All of lore.kernel.org
 help / color / mirror / Atom feed
From: Peter Zijlstra <peterz@infradead.org>
To: Don Zickus <dzickus@redhat.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>,
	Vivek Goyal <vgoyal@redhat.com>, Yinghai Lu <yinghai@kernel.org>,
	Ingo Molnar <mingo@elte.hu>,
	Jason Wessel <jason.wessel@windriver.com>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	Haren Myneni <hbabu@us.ibm.com>
Subject: Re: perf hw  in kexeced kernel broken in tip
Date: Wed, 08 Dec 2010 15:59:16 +0100	[thread overview]
Message-ID: <1291820356.28378.83.camel@laptop> (raw)
In-Reply-To: <1291818005.28378.38.camel@laptop>

On Wed, 2010-12-08 at 15:20 +0100, Peter Zijlstra wrote:

> > I wonder if you should reverse these checks.  If the bios has the perf
> > counter enabled, there might be a high chance that it fails the first
> > check and never gets to the actually bios checks.
> 
> Ah, good point. 

Something like so..

---
Subject: perf, x86: Detect broken BIOSes
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Wed Dec 08 15:56:23 CET 2010

Some BIOSes use PMU resources, this is a bug.

Try to detect this, warn about it, and further refuse to touch the
PMU ourselves.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
---
Index: linux-2.6/arch/x86/kernel/cpu/perf_event.c
===================================================================
--- linux-2.6.orig/arch/x86/kernel/cpu/perf_event.c
+++ linux-2.6/arch/x86/kernel/cpu/perf_event.c
@@ -375,15 +375,51 @@ static void release_pmc_hardware(void) {
 static bool check_hw_exists(void)
 {
 	u64 val, val_new = 0;
-	int ret = 0;
+	int i, reg, ret = 0;
 
+	/*
+	 * Check to see if the BIOS enabled any of the counters, if so
+	 * complain and bail.
+	 */
+	for (i = 0; i < x86_pmu.num_counters; i++) {
+		reg = x86_pmu.eventsel + i;
+		ret = rdmsrl_safe(reg, &val);
+		if (ret)
+			goto msr_fail;
+		if (val & ARCH_PERFMON_EVENTSEL_ENABLE)
+			goto bios_fail;
+	}
+
+	for (i = 0; i < x86_pmu.num_counters_fixed; i++) {
+		reg = MSR_ARCH_PERFMON_FIXED_CTR_CTRL;
+		ret = rdmsrl_safe(reg, &val);
+		if (ret)
+			goto msr_fail;
+		if (val & (0x03 << i*4))
+			goto bios_fail;
+	}
+
+	/*
+	 * Now write a value and read it back to see if it matches,
+	 * this is needed to detect certain hardware emulators (qemu/kvm)
+	 * that don't trap on the MSR access and always return 0s.
+	 */
 	val = 0xabcdUL;
-	ret |= checking_wrmsrl(x86_pmu.perfctr, val);
+	ret = checking_wrmsrl(x86_pmu.perfctr, val);
 	ret |= rdmsrl_safe(x86_pmu.perfctr, &val_new);
 	if (ret || val != val_new)
-		return false;
+		goto msr_fail;
 
 	return true;
+
+bios_fail:
+	printk(KERN_CONT "Broken BIOS detected, software events only.\n");
+	printk(KERN_ERR FW_BUG "invalid MSR: %x=%Lx\n", reg, val);
+	return false;
+
+msr_fail:
+	printk(KERN_CONT "Broken PMU hardware detected, software events only.\n");
+	return false;
 }
 
 static void reserve_ds_buffers(void);
@@ -1378,10 +1414,8 @@ int __init init_hw_perf_events(void)
 	pmu_check_apic();
 
 	/* sanity check that the hardware exists or is emulated */
-	if (!check_hw_exists()) {
-		pr_cont("Broken PMU hardware detected, software events only.\n");
+	if (!check_hw_exists())
 		return 0;
-	}
 
 	pr_cont("%s PMU driver.\n", x86_pmu.name);
 


  parent reply	other threads:[~2010-12-08 14:59 UTC|newest]

Thread overview: 42+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-12-01  8:00 perf hw in kexeced kernel broken in tip Yinghai Lu
2010-12-01 11:27 ` Peter Zijlstra
2010-12-01 16:06   ` Vivek Goyal
2010-12-01 16:11     ` Peter Zijlstra
2010-12-01 16:23       ` Vivek Goyal
2010-12-01 19:38         ` Peter Zijlstra
2010-12-01 19:46           ` Vivek Goyal
2010-12-01 19:49             ` Peter Zijlstra
2010-12-01 19:58               ` Vivek Goyal
2010-12-01 20:07                 ` Peter Zijlstra
2010-12-01 21:48                   ` Eric W. Biederman
2010-12-02  5:23                     ` Don Zickus
2010-12-02  7:34                       ` Peter Zijlstra
2010-12-02 16:15                         ` Don Zickus
2010-12-07 23:30                           ` Peter Zijlstra
2010-12-08 14:01                             ` Don Zickus
2010-12-08 14:20                               ` Peter Zijlstra
2010-12-08 14:42                                 ` Vivek Goyal
2010-12-08 14:48                                   ` Peter Zijlstra
2010-12-08 15:02                                     ` Vivek Goyal
2010-12-08 15:15                                       ` Peter Zijlstra
2010-12-08 15:22                                         ` Vivek Goyal
2010-12-08 21:16                                         ` Eric W. Biederman
2010-12-08 14:59                                 ` Peter Zijlstra [this message]
2010-12-08 18:43                                   ` Yinghai Lu
2010-12-08 19:01                                     ` Don Zickus
2010-12-08 19:05                                       ` Yinghai Lu
2010-12-08 19:17                                         ` Peter Zijlstra
2010-12-08 19:20                                           ` Yinghai Lu
2010-12-08 19:06                                     ` Peter Zijlstra
2010-12-08 19:20                                       ` Yinghai Lu
2010-12-08 22:37                                   ` Don Zickus
2010-12-08 23:20                                     ` Eric W. Biederman
2010-12-09  4:34                                       ` Don Zickus
2010-12-09 20:20                                   ` Don Zickus
2010-12-09 20:44                                     ` Cyrill Gorcunov
2010-12-08 14:33                               ` Peter Zijlstra
2010-12-08 14:39                               ` Vivek Goyal
2010-12-07 21:16                         ` Don Zickus
2010-12-08  0:26                           ` Yinghai Lu
2010-12-08 10:39                             ` Peter Zijlstra
2010-12-01 20:41               ` Eric W. Biederman

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=1291820356.28378.83.camel@laptop \
    --to=peterz@infradead.org \
    --cc=dzickus@redhat.com \
    --cc=ebiederm@xmission.com \
    --cc=hbabu@us.ibm.com \
    --cc=jason.wessel@windriver.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    --cc=vgoyal@redhat.com \
    --cc=yinghai@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.