From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
stable@vger.kernel.org,
Linus Torvalds <torvalds@linux-foundation.org>,
Stephane Eranian <eranian@google.com>,
Shuah Khan <shuah.khan@hp.com>
Subject: [ 22/40] perf,x86: fix kernel crash with PEBS/BTS after suspend/resume
Date: Mon, 18 Mar 2013 14:11:00 -0700 [thread overview]
Message-ID: <20130318211017.837471290@linuxfoundation.org> (raw)
In-Reply-To: <20130318211016.254453918@linuxfoundation.org>
3.0-stable review patch. If anyone has any objections, please let me know.
------------------
From: Stephane Eranian <eranian@google.com>
commit 1d9d8639c063caf6efc2447f5f26aa637f844ff6 upstream.
This patch fixes a kernel crash when using precise sampling (PEBS)
after a suspend/resume. Turns out the CPU notifier code is not invoked
on CPU0 (BP). Therefore, the DS_AREA (used by PEBS) is not restored properly
by the kernel and keeps it power-on/resume value of 0 causing any PEBS
measurement to crash when running on CPU0.
The workaround is to add a hook in the actual resume code to restore
the DS Area MSR value. It is invoked for all CPUS. So for all but CPU0,
the DS_AREA will be restored twice but this is harmless.
Reported-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Stephane Eranian <eranian@google.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Shuah Khan <shuah.khan@hp.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
arch/x86/kernel/cpu/perf_event_intel_ds.c | 8 ++++++++
arch/x86/power/cpu.c | 2 ++
include/linux/perf_event.h | 2 ++
3 files changed, 12 insertions(+)
--- a/arch/x86/kernel/cpu/perf_event_intel_ds.c
+++ b/arch/x86/kernel/cpu/perf_event_intel_ds.c
@@ -754,6 +754,14 @@ static void intel_ds_init(void)
}
}
+void perf_restore_debug_store(void)
+{
+ if (!x86_pmu.bts && !x86_pmu.pebs)
+ return;
+
+ init_debug_store_on_cpu(smp_processor_id());
+}
+
#else /* CONFIG_CPU_SUP_INTEL */
static void reserve_ds_buffers(void)
--- a/arch/x86/power/cpu.c
+++ b/arch/x86/power/cpu.c
@@ -10,6 +10,7 @@
#include <linux/suspend.h>
#include <linux/smp.h>
+#include <linux/perf_event.h>
#include <asm/pgtable.h>
#include <asm/proto.h>
@@ -224,6 +225,7 @@ static void __restore_processor_state(st
do_fpu_end();
mtrr_bp_restore();
+ perf_restore_debug_store();
}
/* Needed by apm.c */
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -1153,6 +1153,7 @@ extern void perf_swevent_put_recursion_c
extern void perf_event_enable(struct perf_event *event);
extern void perf_event_disable(struct perf_event *event);
extern void perf_event_task_tick(void);
+extern void perf_restore_debug_store(void);
#else
static inline void
perf_event_task_sched_in(struct task_struct *task) { }
@@ -1187,6 +1188,7 @@ static inline void perf_swevent_put_recu
static inline void perf_event_enable(struct perf_event *event) { }
static inline void perf_event_disable(struct perf_event *event) { }
static inline void perf_event_task_tick(void) { }
+static inline void perf_restore_debug_store(void) { }
#endif
#define perf_output_put(handle, x) perf_output_copy((handle), &(x), sizeof(x))
next prev parent reply other threads:[~2013-03-18 21:11 UTC|newest]
Thread overview: 43+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-03-18 21:10 [ 00/40] 3.0.70-stable review Greg Kroah-Hartman
2013-03-18 21:10 ` [ 01/40] qcaux: add Franklin U600 Greg Kroah-Hartman
2013-03-18 21:10 ` [ 02/40] virtio: rng: disallow multiple device registrations, fixes crashes Greg Kroah-Hartman
2013-03-18 21:10 ` [ 03/40] USB: option: add Huawei E5331 Greg Kroah-Hartman
2013-03-18 21:10 ` [ 04/40] USB: cdc-wdm: fix buffer overflow Greg Kroah-Hartman
2013-03-18 21:10 ` [ 05/40] usb: cp210x new Vendor/Device IDs Greg Kroah-Hartman
2013-03-18 21:10 ` [ 06/40] USB: added support for Cinterions products AH6 and PLS8 Greg Kroah-Hartman
2013-03-18 21:10 ` [ 07/40] usb: serial: Add Rigblaster Advantage to device table Greg Kroah-Hartman
2013-03-18 21:10 ` [ 08/40] USB: storage: fix Huawei mode switching regression Greg Kroah-Hartman
2013-03-18 21:10 ` [ 09/40] USB: EHCI: dont check DMA values in QH overlays Greg Kroah-Hartman
2013-03-18 21:10 ` [ 10/40] staging: vt6656: Fix oops on resume from suspend Greg Kroah-Hartman
2013-03-18 21:10 ` [ 11/40] signal: always clear sa_restorer on execve Greg Kroah-Hartman
2013-03-18 21:10 ` [ 12/40] ext3: Fix format string issues Greg Kroah-Hartman
2013-03-18 21:10 ` [ 13/40] tty: serial: fix typo "ARCH_S5P6450" Greg Kroah-Hartman
2013-03-18 21:10 ` [ 14/40] TTY: do not reset masters packet mode Greg Kroah-Hartman
2013-03-18 21:10 ` [ 15/40] md: protect against crash upon fsync on ro array Greg Kroah-Hartman
2013-03-18 21:10 ` [ 16/40] Revert duplicated perf commit Greg Kroah-Hartman
2013-03-21 3:31 ` Ben Hutchings
2013-03-18 21:10 ` [ 17/40] ALSA: seq: Fix missing error handling in snd_seq_timer_open() Greg Kroah-Hartman
2013-03-18 21:10 ` [ 18/40] hwmon: (lineage-pem) Add missing terminating entry for pem_[input|fan]_attributes Greg Kroah-Hartman
2013-03-18 21:10 ` [ 19/40] w1: fix oops when w1_search is called from netlink connector Greg Kroah-Hartman
2013-03-18 21:10 ` [ 20/40] powerpc: Fix cputable entry for 970MP rev 1.0 Greg Kroah-Hartman
2013-03-18 21:10 ` [ 21/40] selinux: use GFP_ATOMIC under spin_lock Greg Kroah-Hartman
2013-03-18 21:11 ` Greg Kroah-Hartman [this message]
2013-03-18 21:11 ` [ 23/40] perf,x86: fix wrmsr_on_cpu() warning on suspend/resume Greg Kroah-Hartman
2013-03-18 21:11 ` [ 24/40] perf,x86: fix link failure for non-Intel configs Greg Kroah-Hartman
2013-03-18 21:11 ` [ 25/40] s390/mm: fix flush_tlb_kernel_range() Greg Kroah-Hartman
2013-03-18 21:11 ` [ 26/40] btrfs: use rcu_barrier() to wait for bdev puts at unmount Greg Kroah-Hartman
2013-03-18 21:11 ` [ 27/40] drm/i915: EBUSY status handling added to i915_gem_fault() Greg Kroah-Hartman
2013-03-18 21:11 ` [ 28/40] hwmon: (sht15) Fix memory leak if regulator_enable() fails Greg Kroah-Hartman
2013-03-18 21:11 ` [ 29/40] block: use i_size_write() in bd_set_size() Greg Kroah-Hartman
2013-03-18 21:11 ` [ 30/40] loopdev: fix a deadlock Greg Kroah-Hartman
2013-03-18 21:11 ` [ 31/40] l2tp: Restore socket refcount when sendmsg succeeds Greg Kroah-Hartman
2013-03-18 21:11 ` [ 32/40] rds: limit the size allocated by rds_message_alloc() Greg Kroah-Hartman
2013-03-18 21:11 ` [ 33/40] net: ipv6: Dont purge default router if accept_ra=2 Greg Kroah-Hartman
2013-03-18 21:11 ` [ 34/40] tcp: fix double-counted receiver RTT when leaving receiver fast path Greg Kroah-Hartman
2013-03-18 21:11 ` [ 35/40] tun: add a missing nf_reset() in tun_net_xmit() Greg Kroah-Hartman
2013-03-18 21:11 ` [ 36/40] netlabel: correctly list all the static label mappings Greg Kroah-Hartman
2013-03-18 21:11 ` [ 37/40] bridging: fix rx_handlers return code Greg Kroah-Hartman
2013-03-18 21:11 ` [ 38/40] ipv6: stop multicast forwarding to process interface scoped addresses Greg Kroah-Hartman
2013-03-18 21:11 ` [ 39/40] rtnl: fix info leak on RTM_GETLINK request for VF devices Greg Kroah-Hartman
2013-03-18 21:11 ` [ 40/40] dcbnl: fix various netlink info leaks Greg Kroah-Hartman
2013-03-19 0:47 ` [ 00/40] 3.0.70-stable review Shuah Khan
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=20130318211017.837471290@linuxfoundation.org \
--to=gregkh@linuxfoundation.org \
--cc=eranian@google.com \
--cc=linux-kernel@vger.kernel.org \
--cc=shuah.khan@hp.com \
--cc=stable@vger.kernel.org \
--cc=torvalds@linux-foundation.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).