Skip to contentSkip to conversation

perf/x86/intel/cqm: Check cqm/mbm enabled state in event init

PATCH In review Jiri Olsa 3 participants 3 messages Sep 5, 2016 – Sep 6, 2016

Subsystems (2)
lore
Share & tools
lore

Jiri Olsa

author reviewer
#0 lore raw

PATCH 1/1 perf/x86/intel/cqm: Check cqm/mbm enabled state in event init

Yanqiu Zhang reported kernel panic when using mbm event on system where CQM is detected but without mbm event support, like with perf:

  # perf stat -e 'intel_cqm/event=3/' -a
  BUG: unable to handle kernel NULL pointer dereference at 0000000000000020
  IP: [<ffffffff8100d64c>] update_sample+0xbc/0xe0
  ...
   <IRQ>
   [<ffffffff8100d688>] __intel_mbm_event_init+0x18/0x20
   [<ffffffff81113d6b>] flush_smp_call_function_queue+0x7b/0x160
   [<ffffffff81114853>] generic_smp_call_function_single_interrupt+0x13/0x60
   [<ffffffff81052017>] smp_call_function_interrupt+0x27/0x40
   [<ffffffff816fb06c>] call_function_interrupt+0x8c/0xa0
  ...

The reason is that we currently allow to init mbm event even if mbm support is not detected. Adding checks for both cqm and mbm events and support into cqm's event_init.

---
 arch/x86/events/intel/cqm.c | 9 +++++++++
 1 file changed, 9 insertions(+)
arch/x86/events/intel/cqm.c +9 -0
index 783c49ddef29..8f82b02934fa 100644
@@ -458,6 +458,11 @@ static void __intel_cqm_event_count(void *info);
 static void init_mbm_sample(u32 rmid, u32 evt_type);
 static void __intel_mbm_event_count(void *info);
 
+static bool is_cqm_event(int e)
+{
+	return (e == QOS_L3_OCCUP_EVENT_ID);
+}
+
 static bool is_mbm_event(int e)
 {
 	return (e >= QOS_MBM_TOTAL_EVENT_ID && e <= QOS_MBM_LOCAL_EVENT_ID);
@@ -1366,6 +1371,10 @@ static int intel_cqm_event_init(struct perf_event *event)
 	     (event->attr.config > QOS_MBM_LOCAL_EVENT_ID))
 		return -EINVAL;
 
+	if ((is_cqm_event(event->attr.config) && !cqm_enabled) ||
+	    (is_mbm_event(event->attr.config) && !mbm_enabled))
+		return -EINVAL;
+
 	/* unsupported modes and filters */
 	if (event->attr.exclude_user   ||
 	    event->attr.exclude_kernel ||
Signature · 1 line
2.7.4
maintainer
#1 lore raw
Quoted 17 lines · Jiri Olsa

Yanqiu Zhang reported kernel panic when using mbm event on system where CQM is detected but without mbm event support, like with perf:

  # perf stat -e 'intel_cqm/event=3/' -a
  BUG: unable to handle kernel NULL pointer dereference at 0000000000000020
  IP: [<ffffffff8100d64c>] update_sample+0xbc/0xe0
  ...
   <IRQ>
   [<ffffffff8100d688>] __intel_mbm_event_init+0x18/0x20
   [<ffffffff81113d6b>] flush_smp_call_function_queue+0x7b/0x160
   [<ffffffff81114853>] generic_smp_call_function_single_interrupt+0x13/0x60
   [<ffffffff81052017>] smp_call_function_interrupt+0x27/0x40
   [<ffffffff816fb06c>] call_function_interrupt+0x8c/0xa0
  ...
Quoted 6 lines

The reason is that we currently allow to init mbm event even if mbm support is not detected. Adding checks for both cqm and mbm events and support into cqm's event_init.

Fixes 33c3cc7acfd9 ("perf/x86/mbm: Add Intel Memory B/W Monitoring enumeration and init")
#2 lore raw
In reply to Jiri Olsa
perf/x86/intel/cqm: Check cqm/mbm enabled state in event init
Commit-ID:  79d102cbfd2e9d94257fcc7c82807ef1cdf80322
Gitweb:     http://git.kernel.org/tip/79d102cbfd2e9d94257fcc7c82807ef1cdf80322
Author:     Jiri Olsa <[email protected]>
AuthorDate: Mon, 5 Sep 2016 17:30:07 +0200
Committer:  Thomas Gleixner <[email protected]>
CommitDate: Tue, 6 Sep 2016 10:42:12 +0200

perf/x86/intel/cqm: Check cqm/mbm enabled state in event init

Yanqiu Zhang reported kernel panic when using mbm event on system where CQM is detected but without mbm event support, like with perf:

  # perf stat -e 'intel_cqm/event=3/' -a
  BUG: unable to handle kernel NULL pointer dereference at 0000000000000020
  IP: [<ffffffff8100d64c>] update_sample+0xbc/0xe0
  ...
   <IRQ>
   [<ffffffff8100d688>] __intel_mbm_event_init+0x18/0x20
   [<ffffffff81113d6b>] flush_smp_call_function_queue+0x7b/0x160
   [<ffffffff81114853>] generic_smp_call_function_single_interrupt+0x13/0x60
   [<ffffffff81052017>] smp_call_function_interrupt+0x27/0x40
   [<ffffffff816fb06c>] call_function_interrupt+0x8c/0xa0
  ...

The reason is that we currently allow to init mbm event even if mbm support is not detected. Adding checks for both cqm and mbm events and support into cqm's event_init.

Fixes 33c3cc7acfd9 ("perf/x86/mbm: Add Intel Memory B/W Monitoring enumeration and init")
Cc Vikas Shivappa <[email protected]>
Cc Tony Luck <[email protected]>
---
 arch/x86/events/intel/cqm.c | 9 +++++++++
 1 file changed, 9 insertions(+)
arch/x86/events/intel/cqm.c +9 -0
index 783c49d..8f82b02 100644
@@ -458,6 +458,11 @@ static void __intel_cqm_event_count(void *info);
 static void init_mbm_sample(u32 rmid, u32 evt_type);
 static void __intel_mbm_event_count(void *info);
 
+static bool is_cqm_event(int e)
+{
+	return (e == QOS_L3_OCCUP_EVENT_ID);
+}
+
 static bool is_mbm_event(int e)
 {
 	return (e >= QOS_MBM_TOTAL_EVENT_ID && e <= QOS_MBM_LOCAL_EVENT_ID);
@@ -1366,6 +1371,10 @@ static int intel_cqm_event_init(struct perf_event *event)
 	     (event->attr.config > QOS_MBM_LOCAL_EVENT_ID))
 		return -EINVAL;
 
+	if ((is_cqm_event(event->attr.config) && !cqm_enabled) ||
+	    (is_mbm_event(event->attr.config) && !mbm_enabled))
+		return -EINVAL;
+
 	/* unsupported modes and filters */
 	if (event->attr.exclude_user   ||
 	    event->attr.exclude_kernel ||