Skip to contentSkip to conversation

perf,x86: fix kernel crash with PEBS/BTS after suspend/resume

PATCH In review Shuah Khan 2 participants 2 messages Mar 16, 2013 – Mar 21, 2013 Stable branch

lore
Share & tools
lore

Shuah Khan

author
#0 lore raw

PATCH 1/1 perf,x86: fix kernel crash with PEBS/BTS after suspend/resume

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.

Upstream commit id: 1d9d8639c063caf6efc2447f5f26aa637f844ff6

---
 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(+)
arch/x86/kernel/cpu/perf_event_intel_ds.c +8 -0
index d812fe2..cf82ee5 100644
@@ -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)
arch/x86/power/cpu.c +2 -0
index 87bb35e..0ea8bd2 100644
@@ -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(struct saved_context *ctxt)
 
 	do_fpu_end();
 	mtrr_bp_restore();
+	perf_restore_debug_store();
 }
 
 /* Needed by apm.c */
include/linux/perf_event.h +2 -0
index effadd6..038ad4a 100644
@@ -1153,6 +1153,7 @@ extern void perf_swevent_put_recursion_context(int rctx);
 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_recursion_context(int rctx)		{ }
 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))
Signature · 1 line
1.7.9.5
#1 lore raw
Quoted 10 lines · Shuah Khan

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.

Quoted 7 lines

Upstream commit id: 1d9d8639c063caf6efc2447f5f26aa637f844ff6

[...]

I've queued this up for 3.2 (though context was different again), along with the following fix-ups:

commit 2a6e06b2aed6995af401dcd4feb5e79a0c7ea554

Author: Linus Torvalds <[email protected]>
Date:   Sun Mar 17 15:44:43 2013 -0700

perf,x86: fix wrmsr_on_cpu() warning on suspend/resume

commit 6c4d3bc99b3341067775efd4d9d13cc8e655fd7c

Author: David Rientjes <[email protected]>
Date:   Sun Mar 17 15:49:10 2013 -0700

perf,x86: fix link failure for non-Intel configs

Ben.

Signature · 2 lines
Ben Hutchings
It is easier to write an incorrect program than to understand a correct one.