perf vendor events: Update Intel events
Subsystems (2)
Ian Rogers
reviewerv2 cover perf vendor events: Update Intel events
Update:
- ADL/ADLN v1.29 -> 1.31
- ARL 1.08 -> 1.09
- CLX 1.23 -> 1.25
- EMR 1.11 -> 1.14
- GRR 1.07 -> 1.09
- GNR 1.08 -> 1.10
- ICX 1.27 -> 1.28
- LNL 1.11 -> 1.14
- MTL 1.13 -> 1.14
- PTL new v1.00
- SPR 1.25 -> 1.28
- SRF 1.09 -> 1.11
- SKX 1.36 -> 1.37
- TGL 1.17 -> 1.18
Updates from:
https://github.com/intel/perfmon
with:
https://github.com/intel/perfmon/pull/312
Running the script:
https://github.com/intel/perfmon/blob/main/scripts/create_perf_json.py
v2: Rebase, no conflicts.
Ian Rogers (15):
perf vendor events: Update Alderlake events perf vendor events: Update AlderlakeN events perf vendor events: Update Arrowlake events perf vendor events: Update CascadelakeX events perf vendor events: Update EmeraldRapids events perf vendor events: Update GrandRidge events perf vendor events: Update GraniteRapids events perf vendor events: Update IcelakeX events perf vendor events: Update LunarLake events perf vendor events: Update MeteorLake events perf vendor events: Add PantherLake events perf vendor events: Update SapphireRapids events perf vendor events: Update SierraForest events perf vendor events: Update SkylakeX events perf vendor events: Update TigerLake events
.../pmu-events/arch/x86/alderlake/cache.json | 56 ++- .../arch/x86/alderlake/floating-point.json | 1 - .../pmu-events/arch/x86/alderlake/other.json | 1 - .../arch/x86/alderlake/pipeline.json | 44 +-- .../arch/x86/alderlake/virtual-memory.json | 3 - .../pmu-events/arch/x86/alderlaken/cache.json | 52 +-- .../arch/x86/alderlaken/floating-point.json | 1 - .../pmu-events/arch/x86/alderlaken/other.json | 1 - .../arch/x86/alderlaken/pipeline.json | 42 +-- .../arch/x86/alderlaken/virtual-memory.json | 3 - .../pmu-events/arch/x86/arrowlake/cache.json | 13 +- .../arch/x86/arrowlake/frontend.json | 135 ++++++++ .../arch/x86/cascadelakex/floating-point.json | 6 +- .../arch/x86/cascadelakex/pipeline.json | 2 +- .../arch/x86/emeraldrapids/pipeline.json | 2 +- .../arch/x86/emeraldrapids/uncore-io.json | 12 + .../arch/x86/emeraldrapids/uncore-memory.json | 20 ++ .../arch/x86/grandridge/grr-metrics.json | 30 +- .../x86/grandridge/uncore-interconnect.json | 10 + .../arch/x86/grandridge/uncore-io.json | 12 + .../arch/x86/graniterapids/cache.json | 9 + .../arch/x86/graniterapids/counter.json | 10 +- .../arch/x86/graniterapids/gnr-metrics.json | 36 ++ .../arch/x86/graniterapids/pipeline.json | 2 +- .../graniterapids/uncore-interconnect.json | 19 - .../arch/x86/graniterapids/uncore-io.json | 27 +- .../arch/x86/icelakex/pipeline.json | 2 +- .../arch/x86/icelakex/uncore-cache.json | 2 - .../pmu-events/arch/x86/lunarlake/cache.json | 11 + .../arch/x86/lunarlake/pipeline.json | 18 +- .../arch/x86/lunarlake/virtual-memory.json | 18 - tools/perf/pmu-events/arch/x86/mapfile.csv | 29 +- .../pmu-events/arch/x86/meteorlake/cache.json | 2 +- .../arch/x86/meteorlake/frontend.json | 72 ++++ .../arch/x86/meteorlake/pipeline.json | 2 +- .../arch/x86/pantherlake/cache.json | 278 +++++++++++++++ .../arch/x86/pantherlake/counter.json | 12 + .../arch/x86/pantherlake/frontend.json | 30 ++ .../arch/x86/pantherlake/memory.json | 215 ++++++++++++ .../arch/x86/pantherlake/pipeline.json | 325 ++++++++++++++++++ .../arch/x86/pantherlake/virtual-memory.json | 62 ++++ .../arch/x86/sapphirerapids/pipeline.json | 2 +- .../arch/x86/sapphirerapids/uncore-io.json | 12 + .../x86/sapphirerapids/uncore-memory.json | 20 ++ .../arch/x86/sierraforest/frontend.json | 64 ++++ .../arch/x86/sierraforest/pipeline.json | 8 + .../arch/x86/sierraforest/srf-metrics.json | 48 +++ .../arch/x86/sierraforest/uncore-cache.json | 6 +- .../x86/sierraforest/uncore-interconnect.json | 53 ++- .../arch/x86/sierraforest/uncore-io.json | 27 +- .../arch/x86/skylakex/pipeline.json | 2 +- .../arch/x86/tigerlake/pipeline.json | 2 +- 52 files changed, 1623 insertions(+), 248 deletions(-) create mode 100644 tools/perf/pmu-events/arch/x86/pantherlake/cache.json create mode 100644 tools/perf/pmu-events/arch/x86/pantherlake/counter.json create mode 100644 tools/perf/pmu-events/arch/x86/pantherlake/frontend.json create mode 100644 tools/perf/pmu-events/arch/x86/pantherlake/memory.json create mode 100644 tools/perf/pmu-events/arch/x86/pantherlake/pipeline.json create mode 100644 tools/perf/pmu-events/arch/x86/pantherlake/virtual-memory.json
Signature · 1 line
2.50.0.727.gbf7dc18ff4-goog
Ian Rogers
reviewerPATCH v2 1/15 perf vendor events: Update Alderlake events
Update events from v1.29 to v1.31.
Bring in the event updates v1.31:
https://github.com/intel/perfmon/commit/5a1269c8af70e32a548e74e1fda736189c398ddc
https://github.com/intel/perfmon/commit/76c6d2c348c067e9ae1b616b35ee982da6d873b4
--- .../pmu-events/arch/x86/alderlake/cache.json | 56 +++++++------------ .../arch/x86/alderlake/floating-point.json | 1 - .../pmu-events/arch/x86/alderlake/other.json | 1 - .../arch/x86/alderlake/pipeline.json | 44 ++------------- .../arch/x86/alderlake/virtual-memory.json | 3 - tools/perf/pmu-events/arch/x86/mapfile.csv | 2 +- 6 files changed, 28 insertions(+), 79 deletions(-)
tools/perf/pmu-events/arch/x86/alderlake/cache.json +21 -35
@@ -728,7 +728,6 @@ "Data_LA": "1", "EventCode": "0xd1", "EventName": "MEM_LOAD_UOPS_RETIRED.DRAM_HIT", - "PublicDescription": "Counts the number of load uops retired that hit in DRAM. Available PDIST counters: 0", "SampleAfterValue": "200003", "UMask": "0x80", "Unit": "cpu_atom"
@@ -739,7 +738,6 @@ "Data_LA": "1", "EventCode": "0xd1", "EventName": "MEM_LOAD_UOPS_RETIRED.HITM", - "PublicDescription": "Counts the number of load uops retired that hit in the L3 cache, in which a snoop was required and modified data was forwarded from another core or module. Available PDIST counters: 0", "SampleAfterValue": "200003", "UMask": "0x20", "Unit": "cpu_atom"
@@ -750,7 +748,6 @@ "Data_LA": "1", "EventCode": "0xd1", "EventName": "MEM_LOAD_UOPS_RETIRED.L1_HIT", - "PublicDescription": "Counts the number of load uops retired that hit in the L1 data cache. Available PDIST counters: 0", "SampleAfterValue": "200003", "UMask": "0x1", "Unit": "cpu_atom"
@@ -761,7 +758,6 @@ "Data_LA": "1", "EventCode": "0xd1", "EventName": "MEM_LOAD_UOPS_RETIRED.L1_MISS", - "PublicDescription": "Counts the number of load uops retired that miss in the L1 data cache. Available PDIST counters: 0", "SampleAfterValue": "200003", "UMask": "0x8", "Unit": "cpu_atom"
@@ -772,7 +768,6 @@ "Data_LA": "1", "EventCode": "0xd1", "EventName": "MEM_LOAD_UOPS_RETIRED.L2_HIT", - "PublicDescription": "Counts the number of load uops retired that hit in the L2 cache. Available PDIST counters: 0", "SampleAfterValue": "200003", "UMask": "0x2", "Unit": "cpu_atom"
@@ -783,7 +778,6 @@ "Data_LA": "1", "EventCode": "0xd1", "EventName": "MEM_LOAD_UOPS_RETIRED.L2_MISS", - "PublicDescription": "Counts the number of load uops retired that miss in the L2 cache. Available PDIST counters: 0", "SampleAfterValue": "200003", "UMask": "0x10", "Unit": "cpu_atom"
@@ -794,7 +788,6 @@ "Data_LA": "1", "EventCode": "0xd1", "EventName": "MEM_LOAD_UOPS_RETIRED.L3_HIT", - "PublicDescription": "Counts the number of load uops retired that hit in the L3 cache. Available PDIST counters: 0", "SampleAfterValue": "200003", "UMask": "0x4", "Unit": "cpu_atom"
@@ -805,7 +798,6 @@ "Data_LA": "1", "EventCode": "0xd2", "EventName": "MEM_LOAD_UOPS_RETIRED_MISC.HIT_E_F", - "PublicDescription": "Counts the number of load uops retired that hit in the L3 cache, in which a snoop was required, and non-modified data was forwarded. Available PDIST counters: 0", "SampleAfterValue": "1000003", "UMask": "0x40", "Unit": "cpu_atom"
@@ -816,7 +808,6 @@ "Data_LA": "1", "EventCode": "0xd2", "EventName": "MEM_LOAD_UOPS_RETIRED_MISC.L3_MISS", - "PublicDescription": "Counts the number of load uops retired that miss in the L3 cache. Available PDIST counters: 0", "SampleAfterValue": "1000003", "UMask": "0x20", "Unit": "cpu_atom"
@@ -873,7 +864,7 @@ "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_UOPS_RETIRED.ALL_LOADS", - "PublicDescription": "Counts the total number of load uops retired. Available PDIST counters: 0", + "PublicDescription": "Counts the total number of load uops retired.", "SampleAfterValue": "200003", "UMask": "0x81", "Unit": "cpu_atom"
@@ -884,111 +875,111 @@ "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_UOPS_RETIRED.ALL_STORES", - "PublicDescription": "Counts the total number of store uops retired. Available PDIST counters: 0", + "PublicDescription": "Counts the total number of store uops retired.", "SampleAfterValue": "200003", "UMask": "0x82", "Unit": "cpu_atom" }, { "BriefDescription": "Counts the number of tagged loads with an instruction latency that exceeds or equals the threshold of 128 cycles as defined in MEC_CR_PEBS_LD_LAT_THRESHOLD (3F6H). Only counts with PEBS enabled.", - "Counter": "0,1", + "Counter": "0,1,2,3,4,5", "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_UOPS_RETIRED.LOAD_LATENCY_GT_128", "MSRIndex": "0x3F6", "MSRValue": "0x80", - "PublicDescription": "Counts the number of tagged loads with an instruction latency that exceeds or equals the threshold of 128 cycles as defined in MEC_CR_PEBS_LD_LAT_THRESHOLD (3F6H). Only counts with PEBS enabled. If a PEBS record is generated, will populate the PEBS Latency and PEBS Data Source fields accordingly. Available PDIST counters: 0", + "PublicDescription": "Counts the number of tagged loads with an instruction latency that exceeds or equals the threshold of 128 cycles as defined in MEC_CR_PEBS_LD_LAT_THRESHOLD (3F6H). Only counts with PEBS enabled. If a PEBS record is generated, will populate the PEBS Latency and PEBS Data Source fields accordingly.", "SampleAfterValue": "1000003", "UMask": "0x5", "Unit": "cpu_atom" }, { "BriefDescription": "Counts the number of tagged loads with an instruction latency that exceeds or equals the threshold of 16 cycles as defined in MEC_CR_PEBS_LD_LAT_THRESHOLD (3F6H). Only counts with PEBS enabled.", - "Counter": "0,1", + "Counter": "0,1,2,3,4,5", "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_UOPS_RETIRED.LOAD_LATENCY_GT_16", "MSRIndex": "0x3F6", "MSRValue": "0x10", - "PublicDescription": "Counts the number of tagged loads with an instruction latency that exceeds or equals the threshold of 16 cycles as defined in MEC_CR_PEBS_LD_LAT_THRESHOLD (3F6H). Only counts with PEBS enabled. If a PEBS record is generated, will populate the PEBS Latency and PEBS Data Source fields accordingly. Available PDIST counters: 0", + "PublicDescription": "Counts the number of tagged loads with an instruction latency that exceeds or equals the threshold of 16 cycles as defined in MEC_CR_PEBS_LD_LAT_THRESHOLD (3F6H). Only counts with PEBS enabled. If a PEBS record is generated, will populate the PEBS Latency and PEBS Data Source fields accordingly.", "SampleAfterValue": "1000003", "UMask": "0x5", "Unit": "cpu_atom" }, { "BriefDescription": "Counts the number of tagged loads with an instruction latency that exceeds or equals the threshold of 256 cycles as defined in MEC_CR_PEBS_LD_LAT_THRESHOLD (3F6H). Only counts with PEBS enabled.", - "Counter": "0,1", + "Counter": "0,1,2,3,4,5", "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_UOPS_RETIRED.LOAD_LATENCY_GT_256", "MSRIndex": "0x3F6", "MSRValue": "0x100", - "PublicDescription": "Counts the number of tagged loads with an instruction latency that exceeds or equals the threshold of 256 cycles as defined in MEC_CR_PEBS_LD_LAT_THRESHOLD (3F6H). Only counts with PEBS enabled. If a PEBS record is generated, will populate the PEBS Latency and PEBS Data Source fields accordingly. Available PDIST counters: 0", + "PublicDescription": "Counts the number of tagged loads with an instruction latency that exceeds or equals the threshold of 256 cycles as defined in MEC_CR_PEBS_LD_LAT_THRESHOLD (3F6H). Only counts with PEBS enabled. If a PEBS record is generated, will populate the PEBS Latency and PEBS Data Source fields accordingly.", "SampleAfterValue": "1000003", "UMask": "0x5", "Unit": "cpu_atom" }, { "BriefDescription": "Counts the number of tagged loads with an instruction latency that exceeds or equals the threshold of 32 cycles as defined in MEC_CR_PEBS_LD_LAT_THRESHOLD (3F6H). Only counts with PEBS enabled.", - "Counter": "0,1", + "Counter": "0,1,2,3,4,5", "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_UOPS_RETIRED.LOAD_LATENCY_GT_32", "MSRIndex": "0x3F6", "MSRValue": "0x20", - "PublicDescription": "Counts the number of tagged loads with an instruction latency that exceeds or equals the threshold of 32 cycles as defined in MEC_CR_PEBS_LD_LAT_THRESHOLD (3F6H). Only counts with PEBS enabled. If a PEBS record is generated, will populate the PEBS Latency and PEBS Data Source fields accordingly. Available PDIST counters: 0", + "PublicDescription": "Counts the number of tagged loads with an instruction latency that exceeds or equals the threshold of 32 cycles as defined in MEC_CR_PEBS_LD_LAT_THRESHOLD (3F6H). Only counts with PEBS enabled. If a PEBS record is generated, will populate the PEBS Latency and PEBS Data Source fields accordingly.", "SampleAfterValue": "1000003", "UMask": "0x5", "Unit": "cpu_atom" }, { "BriefDescription": "Counts the number of tagged loads with an instruction latency that exceeds or equals the threshold of 4 cycles as defined in MEC_CR_PEBS_LD_LAT_THRESHOLD (3F6H). Only counts with PEBS enabled.", - "Counter": "0,1", + "Counter": "0,1,2,3,4,5", "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_UOPS_RETIRED.LOAD_LATENCY_GT_4", "MSRIndex": "0x3F6", "MSRValue": "0x4", - "PublicDescription": "Counts the number of tagged loads with an instruction latency that exceeds or equals the threshold of 4 cycles as defined in MEC_CR_PEBS_LD_LAT_THRESHOLD (3F6H). Only counts with PEBS enabled. If a PEBS record is generated, will populate the PEBS Latency and PEBS Data Source fields accordingly. Available PDIST counters: 0", + "PublicDescription": "Counts the number of tagged loads with an instruction latency that exceeds or equals the threshold of 4 cycles as defined in MEC_CR_PEBS_LD_LAT_THRESHOLD (3F6H). Only counts with PEBS enabled. If a PEBS record is generated, will populate the PEBS Latency and PEBS Data Source fields accordingly.", "SampleAfterValue": "1000003", "UMask": "0x5", "Unit": "cpu_atom" }, { "BriefDescription": "Counts the number of tagged loads with an instruction latency that exceeds or equals the threshold of 512 cycles as defined in MEC_CR_PEBS_LD_LAT_THRESHOLD (3F6H). Only counts with PEBS enabled.", - "Counter": "0,1", + "Counter": "0,1,2,3,4,5", "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_UOPS_RETIRED.LOAD_LATENCY_GT_512", "MSRIndex": "0x3F6", "MSRValue": "0x200", - "PublicDescription": "Counts the number of tagged loads with an instruction latency that exceeds or equals the threshold of 512 cycles as defined in MEC_CR_PEBS_LD_LAT_THRESHOLD (3F6H). Only counts with PEBS enabled. If a PEBS record is generated, will populate the PEBS Latency and PEBS Data Source fields accordingly. Available PDIST counters: 0", + "PublicDescription": "Counts the number of tagged loads with an instruction latency that exceeds or equals the threshold of 512 cycles as defined in MEC_CR_PEBS_LD_LAT_THRESHOLD (3F6H). Only counts with PEBS enabled. If a PEBS record is generated, will populate the PEBS Latency and PEBS Data Source fields accordingly.", "SampleAfterValue": "1000003", "UMask": "0x5", "Unit": "cpu_atom" }, { "BriefDescription": "Counts the number of tagged loads with an instruction latency that exceeds or equals the threshold of 64 cycles as defined in MEC_CR_PEBS_LD_LAT_THRESHOLD (3F6H). Only counts with PEBS enabled.", - "Counter": "0,1", + "Counter": "0,1,2,3,4,5", "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_UOPS_RETIRED.LOAD_LATENCY_GT_64", "MSRIndex": "0x3F6", "MSRValue": "0x40", - "PublicDescription": "Counts the number of tagged loads with an instruction latency that exceeds or equals the threshold of 64 cycles as defined in MEC_CR_PEBS_LD_LAT_THRESHOLD (3F6H). Only counts with PEBS enabled. If a PEBS record is generated, will populate the PEBS Latency and PEBS Data Source fields accordingly. Available PDIST counters: 0", + "PublicDescription": "Counts the number of tagged loads with an instruction latency that exceeds or equals the threshold of 64 cycles as defined in MEC_CR_PEBS_LD_LAT_THRESHOLD (3F6H). Only counts with PEBS enabled. If a PEBS record is generated, will populate the PEBS Latency and PEBS Data Source fields accordingly.", "SampleAfterValue": "1000003", "UMask": "0x5", "Unit": "cpu_atom" }, { "BriefDescription": "Counts the number of tagged loads with an instruction latency that exceeds or equals the threshold of 8 cycles as defined in MEC_CR_PEBS_LD_LAT_THRESHOLD (3F6H). Only counts with PEBS enabled.", - "Counter": "0,1", + "Counter": "0,1,2,3,4,5", "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_UOPS_RETIRED.LOAD_LATENCY_GT_8", "MSRIndex": "0x3F6", "MSRValue": "0x8", - "PublicDescription": "Counts the number of tagged loads with an instruction latency that exceeds or equals the threshold of 8 cycles as defined in MEC_CR_PEBS_LD_LAT_THRESHOLD (3F6H). Only counts with PEBS enabled. If a PEBS record is generated, will populate the PEBS Latency and PEBS Data Source fields accordingly. Available PDIST counters: 0", + "PublicDescription": "Counts the number of tagged loads with an instruction latency that exceeds or equals the threshold of 8 cycles as defined in MEC_CR_PEBS_LD_LAT_THRESHOLD (3F6H). Only counts with PEBS enabled. If a PEBS record is generated, will populate the PEBS Latency and PEBS Data Source fields accordingly.", "SampleAfterValue": "1000003", "UMask": "0x5", "Unit": "cpu_atom"
@@ -999,7 +990,6 @@ "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_UOPS_RETIRED.LOCK_LOADS", - "PublicDescription": "Counts the number of load uops retired that performed one or more locks. Available PDIST counters: 0", "SampleAfterValue": "200003", "UMask": "0x21", "Unit": "cpu_atom"
@@ -1010,7 +1000,6 @@ "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_UOPS_RETIRED.SPLIT_LOADS", - "PublicDescription": "Counts the number of retired split load uops. Available PDIST counters: 0", "SampleAfterValue": "200003", "UMask": "0x41", "Unit": "cpu_atom"
@@ -1021,7 +1010,6 @@ "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_UOPS_RETIRED.STLB_MISS", - "PublicDescription": "Counts the total number of load and store uops retired that missed in the second level TLB. Available PDIST counters: 0", "SampleAfterValue": "200003", "UMask": "0x13", "Unit": "cpu_atom"
@@ -1032,7 +1020,6 @@ "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_UOPS_RETIRED.STLB_MISS_LOADS", - "PublicDescription": "Counts the number of load ops retired that miss in the second Level TLB. Available PDIST counters: 0", "SampleAfterValue": "200003", "UMask": "0x11", "Unit": "cpu_atom"
@@ -1043,7 +1030,6 @@ "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_UOPS_RETIRED.STLB_MISS_STORES", - "PublicDescription": "Counts the number of store ops retired that miss in the second level TLB. Available PDIST counters: 0", "SampleAfterValue": "200003", "UMask": "0x12", "Unit": "cpu_atom"
@@ -1054,7 +1040,7 @@ "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_UOPS_RETIRED.STORE_LATENCY", - "PublicDescription": "Counts the number of stores uops retired. Counts with or without PEBS enabled. If PEBS is enabled and a PEBS record is generated, will populate PEBS Latency and PEBS Data Source fields accordingly. Available PDIST counters: 0", + "PublicDescription": "Counts the number of stores uops retired. Counts with or without PEBS enabled. If PEBS is enabled and a PEBS record is generated, will populate PEBS Latency and PEBS Data Source fields accordingly.", "SampleAfterValue": "1000003", "UMask": "0x6", "Unit": "cpu_atom"
@@ -1478,12 +1464,12 @@ "Unit": "cpu_core" }, { - "BriefDescription": "For every cycle where the core is waiting on at least 1 outstanding Demand RFO request, increments by 1.", + "BriefDescription": "Cycles with offcore outstanding demand rfo reads transactions in SuperQueue (SQ), queue to uncore.", "Counter": "0,1,2,3", "CounterMask": "1", "EventCode": "0x20", "EventName": "OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO", - "PublicDescription": "OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO Available PDIST counters: 0", + "PublicDescription": "Counts the number of offcore outstanding demand rfo Reads transactions in the super queue every cycle. The 'Offcore outstanding' state of the transaction lasts from the L2 miss until the sending transaction completion to requestor (SQ deallocation). See the corresponding Umask under OFFCORE_REQUESTS. Available PDIST counters: 0", "SampleAfterValue": "1000003", "UMask": "0x4", "Unit": "cpu_core"
tools/perf/pmu-events/arch/x86/alderlake/floating-point.json +0 -1
@@ -213,7 +213,6 @@ "Counter": "0,1,2,3,4,5", "EventCode": "0xc2", "EventName": "UOPS_RETIRED.FPDIV", - "PublicDescription": "Counts the number of floating point divide uops retired (x87 and SSE, including x87 sqrt). Available PDIST counters: 0", "SampleAfterValue": "2000003", "UMask": "0x8", "Unit": "cpu_atom"
tools/perf/pmu-events/arch/x86/alderlake/other.json +0 -1
@@ -55,7 +55,6 @@ "Deprecated": "1", "EventCode": "0xe4", "EventName": "LBR_INSERTS.ANY", - "PublicDescription": "This event is deprecated. [This event is alias to MISC_RETIRED.LBR_INSERTS] Available PDIST counters: 0", "SampleAfterValue": "1000003", "UMask": "0x1", "Unit": "cpu_atom"
tools/perf/pmu-events/arch/x86/alderlake/pipeline.json +6 -38
272 lines not on this page · raw message
tools/perf/pmu-events/arch/x86/alderlake/virtual-memory.json +0 -3
@@ -266,7 +266,6 @@ "Deprecated": "1", "EventCode": "0xd0", "EventName": "MEM_UOPS_RETIRED.DTLB_MISS", - "PublicDescription": "This event is deprecated. Refer to new event MEM_UOPS_RETIRED.STLB_MISS Available PDIST counters: 0", "SampleAfterValue": "200003", "UMask": "0x13", "Unit": "cpu_atom"
@@ -278,7 +277,6 @@ "Deprecated": "1", "EventCode": "0xd0", "EventName": "MEM_UOPS_RETIRED.DTLB_MISS_LOADS", - "PublicDescription": "This event is deprecated. Refer to new event MEM_UOPS_RETIRED.STLB_MISS_LOADS Available PDIST counters: 0", "SampleAfterValue": "200003", "UMask": "0x11", "Unit": "cpu_atom"
@@ -290,7 +288,6 @@ "Deprecated": "1", "EventCode": "0xd0", "EventName": "MEM_UOPS_RETIRED.DTLB_MISS_STORES", - "PublicDescription": "This event is deprecated. Refer to new event MEM_UOPS_RETIRED.STLB_MISS_STORES Available PDIST counters: 0", "SampleAfterValue": "200003", "UMask": "0x12", "Unit": "cpu_atom"
tools/perf/pmu-events/arch/x86/mapfile.csv +1 -1
@@ -1,5 +1,5 @@ Family-model,Version,Filename,EventType -GenuineIntel-6-(97|9A|B7|BA|BF),v1.29,alderlake,core +GenuineIntel-6-(97|9A|B7|BA|BF),v1.31,alderlake,core GenuineIntel-6-BE,v1.29,alderlaken,core GenuineIntel-6-C[56],v1.08,arrowlake,core GenuineIntel-6-(1C|26|27|35|36),v5,bonnell,core
Signature · 1 line
2.50.0.727.gbf7dc18ff4-goog
Ian Rogers
reviewerPATCH v2 2/15 perf vendor events: Update AlderlakeN events
In reply to Ian RogersUpdate events from v1.29 to v1.31.
Bring in the event updates v1.31:
https://github.com/intel/perfmon/commit/5a1269c8af70e32a548e74e1fda736189c398ddc
https://github.com/intel/perfmon/commit/76c6d2c348c067e9ae1b616b35ee982da6d873b4
--- .../pmu-events/arch/x86/alderlaken/cache.json | 52 +++++++------------ .../arch/x86/alderlaken/floating-point.json | 1 - .../pmu-events/arch/x86/alderlaken/other.json | 1 - .../arch/x86/alderlaken/pipeline.json | 42 ++------------- .../arch/x86/alderlaken/virtual-memory.json | 3 -- tools/perf/pmu-events/arch/x86/mapfile.csv | 2 +- 6 files changed, 25 insertions(+), 76 deletions(-)
tools/perf/pmu-events/arch/x86/alderlaken/cache.json +19 -33
@@ -118,7 +118,6 @@ "Data_LA": "1", "EventCode": "0xd1", "EventName": "MEM_LOAD_UOPS_RETIRED.DRAM_HIT", - "PublicDescription": "Counts the number of load uops retired that hit in DRAM. Available PDIST counters: 0", "SampleAfterValue": "200003", "UMask": "0x80" },
@@ -128,7 +127,6 @@ "Data_LA": "1", "EventCode": "0xd1", "EventName": "MEM_LOAD_UOPS_RETIRED.HITM", - "PublicDescription": "Counts the number of load uops retired that hit in the L3 cache, in which a snoop was required and modified data was forwarded from another core or module. Available PDIST counters: 0", "SampleAfterValue": "200003", "UMask": "0x20" },
@@ -138,7 +136,6 @@ "Data_LA": "1", "EventCode": "0xd1", "EventName": "MEM_LOAD_UOPS_RETIRED.L1_HIT", - "PublicDescription": "Counts the number of load uops retired that hit in the L1 data cache. Available PDIST counters: 0", "SampleAfterValue": "200003", "UMask": "0x1" },
@@ -148,7 +145,6 @@ "Data_LA": "1", "EventCode": "0xd1", "EventName": "MEM_LOAD_UOPS_RETIRED.L1_MISS", - "PublicDescription": "Counts the number of load uops retired that miss in the L1 data cache. Available PDIST counters: 0", "SampleAfterValue": "200003", "UMask": "0x8" },
@@ -158,7 +154,6 @@ "Data_LA": "1", "EventCode": "0xd1", "EventName": "MEM_LOAD_UOPS_RETIRED.L2_HIT", - "PublicDescription": "Counts the number of load uops retired that hit in the L2 cache. Available PDIST counters: 0", "SampleAfterValue": "200003", "UMask": "0x2" },
@@ -168,7 +163,6 @@ "Data_LA": "1", "EventCode": "0xd1", "EventName": "MEM_LOAD_UOPS_RETIRED.L2_MISS", - "PublicDescription": "Counts the number of load uops retired that miss in the L2 cache. Available PDIST counters: 0", "SampleAfterValue": "200003", "UMask": "0x10" },
@@ -178,7 +172,6 @@ "Data_LA": "1", "EventCode": "0xd1", "EventName": "MEM_LOAD_UOPS_RETIRED.L3_HIT", - "PublicDescription": "Counts the number of load uops retired that hit in the L3 cache. Available PDIST counters: 0", "SampleAfterValue": "200003", "UMask": "0x4" },
@@ -188,7 +181,6 @@ "Data_LA": "1", "EventCode": "0xd2", "EventName": "MEM_LOAD_UOPS_RETIRED_MISC.HIT_E_F", - "PublicDescription": "Counts the number of load uops retired that hit in the L3 cache, in which a snoop was required, and non-modified data was forwarded. Available PDIST counters: 0", "SampleAfterValue": "1000003", "UMask": "0x40" },
@@ -198,7 +190,6 @@ "Data_LA": "1", "EventCode": "0xd2", "EventName": "MEM_LOAD_UOPS_RETIRED_MISC.L3_MISS", - "PublicDescription": "Counts the number of load uops retired that miss in the L3 cache. Available PDIST counters: 0", "SampleAfterValue": "1000003", "UMask": "0x20" },
@@ -240,7 +231,7 @@ "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_UOPS_RETIRED.ALL_LOADS", - "PublicDescription": "Counts the total number of load uops retired. Available PDIST counters: 0", + "PublicDescription": "Counts the total number of load uops retired.", "SampleAfterValue": "200003", "UMask": "0x81" },
@@ -250,103 +241,103 @@ "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_UOPS_RETIRED.ALL_STORES", - "PublicDescription": "Counts the total number of store uops retired. Available PDIST counters: 0", + "PublicDescription": "Counts the total number of store uops retired.", "SampleAfterValue": "200003", "UMask": "0x82" }, { "BriefDescription": "Counts the number of tagged loads with an instruction latency that exceeds or equals the threshold of 128 cycles as defined in MEC_CR_PEBS_LD_LAT_THRESHOLD (3F6H). Only counts with PEBS enabled.", - "Counter": "0,1", + "Counter": "0,1,2,3,4,5", "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_UOPS_RETIRED.LOAD_LATENCY_GT_128", "MSRIndex": "0x3F6", "MSRValue": "0x80", - "PublicDescription": "Counts the number of tagged loads with an instruction latency that exceeds or equals the threshold of 128 cycles as defined in MEC_CR_PEBS_LD_LAT_THRESHOLD (3F6H). Only counts with PEBS enabled. If a PEBS record is generated, will populate the PEBS Latency and PEBS Data Source fields accordingly. Available PDIST counters: 0", + "PublicDescription": "Counts the number of tagged loads with an instruction latency that exceeds or equals the threshold of 128 cycles as defined in MEC_CR_PEBS_LD_LAT_THRESHOLD (3F6H). Only counts with PEBS enabled. If a PEBS record is generated, will populate the PEBS Latency and PEBS Data Source fields accordingly.", "SampleAfterValue": "1000003", "UMask": "0x5" }, { "BriefDescription": "Counts the number of tagged loads with an instruction latency that exceeds or equals the threshold of 16 cycles as defined in MEC_CR_PEBS_LD_LAT_THRESHOLD (3F6H). Only counts with PEBS enabled.", - "Counter": "0,1", + "Counter": "0,1,2,3,4,5", "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_UOPS_RETIRED.LOAD_LATENCY_GT_16", "MSRIndex": "0x3F6", "MSRValue": "0x10", - "PublicDescription": "Counts the number of tagged loads with an instruction latency that exceeds or equals the threshold of 16 cycles as defined in MEC_CR_PEBS_LD_LAT_THRESHOLD (3F6H). Only counts with PEBS enabled. If a PEBS record is generated, will populate the PEBS Latency and PEBS Data Source fields accordingly. Available PDIST counters: 0", + "PublicDescription": "Counts the number of tagged loads with an instruction latency that exceeds or equals the threshold of 16 cycles as defined in MEC_CR_PEBS_LD_LAT_THRESHOLD (3F6H). Only counts with PEBS enabled. If a PEBS record is generated, will populate the PEBS Latency and PEBS Data Source fields accordingly.", "SampleAfterValue": "1000003", "UMask": "0x5" }, { "BriefDescription": "Counts the number of tagged loads with an instruction latency that exceeds or equals the threshold of 256 cycles as defined in MEC_CR_PEBS_LD_LAT_THRESHOLD (3F6H). Only counts with PEBS enabled.", - "Counter": "0,1", + "Counter": "0,1,2,3,4,5", "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_UOPS_RETIRED.LOAD_LATENCY_GT_256", "MSRIndex": "0x3F6", "MSRValue": "0x100", - "PublicDescription": "Counts the number of tagged loads with an instruction latency that exceeds or equals the threshold of 256 cycles as defined in MEC_CR_PEBS_LD_LAT_THRESHOLD (3F6H). Only counts with PEBS enabled. If a PEBS record is generated, will populate the PEBS Latency and PEBS Data Source fields accordingly. Available PDIST counters: 0", + "PublicDescription": "Counts the number of tagged loads with an instruction latency that exceeds or equals the threshold of 256 cycles as defined in MEC_CR_PEBS_LD_LAT_THRESHOLD (3F6H). Only counts with PEBS enabled. If a PEBS record is generated, will populate the PEBS Latency and PEBS Data Source fields accordingly.", "SampleAfterValue": "1000003", "UMask": "0x5" }, { "BriefDescription": "Counts the number of tagged loads with an instruction latency that exceeds or equals the threshold of 32 cycles as defined in MEC_CR_PEBS_LD_LAT_THRESHOLD (3F6H). Only counts with PEBS enabled.", - "Counter": "0,1", + "Counter": "0,1,2,3,4,5", "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_UOPS_RETIRED.LOAD_LATENCY_GT_32", "MSRIndex": "0x3F6", "MSRValue": "0x20", - "PublicDescription": "Counts the number of tagged loads with an instruction latency that exceeds or equals the threshold of 32 cycles as defined in MEC_CR_PEBS_LD_LAT_THRESHOLD (3F6H). Only counts with PEBS enabled. If a PEBS record is generated, will populate the PEBS Latency and PEBS Data Source fields accordingly. Available PDIST counters: 0", + "PublicDescription": "Counts the number of tagged loads with an instruction latency that exceeds or equals the threshold of 32 cycles as defined in MEC_CR_PEBS_LD_LAT_THRESHOLD (3F6H). Only counts with PEBS enabled. If a PEBS record is generated, will populate the PEBS Latency and PEBS Data Source fields accordingly.", "SampleAfterValue": "1000003", "UMask": "0x5" }, { "BriefDescription": "Counts the number of tagged loads with an instruction latency that exceeds or equals the threshold of 4 cycles as defined in MEC_CR_PEBS_LD_LAT_THRESHOLD (3F6H). Only counts with PEBS enabled.", - "Counter": "0,1", + "Counter": "0,1,2,3,4,5", "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_UOPS_RETIRED.LOAD_LATENCY_GT_4", "MSRIndex": "0x3F6", "MSRValue": "0x4", - "PublicDescription": "Counts the number of tagged loads with an instruction latency that exceeds or equals the threshold of 4 cycles as defined in MEC_CR_PEBS_LD_LAT_THRESHOLD (3F6H). Only counts with PEBS enabled. If a PEBS record is generated, will populate the PEBS Latency and PEBS Data Source fields accordingly. Available PDIST counters: 0", + "PublicDescription": "Counts the number of tagged loads with an instruction latency that exceeds or equals the threshold of 4 cycles as defined in MEC_CR_PEBS_LD_LAT_THRESHOLD (3F6H). Only counts with PEBS enabled. If a PEBS record is generated, will populate the PEBS Latency and PEBS Data Source fields accordingly.", "SampleAfterValue": "1000003", "UMask": "0x5" }, { "BriefDescription": "Counts the number of tagged loads with an instruction latency that exceeds or equals the threshold of 512 cycles as defined in MEC_CR_PEBS_LD_LAT_THRESHOLD (3F6H). Only counts with PEBS enabled.", - "Counter": "0,1", + "Counter": "0,1,2,3,4,5", "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_UOPS_RETIRED.LOAD_LATENCY_GT_512", "MSRIndex": "0x3F6", "MSRValue": "0x200", - "PublicDescription": "Counts the number of tagged loads with an instruction latency that exceeds or equals the threshold of 512 cycles as defined in MEC_CR_PEBS_LD_LAT_THRESHOLD (3F6H). Only counts with PEBS enabled. If a PEBS record is generated, will populate the PEBS Latency and PEBS Data Source fields accordingly. Available PDIST counters: 0", + "PublicDescription": "Counts the number of tagged loads with an instruction latency that exceeds or equals the threshold of 512 cycles as defined in MEC_CR_PEBS_LD_LAT_THRESHOLD (3F6H). Only counts with PEBS enabled. If a PEBS record is generated, will populate the PEBS Latency and PEBS Data Source fields accordingly.", "SampleAfterValue": "1000003", "UMask": "0x5" }, { "BriefDescription": "Counts the number of tagged loads with an instruction latency that exceeds or equals the threshold of 64 cycles as defined in MEC_CR_PEBS_LD_LAT_THRESHOLD (3F6H). Only counts with PEBS enabled.", - "Counter": "0,1", + "Counter": "0,1,2,3,4,5", "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_UOPS_RETIRED.LOAD_LATENCY_GT_64", "MSRIndex": "0x3F6", "MSRValue": "0x40", - "PublicDescription": "Counts the number of tagged loads with an instruction latency that exceeds or equals the threshold of 64 cycles as defined in MEC_CR_PEBS_LD_LAT_THRESHOLD (3F6H). Only counts with PEBS enabled. If a PEBS record is generated, will populate the PEBS Latency and PEBS Data Source fields accordingly. Available PDIST counters: 0", + "PublicDescription": "Counts the number of tagged loads with an instruction latency that exceeds or equals the threshold of 64 cycles as defined in MEC_CR_PEBS_LD_LAT_THRESHOLD (3F6H). Only counts with PEBS enabled. If a PEBS record is generated, will populate the PEBS Latency and PEBS Data Source fields accordingly.", "SampleAfterValue": "1000003", "UMask": "0x5" }, { "BriefDescription": "Counts the number of tagged loads with an instruction latency that exceeds or equals the threshold of 8 cycles as defined in MEC_CR_PEBS_LD_LAT_THRESHOLD (3F6H). Only counts with PEBS enabled.", - "Counter": "0,1", + "Counter": "0,1,2,3,4,5", "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_UOPS_RETIRED.LOAD_LATENCY_GT_8", "MSRIndex": "0x3F6", "MSRValue": "0x8", - "PublicDescription": "Counts the number of tagged loads with an instruction latency that exceeds or equals the threshold of 8 cycles as defined in MEC_CR_PEBS_LD_LAT_THRESHOLD (3F6H). Only counts with PEBS enabled. If a PEBS record is generated, will populate the PEBS Latency and PEBS Data Source fields accordingly. Available PDIST counters: 0", + "PublicDescription": "Counts the number of tagged loads with an instruction latency that exceeds or equals the threshold of 8 cycles as defined in MEC_CR_PEBS_LD_LAT_THRESHOLD (3F6H). Only counts with PEBS enabled. If a PEBS record is generated, will populate the PEBS Latency and PEBS Data Source fields accordingly.", "SampleAfterValue": "1000003", "UMask": "0x5" },
@@ -356,7 +347,6 @@ "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_UOPS_RETIRED.LOCK_LOADS", - "PublicDescription": "Counts the number of load uops retired that performed one or more locks. Available PDIST counters: 0", "SampleAfterValue": "200003", "UMask": "0x21" },
@@ -366,7 +356,6 @@ "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_UOPS_RETIRED.SPLIT_LOADS", - "PublicDescription": "Counts the number of retired split load uops. Available PDIST counters: 0", "SampleAfterValue": "200003", "UMask": "0x41" },
@@ -376,7 +365,6 @@ "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_UOPS_RETIRED.STLB_MISS", - "PublicDescription": "Counts the total number of load and store uops retired that missed in the second level TLB. Available PDIST counters: 0", "SampleAfterValue": "200003", "UMask": "0x13" },
@@ -386,7 +374,6 @@ "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_UOPS_RETIRED.STLB_MISS_LOADS", - "PublicDescription": "Counts the number of load ops retired that miss in the second Level TLB. Available PDIST counters: 0", "SampleAfterValue": "200003", "UMask": "0x11" },
@@ -396,7 +383,6 @@ "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_UOPS_RETIRED.STLB_MISS_STORES", - "PublicDescription": "Counts the number of store ops retired that miss in the second level TLB. Available PDIST counters: 0", "SampleAfterValue": "200003", "UMask": "0x12" },
@@ -406,7 +392,7 @@ "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_UOPS_RETIRED.STORE_LATENCY", - "PublicDescription": "Counts the number of stores uops retired. Counts with or without PEBS enabled. If PEBS is enabled and a PEBS record is generated, will populate PEBS Latency and PEBS Data Source fields accordingly. Available PDIST counters: 0", + "PublicDescription": "Counts the number of stores uops retired. Counts with or without PEBS enabled. If PEBS is enabled and a PEBS record is generated, will populate PEBS Latency and PEBS Data Source fields accordingly.", "SampleAfterValue": "1000003", "UMask": "0x6" },
tools/perf/pmu-events/arch/x86/alderlaken/floating-point.json +0 -1
@@ -29,7 +29,6 @@ "Counter": "0,1,2,3,4,5", "EventCode": "0xc2", "EventName": "UOPS_RETIRED.FPDIV", - "PublicDescription": "Counts the number of floating point divide uops retired (x87 and SSE, including x87 sqrt). Available PDIST counters: 0", "SampleAfterValue": "2000003", "UMask": "0x8" }
tools/perf/pmu-events/arch/x86/alderlaken/other.json +0 -1
@@ -5,7 +5,6 @@ "Deprecated": "1", "EventCode": "0xe4", "EventName": "LBR_INSERTS.ANY", - "PublicDescription": "This event is deprecated. [This event is alias to MISC_RETIRED.LBR_INSERTS] Available PDIST counters: 0", "SampleAfterValue": "1000003", "UMask": "0x1" },
tools/perf/pmu-events/arch/x86/alderlaken/pipeline.json +5 -37
264 lines not on this page · raw message
tools/perf/pmu-events/arch/x86/alderlaken/virtual-memory.json +0 -3
@@ -57,7 +57,6 @@ "Deprecated": "1", "EventCode": "0xd0", "EventName": "MEM_UOPS_RETIRED.DTLB_MISS", - "PublicDescription": "This event is deprecated. Refer to new event MEM_UOPS_RETIRED.STLB_MISS Available PDIST counters: 0", "SampleAfterValue": "200003", "UMask": "0x13" },
@@ -68,7 +67,6 @@ "Deprecated": "1", "EventCode": "0xd0", "EventName": "MEM_UOPS_RETIRED.DTLB_MISS_LOADS", - "PublicDescription": "This event is deprecated. Refer to new event MEM_UOPS_RETIRED.STLB_MISS_LOADS Available PDIST counters: 0", "SampleAfterValue": "200003", "UMask": "0x11" },
@@ -79,7 +77,6 @@ "Deprecated": "1", "EventCode": "0xd0", "EventName": "MEM_UOPS_RETIRED.DTLB_MISS_STORES", - "PublicDescription": "This event is deprecated. Refer to new event MEM_UOPS_RETIRED.STLB_MISS_STORES Available PDIST counters: 0", "SampleAfterValue": "200003", "UMask": "0x12" }
tools/perf/pmu-events/arch/x86/mapfile.csv +1 -1
@@ -1,6 +1,6 @@ Family-model,Version,Filename,EventType GenuineIntel-6-(97|9A|B7|BA|BF),v1.31,alderlake,core -GenuineIntel-6-BE,v1.29,alderlaken,core +GenuineIntel-6-BE,v1.31,alderlaken,core GenuineIntel-6-C[56],v1.08,arrowlake,core GenuineIntel-6-(1C|26|27|35|36),v5,bonnell,core GenuineIntel-6-(3D|47),v30,broadwell,core
Signature · 1 line
2.50.0.727.gbf7dc18ff4-goog
Ian Rogers
reviewerPATCH v2 3/15 perf vendor events: Update Arrowlake events
In reply to Ian RogersUpdate events from v1.08 to v1.09.
Bring in the event updates v1.09:
https://github.com/intel/perfmon/commit/cf3be6daf0a751ad270b67890dfdb2261dfc75da
--- .../pmu-events/arch/x86/arrowlake/cache.json | 13 +- .../arch/x86/arrowlake/frontend.json | 135 ++++++++++++++++++ tools/perf/pmu-events/arch/x86/mapfile.csv | 2 +- 3 files changed, 148 insertions(+), 2 deletions(-)
tools/perf/pmu-events/arch/x86/arrowlake/cache.json +12 -1
@@ -237,7 +237,7 @@ "Unit": "cpu_lowpower" }, { - "BriefDescription": "Counts the number of L2 Cache Accesses that miss the L2 and get BBL reject short and long rejects (includes those counted in L2_reject_XQ.any), per core event", + "BriefDescription": "Counts the number of L2 Cache Accesses that miss the L2 and get BBL reject short and long rejects, per core event", "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0x24", "EventName": "L2_REQUEST.REJECTS",
@@ -728,6 +728,17 @@ "EventName": "MEM_LOAD_RETIRED.L1_HIT", "PublicDescription": "Counts retired load instructions with at least one uop that hit in the L1 data cache. This event includes all SW prefetches and lock instructions regardless of the data source. Available PDIST counters: 0", "SampleAfterValue": "1000003", + "UMask": "0x101", + "Unit": "cpu_core" + }, + { + "BriefDescription": "Counts retired load instructions with at least one uop that hit in the Level 0 of the L1 data cache. This event includes all SW prefetches and lock instructions regardless of the data source.", + "Counter": "0,1,2,3", + "Data_LA": "1", + "EventCode": "0xd1", + "EventName": "MEM_LOAD_RETIRED.L1_HIT_L0", + "PublicDescription": "Counts retired load instructions with at least one uop that hit in the Level 0 of the L1 data cache. This event includes all SW prefetches and lock instructions regardless of the data source. Available PDIST counters: 0", + "SampleAfterValue": "1000003", "UMask": "0x1", "Unit": "cpu_core" },
tools/perf/pmu-events/arch/x86/arrowlake/frontend.json +135 -0
159 lines not on this page · raw message
tools/perf/pmu-events/arch/x86/mapfile.csv +1 -1
@@ -1,7 +1,7 @@ Family-model,Version,Filename,EventType GenuineIntel-6-(97|9A|B7|BA|BF),v1.31,alderlake,core GenuineIntel-6-BE,v1.31,alderlaken,core -GenuineIntel-6-C[56],v1.08,arrowlake,core +GenuineIntel-6-C[56],v1.09,arrowlake,core GenuineIntel-6-(1C|26|27|35|36),v5,bonnell,core GenuineIntel-6-(3D|47),v30,broadwell,core GenuineIntel-6-56,v12,broadwellde,core
Signature · 1 line
2.50.0.727.gbf7dc18ff4-goog
Ian Rogers
reviewerPATCH v2 4/15 perf vendor events: Update CascadelakeX events
In reply to Ian RogersUpdate events from v1.23 to v1.25.
Bring in the event updates v1.25:
https://github.com/intel/perfmon/commit/86f146e15626b0fd3b032cab4538cafaaf2d0635
https://github.com/intel/perfmon/commit/fef03ffc333ae44d1e9d695b4e67e5bbb4429729
--- .../pmu-events/arch/x86/cascadelakex/floating-point.json | 6 +++--- tools/perf/pmu-events/arch/x86/cascadelakex/pipeline.json | 2 +- tools/perf/pmu-events/arch/x86/mapfile.csv | 2 +- 3 files changed, 5 insertions(+), 5 deletions(-)
tools/perf/pmu-events/arch/x86/cascadelakex/floating-point.json +3 -3
@@ -111,7 +111,7 @@ "Counter": "0,1,2,3", "EventCode": "0xCF", "EventName": "FP_ARITH_INST_RETIRED2.128BIT_PACKED_BF16", - "PublicDescription": "Counts once for each Intel AVX-512 computational 512-bit packed BFloat16 floating-point instruction retired. Applies to the ZMM based VDPBF16PS instruction. Each count represents 64 computation operations. This event is only supported on products formerly named Cooper Lake and is not supported on products formerly named Cascade Lake.", + "PublicDescription": "Counts once for each Intel AVX-512 computational 128-bit packed BFloat16 floating-point instruction retired. Applies to the XMM based VDPBF16PS instruction. Each count represents 16 computation operations. This event is only supported on products formerly named Cooper Lake and is not supported on products formerly named Cascade Lake.", "SampleAfterValue": "2000003", "UMask": "0x20" },
@@ -120,7 +120,7 @@ "Counter": "0,1,2,3", "EventCode": "0xCF", "EventName": "FP_ARITH_INST_RETIRED2.256BIT_PACKED_BF16", - "PublicDescription": "Counts once for each Intel AVX-512 computational 128-bit packed BFloat16 floating-point instruction retired. Applies to the XMM based VDPBF16PS instruction. Each count represents 16 computation operations. This event is only supported on products formerly named Cooper Lake and is not supported on products formerly named Cascade Lake.", + "PublicDescription": "Counts once for each Intel AVX-512 computational 256-bit packed BFloat16 floating-point instruction retired. Applies to the YMM based VDPBF16PS instruction. Each count represents 32 computation operations. This event is only supported on products formerly named Cooper Lake and is not supported on products formerly named Cascade Lake.", "SampleAfterValue": "2000003", "UMask": "0x40" },
@@ -129,7 +129,7 @@ "Counter": "0,1,2,3", "EventCode": "0xCF", "EventName": "FP_ARITH_INST_RETIRED2.512BIT_PACKED_BF16", - "PublicDescription": "Counts once for each Intel AVX-512 computational 256-bit packed BFloat16 floating-point instruction retired. Applies to the YMM based VDPBF16PS instruction. Each count represents 32 computation operations. This event is only supported on products formerly named Cooper Lake and is not supported on products formerly named Cascade Lake.", + "PublicDescription": "Counts once for each Intel AVX-512 computational 512-bit packed BFloat16 floating-point instruction retired. Applies to the ZMM based VDPBF16PS instruction. Each count represents 64 computation operations. This event is only supported on products formerly named Cooper Lake and is not supported on products formerly named Cascade Lake.", "SampleAfterValue": "2000003", "UMask": "0x80" },
tools/perf/pmu-events/arch/x86/cascadelakex/pipeline.json +1 -1
@@ -542,7 +542,7 @@ "Counter": "0,1,2,3", "EventCode": "0x4C", "EventName": "LOAD_HIT_PRE.SW_PF", - "PublicDescription": "Counts all not software-prefetch load dispatches that hit the fill buffer (FB) allocated for the software prefetch. It can also be incremented by some lock instructions. So it should only be used with profiling so that the locks can be excluded by ASM (Assembly File) inspection of the nearby instructions.", + "PublicDescription": "Counts all software-prefetch load dispatches that hit the fill buffer (FB) allocated for the software prefetch. It can also be incremented by some lock instructions. So it should only be used with profiling so that the locks can be excluded by ASM (Assembly File) inspection of the nearby instructions.", "SampleAfterValue": "100003", "UMask": "0x1" },
tools/perf/pmu-events/arch/x86/mapfile.csv +1 -1
@@ -6,7 +6,7 @@ GenuineIntel-6-(1C|26|27|35|36),v5,bonnell,core GenuineIntel-6-(3D|47),v30,broadwell,core GenuineIntel-6-56,v12,broadwellde,core GenuineIntel-6-4F,v23,broadwellx,core -GenuineIntel-6-55-[56789ABCDEF],v1.23,cascadelakex,core +GenuineIntel-6-55-[56789ABCDEF],v1.25,cascadelakex,core GenuineIntel-6-DD,v1.00,clearwaterforest,core GenuineIntel-6-9[6C],v1.05,elkhartlake,core GenuineIntel-6-CF,v1.11,emeraldrapids,core
Signature · 1 line
2.50.0.727.gbf7dc18ff4-goog
Ian Rogers
reviewerPATCH v2 5/15 perf vendor events: Update EmeraldRapids events
In reply to Ian RogersUpdate events from v1.11 to v1.14.
Bring in the event updates v1.14:
https://github.com/intel/perfmon/commit/6f6e4c8c906992b450cb2014d0501a9ec1cda0d0
https://github.com/intel/perfmon/commit/e363f82276c129aec60402a1d64efbbd41af844e
--- .../arch/x86/emeraldrapids/pipeline.json | 2 +- .../arch/x86/emeraldrapids/uncore-io.json | 12 +++++++++++ .../arch/x86/emeraldrapids/uncore-memory.json | 20 +++++++++++++++++++ tools/perf/pmu-events/arch/x86/mapfile.csv | 2 +- 4 files changed, 34 insertions(+), 2 deletions(-)
tools/perf/pmu-events/arch/x86/emeraldrapids/pipeline.json +1 -1
@@ -684,7 +684,7 @@ "Counter": "0,1,2,3", "EventCode": "0x4c", "EventName": "LOAD_HIT_PREFETCH.SWPF", - "PublicDescription": "Counts all not software-prefetch load dispatches that hit the fill buffer (FB) allocated for the software prefetch. It can also be incremented by some lock instructions. So it should only be used with profiling so that the locks can be excluded by ASM (Assembly File) inspection of the nearby instructions. Available PDIST counters: 0", + "PublicDescription": "Counts all software-prefetch load dispatches that hit the fill buffer (FB) allocated for the software prefetch. It can also be incremented by some lock instructions. So it should only be used with profiling so that the locks can be excluded by ASM (Assembly File) inspection of the nearby instructions. Available PDIST counters: 0", "SampleAfterValue": "100003", "UMask": "0x1" },
tools/perf/pmu-events/arch/x86/emeraldrapids/uncore-io.json +12 -0
@@ -1821,6 +1821,18 @@ "UMask": "0x4", "Unit": "IIO" }, + { + "BriefDescription": "Posted requests sent by the integrated IO (IIO) controller to the Ubox, useful for counting message signaled interrupts (MSI).", + "Counter": "0,1,2,3", + "EventCode": "0x8e", + "EventName": "UNC_IIO_NUM_REQ_OF_CPU_BY_TGT.UBOX_POSTED", + "Experimental": "1", + "FCMask": "0x01", + "PerPkg": "1", + "PortMask": "0x00FF", + "UMask": "0x4", + "Unit": "IIO" + }, { "BriefDescription": "ITC address map 1", "Counter": "0,1,2,3",
tools/perf/pmu-events/arch/x86/emeraldrapids/uncore-memory.json +20 -0
@@ -2145,6 +2145,16 @@ "UMask": "0x1", "Unit": "MCHBM" }, + { + "BriefDescription": "ECC Correctable Errors", + "Counter": "0,1,2,3", + "EventCode": "0x09", + "EventName": "UNC_MCHBM_ECC_CORRECTABLE_ERRORS", + "Experimental": "1", + "PerPkg": "1", + "PublicDescription": "ECC Correctable Errors. Counts the number of ECC errors detected and corrected by the iMC on this channel. This counter is only useful with ECC devices. This count will increment one time for each correction regardless of the number of bits corrected. The iMC can correct up to 4 bit errors in independent channel mode and 8 bit errors in lockstep mode.", + "Unit": "MCHBM" + }, { "BriefDescription": "HBM Precharge All Commands", "Counter": "0,1,2,3",
@@ -2759,6 +2769,16 @@ "UMask": "0x3", "Unit": "iMC" }, + { + "BriefDescription": "ECC Correctable Errors", + "Counter": "0,1,2,3", + "EventCode": "0x09", + "EventName": "UNC_M_ECC_CORRECTABLE_ERRORS", + "Experimental": "1", + "PerPkg": "1", + "PublicDescription": "ECC Correctable Errors : Counts the number of ECC errors detected and corrected by the iMC on this channel. This counter is only useful with ECC DRAM devices. This count will increment one time for each correction regardless of the number of bits corrected. The iMC can correct up to 4 bit errors in independent channel mode and 8 bit errors in lockstep mode.", + "Unit": "iMC" + }, { "BriefDescription": "IMC Clockticks at HCLK frequency", "Counter": "0,1,2,3",
tools/perf/pmu-events/arch/x86/mapfile.csv +1 -1
@@ -9,7 +9,7 @@ GenuineIntel-6-4F,v23,broadwellx,core GenuineIntel-6-55-[56789ABCDEF],v1.25,cascadelakex,core GenuineIntel-6-DD,v1.00,clearwaterforest,core GenuineIntel-6-9[6C],v1.05,elkhartlake,core -GenuineIntel-6-CF,v1.11,emeraldrapids,core +GenuineIntel-6-CF,v1.14,emeraldrapids,core GenuineIntel-6-5[CF],v13,goldmont,core GenuineIntel-6-7A,v1.01,goldmontplus,core GenuineIntel-6-B6,v1.07,grandridge,core
Signature · 1 line
2.50.0.727.gbf7dc18ff4-goog
Ian Rogers
reviewerPATCH v2 6/15 perf vendor events: Update GrandRidge events
In reply to Ian RogersUpdate events from v1.07 to v1.09.
Bring in the event updates v1.09:
https://github.com/intel/perfmon/commit/8c74d09c8544421256a79f4f21e548ad756f5b7f
https://github.com/intel/perfmon/commit/18c7d2a75e45eacf5553f900ae2097a1290f5bed
--- .../arch/x86/grandridge/grr-metrics.json | 30 +++++++++++++++---- .../x86/grandridge/uncore-interconnect.json | 10 +++++++ .../arch/x86/grandridge/uncore-io.json | 12 ++++++++ tools/perf/pmu-events/arch/x86/mapfile.csv | 2 +- 4 files changed, 47 insertions(+), 7 deletions(-)
tools/perf/pmu-events/arch/x86/grandridge/grr-metrics.json +24 -6
@@ -106,6 +106,30 @@ "MetricName": "io_bandwidth_write", "ScaleUnit": "1MB/s" }, + { + "BriefDescription": "The percent of inbound full cache line writes initiated by IO that miss the L3 cache", + "MetricExpr": "UNC_CHA_TOR_INSERTS.IO_MISS_ITOM / UNC_CHA_TOR_INSERTS.IO_ITOM", + "MetricName": "io_full_write_l3_miss", + "ScaleUnit": "100%" + }, + { + "BriefDescription": "Message Signaled Interrupts (MSI) per second sent by the integrated I/O traffic controller (IIO) to System Configuration Controller (Ubox)", + "MetricExpr": "UNC_IIO_NUM_REQ_OF_CPU_BY_TGT.UBOX_POSTED / duration_time", + "MetricName": "io_msi", + "ScaleUnit": "1per_sec" + }, + { + "BriefDescription": "The percent of inbound partial writes initiated by IO that miss the L3 cache", + "MetricExpr": "(UNC_CHA_TOR_INSERTS.IO_MISS_ITOMCACHENEAR + UNC_CHA_TOR_INSERTS.IO_MISS_RFO) / (UNC_CHA_TOR_INSERTS.IO_ITOMCACHENEAR + UNC_CHA_TOR_INSERTS.IO_RFO)", + "MetricName": "io_partial_write_l3_miss", + "ScaleUnit": "100%" + }, + { + "BriefDescription": "The percent of inbound reads initiated by IO that miss the L3 cache", + "MetricExpr": "UNC_CHA_TOR_INSERTS.IO_MISS_PCIRDCUR / UNC_CHA_TOR_INSERTS.IO_PCIRDCUR", + "MetricName": "io_read_l3_miss", + "ScaleUnit": "100%" + }, { "BriefDescription": "Ratio of number of completed page walks (for 2 megabyte and 4 megabyte page sizes) caused by a code fetch to the total number of completed instructions", "MetricExpr": "ITLB_MISSES.WALK_COMPLETED_2M_4M / INST_RETIRED.ANY",
@@ -162,12 +186,6 @@ "MetricName": "llc_data_read_mpi_demand_plus_prefetch", "ScaleUnit": "1per_instr" }, - { - "BriefDescription": "Average latency of a last level cache (LLC) demand data read miss (read memory access) in nano seconds", - "MetricExpr": "1e9 * (UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_OPT / UNC_CHA_TOR_INSERTS.IA_MISS_DRD_OPT) / (UNC_CHA_CLOCKTICKS / (source_count(UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_OPT) * #num_packages)) * duration_time", - "MetricName": "llc_demand_data_read_miss_latency", - "ScaleUnit": "1ns" - }, { "BriefDescription": "Load operations retired per instruction", "MetricExpr": "MEM_UOPS_RETIRED.ALL_LOADS / INST_RETIRED.ANY",
tools/perf/pmu-events/arch/x86/grandridge/uncore-interconnect.json +10 -0
@@ -261,5 +261,15 @@ "PerPkg": "1", "UMask": "0x8", "Unit": "IRP" + }, + { + "BriefDescription": "Message Received : MSI", + "Counter": "0,1", + "EventCode": "0x42", + "EventName": "UNC_U_EVENT_MSG.MSI_RCVD", + "PerPkg": "1", + "PublicDescription": "Message Received : MSI : Message Signaled Interrupts - interrupts sent by devices (including PCIe via IOxAPIC) (Socket Mode only)", + "UMask": "0x2", + "Unit": "UBOX" } ]
tools/perf/pmu-events/arch/x86/grandridge/uncore-io.json +12 -0
@@ -907,6 +907,18 @@ "UMask": "0x4", "Unit": "IIO" }, + { + "BriefDescription": "Posted requests sent by the integrated IO (IIO) controller to the Ubox, useful for counting message signaled interrupts (MSI).", + "Counter": "0,1,2,3", + "EventCode": "0x8e", + "EventName": "UNC_IIO_NUM_REQ_OF_CPU_BY_TGT.UBOX_POSTED", + "FCMask": "0x01", + "PerPkg": "1", + "PortMask": "0x0FF", + "PublicDescription": "-", + "UMask": "0x4", + "Unit": "IIO" + }, { "BriefDescription": "All 9 bits of Page Walk Tracker Occupancy", "Counter": "0,1,2,3",
tools/perf/pmu-events/arch/x86/mapfile.csv +1 -1
@@ -12,7 +12,7 @@ GenuineIntel-6-9[6C],v1.05,elkhartlake,core GenuineIntel-6-CF,v1.14,emeraldrapids,core GenuineIntel-6-5[CF],v13,goldmont,core GenuineIntel-6-7A,v1.01,goldmontplus,core -GenuineIntel-6-B6,v1.07,grandridge,core +GenuineIntel-6-B6,v1.09,grandridge,core GenuineIntel-6-A[DE],v1.08,graniterapids,core GenuineIntel-6-(3C|45|46),v36,haswell,core GenuineIntel-6-3F,v29,haswellx,core
Signature · 1 line
2.50.0.727.gbf7dc18ff4-goog
Ian Rogers
reviewerPATCH v2 7/15 perf vendor events: Update GraniteRapids events
In reply to Ian RogersUpdate events from v1.08 to v1.10.
Bring in the event updates v1.10
https://github.com/intel/perfmon/commit/96259a932e2ce5f70ed7d347ca92fdeb78f83aa5
https://github.com/intel/perfmon/commit/19e315c8d2e0b44e170a6e60de44c9359062a6aa
--- .../arch/x86/graniterapids/cache.json | 9 +++++ .../arch/x86/graniterapids/counter.json | 10 +++--- .../arch/x86/graniterapids/gnr-metrics.json | 36 +++++++++++++++++++ .../arch/x86/graniterapids/pipeline.json | 2 +- .../graniterapids/uncore-interconnect.json | 19 ---------- .../arch/x86/graniterapids/uncore-io.json | 27 +++++++++++++- tools/perf/pmu-events/arch/x86/mapfile.csv | 2 +- 7 files changed, 78 insertions(+), 27 deletions(-)
tools/perf/pmu-events/arch/x86/graniterapids/cache.json +9 -0
@@ -977,6 +977,15 @@ "SampleAfterValue": "100003", "UMask": "0x4" }, + { + "BriefDescription": "Offcore Uncacheable memory data read transactions.", + "Counter": "0,1,2,3", + "EventCode": "0x21", + "EventName": "OFFCORE_REQUESTS.MEM_UC", + "PublicDescription": "This event counts noncacheable memory data read transactions. Available PDIST counters: 0", + "SampleAfterValue": "100003", + "UMask": "0x20" + }, { "BriefDescription": "Cycles when offcore outstanding cacheable Core Data Read transactions are present in SuperQueue (SQ), queue to uncore.", "Counter": "0,1,2,3",
tools/perf/pmu-events/arch/x86/graniterapids/counter.json +5 -5
@@ -59,6 +59,11 @@ "CountersNumFixed": "0", "CountersNumGeneric": 4 }, + { + "Unit": "UBOX", + "CountersNumFixed": "0", + "CountersNumGeneric": "2" + }, { "Unit": "PCU", "CountersNumFixed": "0",
@@ -73,10 +78,5 @@ "Unit": "MDF", "CountersNumFixed": "0", "CountersNumGeneric": "4" - }, - { - "Unit": "UBOX", - "CountersNumFixed": "0", - "CountersNumGeneric": "2" } ]
tools/perf/pmu-events/arch/x86/graniterapids/gnr-metrics.json +36 -0
@@ -95,6 +95,12 @@ "MetricName": "io_bandwidth_read", "ScaleUnit": "1MB/s" }, + { + "BriefDescription": "Bandwidth of inbound IO reads that are initiated by end device controllers that are requesting memory from the CPU and miss the L3 cache", + "MetricExpr": "UNC_CHA_TOR_INSERTS.IO_MISS_PCIRDCUR * 64 / 1e6 / duration_time", + "MetricName": "io_bandwidth_read_l3_miss", + "ScaleUnit": "1MB/s" + }, { "BriefDescription": "Bandwidth of IO reads that are initiated by end device controllers that are requesting memory from the local CPU socket", "MetricExpr": "UNC_CHA_TOR_INSERTS.IO_PCIRDCUR_LOCAL * 64 / 1e6 / duration_time",
@@ -113,6 +119,12 @@ "MetricName": "io_bandwidth_write", "ScaleUnit": "1MB/s" }, + { + "BriefDescription": "Bandwidth of inbound IO writes that are initiated by end device controllers that are writing memory to the CPU", + "MetricExpr": "(UNC_CHA_TOR_INSERTS.IO_MISS_ITOM + UNC_CHA_TOR_INSERTS.IO_MISS_ITOMCACHENEAR) * 64 / 1e6 / duration_time", + "MetricName": "io_bandwidth_write_l3_miss", + "ScaleUnit": "1MB/s" + }, { "BriefDescription": "Bandwidth of IO writes that are initiated by end device controllers that are writing memory to the local CPU socket", "MetricExpr": "(UNC_CHA_TOR_INSERTS.IO_ITOM_LOCAL + UNC_CHA_TOR_INSERTS.IO_ITOMCACHENEAR_LOCAL) * 64 / 1e6 / duration_time",
@@ -125,6 +137,30 @@ "MetricName": "io_bandwidth_write_remote", "ScaleUnit": "1MB/s" }, + { + "BriefDescription": "The percent of inbound full cache line writes initiated by IO that miss the L3 cache", + "MetricExpr": "UNC_CHA_TOR_INSERTS.IO_MISS_ITOM / UNC_CHA_TOR_INSERTS.IO_ITOM", + "MetricName": "io_full_write_l3_miss", + "ScaleUnit": "100%" + }, + { + "BriefDescription": "Message Signaled Interrupts (MSI) per second sent by the integrated I/O traffic controller (IIO) to System Configuration Controller (Ubox)", + "MetricExpr": "UNC_IIO_NUM_REQ_OF_CPU_BY_TGT.UBOX_POSTED / duration_time", + "MetricName": "io_msi", + "ScaleUnit": "1per_sec" + }, + { + "BriefDescription": "The percent of inbound partial writes initiated by IO that miss the L3 cache", + "MetricExpr": "(UNC_CHA_TOR_INSERTS.IO_MISS_ITOMCACHENEAR + UNC_CHA_TOR_INSERTS.IO_MISS_RFO) / (UNC_CHA_TOR_INSERTS.IO_ITOMCACHENEAR + UNC_CHA_TOR_INSERTS.IO_RFO)", + "MetricName": "io_partial_write_l3_miss", + "ScaleUnit": "100%" + }, + { + "BriefDescription": "The percent of inbound reads initiated by IO that miss the L3 cache", + "MetricExpr": "UNC_CHA_TOR_INSERTS.IO_MISS_PCIRDCUR / UNC_CHA_TOR_INSERTS.IO_PCIRDCUR", + "MetricName": "io_read_l3_miss", + "ScaleUnit": "100%" + }, { "BriefDescription": "Ratio of number of completed page walks (for all page sizes) caused by a code fetch to the total number of completed instructions", "MetricExpr": "ITLB_MISSES.WALK_COMPLETED / INST_RETIRED.ANY",
tools/perf/pmu-events/arch/x86/graniterapids/pipeline.json +1 -1
@@ -738,7 +738,7 @@ "Counter": "0,1,2,3", "EventCode": "0x4c", "EventName": "LOAD_HIT_PREFETCH.SWPF", - "PublicDescription": "Counts all not software-prefetch load dispatches that hit the fill buffer (FB) allocated for the software prefetch. It can also be incremented by some lock instructions. So it should only be used with profiling so that the locks can be excluded by ASM (Assembly File) inspection of the nearby instructions. Available PDIST counters: 0", + "PublicDescription": "Counts all software-prefetch load dispatches that hit the fill buffer (FB) allocated for the software prefetch. It can also be incremented by some lock instructions. So it should only be used with profiling so that the locks can be excluded by ASM (Assembly File) inspection of the nearby instructions. Available PDIST counters: 0", "SampleAfterValue": "100003", "UMask": "0x1" },
tools/perf/pmu-events/arch/x86/graniterapids/uncore-interconnect.json +0 -19
@@ -1915,24 +1915,6 @@ "UMask": "0x4", "Unit": "UPI" }, - { - "BriefDescription": "Tx Flit Buffer Allocations : Number of allocations into the UPI Tx Flit Buffer. Generally, when data is transmitted across UPI, it will bypass the TxQ and pass directly to the link. However, the TxQ will be used with L0p and when LLR occurs, increasing latency to transfer out to the link. This event can be used in conjunction with the Flit Buffer Occupancy event in order to calculate the average flit buffer lifetime.", - "Counter": "0,1,2,3", - "EventCode": "0x40", - "EventName": "UNC_UPI_TxL_INSERTS", - "Experimental": "1", - "PerPkg": "1", - "Unit": "UPI" - }, - { - "BriefDescription": "Tx Flit Buffer Occupancy : Accumulates the number of flits in the TxQ. Generally, when data is transmitted across UPI, it will bypass the TxQ and pass directly to the link. However, the TxQ will be used with L0p and when LLR occurs, increasing latency to transfer out to the link. This can be used with the cycles not empty event to track average occupancy, or the allocations event to track average lifetime in the TxQ.", - "Counter": "0,1,2,3", - "EventCode": "0x42", - "EventName": "UNC_UPI_TxL_OCCUPANCY", - "Experimental": "1", - "PerPkg": "1", - "Unit": "UPI" - }, { "BriefDescription": "Message Received : Doorbell", "Counter": "0,1",
@@ -1970,7 +1952,6 @@ "Counter": "0,1", "EventCode": "0x42", "EventName": "UNC_U_EVENT_MSG.MSI_RCVD", - "Experimental": "1", "PerPkg": "1", "PublicDescription": "Message Received : MSI : Message Signaled Interrupts - interrupts sent by devices (including PCIe via IOxAPIC) (Socket Mode only)", "UMask": "0x2",
tools/perf/pmu-events/arch/x86/graniterapids/uncore-io.json +26 -1
@@ -1121,8 +1121,9 @@ "Unit": "IIO" }, { - "BriefDescription": "Occupancy of outbound request queue : To device : Counts number of outbound requests/completions IIO is currently processing", + "BriefDescription": "This event is deprecated. [This event is alias to UNC_IIO_NUM_OUTSTANDING_REQ_FROM_CPU.TO_IO]", "Counter": "2,3", + "Deprecated": "1", "EventCode": "0xc5", "EventName": "UNC_IIO_NUM_OUSTANDING_REQ_FROM_CPU.TO_IO", "Experimental": "1",
@@ -1132,6 +1133,18 @@ "UMask": "0x8", "Unit": "IIO" }, + { + "BriefDescription": "Occupancy of outbound request queue : To device : Counts number of outbound requests/completions IIO is currently processing [This event is alias to UNC_IIO_NUM_OUSTANDING_REQ_FROM_CPU.TO_IO]", + "Counter": "2,3", + "EventCode": "0xc5", + "EventName": "UNC_IIO_NUM_OUTSTANDING_REQ_FROM_CPU.TO_IO", + "Experimental": "1", + "FCMask": "0x07", + "PerPkg": "1", + "PortMask": "0x0FF", + "UMask": "0x8", + "Unit": "IIO" + }, { "BriefDescription": "Passing data to be written", "Counter": "0,1,2,3",
@@ -1300,6 +1313,18 @@ "UMask": "0x4", "Unit": "IIO" }, + { + "BriefDescription": "Posted requests sent by the integrated IO (IIO) controller to the Ubox, useful for counting message signaled interrupts (MSI).", + "Counter": "0,1,2,3", + "EventCode": "0x8e", + "EventName": "UNC_IIO_NUM_REQ_OF_CPU_BY_TGT.UBOX_POSTED", + "FCMask": "0x01", + "PerPkg": "1", + "PortMask": "0x0FF", + "PublicDescription": "-", + "UMask": "0x4", + "Unit": "IIO" + }, { "BriefDescription": "All 9 bits of Page Walk Tracker Occupancy", "Counter": "0,1,2,3",
tools/perf/pmu-events/arch/x86/mapfile.csv +1 -1
@@ -13,7 +13,7 @@ GenuineIntel-6-CF,v1.14,emeraldrapids,core GenuineIntel-6-5[CF],v13,goldmont,core GenuineIntel-6-7A,v1.01,goldmontplus,core GenuineIntel-6-B6,v1.09,grandridge,core -GenuineIntel-6-A[DE],v1.08,graniterapids,core +GenuineIntel-6-A[DE],v1.10,graniterapids,core GenuineIntel-6-(3C|45|46),v36,haswell,core GenuineIntel-6-3F,v29,haswellx,core GenuineIntel-6-7[DE],v1.24,icelake,core
Signature · 1 line
2.50.0.727.gbf7dc18ff4-goog
Ian Rogers
reviewerPATCH v2 8/15 perf vendor events: Update IcelakeX events
In reply to Ian RogersUpdate events from v1.27 to v1.28.
Bring in the event updates v1.28:
https://github.com/intel/perfmon/commit/c52728a46cf37ba271c09b1eb7093cfc82dfbf29
--- tools/perf/pmu-events/arch/x86/icelakex/pipeline.json | 2 +- tools/perf/pmu-events/arch/x86/icelakex/uncore-cache.json | 2 -- tools/perf/pmu-events/arch/x86/mapfile.csv | 2 +- 3 files changed, 2 insertions(+), 4 deletions(-)
tools/perf/pmu-events/arch/x86/icelakex/pipeline.json +1 -1
@@ -477,7 +477,7 @@ "Counter": "0,1,2,3", "EventCode": "0x4c", "EventName": "LOAD_HIT_PREFETCH.SWPF", - "PublicDescription": "Counts all not software-prefetch load dispatches that hit the fill buffer (FB) allocated for the software prefetch. It can also be incremented by some lock instructions. So it should only be used with profiling so that the locks can be excluded by ASM (Assembly File) inspection of the nearby instructions.", + "PublicDescription": "Counts all software-prefetch load dispatches that hit the fill buffer (FB) allocated for the software prefetch. It can also be incremented by some lock instructions. So it should only be used with profiling so that the locks can be excluded by ASM (Assembly File) inspection of the nearby instructions.", "SampleAfterValue": "100003", "UMask": "0x1" },
tools/perf/pmu-events/arch/x86/icelakex/uncore-cache.json +0 -2
@@ -8193,7 +8193,6 @@ "Counter": "0,1,2,3", "EventCode": "0x35", "EventName": "UNC_CHA_TOR_INSERTS.IO_MISS_RFO", - "Experimental": "1", "PerPkg": "1", "PublicDescription": "TOR Inserts : RFOs issued by IO Devices that missed the LLC : Counts the number of entries successfully inserted into the TOR that match qualifications specified by the subevent. Does not include addressless requests such as locks and interrupts.", "UMask": "0xc803fe04",
@@ -8234,7 +8233,6 @@ "Counter": "0,1,2,3", "EventCode": "0x35", "EventName": "UNC_CHA_TOR_INSERTS.IO_RFO", - "Experimental": "1", "PerPkg": "1", "PublicDescription": "TOR Inserts : RFOs issued by IO Devices : Counts the number of entries successfully inserted into the TOR that match qualifications specified by the subevent. Does not include addressless requests such as locks and interrupts.", "UMask": "0xc803ff04",
tools/perf/pmu-events/arch/x86/mapfile.csv +1 -1
@@ -17,7 +17,7 @@ GenuineIntel-6-A[DE],v1.10,graniterapids,core GenuineIntel-6-(3C|45|46),v36,haswell,core GenuineIntel-6-3F,v29,haswellx,core GenuineIntel-6-7[DE],v1.24,icelake,core -GenuineIntel-6-6[AC],v1.27,icelakex,core +GenuineIntel-6-6[AC],v1.28,icelakex,core GenuineIntel-6-3A,v24,ivybridge,core GenuineIntel-6-3E,v24,ivytown,core GenuineIntel-6-2D,v24,jaketown,core
Signature · 1 line
2.50.0.727.gbf7dc18ff4-goog
Ian Rogers
reviewerPATCH v2 9/15 perf vendor events: Update LunarLake events
In reply to Ian RogersUpdate events from v1.11 to v1.14.
Bring in the event updates v1.14:
https://github.com/intel/perfmon/commit/95634fec10542c0c466eb2c6d9a81e0c24fb1123
https://github.com/intel/perfmon/commit/84a49938387ac592af0a622273e4e8e4997e987d
--- .../pmu-events/arch/x86/lunarlake/cache.json | 11 +++++++++++ .../arch/x86/lunarlake/pipeline.json | 18 ++++++++++++++---- .../arch/x86/lunarlake/virtual-memory.json | 18 ------------------ tools/perf/pmu-events/arch/x86/mapfile.csv | 2 +- 4 files changed, 26 insertions(+), 23 deletions(-)
tools/perf/pmu-events/arch/x86/lunarlake/cache.json +11 -0
@@ -790,6 +790,17 @@ "EventName": "MEM_LOAD_RETIRED.L1_HIT", "PublicDescription": "Counts retired load instructions with at least one uop that hit in the L1 data cache. This event includes all SW prefetches and lock instructions regardless of the data source. Available PDIST counters: 0", "SampleAfterValue": "1000003", + "UMask": "0x101", + "Unit": "cpu_core" + }, + { + "BriefDescription": "Counts retired load instructions with at least one uop that hit in the Level 0 of the L1 data cache. This event includes all SW prefetches and lock instructions regardless of the data source.", + "Counter": "0,1,2,3", + "Data_LA": "1", + "EventCode": "0xd1", + "EventName": "MEM_LOAD_RETIRED.L1_HIT_L0", + "PublicDescription": "Counts retired load instructions with at least one uop that hit in the Level 0 of the L1 data cache. This event includes all SW prefetches and lock instructions regardless of the data source. Available PDIST counters: 0", + "SampleAfterValue": "1000003", "UMask": "0x1", "Unit": "cpu_core" },
tools/perf/pmu-events/arch/x86/lunarlake/pipeline.json +14 -4
@@ -1247,9 +1247,19 @@ "Unit": "cpu_core" }, { - "BriefDescription": "Counts the number of demand loads that match on a wcb (request buffer) allocated by an L1 hardware prefetch", + "BriefDescription": "Counts the number of demand loads that match on a wcb (request buffer) allocated by an L1 hardware prefetch [This event is alias to LOAD_HIT_PREFETCH.HW_PF]", "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0x4c", + "EventName": "LOAD_HIT_PREFETCH.HWPF", + "SampleAfterValue": "1000003", + "UMask": "0x2", + "Unit": "cpu_atom" + }, + { + "BriefDescription": "This event is deprecated. [This event is alias to LOAD_HIT_PREFETCH.HWPF]", + "Counter": "0,1,2,3,4,5,6,7", + "Deprecated": "1", + "EventCode": "0x4c", "EventName": "LOAD_HIT_PREFETCH.HW_PF", "SampleAfterValue": "1000003", "UMask": "0x2",
@@ -1664,7 +1674,7 @@ }, { "BriefDescription": "Fixed Counter: Counts the number of issue slots not consumed by the backend because allocation is stalled due to a mispredicted jump or a machine clear.", - "Counter": "36", + "Counter": "Fixed counter 4", "EventName": "TOPDOWN_BAD_SPECULATION.ALL", "PublicDescription": "Fixed Counter: Counts the number of issue slots that were not consumed by the backend because allocation is stalled due to a mispredicted jump or a machine clear. Counts all issue slots blocked during this recovery window including relevant microcode flows and while uops are not yet available in the IQ. Also, includes the issue slots that were consumed by the backend but were thrown away because they were younger than the mispredict or machine clear.", "SampleAfterValue": "1000003",
@@ -1797,7 +1807,7 @@ }, { "BriefDescription": "Fixed Counter: Counts the number of retirement slots not consumed due to front end stalls.", - "Counter": "37", + "Counter": "Fixed counter 5", "EventName": "TOPDOWN_FE_BOUND.ALL", "SampleAfterValue": "1000003", "UMask": "0x6",
@@ -1903,7 +1913,7 @@ }, { "BriefDescription": "Fixed Counter: Counts the number of consumed retirement slots.", - "Counter": "38", + "Counter": "Fixed counter 6", "EventName": "TOPDOWN_RETIRING.ALL", "SampleAfterValue": "1000003", "UMask": "0x7",
tools/perf/pmu-events/arch/x86/lunarlake/virtual-memory.json +0 -18
@@ -36,24 +36,6 @@ "UMask": "0x320", "Unit": "cpu_core" }, - { - "BriefDescription": "Counts the number of first level TLB misses but second level hits due to a demand load that did not start a page walk. Account for 4k page size only. Will result in a DTLB write from STLB.", - "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0x08", - "EventName": "DTLB_LOAD_MISSES.STLB_HIT_4K", - "SampleAfterValue": "200003", - "UMask": "0x20", - "Unit": "cpu_atom" - }, - { - "BriefDescription": "Counts the number of first level TLB misses but second level hits due to a demand load that did not start a page walk. Account for large page sizes only. Will result in a DTLB write from STLB.", - "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0x08", - "EventName": "DTLB_LOAD_MISSES.STLB_HIT_LGPG", - "SampleAfterValue": "200003", - "UMask": "0x40", - "Unit": "cpu_atom" - }, { "BriefDescription": "Cycles when at least one PMH is busy with a page walk for a demand load.", "Counter": "0,1,2,3,4,5,6,7,8,9",
tools/perf/pmu-events/arch/x86/mapfile.csv +1 -1
@@ -22,7 +22,7 @@ GenuineIntel-6-3A,v24,ivybridge,core GenuineIntel-6-3E,v24,ivytown,core GenuineIntel-6-2D,v24,jaketown,core GenuineIntel-6-(57|85),v16,knightslanding,core -GenuineIntel-6-BD,v1.11,lunarlake,core +GenuineIntel-6-BD,v1.14,lunarlake,core GenuineIntel-6-(AA|AC|B5),v1.13,meteorlake,core GenuineIntel-6-1[AEF],v4,nehalemep,core GenuineIntel-6-2E,v4,nehalemex,core
Signature · 1 line
2.50.0.727.gbf7dc18ff4-goog
Ian Rogers
reviewerPATCH v2 10/15 perf vendor events: Update MeteorLake events
In reply to Ian RogersUpdate events from v1.13 to v1.14.
Bring in the event updates v1.14:
https://github.com/intel/perfmon/commit/6c53969b8d1a83afe6ae90149c8dd4ee416027ef
--- tools/perf/pmu-events/arch/x86/mapfile.csv | 2 +- .../pmu-events/arch/x86/meteorlake/cache.json | 2 +- .../arch/x86/meteorlake/frontend.json | 72 +++++++++++++++++++ .../arch/x86/meteorlake/pipeline.json | 2 +- 4 files changed, 75 insertions(+), 3 deletions(-)
tools/perf/pmu-events/arch/x86/mapfile.csv +1 -1
@@ -23,7 +23,7 @@ GenuineIntel-6-3E,v24,ivytown,core GenuineIntel-6-2D,v24,jaketown,core GenuineIntel-6-(57|85),v16,knightslanding,core GenuineIntel-6-BD,v1.14,lunarlake,core -GenuineIntel-6-(AA|AC|B5),v1.13,meteorlake,core +GenuineIntel-6-(AA|AC|B5),v1.14,meteorlake,core GenuineIntel-6-1[AEF],v4,nehalemep,core GenuineIntel-6-2E,v4,nehalemex,core GenuineIntel-6-A7,v1.04,rocketlake,core
tools/perf/pmu-events/arch/x86/meteorlake/cache.json +1 -1
@@ -231,7 +231,7 @@ "Unit": "cpu_core" }, { - "BriefDescription": "Counts the number of L2 Cache Accesses that miss the L2 and get BBL reject short and long rejects (includes those counted in L2_reject_XQ.any), per core event", + "BriefDescription": "Counts the number of L2 Cache Accesses that miss the L2 and get BBL reject short and long rejects, per core event", "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0x24", "EventName": "L2_REQUEST.REJECTS",
tools/perf/pmu-events/arch/x86/meteorlake/frontend.json +72 -0
96 lines not on this page · raw message
tools/perf/pmu-events/arch/x86/meteorlake/pipeline.json +1 -1
@@ -1076,7 +1076,7 @@ "Counter": "0,1,2,3", "EventCode": "0x4c", "EventName": "LOAD_HIT_PREFETCH.SWPF", - "PublicDescription": "Counts all not software-prefetch load dispatches that hit the fill buffer (FB) allocated for the software prefetch. It can also be incremented by some lock instructions. So it should only be used with profiling so that the locks can be excluded by ASM (Assembly File) inspection of the nearby instructions. Available PDIST counters: 0", + "PublicDescription": "Counts all software-prefetch load dispatches that hit the fill buffer (FB) allocated for the software prefetch. It can also be incremented by some lock instructions. So it should only be used with profiling so that the locks can be excluded by ASM (Assembly File) inspection of the nearby instructions. Available PDIST counters: 0", "SampleAfterValue": "100003", "UMask": "0x1", "Unit": "cpu_core"
Signature · 1 line
2.50.0.727.gbf7dc18ff4-goog
Ian Rogers
reviewerPATCH v2 11/15 perf vendor events: Add PantherLake events
In reply to Ian RogersBring in the events at v1.00:
https://github.com/intel/perfmon/commit/d90a6737d0e4e6fbea4a5951e829615fd8317c24
--- tools/perf/pmu-events/arch/x86/mapfile.csv | 1 + .../arch/x86/pantherlake/cache.json | 278 +++++++++++++++ .../arch/x86/pantherlake/counter.json | 12 + .../arch/x86/pantherlake/frontend.json | 30 ++ .../arch/x86/pantherlake/memory.json | 215 ++++++++++++ .../arch/x86/pantherlake/pipeline.json | 325 ++++++++++++++++++ .../arch/x86/pantherlake/virtual-memory.json | 62 ++++ 7 files changed, 923 insertions(+) create mode 100644 tools/perf/pmu-events/arch/x86/pantherlake/cache.json create mode 100644 tools/perf/pmu-events/arch/x86/pantherlake/counter.json create mode 100644 tools/perf/pmu-events/arch/x86/pantherlake/frontend.json create mode 100644 tools/perf/pmu-events/arch/x86/pantherlake/memory.json create mode 100644 tools/perf/pmu-events/arch/x86/pantherlake/pipeline.json create mode 100644 tools/perf/pmu-events/arch/x86/pantherlake/virtual-memory.json
tools/perf/pmu-events/arch/x86/mapfile.csv +1 -0
@@ -26,6 +26,7 @@ GenuineIntel-6-BD,v1.14,lunarlake,core GenuineIntel-6-(AA|AC|B5),v1.14,meteorlake,core GenuineIntel-6-1[AEF],v4,nehalemep,core GenuineIntel-6-2E,v4,nehalemex,core +GenuineIntel-6-CC,v1.00,pantherlake,core GenuineIntel-6-A7,v1.04,rocketlake,core GenuineIntel-6-2A,v19,sandybridge,core GenuineIntel-6-8F,v1.25,sapphirerapids,core
tools/perf/pmu-events/arch/x86/pantherlake/cache.json new file +278 -0
278 lines not on this page · raw message
tools/perf/pmu-events/arch/x86/pantherlake/counter.json new file +12 -0
@@ -0,0 +1,12 @@ +[ + { + "Unit": "cpu_atom", + "CountersNumFixed": "3", + "CountersNumGeneric": "39" + }, + { + "Unit": "cpu_core", + "CountersNumFixed": "4", + "CountersNumGeneric": "10" + } +]
tools/perf/pmu-events/arch/x86/pantherlake/frontend.json new file +30 -0
@@ -0,0 +1,30 @@ +[ + { + "BriefDescription": "Counts every time the code stream enters into a new cache line by walking sequential from the previous line or being redirected by a jump.", + "Counter": "0,1,2,3,4,5,6,7", + "EventCode": "0x80", + "EventName": "ICACHE.ACCESSES", + "SampleAfterValue": "1000003", + "UMask": "0x3", + "Unit": "cpu_atom" + }, + { + "BriefDescription": "Counts every time the code stream enters into a new cache line by walking sequential from the previous line or being redirected by a jump and the instruction cache registers bytes are not present. -", + "Counter": "0,1,2,3,4,5,6,7", + "EventCode": "0x80", + "EventName": "ICACHE.MISSES", + "SampleAfterValue": "1000003", + "UMask": "0x2", + "Unit": "cpu_atom" + }, + { + "BriefDescription": "This event counts a subset of the Topdown Slots event that when no operation was delivered to the back-end pipeline due to instruction fetch limitations when the back-end could have accepted more operations. Common examples include instruction cache misses or x86 instruction decode limitations.", + "Counter": "0,1,2,3,4,5,6,7,8,9", + "EventCode": "0x9c", + "EventName": "IDQ_BUBBLES.CORE", + "PublicDescription": "This event counts a subset of the Topdown Slots event that when no operation was delivered to the back-end pipeline due to instruction fetch limitations when the back-end could have accepted more operations. Common examples include instruction cache misses or x86 instruction decode limitations. Software can use this event as the numerator for the Frontend Bound metric (or top-level category) of the Top-down Microarchitecture Analysis method.", + "SampleAfterValue": "1000003", + "UMask": "0x1", + "Unit": "cpu_core" + } +]
tools/perf/pmu-events/arch/x86/pantherlake/memory.json new file +215 -0
215 lines not on this page · raw message
tools/perf/pmu-events/arch/x86/pantherlake/pipeline.json new file +325 -0
325 lines not on this page · raw message
tools/perf/pmu-events/arch/x86/pantherlake/virtual-memory.json new file +62 -0
62 lines not on this page · raw message
Signature · 1 line
2.50.0.727.gbf7dc18ff4-goog
Ian Rogers
reviewerPATCH v2 12/15 perf vendor events: Update SapphireRapids events
In reply to Ian RogersUpdate events from v1.25 to v1.28.
Bring in the event updates v1.28:
https://github.com/intel/perfmon/commit/990bfdff270adf08d408534d6d66ba47ec6adb34
https://github.com/intel/perfmon/commit/b7b4d7f18cf9a893438777a571abc7ecc087368b
--- tools/perf/pmu-events/arch/x86/mapfile.csv | 2 +- .../arch/x86/sapphirerapids/pipeline.json | 2 +- .../arch/x86/sapphirerapids/uncore-io.json | 12 +++++++++++ .../x86/sapphirerapids/uncore-memory.json | 20 +++++++++++++++++++ 4 files changed, 34 insertions(+), 2 deletions(-)
tools/perf/pmu-events/arch/x86/mapfile.csv +1 -1
@@ -29,7 +29,7 @@ GenuineIntel-6-2E,v4,nehalemex,core GenuineIntel-6-CC,v1.00,pantherlake,core GenuineIntel-6-A7,v1.04,rocketlake,core GenuineIntel-6-2A,v19,sandybridge,core -GenuineIntel-6-8F,v1.25,sapphirerapids,core +GenuineIntel-6-8F,v1.28,sapphirerapids,core GenuineIntel-6-AF,v1.09,sierraforest,core GenuineIntel-6-(37|4A|4C|4D|5A),v15,silvermont,core GenuineIntel-6-(4E|5E|8E|9E|A5|A6),v59,skylake,core
tools/perf/pmu-events/arch/x86/sapphirerapids/pipeline.json +1 -1
@@ -684,7 +684,7 @@ "Counter": "0,1,2,3", "EventCode": "0x4c", "EventName": "LOAD_HIT_PREFETCH.SWPF", - "PublicDescription": "Counts all not software-prefetch load dispatches that hit the fill buffer (FB) allocated for the software prefetch. It can also be incremented by some lock instructions. So it should only be used with profiling so that the locks can be excluded by ASM (Assembly File) inspection of the nearby instructions. Available PDIST counters: 0", + "PublicDescription": "Counts all software-prefetch load dispatches that hit the fill buffer (FB) allocated for the software prefetch. It can also be incremented by some lock instructions. So it should only be used with profiling so that the locks can be excluded by ASM (Assembly File) inspection of the nearby instructions. Available PDIST counters: 0", "SampleAfterValue": "100003", "UMask": "0x1" },
tools/perf/pmu-events/arch/x86/sapphirerapids/uncore-io.json +12 -0
@@ -1901,6 +1901,18 @@ "UMask": "0x4", "Unit": "IIO" }, + { + "BriefDescription": "Posted requests sent by the integrated IO (IIO) controller to the Ubox, useful for counting message signaled interrupts (MSI).", + "Counter": "0,1,2,3", + "EventCode": "0x8e", + "EventName": "UNC_IIO_NUM_REQ_OF_CPU_BY_TGT.UBOX_POSTED", + "Experimental": "1", + "FCMask": "0x01", + "PerPkg": "1", + "PortMask": "0x00FF", + "UMask": "0x4", + "Unit": "IIO" + }, { "BriefDescription": "ITC address map 1", "Counter": "0,1,2,3",
tools/perf/pmu-events/arch/x86/sapphirerapids/uncore-memory.json +20 -0
@@ -2145,6 +2145,16 @@ "UMask": "0x1", "Unit": "MCHBM" }, + { + "BriefDescription": "ECC Correctable Errors", + "Counter": "0,1,2,3", + "EventCode": "0x09", + "EventName": "UNC_MCHBM_ECC_CORRECTABLE_ERRORS", + "Experimental": "1", + "PerPkg": "1", + "PublicDescription": "ECC Correctable Errors. Counts the number of ECC errors detected and corrected by the iMC on this channel. This counter is only useful with ECC devices. This count will increment one time for each correction regardless of the number of bits corrected. The iMC can correct up to 4 bit errors in independent channel mode and 8 bit errors in lockstep mode.", + "Unit": "MCHBM" + }, { "BriefDescription": "HBM Precharge All Commands", "Counter": "0,1,2,3",
@@ -2759,6 +2769,16 @@ "UMask": "0x3", "Unit": "iMC" }, + { + "BriefDescription": "ECC Correctable Errors", + "Counter": "0,1,2,3", + "EventCode": "0x09", + "EventName": "UNC_M_ECC_CORRECTABLE_ERRORS", + "Experimental": "1", + "PerPkg": "1", + "PublicDescription": "ECC Correctable Errors : Counts the number of ECC errors detected and corrected by the iMC on this channel. This counter is only useful with ECC DRAM devices. This count will increment one time for each correction regardless of the number of bits corrected. The iMC can correct up to 4 bit errors in independent channel mode and 8 bit errors in lockstep mode.", + "Unit": "iMC" + }, { "BriefDescription": "IMC Clockticks at HCLK frequency", "Counter": "0,1,2,3",
Signature · 1 line
2.50.0.727.gbf7dc18ff4-goog
Ian Rogers
reviewerPATCH v2 13/15 perf vendor events: Update SierraForest events
In reply to Ian RogersUpdate events from v1.09 to v1.11.
Bring in the event updates v1.11:
https://github.com/intel/perfmon/commit/6b824df1dba3948146281c8ba2a8c3e7bf7f7c51
https://github.com/intel/perfmon/commit/4b0346fbee2b04dd34526522250116aee525c922
--- tools/perf/pmu-events/arch/x86/mapfile.csv | 2 +- .../arch/x86/sierraforest/frontend.json | 64 +++++++++++++++++++ .../arch/x86/sierraforest/pipeline.json | 8 +++ .../arch/x86/sierraforest/srf-metrics.json | 48 ++++++++++++++ .../arch/x86/sierraforest/uncore-cache.json | 6 +- .../x86/sierraforest/uncore-interconnect.json | 53 ++++++++++++--- .../arch/x86/sierraforest/uncore-io.json | 27 +++++++- 7 files changed, 194 insertions(+), 14 deletions(-)
tools/perf/pmu-events/arch/x86/mapfile.csv +1 -1
@@ -30,7 +30,7 @@ GenuineIntel-6-CC,v1.00,pantherlake,core GenuineIntel-6-A7,v1.04,rocketlake,core GenuineIntel-6-2A,v19,sandybridge,core GenuineIntel-6-8F,v1.28,sapphirerapids,core -GenuineIntel-6-AF,v1.09,sierraforest,core +GenuineIntel-6-AF,v1.11,sierraforest,core GenuineIntel-6-(37|4A|4C|4D|5A),v15,silvermont,core GenuineIntel-6-(4E|5E|8E|9E|A5|A6),v59,skylake,core GenuineIntel-6-55-[01234],v1.36,skylakex,core
tools/perf/pmu-events/arch/x86/sierraforest/frontend.json +64 -0
76 lines not on this page · raw message
tools/perf/pmu-events/arch/x86/sierraforest/pipeline.json +8 -0
@@ -11,6 +11,7 @@ { "BriefDescription": "Counts the total number of branch instructions retired for all branch types.", "Counter": "0,1,2,3,4,5,6,7", + "Errata": "SRF6, SRF7", "EventCode": "0xc4", "EventName": "BR_INST_RETIRED.ALL_BRANCHES", "PublicDescription": "Counts the total number of instructions in which the instruction pointer (IP) of the processor is resteered due to a branch instruction and the branch instruction successfully retires. All branch type instructions are accounted for.",
@@ -19,6 +20,7 @@ { "BriefDescription": "Counts the number of retired JCC (Jump on Conditional Code) branch instructions retired, includes both taken and not taken branches.", "Counter": "0,1,2,3,4,5,6,7", + "Errata": "SRF7", "EventCode": "0xc4", "EventName": "BR_INST_RETIRED.COND", "SampleAfterValue": "200003",
@@ -35,6 +37,7 @@ { "BriefDescription": "Counts the number of far branch instructions retired, includes far jump, far call and return, and interrupt call and return.", "Counter": "0,1,2,3,4,5,6,7", + "Errata": "SRF7", "EventCode": "0xc4", "EventName": "BR_INST_RETIRED.FAR_BRANCH", "SampleAfterValue": "200003",
@@ -43,6 +46,7 @@ { "BriefDescription": "Counts the number of near indirect JMP and near indirect CALL branch instructions retired.", "Counter": "0,1,2,3,4,5,6,7", + "Errata": "SRF7", "EventCode": "0xc4", "EventName": "BR_INST_RETIRED.INDIRECT", "SampleAfterValue": "200003",
@@ -51,6 +55,7 @@ { "BriefDescription": "Counts the number of near indirect CALL branch instructions retired.", "Counter": "0,1,2,3,4,5,6,7", + "Errata": "SRF7", "EventCode": "0xc4", "EventName": "BR_INST_RETIRED.INDIRECT_CALL", "SampleAfterValue": "200003",
@@ -68,6 +73,7 @@ "BriefDescription": "This event is deprecated. Refer to new event BR_INST_RETIRED.INDIRECT_CALL", "Counter": "0,1,2,3,4,5,6,7", "Deprecated": "1", + "Errata": "SRF7", "EventCode": "0xc4", "EventName": "BR_INST_RETIRED.IND_CALL", "SampleAfterValue": "200003",
@@ -76,6 +82,7 @@ { "BriefDescription": "Counts the number of near CALL branch instructions retired.", "Counter": "0,1,2,3,4,5,6,7", + "Errata": "SRF6, SRF7", "EventCode": "0xc4", "EventName": "BR_INST_RETIRED.NEAR_CALL", "SampleAfterValue": "200003",
@@ -92,6 +99,7 @@ { "BriefDescription": "Counts the number of near taken branch instructions retired.", "Counter": "0,1,2,3,4,5,6,7", + "Errata": "SRF7", "EventCode": "0xc4", "EventName": "BR_INST_RETIRED.NEAR_TAKEN", "SampleAfterValue": "200003",
tools/perf/pmu-events/arch/x86/sierraforest/srf-metrics.json +48 -0
72 lines not on this page · raw message
tools/perf/pmu-events/arch/x86/sierraforest/uncore-cache.json +3 -3
@@ -874,7 +874,7 @@ "Unit": "CHA" }, { - "BriefDescription": "Counts snoop filter capacity evictions for entries tracking exclusive lines in the cores? cache.? Snoop filter capacity evictions occur when the snoop filter is full and evicts an existing entry to track a new entry.? Does not count clean evictions such as when a core?s cache replaces a tracked cacheline with a new cacheline.", + "BriefDescription": "Counts snoop filter capacity evictions for entries tracking exclusive lines in the core's cache. Snoop filter capacity evictions occur when the snoop filter is full and evicts an existing entry to track a new entry. Does not count clean evictions such as when a core's cache replaces a tracked cacheline with a new cacheline.", "Counter": "0,1,2,3", "EventCode": "0x3d", "EventName": "UNC_CHA_SF_EVICTION.E_STATE",
@@ -885,7 +885,7 @@ "Unit": "CHA" }, { - "BriefDescription": "Counts snoop filter capacity evictions for entries tracking modified lines in the cores? cache.? Snoop filter capacity evictions occur when the snoop filter is full and evicts an existing entry to track a new entry.? Does not count clean evictions such as when a core?s cache replaces a tracked cacheline with a new cacheline.", + "BriefDescription": "Counts snoop filter capacity evictions for entries tracking modified lines in the core's cache. Snoop filter capacity evictions occur when the snoop filter is full and evicts an existing entry to track a new entry. Does not count clean evictions such as when a core's cache replaces a tracked cacheline with a new cacheline.", "Counter": "0,1,2,3", "EventCode": "0x3d", "EventName": "UNC_CHA_SF_EVICTION.M_STATE",
@@ -895,7 +895,7 @@ "Unit": "CHA" }, { - "BriefDescription": "Counts snoop filter capacity evictions for entries tracking shared lines in the cores? cache.? Snoop filter capacity evictions occur when the snoop filter is full and evicts an existing entry to track a new entry.? Does not count clean evictions such as when a core?s cache replaces a tracked cacheline with a new cacheline.", + "BriefDescription": "Counts snoop filter capacity evictions for entries tracking shared lines in the core's cache. Snoop filter capacity evictions occur when the snoop filter is full and evicts an existing entry to track a new entry. Does not count clean evictions such as when a core's cache replaces a tracked cacheline with a new cacheline.", "Counter": "0,1,2,3", "EventCode": "0x3d", "EventName": "UNC_CHA_SF_EVICTION.S_STATE",
tools/perf/pmu-events/arch/x86/sierraforest/uncore-interconnect.json +44 -9
65 lines not on this page · raw message
tools/perf/pmu-events/arch/x86/sierraforest/uncore-io.json +26 -1
@@ -1121,8 +1121,9 @@ "Unit": "IIO" }, { - "BriefDescription": "Occupancy of outbound request queue : To device : Counts number of outbound requests/completions IIO is currently processing", + "BriefDescription": "This event is deprecated. [This event is alias to UNC_IIO_NUM_OUTSTANDING_REQ_FROM_CPU.TO_IO]", "Counter": "2,3", + "Deprecated": "1", "EventCode": "0xc5", "EventName": "UNC_IIO_NUM_OUSTANDING_REQ_FROM_CPU.TO_IO", "Experimental": "1",
@@ -1132,6 +1133,18 @@ "UMask": "0x8", "Unit": "IIO" }, + { + "BriefDescription": "Occupancy of outbound request queue : To device : Counts number of outbound requests/completions IIO is currently processing [This event is alias to UNC_IIO_NUM_OUSTANDING_REQ_FROM_CPU.TO_IO]", + "Counter": "2,3", + "EventCode": "0xc5", + "EventName": "UNC_IIO_NUM_OUTSTANDING_REQ_FROM_CPU.TO_IO", + "Experimental": "1", + "FCMask": "0x07", + "PerPkg": "1", + "PortMask": "0x0FF", + "UMask": "0x8", + "Unit": "IIO" + }, { "BriefDescription": "Passing data to be written", "Counter": "0,1,2,3",
@@ -1300,6 +1313,18 @@ "UMask": "0x4", "Unit": "IIO" }, + { + "BriefDescription": "Posted requests sent by the integrated IO (IIO) controller to the Ubox, useful for counting message signaled interrupts (MSI).", + "Counter": "0,1,2,3", + "EventCode": "0x8e", + "EventName": "UNC_IIO_NUM_REQ_OF_CPU_BY_TGT.UBOX_POSTED", + "FCMask": "0x01", + "PerPkg": "1", + "PortMask": "0x0FF", + "PublicDescription": "-", + "UMask": "0x4", + "Unit": "IIO" + }, { "BriefDescription": "All 9 bits of Page Walk Tracker Occupancy", "Counter": "0,1,2,3",
Signature · 1 line
2.50.0.727.gbf7dc18ff4-goog
Ian Rogers
reviewerPATCH v2 14/15 perf vendor events: Update SkylakeX events
In reply to Ian RogersUpdate events from v1.36 to v1.37.
Bring in the event updates v1.37:
https://github.com/intel/perfmon/commit/6ee8e4cadda8b6954bd84236e20fab95e345578f
--- tools/perf/pmu-events/arch/x86/mapfile.csv | 2 +- tools/perf/pmu-events/arch/x86/skylakex/pipeline.json | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-)
tools/perf/pmu-events/arch/x86/mapfile.csv +1 -1
@@ -33,7 +33,7 @@ GenuineIntel-6-8F,v1.28,sapphirerapids,core GenuineIntel-6-AF,v1.11,sierraforest,core GenuineIntel-6-(37|4A|4C|4D|5A),v15,silvermont,core GenuineIntel-6-(4E|5E|8E|9E|A5|A6),v59,skylake,core -GenuineIntel-6-55-[01234],v1.36,skylakex,core +GenuineIntel-6-55-[01234],v1.37,skylakex,core GenuineIntel-6-86,v1.23,snowridgex,core GenuineIntel-6-8[CD],v1.17,tigerlake,core GenuineIntel-6-2C,v5,westmereep-dp,core
tools/perf/pmu-events/arch/x86/skylakex/pipeline.json +1 -1
@@ -542,7 +542,7 @@ "Counter": "0,1,2,3", "EventCode": "0x4C", "EventName": "LOAD_HIT_PRE.SW_PF", - "PublicDescription": "Counts all not software-prefetch load dispatches that hit the fill buffer (FB) allocated for the software prefetch. It can also be incremented by some lock instructions. So it should only be used with profiling so that the locks can be excluded by ASM (Assembly File) inspection of the nearby instructions.", + "PublicDescription": "Counts all software-prefetch load dispatches that hit the fill buffer (FB) allocated for the software prefetch. It can also be incremented by some lock instructions. So it should only be used with profiling so that the locks can be excluded by ASM (Assembly File) inspection of the nearby instructions.", "SampleAfterValue": "100003", "UMask": "0x1" },
Signature · 1 line
2.50.0.727.gbf7dc18ff4-goog
Ian Rogers
reviewerPATCH v2 15/15 perf vendor events: Update TigerLake events
In reply to Ian RogersUpdate events from v1.17 to v1.18.
Bring in the event updates v1.18:
https://github.com/intel/perfmon/commit/943fea37d0d54232605f12abf72a812ac314cd1d
--- tools/perf/pmu-events/arch/x86/mapfile.csv | 2 +- tools/perf/pmu-events/arch/x86/tigerlake/pipeline.json | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-)
tools/perf/pmu-events/arch/x86/mapfile.csv +1 -1
@@ -35,7 +35,7 @@ GenuineIntel-6-(37|4A|4C|4D|5A),v15,silvermont,core GenuineIntel-6-(4E|5E|8E|9E|A5|A6),v59,skylake,core GenuineIntel-6-55-[01234],v1.37,skylakex,core GenuineIntel-6-86,v1.23,snowridgex,core -GenuineIntel-6-8[CD],v1.17,tigerlake,core +GenuineIntel-6-8[CD],v1.18,tigerlake,core GenuineIntel-6-2C,v5,westmereep-dp,core GenuineIntel-6-25,v4,westmereep-sp,core GenuineIntel-6-2F,v4,westmereex,core
tools/perf/pmu-events/arch/x86/tigerlake/pipeline.json +1 -1
@@ -497,7 +497,7 @@ "Counter": "0,1,2,3", "EventCode": "0x4c", "EventName": "LOAD_HIT_PREFETCH.SWPF", - "PublicDescription": "Counts all not software-prefetch load dispatches that hit the fill buffer (FB) allocated for the software prefetch. It can also be incremented by some lock instructions. So it should only be used with profiling so that the locks can be excluded by ASM (Assembly File) inspection of the nearby instructions.", + "PublicDescription": "Counts all software-prefetch load dispatches that hit the fill buffer (FB) allocated for the software prefetch. It can also be incremented by some lock instructions. So it should only be used with profiling so that the locks can be excluded by ASM (Assembly File) inspection of the nearby instructions.", "SampleAfterValue": "100003", "UMask": "0x1" },
Signature · 1 line
2.50.0.727.gbf7dc18ff4-goog
Quoted 16 lines · Ian Rogers
Update:
- ADL/ADLN v1.29 -> 1.31
- ARL 1.08 -> 1.09
- CLX 1.23 -> 1.25
- EMR 1.11 -> 1.14
- GRR 1.07 -> 1.09
- GNR 1.08 -> 1.10
- ICX 1.27 -> 1.28
- LNL 1.11 -> 1.14
- MTL 1.13 -> 1.14
- PTL new v1.00
- SPR 1.25 -> 1.28
- SRF 1.09 -> 1.11
- SKX 1.36 -> 1.37
- TGL 1.17 -> 1.18
Quoted 8 lines
Updates from:
https://github.com/intel/perfmon
with:
https://github.com/intel/perfmon/pull/312
Running the script:
https://github.com/intel/perfmon/blob/main/scripts/create_perf_json.pyv2: Rebase, no conflicts.
Tested on ADL and ARL, LGTM.
Thanks,
Tom
Quoted 78 lines
Ian Rogers (15):
perf vendor events: Update Alderlake events perf vendor events: Update AlderlakeN events perf vendor events: Update Arrowlake events perf vendor events: Update CascadelakeX events perf vendor events: Update EmeraldRapids events perf vendor events: Update GrandRidge events perf vendor events: Update GraniteRapids events perf vendor events: Update IcelakeX events perf vendor events: Update LunarLake events perf vendor events: Update MeteorLake events perf vendor events: Add PantherLake events perf vendor events: Update SapphireRapids events perf vendor events: Update SierraForest events perf vendor events: Update SkylakeX events perf vendor events: Update TigerLake events.../pmu-events/arch/x86/alderlake/cache.json | 56 ++- .../arch/x86/alderlake/floating-point.json | 1 - .../pmu-events/arch/x86/alderlake/other.json | 1 - .../arch/x86/alderlake/pipeline.json | 44 +-- .../arch/x86/alderlake/virtual-memory.json | 3 - .../pmu-events/arch/x86/alderlaken/cache.json | 52 +-- .../arch/x86/alderlaken/floating-point.json | 1 - .../pmu-events/arch/x86/alderlaken/other.json | 1 - .../arch/x86/alderlaken/pipeline.json | 42 +-- .../arch/x86/alderlaken/virtual-memory.json | 3 - .../pmu-events/arch/x86/arrowlake/cache.json | 13 +- .../arch/x86/arrowlake/frontend.json | 135 ++++++++ .../arch/x86/cascadelakex/floating-point.json | 6 +- .../arch/x86/cascadelakex/pipeline.json | 2 +- .../arch/x86/emeraldrapids/pipeline.json | 2 +- .../arch/x86/emeraldrapids/uncore-io.json | 12 + .../arch/x86/emeraldrapids/uncore-memory.json | 20 ++ .../arch/x86/grandridge/grr-metrics.json | 30 +- .../x86/grandridge/uncore-interconnect.json | 10 + .../arch/x86/grandridge/uncore-io.json | 12 + .../arch/x86/graniterapids/cache.json | 9 + .../arch/x86/graniterapids/counter.json | 10 +- .../arch/x86/graniterapids/gnr-metrics.json | 36 ++ .../arch/x86/graniterapids/pipeline.json | 2 +- .../graniterapids/uncore-interconnect.json | 19 - .../arch/x86/graniterapids/uncore-io.json | 27 +- .../arch/x86/icelakex/pipeline.json | 2 +- .../arch/x86/icelakex/uncore-cache.json | 2 - .../pmu-events/arch/x86/lunarlake/cache.json | 11 + .../arch/x86/lunarlake/pipeline.json | 18 +- .../arch/x86/lunarlake/virtual-memory.json | 18 - tools/perf/pmu-events/arch/x86/mapfile.csv | 29 +- .../pmu-events/arch/x86/meteorlake/cache.json | 2 +- .../arch/x86/meteorlake/frontend.json | 72 ++++ .../arch/x86/meteorlake/pipeline.json | 2 +- .../arch/x86/pantherlake/cache.json | 278 +++++++++++++++ .../arch/x86/pantherlake/counter.json | 12 + .../arch/x86/pantherlake/frontend.json | 30 ++ .../arch/x86/pantherlake/memory.json | 215 ++++++++++++ .../arch/x86/pantherlake/pipeline.json | 325 ++++++++++++++++++ .../arch/x86/pantherlake/virtual-memory.json | 62 ++++ .../arch/x86/sapphirerapids/pipeline.json | 2 +- .../arch/x86/sapphirerapids/uncore-io.json | 12 + .../x86/sapphirerapids/uncore-memory.json | 20 ++ .../arch/x86/sierraforest/frontend.json | 64 ++++ .../arch/x86/sierraforest/pipeline.json | 8 + .../arch/x86/sierraforest/srf-metrics.json | 48 +++ .../arch/x86/sierraforest/uncore-cache.json | 6 +- .../x86/sierraforest/uncore-interconnect.json | 53 ++- .../arch/x86/sierraforest/uncore-io.json | 27 +- .../arch/x86/skylakex/pipeline.json | 2 +- .../arch/x86/tigerlake/pipeline.json | 2 +- 52 files changed, 1623 insertions(+), 248 deletions(-) create mode 100644 tools/perf/pmu-events/arch/x86/pantherlake/cache.json create mode 100644 tools/perf/pmu-events/arch/x86/pantherlake/counter.json create mode 100644 tools/perf/pmu-events/arch/x86/pantherlake/frontend.json create mode 100644 tools/perf/pmu-events/arch/x86/pantherlake/memory.json create mode 100644 tools/perf/pmu-events/arch/x86/pantherlake/pipeline.json create mode 100644 tools/perf/pmu-events/arch/x86/pantherlake/virtual-memory.json
Quoted 16 lines · Ian Rogers
Update:
- ADL/ADLN v1.29 -> 1.31
- ARL 1.08 -> 1.09
- CLX 1.23 -> 1.25
- EMR 1.11 -> 1.14
- GRR 1.07 -> 1.09
- GNR 1.08 -> 1.10
- ICX 1.27 -> 1.28
- LNL 1.11 -> 1.14
- MTL 1.13 -> 1.14
- PTL new v1.00
- SPR 1.25 -> 1.28
- SRF 1.09 -> 1.11
- SKX 1.36 -> 1.37
- TGL 1.17 -> 1.18
Quoted 1 line
[...]
Applied to perf-tools-next, thanks!
Best regards,
Namhyung