* [PATCH v1 04/15] perf vendor events intel: Update elkhartlake events
2023-05-15 21:58 [PATCH v1 00/15] Event, metric and metric group improvements Ian Rogers
@ 2023-05-15 21:58 ` Ian Rogers
2023-05-15 21:58 ` [PATCH v1 08/15] perf vendor events intel: Update jaketown metrics Ian Rogers
` (4 subsequent siblings)
5 siblings, 0 replies; 9+ messages in thread
From: Ian Rogers @ 2023-05-15 21:58 UTC (permalink / raw)
To: Peter Zijlstra, Ingo Molnar, Arnaldo Carvalho de Melo,
Mark Rutland, Alexander Shishkin, Jiri Olsa, Namhyung Kim,
Ian Rogers, Adrian Hunter, Kan Liang, Zhengjun Xing, John Garry,
Kajol Jain, Thomas Richter, linux-kernel, linux-perf-users
Update elkhartlake to v1.04 that marks deprecated a number of events
and adds additional description to MEM_BOUND_STALLS.IFETCH. The events
data was generated by:
https://github.com/intel/perfmon/blob/main/scripts/create_perf_json.py
Signed-off-by: Ian Rogers <irogers@google.com>
---
tools/perf/pmu-events/arch/x86/elkhartlake/cache.json | 7 +++++++
tools/perf/pmu-events/arch/x86/elkhartlake/memory.json | 2 ++
tools/perf/pmu-events/arch/x86/elkhartlake/other.json | 10 ++++++++++
.../perf/pmu-events/arch/x86/elkhartlake/pipeline.json | 3 +++
tools/perf/pmu-events/arch/x86/mapfile.csv | 2 +-
5 files changed, 23 insertions(+), 1 deletion(-)
diff --git a/tools/perf/pmu-events/arch/x86/elkhartlake/cache.json b/tools/perf/pmu-events/arch/x86/elkhartlake/cache.json
index 0ab90e3bf76b..c6be60584522 100644
--- a/tools/perf/pmu-events/arch/x86/elkhartlake/cache.json
+++ b/tools/perf/pmu-events/arch/x86/elkhartlake/cache.json
@@ -72,6 +72,7 @@
"BriefDescription": "Counts the number of cycles the core is stalled due to an instruction cache or TLB miss which hit in the L2, LLC, DRAM or MMIO (Non-DRAM).",
"EventCode": "0x34",
"EventName": "MEM_BOUND_STALLS.IFETCH",
+ "PublicDescription": "Counts the number of cycles the core is stalled due to an instruction cache or translation lookaside buffer (TLB) miss which hit in the L2, LLC, DRAM or MMIO (Non-DRAM).",
"SampleAfterValue": "200003",
"UMask": "0x38"
},
@@ -437,6 +438,7 @@
},
{
"BriefDescription": "This event is deprecated. Refer to new event OCR.DEMAND_DATA_AND_L1PF_RD.L3_HIT",
+ "Deprecated": "1",
"EventCode": "0XB7",
"EventName": "OCR.DEMAND_DATA_RD.L3_HIT",
"MSRIndex": "0x1a6,0x1a7",
@@ -446,6 +448,7 @@
},
{
"BriefDescription": "This event is deprecated. Refer to new event OCR.DEMAND_DATA_AND_L1PF_RD.L3_HIT.SNOOP_HITM",
+ "Deprecated": "1",
"EventCode": "0XB7",
"EventName": "OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HITM",
"MSRIndex": "0x1a6,0x1a7",
@@ -455,6 +458,7 @@
},
{
"BriefDescription": "This event is deprecated. Refer to new event OCR.DEMAND_DATA_AND_L1PF_RD.L3_HIT.SNOOP_HIT_NO_FWD",
+ "Deprecated": "1",
"EventCode": "0XB7",
"EventName": "OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HIT_NO_FWD",
"MSRIndex": "0x1a6,0x1a7",
@@ -464,6 +468,7 @@
},
{
"BriefDescription": "This event is deprecated. Refer to new event OCR.DEMAND_DATA_AND_L1PF_RD.L3_HIT.SNOOP_HIT_WITH_FWD",
+ "Deprecated": "1",
"EventCode": "0XB7",
"EventName": "OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HIT_WITH_FWD",
"MSRIndex": "0x1a6,0x1a7",
@@ -473,6 +478,7 @@
},
{
"BriefDescription": "This event is deprecated. Refer to new event OCR.DEMAND_DATA_AND_L1PF_RD.L3_HIT.SNOOP_MISS",
+ "Deprecated": "1",
"EventCode": "0XB7",
"EventName": "OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_MISS",
"MSRIndex": "0x1a6,0x1a7",
@@ -482,6 +488,7 @@
},
{
"BriefDescription": "This event is deprecated. Refer to new event OCR.DEMAND_DATA_AND_L1PF_RD.L3_HIT.SNOOP_NOT_NEEDED",
+ "Deprecated": "1",
"EventCode": "0XB7",
"EventName": "OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_NOT_NEEDED",
"MSRIndex": "0x1a6,0x1a7",
diff --git a/tools/perf/pmu-events/arch/x86/elkhartlake/memory.json b/tools/perf/pmu-events/arch/x86/elkhartlake/memory.json
index 18621909d1a9..c02eb0e836ad 100644
--- a/tools/perf/pmu-events/arch/x86/elkhartlake/memory.json
+++ b/tools/perf/pmu-events/arch/x86/elkhartlake/memory.json
@@ -96,6 +96,7 @@
},
{
"BriefDescription": "This event is deprecated. Refer to new event OCR.DEMAND_DATA_AND_L1PF_RD.L3_MISS",
+ "Deprecated": "1",
"EventCode": "0XB7",
"EventName": "OCR.DEMAND_DATA_RD.L3_MISS",
"MSRIndex": "0x1a6,0x1a7",
@@ -105,6 +106,7 @@
},
{
"BriefDescription": "This event is deprecated. Refer to new event OCR.DEMAND_DATA_AND_L1PF_RD.L3_MISS_LOCAL",
+ "Deprecated": "1",
"EventCode": "0XB7",
"EventName": "OCR.DEMAND_DATA_RD.L3_MISS_LOCAL",
"MSRIndex": "0x1a6,0x1a7",
diff --git a/tools/perf/pmu-events/arch/x86/elkhartlake/other.json b/tools/perf/pmu-events/arch/x86/elkhartlake/other.json
index 00ae180ded25..fefbc383b840 100644
--- a/tools/perf/pmu-events/arch/x86/elkhartlake/other.json
+++ b/tools/perf/pmu-events/arch/x86/elkhartlake/other.json
@@ -1,6 +1,7 @@
[
{
"BriefDescription": "This event is deprecated. Refer to new event BUS_LOCK.SELF_LOCKS",
+ "Deprecated": "1",
"EdgeDetect": "1",
"EventCode": "0x63",
"EventName": "BUS_LOCK.ALL",
@@ -16,6 +17,7 @@
},
{
"BriefDescription": "This event is deprecated. Refer to new event BUS_LOCK.BLOCK_CYCLES",
+ "Deprecated": "1",
"EventCode": "0x63",
"EventName": "BUS_LOCK.CYCLES_OTHER_BLOCK",
"SampleAfterValue": "200003",
@@ -23,6 +25,7 @@
},
{
"BriefDescription": "This event is deprecated. Refer to new event BUS_LOCK.LOCK_CYCLES",
+ "Deprecated": "1",
"EventCode": "0x63",
"EventName": "BUS_LOCK.CYCLES_SELF_BLOCK",
"SampleAfterValue": "200003",
@@ -46,6 +49,7 @@
},
{
"BriefDescription": "This event is deprecated. Refer to new event MEM_BOUND_STALLS.LOAD_DRAM_HIT",
+ "Deprecated": "1",
"EventCode": "0x34",
"EventName": "C0_STALLS.LOAD_DRAM_HIT",
"SampleAfterValue": "200003",
@@ -53,6 +57,7 @@
},
{
"BriefDescription": "This event is deprecated. Refer to new event MEM_BOUND_STALLS.LOAD_L2_HIT",
+ "Deprecated": "1",
"EventCode": "0x34",
"EventName": "C0_STALLS.LOAD_L2_HIT",
"SampleAfterValue": "200003",
@@ -60,6 +65,7 @@
},
{
"BriefDescription": "This event is deprecated. Refer to new event MEM_BOUND_STALLS.LOAD_LLC_HIT",
+ "Deprecated": "1",
"EventCode": "0x34",
"EventName": "C0_STALLS.LOAD_LLC_HIT",
"SampleAfterValue": "200003",
@@ -207,6 +213,7 @@
},
{
"BriefDescription": "This event is deprecated. Refer to new event OCR.DEMAND_DATA_AND_L1PF_RD.ANY_RESPONSE",
+ "Deprecated": "1",
"EventCode": "0XB7",
"EventName": "OCR.DEMAND_DATA_RD.ANY_RESPONSE",
"MSRIndex": "0x1a6,0x1a7",
@@ -216,6 +223,7 @@
},
{
"BriefDescription": "This event is deprecated. Refer to new event OCR.DEMAND_DATA_AND_L1PF_RD.DRAM",
+ "Deprecated": "1",
"EventCode": "0XB7",
"EventName": "OCR.DEMAND_DATA_RD.DRAM",
"MSRIndex": "0x1a6,0x1a7",
@@ -225,6 +233,7 @@
},
{
"BriefDescription": "This event is deprecated. Refer to new event OCR.DEMAND_DATA_AND_L1PF_RD.LOCAL_DRAM",
+ "Deprecated": "1",
"EventCode": "0XB7",
"EventName": "OCR.DEMAND_DATA_RD.LOCAL_DRAM",
"MSRIndex": "0x1a6,0x1a7",
@@ -234,6 +243,7 @@
},
{
"BriefDescription": "This event is deprecated. Refer to new event OCR.DEMAND_DATA_AND_L1PF_RD.OUTSTANDING",
+ "Deprecated": "1",
"EventCode": "0XB7",
"EventName": "OCR.DEMAND_DATA_RD.OUTSTANDING",
"MSRIndex": "0x1a6",
diff --git a/tools/perf/pmu-events/arch/x86/elkhartlake/pipeline.json b/tools/perf/pmu-events/arch/x86/elkhartlake/pipeline.json
index 9dd8c909facc..c483c0838e08 100644
--- a/tools/perf/pmu-events/arch/x86/elkhartlake/pipeline.json
+++ b/tools/perf/pmu-events/arch/x86/elkhartlake/pipeline.json
@@ -165,6 +165,7 @@
},
{
"BriefDescription": "This event is deprecated.",
+ "Deprecated": "1",
"EventCode": "0xcd",
"EventName": "CYCLES_DIV_BUSY.ANY",
"SampleAfterValue": "2000003"
@@ -283,6 +284,7 @@
},
{
"BriefDescription": "This event is deprecated. Refer to new event TOPDOWN_BAD_SPECULATION.FASTNUKE",
+ "Deprecated": "1",
"EventCode": "0x73",
"EventName": "TOPDOWN_BAD_SPECULATION.MONUKE",
"SampleAfterValue": "1000003",
@@ -338,6 +340,7 @@
},
{
"BriefDescription": "This event is deprecated.",
+ "Deprecated": "1",
"EventCode": "0x74",
"EventName": "TOPDOWN_BE_BOUND.STORE_BUFFER",
"SampleAfterValue": "1000003",
diff --git a/tools/perf/pmu-events/arch/x86/mapfile.csv b/tools/perf/pmu-events/arch/x86/mapfile.csv
index 6b132eecd2a7..f3ae41e28ed2 100644
--- a/tools/perf/pmu-events/arch/x86/mapfile.csv
+++ b/tools/perf/pmu-events/arch/x86/mapfile.csv
@@ -6,7 +6,7 @@ GenuineIntel-6-(3D|47),v28,broadwell,core
GenuineIntel-6-56,v10,broadwellde,core
GenuineIntel-6-4F,v21,broadwellx,core
GenuineIntel-6-55-[56789ABCDEF],v1.18,cascadelakex,core
-GenuineIntel-6-9[6C],v1.03,elkhartlake,core
+GenuineIntel-6-9[6C],v1.04,elkhartlake,core
GenuineIntel-6-5[CF],v13,goldmont,core
GenuineIntel-6-7A,v1.01,goldmontplus,core
GenuineIntel-6-B6,v1.00,grandridge,core
--
2.40.1.606.ga4b1b128d6-goog
^ permalink raw reply related [flat|nested] 9+ messages in thread* [PATCH v1 08/15] perf vendor events intel: Update jaketown metrics
2023-05-15 21:58 [PATCH v1 00/15] Event, metric and metric group improvements Ian Rogers
2023-05-15 21:58 ` [PATCH v1 04/15] perf vendor events intel: Update elkhartlake events Ian Rogers
@ 2023-05-15 21:58 ` Ian Rogers
2023-05-15 21:58 ` [PATCH v1 09/15] perf vendor events intel: Update sandybridge metrics Ian Rogers
` (3 subsequent siblings)
5 siblings, 0 replies; 9+ messages in thread
From: Ian Rogers @ 2023-05-15 21:58 UTC (permalink / raw)
To: Peter Zijlstra, Ingo Molnar, Arnaldo Carvalho de Melo,
Mark Rutland, Alexander Shishkin, Jiri Olsa, Namhyung Kim,
Ian Rogers, Adrian Hunter, Kan Liang, Zhengjun Xing, John Garry,
Kajol Jain, Thomas Richter, linux-kernel, linux-perf-users
Metrics are updated to make TMA info metric names
synchronized. Metrics were generated by:
https://github.com/intel/perfmon/blob/main/scripts/create_perf_json.py
Signed-off-by: Ian Rogers <irogers@google.com>
---
.../arch/x86/jaketown/jkt-metrics.json | 224 +++++++++---------
1 file changed, 112 insertions(+), 112 deletions(-)
diff --git a/tools/perf/pmu-events/arch/x86/jaketown/jkt-metrics.json b/tools/perf/pmu-events/arch/x86/jaketown/jkt-metrics.json
index 66a6f657bd6f..35b1a3aa728d 100644
--- a/tools/perf/pmu-events/arch/x86/jaketown/jkt-metrics.json
+++ b/tools/perf/pmu-events/arch/x86/jaketown/jkt-metrics.json
@@ -50,7 +50,7 @@
},
{
"BriefDescription": "Uncore frequency per die [GHZ]",
- "MetricExpr": "tma_info_socket_clks / #num_dies / duration_time / 1e9",
+ "MetricExpr": "tma_info_system_socket_clks / #num_dies / duration_time / 1e9",
"MetricGroup": "SoC",
"MetricName": "UNCORE_FREQ"
},
@@ -82,7 +82,7 @@
},
{
"BriefDescription": "This category represents fraction of slots wasted due to incorrect speculations",
- "MetricExpr": "(UOPS_ISSUED.ANY - UOPS_RETIRED.RETIRE_SLOTS + 4 * (INT_MISC.RECOVERY_CYCLES_ANY / 2 if #SMT_on else INT_MISC.RECOVERY_CYCLES)) / tma_info_slots",
+ "MetricExpr": "(UOPS_ISSUED.ANY - UOPS_RETIRED.RETIRE_SLOTS + 4 * (INT_MISC.RECOVERY_CYCLES_ANY / 2 if #SMT_on else INT_MISC.RECOVERY_CYCLES)) / tma_info_thread_slots",
"MetricGroup": "TmaL1;TopdownL1;tma_L1_group",
"MetricName": "tma_bad_speculation",
"MetricThreshold": "tma_bad_speculation > 0.15",
@@ -98,12 +98,12 @@
"MetricName": "tma_branch_mispredicts",
"MetricThreshold": "tma_branch_mispredicts > 0.1 & tma_bad_speculation > 0.15",
"MetricgroupNoGroup": "TopdownL2",
- "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Branch Misprediction. These slots are either wasted by uops fetched from an incorrectly speculated program path; or stalls when the out-of-order part of the machine needs to recover its state from a speculative path. Sample with: BR_MISP_RETIRED.ALL_BRANCHES. Related metrics: tma_info_branch_misprediction_cost, tma_mispredicts_resteers",
+ "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Branch Misprediction. These slots are either wasted by uops fetched from an incorrectly speculated program path; or stalls when the out-of-order part of the machine needs to recover its state from a speculative path. Sample with: BR_MISP_RETIRED.ALL_BRANCHES. Related metrics: tma_info_bad_spec_branch_misprediction_cost, tma_mispredicts_resteers",
"ScaleUnit": "100%"
},
{
"BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers",
- "MetricExpr": "12 * (BR_MISP_RETIRED.ALL_BRANCHES + MACHINE_CLEARS.COUNT + BACLEARS.ANY) / tma_info_clks",
+ "MetricExpr": "12 * (BR_MISP_RETIRED.ALL_BRANCHES + MACHINE_CLEARS.COUNT + BACLEARS.ANY) / tma_info_thread_clks",
"MetricGroup": "FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group",
"MetricName": "tma_branch_resteers",
"MetricThreshold": "tma_branch_resteers > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)",
@@ -123,7 +123,7 @@
},
{
"BriefDescription": "This metric represents fraction of cycles where the Divider unit was active",
- "MetricExpr": "ARITH.FPU_DIV_ACTIVE / tma_info_core_clks",
+ "MetricExpr": "ARITH.FPU_DIV_ACTIVE / tma_info_core_core_clks",
"MetricGroup": "TopdownL3;tma_L3_group;tma_core_bound_group",
"MetricName": "tma_divider",
"MetricThreshold": "tma_divider > 0.2 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2)",
@@ -133,7 +133,7 @@
{
"BriefDescription": "This metric estimates how often the CPU was stalled on accesses to external memory (DRAM) by loads",
"MetricConstraint": "NO_GROUP_EVENTS_SMT",
- "MetricExpr": "(1 - MEM_LOAD_UOPS_RETIRED.LLC_HIT / (MEM_LOAD_UOPS_RETIRED.LLC_HIT + 7 * MEM_LOAD_UOPS_RETIRED.LLC_MISS)) * CYCLE_ACTIVITY.STALLS_L2_PENDING / tma_info_clks",
+ "MetricExpr": "(1 - MEM_LOAD_UOPS_RETIRED.LLC_HIT / (MEM_LOAD_UOPS_RETIRED.LLC_HIT + 7 * MEM_LOAD_UOPS_RETIRED.LLC_MISS)) * CYCLE_ACTIVITY.STALLS_L2_PENDING / tma_info_thread_clks",
"MetricGroup": "MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group",
"MetricName": "tma_dram_bound",
"MetricThreshold": "tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)",
@@ -142,16 +142,16 @@
},
{
"BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines",
- "MetricExpr": "DSB2MITE_SWITCHES.PENALTY_CYCLES / tma_info_clks",
+ "MetricExpr": "DSB2MITE_SWITCHES.PENALTY_CYCLES / tma_info_thread_clks",
"MetricGroup": "DSBmiss;FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueFB",
"MetricName": "tma_dsb_switches",
"MetricThreshold": "tma_dsb_switches > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)",
- "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines. The DSB (decoded i-cache) is a Uop Cache where the front-end directly delivers Uops (micro operations) avoiding heavy x86 decoding. The DSB pipeline has shorter latency and delivered higher bandwidth than the MITE (legacy instruction decode pipeline). Switching between the two pipelines can cause penalties hence this metric measures the exposed penalty. Related metrics: tma_fetch_bandwidth, tma_info_dsb_coverage, tma_lcp",
+ "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines. The DSB (decoded i-cache) is a Uop Cache where the front-end directly delivers Uops (micro operations) avoiding heavy x86 decoding. The DSB pipeline has shorter latency and delivered higher bandwidth than the MITE (legacy instruction decode pipeline). Switching between the two pipelines can cause penalties hence this metric measures the exposed penalty. Related metrics: tma_fetch_bandwidth, tma_info_frontend_dsb_coverage, tma_lcp",
"ScaleUnit": "100%"
},
{
"BriefDescription": "This metric roughly estimates the fraction of cycles where the Data TLB (DTLB) was missed by load accesses",
- "MetricExpr": "(7 * DTLB_LOAD_MISSES.STLB_HIT + DTLB_LOAD_MISSES.WALK_DURATION) / tma_info_clks",
+ "MetricExpr": "(7 * DTLB_LOAD_MISSES.STLB_HIT + DTLB_LOAD_MISSES.WALK_DURATION) / tma_info_thread_clks",
"MetricGroup": "MemoryTLB;TopdownL4;tma_L4_group;tma_issueTLB;tma_l1_bound_group",
"MetricName": "tma_dtlb_load",
"MetricThreshold": "tma_dtlb_load > 0.1",
@@ -163,14 +163,14 @@
"MetricExpr": "tma_frontend_bound - tma_fetch_latency",
"MetricGroup": "FetchBW;Frontend;TmaL2;TopdownL2;tma_L2_group;tma_frontend_bound_group;tma_issueFB",
"MetricName": "tma_fetch_bandwidth",
- "MetricThreshold": "tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 4 > 0.35",
+ "MetricThreshold": "tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 4 > 0.35",
"MetricgroupNoGroup": "TopdownL2",
- "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend bandwidth issues. For example; inefficiencies at the instruction decoders; or restrictions for caching in the DSB (decoded uops cache) are categorized under Fetch Bandwidth. In such cases; the Frontend typically delivers suboptimal amount of uops to the Backend. Related metrics: tma_dsb_switches, tma_info_dsb_coverage, tma_lcp",
+ "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend bandwidth issues. For example; inefficiencies at the instruction decoders; or restrictions for caching in the DSB (decoded uops cache) are categorized under Fetch Bandwidth. In such cases; the Frontend typically delivers suboptimal amount of uops to the Backend. Related metrics: tma_dsb_switches, tma_info_frontend_dsb_coverage, tma_lcp",
"ScaleUnit": "100%"
},
{
"BriefDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend latency issues",
- "MetricExpr": "4 * min(CPU_CLK_UNHALTED.THREAD, IDQ_UOPS_NOT_DELIVERED.CYCLES_0_UOPS_DELIV.CORE) / tma_info_slots",
+ "MetricExpr": "4 * min(CPU_CLK_UNHALTED.THREAD, IDQ_UOPS_NOT_DELIVERED.CYCLES_0_UOPS_DELIV.CORE) / tma_info_thread_slots",
"MetricGroup": "Frontend;TmaL2;TopdownL2;tma_L2_group;tma_frontend_bound_group",
"MetricName": "tma_fetch_latency",
"MetricThreshold": "tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15",
@@ -207,7 +207,7 @@
},
{
"BriefDescription": "This category represents fraction of slots where the processor's Frontend undersupplies its Backend",
- "MetricExpr": "IDQ_UOPS_NOT_DELIVERED.CORE / tma_info_slots",
+ "MetricExpr": "IDQ_UOPS_NOT_DELIVERED.CORE / tma_info_thread_slots",
"MetricGroup": "PGO;TmaL1;TopdownL1;tma_L1_group",
"MetricName": "tma_frontend_bound",
"MetricThreshold": "tma_frontend_bound > 0.15",
@@ -225,170 +225,170 @@
"PublicDescription": "This metric represents fraction of slots where the CPU was retiring heavy-weight operations -- instructions that require two or more uops or micro-coded sequences. This highly-correlates with the uop length of these instructions/sequences.",
"ScaleUnit": "100%"
},
- {
- "BriefDescription": "Measured Average Frequency for unhalted processors [GHz]",
- "MetricExpr": "tma_info_turbo_utilization * TSC / 1e9 / duration_time",
- "MetricGroup": "Power;Summary",
- "MetricName": "tma_info_average_frequency"
- },
- {
- "BriefDescription": "Per-Logical Processor actual clocks when the Logical Processor is active.",
- "MetricExpr": "CPU_CLK_UNHALTED.THREAD",
- "MetricGroup": "Pipeline",
- "MetricName": "tma_info_clks"
- },
{
"BriefDescription": "Core actual clocks when any Logical Processor is active on the Physical Core",
- "MetricExpr": "(CPU_CLK_UNHALTED.THREAD / 2 * (1 + CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / CPU_CLK_UNHALTED.REF_XCLK) if #core_wide < 1 else (CPU_CLK_UNHALTED.THREAD_ANY / 2 if #SMT_on else tma_info_clks))",
+ "MetricExpr": "(CPU_CLK_UNHALTED.THREAD / 2 * (1 + CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / CPU_CLK_UNHALTED.REF_XCLK) if #core_wide < 1 else (CPU_CLK_UNHALTED.THREAD_ANY / 2 if #SMT_on else tma_info_thread_clks))",
"MetricGroup": "SMT",
- "MetricName": "tma_info_core_clks"
+ "MetricName": "tma_info_core_core_clks"
},
{
"BriefDescription": "Instructions Per Cycle across hyper-threads (per physical core)",
- "MetricExpr": "INST_RETIRED.ANY / tma_info_core_clks",
+ "MetricExpr": "INST_RETIRED.ANY / tma_info_core_core_clks",
"MetricGroup": "Ret;SMT;TmaL1;tma_L1_group",
- "MetricName": "tma_info_coreipc"
- },
- {
- "BriefDescription": "Cycles Per Instruction (per Logical Processor)",
- "MetricExpr": "1 / tma_info_ipc",
- "MetricGroup": "Mem;Pipeline",
- "MetricName": "tma_info_cpi"
+ "MetricName": "tma_info_core_coreipc"
},
{
- "BriefDescription": "Average CPU Utilization",
- "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC / TSC",
- "MetricGroup": "HPC;Summary",
- "MetricName": "tma_info_cpu_utilization"
+ "BriefDescription": "Floating Point Operations Per Cycle",
+ "MetricExpr": "(FP_COMP_OPS_EXE.SSE_SCALAR_SINGLE + FP_COMP_OPS_EXE.SSE_SCALAR_DOUBLE + 2 * FP_COMP_OPS_EXE.SSE_PACKED_DOUBLE + 4 * (FP_COMP_OPS_EXE.SSE_PACKED_SINGLE + SIMD_FP_256.PACKED_DOUBLE) + 8 * SIMD_FP_256.PACKED_SINGLE) / tma_info_core_core_clks",
+ "MetricGroup": "Flops;Ret",
+ "MetricName": "tma_info_core_flopc"
},
{
- "BriefDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]",
- "MetricExpr": "64 * (UNC_M_CAS_COUNT.RD + UNC_M_CAS_COUNT.WR) / 1e9 / duration_time",
- "MetricGroup": "HPC;Mem;MemoryBW;SoC;tma_issueBW",
- "MetricName": "tma_info_dram_bw_use",
- "PublicDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]. Related metrics: tma_mem_bandwidth"
+ "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-core",
+ "MetricExpr": "UOPS_DISPATCHED.THREAD / (cpu@UOPS_DISPATCHED.CORE\\,cmask\\=1@ / 2 if #SMT_on else cpu@UOPS_DISPATCHED.CORE\\,cmask\\=1@)",
+ "MetricGroup": "Backend;Cor;Pipeline;PortsUtil",
+ "MetricName": "tma_info_core_ilp"
},
{
"BriefDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache)",
"MetricExpr": "IDQ.DSB_UOPS / (IDQ.DSB_UOPS + LSD.UOPS + IDQ.MITE_UOPS + IDQ.MS_UOPS)",
"MetricGroup": "DSB;Fed;FetchBW;tma_issueFB",
- "MetricName": "tma_info_dsb_coverage",
- "MetricThreshold": "tma_info_dsb_coverage < 0.7 & tma_info_ipc / 4 > 0.35",
+ "MetricName": "tma_info_frontend_dsb_coverage",
+ "MetricThreshold": "tma_info_frontend_dsb_coverage < 0.7 & tma_info_thread_ipc / 4 > 0.35",
"PublicDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache). Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_lcp"
},
{
- "BriefDescription": "The ratio of Executed- by Issued-Uops",
- "MetricExpr": "UOPS_DISPATCHED.THREAD / UOPS_ISSUED.ANY",
- "MetricGroup": "Cor;Pipeline",
- "MetricName": "tma_info_execute_per_issue",
- "PublicDescription": "The ratio of Executed- by Issued-Uops. Ratio > 1 suggests high rate of uop micro-fusions. Ratio < 1 suggest high rate of \"execute\" at rename stage."
+ "BriefDescription": "Total number of retired Instructions",
+ "MetricExpr": "INST_RETIRED.ANY",
+ "MetricGroup": "Summary;TmaL1;tma_L1_group",
+ "MetricName": "tma_info_inst_mix_instructions",
+ "PublicDescription": "Total number of retired Instructions. Sample with: INST_RETIRED.PREC_DIST"
},
{
- "BriefDescription": "Floating Point Operations Per Cycle",
- "MetricExpr": "(FP_COMP_OPS_EXE.SSE_SCALAR_SINGLE + FP_COMP_OPS_EXE.SSE_SCALAR_DOUBLE + 2 * FP_COMP_OPS_EXE.SSE_PACKED_DOUBLE + 4 * (FP_COMP_OPS_EXE.SSE_PACKED_SINGLE + SIMD_FP_256.PACKED_DOUBLE) + 8 * SIMD_FP_256.PACKED_SINGLE) / tma_info_core_clks",
- "MetricGroup": "Flops;Ret",
- "MetricName": "tma_info_flopc"
+ "BriefDescription": "Average number of Uops retired in cycles where at least one uop has retired.",
+ "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / cpu@UOPS_RETIRED.RETIRE_SLOTS\\,cmask\\=1@",
+ "MetricGroup": "Pipeline;Ret",
+ "MetricName": "tma_info_pipeline_retire"
},
{
- "BriefDescription": "Giga Floating Point Operations Per Second",
- "MetricExpr": "(FP_COMP_OPS_EXE.SSE_SCALAR_SINGLE + FP_COMP_OPS_EXE.SSE_SCALAR_DOUBLE + 2 * FP_COMP_OPS_EXE.SSE_PACKED_DOUBLE + 4 * (FP_COMP_OPS_EXE.SSE_PACKED_SINGLE + SIMD_FP_256.PACKED_DOUBLE) + 8 * SIMD_FP_256.PACKED_SINGLE) / 1e9 / duration_time",
- "MetricGroup": "Cor;Flops;HPC",
- "MetricName": "tma_info_gflops",
- "PublicDescription": "Giga Floating Point Operations Per Second. Aggregate across all supported options of: FP precisions, scalar and vector instructions, vector-width and AMX engine."
+ "BriefDescription": "Measured Average Frequency for unhalted processors [GHz]",
+ "MetricExpr": "tma_info_system_turbo_utilization * TSC / 1e9 / duration_time",
+ "MetricGroup": "Power;Summary",
+ "MetricName": "tma_info_system_average_frequency"
},
{
- "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-core",
- "MetricExpr": "UOPS_DISPATCHED.THREAD / (cpu@UOPS_DISPATCHED.CORE\\,cmask\\=1@ / 2 if #SMT_on else cpu@UOPS_DISPATCHED.CORE\\,cmask\\=1@)",
- "MetricGroup": "Backend;Cor;Pipeline;PortsUtil",
- "MetricName": "tma_info_ilp"
+ "BriefDescription": "Average CPU Utilization",
+ "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC / TSC",
+ "MetricGroup": "HPC;Summary",
+ "MetricName": "tma_info_system_cpu_utilization"
},
{
- "BriefDescription": "Total number of retired Instructions",
- "MetricExpr": "INST_RETIRED.ANY",
- "MetricGroup": "Summary;TmaL1;tma_L1_group",
- "MetricName": "tma_info_instructions",
- "PublicDescription": "Total number of retired Instructions. Sample with: INST_RETIRED.PREC_DIST"
+ "BriefDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]",
+ "MetricExpr": "64 * (UNC_M_CAS_COUNT.RD + UNC_M_CAS_COUNT.WR) / 1e9 / duration_time",
+ "MetricGroup": "HPC;Mem;MemoryBW;SoC;tma_issueBW",
+ "MetricName": "tma_info_system_dram_bw_use",
+ "PublicDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]. Related metrics: tma_mem_bandwidth"
},
{
- "BriefDescription": "Instructions Per Cycle (per Logical Processor)",
- "MetricExpr": "INST_RETIRED.ANY / tma_info_clks",
- "MetricGroup": "Ret;Summary",
- "MetricName": "tma_info_ipc"
+ "BriefDescription": "Giga Floating Point Operations Per Second",
+ "MetricExpr": "(FP_COMP_OPS_EXE.SSE_SCALAR_SINGLE + FP_COMP_OPS_EXE.SSE_SCALAR_DOUBLE + 2 * FP_COMP_OPS_EXE.SSE_PACKED_DOUBLE + 4 * (FP_COMP_OPS_EXE.SSE_PACKED_SINGLE + SIMD_FP_256.PACKED_DOUBLE) + 8 * SIMD_FP_256.PACKED_SINGLE) / 1e9 / duration_time",
+ "MetricGroup": "Cor;Flops;HPC",
+ "MetricName": "tma_info_system_gflops",
+ "PublicDescription": "Giga Floating Point Operations Per Second. Aggregate across all supported options of: FP precisions, scalar and vector instructions, vector-width and AMX engine."
},
{
"BriefDescription": "Instructions per Far Branch ( Far Branches apply upon transition from application to operating system, handling interrupts, exceptions) [lower number means higher occurrence rate]",
"MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.FAR_BRANCH:u",
"MetricGroup": "Branches;OS",
- "MetricName": "tma_info_ipfarbranch",
- "MetricThreshold": "tma_info_ipfarbranch < 1e6"
+ "MetricName": "tma_info_system_ipfarbranch",
+ "MetricThreshold": "tma_info_system_ipfarbranch < 1e6"
},
{
"BriefDescription": "Cycles Per Instruction for the Operating System (OS) Kernel mode",
"MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / INST_RETIRED.ANY_P:k",
"MetricGroup": "OS",
- "MetricName": "tma_info_kernel_cpi"
+ "MetricName": "tma_info_system_kernel_cpi"
},
{
"BriefDescription": "Fraction of cycles spent in the Operating System (OS) Kernel mode",
"MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / CPU_CLK_UNHALTED.THREAD",
"MetricGroup": "OS",
- "MetricName": "tma_info_kernel_utilization",
- "MetricThreshold": "tma_info_kernel_utilization > 0.05"
+ "MetricName": "tma_info_system_kernel_utilization",
+ "MetricThreshold": "tma_info_system_kernel_utilization > 0.05"
},
{
"BriefDescription": "Average number of parallel data read requests to external memory",
"MetricExpr": "UNC_C_TOR_OCCUPANCY.MISS_OPCODE@filter_opc\\=0x182@ / UNC_C_TOR_OCCUPANCY.MISS_OPCODE@filter_opc\\=0x182\\,thresh\\=1@",
"MetricGroup": "Mem;MemoryBW;SoC",
- "MetricName": "tma_info_mem_parallel_reads",
+ "MetricName": "tma_info_system_mem_parallel_reads",
"PublicDescription": "Average number of parallel data read requests to external memory. Accounts for demand loads and L1/L2 prefetches"
},
{
"BriefDescription": "Average latency of data read request to external memory (in nanoseconds)",
- "MetricExpr": "1e9 * (UNC_C_TOR_OCCUPANCY.MISS_OPCODE@filter_opc\\=0x182@ / UNC_C_TOR_INSERTS.MISS_OPCODE@filter_opc\\=0x182@) / (tma_info_socket_clks / duration_time)",
+ "MetricExpr": "1e9 * (UNC_C_TOR_OCCUPANCY.MISS_OPCODE@filter_opc\\=0x182@ / UNC_C_TOR_INSERTS.MISS_OPCODE@filter_opc\\=0x182@) / (tma_info_system_socket_clks / duration_time)",
"MetricGroup": "Mem;MemoryLat;SoC",
- "MetricName": "tma_info_mem_read_latency",
+ "MetricName": "tma_info_system_mem_read_latency",
"PublicDescription": "Average latency of data read request to external memory (in nanoseconds). Accounts for demand loads and L1/L2 prefetches. ([RKL+]memory-controller only)"
},
- {
- "BriefDescription": "Average number of Uops retired in cycles where at least one uop has retired.",
- "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / cpu@UOPS_RETIRED.RETIRE_SLOTS\\,cmask\\=1@",
- "MetricGroup": "Pipeline;Ret",
- "MetricName": "tma_info_retire"
- },
- {
- "BriefDescription": "Total issue-pipeline slots (per-Physical Core till ICL; per-Logical Processor ICL onward)",
- "MetricExpr": "4 * tma_info_core_clks",
- "MetricGroup": "TmaL1;tma_L1_group",
- "MetricName": "tma_info_slots"
- },
{
"BriefDescription": "Fraction of cycles where both hardware Logical Processors were active",
"MetricExpr": "(1 - CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / (CPU_CLK_UNHALTED.REF_XCLK_ANY / 2) if #SMT_on else 0)",
"MetricGroup": "SMT",
- "MetricName": "tma_info_smt_2t_utilization"
+ "MetricName": "tma_info_system_smt_2t_utilization"
},
{
"BriefDescription": "Socket actual clocks when any core is active on that socket",
"MetricExpr": "cbox_0@event\\=0x0@",
"MetricGroup": "SoC",
- "MetricName": "tma_info_socket_clks"
+ "MetricName": "tma_info_system_socket_clks"
},
{
"BriefDescription": "Average Frequency Utilization relative nominal frequency",
- "MetricExpr": "tma_info_clks / CPU_CLK_UNHALTED.REF_TSC",
+ "MetricExpr": "tma_info_thread_clks / CPU_CLK_UNHALTED.REF_TSC",
"MetricGroup": "Power",
- "MetricName": "tma_info_turbo_utilization"
+ "MetricName": "tma_info_system_turbo_utilization"
+ },
+ {
+ "BriefDescription": "Per-Logical Processor actual clocks when the Logical Processor is active.",
+ "MetricExpr": "CPU_CLK_UNHALTED.THREAD",
+ "MetricGroup": "Pipeline",
+ "MetricName": "tma_info_thread_clks"
+ },
+ {
+ "BriefDescription": "Cycles Per Instruction (per Logical Processor)",
+ "MetricExpr": "1 / tma_info_thread_ipc",
+ "MetricGroup": "Mem;Pipeline",
+ "MetricName": "tma_info_thread_cpi"
+ },
+ {
+ "BriefDescription": "The ratio of Executed- by Issued-Uops",
+ "MetricExpr": "UOPS_DISPATCHED.THREAD / UOPS_ISSUED.ANY",
+ "MetricGroup": "Cor;Pipeline",
+ "MetricName": "tma_info_thread_execute_per_issue",
+ "PublicDescription": "The ratio of Executed- by Issued-Uops. Ratio > 1 suggests high rate of uop micro-fusions. Ratio < 1 suggest high rate of \"execute\" at rename stage."
+ },
+ {
+ "BriefDescription": "Instructions Per Cycle (per Logical Processor)",
+ "MetricExpr": "INST_RETIRED.ANY / tma_info_thread_clks",
+ "MetricGroup": "Ret;Summary",
+ "MetricName": "tma_info_thread_ipc"
+ },
+ {
+ "BriefDescription": "Total issue-pipeline slots (per-Physical Core till ICL; per-Logical Processor ICL onward)",
+ "MetricExpr": "4 * tma_info_core_core_clks",
+ "MetricGroup": "TmaL1;tma_L1_group",
+ "MetricName": "tma_info_thread_slots"
},
{
"BriefDescription": "Uops Per Instruction",
"MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / INST_RETIRED.ANY",
"MetricGroup": "Pipeline;Ret;Retire",
- "MetricName": "tma_info_uoppi",
- "MetricThreshold": "tma_info_uoppi > 1.05"
+ "MetricName": "tma_info_thread_uoppi",
+ "MetricThreshold": "tma_info_thread_uoppi > 1.05"
},
{
"BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Instruction TLB (ITLB) misses",
- "MetricExpr": "(12 * ITLB_MISSES.STLB_HIT + ITLB_MISSES.WALK_DURATION) / tma_info_clks",
+ "MetricExpr": "(12 * ITLB_MISSES.STLB_HIT + ITLB_MISSES.WALK_DURATION) / tma_info_thread_clks",
"MetricGroup": "BigFoot;FetchLat;MemoryTLB;TopdownL3;tma_L3_group;tma_fetch_latency_group",
"MetricName": "tma_itlb_misses",
"MetricThreshold": "tma_itlb_misses > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)",
@@ -398,7 +398,7 @@
{
"BriefDescription": "This metric estimates how often the CPU was stalled due to loads accesses to L3 cache or contended with a sibling Core",
"MetricConstraint": "NO_GROUP_EVENTS_SMT",
- "MetricExpr": "MEM_LOAD_UOPS_RETIRED.LLC_HIT / (MEM_LOAD_UOPS_RETIRED.LLC_HIT + 7 * MEM_LOAD_UOPS_RETIRED.LLC_MISS) * CYCLE_ACTIVITY.STALLS_L2_PENDING / tma_info_clks",
+ "MetricExpr": "MEM_LOAD_UOPS_RETIRED.LLC_HIT / (MEM_LOAD_UOPS_RETIRED.LLC_HIT + 7 * MEM_LOAD_UOPS_RETIRED.LLC_MISS) * CYCLE_ACTIVITY.STALLS_L2_PENDING / tma_info_thread_clks",
"MetricGroup": "CacheMisses;MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group",
"MetricName": "tma_l3_bound",
"MetricThreshold": "tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)",
@@ -407,11 +407,11 @@
},
{
"BriefDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs)",
- "MetricExpr": "ILD_STALL.LCP / tma_info_clks",
+ "MetricExpr": "ILD_STALL.LCP / tma_info_thread_clks",
"MetricGroup": "FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueFB",
"MetricName": "tma_lcp",
"MetricThreshold": "tma_lcp > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)",
- "PublicDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs). Using proper compiler flags or Intel Compiler by default will certainly avoid this. #Link: Optimization Guide about LCP BKMs. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_dsb_coverage",
+ "PublicDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs). Using proper compiler flags or Intel Compiler by default will certainly avoid this. #Link: Optimization Guide about LCP BKMs. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_frontend_dsb_coverage",
"ScaleUnit": "100%"
},
{
@@ -437,16 +437,16 @@
},
{
"BriefDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM)",
- "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, cpu@OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD\\,cmask\\=6@) / tma_info_clks",
+ "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, cpu@OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD\\,cmask\\=6@) / tma_info_thread_clks",
"MetricGroup": "MemoryBW;Offcore;TopdownL4;tma_L4_group;tma_dram_bound_group;tma_issueBW",
"MetricName": "tma_mem_bandwidth",
"MetricThreshold": "tma_mem_bandwidth > 0.2 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))",
- "PublicDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM). The underlying heuristic assumes that a similar off-core traffic is generated by all IA cores. This metric does not aggregate non-data-read requests by this logical processor; requests from other IA Logical Processors/Physical Cores/sockets; or other non-IA devices like GPU; hence the maximum external memory bandwidth limits may or may not be approached when this metric is flagged (see Uncore counters for that). Related metrics: tma_info_dram_bw_use",
+ "PublicDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM). The underlying heuristic assumes that a similar off-core traffic is generated by all IA cores. This metric does not aggregate non-data-read requests by this logical processor; requests from other IA Logical Processors/Physical Cores/sockets; or other non-IA devices like GPU; hence the maximum external memory bandwidth limits may or may not be approached when this metric is flagged (see Uncore counters for that). Related metrics: tma_info_system_dram_bw_use",
"ScaleUnit": "100%"
},
{
"BriefDescription": "This metric estimates fraction of cycles where the performance was likely hurt due to latency from external memory (DRAM)",
- "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD) / tma_info_clks - tma_mem_bandwidth",
+ "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD) / tma_info_thread_clks - tma_mem_bandwidth",
"MetricGroup": "MemoryLat;Offcore;TopdownL4;tma_L4_group;tma_dram_bound_group;tma_issueLat",
"MetricName": "tma_mem_latency",
"MetricThreshold": "tma_mem_latency > 0.1 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))",
@@ -456,7 +456,7 @@
{
"BriefDescription": "This metric represents fraction of slots the Memory subsystem within the Backend was a bottleneck",
"MetricConstraint": "NO_GROUP_EVENTS",
- "MetricExpr": "(min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.STALLS_L1D_PENDING) + RESOURCE_STALLS.SB) / (min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.CYCLES_NO_DISPATCH) + cpu@UOPS_DISPATCHED.THREAD\\,cmask\\=1@ - (cpu@UOPS_DISPATCHED.THREAD\\,cmask\\=3@ if tma_info_ipc > 1.8 else cpu@UOPS_DISPATCHED.THREAD\\,cmask\\=2@) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB) * tma_backend_bound",
+ "MetricExpr": "(min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.STALLS_L1D_PENDING) + RESOURCE_STALLS.SB) / (min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.CYCLES_NO_DISPATCH) + cpu@UOPS_DISPATCHED.THREAD\\,cmask\\=1@ - (cpu@UOPS_DISPATCHED.THREAD\\,cmask\\=3@ if tma_info_thread_ipc > 1.8 else cpu@UOPS_DISPATCHED.THREAD\\,cmask\\=2@) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB) * tma_backend_bound",
"MetricGroup": "Backend;TmaL2;TopdownL2;tma_L2_group;tma_backend_bound_group",
"MetricName": "tma_memory_bound",
"MetricThreshold": "tma_memory_bound > 0.2 & tma_backend_bound > 0.2",
@@ -466,7 +466,7 @@
},
{
"BriefDescription": "This metric represents fraction of slots the CPU was retiring uops fetched by the Microcode Sequencer (MS) unit",
- "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / UOPS_ISSUED.ANY * IDQ.MS_UOPS / tma_info_slots",
+ "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / UOPS_ISSUED.ANY * IDQ.MS_UOPS / tma_info_thread_slots",
"MetricGroup": "MicroSeq;TopdownL3;tma_L3_group;tma_heavy_operations_group;tma_issueMC;tma_issueMS",
"MetricName": "tma_microcode_sequencer",
"MetricThreshold": "tma_microcode_sequencer > 0.05 & tma_heavy_operations > 0.1",
@@ -475,7 +475,7 @@
},
{
"BriefDescription": "This metric estimates the fraction of cycles when the CPU was stalled due to switches of uop delivery to the Microcode Sequencer (MS)",
- "MetricExpr": "3 * IDQ.MS_SWITCHES / tma_info_clks",
+ "MetricExpr": "3 * IDQ.MS_SWITCHES / tma_info_thread_clks",
"MetricGroup": "FetchLat;MicroSeq;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueMC;tma_issueMS;tma_issueMV;tma_issueSO",
"MetricName": "tma_ms_switches",
"MetricThreshold": "tma_ms_switches > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)",
@@ -485,7 +485,7 @@
{
"BriefDescription": "This metric estimates fraction of cycles the CPU performance was potentially limited due to Core computation issues (non divider-related)",
"MetricConstraint": "NO_GROUP_EVENTS",
- "MetricExpr": "(min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.CYCLES_NO_DISPATCH) + cpu@UOPS_DISPATCHED.THREAD\\,cmask\\=1@ - (cpu@UOPS_DISPATCHED.THREAD\\,cmask\\=3@ if tma_info_ipc > 1.8 else cpu@UOPS_DISPATCHED.THREAD\\,cmask\\=2@) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB - RESOURCE_STALLS.SB - min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.STALLS_L1D_PENDING)) / tma_info_clks",
+ "MetricExpr": "(min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.CYCLES_NO_DISPATCH) + cpu@UOPS_DISPATCHED.THREAD\\,cmask\\=1@ - (cpu@UOPS_DISPATCHED.THREAD\\,cmask\\=3@ if tma_info_thread_ipc > 1.8 else cpu@UOPS_DISPATCHED.THREAD\\,cmask\\=2@) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB - RESOURCE_STALLS.SB - min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.STALLS_L1D_PENDING)) / tma_info_thread_clks",
"MetricGroup": "PortsUtil;TopdownL3;tma_L3_group;tma_core_bound_group",
"MetricName": "tma_ports_utilization",
"MetricThreshold": "tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2)",
@@ -494,7 +494,7 @@
},
{
"BriefDescription": "This category represents fraction of slots utilized by useful work i.e. issued uops that eventually get retired",
- "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / tma_info_slots",
+ "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / tma_info_thread_slots",
"MetricGroup": "TmaL1;TopdownL1;tma_L1_group",
"MetricName": "tma_retiring",
"MetricThreshold": "tma_retiring > 0.7 | tma_heavy_operations > 0.1",
@@ -504,7 +504,7 @@
},
{
"BriefDescription": "This metric estimates how often CPU was stalled due to RFO store memory accesses; RFO store issue a read-for-ownership request before the write",
- "MetricExpr": "RESOURCE_STALLS.SB / tma_info_clks",
+ "MetricExpr": "RESOURCE_STALLS.SB / tma_info_thread_clks",
"MetricGroup": "MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group",
"MetricName": "tma_store_bound",
"MetricThreshold": "tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)",
--
2.40.1.606.ga4b1b128d6-goog
^ permalink raw reply related [flat|nested] 9+ messages in thread* [PATCH v1 09/15] perf vendor events intel: Update sandybridge metrics
2023-05-15 21:58 [PATCH v1 00/15] Event, metric and metric group improvements Ian Rogers
2023-05-15 21:58 ` [PATCH v1 04/15] perf vendor events intel: Update elkhartlake events Ian Rogers
2023-05-15 21:58 ` [PATCH v1 08/15] perf vendor events intel: Update jaketown metrics Ian Rogers
@ 2023-05-15 21:58 ` Ian Rogers
2023-05-15 21:58 ` [PATCH v1 12/15] perf vendor events intel: Update snowridgex events Ian Rogers
` (2 subsequent siblings)
5 siblings, 0 replies; 9+ messages in thread
From: Ian Rogers @ 2023-05-15 21:58 UTC (permalink / raw)
To: Peter Zijlstra, Ingo Molnar, Arnaldo Carvalho de Melo,
Mark Rutland, Alexander Shishkin, Jiri Olsa, Namhyung Kim,
Ian Rogers, Adrian Hunter, Kan Liang, Zhengjun Xing, John Garry,
Kajol Jain, Thomas Richter, linux-kernel, linux-perf-users
Metrics are updated to make TMA info metric names
synchronized. Metrics were generated by:
https://github.com/intel/perfmon/blob/main/scripts/create_perf_json.py
Signed-off-by: Ian Rogers <irogers@google.com>
---
.../arch/x86/sandybridge/snb-metrics.json | 222 +++++++++---------
1 file changed, 111 insertions(+), 111 deletions(-)
diff --git a/tools/perf/pmu-events/arch/x86/sandybridge/snb-metrics.json b/tools/perf/pmu-events/arch/x86/sandybridge/snb-metrics.json
index 4b8bc19392a4..8898b6fd0dea 100644
--- a/tools/perf/pmu-events/arch/x86/sandybridge/snb-metrics.json
+++ b/tools/perf/pmu-events/arch/x86/sandybridge/snb-metrics.json
@@ -50,7 +50,7 @@
},
{
"BriefDescription": "Uncore frequency per die [GHZ]",
- "MetricExpr": "tma_info_socket_clks / #num_dies / duration_time / 1e9",
+ "MetricExpr": "tma_info_system_socket_clks / #num_dies / duration_time / 1e9",
"MetricGroup": "SoC",
"MetricName": "UNCORE_FREQ"
},
@@ -82,7 +82,7 @@
},
{
"BriefDescription": "This category represents fraction of slots wasted due to incorrect speculations",
- "MetricExpr": "(UOPS_ISSUED.ANY - UOPS_RETIRED.RETIRE_SLOTS + 4 * (INT_MISC.RECOVERY_CYCLES_ANY / 2 if #SMT_on else INT_MISC.RECOVERY_CYCLES)) / tma_info_slots",
+ "MetricExpr": "(UOPS_ISSUED.ANY - UOPS_RETIRED.RETIRE_SLOTS + 4 * (INT_MISC.RECOVERY_CYCLES_ANY / 2 if #SMT_on else INT_MISC.RECOVERY_CYCLES)) / tma_info_thread_slots",
"MetricGroup": "TmaL1;TopdownL1;tma_L1_group",
"MetricName": "tma_bad_speculation",
"MetricThreshold": "tma_bad_speculation > 0.15",
@@ -98,12 +98,12 @@
"MetricName": "tma_branch_mispredicts",
"MetricThreshold": "tma_branch_mispredicts > 0.1 & tma_bad_speculation > 0.15",
"MetricgroupNoGroup": "TopdownL2",
- "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Branch Misprediction. These slots are either wasted by uops fetched from an incorrectly speculated program path; or stalls when the out-of-order part of the machine needs to recover its state from a speculative path. Sample with: BR_MISP_RETIRED.ALL_BRANCHES. Related metrics: tma_info_branch_misprediction_cost, tma_mispredicts_resteers",
+ "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Branch Misprediction. These slots are either wasted by uops fetched from an incorrectly speculated program path; or stalls when the out-of-order part of the machine needs to recover its state from a speculative path. Sample with: BR_MISP_RETIRED.ALL_BRANCHES. Related metrics: tma_info_bad_spec_branch_misprediction_cost, tma_mispredicts_resteers",
"ScaleUnit": "100%"
},
{
"BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers",
- "MetricExpr": "12 * (BR_MISP_RETIRED.ALL_BRANCHES + MACHINE_CLEARS.COUNT + BACLEARS.ANY) / tma_info_clks",
+ "MetricExpr": "12 * (BR_MISP_RETIRED.ALL_BRANCHES + MACHINE_CLEARS.COUNT + BACLEARS.ANY) / tma_info_thread_clks",
"MetricGroup": "FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group",
"MetricName": "tma_branch_resteers",
"MetricThreshold": "tma_branch_resteers > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)",
@@ -123,7 +123,7 @@
},
{
"BriefDescription": "This metric represents fraction of cycles where the Divider unit was active",
- "MetricExpr": "ARITH.FPU_DIV_ACTIVE / tma_info_core_clks",
+ "MetricExpr": "ARITH.FPU_DIV_ACTIVE / tma_info_core_core_clks",
"MetricGroup": "TopdownL3;tma_L3_group;tma_core_bound_group",
"MetricName": "tma_divider",
"MetricThreshold": "tma_divider > 0.2 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2)",
@@ -133,7 +133,7 @@
{
"BriefDescription": "This metric estimates how often the CPU was stalled on accesses to external memory (DRAM) by loads",
"MetricConstraint": "NO_GROUP_EVENTS_SMT",
- "MetricExpr": "(1 - MEM_LOAD_UOPS_RETIRED.LLC_HIT / (MEM_LOAD_UOPS_RETIRED.LLC_HIT + 7 * MEM_LOAD_UOPS_MISC_RETIRED.LLC_MISS)) * CYCLE_ACTIVITY.STALLS_L2_PENDING / tma_info_clks",
+ "MetricExpr": "(1 - MEM_LOAD_UOPS_RETIRED.LLC_HIT / (MEM_LOAD_UOPS_RETIRED.LLC_HIT + 7 * MEM_LOAD_UOPS_MISC_RETIRED.LLC_MISS)) * CYCLE_ACTIVITY.STALLS_L2_PENDING / tma_info_thread_clks",
"MetricGroup": "MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group",
"MetricName": "tma_dram_bound",
"MetricThreshold": "tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)",
@@ -142,16 +142,16 @@
},
{
"BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines",
- "MetricExpr": "DSB2MITE_SWITCHES.PENALTY_CYCLES / tma_info_clks",
+ "MetricExpr": "DSB2MITE_SWITCHES.PENALTY_CYCLES / tma_info_thread_clks",
"MetricGroup": "DSBmiss;FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueFB",
"MetricName": "tma_dsb_switches",
"MetricThreshold": "tma_dsb_switches > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)",
- "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines. The DSB (decoded i-cache) is a Uop Cache where the front-end directly delivers Uops (micro operations) avoiding heavy x86 decoding. The DSB pipeline has shorter latency and delivered higher bandwidth than the MITE (legacy instruction decode pipeline). Switching between the two pipelines can cause penalties hence this metric measures the exposed penalty. Related metrics: tma_fetch_bandwidth, tma_info_dsb_coverage, tma_lcp",
+ "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines. The DSB (decoded i-cache) is a Uop Cache where the front-end directly delivers Uops (micro operations) avoiding heavy x86 decoding. The DSB pipeline has shorter latency and delivered higher bandwidth than the MITE (legacy instruction decode pipeline). Switching between the two pipelines can cause penalties hence this metric measures the exposed penalty. Related metrics: tma_fetch_bandwidth, tma_info_frontend_dsb_coverage, tma_lcp",
"ScaleUnit": "100%"
},
{
"BriefDescription": "This metric roughly estimates the fraction of cycles where the Data TLB (DTLB) was missed by load accesses",
- "MetricExpr": "(7 * DTLB_LOAD_MISSES.STLB_HIT + DTLB_LOAD_MISSES.WALK_DURATION) / tma_info_clks",
+ "MetricExpr": "(7 * DTLB_LOAD_MISSES.STLB_HIT + DTLB_LOAD_MISSES.WALK_DURATION) / tma_info_thread_clks",
"MetricGroup": "MemoryTLB;TopdownL4;tma_L4_group;tma_issueTLB;tma_l1_bound_group",
"MetricName": "tma_dtlb_load",
"MetricThreshold": "tma_dtlb_load > 0.1",
@@ -163,14 +163,14 @@
"MetricExpr": "tma_frontend_bound - tma_fetch_latency",
"MetricGroup": "FetchBW;Frontend;TmaL2;TopdownL2;tma_L2_group;tma_frontend_bound_group;tma_issueFB",
"MetricName": "tma_fetch_bandwidth",
- "MetricThreshold": "tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 4 > 0.35",
+ "MetricThreshold": "tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 4 > 0.35",
"MetricgroupNoGroup": "TopdownL2",
- "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend bandwidth issues. For example; inefficiencies at the instruction decoders; or restrictions for caching in the DSB (decoded uops cache) are categorized under Fetch Bandwidth. In such cases; the Frontend typically delivers suboptimal amount of uops to the Backend. Related metrics: tma_dsb_switches, tma_info_dsb_coverage, tma_lcp",
+ "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend bandwidth issues. For example; inefficiencies at the instruction decoders; or restrictions for caching in the DSB (decoded uops cache) are categorized under Fetch Bandwidth. In such cases; the Frontend typically delivers suboptimal amount of uops to the Backend. Related metrics: tma_dsb_switches, tma_info_frontend_dsb_coverage, tma_lcp",
"ScaleUnit": "100%"
},
{
"BriefDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend latency issues",
- "MetricExpr": "4 * min(CPU_CLK_UNHALTED.THREAD, IDQ_UOPS_NOT_DELIVERED.CYCLES_0_UOPS_DELIV.CORE) / tma_info_slots",
+ "MetricExpr": "4 * min(CPU_CLK_UNHALTED.THREAD, IDQ_UOPS_NOT_DELIVERED.CYCLES_0_UOPS_DELIV.CORE) / tma_info_thread_slots",
"MetricGroup": "Frontend;TmaL2;TopdownL2;tma_L2_group;tma_frontend_bound_group",
"MetricName": "tma_fetch_latency",
"MetricThreshold": "tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15",
@@ -207,7 +207,7 @@
},
{
"BriefDescription": "This category represents fraction of slots where the processor's Frontend undersupplies its Backend",
- "MetricExpr": "IDQ_UOPS_NOT_DELIVERED.CORE / tma_info_slots",
+ "MetricExpr": "IDQ_UOPS_NOT_DELIVERED.CORE / tma_info_thread_slots",
"MetricGroup": "PGO;TmaL1;TopdownL1;tma_L1_group",
"MetricName": "tma_frontend_bound",
"MetricThreshold": "tma_frontend_bound > 0.15",
@@ -225,169 +225,169 @@
"PublicDescription": "This metric represents fraction of slots where the CPU was retiring heavy-weight operations -- instructions that require two or more uops or micro-coded sequences. This highly-correlates with the uop length of these instructions/sequences.",
"ScaleUnit": "100%"
},
- {
- "BriefDescription": "Measured Average Frequency for unhalted processors [GHz]",
- "MetricExpr": "tma_info_turbo_utilization * TSC / 1e9 / duration_time",
- "MetricGroup": "Power;Summary",
- "MetricName": "tma_info_average_frequency"
- },
- {
- "BriefDescription": "Per-Logical Processor actual clocks when the Logical Processor is active.",
- "MetricExpr": "CPU_CLK_UNHALTED.THREAD",
- "MetricGroup": "Pipeline",
- "MetricName": "tma_info_clks"
- },
{
"BriefDescription": "Core actual clocks when any Logical Processor is active on the Physical Core",
- "MetricExpr": "(CPU_CLK_UNHALTED.THREAD / 2 * (1 + CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / CPU_CLK_UNHALTED.REF_XCLK) if #core_wide < 1 else (CPU_CLK_UNHALTED.THREAD_ANY / 2 if #SMT_on else tma_info_clks))",
+ "MetricExpr": "(CPU_CLK_UNHALTED.THREAD / 2 * (1 + CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / CPU_CLK_UNHALTED.REF_XCLK) if #core_wide < 1 else (CPU_CLK_UNHALTED.THREAD_ANY / 2 if #SMT_on else tma_info_thread_clks))",
"MetricGroup": "SMT",
- "MetricName": "tma_info_core_clks"
+ "MetricName": "tma_info_core_core_clks"
},
{
"BriefDescription": "Instructions Per Cycle across hyper-threads (per physical core)",
- "MetricExpr": "INST_RETIRED.ANY / tma_info_core_clks",
+ "MetricExpr": "INST_RETIRED.ANY / tma_info_core_core_clks",
"MetricGroup": "Ret;SMT;TmaL1;tma_L1_group",
- "MetricName": "tma_info_coreipc"
- },
- {
- "BriefDescription": "Cycles Per Instruction (per Logical Processor)",
- "MetricExpr": "1 / tma_info_ipc",
- "MetricGroup": "Mem;Pipeline",
- "MetricName": "tma_info_cpi"
+ "MetricName": "tma_info_core_coreipc"
},
{
- "BriefDescription": "Average CPU Utilization",
- "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC / TSC",
- "MetricGroup": "HPC;Summary",
- "MetricName": "tma_info_cpu_utilization"
+ "BriefDescription": "Floating Point Operations Per Cycle",
+ "MetricExpr": "(FP_COMP_OPS_EXE.SSE_SCALAR_SINGLE + FP_COMP_OPS_EXE.SSE_SCALAR_DOUBLE + 2 * FP_COMP_OPS_EXE.SSE_PACKED_DOUBLE + 4 * (FP_COMP_OPS_EXE.SSE_PACKED_SINGLE + SIMD_FP_256.PACKED_DOUBLE) + 8 * SIMD_FP_256.PACKED_SINGLE) / tma_info_core_core_clks",
+ "MetricGroup": "Flops;Ret",
+ "MetricName": "tma_info_core_flopc"
},
{
- "BriefDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]",
- "MetricExpr": "64 * (UNC_ARB_TRK_REQUESTS.ALL + UNC_ARB_COH_TRK_REQUESTS.ALL) / 1e6 / duration_time / 1e3",
- "MetricGroup": "HPC;Mem;MemoryBW;SoC;tma_issueBW",
- "MetricName": "tma_info_dram_bw_use",
- "PublicDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]. Related metrics: tma_mem_bandwidth"
+ "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-core",
+ "MetricExpr": "UOPS_DISPATCHED.THREAD / (cpu@UOPS_DISPATCHED.CORE\\,cmask\\=1@ / 2 if #SMT_on else cpu@UOPS_DISPATCHED.CORE\\,cmask\\=1@)",
+ "MetricGroup": "Backend;Cor;Pipeline;PortsUtil",
+ "MetricName": "tma_info_core_ilp"
},
{
"BriefDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache)",
"MetricExpr": "IDQ.DSB_UOPS / (IDQ.DSB_UOPS + LSD.UOPS + IDQ.MITE_UOPS + IDQ.MS_UOPS)",
"MetricGroup": "DSB;Fed;FetchBW;tma_issueFB",
- "MetricName": "tma_info_dsb_coverage",
- "MetricThreshold": "tma_info_dsb_coverage < 0.7 & tma_info_ipc / 4 > 0.35",
+ "MetricName": "tma_info_frontend_dsb_coverage",
+ "MetricThreshold": "tma_info_frontend_dsb_coverage < 0.7 & tma_info_thread_ipc / 4 > 0.35",
"PublicDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache). Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_lcp"
},
{
- "BriefDescription": "The ratio of Executed- by Issued-Uops",
- "MetricExpr": "UOPS_DISPATCHED.THREAD / UOPS_ISSUED.ANY",
- "MetricGroup": "Cor;Pipeline",
- "MetricName": "tma_info_execute_per_issue",
- "PublicDescription": "The ratio of Executed- by Issued-Uops. Ratio > 1 suggests high rate of uop micro-fusions. Ratio < 1 suggest high rate of \"execute\" at rename stage."
+ "BriefDescription": "Total number of retired Instructions",
+ "MetricExpr": "INST_RETIRED.ANY",
+ "MetricGroup": "Summary;TmaL1;tma_L1_group",
+ "MetricName": "tma_info_inst_mix_instructions",
+ "PublicDescription": "Total number of retired Instructions. Sample with: INST_RETIRED.PREC_DIST"
},
{
- "BriefDescription": "Floating Point Operations Per Cycle",
- "MetricExpr": "(FP_COMP_OPS_EXE.SSE_SCALAR_SINGLE + FP_COMP_OPS_EXE.SSE_SCALAR_DOUBLE + 2 * FP_COMP_OPS_EXE.SSE_PACKED_DOUBLE + 4 * (FP_COMP_OPS_EXE.SSE_PACKED_SINGLE + SIMD_FP_256.PACKED_DOUBLE) + 8 * SIMD_FP_256.PACKED_SINGLE) / tma_info_core_clks",
- "MetricGroup": "Flops;Ret",
- "MetricName": "tma_info_flopc"
+ "BriefDescription": "Average number of Uops retired in cycles where at least one uop has retired.",
+ "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / cpu@UOPS_RETIRED.RETIRE_SLOTS\\,cmask\\=1@",
+ "MetricGroup": "Pipeline;Ret",
+ "MetricName": "tma_info_pipeline_retire"
},
{
- "BriefDescription": "Giga Floating Point Operations Per Second",
- "MetricExpr": "(FP_COMP_OPS_EXE.SSE_SCALAR_SINGLE + FP_COMP_OPS_EXE.SSE_SCALAR_DOUBLE + 2 * FP_COMP_OPS_EXE.SSE_PACKED_DOUBLE + 4 * (FP_COMP_OPS_EXE.SSE_PACKED_SINGLE + SIMD_FP_256.PACKED_DOUBLE) + 8 * SIMD_FP_256.PACKED_SINGLE) / 1e9 / duration_time",
- "MetricGroup": "Cor;Flops;HPC",
- "MetricName": "tma_info_gflops",
- "PublicDescription": "Giga Floating Point Operations Per Second. Aggregate across all supported options of: FP precisions, scalar and vector instructions, vector-width and AMX engine."
+ "BriefDescription": "Measured Average Frequency for unhalted processors [GHz]",
+ "MetricExpr": "tma_info_system_turbo_utilization * TSC / 1e9 / duration_time",
+ "MetricGroup": "Power;Summary",
+ "MetricName": "tma_info_system_average_frequency"
},
{
- "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-core",
- "MetricExpr": "UOPS_DISPATCHED.THREAD / (cpu@UOPS_DISPATCHED.CORE\\,cmask\\=1@ / 2 if #SMT_on else cpu@UOPS_DISPATCHED.CORE\\,cmask\\=1@)",
- "MetricGroup": "Backend;Cor;Pipeline;PortsUtil",
- "MetricName": "tma_info_ilp"
+ "BriefDescription": "Average CPU Utilization",
+ "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC / TSC",
+ "MetricGroup": "HPC;Summary",
+ "MetricName": "tma_info_system_cpu_utilization"
},
{
- "BriefDescription": "Total number of retired Instructions",
- "MetricExpr": "INST_RETIRED.ANY",
- "MetricGroup": "Summary;TmaL1;tma_L1_group",
- "MetricName": "tma_info_instructions",
- "PublicDescription": "Total number of retired Instructions. Sample with: INST_RETIRED.PREC_DIST"
+ "BriefDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]",
+ "MetricExpr": "64 * (UNC_ARB_TRK_REQUESTS.ALL + UNC_ARB_COH_TRK_REQUESTS.ALL) / 1e6 / duration_time / 1e3",
+ "MetricGroup": "HPC;Mem;MemoryBW;SoC;tma_issueBW",
+ "MetricName": "tma_info_system_dram_bw_use",
+ "PublicDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]. Related metrics: tma_mem_bandwidth"
},
{
- "BriefDescription": "Instructions Per Cycle (per Logical Processor)",
- "MetricExpr": "INST_RETIRED.ANY / tma_info_clks",
- "MetricGroup": "Ret;Summary",
- "MetricName": "tma_info_ipc"
+ "BriefDescription": "Giga Floating Point Operations Per Second",
+ "MetricExpr": "(FP_COMP_OPS_EXE.SSE_SCALAR_SINGLE + FP_COMP_OPS_EXE.SSE_SCALAR_DOUBLE + 2 * FP_COMP_OPS_EXE.SSE_PACKED_DOUBLE + 4 * (FP_COMP_OPS_EXE.SSE_PACKED_SINGLE + SIMD_FP_256.PACKED_DOUBLE) + 8 * SIMD_FP_256.PACKED_SINGLE) / 1e9 / duration_time",
+ "MetricGroup": "Cor;Flops;HPC",
+ "MetricName": "tma_info_system_gflops",
+ "PublicDescription": "Giga Floating Point Operations Per Second. Aggregate across all supported options of: FP precisions, scalar and vector instructions, vector-width and AMX engine."
},
{
"BriefDescription": "Instructions per Far Branch ( Far Branches apply upon transition from application to operating system, handling interrupts, exceptions) [lower number means higher occurrence rate]",
"MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.FAR_BRANCH:u",
"MetricGroup": "Branches;OS",
- "MetricName": "tma_info_ipfarbranch",
- "MetricThreshold": "tma_info_ipfarbranch < 1e6"
+ "MetricName": "tma_info_system_ipfarbranch",
+ "MetricThreshold": "tma_info_system_ipfarbranch < 1e6"
},
{
"BriefDescription": "Cycles Per Instruction for the Operating System (OS) Kernel mode",
"MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / INST_RETIRED.ANY_P:k",
"MetricGroup": "OS",
- "MetricName": "tma_info_kernel_cpi"
+ "MetricName": "tma_info_system_kernel_cpi"
},
{
"BriefDescription": "Fraction of cycles spent in the Operating System (OS) Kernel mode",
"MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / CPU_CLK_UNHALTED.THREAD",
"MetricGroup": "OS",
- "MetricName": "tma_info_kernel_utilization",
- "MetricThreshold": "tma_info_kernel_utilization > 0.05"
+ "MetricName": "tma_info_system_kernel_utilization",
+ "MetricThreshold": "tma_info_system_kernel_utilization > 0.05"
},
{
"BriefDescription": "Average number of parallel requests to external memory",
"MetricExpr": "UNC_ARB_TRK_OCCUPANCY.ALL / UNC_ARB_TRK_OCCUPANCY.CYCLES_WITH_ANY_REQUEST",
"MetricGroup": "Mem;SoC",
- "MetricName": "tma_info_mem_parallel_requests",
+ "MetricName": "tma_info_system_mem_parallel_requests",
"PublicDescription": "Average number of parallel requests to external memory. Accounts for all requests"
},
{
"BriefDescription": "Average latency of all requests to external memory (in Uncore cycles)",
"MetricExpr": "UNC_ARB_TRK_OCCUPANCY.ALL / UNC_ARB_TRK_REQUESTS.ALL",
"MetricGroup": "Mem;SoC",
- "MetricName": "tma_info_mem_request_latency"
- },
- {
- "BriefDescription": "Average number of Uops retired in cycles where at least one uop has retired.",
- "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / cpu@UOPS_RETIRED.RETIRE_SLOTS\\,cmask\\=1@",
- "MetricGroup": "Pipeline;Ret",
- "MetricName": "tma_info_retire"
- },
- {
- "BriefDescription": "Total issue-pipeline slots (per-Physical Core till ICL; per-Logical Processor ICL onward)",
- "MetricExpr": "4 * tma_info_core_clks",
- "MetricGroup": "TmaL1;tma_L1_group",
- "MetricName": "tma_info_slots"
+ "MetricName": "tma_info_system_mem_request_latency"
},
{
"BriefDescription": "Fraction of cycles where both hardware Logical Processors were active",
"MetricExpr": "(1 - CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / (CPU_CLK_UNHALTED.REF_XCLK_ANY / 2) if #SMT_on else 0)",
"MetricGroup": "SMT",
- "MetricName": "tma_info_smt_2t_utilization"
+ "MetricName": "tma_info_system_smt_2t_utilization"
},
{
"BriefDescription": "Socket actual clocks when any core is active on that socket",
"MetricExpr": "UNC_CLOCK.SOCKET",
"MetricGroup": "SoC",
- "MetricName": "tma_info_socket_clks"
+ "MetricName": "tma_info_system_socket_clks"
},
{
"BriefDescription": "Average Frequency Utilization relative nominal frequency",
- "MetricExpr": "tma_info_clks / CPU_CLK_UNHALTED.REF_TSC",
+ "MetricExpr": "tma_info_thread_clks / CPU_CLK_UNHALTED.REF_TSC",
"MetricGroup": "Power",
- "MetricName": "tma_info_turbo_utilization"
+ "MetricName": "tma_info_system_turbo_utilization"
+ },
+ {
+ "BriefDescription": "Per-Logical Processor actual clocks when the Logical Processor is active.",
+ "MetricExpr": "CPU_CLK_UNHALTED.THREAD",
+ "MetricGroup": "Pipeline",
+ "MetricName": "tma_info_thread_clks"
+ },
+ {
+ "BriefDescription": "Cycles Per Instruction (per Logical Processor)",
+ "MetricExpr": "1 / tma_info_thread_ipc",
+ "MetricGroup": "Mem;Pipeline",
+ "MetricName": "tma_info_thread_cpi"
+ },
+ {
+ "BriefDescription": "The ratio of Executed- by Issued-Uops",
+ "MetricExpr": "UOPS_DISPATCHED.THREAD / UOPS_ISSUED.ANY",
+ "MetricGroup": "Cor;Pipeline",
+ "MetricName": "tma_info_thread_execute_per_issue",
+ "PublicDescription": "The ratio of Executed- by Issued-Uops. Ratio > 1 suggests high rate of uop micro-fusions. Ratio < 1 suggest high rate of \"execute\" at rename stage."
+ },
+ {
+ "BriefDescription": "Instructions Per Cycle (per Logical Processor)",
+ "MetricExpr": "INST_RETIRED.ANY / tma_info_thread_clks",
+ "MetricGroup": "Ret;Summary",
+ "MetricName": "tma_info_thread_ipc"
+ },
+ {
+ "BriefDescription": "Total issue-pipeline slots (per-Physical Core till ICL; per-Logical Processor ICL onward)",
+ "MetricExpr": "4 * tma_info_core_core_clks",
+ "MetricGroup": "TmaL1;tma_L1_group",
+ "MetricName": "tma_info_thread_slots"
},
{
"BriefDescription": "Uops Per Instruction",
"MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / INST_RETIRED.ANY",
"MetricGroup": "Pipeline;Ret;Retire",
- "MetricName": "tma_info_uoppi",
- "MetricThreshold": "tma_info_uoppi > 1.05"
+ "MetricName": "tma_info_thread_uoppi",
+ "MetricThreshold": "tma_info_thread_uoppi > 1.05"
},
{
"BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Instruction TLB (ITLB) misses",
- "MetricExpr": "(12 * ITLB_MISSES.STLB_HIT + ITLB_MISSES.WALK_DURATION) / tma_info_clks",
+ "MetricExpr": "(12 * ITLB_MISSES.STLB_HIT + ITLB_MISSES.WALK_DURATION) / tma_info_thread_clks",
"MetricGroup": "BigFoot;FetchLat;MemoryTLB;TopdownL3;tma_L3_group;tma_fetch_latency_group",
"MetricName": "tma_itlb_misses",
"MetricThreshold": "tma_itlb_misses > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)",
@@ -397,7 +397,7 @@
{
"BriefDescription": "This metric estimates how often the CPU was stalled due to loads accesses to L3 cache or contended with a sibling Core",
"MetricConstraint": "NO_GROUP_EVENTS_SMT",
- "MetricExpr": "MEM_LOAD_UOPS_RETIRED.LLC_HIT / (MEM_LOAD_UOPS_RETIRED.LLC_HIT + 7 * MEM_LOAD_UOPS_MISC_RETIRED.LLC_MISS) * CYCLE_ACTIVITY.STALLS_L2_PENDING / tma_info_clks",
+ "MetricExpr": "MEM_LOAD_UOPS_RETIRED.LLC_HIT / (MEM_LOAD_UOPS_RETIRED.LLC_HIT + 7 * MEM_LOAD_UOPS_MISC_RETIRED.LLC_MISS) * CYCLE_ACTIVITY.STALLS_L2_PENDING / tma_info_thread_clks",
"MetricGroup": "CacheMisses;MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group",
"MetricName": "tma_l3_bound",
"MetricThreshold": "tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)",
@@ -406,11 +406,11 @@
},
{
"BriefDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs)",
- "MetricExpr": "ILD_STALL.LCP / tma_info_clks",
+ "MetricExpr": "ILD_STALL.LCP / tma_info_thread_clks",
"MetricGroup": "FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueFB",
"MetricName": "tma_lcp",
"MetricThreshold": "tma_lcp > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)",
- "PublicDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs). Using proper compiler flags or Intel Compiler by default will certainly avoid this. #Link: Optimization Guide about LCP BKMs. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_dsb_coverage",
+ "PublicDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs). Using proper compiler flags or Intel Compiler by default will certainly avoid this. #Link: Optimization Guide about LCP BKMs. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_frontend_dsb_coverage",
"ScaleUnit": "100%"
},
{
@@ -436,16 +436,16 @@
},
{
"BriefDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM)",
- "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, cpu@OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD\\,cmask\\=6@) / tma_info_clks",
+ "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, cpu@OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD\\,cmask\\=6@) / tma_info_thread_clks",
"MetricGroup": "MemoryBW;Offcore;TopdownL4;tma_L4_group;tma_dram_bound_group;tma_issueBW",
"MetricName": "tma_mem_bandwidth",
"MetricThreshold": "tma_mem_bandwidth > 0.2 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))",
- "PublicDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM). The underlying heuristic assumes that a similar off-core traffic is generated by all IA cores. This metric does not aggregate non-data-read requests by this logical processor; requests from other IA Logical Processors/Physical Cores/sockets; or other non-IA devices like GPU; hence the maximum external memory bandwidth limits may or may not be approached when this metric is flagged (see Uncore counters for that). Related metrics: tma_info_dram_bw_use",
+ "PublicDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM). The underlying heuristic assumes that a similar off-core traffic is generated by all IA cores. This metric does not aggregate non-data-read requests by this logical processor; requests from other IA Logical Processors/Physical Cores/sockets; or other non-IA devices like GPU; hence the maximum external memory bandwidth limits may or may not be approached when this metric is flagged (see Uncore counters for that). Related metrics: tma_info_system_dram_bw_use",
"ScaleUnit": "100%"
},
{
"BriefDescription": "This metric estimates fraction of cycles where the performance was likely hurt due to latency from external memory (DRAM)",
- "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD) / tma_info_clks - tma_mem_bandwidth",
+ "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD) / tma_info_thread_clks - tma_mem_bandwidth",
"MetricGroup": "MemoryLat;Offcore;TopdownL4;tma_L4_group;tma_dram_bound_group;tma_issueLat",
"MetricName": "tma_mem_latency",
"MetricThreshold": "tma_mem_latency > 0.1 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))",
@@ -455,7 +455,7 @@
{
"BriefDescription": "This metric represents fraction of slots the Memory subsystem within the Backend was a bottleneck",
"MetricConstraint": "NO_GROUP_EVENTS",
- "MetricExpr": "(min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.STALLS_L1D_PENDING) + RESOURCE_STALLS.SB) / (min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.CYCLES_NO_DISPATCH) + cpu@UOPS_DISPATCHED.THREAD\\,cmask\\=1@ - (cpu@UOPS_DISPATCHED.THREAD\\,cmask\\=3@ if tma_info_ipc > 1.8 else cpu@UOPS_DISPATCHED.THREAD\\,cmask\\=2@) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB) * tma_backend_bound",
+ "MetricExpr": "(min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.STALLS_L1D_PENDING) + RESOURCE_STALLS.SB) / (min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.CYCLES_NO_DISPATCH) + cpu@UOPS_DISPATCHED.THREAD\\,cmask\\=1@ - (cpu@UOPS_DISPATCHED.THREAD\\,cmask\\=3@ if tma_info_thread_ipc > 1.8 else cpu@UOPS_DISPATCHED.THREAD\\,cmask\\=2@) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB) * tma_backend_bound",
"MetricGroup": "Backend;TmaL2;TopdownL2;tma_L2_group;tma_backend_bound_group",
"MetricName": "tma_memory_bound",
"MetricThreshold": "tma_memory_bound > 0.2 & tma_backend_bound > 0.2",
@@ -465,7 +465,7 @@
},
{
"BriefDescription": "This metric represents fraction of slots the CPU was retiring uops fetched by the Microcode Sequencer (MS) unit",
- "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / UOPS_ISSUED.ANY * IDQ.MS_UOPS / tma_info_slots",
+ "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / UOPS_ISSUED.ANY * IDQ.MS_UOPS / tma_info_thread_slots",
"MetricGroup": "MicroSeq;TopdownL3;tma_L3_group;tma_heavy_operations_group;tma_issueMC;tma_issueMS",
"MetricName": "tma_microcode_sequencer",
"MetricThreshold": "tma_microcode_sequencer > 0.05 & tma_heavy_operations > 0.1",
@@ -474,7 +474,7 @@
},
{
"BriefDescription": "This metric estimates the fraction of cycles when the CPU was stalled due to switches of uop delivery to the Microcode Sequencer (MS)",
- "MetricExpr": "3 * IDQ.MS_SWITCHES / tma_info_clks",
+ "MetricExpr": "3 * IDQ.MS_SWITCHES / tma_info_thread_clks",
"MetricGroup": "FetchLat;MicroSeq;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueMC;tma_issueMS;tma_issueMV;tma_issueSO",
"MetricName": "tma_ms_switches",
"MetricThreshold": "tma_ms_switches > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)",
@@ -484,7 +484,7 @@
{
"BriefDescription": "This metric estimates fraction of cycles the CPU performance was potentially limited due to Core computation issues (non divider-related)",
"MetricConstraint": "NO_GROUP_EVENTS",
- "MetricExpr": "(min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.CYCLES_NO_DISPATCH) + cpu@UOPS_DISPATCHED.THREAD\\,cmask\\=1@ - (cpu@UOPS_DISPATCHED.THREAD\\,cmask\\=3@ if tma_info_ipc > 1.8 else cpu@UOPS_DISPATCHED.THREAD\\,cmask\\=2@) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB - RESOURCE_STALLS.SB - min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.STALLS_L1D_PENDING)) / tma_info_clks",
+ "MetricExpr": "(min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.CYCLES_NO_DISPATCH) + cpu@UOPS_DISPATCHED.THREAD\\,cmask\\=1@ - (cpu@UOPS_DISPATCHED.THREAD\\,cmask\\=3@ if tma_info_thread_ipc > 1.8 else cpu@UOPS_DISPATCHED.THREAD\\,cmask\\=2@) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB - RESOURCE_STALLS.SB - min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.STALLS_L1D_PENDING)) / tma_info_thread_clks",
"MetricGroup": "PortsUtil;TopdownL3;tma_L3_group;tma_core_bound_group",
"MetricName": "tma_ports_utilization",
"MetricThreshold": "tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2)",
@@ -493,7 +493,7 @@
},
{
"BriefDescription": "This category represents fraction of slots utilized by useful work i.e. issued uops that eventually get retired",
- "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / tma_info_slots",
+ "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / tma_info_thread_slots",
"MetricGroup": "TmaL1;TopdownL1;tma_L1_group",
"MetricName": "tma_retiring",
"MetricThreshold": "tma_retiring > 0.7 | tma_heavy_operations > 0.1",
@@ -503,7 +503,7 @@
},
{
"BriefDescription": "This metric estimates how often CPU was stalled due to RFO store memory accesses; RFO store issue a read-for-ownership request before the write",
- "MetricExpr": "RESOURCE_STALLS.SB / tma_info_clks",
+ "MetricExpr": "RESOURCE_STALLS.SB / tma_info_thread_clks",
"MetricGroup": "MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group",
"MetricName": "tma_store_bound",
"MetricThreshold": "tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)",
--
2.40.1.606.ga4b1b128d6-goog
^ permalink raw reply related [flat|nested] 9+ messages in thread* [PATCH v1 12/15] perf vendor events intel: Update snowridgex events
2023-05-15 21:58 [PATCH v1 00/15] Event, metric and metric group improvements Ian Rogers
` (2 preceding siblings ...)
2023-05-15 21:58 ` [PATCH v1 09/15] perf vendor events intel: Update sandybridge metrics Ian Rogers
@ 2023-05-15 21:58 ` Ian Rogers
2023-05-15 21:58 ` [PATCH v1 14/15] perf jevents: Add support for metricgroup descriptions Ian Rogers
[not found] ` <20230515215844.653610-16-irogers@google.com>
5 siblings, 0 replies; 9+ messages in thread
From: Ian Rogers @ 2023-05-15 21:58 UTC (permalink / raw)
To: Peter Zijlstra, Ingo Molnar, Arnaldo Carvalho de Melo,
Mark Rutland, Alexander Shishkin, Jiri Olsa, Namhyung Kim,
Ian Rogers, Adrian Hunter, Kan Liang, Zhengjun Xing, John Garry,
Kajol Jain, Thomas Richter, linux-kernel, linux-perf-users
Update snowridgex to v1.21 that marks deprecated a number of events
and adds improves descriptions. The events data was generated by:
https://github.com/intel/perfmon/blob/main/scripts/create_perf_json.py
Signed-off-by: Ian Rogers <irogers@google.com>
---
tools/perf/pmu-events/arch/x86/mapfile.csv | 2 +-
.../perf/pmu-events/arch/x86/snowridgex/cache.json | 7 +++++++
.../pmu-events/arch/x86/snowridgex/memory.json | 2 ++
.../perf/pmu-events/arch/x86/snowridgex/other.json | 10 ++++++++++
.../pmu-events/arch/x86/snowridgex/pipeline.json | 3 +++
.../arch/x86/snowridgex/uncore-interconnect.json | 14 +++++++-------
.../pmu-events/arch/x86/snowridgex/uncore-io.json | 8 --------
.../arch/x86/snowridgex/uncore-memory.json | 7 +++----
.../arch/x86/snowridgex/uncore-power.json | 6 +++---
9 files changed, 36 insertions(+), 23 deletions(-)
diff --git a/tools/perf/pmu-events/arch/x86/mapfile.csv b/tools/perf/pmu-events/arch/x86/mapfile.csv
index 4731a92af9f9..4a1a2b8d6201 100644
--- a/tools/perf/pmu-events/arch/x86/mapfile.csv
+++ b/tools/perf/pmu-events/arch/x86/mapfile.csv
@@ -28,7 +28,7 @@ GenuineIntel-6-AF,v1.00,sierraforest,core
GenuineIntel-6-(37|4A|4C|4D|5A),v15,silvermont,core
GenuineIntel-6-(4E|5E|8E|9E|A5|A6),v56,skylake,core
GenuineIntel-6-55-[01234],v1.30,skylakex,core
-GenuineIntel-6-86,v1.20,snowridgex,core
+GenuineIntel-6-86,v1.21,snowridgex,core
GenuineIntel-6-8[CD],v1.10,tigerlake,core
GenuineIntel-6-2C,v4,westmereep-dp,core
GenuineIntel-6-25,v3,westmereep-sp,core
diff --git a/tools/perf/pmu-events/arch/x86/snowridgex/cache.json b/tools/perf/pmu-events/arch/x86/snowridgex/cache.json
index 0ab90e3bf76b..c6be60584522 100644
--- a/tools/perf/pmu-events/arch/x86/snowridgex/cache.json
+++ b/tools/perf/pmu-events/arch/x86/snowridgex/cache.json
@@ -72,6 +72,7 @@
"BriefDescription": "Counts the number of cycles the core is stalled due to an instruction cache or TLB miss which hit in the L2, LLC, DRAM or MMIO (Non-DRAM).",
"EventCode": "0x34",
"EventName": "MEM_BOUND_STALLS.IFETCH",
+ "PublicDescription": "Counts the number of cycles the core is stalled due to an instruction cache or translation lookaside buffer (TLB) miss which hit in the L2, LLC, DRAM or MMIO (Non-DRAM).",
"SampleAfterValue": "200003",
"UMask": "0x38"
},
@@ -437,6 +438,7 @@
},
{
"BriefDescription": "This event is deprecated. Refer to new event OCR.DEMAND_DATA_AND_L1PF_RD.L3_HIT",
+ "Deprecated": "1",
"EventCode": "0XB7",
"EventName": "OCR.DEMAND_DATA_RD.L3_HIT",
"MSRIndex": "0x1a6,0x1a7",
@@ -446,6 +448,7 @@
},
{
"BriefDescription": "This event is deprecated. Refer to new event OCR.DEMAND_DATA_AND_L1PF_RD.L3_HIT.SNOOP_HITM",
+ "Deprecated": "1",
"EventCode": "0XB7",
"EventName": "OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HITM",
"MSRIndex": "0x1a6,0x1a7",
@@ -455,6 +458,7 @@
},
{
"BriefDescription": "This event is deprecated. Refer to new event OCR.DEMAND_DATA_AND_L1PF_RD.L3_HIT.SNOOP_HIT_NO_FWD",
+ "Deprecated": "1",
"EventCode": "0XB7",
"EventName": "OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HIT_NO_FWD",
"MSRIndex": "0x1a6,0x1a7",
@@ -464,6 +468,7 @@
},
{
"BriefDescription": "This event is deprecated. Refer to new event OCR.DEMAND_DATA_AND_L1PF_RD.L3_HIT.SNOOP_HIT_WITH_FWD",
+ "Deprecated": "1",
"EventCode": "0XB7",
"EventName": "OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HIT_WITH_FWD",
"MSRIndex": "0x1a6,0x1a7",
@@ -473,6 +478,7 @@
},
{
"BriefDescription": "This event is deprecated. Refer to new event OCR.DEMAND_DATA_AND_L1PF_RD.L3_HIT.SNOOP_MISS",
+ "Deprecated": "1",
"EventCode": "0XB7",
"EventName": "OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_MISS",
"MSRIndex": "0x1a6,0x1a7",
@@ -482,6 +488,7 @@
},
{
"BriefDescription": "This event is deprecated. Refer to new event OCR.DEMAND_DATA_AND_L1PF_RD.L3_HIT.SNOOP_NOT_NEEDED",
+ "Deprecated": "1",
"EventCode": "0XB7",
"EventName": "OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_NOT_NEEDED",
"MSRIndex": "0x1a6,0x1a7",
diff --git a/tools/perf/pmu-events/arch/x86/snowridgex/memory.json b/tools/perf/pmu-events/arch/x86/snowridgex/memory.json
index 18621909d1a9..c02eb0e836ad 100644
--- a/tools/perf/pmu-events/arch/x86/snowridgex/memory.json
+++ b/tools/perf/pmu-events/arch/x86/snowridgex/memory.json
@@ -96,6 +96,7 @@
},
{
"BriefDescription": "This event is deprecated. Refer to new event OCR.DEMAND_DATA_AND_L1PF_RD.L3_MISS",
+ "Deprecated": "1",
"EventCode": "0XB7",
"EventName": "OCR.DEMAND_DATA_RD.L3_MISS",
"MSRIndex": "0x1a6,0x1a7",
@@ -105,6 +106,7 @@
},
{
"BriefDescription": "This event is deprecated. Refer to new event OCR.DEMAND_DATA_AND_L1PF_RD.L3_MISS_LOCAL",
+ "Deprecated": "1",
"EventCode": "0XB7",
"EventName": "OCR.DEMAND_DATA_RD.L3_MISS_LOCAL",
"MSRIndex": "0x1a6,0x1a7",
diff --git a/tools/perf/pmu-events/arch/x86/snowridgex/other.json b/tools/perf/pmu-events/arch/x86/snowridgex/other.json
index 00ae180ded25..fefbc383b840 100644
--- a/tools/perf/pmu-events/arch/x86/snowridgex/other.json
+++ b/tools/perf/pmu-events/arch/x86/snowridgex/other.json
@@ -1,6 +1,7 @@
[
{
"BriefDescription": "This event is deprecated. Refer to new event BUS_LOCK.SELF_LOCKS",
+ "Deprecated": "1",
"EdgeDetect": "1",
"EventCode": "0x63",
"EventName": "BUS_LOCK.ALL",
@@ -16,6 +17,7 @@
},
{
"BriefDescription": "This event is deprecated. Refer to new event BUS_LOCK.BLOCK_CYCLES",
+ "Deprecated": "1",
"EventCode": "0x63",
"EventName": "BUS_LOCK.CYCLES_OTHER_BLOCK",
"SampleAfterValue": "200003",
@@ -23,6 +25,7 @@
},
{
"BriefDescription": "This event is deprecated. Refer to new event BUS_LOCK.LOCK_CYCLES",
+ "Deprecated": "1",
"EventCode": "0x63",
"EventName": "BUS_LOCK.CYCLES_SELF_BLOCK",
"SampleAfterValue": "200003",
@@ -46,6 +49,7 @@
},
{
"BriefDescription": "This event is deprecated. Refer to new event MEM_BOUND_STALLS.LOAD_DRAM_HIT",
+ "Deprecated": "1",
"EventCode": "0x34",
"EventName": "C0_STALLS.LOAD_DRAM_HIT",
"SampleAfterValue": "200003",
@@ -53,6 +57,7 @@
},
{
"BriefDescription": "This event is deprecated. Refer to new event MEM_BOUND_STALLS.LOAD_L2_HIT",
+ "Deprecated": "1",
"EventCode": "0x34",
"EventName": "C0_STALLS.LOAD_L2_HIT",
"SampleAfterValue": "200003",
@@ -60,6 +65,7 @@
},
{
"BriefDescription": "This event is deprecated. Refer to new event MEM_BOUND_STALLS.LOAD_LLC_HIT",
+ "Deprecated": "1",
"EventCode": "0x34",
"EventName": "C0_STALLS.LOAD_LLC_HIT",
"SampleAfterValue": "200003",
@@ -207,6 +213,7 @@
},
{
"BriefDescription": "This event is deprecated. Refer to new event OCR.DEMAND_DATA_AND_L1PF_RD.ANY_RESPONSE",
+ "Deprecated": "1",
"EventCode": "0XB7",
"EventName": "OCR.DEMAND_DATA_RD.ANY_RESPONSE",
"MSRIndex": "0x1a6,0x1a7",
@@ -216,6 +223,7 @@
},
{
"BriefDescription": "This event is deprecated. Refer to new event OCR.DEMAND_DATA_AND_L1PF_RD.DRAM",
+ "Deprecated": "1",
"EventCode": "0XB7",
"EventName": "OCR.DEMAND_DATA_RD.DRAM",
"MSRIndex": "0x1a6,0x1a7",
@@ -225,6 +233,7 @@
},
{
"BriefDescription": "This event is deprecated. Refer to new event OCR.DEMAND_DATA_AND_L1PF_RD.LOCAL_DRAM",
+ "Deprecated": "1",
"EventCode": "0XB7",
"EventName": "OCR.DEMAND_DATA_RD.LOCAL_DRAM",
"MSRIndex": "0x1a6,0x1a7",
@@ -234,6 +243,7 @@
},
{
"BriefDescription": "This event is deprecated. Refer to new event OCR.DEMAND_DATA_AND_L1PF_RD.OUTSTANDING",
+ "Deprecated": "1",
"EventCode": "0XB7",
"EventName": "OCR.DEMAND_DATA_RD.OUTSTANDING",
"MSRIndex": "0x1a6",
diff --git a/tools/perf/pmu-events/arch/x86/snowridgex/pipeline.json b/tools/perf/pmu-events/arch/x86/snowridgex/pipeline.json
index 9dd8c909facc..c483c0838e08 100644
--- a/tools/perf/pmu-events/arch/x86/snowridgex/pipeline.json
+++ b/tools/perf/pmu-events/arch/x86/snowridgex/pipeline.json
@@ -165,6 +165,7 @@
},
{
"BriefDescription": "This event is deprecated.",
+ "Deprecated": "1",
"EventCode": "0xcd",
"EventName": "CYCLES_DIV_BUSY.ANY",
"SampleAfterValue": "2000003"
@@ -283,6 +284,7 @@
},
{
"BriefDescription": "This event is deprecated. Refer to new event TOPDOWN_BAD_SPECULATION.FASTNUKE",
+ "Deprecated": "1",
"EventCode": "0x73",
"EventName": "TOPDOWN_BAD_SPECULATION.MONUKE",
"SampleAfterValue": "1000003",
@@ -338,6 +340,7 @@
},
{
"BriefDescription": "This event is deprecated.",
+ "Deprecated": "1",
"EventCode": "0x74",
"EventName": "TOPDOWN_BE_BOUND.STORE_BUFFER",
"SampleAfterValue": "1000003",
diff --git a/tools/perf/pmu-events/arch/x86/snowridgex/uncore-interconnect.json b/tools/perf/pmu-events/arch/x86/snowridgex/uncore-interconnect.json
index de3840078e21..7e2895f7fe3d 100644
--- a/tools/perf/pmu-events/arch/x86/snowridgex/uncore-interconnect.json
+++ b/tools/perf/pmu-events/arch/x86/snowridgex/uncore-interconnect.json
@@ -590,7 +590,7 @@
"EventCode": "0x0C",
"EventName": "UNC_I_TxS_REQUEST_OCCUPANCY",
"PerPkg": "1",
- "PublicDescription": "Outbound Request Queue Occupancy : Accumultes the number of outstanding outbound requests from the IRP to the switch (towards the devices). This can be used in conjuection with the allocations event in order to calculate average latency of outbound requests.",
+ "PublicDescription": "Outbound Request Queue Occupancy : Accumulates the number of outstanding outbound requests from the IRP to the switch (towards the devices). This can be used in conjunction with the allocations event in order to calculate average latency of outbound requests.",
"Unit": "IRP"
},
{
@@ -5570,7 +5570,7 @@
"Unit": "M2M"
},
{
- "BriefDescription": "M2M->iMC WPQ Cycles w/Credits - Regular : Channel 0",
+ "BriefDescription": "M2M->iMC WPQ Cycles w/Credits - Regular : Channel 0",
"EventCode": "0x4D",
"EventName": "UNC_M2M_WPQ_NO_REG_CRD.CHN0",
"PerPkg": "1",
@@ -5578,7 +5578,7 @@
"Unit": "M2M"
},
{
- "BriefDescription": "M2M->iMC WPQ Cycles w/Credits - Regular : Channel 1",
+ "BriefDescription": "M2M->iMC WPQ Cycles w/Credits - Regular : Channel 1",
"EventCode": "0x4D",
"EventName": "UNC_M2M_WPQ_NO_REG_CRD.CHN1",
"PerPkg": "1",
@@ -5586,7 +5586,7 @@
"Unit": "M2M"
},
{
- "BriefDescription": "M2M->iMC WPQ Cycles w/Credits - Regular : Channel 2",
+ "BriefDescription": "M2M->iMC WPQ Cycles w/Credits - Regular : Channel 2",
"EventCode": "0x4D",
"EventName": "UNC_M2M_WPQ_NO_REG_CRD.CHN2",
"PerPkg": "1",
@@ -5594,7 +5594,7 @@
"Unit": "M2M"
},
{
- "BriefDescription": "M2M->iMC WPQ Cycles w/Credits - Special : Channel 0",
+ "BriefDescription": "M2M->iMC WPQ Cycles w/Credits - Special : Channel 0",
"EventCode": "0x4E",
"EventName": "UNC_M2M_WPQ_NO_SPEC_CRD.CHN0",
"PerPkg": "1",
@@ -5602,7 +5602,7 @@
"Unit": "M2M"
},
{
- "BriefDescription": "M2M->iMC WPQ Cycles w/Credits - Special : Channel 1",
+ "BriefDescription": "M2M->iMC WPQ Cycles w/Credits - Special : Channel 1",
"EventCode": "0x4E",
"EventName": "UNC_M2M_WPQ_NO_SPEC_CRD.CHN1",
"PerPkg": "1",
@@ -5610,7 +5610,7 @@
"Unit": "M2M"
},
{
- "BriefDescription": "M2M->iMC WPQ Cycles w/Credits - Special : Channel 2",
+ "BriefDescription": "M2M->iMC WPQ Cycles w/Credits - Special : Channel 2",
"EventCode": "0x4E",
"EventName": "UNC_M2M_WPQ_NO_SPEC_CRD.CHN2",
"PerPkg": "1",
diff --git a/tools/perf/pmu-events/arch/x86/snowridgex/uncore-io.json b/tools/perf/pmu-events/arch/x86/snowridgex/uncore-io.json
index 996028071ee4..ecdd6f0f8e8f 100644
--- a/tools/perf/pmu-events/arch/x86/snowridgex/uncore-io.json
+++ b/tools/perf/pmu-events/arch/x86/snowridgex/uncore-io.json
@@ -34,7 +34,6 @@
"EventCode": "0xff",
"EventName": "UNC_IIO_BANDWIDTH_IN.PART0_FREERUN",
"PerPkg": "1",
- "PublicDescription": "UNC_IIO_BANDWIDTH_IN.PART0_FREERUN",
"UMask": "0x20",
"Unit": "iio_free_running"
},
@@ -43,7 +42,6 @@
"EventCode": "0xff",
"EventName": "UNC_IIO_BANDWIDTH_IN.PART1_FREERUN",
"PerPkg": "1",
- "PublicDescription": "UNC_IIO_BANDWIDTH_IN.PART1_FREERUN",
"UMask": "0x21",
"Unit": "iio_free_running"
},
@@ -52,7 +50,6 @@
"EventCode": "0xff",
"EventName": "UNC_IIO_BANDWIDTH_IN.PART2_FREERUN",
"PerPkg": "1",
- "PublicDescription": "UNC_IIO_BANDWIDTH_IN.PART2_FREERUN",
"UMask": "0x22",
"Unit": "iio_free_running"
},
@@ -61,7 +58,6 @@
"EventCode": "0xff",
"EventName": "UNC_IIO_BANDWIDTH_IN.PART3_FREERUN",
"PerPkg": "1",
- "PublicDescription": "UNC_IIO_BANDWIDTH_IN.PART3_FREERUN",
"UMask": "0x23",
"Unit": "iio_free_running"
},
@@ -70,7 +66,6 @@
"EventCode": "0xff",
"EventName": "UNC_IIO_BANDWIDTH_IN.PART4_FREERUN",
"PerPkg": "1",
- "PublicDescription": "UNC_IIO_BANDWIDTH_IN.PART4_FREERUN",
"UMask": "0x24",
"Unit": "iio_free_running"
},
@@ -79,7 +74,6 @@
"EventCode": "0xff",
"EventName": "UNC_IIO_BANDWIDTH_IN.PART5_FREERUN",
"PerPkg": "1",
- "PublicDescription": "UNC_IIO_BANDWIDTH_IN.PART5_FREERUN",
"UMask": "0x25",
"Unit": "iio_free_running"
},
@@ -88,7 +82,6 @@
"EventCode": "0xff",
"EventName": "UNC_IIO_BANDWIDTH_IN.PART6_FREERUN",
"PerPkg": "1",
- "PublicDescription": "UNC_IIO_BANDWIDTH_IN.PART6_FREERUN",
"UMask": "0x26",
"Unit": "iio_free_running"
},
@@ -97,7 +90,6 @@
"EventCode": "0xff",
"EventName": "UNC_IIO_BANDWIDTH_IN.PART7_FREERUN",
"PerPkg": "1",
- "PublicDescription": "UNC_IIO_BANDWIDTH_IN.PART7_FREERUN",
"UMask": "0x27",
"Unit": "iio_free_running"
},
diff --git a/tools/perf/pmu-events/arch/x86/snowridgex/uncore-memory.json b/tools/perf/pmu-events/arch/x86/snowridgex/uncore-memory.json
index 530e9b71b92a..b80911d498dd 100644
--- a/tools/perf/pmu-events/arch/x86/snowridgex/uncore-memory.json
+++ b/tools/perf/pmu-events/arch/x86/snowridgex/uncore-memory.json
@@ -130,7 +130,6 @@
"EventCode": "0xff",
"EventName": "UNC_M_CLOCKTICKS_FREERUN",
"PerPkg": "1",
- "PublicDescription": "UNC_M_CLOCKTICKS_FREERUN",
"UMask": "0x10",
"Unit": "imc_free_running"
},
@@ -322,7 +321,7 @@
"EventCode": "0x02",
"EventName": "UNC_M_PRE_COUNT.PGT",
"PerPkg": "1",
- "PublicDescription": "DRAM Precharge commands. : Precharge due to page table : Counts the number of DRAM Precharge commands sent on this channel. : Prechages from Page Table",
+ "PublicDescription": "DRAM Precharge commands. : Precharge due to page table : Counts the number of DRAM Precharge commands sent on this channel. : Precharges from Page Table",
"UMask": "0x10",
"Unit": "iMC"
},
@@ -497,7 +496,7 @@
"EventCode": "0x82",
"EventName": "UNC_M_WPQ_OCCUPANCY_PCH0",
"PerPkg": "1",
- "PublicDescription": "Write Pending Queue Occupancy : Accumulates the occupancies of the Write Pending Queue each cycle. This can then be used to calculate both the average queue occupancy (in conjunction with the number of cycles not empty) and the average latency (in conjunction with the number of allocations). The WPQ is used to schedule write out to the memory controller and to track the writes. Requests allocate into the WPQ soon after they enter the memory controller, and need credits for an entry in this buffer before being sent from the HA to the iMC. They deallocate after being issued to DRAM. Write requests themselves are able to complete (from the perspective of the rest of the system) as soon they have posted to the iMC. This is not to be confused with actually performing the write to DRAM. Therefore, the average latency for this queue is actually not useful for deconstruction intermediate write latencies. So, we provide filtering based on if the request has posted or not. By using the not posted filter, we can track how long writes spent in the iMC before completions were sent to the HA. The posted filter, on the other hand, provides information about how much queueing is actually happenning in the iMC for writes before they are actually issued to memory. High average occupancies will generally coincide with high write major mode counts.",
+ "PublicDescription": "Write Pending Queue Occupancy : Accumulates the occupancies of the Write Pending Queue each cycle. This can then be used to calculate both the average queue occupancy (in conjunction with the number of cycles not empty) and the average latency (in conjunction with the number of allocations). The WPQ is used to schedule write out to the memory controller and to track the writes. Requests allocate into the WPQ soon after they enter the memory controller, and need credits for an entry in this buffer before being sent from the HA to the iMC. They deallocate after being issued to DRAM. Write requests themselves are able to complete (from the perspective of the rest of the system) as soon they have posted to the iMC. This is not to be confused with actually performing the write to DRAM. Therefore, the average latency for this queue is actually not useful for deconstruction intermediate write latencies. So, we provide filtering based on if the request has posted or not. By using the not posted filter, we can track how long writes spent in the iMC before completions were sent to the HA. The posted filter, on the other hand, provides information about how much queueing is actually happening in the iMC for writes before they are actually issued to memory. High average occupancies will generally coincide with high write major mode counts.",
"Unit": "iMC"
},
{
@@ -505,7 +504,7 @@
"EventCode": "0x83",
"EventName": "UNC_M_WPQ_OCCUPANCY_PCH1",
"PerPkg": "1",
- "PublicDescription": "Write Pending Queue Occupancy : Accumulates the occupancies of the Write Pending Queue each cycle. This can then be used to calculate both the average queue occupancy (in conjunction with the number of cycles not empty) and the average latency (in conjunction with the number of allocations). The WPQ is used to schedule write out to the memory controller and to track the writes. Requests allocate into the WPQ soon after they enter the memory controller, and need credits for an entry in this buffer before being sent from the HA to the iMC. They deallocate after being issued to DRAM. Write requests themselves are able to complete (from the perspective of the rest of the system) as soon they have posted to the iMC. This is not to be confused with actually performing the write to DRAM. Therefore, the average latency for this queue is actually not useful for deconstruction intermediate write latencies. So, we provide filtering based on if the request has posted or not. By using the not posted filter, we can track how long writes spent in the iMC before completions were sent to the HA. The posted filter, on the other hand, provides information about how much queueing is actually happenning in the iMC for writes before they are actually issued to memory. High average occupancies will generally coincide with high write major mode counts.",
+ "PublicDescription": "Write Pending Queue Occupancy : Accumulates the occupancies of the Write Pending Queue each cycle. This can then be used to calculate both the average queue occupancy (in conjunction with the number of cycles not empty) and the average latency (in conjunction with the number of allocations). The WPQ is used to schedule write out to the memory controller and to track the writes. Requests allocate into the WPQ soon after they enter the memory controller, and need credits for an entry in this buffer before being sent from the HA to the iMC. They deallocate after being issued to DRAM. Write requests themselves are able to complete (from the perspective of the rest of the system) as soon they have posted to the iMC. This is not to be confused with actually performing the write to DRAM. Therefore, the average latency for this queue is actually not useful for deconstruction intermediate write latencies. So, we provide filtering based on if the request has posted or not. By using the not posted filter, we can track how long writes spent in the iMC before completions were sent to the HA. The posted filter, on the other hand, provides information about how much queueing is actually happening in the iMC for writes before they are actually issued to memory. High average occupancies will generally coincide with high write major mode counts.",
"Unit": "iMC"
},
{
diff --git a/tools/perf/pmu-events/arch/x86/snowridgex/uncore-power.json b/tools/perf/pmu-events/arch/x86/snowridgex/uncore-power.json
index 27fc155f1223..a61ffca2dfea 100644
--- a/tools/perf/pmu-events/arch/x86/snowridgex/uncore-power.json
+++ b/tools/perf/pmu-events/arch/x86/snowridgex/uncore-power.json
@@ -149,7 +149,7 @@
"EventCode": "0x80",
"EventName": "UNC_P_POWER_STATE_OCCUPANCY.CORES_C0",
"PerPkg": "1",
- "PublicDescription": "Number of cores in C-State : C0 and C1 : This is an occupancy event that tracks the number of cores that are in the chosen C-State. It can be used by itself to get the average number of cores in that C-state with threshholding to generate histograms, or with other PCU events and occupancy triggering to capture other details.",
+ "PublicDescription": "Number of cores in C-State : C0 and C1 : This is an occupancy event that tracks the number of cores that are in the chosen C-State. It can be used by itself to get the average number of cores in that C-state with thresholding to generate histograms, or with other PCU events and occupancy triggering to capture other details.",
"Unit": "PCU"
},
{
@@ -157,7 +157,7 @@
"EventCode": "0x80",
"EventName": "UNC_P_POWER_STATE_OCCUPANCY.CORES_C3",
"PerPkg": "1",
- "PublicDescription": "Number of cores in C-State : C3 : This is an occupancy event that tracks the number of cores that are in the chosen C-State. It can be used by itself to get the average number of cores in that C-state with threshholding to generate histograms, or with other PCU events and occupancy triggering to capture other details.",
+ "PublicDescription": "Number of cores in C-State : C3 : This is an occupancy event that tracks the number of cores that are in the chosen C-State. It can be used by itself to get the average number of cores in that C-state with thresholding to generate histograms, or with other PCU events and occupancy triggering to capture other details.",
"Unit": "PCU"
},
{
@@ -165,7 +165,7 @@
"EventCode": "0x80",
"EventName": "UNC_P_POWER_STATE_OCCUPANCY.CORES_C6",
"PerPkg": "1",
- "PublicDescription": "Number of cores in C-State : C6 and C7 : This is an occupancy event that tracks the number of cores that are in the chosen C-State. It can be used by itself to get the average number of cores in that C-state with threshholding to generate histograms, or with other PCU events and occupancy triggering to capture other details.",
+ "PublicDescription": "Number of cores in C-State : C6 and C7 : This is an occupancy event that tracks the number of cores that are in the chosen C-State. It can be used by itself to get the average number of cores in that C-state with thresholding to generate histograms, or with other PCU events and occupancy triggering to capture other details.",
"Unit": "PCU"
},
{
--
2.40.1.606.ga4b1b128d6-goog
^ permalink raw reply related [flat|nested] 9+ messages in thread* [PATCH v1 14/15] perf jevents: Add support for metricgroup descriptions
2023-05-15 21:58 [PATCH v1 00/15] Event, metric and metric group improvements Ian Rogers
` (3 preceding siblings ...)
2023-05-15 21:58 ` [PATCH v1 12/15] perf vendor events intel: Update snowridgex events Ian Rogers
@ 2023-05-15 21:58 ` Ian Rogers
[not found] ` <20230515215844.653610-16-irogers@google.com>
5 siblings, 0 replies; 9+ messages in thread
From: Ian Rogers @ 2023-05-15 21:58 UTC (permalink / raw)
To: Peter Zijlstra, Ingo Molnar, Arnaldo Carvalho de Melo,
Mark Rutland, Alexander Shishkin, Jiri Olsa, Namhyung Kim,
Ian Rogers, Adrian Hunter, Kan Liang, Zhengjun Xing, John Garry,
Kajol Jain, Thomas Richter, linux-kernel, linux-perf-users
Metrics have a field where the groups they belong to are listed like
the following from
tools/perf/pmu-events/arch/x86/skylakex/skx-metrics.json:
"MetricGroup": "PGO;TmaL1;TopdownL1;tma_L1_group",
"MetricName": "tma_frontend_bound",
The metric groups are shown in 'perf list' like the following where
TopdownL1 is a metric group:
TopdownL1:
tma_backend_bound
[This category represents fraction of slots where no uops are being
delivered due to a lack of required resources for accepting new uops
in the Backend]
tma_bad_speculation
[This category represents fraction of slots wasted due to incorrect
speculations]
tma_frontend_bound
[This category represents fraction of slots where the processor's
Frontend undersupplies its Backend]
tma_retiring
[This category represents fraction of slots utilized by useful work
i.e. issued uops that eventually get retired]
This patch adds support for a new json file in each model directory
called metricgroups.json that comprises a dictionary containing
entries that map from a metric group to a description:
{
...
"TopdownL1": "Metrics for top-down breakdown at level 1",
...
}
perf list is then updated to support this changing the above output
to:
TopdownL1: [Metrics for top-down breakdown at level 1]
Signed-off-by: Ian Rogers <irogers@google.com>
---
tools/perf/builtin-list.c | 11 ++++--
tools/perf/pmu-events/empty-pmu-events.c | 5 +++
tools/perf/pmu-events/jevents.py | 49 +++++++++++++++++++++++-
tools/perf/pmu-events/pmu-events.h | 2 +
4 files changed, 62 insertions(+), 5 deletions(-)
diff --git a/tools/perf/builtin-list.c b/tools/perf/builtin-list.c
index c6bd0aa4a56e..e8520a027b45 100644
--- a/tools/perf/builtin-list.c
+++ b/tools/perf/builtin-list.c
@@ -192,9 +192,14 @@ static void default_print_metric(void *ps,
if (group && print_state->metricgroups) {
if (print_state->name_only)
printf("%s ", group);
- else if (print_state->metrics)
- printf("\n%s:\n", group);
- else
+ else if (print_state->metrics) {
+ const char *gdesc = describe_metricgroup(group);
+
+ if (gdesc)
+ printf("\n%s: [%s]\n", group, gdesc);
+ else
+ printf("\n%s:\n", group);
+ } else
printf("%s\n", group);
}
zfree(&print_state->last_metricgroups);
diff --git a/tools/perf/pmu-events/empty-pmu-events.c b/tools/perf/pmu-events/empty-pmu-events.c
index e74defb5284f..a630c617e879 100644
--- a/tools/perf/pmu-events/empty-pmu-events.c
+++ b/tools/perf/pmu-events/empty-pmu-events.c
@@ -420,3 +420,8 @@ int pmu_for_each_sys_metric(pmu_metric_iter_fn fn __maybe_unused, void *data __m
{
return 0;
}
+
+const char *describe_metricgroup(const char *group __maybe_unused)
+{
+ return NULL;
+}
diff --git a/tools/perf/pmu-events/jevents.py b/tools/perf/pmu-events/jevents.py
index 487ff01baf1b..8fca7c9adee0 100755
--- a/tools/perf/pmu-events/jevents.py
+++ b/tools/perf/pmu-events/jevents.py
@@ -37,6 +37,8 @@ _pending_metrics = []
_pending_metrics_tblname = None
# Global BigCString shared by all structures.
_bcs = None
+# Map from the name of a metric group to a description of the group.
+_metricgroups = {}
# Order specific JsonEvent attributes will be visited.
_json_event_attributes = [
# cmp_sevent related attributes.
@@ -512,6 +514,17 @@ def preprocess_one_file(parents: Sequence[str], item: os.DirEntry) -> None:
if not item.is_file() or not item.name.endswith('.json'):
return
+ if item.name == 'metricgroups.json':
+ metricgroup_descriptions = json.load(open(item.path))
+ for mgroup in metricgroup_descriptions:
+ assert len(mgroup) > 1, parents
+ description = f"{metricgroup_descriptions[mgroup]}\\000"
+ mgroup = f"{mgroup}\\000"
+ _bcs.add(mgroup)
+ _bcs.add(description)
+ _metricgroups[mgroup] = description
+ return
+
topic = get_topic(item.name)
for event in read_json_events(item.path, topic):
if event.name:
@@ -548,7 +561,7 @@ def process_one_file(parents: Sequence[str], item: os.DirEntry) -> None:
# Ignore other directories. If the file name does not have a .json
# extension, ignore it. It could be a readme.txt for instance.
- if not item.is_file() or not item.name.endswith('.json'):
+ if not item.is_file() or not item.name.endswith('.json') or item.name == 'metricgroups.json':
return
add_events_table_entries(item, get_topic(item.name))
@@ -911,6 +924,38 @@ int pmu_for_each_sys_metric(pmu_metric_iter_fn fn, void *data)
}
""")
+def print_metricgroups() -> None:
+ _args.output_file.write("""
+static const int metricgroups[][2] = {
+""")
+ for mgroup in sorted(_metricgroups):
+ description = _metricgroups[mgroup]
+ _args.output_file.write(
+ f'\t{{ {_bcs.offsets[mgroup]}, {_bcs.offsets[description]} }}, /* {mgroup} => {description} */\n'
+ )
+ _args.output_file.write("""
+};
+
+const char *describe_metricgroup(const char *group)
+{
+ int low = 0, high = ARRAY_SIZE(metricgroups) - 1;
+
+ while (low <= high) {
+ int mid = (low + high) / 2;
+ const char *mgroup = &big_c_string[metricgroups[mid][0]];
+ int cmp = strcmp(mgroup, group);
+
+ if (cmp == 0) {
+ return &big_c_string[metricgroups[mid][1]];
+ } else if (cmp < 0) {
+ low = mid + 1;
+ } else {
+ high = mid - 1;
+ }
+ }
+ return NULL;
+}
+""")
def main() -> None:
global _args
@@ -993,7 +1038,7 @@ struct compact_pmu_event {
print_mapping_table(archs)
print_system_mapping_table()
-
+ print_metricgroups()
if __name__ == '__main__':
main()
diff --git a/tools/perf/pmu-events/pmu-events.h b/tools/perf/pmu-events/pmu-events.h
index 3549e6971a4d..8cd23d656a5d 100644
--- a/tools/perf/pmu-events/pmu-events.h
+++ b/tools/perf/pmu-events/pmu-events.h
@@ -93,4 +93,6 @@ const struct pmu_metrics_table *find_sys_metrics_table(const char *name);
int pmu_for_each_sys_event(pmu_event_iter_fn fn, void *data);
int pmu_for_each_sys_metric(pmu_metric_iter_fn fn, void *data);
+const char *describe_metricgroup(const char *group);
+
#endif
--
2.40.1.606.ga4b1b128d6-goog
^ permalink raw reply related [flat|nested] 9+ messages in thread[parent not found: <20230515215844.653610-16-irogers@google.com>]
* Re: [PATCH v1 15/15] perf vendor events intel: Add metricgroup descriptions for all models
[not found] ` <20230515215844.653610-16-irogers@google.com>
@ 2023-05-16 17:53 ` Liang, Kan
2023-05-16 18:19 ` Ian Rogers
0 siblings, 1 reply; 9+ messages in thread
From: Liang, Kan @ 2023-05-16 17:53 UTC (permalink / raw)
To: Ian Rogers, Peter Zijlstra, Ingo Molnar, Arnaldo Carvalho de Melo,
Mark Rutland, Alexander Shishkin, Jiri Olsa, Namhyung Kim,
Adrian Hunter, John Garry, Kajol Jain, Thomas Richter,
linux-kernel, linux-perf-users
On 2023-05-15 5:58 p.m., Ian Rogers wrote:
> Add metric group descriptions created by:
> https://github.com/intel/perfmon/blob/main/scripts/create_perf_json.py
> The descriptions add some additional detail in perf list.
>
> Signed-off-by: Ian Rogers <irogers@google.com>
> ---
> .../arch/x86/alderlake/metricgroups.json | 122 ++++++++++++++++++
> .../arch/x86/alderlaken/metricgroups.json | 26 ++++
> .../arch/x86/broadwell/metricgroups.json | 107 +++++++++++++++
> .../arch/x86/broadwellde/metricgroups.json | 107 +++++++++++++++
> .../arch/x86/broadwellx/metricgroups.json | 107 +++++++++++++++
> .../arch/x86/cascadelakex/metricgroups.json | 114 ++++++++++++++++
> .../arch/x86/haswell/metricgroups.json | 107 +++++++++++++++
> .../arch/x86/haswellx/metricgroups.json | 107 +++++++++++++++
> .../arch/x86/icelake/metricgroups.json | 113 ++++++++++++++++
> .../arch/x86/icelakex/metricgroups.json | 114 ++++++++++++++++
> .../arch/x86/ivybridge/metricgroups.json | 107 +++++++++++++++
> .../arch/x86/ivytown/metricgroups.json | 107 +++++++++++++++
> .../arch/x86/jaketown/metricgroups.json | 100 ++++++++++++++
> .../arch/x86/sandybridge/metricgroups.json | 100 ++++++++++++++
> .../arch/x86/sapphirerapids/metricgroups.json | 118 +++++++++++++++++
> .../arch/x86/skylake/metricgroups.json | 113 ++++++++++++++++
> .../arch/x86/skylakex/metricgroups.json | 114 ++++++++++++++++
> .../arch/x86/tigerlake/metricgroups.json | 113 ++++++++++++++++
> 18 files changed, 1896 insertions(+)
> create mode 100644 tools/perf/pmu-events/arch/x86/alderlake/metricgroups.json
> create mode 100644 tools/perf/pmu-events/arch/x86/alderlaken/metricgroups.json
> create mode 100644 tools/perf/pmu-events/arch/x86/broadwell/metricgroups.json
> create mode 100644 tools/perf/pmu-events/arch/x86/broadwellde/metricgroups.json
> create mode 100644 tools/perf/pmu-events/arch/x86/broadwellx/metricgroups.json
> create mode 100644 tools/perf/pmu-events/arch/x86/cascadelakex/metricgroups.json
> create mode 100644 tools/perf/pmu-events/arch/x86/haswell/metricgroups.json
> create mode 100644 tools/perf/pmu-events/arch/x86/haswellx/metricgroups.json
> create mode 100644 tools/perf/pmu-events/arch/x86/icelake/metricgroups.json
> create mode 100644 tools/perf/pmu-events/arch/x86/icelakex/metricgroups.json
> create mode 100644 tools/perf/pmu-events/arch/x86/ivybridge/metricgroups.json
> create mode 100644 tools/perf/pmu-events/arch/x86/ivytown/metricgroups.json
> create mode 100644 tools/perf/pmu-events/arch/x86/jaketown/metricgroups.json
> create mode 100644 tools/perf/pmu-events/arch/x86/sandybridge/metricgroups.json
> create mode 100644 tools/perf/pmu-events/arch/x86/sapphirerapids/metricgroups.json
> create mode 100644 tools/perf/pmu-events/arch/x86/skylake/metricgroups.json
> create mode 100644 tools/perf/pmu-events/arch/x86/skylakex/metricgroups.json
> create mode 100644 tools/perf/pmu-events/arch/x86/tigerlake/metricgroups.json
>
> diff --git a/tools/perf/pmu-events/arch/x86/alderlake/metricgroups.json b/tools/perf/pmu-events/arch/x86/alderlake/metricgroups.json
> new file mode 100644
> index 000000000000..273ccfb0ed6f
> --- /dev/null
> +++ b/tools/perf/pmu-events/arch/x86/alderlake/metricgroups.json
> @@ -0,0 +1,122 @@
> +{
> + "Backend": "Grouping from metrics spreadsheet",
> + "Bad": "Grouping from metrics spreadsheet",
> + "BadSpec": "Grouping from metrics spreadsheet",
> + "BigFoot": "Grouping from metrics spreadsheet",
> + "BrMispredicts": "Grouping from metrics spreadsheet",
> + "Branches": "Grouping from metrics spreadsheet",
The description "Grouping from metrics spreadsheet" seems meaningless,
and may bring confusion information, e.g., what is metrics spreadsheet?
I think it may be better to leave it empty if we don't have a proper
description for the metricgroup.
Other descriptions look good to me.
Thanks,
Kan
^ permalink raw reply [flat|nested] 9+ messages in thread* Re: [PATCH v1 15/15] perf vendor events intel: Add metricgroup descriptions for all models
2023-05-16 17:53 ` [PATCH v1 15/15] perf vendor events intel: Add metricgroup descriptions for all models Liang, Kan
@ 2023-05-16 18:19 ` Ian Rogers
2023-05-16 21:41 ` Liang, Kan
0 siblings, 1 reply; 9+ messages in thread
From: Ian Rogers @ 2023-05-16 18:19 UTC (permalink / raw)
To: Liang, Kan
Cc: Peter Zijlstra, Ingo Molnar, Arnaldo Carvalho de Melo,
Mark Rutland, Alexander Shishkin, Jiri Olsa, Namhyung Kim,
Adrian Hunter, John Garry, Kajol Jain, Thomas Richter,
linux-kernel, linux-perf-users
On Tue, May 16, 2023 at 10:54 AM Liang, Kan <kan.liang@linux.intel.com> wrote:
>
>
>
> On 2023-05-15 5:58 p.m., Ian Rogers wrote:
> > Add metric group descriptions created by:
> > https://github.com/intel/perfmon/blob/main/scripts/create_perf_json.py
> > The descriptions add some additional detail in perf list.
> >
> > Signed-off-by: Ian Rogers <irogers@google.com>
> > ---
> > .../arch/x86/alderlake/metricgroups.json | 122 ++++++++++++++++++
> > .../arch/x86/alderlaken/metricgroups.json | 26 ++++
> > .../arch/x86/broadwell/metricgroups.json | 107 +++++++++++++++
> > .../arch/x86/broadwellde/metricgroups.json | 107 +++++++++++++++
> > .../arch/x86/broadwellx/metricgroups.json | 107 +++++++++++++++
> > .../arch/x86/cascadelakex/metricgroups.json | 114 ++++++++++++++++
> > .../arch/x86/haswell/metricgroups.json | 107 +++++++++++++++
> > .../arch/x86/haswellx/metricgroups.json | 107 +++++++++++++++
> > .../arch/x86/icelake/metricgroups.json | 113 ++++++++++++++++
> > .../arch/x86/icelakex/metricgroups.json | 114 ++++++++++++++++
> > .../arch/x86/ivybridge/metricgroups.json | 107 +++++++++++++++
> > .../arch/x86/ivytown/metricgroups.json | 107 +++++++++++++++
> > .../arch/x86/jaketown/metricgroups.json | 100 ++++++++++++++
> > .../arch/x86/sandybridge/metricgroups.json | 100 ++++++++++++++
> > .../arch/x86/sapphirerapids/metricgroups.json | 118 +++++++++++++++++
> > .../arch/x86/skylake/metricgroups.json | 113 ++++++++++++++++
> > .../arch/x86/skylakex/metricgroups.json | 114 ++++++++++++++++
> > .../arch/x86/tigerlake/metricgroups.json | 113 ++++++++++++++++
> > 18 files changed, 1896 insertions(+)
> > create mode 100644 tools/perf/pmu-events/arch/x86/alderlake/metricgroups.json
> > create mode 100644 tools/perf/pmu-events/arch/x86/alderlaken/metricgroups.json
> > create mode 100644 tools/perf/pmu-events/arch/x86/broadwell/metricgroups.json
> > create mode 100644 tools/perf/pmu-events/arch/x86/broadwellde/metricgroups.json
> > create mode 100644 tools/perf/pmu-events/arch/x86/broadwellx/metricgroups.json
> > create mode 100644 tools/perf/pmu-events/arch/x86/cascadelakex/metricgroups.json
> > create mode 100644 tools/perf/pmu-events/arch/x86/haswell/metricgroups.json
> > create mode 100644 tools/perf/pmu-events/arch/x86/haswellx/metricgroups.json
> > create mode 100644 tools/perf/pmu-events/arch/x86/icelake/metricgroups.json
> > create mode 100644 tools/perf/pmu-events/arch/x86/icelakex/metricgroups.json
> > create mode 100644 tools/perf/pmu-events/arch/x86/ivybridge/metricgroups.json
> > create mode 100644 tools/perf/pmu-events/arch/x86/ivytown/metricgroups.json
> > create mode 100644 tools/perf/pmu-events/arch/x86/jaketown/metricgroups.json
> > create mode 100644 tools/perf/pmu-events/arch/x86/sandybridge/metricgroups.json
> > create mode 100644 tools/perf/pmu-events/arch/x86/sapphirerapids/metricgroups.json
> > create mode 100644 tools/perf/pmu-events/arch/x86/skylake/metricgroups.json
> > create mode 100644 tools/perf/pmu-events/arch/x86/skylakex/metricgroups.json
> > create mode 100644 tools/perf/pmu-events/arch/x86/tigerlake/metricgroups.json
> >
> > diff --git a/tools/perf/pmu-events/arch/x86/alderlake/metricgroups.json b/tools/perf/pmu-events/arch/x86/alderlake/metricgroups.json
> > new file mode 100644
> > index 000000000000..273ccfb0ed6f
> > --- /dev/null
> > +++ b/tools/perf/pmu-events/arch/x86/alderlake/metricgroups.json
> > @@ -0,0 +1,122 @@
> > +{
> > + "Backend": "Grouping from metrics spreadsheet",
> > + "Bad": "Grouping from metrics spreadsheet",
> > + "BadSpec": "Grouping from metrics spreadsheet",
> > + "BigFoot": "Grouping from metrics spreadsheet",
> > + "BrMispredicts": "Grouping from metrics spreadsheet",
> > + "Branches": "Grouping from metrics spreadsheet",
>
> The description "Grouping from metrics spreadsheet" seems meaningless,
> and may bring confusion information, e.g., what is metrics spreadsheet?
> I think it may be better to leave it empty if we don't have a proper
> description for the metricgroup.
Thanks Kan, the TMA spreadsheet has a column (AG in version 4.5-full)
called Metric Group. For the level 3 ICache_Misses metric the value of
the cell is "BigFoot;FetchLat;IcMiss". The converter will put these
groups alongside the groups derived by level, parent/child
relationship. A group like "BigFoot" could use some explanation. The
intent with saying "Grouping from metrics spreadsheet" is to at least
give the origin of the metric group which I think is at least a little
bit better than saying nothing and not knowing the origin of where the
grouping comes from. I'd like to do more, so I think we can iterate
and improve, but I think this first step has some value.
Thanks,
Ian
> Other descriptions look good to me.
>
> Thanks,
> Kan
^ permalink raw reply [flat|nested] 9+ messages in thread* Re: [PATCH v1 15/15] perf vendor events intel: Add metricgroup descriptions for all models
2023-05-16 18:19 ` Ian Rogers
@ 2023-05-16 21:41 ` Liang, Kan
0 siblings, 0 replies; 9+ messages in thread
From: Liang, Kan @ 2023-05-16 21:41 UTC (permalink / raw)
To: Ian Rogers
Cc: Peter Zijlstra, Ingo Molnar, Arnaldo Carvalho de Melo,
Mark Rutland, Alexander Shishkin, Jiri Olsa, Namhyung Kim,
Adrian Hunter, John Garry, Kajol Jain, Thomas Richter,
linux-kernel, linux-perf-users
On 2023-05-16 2:19 p.m., Ian Rogers wrote:
> On Tue, May 16, 2023 at 10:54 AM Liang, Kan <kan.liang@linux.intel.com> wrote:
>>
>>
>>
>> On 2023-05-15 5:58 p.m., Ian Rogers wrote:
>>> Add metric group descriptions created by:
>>> https://github.com/intel/perfmon/blob/main/scripts/create_perf_json.py
>>> The descriptions add some additional detail in perf list.
>>>
>>> Signed-off-by: Ian Rogers <irogers@google.com>
>>> ---
>>> .../arch/x86/alderlake/metricgroups.json | 122 ++++++++++++++++++
>>> .../arch/x86/alderlaken/metricgroups.json | 26 ++++
>>> .../arch/x86/broadwell/metricgroups.json | 107 +++++++++++++++
>>> .../arch/x86/broadwellde/metricgroups.json | 107 +++++++++++++++
>>> .../arch/x86/broadwellx/metricgroups.json | 107 +++++++++++++++
>>> .../arch/x86/cascadelakex/metricgroups.json | 114 ++++++++++++++++
>>> .../arch/x86/haswell/metricgroups.json | 107 +++++++++++++++
>>> .../arch/x86/haswellx/metricgroups.json | 107 +++++++++++++++
>>> .../arch/x86/icelake/metricgroups.json | 113 ++++++++++++++++
>>> .../arch/x86/icelakex/metricgroups.json | 114 ++++++++++++++++
>>> .../arch/x86/ivybridge/metricgroups.json | 107 +++++++++++++++
>>> .../arch/x86/ivytown/metricgroups.json | 107 +++++++++++++++
>>> .../arch/x86/jaketown/metricgroups.json | 100 ++++++++++++++
>>> .../arch/x86/sandybridge/metricgroups.json | 100 ++++++++++++++
>>> .../arch/x86/sapphirerapids/metricgroups.json | 118 +++++++++++++++++
>>> .../arch/x86/skylake/metricgroups.json | 113 ++++++++++++++++
>>> .../arch/x86/skylakex/metricgroups.json | 114 ++++++++++++++++
>>> .../arch/x86/tigerlake/metricgroups.json | 113 ++++++++++++++++
>>> 18 files changed, 1896 insertions(+)
>>> create mode 100644 tools/perf/pmu-events/arch/x86/alderlake/metricgroups.json
>>> create mode 100644 tools/perf/pmu-events/arch/x86/alderlaken/metricgroups.json
>>> create mode 100644 tools/perf/pmu-events/arch/x86/broadwell/metricgroups.json
>>> create mode 100644 tools/perf/pmu-events/arch/x86/broadwellde/metricgroups.json
>>> create mode 100644 tools/perf/pmu-events/arch/x86/broadwellx/metricgroups.json
>>> create mode 100644 tools/perf/pmu-events/arch/x86/cascadelakex/metricgroups.json
>>> create mode 100644 tools/perf/pmu-events/arch/x86/haswell/metricgroups.json
>>> create mode 100644 tools/perf/pmu-events/arch/x86/haswellx/metricgroups.json
>>> create mode 100644 tools/perf/pmu-events/arch/x86/icelake/metricgroups.json
>>> create mode 100644 tools/perf/pmu-events/arch/x86/icelakex/metricgroups.json
>>> create mode 100644 tools/perf/pmu-events/arch/x86/ivybridge/metricgroups.json
>>> create mode 100644 tools/perf/pmu-events/arch/x86/ivytown/metricgroups.json
>>> create mode 100644 tools/perf/pmu-events/arch/x86/jaketown/metricgroups.json
>>> create mode 100644 tools/perf/pmu-events/arch/x86/sandybridge/metricgroups.json
>>> create mode 100644 tools/perf/pmu-events/arch/x86/sapphirerapids/metricgroups.json
>>> create mode 100644 tools/perf/pmu-events/arch/x86/skylake/metricgroups.json
>>> create mode 100644 tools/perf/pmu-events/arch/x86/skylakex/metricgroups.json
>>> create mode 100644 tools/perf/pmu-events/arch/x86/tigerlake/metricgroups.json
>>>
>>> diff --git a/tools/perf/pmu-events/arch/x86/alderlake/metricgroups.json b/tools/perf/pmu-events/arch/x86/alderlake/metricgroups.json
>>> new file mode 100644
>>> index 000000000000..273ccfb0ed6f
>>> --- /dev/null
>>> +++ b/tools/perf/pmu-events/arch/x86/alderlake/metricgroups.json
>>> @@ -0,0 +1,122 @@
>>> +{
>>> + "Backend": "Grouping from metrics spreadsheet",
>>> + "Bad": "Grouping from metrics spreadsheet",
>>> + "BadSpec": "Grouping from metrics spreadsheet",
>>> + "BigFoot": "Grouping from metrics spreadsheet",
>>> + "BrMispredicts": "Grouping from metrics spreadsheet",
>>> + "Branches": "Grouping from metrics spreadsheet",
>>
>> The description "Grouping from metrics spreadsheet" seems meaningless,
>> and may bring confusion information, e.g., what is metrics spreadsheet?
>> I think it may be better to leave it empty if we don't have a proper
>> description for the metricgroup.
>
> Thanks Kan, the TMA spreadsheet has a column (AG in version 4.5-full)
> called Metric Group. For the level 3 ICache_Misses metric the value of
> the cell is "BigFoot;FetchLat;IcMiss". The converter will put these
> groups alongside the groups derived by level, parent/child
> relationship. A group like "BigFoot" could use some explanation. The
> intent with saying "Grouping from metrics spreadsheet" is to at least
> give the origin of the metric group which I think is at least a little
> bit better than saying nothing and not knowing the origin of where the
> grouping comes from. I'd like to do more, so I think we can iterate
> and improve, but I think this first step has some value.
>
So the groups with "Grouping from metrics spreadsheet" are all from the
"Metric Group" column, right?
I think the official name of TMA is Top-down Microarchitecture Analysis
Metrics. Can we change the description to "Grouping from Top-down
Microarchitecture Analysis Metrics spreadsheet"? If the end user
searches the name, they can be pointed to the correct spreadsheet file.
Thanks,
Kan
^ permalink raw reply [flat|nested] 9+ messages in thread