mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: tip-bot for Andi Kleen <tipbot@zytor.com>
To: linux-tip-commits@vger.kernel.org
Cc: tglx@linutronix.de, patrick.lu@intel.com, ak@linux.intel.com,
	alexei.starovoitov@gmail.com, mingo@kernel.org, acme@kernel.org,
	peterz@infradead.org, linux-kernel@vger.kernel.org,
	hpa@zytor.com
Subject: [tip:perf/urgent] perf/x86/intel/uncore: Fix boot crash on SBOX PMU on Haswell-EP
Date: Sun, 16 Nov 2014 01:49:42 -0800	[thread overview]
Message-ID: <tip-68055915c1c22489f9658bd2b7391bb11b2cf4e4@git.kernel.org> (raw)
In-Reply-To: <1415062828-19759-4-git-send-email-andi@firstfloor.org>

Commit-ID:  68055915c1c22489f9658bd2b7391bb11b2cf4e4
Gitweb:     http://git.kernel.org/tip/68055915c1c22489f9658bd2b7391bb11b2cf4e4
Author:     Andi Kleen <ak@linux.intel.com>
AuthorDate: Mon, 3 Nov 2014 17:00:28 -0800
Committer:  Ingo Molnar <mingo@kernel.org>
CommitDate: Sun, 16 Nov 2014 09:53:36 +0100

perf/x86/intel/uncore: Fix boot crash on SBOX PMU on Haswell-EP

There were several reports that on some systems writing the SBOX0 PMU
initialization MSR would #GP at boot. This did not happen on all
systems -- my two test systems booted fine.

Writing the three initialization bits bit-by-bit seems to avoid the
problem. So add a special callback to do just that.

This replaces an earlier patch that disabled the SBOX.

Reported-by: Alexei Starovoitov <alexei.starovoitov@gmail.com>
Reported-and-Tested-by: Patrick Lu <patrick.lu@intel.com>
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
Link: http://lkml.kernel.org/r/1415062828-19759-4-git-send-email-andi@firstfloor.org
[ Fixed a whitespace error and added attribution tags that were left out inexplicably. ]
Signed-off-by: Ingo Molnar <mingo@kernel.org>
---
 .../x86/kernel/cpu/perf_event_intel_uncore_snbep.c | 33 ++++++++++++++++++++--
 1 file changed, 30 insertions(+), 3 deletions(-)

diff --git a/arch/x86/kernel/cpu/perf_event_intel_uncore_snbep.c b/arch/x86/kernel/cpu/perf_event_intel_uncore_snbep.c
index 0af1c93..f9ed429 100644
--- a/arch/x86/kernel/cpu/perf_event_intel_uncore_snbep.c
+++ b/arch/x86/kernel/cpu/perf_event_intel_uncore_snbep.c
@@ -486,14 +486,17 @@ static struct attribute_group snbep_uncore_qpi_format_group = {
 	.attrs = snbep_uncore_qpi_formats_attr,
 };
 
-#define SNBEP_UNCORE_MSR_OPS_COMMON_INIT()			\
-	.init_box	= snbep_uncore_msr_init_box,		\
+#define __SNBEP_UNCORE_MSR_OPS_COMMON_INIT()			\
 	.disable_box	= snbep_uncore_msr_disable_box,		\
 	.enable_box	= snbep_uncore_msr_enable_box,		\
 	.disable_event	= snbep_uncore_msr_disable_event,	\
 	.enable_event	= snbep_uncore_msr_enable_event,	\
 	.read_counter	= uncore_msr_read_counter
 
+#define SNBEP_UNCORE_MSR_OPS_COMMON_INIT()			\
+	__SNBEP_UNCORE_MSR_OPS_COMMON_INIT(),			\
+	.init_box	= snbep_uncore_msr_init_box		\
+
 static struct intel_uncore_ops snbep_uncore_msr_ops = {
 	SNBEP_UNCORE_MSR_OPS_COMMON_INIT(),
 };
@@ -1919,6 +1922,30 @@ static struct intel_uncore_type hswep_uncore_cbox = {
 	.format_group		= &hswep_uncore_cbox_format_group,
 };
 
+/*
+ * Write SBOX Initialization register bit by bit to avoid spurious #GPs
+ */
+static void hswep_uncore_sbox_msr_init_box(struct intel_uncore_box *box)
+{
+	unsigned msr = uncore_msr_box_ctl(box);
+
+	if (msr) {
+		u64 init = SNBEP_PMON_BOX_CTL_INT;
+		u64 flags = 0;
+		int i;
+
+		for_each_set_bit(i, (unsigned long *)&init, 64) {
+			flags |= (1ULL << i);
+			wrmsrl(msr, flags);
+		}
+	}
+}
+
+static struct intel_uncore_ops hswep_uncore_sbox_msr_ops = {
+	__SNBEP_UNCORE_MSR_OPS_COMMON_INIT(),
+	.init_box		= hswep_uncore_sbox_msr_init_box
+};
+
 static struct attribute *hswep_uncore_sbox_formats_attr[] = {
 	&format_attr_event.attr,
 	&format_attr_umask.attr,
@@ -1944,7 +1971,7 @@ static struct intel_uncore_type hswep_uncore_sbox = {
 	.event_mask		= HSWEP_S_MSR_PMON_RAW_EVENT_MASK,
 	.box_ctl		= HSWEP_S0_MSR_PMON_BOX_CTL,
 	.msr_offset		= HSWEP_SBOX_MSR_OFFSET,
-	.ops			= &snbep_uncore_msr_ops,
+	.ops			= &hswep_uncore_sbox_msr_ops,
 	.format_group		= &hswep_uncore_sbox_format_group,
 };
 

      reply	other threads:[~2014-11-16  9:50 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-11-04  1:00 Fixes for the Haswell EP uncore driver Andi Kleen
2014-11-04  1:00 ` [PATCH 1/3] perf, x86: Add scaling units to the EP iMC events Andi Kleen
2014-11-16 12:34   ` [tip:perf/core] perf/x86/intel/uncore: " tip-bot for Andi Kleen
2014-11-04  1:00 ` [PATCH 2/3] perf, x86, uncore: Fix IRP uncore register offsets on Haswell EP Andi Kleen
2014-11-16  9:49   ` [tip:perf/urgent] perf/x86/intel/uncore: " tip-bot for Andi Kleen
2014-11-04  1:00 ` [PATCH 3/3] perf, x86, uncore: Fix SBOX PMU on Haswell-EP to fix boot crash Andi Kleen
2014-11-16  9:49   ` tip-bot for Andi Kleen [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=tip-68055915c1c22489f9658bd2b7391bb11b2cf4e4@git.kernel.org \
    --to=tipbot@zytor.com \
    --cc=acme@kernel.org \
    --cc=ak@linux.intel.com \
    --cc=alexei.starovoitov@gmail.com \
    --cc=hpa@zytor.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-tip-commits@vger.kernel.org \
    --cc=mingo@kernel.org \
    --cc=patrick.lu@intel.com \
    --cc=peterz@infradead.org \
    --cc=tglx@linutronix.de \
    /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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

Powered by JetHome