[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20120309114112.GR16322@erda.amd.com>
Date: Fri, 9 Mar 2012 12:41:12 +0100
From: Robert Richter <robert.richter@....com>
To: Ingo Molnar <mingo@...e.hu>
CC: Peter Zijlstra <peterz@...radead.org>,
Arnaldo Carvalho de Melo <acme@...hat.com>,
Stephane Eranian <eranian@...gle.com>,
LKML <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH 4/4] perf script: Add script to collect and display IBS
samples
Ingo,
On 08.03.12 13:19:24, Ingo Molnar wrote:
> I've applied the following 4 kernel patches to perf/x86-ibs:
>
> db98c5faf8cb: perf/x86: Implement 64-bit counter support for IBS
> 4db2e8e6500d: perf/x86: Implement IBS pmu control ops
> b7074f1fbd61: perf/x86: Implement IBS interrupt handler
> 510419435c69: perf/x86: Implement IBS event configuration
>
> and pushed it out.
>
> We can apply any tooling changes to this branch as well.
>
> I suspect being able to skid-less profile on AMD CPUs, using
> IBS, would be a useful feature to users. What would be required
> for that to work?
thanks for putting this into tip. This helps a lot.
I am just looking into skid-less sampling. The effort of attaching the
rip taken from the IBS sample for events 0x76/0xc1 should be feasable.
In between I started the implementation of ibs pseudo events as
described here:
Appendix F Guide to Instruction-Based Sampling
http://support.amd.com/us/Processor_TechDocs/47414_15h_sw_opt_guide.pdf
This includes kernel side filtering and ibs support for the perf tool.
-Robert
--
Advanced Micro Devices, Inc.
Operating System Research Center
--
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