From owner-freebsd-hackers@freebsd.org  Fri May  5 11:16:12 2017
Return-Path: <owner-freebsd-hackers@freebsd.org>
Delivered-To: freebsd-hackers@mailman.ysv.freebsd.org
Received: from mx1.freebsd.org (mx1.freebsd.org
 [IPv6:2001:1900:2254:206a::19:1])
 by mailman.ysv.freebsd.org (Postfix) with ESMTP id 14B8AD5FEE7
 for <freebsd-hackers@mailman.ysv.freebsd.org>;
 Fri,  5 May 2017 11:16:12 +0000 (UTC) (envelope-from zec@fer.hr)
Received: from mailman.ysv.freebsd.org (unknown [127.0.1.3])
 by mx1.freebsd.org (Postfix) with ESMTP id 057F01FBD
 for <freebsd-hackers@freebsd.org>; Fri,  5 May 2017 11:16:12 +0000 (UTC)
 (envelope-from zec@fer.hr)
Received: by mailman.ysv.freebsd.org (Postfix)
 id 01DDFD5FEE6; Fri,  5 May 2017 11:16:12 +0000 (UTC)
Delivered-To: hackers@mailman.ysv.freebsd.org
Received: from mx1.freebsd.org (mx1.freebsd.org
 [IPv6:2001:1900:2254:206a::19:1])
 by mailman.ysv.freebsd.org (Postfix) with ESMTP id 0181BD5FEE5
 for <hackers@mailman.ysv.freebsd.org>; Fri,  5 May 2017 11:16:12 +0000 (UTC)
 (envelope-from zec@fer.hr)
Received: from mail.fer.hr (mail.fer.hr [161.53.72.233])
 (using TLSv1.2 with cipher ECDHE-RSA-AES256-SHA384 (256/256 bits))
 (Client CN "mail.fer.hr", Issuer "TERENA SSL CA 3" (verified OK))
 by mx1.freebsd.org (Postfix) with ESMTPS id 9101D1FBC
 for <hackers@freebsd.org>; Fri,  5 May 2017 11:16:11 +0000 (UTC)
 (envelope-from zec@fer.hr)
Received: from x23 (161.53.63.210) by MAIL.fer.hr (161.53.72.233) with
 Microsoft SMTP Server (TLS) id 14.3.319.2; Fri, 5 May 2017 13:16:01 +0200
Date: Fri, 5 May 2017 13:16:38 +0200
From: Marko Zec <zec@fer.hr>
To: <hackers@freebsd.org>
Subject: AMD CPU cache monitoring
Message-ID: <20170505131638.4835dfce@x23>
X-Mailer: Claws Mail 3.14.1 (GTK+ 2.24.29; amd64-portbld-freebsd11.0)
MIME-Version: 1.0
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit
X-Originating-IP: [161.53.63.210]
X-BeenThere: freebsd-hackers@freebsd.org
X-Mailman-Version: 2.1.23
Precedence: list
List-Id: Technical Discussions relating to FreeBSD
 <freebsd-hackers.freebsd.org>
List-Unsubscribe: <https://lists.freebsd.org/mailman/options/freebsd-hackers>, 
 <mailto:freebsd-hackers-request@freebsd.org?subject=unsubscribe>
List-Archive: <http://lists.freebsd.org/pipermail/freebsd-hackers/>
List-Post: <mailto:freebsd-hackers@freebsd.org>
List-Help: <mailto:freebsd-hackers-request@freebsd.org?subject=help>
List-Subscribe: <https://lists.freebsd.org/mailman/listinfo/freebsd-hackers>, 
 <mailto:freebsd-hackers-request@freebsd.org?subject=subscribe>
X-List-Received-Date: Fri, 05 May 2017 11:16:12 -0000

I'm looking for a simple, vendor-neutral tool for monitoring
per-CPU-core data cache hits and misses (L2, L3).

Adrian mentioned https://github.com/opcm/pcm in his article on
profiling:
http://adrianchadd.blogspot.hr/2013/08/profiling-on-superscalar-architectures.html

This tool works fine on FBSD 11 and Intel silicon, and would be almost
ideal for my needs, but unfortunately it doesn't work with AMD CPUs.

OTOH, pmc(3) seems to be a more generic mechanism for working with
performance counters, but I couldn't find useful examples on how to
harvest per-cpu counters.  Most of the pmc-related examples I bumped
itno seem to be focused on call stack profiling, which isn't what I'm
looking for.

Any ideas on how to monitor CPU cache hits / misses on AMD CPUs?

Thanks,

Marko