang <fengguang.wu@intel.com>
Subject: Re: [PATCH 5/6] perf_counter: kerneltop: mmap_pages argument
In-Reply-To: <1237984033.7972.865.camel@twins>
References: <20090325113021.781490788@chello.nl>
	<20090325113317.104545398@chello.nl>
	<20090325121811.GC11571@elte.hu>
	<1237984033.7972.865.camel@twins>
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 948
Lines: 25

Peter Zijlstra writes:

> One possible solution is making the control page writable and writing
> the userspace read position to it, then the kernel can, on
> perf_output_begin() detect the overflow and count the number of
> overwritten events.
> 
> This overflow count could then be published back into the control page.

We could in principle have many different processes mmapping the same
counter and reading the ring buffer, couldn't we?  So which process
gets to put its read position in the control page?

> TBH I'm not much of a fan, making all these pages writable just opens a
> whole can of worms, and that accurate overflow tracking will put more
> code in the output path.

I agree.

Paul.
--
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/