Re: [patch] oprofile for ppc

From: Albert Cahalan <albert@users.sf.net>
To: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: "Albert D. Cahalan" <acahalan@cs.uml.edu>,
	oprofile-list@lists.sourceforge.net,
	linuxppc-dev@lists.linuxppc.org,
	Segher Boessenkool <segher@koffie.nl>,
	o.oppitz@web.de, afleming@motorola.com,
	linux-kernel@vger.kernel.org
Subject: Re: [patch] oprofile for ppc
Date: 07 Mar 2003 13:31:01 -0500	[thread overview]
Message-ID: <1047061862.1900.67.camel@cube> (raw)
In-Reply-To: <1047032003.12206.5.camel@zion.wanadoo.fr>

On Fri, 2003-03-07 at 05:13, Benjamin Herrenschmidt wrote:
> On Fri, 2003-03-07 at 10:29, Albert D. Cahalan wrote:

>> This is basic timer profiling for ppc, tested on the
>> 2.5.62 linuxppc kernel. It's a port of the ppc64 code.
>
> I'm sure I missed something... but I fail to see the the
> interest in profiling based on sampling the instruction ptr
> on a 100 Hz basis. This is way to slow to give any useful
> results imho

This is just the first part of the code. Please merge it
into any tree you have, unless it's obviously broken.
It is useful for long-running processes that don't do
much that is tied to the clock tick. (number crunching,
maybe X, web browsers without animations, /tmp cleaner...)

The i386 port is already using 1000 Hz in the kernel,
and has 100 Hz as a non-default option. I'd really like
to have this on my Mac; lots of things would improve.

I intend to allow sampling based on the performance counter
interrupt/trap/exception and the external interrupt signal.