[PATCH v1 05/10] perf vendor events intel: Update graniterapids events from 1.16 to 1.17

[Ian Rogers]

The updated events were published in:
https://github.com/intel/perfmon/commit/c9ebc3ff9c3d408a888fbfbe73d386ef86c7306f

With new IO and SNC metrics in:
https://github.com/intel/perfmon/commit/04cf5e1e804afd775401167870d48cd25864be7b
https://github.com/intel/perfmon/commit/98b2602d83de6625bae1e6fcaab3a39b0a341255

Signed-off-by: Ian Rogers <irogers at google.com>
---
 .../arch/x86/graniterapids/frontend.json      | 16 +++++++++++
 .../arch/x86/graniterapids/gnr-metrics.json   | 27 +++++++++++++++++++
 tools/perf/pmu-events/arch/x86/mapfile.csv    |  2 +-
 3 files changed, 44 insertions(+), 1 deletion(-)

diff --git a/tools/perf/pmu-events/arch/x86/graniterapids/frontend.json b/tools/perf/pmu-events/arch/x86/graniterapids/frontend.json
index d580d305c926..1fdeaebb739f 100644
--- a/tools/perf/pmu-events/arch/x86/graniterapids/frontend.json
+++ b/tools/perf/pmu-events/arch/x86/graniterapids/frontend.json
@@ -325,6 +325,22 @@
         "SampleAfterValue": "200003",
         "UMask": "0x4"
     },
+    {
+        "BriefDescription": "ICACHE_TAG.STALLS_INUSE",
+        "Counter": "0,1,2,3",
+        "EventCode": "0x83",
+        "EventName": "ICACHE_TAG.STALLS_INUSE",
+        "SampleAfterValue": "200003",
+        "UMask": "0x10"
+    },
+    {
+        "BriefDescription": "ICACHE_TAG.STALLS_ISB",
+        "Counter": "0,1,2,3",
+        "EventCode": "0x83",
+        "EventName": "ICACHE_TAG.STALLS_ISB",
+        "SampleAfterValue": "200003",
+        "UMask": "0x8"
+    },
     {
         "BriefDescription": "Cycles Decode Stream Buffer (DSB) is delivering any Uop",
         "Counter": "0,1,2,3",
diff --git a/tools/perf/pmu-events/arch/x86/graniterapids/gnr-metrics.json b/tools/perf/pmu-events/arch/x86/graniterapids/gnr-metrics.json
index cc3c834ca286..299631fb8d53 100644
--- a/tools/perf/pmu-events/arch/x86/graniterapids/gnr-metrics.json
+++ b/tools/perf/pmu-events/arch/x86/graniterapids/gnr-metrics.json
@@ -143,6 +143,12 @@
         "MetricName": "io_full_write_l3_miss",
         "ScaleUnit": "100%"
     },
+    {
+        "BriefDescription": "The number of times per second that ownership of a cacheline was stolen from the integrated IO controller before it was able to write back the modified line",
+        "MetricExpr": "(UNC_I_MISC1.LOST_FWD + UNC_I_MISC1.SEC_RCVD_INVLD) / duration_time",
+        "MetricName": "io_lost_fwd",
+        "ScaleUnit": "1per_sec"
+    },
     {
         "BriefDescription": "Message Signaled Interrupts (MSI) per second sent by the integrated I/O traffic controller (IIO) to System Configuration Controller (Ubox)",
         "MetricExpr": "UNC_IIO_NUM_REQ_OF_CPU_BY_TGT.UBOX_POSTED / duration_time",
@@ -294,6 +300,27 @@
         "MetricName": "memory_bandwidth_write",
         "ScaleUnit": "1MB/s"
     },
+    {
+        "BriefDescription": "All reads to the local sub-numa cluster cache as a percentage of total memory read accesses",
+        "MetricExpr": "(L2_LINES_IN.ALL - (OCR.READS_TO_CORE.SNC_CACHE.HITM + OCR.READS_TO_CORE.SNC_CACHE.HIT_WITH_FWD + OCR.READS_TO_CORE.REMOTE_CACHE.SNOOP_FWD + OCR.READS_TO_CORE.REMOTE_MEMORY + OCR.READS_TO_CORE.L3_MISS_LOCAL)) / L2_LINES_IN.ALL",
+        "MetricName": "numa_percent_all_reads_to_local_cluster_cache",
+        "PublicDescription": "All reads to the local sub-numa cluster cache as a percentage of total memory read accesses. Includes demand and prefetch requests for data reads, code reads, read for ownerships (RFO), does not include LLC prefetches",
+        "ScaleUnit": "100%"
+    },
+    {
+        "BriefDescription": "All reads to the local sub-numa cluster memory as a percentage of total memory read accesses",
+        "MetricExpr": "OCR.READS_TO_CORE.L3_MISS_LOCAL / L2_LINES_IN.ALL",
+        "MetricName": "numa_percent_all_reads_to_local_cluster_memory",
+        "PublicDescription": "All reads to the local sub-numa cluster memory as a percentage of total memory read accesses. Includes demand and prefetch requests for data reads, code reads, read for ownerships (RFO), does not include LLC prefetches",
+        "ScaleUnit": "100%"
+    },
+    {
+        "BriefDescription": "All reads to a remote sub-numa cluster cache as a percentage of total memory read accesses",
+        "MetricExpr": "(OCR.READS_TO_CORE.SNC_CACHE.HIT_WITH_FWD + OCR.READS_TO_CORE.SNC_CACHE.HITM) / L2_LINES_IN.ALL",
+        "MetricName": "numa_percent_all_reads_to_remote_cluster_cache",
+        "PublicDescription": "All reads to a remote sub-numa cluster cache as a percentage of total memory read accesses. Includes demand and prefetch requests for data reads, code reads, read for ownerships (RFO), does not include LLC prefetches",
+        "ScaleUnit": "100%"
+    },
     {
         "BriefDescription": "Memory read that miss the last level cache (LLC) addressed to local DRAM as a percentage of total memory read accesses, does not include LLC prefetches",
         "MetricExpr": "(UNC_CHA_TOR_INSERTS.IA_MISS_DRD_LOCAL + UNC_CHA_TOR_INSERTS.IA_MISS_DRD_PREF_LOCAL) / (UNC_CHA_TOR_INSERTS.IA_MISS_DRD_LOCAL + UNC_CHA_TOR_INSERTS.IA_MISS_DRD_PREF_LOCAL + UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE + UNC_CHA_TOR_INSERTS.IA_MISS_DRD_PREF_REMOTE)",
diff --git a/tools/perf/pmu-events/arch/x86/mapfile.csv b/tools/perf/pmu-events/arch/x86/mapfile.csv
index b84035dc5b4f..96580ffda7bf 100644
--- a/tools/perf/pmu-events/arch/x86/mapfile.csv
+++ b/tools/perf/pmu-events/arch/x86/mapfile.csv
@@ -13,7 +13,7 @@ GenuineIntel-6-CF,v1.21,emeraldrapids,core
 GenuineIntel-6-5[CF],v13,goldmont,core
 GenuineIntel-6-7A,v1.01,goldmontplus,core
 GenuineIntel-6-B6,v1.11,grandridge,core
-GenuineIntel-6-A[DE],v1.16,graniterapids,core
+GenuineIntel-6-A[DE],v1.17,graniterapids,core
 GenuineIntel-6-(3C|45|46),v36,haswell,core
 GenuineIntel-6-3F,v29,haswellx,core
 GenuineIntel-6-7[DE],v1.24,icelake,core
-- 
2.53.0.414.gf7e9f6c205-goog