perf/x86/intel: Correct wrong 'or' operation for pmu capabilities

[PATCH] perf/x86/intel: Correct wrong 'or' operation for pmu capabilities

Posted by Dapeng Mi 2 years ago

When running perf-stat command on Intel hybrid platform, perf-stat
reports the following errors.

sudo taskset -c 7 ./perf stat -vvvv -e cpu_atom/instructions/ sleep 1

Opening: cpu/cycles/:HG
------------------------------------------------------------
perf_event_attr:
  type                             0 (PERF_TYPE_HARDWARE)
  config                           0xa00000000
  disabled                         1
------------------------------------------------------------
sys_perf_event_open: pid 0  cpu -1  group_fd -1  flags 0x8
sys_perf_event_open failed, error -16

 Performance counter stats for 'sleep 1':

     <not counted>      cpu_atom/instructions/

It looks the cpu_atom/instructions/ event can't be enabled on atom pmu
even the process is pinned on atom core. The investigation shows that
exclusive_event_init() helper always returns -EBUSY error in the perf
event creation. That's strange since the atom pmu should not be a
exclusive PMU.

Further investigation shows the issue is introduced by commit 97588df87b56
("perf/x86/intel: Add common intel_pmu_init_hybrid()"). The commit
originally intents to clear the bit PERF_PMU_CAP_AUX_OUTPUT from pmu
capabilities if intel_cap.pebs_output_pt_available is not set, but it
incorrectly uses 'or' operation and leads to all pmu capabilities bits
are set to 1 except bit PERF_PMU_CAP_AUX_OUTPUT.

Testing this fix on Intel hybrid platforms, the observed issues
disappear.

Fixes: 97588df87b56 ("perf/x86/intel: Add common intel_pmu_init_hybrid()")
Cc: stable@vger.kernel.org
Signed-off-by: Dapeng Mi <dapeng1.mi@linux.intel.com>
---
 arch/x86/events/intel/core.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/x86/events/intel/core.c b/arch/x86/events/intel/core.c
index a08f794a0e79..ce1c777227b4 100644
--- a/arch/x86/events/intel/core.c
+++ b/arch/x86/events/intel/core.c
@@ -4660,7 +4660,7 @@ static void intel_pmu_check_hybrid_pmus(struct x86_hybrid_pmu *pmu)
 	if (pmu->intel_cap.pebs_output_pt_available)
 		pmu->pmu.capabilities |= PERF_PMU_CAP_AUX_OUTPUT;
 	else
-		pmu->pmu.capabilities |= ~PERF_PMU_CAP_AUX_OUTPUT;
+		pmu->pmu.capabilities &= ~PERF_PMU_CAP_AUX_OUTPUT;
 
 	intel_pmu_check_event_constraints(pmu->event_constraints,
 					  pmu->num_counters,

base-commit: b85ea95d086471afb4ad062012a4d73cd328fa86
-- 
2.34.1

Re: [PATCH] perf/x86/intel: Correct wrong 'or' operation for pmu capabilities

Posted by Liang, Kan 2 years ago


On 2023-11-20 8:46 p.m., Dapeng Mi wrote:
> When running perf-stat command on Intel hybrid platform, perf-stat
> reports the following errors.
> 
> sudo taskset -c 7 ./perf stat -vvvv -e cpu_atom/instructions/ sleep 1
> 
> Opening: cpu/cycles/:HG
> ------------------------------------------------------------
> perf_event_attr:
>   type                             0 (PERF_TYPE_HARDWARE)
>   config                           0xa00000000
>   disabled                         1
> ------------------------------------------------------------
> sys_perf_event_open: pid 0  cpu -1  group_fd -1  flags 0x8
> sys_perf_event_open failed, error -16
> 
>  Performance counter stats for 'sleep 1':
> 
>      <not counted>      cpu_atom/instructions/
> 
> It looks the cpu_atom/instructions/ event can't be enabled on atom pmu
> even the process is pinned on atom core. The investigation shows that
> exclusive_event_init() helper always returns -EBUSY error in the perf
> event creation. That's strange since the atom pmu should not be a
> exclusive PMU.
> 
> Further investigation shows the issue is introduced by commit 97588df87b56
> ("perf/x86/intel: Add common intel_pmu_init_hybrid()"). The commit
> originally intents to clear the bit PERF_PMU_CAP_AUX_OUTPUT from pmu
> capabilities if intel_cap.pebs_output_pt_available is not set, but it
> incorrectly uses 'or' operation and leads to all pmu capabilities bits
> are set to 1 except bit PERF_PMU_CAP_AUX_OUTPUT.
> 
> Testing this fix on Intel hybrid platforms, the observed issues
> disappear.
> 
> Fixes: 97588df87b56 ("perf/x86/intel: Add common intel_pmu_init_hybrid()")
> Cc: stable@vger.kernel.org
> Signed-off-by: Dapeng Mi <dapeng1.mi@linux.intel.com>

Reviewed-by: Kan Liang <kan.liang@linux.intel.com>

Thanks,
Kan

> ---
>  arch/x86/events/intel/core.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/arch/x86/events/intel/core.c b/arch/x86/events/intel/core.c
> index a08f794a0e79..ce1c777227b4 100644
> --- a/arch/x86/events/intel/core.c
> +++ b/arch/x86/events/intel/core.c
> @@ -4660,7 +4660,7 @@ static void intel_pmu_check_hybrid_pmus(struct x86_hybrid_pmu *pmu)
>  	if (pmu->intel_cap.pebs_output_pt_available)
>  		pmu->pmu.capabilities |= PERF_PMU_CAP_AUX_OUTPUT;
>  	else
> -		pmu->pmu.capabilities |= ~PERF_PMU_CAP_AUX_OUTPUT;
> +		pmu->pmu.capabilities &= ~PERF_PMU_CAP_AUX_OUTPUT;
>  
>  	intel_pmu_check_event_constraints(pmu->event_constraints,
>  					  pmu->num_counters,
> 
> base-commit: b85ea95d086471afb4ad062012a4d73cd328fa86

[tip: perf/urgent] perf/x86/intel: Correct incorrect 'or' operation for PMU capabilities

Posted by tip-bot2 for Dapeng Mi 2 years ago

The following commit has been merged into the perf/urgent branch of tip:

Commit-ID:     e8df9d9f4209c04161321d8c12640ae560f65939
Gitweb:        https://git.kernel.org/tip/e8df9d9f4209c04161321d8c12640ae560f65939
Author:        Dapeng Mi <dapeng1.mi@linux.intel.com>
AuthorDate:    Tue, 21 Nov 2023 09:46:28 +08:00
Committer:     Ingo Molnar <mingo@kernel.org>
CommitterDate: Tue, 21 Nov 2023 13:44:36 +01:00

perf/x86/intel: Correct incorrect 'or' operation for PMU capabilities

When running perf-stat command on Intel hybrid platform, perf-stat
reports the following errors:

  sudo taskset -c 7 ./perf stat -vvvv -e cpu_atom/instructions/ sleep 1

  Opening: cpu/cycles/:HG
  ------------------------------------------------------------
  perf_event_attr:
    type                             0 (PERF_TYPE_HARDWARE)
    config                           0xa00000000
    disabled                         1
  ------------------------------------------------------------
  sys_perf_event_open: pid 0  cpu -1  group_fd -1  flags 0x8
  sys_perf_event_open failed, error -16

   Performance counter stats for 'sleep 1':

       <not counted>      cpu_atom/instructions/

It looks the cpu_atom/instructions/ event can't be enabled on atom PMU
even when the process is pinned on atom core. Investigation shows that
exclusive_event_init() helper always returns -EBUSY error in the perf
event creation. That's strange since the atom PMU should not be an
exclusive PMU.

Further investigation shows the issue was introduced by commit:

  97588df87b56 ("perf/x86/intel: Add common intel_pmu_init_hybrid()")

The commit originally intents to clear the bit PERF_PMU_CAP_AUX_OUTPUT
from PMU capabilities if intel_cap.pebs_output_pt_available is not set,
but it incorrectly uses 'or' operation and leads to all PMU capabilities
bits are set to 1 except bit PERF_PMU_CAP_AUX_OUTPUT.

Testing this fix on Intel hybrid platforms, the observed issues
disappear.

Fixes: 97588df87b56 ("perf/x86/intel: Add common intel_pmu_init_hybrid()")
Signed-off-by: Dapeng Mi <dapeng1.mi@linux.intel.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20231121014628.729989-1-dapeng1.mi@linux.intel.com
---
 arch/x86/events/intel/core.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/x86/events/intel/core.c b/arch/x86/events/intel/core.c
index a08f794..ce1c777 100644
--- a/arch/x86/events/intel/core.c
+++ b/arch/x86/events/intel/core.c
@@ -4660,7 +4660,7 @@ static void intel_pmu_check_hybrid_pmus(struct x86_hybrid_pmu *pmu)
 	if (pmu->intel_cap.pebs_output_pt_available)
 		pmu->pmu.capabilities |= PERF_PMU_CAP_AUX_OUTPUT;
 	else
-		pmu->pmu.capabilities |= ~PERF_PMU_CAP_AUX_OUTPUT;
+		pmu->pmu.capabilities &= ~PERF_PMU_CAP_AUX_OUTPUT;
 
 	intel_pmu_check_event_constraints(pmu->event_constraints,
 					  pmu->num_counters,