All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Robert Richter" <robert.richter@amd.com>
To: "Stephane Eranian" <eranian@hpl.hp.com>
Cc: "Andi Kleen" <andi@firstfloor.org>, linux-kernel@vger.kernel.org
Subject: Re: [PATCH 20/22] 2.6.22-rc3 perfmon2 : new x86_64 files
Date: Thu, 7 Jun 2007 23:55:03 +0200	[thread overview]
Message-ID: <20070607215503.GC29058@erda.amd.com> (raw)
In-Reply-To: <20070531164638.GB23939@frankl.hpl.hp.com>

Stephane,

> > > +
> > > +	if (current_cpu_data.x86 != 15) {
> > > +		PFM_INFO("unsupported family=%d", current_cpu_data.x86);
> > > +		return -1;
> > > +	}
> > 
> > Already obsolete with Barcelona and Griffin (16 and 17) but very similar
> > counters. 
> > 
> Yes, but those will get new description modules.

see enclosed patch that also enables Barcelona cores. I have tested
your releases on K8 and Barcelona. 2.6.21 release
(perfmon-new-base-070509) works fine. In 2.6.22-rc3 syscalls seems to
be broken. I receive a EBADF error when trying to create a context. It
seems that sys_pfm_create_context() is never called. I could not yet
dig into this deeper.

There are also patches available for new family 10 features. I will
release these patches next week. I already implemented a generic AMD64
PMU description that detects features for different families and CPU
revisions. Enclosed patch is part of the patch series.

-Robert



Index: linux-2.6.22-rc3/arch/i386/perfmon/perfmon.c
===================================================================
--- linux-2.6.22-rc3.orig/arch/i386/perfmon/perfmon.c
+++ linux-2.6.22-rc3/arch/i386/perfmon/perfmon.c
@@ -1059,6 +1059,7 @@ char *pfm_arch_get_pmu_module_name(void)
 			goto try_arch;
 		}
 	case 15:
+	case 16:
 		/* All Opteron processors */
 		if (cpu_data->x86_vendor == X86_VENDOR_AMD)
 			return "perfmon_k8";
Index: linux-2.6.22-rc3/arch/x86_64/perfmon/perfmon_k8.c
===================================================================
--- linux-2.6.22-rc3.orig/arch/x86_64/perfmon/perfmon_k8.c
+++ linux-2.6.22-rc3/arch/x86_64/perfmon/perfmon_k8.c
@@ -307,7 +307,12 @@ static int pfm_k8_probe_pmu(void)
 		return -1;
 	}
 
-	if (current_cpu_data.x86 != 15) {
+	switch (current_cpu_data.x86) {
+	case 15:
+	case 16:
+		PFM_INFO("found family=%d", current_cpu_data.x86);
+		break;
+	default:
 		PFM_INFO("unsupported family=%d", current_cpu_data.x86);
 		return -1;
 	}

-- 
AMD Saxony, Dresden, Germany
Operating System Research Center
email: robert.richter@amd.com




      reply	other threads:[~2007-06-07 21:55 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-05-29 13:48 [PATCH 20/22] 2.6.22-rc3 perfmon2 : new x86_64 files Stephane Eranian
2007-05-31 14:38 ` Andi Kleen
2007-05-31 16:46   ` Stephane Eranian
2007-06-07 21:55     ` Robert Richter [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20070607215503.GC29058@erda.amd.com \
    --to=robert.richter@amd.com \
    --cc=andi@firstfloor.org \
    --cc=eranian@hpl.hp.com \
    --cc=linux-kernel@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.