[PATCH V4 1/2] perf/x86/intel/ds: Flush the PEBS buffer in PEBS enable

* [PATCH V4 1/2] perf/x86/intel/ds: Flush the PEBS buffer in PEBS enable
@ 2023-04-21 18:45 kan.liang
  2023-04-21 18:45 ` [PATCH V4 2/2] perf/x86/intel/ds: Delay the threshold update kan.liang
                   ` (2 more replies)
  0 siblings, 3 replies; 10+ messages in thread
From: kan.liang @ 2023-04-21 18:45 UTC (permalink / raw)
  To: peterz, mingo, linux-kernel; +Cc: eranian, ak, Kan Liang

From: Kan Liang <kan.liang@linux.intel.com>

Several similar kernel warnings can be triggered,

  [56605.607840] CPU0 PEBS record size 0, expected 32, config 0
  cpuc->record_size=208

when the below commands are running in parallel for a while on SPR.

  while true; do perf record --no-buildid -a --intr-regs=AX -e
  cpu/event=0xd0,umask=0x81/pp -c 10003 -o /dev/null ./triad; done &

  while true; do perf record -o /tmp/out -W -d -e
  '{ld_blocks.store_forward:period=1000000,
  MEM_TRANS_RETIRED.LOAD_LATENCY:u:precise=2:ldlat=4}'
  -c 1037 ./triad; done
  *The triad program is just the generation of loads/stores.

The warnings are triggered when an unexpected PEBS record (with a
different config and size) is found.

A system-wide PEBS event with the large PEBS config may be enabled
during a context switch. Some PEBS records for the system-wide PEBS may
be generated while the old task is sched out but the new one hasn't been
sched in yet. When the new task is sched in, the cpuc->pebs_record_size
may be updated for the per-task PEBS events. So the existing system-wide
PEBS records have a different size from the later PEBS records.

The PEBS buffer should be flushed right before the hardware is
reprogrammed. The new size and threshold should be updated after the old
buffer has been flushed.

Reported-by: Stephane Eranian <eranian@google.com>
Suggested-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
---

Changes since V3:
- update comments

 arch/x86/events/intel/ds.c | 41 +++++++++++++++++++++++++++-----------
 1 file changed, 29 insertions(+), 12 deletions(-)

diff --git a/arch/x86/events/intel/ds.c b/arch/x86/events/intel/ds.c
index a2e566e53076..94043232991c 100644
--- a/arch/x86/events/intel/ds.c
+++ b/arch/x86/events/intel/ds.c
@@ -1252,22 +1252,26 @@ pebs_update_state(bool needed_cb, struct cpu_hw_events *cpuc,
 	if (x86_pmu.intel_cap.pebs_baseline && add) {
 		u64 pebs_data_cfg;
 
-		/* Clear pebs_data_cfg and pebs_record_size for first PEBS. */
-		if (cpuc->n_pebs == 1) {
+		/* Clear pebs_data_cfg for first PEBS. */
+		if (cpuc->n_pebs == 1)
 			cpuc->pebs_data_cfg = 0;
-			cpuc->pebs_record_size = sizeof(struct pebs_basic);
-		}
 
 		pebs_data_cfg = pebs_update_adaptive_cfg(event);
 
-		/* Update pebs_record_size if new event requires more data. */
-		if (pebs_data_cfg & ~cpuc->pebs_data_cfg) {
+		/*
+		 * Only update the pebs_data_cfg here. The pebs_record_size
+		 * will be updated later when the new pebs_data_cfg takes effect.
+		 */
+		if (pebs_data_cfg & ~cpuc->pebs_data_cfg)
 			cpuc->pebs_data_cfg |= pebs_data_cfg;
-			adaptive_pebs_record_size_update();
-			update = true;
-		}
 	}
 
+	/*
+	 * For the adaptive PEBS, the threshold will be updated later
+	 * when the new pebs_data_cfg takes effect.
+	 * The threshold may not be accurate before that, but that
+	 * does not hurt.
+	 */
 	if (update)
 		pebs_update_threshold(cpuc);
 }
@@ -1326,6 +1330,13 @@ static void intel_pmu_pebs_via_pt_enable(struct perf_event *event)
 	wrmsrl(base + idx, value);
 }
 
+static inline void intel_pmu_drain_large_pebs(struct cpu_hw_events *cpuc)
+{
+	if (cpuc->n_pebs == cpuc->n_large_pebs &&
+	    cpuc->n_pebs != cpuc->n_pebs_via_pt)
+		intel_pmu_drain_pebs_buffer();
+}
+
 void intel_pmu_pebs_enable(struct perf_event *event)
 {
 	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
@@ -1345,6 +1356,14 @@ void intel_pmu_pebs_enable(struct perf_event *event)
 	if (x86_pmu.intel_cap.pebs_baseline) {
 		hwc->config |= ICL_EVENTSEL_ADAPTIVE;
 		if (cpuc->pebs_data_cfg != cpuc->active_pebs_data_cfg) {
+			/*
+			 * drain_pebs() assumes uniform record size;
+			 * hence we need to drain when changing said
+			 * size.
+			 */
+			intel_pmu_drain_large_pebs(cpuc);
+			adaptive_pebs_record_size_update();
+			pebs_update_threshold(cpuc);
 			wrmsrl(MSR_PEBS_DATA_CFG, cpuc->pebs_data_cfg);
 			cpuc->active_pebs_data_cfg = cpuc->pebs_data_cfg;
 		}
@@ -1391,9 +1410,7 @@ void intel_pmu_pebs_disable(struct perf_event *event)
 	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
 	struct hw_perf_event *hwc = &event->hw;
 
-	if (cpuc->n_pebs == cpuc->n_large_pebs &&
-	    cpuc->n_pebs != cpuc->n_pebs_via_pt)
-		intel_pmu_drain_pebs_buffer();
+	intel_pmu_drain_large_pebs(cpuc);
 
 	cpuc->pebs_enabled &= ~(1ULL << hwc->idx);
 
-- 
2.35.1


^ permalink raw reply related	[flat|nested] 10+ messages in thread