perf vendor events intel: Update alderlake/alderlake events to v1.23
Subsystems (2)
Ian Rogers
reviewerPATCH 1/9 perf vendor events intel: Update alderlake/alderlake events to v1.23
Update alderlake and alderlaken events from v1.21 to v1.23 adding the changes from:
https://github.com/intel/perfmon/commit/8df4db9433a2aab59dbbac1a70281032d1af7734
https://github.com/intel/perfmon/commit/846bd247c6e04acc572ca56c992e9e65852bbe63
The tsx_cycles_per_elision metric is updated from PR:
https://github.com/intel/perfmon/pull/116
--- .../arch/x86/alderlake/adl-metrics.json | 2 +- .../arch/x86/alderlake/frontend.json | 42 ++++++++++-- .../pmu-events/arch/x86/alderlake/memory.json | 4 +- .../arch/x86/alderlake/pipeline.json | 20 +++++- .../x86/alderlake/uncore-interconnect.json | 2 + .../arch/x86/alderlaken/memory.json | 4 +- .../arch/x86/alderlaken/pipeline.json | 16 +++++ .../x86/alderlaken/uncore-interconnect.json | 66 +++++++++++++++++++ tools/perf/pmu-events/arch/x86/mapfile.csv | 4 +- 9 files changed, 146 insertions(+), 14 deletions(-)
tools/perf/pmu-events/arch/x86/alderlake/adl-metrics.json +1 -1
@@ -99,7 +99,7 @@ }, { "BriefDescription": "Number of cycles within a transaction divided by the number of elisions.", - "MetricExpr": "(cycles\\-t / el\\-start if has_event(cycles\\-t) else 0)", + "MetricExpr": "(cycles\\-t / el\\-start if has_event(el\\-start) else 0)", "MetricGroup": "transaction", "MetricName": "tsx_cycles_per_elision", "ScaleUnit": "1cycles / elision"
tools/perf/pmu-events/arch/x86/alderlake/frontend.json +36 -6
67 lines not on this page · raw message
tools/perf/pmu-events/arch/x86/alderlake/memory.json +2 -2
@@ -248,7 +248,7 @@ "Unit": "cpu_core" }, { - "BriefDescription": "Counts demand data reads that were not supplied by the L3 cache.", + "BriefDescription": "Counts demand data reads that were not supplied by the L3 cache. [L3_MISS_LOCAL is alias to L3_MISS]", "EventCode": "0xB7", "EventName": "OCR.DEMAND_DATA_RD.L3_MISS_LOCAL", "MSRIndex": "0x1a6,0x1a7",
@@ -278,7 +278,7 @@ "Unit": "cpu_core" }, { - "BriefDescription": "Counts demand reads for ownership (RFO) and software prefetches for exclusive ownership (PREFETCHW) that were not supplied by the L3 cache.", + "BriefDescription": "Counts demand reads for ownership (RFO) and software prefetches for exclusive ownership (PREFETCHW) that were not supplied by the L3 cache. [L3_MISS_LOCAL is alias to L3_MISS]", "EventCode": "0xB7", "EventName": "OCR.DEMAND_RFO.L3_MISS_LOCAL", "MSRIndex": "0x1a6,0x1a7",
tools/perf/pmu-events/arch/x86/alderlake/pipeline.json +19 -1
@@ -238,6 +238,15 @@ "UMask": "0x8", "Unit": "cpu_core" }, + { + "BriefDescription": "Counts the number of near taken branch instructions retired.", + "EventCode": "0xc4", + "EventName": "BR_INST_RETIRED.NEAR_TAKEN", + "PEBS": "1", + "SampleAfterValue": "200003", + "UMask": "0xc0", + "Unit": "cpu_atom" + }, { "BriefDescription": "Taken branch instructions retired.", "EventCode": "0xc4",
@@ -411,6 +420,15 @@ "UMask": "0x7e", "Unit": "cpu_atom" }, + { + "BriefDescription": "Counts the number of mispredicted near taken branch instructions retired.", + "EventCode": "0xc5", + "EventName": "BR_MISP_RETIRED.NEAR_TAKEN", + "PEBS": "1", + "SampleAfterValue": "200003", + "UMask": "0x80", + "Unit": "cpu_atom" + }, { "BriefDescription": "Number of near branch instructions retired that were mispredicted and taken.", "EventCode": "0xc5",
@@ -842,7 +860,7 @@ "Unit": "cpu_core" }, { - "BriefDescription": "INT_MISC.UNKNOWN_BRANCH_CYCLES", + "BriefDescription": "Bubble cycles of BAClear (Unknown Branch).", "EventCode": "0xad", "EventName": "INT_MISC.UNKNOWN_BRANCH_CYCLES", "MSRIndex": "0x3F7",
tools/perf/pmu-events/arch/x86/alderlake/uncore-interconnect.json +2 -0
@@ -25,6 +25,7 @@ }, { "BriefDescription": "This event is deprecated. Refer to new event UNC_ARB_REQ_TRK_REQUEST.DRD", + "Deprecated": "1", "EventCode": "0x81", "EventName": "UNC_ARB_DAT_REQUESTS.RD", "PerPkg": "1",
@@ -33,6 +34,7 @@ }, { "BriefDescription": "This event is deprecated. Refer to new event UNC_ARB_DAT_OCCUPANCY.ALL", + "Deprecated": "1", "EventCode": "0x85", "EventName": "UNC_ARB_IFA_OCCUPANCY.ALL", "PerPkg": "1",
tools/perf/pmu-events/arch/x86/alderlaken/memory.json +2 -2
@@ -59,7 +59,7 @@ "UMask": "0x1" }, { - "BriefDescription": "Counts demand data reads that were not supplied by the L3 cache.", + "BriefDescription": "Counts demand data reads that were not supplied by the L3 cache. [L3_MISS_LOCAL is alias to L3_MISS]", "EventCode": "0xB7", "EventName": "OCR.DEMAND_DATA_RD.L3_MISS_LOCAL", "MSRIndex": "0x1a6,0x1a7",
@@ -77,7 +77,7 @@ "UMask": "0x1" }, { - "BriefDescription": "Counts demand reads for ownership (RFO) and software prefetches for exclusive ownership (PREFETCHW) that were not supplied by the L3 cache.", + "BriefDescription": "Counts demand reads for ownership (RFO) and software prefetches for exclusive ownership (PREFETCHW) that were not supplied by the L3 cache. [L3_MISS_LOCAL is alias to L3_MISS]", "EventCode": "0xB7", "EventName": "OCR.DEMAND_RFO.L3_MISS_LOCAL", "MSRIndex": "0x1a6,0x1a7",
tools/perf/pmu-events/arch/x86/alderlaken/pipeline.json +16 -0
@@ -90,6 +90,14 @@ "SampleAfterValue": "200003", "UMask": "0xf7" }, + { + "BriefDescription": "Counts the number of near taken branch instructions retired.", + "EventCode": "0xc4", + "EventName": "BR_INST_RETIRED.NEAR_TAKEN", + "PEBS": "1", + "SampleAfterValue": "200003", + "UMask": "0xc0" + }, { "BriefDescription": "This event is deprecated. Refer to new event BR_INST_RETIRED.INDIRECT", "Deprecated": "1",
@@ -183,6 +191,14 @@ "SampleAfterValue": "200003", "UMask": "0x7e" }, + { + "BriefDescription": "Counts the number of mispredicted near taken branch instructions retired.", + "EventCode": "0xc5", + "EventName": "BR_MISP_RETIRED.NEAR_TAKEN", + "PEBS": "1", + "SampleAfterValue": "200003", + "UMask": "0x80" + }, { "BriefDescription": "This event is deprecated. Refer to new event BR_MISP_RETIRED.INDIRECT", "Deprecated": "1",
tools/perf/pmu-events/arch/x86/alderlaken/uncore-interconnect.json +66 -0
83 lines not on this page · raw message
tools/perf/pmu-events/arch/x86/mapfile.csv +2 -2
@@ -1,6 +1,6 @@ Family-model,Version,Filename,EventType -GenuineIntel-6-(97|9A|B7|BA|BF),v1.21,alderlake,core -GenuineIntel-6-BE,v1.21,alderlaken,core +GenuineIntel-6-(97|9A|B7|BA|BF),v1.23,alderlake,core +GenuineIntel-6-BE,v1.23,alderlaken,core GenuineIntel-6-(1C|26|27|35|36),v4,bonnell,core GenuineIntel-6-(3D|47),v28,broadwell,core GenuineIntel-6-56,v11,broadwellde,core
Signature · 1 line
2.42.0.758.gaed0368e0e-goog
Ian Rogers
reviewerPATCH 2/9 perf vendor events intel: Update emeraldrapids to v1.01
Update emeraldrapids to v1.01 from v1.00 adding the changes from:
https://github.com/intel/perfmon/commit/3993b600e032a9fd443ffd828aab73de7cb167e5
--- .../arch/x86/emeraldrapids/uncore-cache.json | 36 +++++++++++++++++++ tools/perf/pmu-events/arch/x86/mapfile.csv | 2 +- 2 files changed, 37 insertions(+), 1 deletion(-)
tools/perf/pmu-events/arch/x86/emeraldrapids/uncore-cache.json +36 -0
@@ -3448,6 +3448,15 @@ "UMask": "0x10c8168201", "Unit": "CHA" }, + { + "BriefDescription": "UNC_CHA_TOR_INSERTS.IA_MISS_DRD_CXL_EXP_LOCAL", + "EventCode": "0x35", + "EventName": "UNC_CHA_TOR_INSERTS.IA_MISS_DRD_CXL_EXP_LOCAL", + "PerPkg": "1", + "PortMask": "0x000", + "UMask": "0x20c8168201", + "Unit": "CHA" + }, { "BriefDescription": "TOR Inserts for DRds issued by IA Cores targeting DDR Mem that Missed the LLC", "EventCode": "0x35",
@@ -3838,6 +3847,15 @@ "UMask": "0x10c8068201", "Unit": "CHA" }, + { + "BriefDescription": "UNC_CHA_TOR_INSERTS.IA_MISS_RFO_CXL_EXP_LOCAL", + "EventCode": "0x35", + "EventName": "UNC_CHA_TOR_INSERTS.IA_MISS_RFO_CXL_EXP_LOCAL", + "PerPkg": "1", + "PortMask": "0x000", + "UMask": "0x20c8068201", + "Unit": "CHA" + }, { "BriefDescription": "TOR Inserts RFO misses from local IA", "EventCode": "0x35",
@@ -4877,6 +4895,15 @@ "UMask": "0x10c8168201", "Unit": "CHA" }, + { + "BriefDescription": "UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_CXL_EXP_LOCAL", + "EventCode": "0x36", + "EventName": "UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_CXL_EXP_LOCAL", + "PerPkg": "1", + "PortMask": "0x000", + "UMask": "0x20c8168201", + "Unit": "CHA" + }, { "BriefDescription": "TOR Occupancy for DRds issued by iA Cores targeting DDR Mem that Missed the LLC", "EventCode": "0x36",
@@ -5267,6 +5294,15 @@ "UMask": "0x10c8068201", "Unit": "CHA" }, + { + "BriefDescription": "UNC_CHA_TOR_OCCUPANCY.IA_MISS_RFO_CXL_EXP_LOCAL", + "EventCode": "0x36", + "EventName": "UNC_CHA_TOR_OCCUPANCY.IA_MISS_RFO_CXL_EXP_LOCAL", + "PerPkg": "1", + "PortMask": "0x000", + "UMask": "0x20c8068201", + "Unit": "CHA" + }, { "BriefDescription": "TOR Occupancy; RFO misses from local IA", "EventCode": "0x36",
tools/perf/pmu-events/arch/x86/mapfile.csv +1 -1
@@ -7,7 +7,7 @@ GenuineIntel-6-56,v11,broadwellde,core GenuineIntel-6-4F,v22,broadwellx,core GenuineIntel-6-55-[56789ABCDEF],v1.20,cascadelakex,core GenuineIntel-6-9[6C],v1.04,elkhartlake,core -GenuineIntel-6-CF,v1.00,emeraldrapids,core +GenuineIntel-6-CF,v1.01,emeraldrapids,core GenuineIntel-6-5[CF],v13,goldmont,core GenuineIntel-6-7A,v1.01,goldmontplus,core GenuineIntel-6-B6,v1.00,grandridge,core
Signature · 1 line
2.42.0.758.gaed0368e0e-goog
Ian Rogers
reviewerPATCH 3/9 perf vendor events intel: Update a spelling in haswell/haswellx
In reply to Ian RogersThe spelling of "in-flight" was switched to "inflight".
--- tools/perf/pmu-events/arch/x86/haswell/memory.json | 2 +- tools/perf/pmu-events/arch/x86/haswellx/memory.json | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-)
tools/perf/pmu-events/arch/x86/haswell/memory.json +1 -1
@@ -62,7 +62,7 @@ "BriefDescription": "Counts the number of machine clears due to memory order conflicts.", "EventCode": "0xC3", "EventName": "MACHINE_CLEARS.MEMORY_ORDERING", - "PublicDescription": "This event counts the number of memory ordering machine clears detected. Memory ordering machine clears can result from memory address aliasing or snoops from another hardware thread or core to data in-flight in the pipeline. Machine clears can have a significant performance impact if they are happening frequently.", + "PublicDescription": "This event counts the number of memory ordering machine clears detected. Memory ordering machine clears can result from memory address aliasing or snoops from another hardware thread or core to data inflight in the pipeline. Machine clears can have a significant performance impact if they are happening frequently.", "SampleAfterValue": "100003", "UMask": "0x2" },
tools/perf/pmu-events/arch/x86/haswellx/memory.json +1 -1
@@ -62,7 +62,7 @@ "BriefDescription": "Counts the number of machine clears due to memory order conflicts.", "EventCode": "0xC3", "EventName": "MACHINE_CLEARS.MEMORY_ORDERING", - "PublicDescription": "This event counts the number of memory ordering machine clears detected. Memory ordering machine clears can result from memory address aliasing or snoops from another hardware thread or core to data in-flight in the pipeline. Machine clears can have a significant performance impact if they are happening frequently.", + "PublicDescription": "This event counts the number of memory ordering machine clears detected. Memory ordering machine clears can result from memory address aliasing or snoops from another hardware thread or core to data inflight in the pipeline. Machine clears can have a significant performance impact if they are happening frequently.", "SampleAfterValue": "100003", "UMask": "0x2" },
Signature · 1 line
2.42.0.758.gaed0368e0e-goog
Ian Rogers
reviewerPATCH 4/9 perf vendor events intel: Add typo fix for ivybridge FP
In reply to Ian RogersAdd a missed space.
--- tools/perf/pmu-events/arch/x86/ivybridge/floating-point.json | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
tools/perf/pmu-events/arch/x86/ivybridge/floating-point.json +1 -1
@@ -73,7 +73,7 @@ "UMask": "0x20" }, { - "BriefDescription": "Number of FP Computational Uops Executed this cycle. The number of FADD, FSUB, FCOM, FMULs, integer MULsand IMULs, FDIVs, FPREMs, FSQRTS, integer DIVs, and IDIVs. This event does not distinguish an FADD used in the middle of a transcendental flow from a s", + "BriefDescription": "Number of FP Computational Uops Executed this cycle. The number of FADD, FSUB, FCOM, FMULs, integer MULs and IMULs, FDIVs, FPREMs, FSQRTS, integer DIVs, and IDIVs. This event does not distinguish an FADD used in the middle of a transcendental flow from a s", "EventCode": "0x10", "EventName": "FP_COMP_OPS_EXE.X87", "PublicDescription": "Counts number of X87 uops executed.",
Signature · 1 line
2.42.0.758.gaed0368e0e-goog
Ian Rogers
reviewerPATCH 5/9 perf vendor events intel: Update knightslanding events to v16
In reply to Ian RogersUpdate knightslanding from v10 to v16 adding the changes from:
https://github.com/intel/perfmon/commit/6c1f169f6ed63ee1fd75ebb303d0fd06d71196f5
https://github.com/intel/perfmon/commit/b22ca587ec8b5ac20471ea2f14924f63e63afe9d
https://github.com/intel/perfmon/commit/e685286f083ee81cb7dafd0cd8546c79ee433187
--- .../arch/x86/knightslanding/cache.json | 39 ++++++++----- .../x86/knightslanding/floating-point.json | 8 +-- .../arch/x86/knightslanding/pipeline.json | 55 +++++++++++-------- .../arch/x86/knightslanding/uncore-cache.json | 26 ++++----- .../x86/knightslanding/virtual-memory.json | 2 +- tools/perf/pmu-events/arch/x86/mapfile.csv | 2 +- 6 files changed, 75 insertions(+), 57 deletions(-)
tools/perf/pmu-events/arch/x86/knightslanding/cache.json +24 -15
@@ -6,13 +6,19 @@ "SampleAfterValue": "200003" }, { - "BriefDescription": "Counts the number of core cycles the fetch stalls because of an icache miss. This is a cumulative count of core cycles the fetch stalled for all icache misses.", + "BriefDescription": "This event counts the number of core cycles the fetch stalls because of an icache miss. This is a cumulative count of cycles the NIP stalled for all icache misses.", "EventCode": "0x86", "EventName": "FETCH_STALL.ICACHE_FILL_PENDING_CYCLES", - "PublicDescription": "This event counts the number of core cycles the fetch stalls because of an icache miss. This is a cumulative count of cycles the NIP stalled for all icache misses.", "SampleAfterValue": "200003", "UMask": "0x4" }, + { + "BriefDescription": "Counts the number of L2HWP allocated into XQ GP", + "EventCode": "0x3E", + "EventName": "L2_PREFETCHER.ALLOC_XQ", + "SampleAfterValue": "100007", + "UMask": "0x4" + }, { "BriefDescription": "Counts the number of L2 cache misses", "EventCode": "0x2E",
@@ -28,7 +34,7 @@ "UMask": "0x4f" }, { - "BriefDescription": "Counts the number of MEC requests from the L2Q that reference a cache line (cacheable requests) excluding SW prefetches filling only to L2 cache and L1 evictions (automatically exlcudes L2HWP, UC, WC) that were rejected - Multiple repeated rejects should be counted multiple times", + "BriefDescription": "Counts the number of MEC requests from the L2Q that reference a cache line (cacheable requests) excluding SW prefetches filling only to L2 cache and L1 evictions (automatically excludes L2HWP, UC, WC) that were rejected - Multiple repeated rejects should be counted multiple times", "EventCode": "0x30", "EventName": "L2_REQUESTS_REJECT.ALL", "SampleAfterValue": "200003"
@@ -50,11 +56,12 @@ "UMask": "0x80" }, { - "BriefDescription": "Counts the loads retired that get the data from the other core in the same tile in M state", + "BriefDescription": "Counts the loads retired that get the data from the other core in the same tile in M state (Precise Event)", "Data_LA": "1", "EventCode": "0x04", "EventName": "MEM_UOPS_RETIRED.HITM", "PEBS": "1", + "PublicDescription": "This event counts the number of load micro-ops retired that got data from another core's cache. (Precise Event).", "SampleAfterValue": "200003", "UMask": "0x20" },
@@ -67,20 +74,22 @@ "UMask": "0x1" }, { - "BriefDescription": "Counts the number of load micro-ops retired that hit in the L2", + "BriefDescription": "Counts the number of load micro-ops retired that hit in the L2 (Precise Event)", "Data_LA": "1", "EventCode": "0x04", "EventName": "MEM_UOPS_RETIRED.L2_HIT_LOADS", "PEBS": "1", + "PublicDescription": "This event counts the number of load micro-uops retired that hit in the L2 (Precise Event)", "SampleAfterValue": "200003", "UMask": "0x2" }, { - "BriefDescription": "Counts the number of load micro-ops retired that miss in the L2", + "BriefDescription": "Counts the number of load micro-ops retired that miss in the L2 (Precise Event)", "Data_LA": "1", "EventCode": "0x04", "EventName": "MEM_UOPS_RETIRED.L2_MISS_LOADS", "PEBS": "1", + "PublicDescription": "This event counts the number of load micro-ops retired that miss in the L2 (Precise Event)", "SampleAfterValue": "100007", "UMask": "0x4" },
@@ -620,6 +629,15 @@ "SampleAfterValue": "100007", "UMask": "0x1" }, + { + "BriefDescription": "Accounts for responses which miss its own tile's L2.", + "EventCode": "0xB7", + "EventName": "OFFCORE_RESPONSE.ANY_REQUEST.L2_MISS", + "MSRIndex": "0x1a6,0x1a7", + "MSRValue": "0x18001981F8", + "SampleAfterValue": "100007", + "UMask": "0x1" + }, { "BriefDescription": "Counts any request that are outstanding, per weighted cycle, from the time of the request to when any response is received. The outstanding response should be programmed only on PMC0.", "EventCode": "0xB7",
@@ -1664,15 +1682,6 @@ "SampleAfterValue": "100007", "UMask": "0x1" }, - { - "BriefDescription": "Counts L2 data RFO prefetches (includes PREFETCHW instruction) that provides no supplier details", - "EventCode": "0xB7", - "EventName": "OFFCORE_RESPONSE.PF_L2_RFO.SUPPLIER_NONE", - "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x0000020020", - "SampleAfterValue": "100007", - "UMask": "0x1" - }, { "BriefDescription": "Counts Software Prefetches that accounts for any response", "EventCode": "0xB7",
tools/perf/pmu-events/arch/x86/knightslanding/floating-point.json +4 -4
@@ -8,18 +8,18 @@ "UMask": "0x4" }, { - "BriefDescription": "Counts the number of vector SSE, AVX, AVX2, AVX-512 micro-ops retired. More specifically, it counts packed SSE, AVX, AVX2, AVX-512 micro-ops (both floating point and integer) except for loads (memory-to-register mov-type micro-ops), packed byte and word multiplies.", + "BriefDescription": "Counts the number of packed SSE, AVX, AVX2, AVX-512 micro-ops (both floating point and integer) except for loads (memory-to-register mov-type micro-ops), packed byte and word multiplies.", "EventCode": "0xC2", "EventName": "UOPS_RETIRED.PACKED_SIMD", - "PublicDescription": "This event counts the number of packed vector SSE, AVX, AVX2, and AVX-512 micro-ops retired (floating point, integer and store) except for loads (memory-to-register mov-type micro-ops), packed byte and word multiplies.", + "PublicDescription": "The length of the packed operation (128bits, 256bits or 512bits) is not taken into account when updating the counter; all count the same (+1). \r\nMask (k) registers are ignored. For example: a micro-op operating with a mask that only enables one element or even zero elements will still trigger this counter (+1)\r\nThis event is defined at the micro-op level and not instruction level. Most instructions are implemented with one micro-op but not all.", "SampleAfterValue": "200003", "UMask": "0x40" }, { - "BriefDescription": "Counts the number of scalar SSE, AVX, AVX2, AVX-512 micro-ops retired. More specifically, it counts scalar SSE, AVX, AVX2, AVX-512 micro-ops except for loads (memory-to-register mov-type micro ops), division, sqrt.", + "BriefDescription": "Counts the number of scalar SSE, AVX, AVX2, AVX-512 micro-ops except for loads (memory-to-register mov-type micro ops), division, sqrt.", "EventCode": "0xC2", "EventName": "UOPS_RETIRED.SCALAR_SIMD", - "PublicDescription": "This event counts the number of scalar SSE, AVX, AVX2, AVX-512 micro-ops retired (floating point, integer and store) except for loads (memory-to-register mov-type micro ops), division, sqrt.", + "PublicDescription": "This event is defined at the micro-op level and not instruction level. Most instructions are implemented with one micro-op but not all.", "SampleAfterValue": "200003", "UMask": "0x20" }
tools/perf/pmu-events/arch/x86/knightslanding/pipeline.json +32 -23
198 lines not on this page · raw message
tools/perf/pmu-events/arch/x86/knightslanding/uncore-cache.json +13 -13
@@ -2558,7 +2558,7 @@ "Unit": "CHA" }, { - "BriefDescription": "REQUESTQ'' includes: IRQ, PRQ, IPQ, RRQ, WBQ (everything except for ISMQ)", + "BriefDescription": "REQUESTQ includes: IRQ, PRQ, IPQ, RRQ, WBQ (everything except for ISMQ)", "EventCode": "0x2A", "EventName": "UNC_H_INGRESS_RETRY_REQ_Q0_RETRY.AD_REQ_VN0", "PerPkg": "1",
@@ -2566,7 +2566,7 @@ "Unit": "CHA" }, { - "BriefDescription": "REQUESTQ'' includes: IRQ, PRQ, IPQ, RRQ, WBQ (everything except for ISMQ)", + "BriefDescription": "REQUESTQ includes: IRQ, PRQ, IPQ, RRQ, WBQ (everything except for ISMQ)", "EventCode": "0x2A", "EventName": "UNC_H_INGRESS_RETRY_REQ_Q0_RETRY.AD_RSP_VN0", "PerPkg": "1",
@@ -2574,7 +2574,7 @@ "Unit": "CHA" }, { - "BriefDescription": "REQUESTQ'' includes: IRQ, PRQ, IPQ, RRQ, WBQ (everything except for ISMQ)", + "BriefDescription": "REQUESTQ includes: IRQ, PRQ, IPQ, RRQ, WBQ (everything except for ISMQ)", "EventCode": "0x2A", "EventName": "UNC_H_INGRESS_RETRY_REQ_Q0_RETRY.AK_NON_UPI", "PerPkg": "1",
@@ -2582,7 +2582,7 @@ "Unit": "CHA" }, { - "BriefDescription": "REQUESTQ'' includes: IRQ, PRQ, IPQ, RRQ, WBQ (everything except for ISMQ)", + "BriefDescription": "REQUESTQ includes: IRQ, PRQ, IPQ, RRQ, WBQ (everything except for ISMQ)", "EventCode": "0x2A", "EventName": "UNC_H_INGRESS_RETRY_REQ_Q0_RETRY.BL_NCB_VN0", "PerPkg": "1",
@@ -2590,7 +2590,7 @@ "Unit": "CHA" }, { - "BriefDescription": "REQUESTQ'' includes: IRQ, PRQ, IPQ, RRQ, WBQ (everything except for ISMQ)", + "BriefDescription": "REQUESTQ includes: IRQ, PRQ, IPQ, RRQ, WBQ (everything except for ISMQ)", "EventCode": "0x2A", "EventName": "UNC_H_INGRESS_RETRY_REQ_Q0_RETRY.BL_NCS_VN0", "PerPkg": "1",
@@ -2598,7 +2598,7 @@ "Unit": "CHA" }, { - "BriefDescription": "REQUESTQ'' includes: IRQ, PRQ, IPQ, RRQ, WBQ (everything except for ISMQ)", + "BriefDescription": "REQUESTQ includes: IRQ, PRQ, IPQ, RRQ, WBQ (everything except for ISMQ)", "EventCode": "0x2A", "EventName": "UNC_H_INGRESS_RETRY_REQ_Q0_RETRY.BL_RSP_VN0", "PerPkg": "1",
@@ -2606,7 +2606,7 @@ "Unit": "CHA" }, { - "BriefDescription": "REQUESTQ'' includes: IRQ, PRQ, IPQ, RRQ, WBQ (everything except for ISMQ)", + "BriefDescription": "REQUESTQ includes: IRQ, PRQ, IPQ, RRQ, WBQ (everything except for ISMQ)", "EventCode": "0x2A", "EventName": "UNC_H_INGRESS_RETRY_REQ_Q0_RETRY.BL_WB_VN0", "PerPkg": "1",
@@ -2614,7 +2614,7 @@ "Unit": "CHA" }, { - "BriefDescription": "REQUESTQ'' includes: IRQ, PRQ, IPQ, RRQ, WBQ (everything except for ISMQ)", + "BriefDescription": "REQUESTQ includes: IRQ, PRQ, IPQ, RRQ, WBQ (everything except for ISMQ)", "EventCode": "0x2A", "EventName": "UNC_H_INGRESS_RETRY_REQ_Q0_RETRY.IV_NON_UPI", "PerPkg": "1",
@@ -2622,7 +2622,7 @@ "Unit": "CHA" }, { - "BriefDescription": "REQUESTQ'' includes: IRQ, PRQ, IPQ, RRQ, WBQ (everything except for ISMQ)", + "BriefDescription": "REQUESTQ includes: IRQ, PRQ, IPQ, RRQ, WBQ (everything except for ISMQ)", "EventCode": "0x2B", "EventName": "UNC_H_INGRESS_RETRY_REQ_Q1_RETRY.ALLOW_SNP", "PerPkg": "1",
@@ -2630,7 +2630,7 @@ "Unit": "CHA" }, { - "BriefDescription": "REQUESTQ'' includes: IRQ, PRQ, IPQ, RRQ, WBQ (everything except for ISMQ)", + "BriefDescription": "REQUESTQ includes: IRQ, PRQ, IPQ, RRQ, WBQ (everything except for ISMQ)", "EventCode": "0x2B", "EventName": "UNC_H_INGRESS_RETRY_REQ_Q1_RETRY.ANY_REJECT_IRQ0", "PerPkg": "1",
@@ -2638,7 +2638,7 @@ "Unit": "CHA" }, { - "BriefDescription": "REQUESTQ'' includes: IRQ, PRQ, IPQ, RRQ, WBQ (everything except for ISMQ)", + "BriefDescription": "REQUESTQ includes: IRQ, PRQ, IPQ, RRQ, WBQ (everything except for ISMQ)", "EventCode": "0x2B", "EventName": "UNC_H_INGRESS_RETRY_REQ_Q1_RETRY.PA_MATCH", "PerPkg": "1",
@@ -2646,7 +2646,7 @@ "Unit": "CHA" }, { - "BriefDescription": "REQUESTQ'' includes: IRQ, PRQ, IPQ, RRQ, WBQ (everything except for ISMQ)", + "BriefDescription": "REQUESTQ includes: IRQ, PRQ, IPQ, RRQ, WBQ (everything except for ISMQ)", "EventCode": "0x2B", "EventName": "UNC_H_INGRESS_RETRY_REQ_Q1_RETRY.SF_VICTIM", "PerPkg": "1",
@@ -2654,7 +2654,7 @@ "Unit": "CHA" }, { - "BriefDescription": "REQUESTQ'' includes: IRQ, PRQ, IPQ, RRQ, WBQ (everything except for ISMQ)", + "BriefDescription": "REQUESTQ includes: IRQ, PRQ, IPQ, RRQ, WBQ (everything except for ISMQ)", "EventCode": "0x2B", "EventName": "UNC_H_INGRESS_RETRY_REQ_Q1_RETRY.SF_WAY", "PerPkg": "1",
tools/perf/pmu-events/arch/x86/knightslanding/virtual-memory.json +1 -1
@@ -1,6 +1,6 @@ [ { - "BriefDescription": "Counts the number of load micro-ops retired that cause a DTLB miss", + "BriefDescription": "Counts the number of load micro-ops retired that cause a DTLB miss (Precise Event)", "Data_LA": "1", "EventCode": "0x04", "EventName": "MEM_UOPS_RETIRED.DTLB_MISS_LOADS",
tools/perf/pmu-events/arch/x86/mapfile.csv +1 -1
@@ -19,7 +19,7 @@ GenuineIntel-6-6[AC],v1.21,icelakex,core GenuineIntel-6-3A,v24,ivybridge,core GenuineIntel-6-3E,v24,ivytown,core GenuineIntel-6-2D,v24,jaketown,core -GenuineIntel-6-(57|85),v10,knightslanding,core +GenuineIntel-6-(57|85),v16,knightslanding,core GenuineIntel-6-BD,v1.00,lunarlake,core GenuineIntel-6-A[AC],v1.04,meteorlake,core GenuineIntel-6-1[AEF],v4,nehalemep,core
Signature · 1 line
2.42.0.758.gaed0368e0e-goog
Ian Rogers
reviewerPATCH 6/9 perf vendor events intel: Update meteorlake events to v1.06
In reply to Ian RogersUpdate meteorlake from v1.04 to v1.06 adding the changes from:
https://github.com/intel/perfmon/commit/bc84df043091ec7c98c0629f3d074d9d7a108194
https://github.com/intel/perfmon/commit/405d3ee987d756b5b5d9a64d8a8fa77559822ecf
--- tools/perf/pmu-events/arch/x86/mapfile.csv | 2 +- .../pmu-events/arch/x86/meteorlake/cache.json | 30 ++++++++ .../arch/x86/meteorlake/frontend.json | 29 ++++++-- .../arch/x86/meteorlake/memory.json | 37 ++++++++++ .../pmu-events/arch/x86/meteorlake/other.json | 40 +++++++++++ .../arch/x86/meteorlake/pipeline.json | 68 ++++++++++++++++++- .../arch/x86/meteorlake/uncore-other.json | 9 +++ 7 files changed, 209 insertions(+), 6 deletions(-) create mode 100644 tools/perf/pmu-events/arch/x86/meteorlake/uncore-other.json
tools/perf/pmu-events/arch/x86/mapfile.csv +1 -1
@@ -21,7 +21,7 @@ GenuineIntel-6-3E,v24,ivytown,core GenuineIntel-6-2D,v24,jaketown,core GenuineIntel-6-(57|85),v16,knightslanding,core GenuineIntel-6-BD,v1.00,lunarlake,core -GenuineIntel-6-A[AC],v1.04,meteorlake,core +GenuineIntel-6-A[AC],v1.06,meteorlake,core GenuineIntel-6-1[AEF],v4,nehalemep,core GenuineIntel-6-2E,v4,nehalemex,core GenuineIntel-6-A7,v1.01,rocketlake,core
tools/perf/pmu-events/arch/x86/meteorlake/cache.json +30 -0
@@ -966,6 +966,16 @@ "UMask": "0x3", "Unit": "cpu_core" }, + { + "BriefDescription": "Counts demand data reads that were supplied by the L3 cache where a snoop was sent, the snoop hit, and modified data was forwarded.", + "EventCode": "0xB7", + "EventName": "OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HITM", + "MSRIndex": "0x1a6,0x1a7", + "MSRValue": "0x10003C0001", + "SampleAfterValue": "100003", + "UMask": "0x1", + "Unit": "cpu_atom" + }, { "BriefDescription": "Counts demand data reads that resulted in a snoop hit in another cores caches, data forwarding is required as the data is modified.", "EventCode": "0x2A,0x2B",
@@ -976,6 +986,16 @@ "UMask": "0x1", "Unit": "cpu_core" }, + { + "BriefDescription": "Counts demand data reads that were supplied by the L3 cache where a snoop was sent, the snoop hit, and non-modified data was forwarded.", + "EventCode": "0xB7", + "EventName": "OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HIT_WITH_FWD", + "MSRIndex": "0x1a6,0x1a7", + "MSRValue": "0x8003C0001", + "SampleAfterValue": "100003", + "UMask": "0x1", + "Unit": "cpu_atom" + }, { "BriefDescription": "Counts demand data reads that resulted in a snoop hit in another cores caches which forwarded the unmodified data to the requesting core.", "EventCode": "0x2A,0x2B",
@@ -986,6 +1006,16 @@ "UMask": "0x1", "Unit": "cpu_core" }, + { + "BriefDescription": "Counts demand reads for ownership (RFO) and software prefetches for exclusive ownership (PREFETCHW) that were supplied by the L3 cache where a snoop was sent, the snoop hit, and modified data was forwarded.", + "EventCode": "0xB7", + "EventName": "OCR.DEMAND_RFO.L3_HIT.SNOOP_HITM", + "MSRIndex": "0x1a6,0x1a7", + "MSRValue": "0x10003C0002", + "SampleAfterValue": "100003", + "UMask": "0x1", + "Unit": "cpu_atom" + }, { "BriefDescription": "Counts demand read for ownership (RFO) requests and software prefetches for exclusive ownership (PREFETCHW) that resulted in a snoop hit in another cores caches, data forwarding is required as the data is modified.", "EventCode": "0x2A,0x2B",
tools/perf/pmu-events/arch/x86/meteorlake/frontend.json +25 -4
@@ -460,6 +460,27 @@ "UMask": "0x1", "Unit": "cpu_core" }, + { + "BriefDescription": "Cycles when no uops are not delivered by the IDQ when backend of the machine is not stalled [This event is alias to IDQ_UOPS_NOT_DELIVERED.CYCLES_0_UOPS_DELIV.CORE]", + "CounterMask": "6", + "EventCode": "0x9c", + "EventName": "IDQ_BUBBLES.CYCLES_0_UOPS_DELIV.CORE", + "PublicDescription": "Counts the number of cycles when no uops were delivered by the Instruction Decode Queue (IDQ) to the back-end of the pipeline when there was no back-end stalls. This event counts for one SMT thread in a given cycle. [This event is alias to IDQ_UOPS_NOT_DELIVERED.CYCLES_0_UOPS_DELIV.CORE]", + "SampleAfterValue": "1000003", + "UMask": "0x1", + "Unit": "cpu_core" + }, + { + "BriefDescription": "Cycles when optimal number of uops was delivered to the back-end when the back-end is not stalled [This event is alias to IDQ_UOPS_NOT_DELIVERED.CYCLES_FE_WAS_OK]", + "CounterMask": "1", + "EventCode": "0x9c", + "EventName": "IDQ_BUBBLES.CYCLES_FE_WAS_OK", + "Invert": "1", + "PublicDescription": "Counts the number of cycles when the optimal number of uops were delivered by the Instruction Decode Queue (IDQ) to the back-end of the pipeline when there was no back-end stalls. This event counts for one SMT thread in a given cycle. [This event is alias to IDQ_UOPS_NOT_DELIVERED.CYCLES_FE_WAS_OK]", + "SampleAfterValue": "1000003", + "UMask": "0x1", + "Unit": "cpu_core" + }, { "BriefDescription": "Uops not delivered by IDQ when backend of the machine is not stalled", "EventCode": "0x9c",
@@ -470,22 +491,22 @@ "Unit": "cpu_core" }, { - "BriefDescription": "Cycles when no uops are not delivered by the IDQ when backend of the machine is not stalled", + "BriefDescription": "Cycles when no uops are not delivered by the IDQ when backend of the machine is not stalled [This event is alias to IDQ_BUBBLES.CYCLES_0_UOPS_DELIV.CORE]", "CounterMask": "6", "EventCode": "0x9c", "EventName": "IDQ_UOPS_NOT_DELIVERED.CYCLES_0_UOPS_DELIV.CORE", - "PublicDescription": "Counts the number of cycles when no uops were delivered by the Instruction Decode Queue (IDQ) to the back-end of the pipeline when there was no back-end stalls. This event counts for one SMT thread in a given cycle.", + "PublicDescription": "Counts the number of cycles when no uops were delivered by the Instruction Decode Queue (IDQ) to the back-end of the pipeline when there was no back-end stalls. This event counts for one SMT thread in a given cycle. [This event is alias to IDQ_BUBBLES.CYCLES_0_UOPS_DELIV.CORE]", "SampleAfterValue": "1000003", "UMask": "0x1", "Unit": "cpu_core" }, { - "BriefDescription": "Cycles when optimal number of uops was delivered to the back-end when the back-end is not stalled", + "BriefDescription": "Cycles when optimal number of uops was delivered to the back-end when the back-end is not stalled [This event is alias to IDQ_BUBBLES.CYCLES_FE_WAS_OK]", "CounterMask": "1", "EventCode": "0x9c", "EventName": "IDQ_UOPS_NOT_DELIVERED.CYCLES_FE_WAS_OK", "Invert": "1", - "PublicDescription": "Counts the number of cycles when the optimal number of uops were delivered by the Instruction Decode Queue (IDQ) to the back-end of the pipeline when there was no back-end stalls. This event counts for one SMT thread in a given cycle.", + "PublicDescription": "Counts the number of cycles when the optimal number of uops were delivered by the Instruction Decode Queue (IDQ) to the back-end of the pipeline when there was no back-end stalls. This event counts for one SMT thread in a given cycle. [This event is alias to IDQ_BUBBLES.CYCLES_FE_WAS_OK]", "SampleAfterValue": "1000003", "UMask": "0x1", "Unit": "cpu_core"
tools/perf/pmu-events/arch/x86/meteorlake/memory.json +37 -0
@@ -66,6 +66,14 @@ "UMask": "0x84", "Unit": "cpu_atom" }, + { + "BriefDescription": "Counts the number of machine clears due to memory ordering caused by a snoop from an external agent. Does not count internally generated machine clears such as those due to memory disambiguation.", + "EventCode": "0xc3", + "EventName": "MACHINE_CLEARS.MEMORY_ORDERING", + "SampleAfterValue": "20003", + "UMask": "0x2", + "Unit": "cpu_atom" + }, { "BriefDescription": "Number of machine clears due to memory ordering conflicts.", "EventCode": "0xc3",
@@ -75,6 +83,15 @@ "UMask": "0x2", "Unit": "cpu_core" }, + { + "BriefDescription": "Cycles while L1 cache miss demand load is outstanding.", + "CounterMask": "2", + "EventCode": "0x47", + "EventName": "MEMORY_ACTIVITY.CYCLES_L1D_MISS", + "SampleAfterValue": "1000003", + "UMask": "0x2", + "Unit": "cpu_core" + }, { "BriefDescription": "Execution stalls while L1 cache miss demand load is outstanding.", "CounterMask": "3",
@@ -279,6 +296,26 @@ "UMask": "0x4", "Unit": "cpu_atom" }, + { + "BriefDescription": "Counts demand data reads that were not supplied by the L3 cache.", + "EventCode": "0x2A,0x2B", + "EventName": "OCR.DEMAND_DATA_RD.L3_MISS", + "MSRIndex": "0x1a6,0x1a7", + "MSRValue": "0x3FBFC00001", + "SampleAfterValue": "100003", + "UMask": "0x1", + "Unit": "cpu_core" + }, + { + "BriefDescription": "Counts demand read for ownership (RFO) requests and software prefetches for exclusive ownership (PREFETCHW) that were not supplied by the L3 cache.", + "EventCode": "0x2A,0x2B", + "EventName": "OCR.DEMAND_RFO.L3_MISS", + "MSRIndex": "0x1a6,0x1a7", + "MSRValue": "0x3FBFC00002", + "SampleAfterValue": "100003", + "UMask": "0x1", + "Unit": "cpu_core" + }, { "BriefDescription": "Counts demand data read requests that miss the L3 cache.", "EventCode": "0x21",
tools/perf/pmu-events/arch/x86/meteorlake/other.json +40 -0
@@ -7,6 +7,46 @@ "UMask": "0x8", "Unit": "cpu_core" }, + { + "BriefDescription": "Counts demand data reads that have any type of response.", + "EventCode": "0x2A,0x2B", + "EventName": "OCR.DEMAND_DATA_RD.ANY_RESPONSE", + "MSRIndex": "0x1a6,0x1a7", + "MSRValue": "0x10001", + "SampleAfterValue": "100003", + "UMask": "0x1", + "Unit": "cpu_core" + }, + { + "BriefDescription": "Counts demand data reads that were supplied by DRAM.", + "EventCode": "0x2A,0x2B", + "EventName": "OCR.DEMAND_DATA_RD.DRAM", + "MSRIndex": "0x1a6,0x1a7", + "MSRValue": "0x184000001", + "SampleAfterValue": "100003", + "UMask": "0x1", + "Unit": "cpu_core" + }, + { + "BriefDescription": "Counts demand read for ownership (RFO) requests and software prefetches for exclusive ownership (PREFETCHW) that have any type of response.", + "EventCode": "0x2A,0x2B", + "EventName": "OCR.DEMAND_RFO.ANY_RESPONSE", + "MSRIndex": "0x1a6,0x1a7", + "MSRValue": "0x10002", + "SampleAfterValue": "100003", + "UMask": "0x1", + "Unit": "cpu_core" + }, + { + "BriefDescription": "Counts streaming stores that have any type of response.", + "EventCode": "0xB7", + "EventName": "OCR.STREAMING_WR.ANY_RESPONSE", + "MSRIndex": "0x1a6,0x1a7", + "MSRValue": "0x10800", + "SampleAfterValue": "100003", + "UMask": "0x1", + "Unit": "cpu_atom" + }, { "BriefDescription": "Counts streaming stores that have any type of response.", "EventCode": "0x2A,0x2B",
tools/perf/pmu-events/arch/x86/meteorlake/pipeline.json +67 -1
116 lines not on this page · raw message
tools/perf/pmu-events/arch/x86/meteorlake/uncore-other.json new file +9 -0
@@ -0,0 +1,9 @@ +[ + { + "BriefDescription": "This 48-bit fixed counter counts the UCLK cycles.", + "EventCode": "0xff", + "EventName": "UNC_CLOCK.SOCKET", + "PerPkg": "1", + "Unit": "CLOCK" + } +]
Signature · 1 line
2.42.0.758.gaed0368e0e-goog
Ian Rogers
reviewerPATCH 7/9 perf vendor events intel: Update westmereex events to v4
In reply to Ian RogersUpdate westmereex events from v3 to v4 fixing a spelling issue.
--- tools/perf/pmu-events/arch/x86/mapfile.csv | 2 +- tools/perf/pmu-events/arch/x86/westmereex/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-6-86,v1.21,snowridgex,core GenuineIntel-6-8[CD],v1.13,tigerlake,core GenuineIntel-6-2C,v5,westmereep-dp,core GenuineIntel-6-25,v4,westmereep-sp,core -GenuineIntel-6-2F,v3,westmereex,core +GenuineIntel-6-2F,v4,westmereex,core AuthenticAMD-23-([12][0-9A-F]|[0-9A-F]),v2,amdzen1,core AuthenticAMD-23-[[:xdigit:]]+,v1,amdzen2,core AuthenticAMD-25-([245][[:xdigit:]]|[[:xdigit:]]),v1,amdzen3,core
tools/perf/pmu-events/arch/x86/westmereex/pipeline.json +1 -1
@@ -45,7 +45,7 @@ "UMask": "0x1" }, { - "BriefDescription": "Early Branch Prediciton Unit clears", + "BriefDescription": "Early Branch Prediction Unit clears", "EventCode": "0xE8", "EventName": "BPU_CLEARS.EARLY", "SampleAfterValue": "2000000",
Signature · 1 line
2.42.0.758.gaed0368e0e-goog
Ian Rogers
reviewerPATCH 8/9 perf vendor events intel: Update bonnell version number to v5
In reply to Ian RogersSpelling fixes were already incorporated in the Linux perf tree, update the version number to reflect this.
--- tools/perf/pmu-events/arch/x86/mapfile.csv | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
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),v1.23,alderlake,core GenuineIntel-6-BE,v1.23,alderlaken,core -GenuineIntel-6-(1C|26|27|35|36),v4,bonnell,core +GenuineIntel-6-(1C|26|27|35|36),v5,bonnell,core GenuineIntel-6-(3D|47),v28,broadwell,core GenuineIntel-6-56,v11,broadwellde,core GenuineIntel-6-4F,v22,broadwellx,core
Signature · 1 line
2.42.0.758.gaed0368e0e-goog
Ian Rogers
reviewerPATCH 9/9 perf vendor events intel: Update tsx_cycles_per_elision metrics
In reply to Ian RogersUpdate tsx_cycles_per_elision as per:
https://github.com/intel/perfmon/pull/116
Prefer the el-start event rather than cycles-t for detecting whether the metric will work as HLE may be disabled. Remove the metric from sapphirerapids that has no el-start event.
--- .../perf/pmu-events/arch/x86/cascadelakex/clx-metrics.json | 2 +- tools/perf/pmu-events/arch/x86/icelake/icl-metrics.json | 2 +- tools/perf/pmu-events/arch/x86/icelakex/icx-metrics.json | 2 +- tools/perf/pmu-events/arch/x86/rocketlake/rkl-metrics.json | 2 +- .../pmu-events/arch/x86/sapphirerapids/spr-metrics.json | 7 ------- tools/perf/pmu-events/arch/x86/skylake/skl-metrics.json | 2 +- tools/perf/pmu-events/arch/x86/skylakex/skx-metrics.json | 2 +- tools/perf/pmu-events/arch/x86/tigerlake/tgl-metrics.json | 2 +- 8 files changed, 7 insertions(+), 14 deletions(-)
tools/perf/pmu-events/arch/x86/cascadelakex/clx-metrics.json +1 -1
@@ -1837,7 +1837,7 @@ }, { "BriefDescription": "Number of cycles within a transaction divided by the number of elisions.", - "MetricExpr": "(cycles\\-t / el\\-start if has_event(cycles\\-t) else 0)", + "MetricExpr": "(cycles\\-t / el\\-start if has_event(el\\-start) else 0)", "MetricGroup": "transaction", "MetricName": "tsx_cycles_per_elision", "ScaleUnit": "1cycles / elision"
tools/perf/pmu-events/arch/x86/icelake/icl-metrics.json +1 -1
@@ -1525,7 +1525,7 @@ }, { "BriefDescription": "Number of cycles within a transaction divided by the number of elisions.", - "MetricExpr": "(cycles\\-t / el\\-start if has_event(cycles\\-t) else 0)", + "MetricExpr": "(cycles\\-t / el\\-start if has_event(el\\-start) else 0)", "MetricGroup": "transaction", "MetricName": "tsx_cycles_per_elision", "ScaleUnit": "1cycles / elision"
tools/perf/pmu-events/arch/x86/icelakex/icx-metrics.json +1 -1
@@ -1821,7 +1821,7 @@ }, { "BriefDescription": "Number of cycles within a transaction divided by the number of elisions.", - "MetricExpr": "(cycles\\-t / el\\-start if has_event(cycles\\-t) else 0)", + "MetricExpr": "(cycles\\-t / el\\-start if has_event(el\\-start) else 0)", "MetricGroup": "transaction", "MetricName": "tsx_cycles_per_elision", "ScaleUnit": "1cycles / elision"
tools/perf/pmu-events/arch/x86/rocketlake/rkl-metrics.json +1 -1
@@ -1551,7 +1551,7 @@ }, { "BriefDescription": "Number of cycles within a transaction divided by the number of elisions.", - "MetricExpr": "(cycles\\-t / el\\-start if has_event(cycles\\-t) else 0)", + "MetricExpr": "(cycles\\-t / el\\-start if has_event(el\\-start) else 0)", "MetricGroup": "transaction", "MetricName": "tsx_cycles_per_elision", "ScaleUnit": "1cycles / elision"
tools/perf/pmu-events/arch/x86/sapphirerapids/spr-metrics.json +0 -7
@@ -1944,13 +1944,6 @@ "MetricName": "tsx_aborted_cycles", "ScaleUnit": "100%" }, - { - "BriefDescription": "Number of cycles within a transaction divided by the number of elisions.", - "MetricExpr": "(cycles\\-t / el\\-start if has_event(cycles\\-t) else 0)", - "MetricGroup": "transaction", - "MetricName": "tsx_cycles_per_elision", - "ScaleUnit": "1cycles / elision" - }, { "BriefDescription": "Number of cycles within a transaction divided by the number of transactions.", "MetricExpr": "(cycles\\-t / tx\\-start if has_event(cycles\\-t) else 0)",
tools/perf/pmu-events/arch/x86/skylake/skl-metrics.json +1 -1
@@ -1473,7 +1473,7 @@ }, { "BriefDescription": "Number of cycles within a transaction divided by the number of elisions.", - "MetricExpr": "(cycles\\-t / el\\-start if has_event(cycles\\-t) else 0)", + "MetricExpr": "(cycles\\-t / el\\-start if has_event(el\\-start) else 0)", "MetricGroup": "transaction", "MetricName": "tsx_cycles_per_elision", "ScaleUnit": "1cycles / elision"
tools/perf/pmu-events/arch/x86/skylakex/skx-metrics.json +1 -1
@@ -1781,7 +1781,7 @@ }, { "BriefDescription": "Number of cycles within a transaction divided by the number of elisions.", - "MetricExpr": "(cycles\\-t / el\\-start if has_event(cycles\\-t) else 0)", + "MetricExpr": "(cycles\\-t / el\\-start if has_event(el\\-start) else 0)", "MetricGroup": "transaction", "MetricName": "tsx_cycles_per_elision", "ScaleUnit": "1cycles / elision"
tools/perf/pmu-events/arch/x86/tigerlake/tgl-metrics.json +1 -1
@@ -1539,7 +1539,7 @@ }, { "BriefDescription": "Number of cycles within a transaction divided by the number of elisions.", - "MetricExpr": "(cycles\\-t / el\\-start if has_event(cycles\\-t) else 0)", + "MetricExpr": "(cycles\\-t / el\\-start if has_event(el\\-start) else 0)", "MetricGroup": "transaction", "MetricName": "tsx_cycles_per_elision", "ScaleUnit": "1cycles / elision"
Signature · 1 line
2.42.0.758.gaed0368e0e-goog
Quoted 9 lines · Ian Rogers
Update alderlake and alderlaken events from v1.21 to v1.23 adding the changes from:
https://github.com/intel/perfmon/commit/8df4db9433a2aab59dbbac1a70281032d1af7734
https://github.com/intel/perfmon/commit/846bd247c6e04acc572ca56c992e9e65852bbe63The
tsx_cycles_per_elisionmetric is updated from PR:
https://github.com/intel/perfmon/pull/116Signed-off-by Ian Rogers <[email protected]>
Thanks Ian. The whole patch series looks good to me.
Thanks,
Kan
Quoted 343 lines
--- .../arch/x86/alderlake/adl-metrics.json | 2 +- .../arch/x86/alderlake/frontend.json | 42 ++++++++++-- .../pmu-events/arch/x86/alderlake/memory.json | 4 +- .../arch/x86/alderlake/pipeline.json | 20 +++++- .../x86/alderlake/uncore-interconnect.json | 2 + .../arch/x86/alderlaken/memory.json | 4 +- .../arch/x86/alderlaken/pipeline.json | 16 +++++ .../x86/alderlaken/uncore-interconnect.json | 66 +++++++++++++++++++ tools/perf/pmu-events/arch/x86/mapfile.csv | 4 +- 9 files changed, 146 insertions(+), 14 deletions(-)tools/perf/pmu-events/arch/x86/alderlake/adl-metrics.json +1
-1@@ -99,7 +99,7 @@ }, { "BriefDescription": "Number of cycles within a transaction divided by the number of elisions.", - "MetricExpr": "(cycles\\-t / el\\-start if has_event(cycles\\-t) else 0)", + "MetricExpr": "(cycles\\-t / el\\-start if has_event(el\\-start) else 0)", "MetricGroup": "transaction", "MetricName": "tsx_cycles_per_elision", "ScaleUnit": "1cycles / elision"tools/perf/pmu-events/arch/x86/alderlake/frontend.json +36
-6@@ -394,31 +394,61 @@ "Unit": "cpu_core" }, { - "BriefDescription": "Uops not delivered by IDQ when backend of the machine is not stalled", + "BriefDescription": "Uops not delivered by IDQ when backend of the machine is not stalled [This event is alias to IDQ_UOPS_NOT_DELIVERED.CORE]", + "EventCode": "0x9c", + "EventName": "IDQ_BUBBLES.CORE", + "PublicDescription": "Counts the number of uops not delivered to by the Instruction Decode Queue (IDQ) to the back-end of the pipeline when there was no back-end stalls. This event counts for one SMT thread in a given cycle. [This event is alias to IDQ_UOPS_NOT_DELIVERED.CORE]", + "SampleAfterValue": "1000003", + "UMask": "0x1", + "Unit": "cpu_core" + }, + { + "BriefDescription": "Cycles when no uops are not delivered by the IDQ when backend of the machine is not stalled [This event is alias to IDQ_UOPS_NOT_DELIVERED.CYCLES_0_UOPS_DELIV.CORE]", + "CounterMask": "6", + "EventCode": "0x9c", + "EventName": "IDQ_BUBBLES.CYCLES_0_UOPS_DELIV.CORE", + "PublicDescription": "Counts the number of cycles when no uops were delivered by the Instruction Decode Queue (IDQ) to the back-end of the pipeline when there was no back-end stalls. This event counts for one SMT thread in a given cycle. [This event is alias to IDQ_UOPS_NOT_DELIVERED.CYCLES_0_UOPS_DELIV.CORE]", + "SampleAfterValue": "1000003", + "UMask": "0x1", + "Unit": "cpu_core" + }, + { + "BriefDescription": "Cycles when optimal number of uops was delivered to the back-end when the back-end is not stalled [This event is alias to IDQ_UOPS_NOT_DELIVERED.CYCLES_FE_WAS_OK]", + "CounterMask": "1", + "EventCode": "0x9c", + "EventName": "IDQ_BUBBLES.CYCLES_FE_WAS_OK", + "Invert": "1", + "PublicDescription": "Counts the number of cycles when the optimal number of uops were delivered by the Instruction Decode Queue (IDQ) to the back-end of the pipeline when there was no back-end stalls. This event counts for one SMT thread in a given cycle. [This event is alias to IDQ_UOPS_NOT_DELIVERED.CYCLES_FE_WAS_OK]", + "SampleAfterValue": "1000003", + "UMask": "0x1", + "Unit": "cpu_core" + }, + { + "BriefDescription": "Uops not delivered by IDQ when backend of the machine is not stalled [This event is alias to IDQ_BUBBLES.CORE]", "EventCode": "0x9c", "EventName": "IDQ_UOPS_NOT_DELIVERED.CORE", - "PublicDescription": "Counts the number of uops not delivered to by the Instruction Decode Queue (IDQ) to the back-end of the pipeline when there was no back-end stalls. This event counts for one SMT thread in a given cycle.", + "PublicDescription": "Counts the number of uops not delivered to by the Instruction Decode Queue (IDQ) to the back-end of the pipeline when there was no back-end stalls. This event counts for one SMT thread in a given cycle. [This event is alias to IDQ_BUBBLES.CORE]", "SampleAfterValue": "1000003", "UMask": "0x1", "Unit": "cpu_core" }, { - "BriefDescription": "Cycles when no uops are not delivered by the IDQ when backend of the machine is not stalled", + "BriefDescription": "Cycles when no uops are not delivered by the IDQ when backend of the machine is not stalled [This event is alias to IDQ_BUBBLES.CYCLES_0_UOPS_DELIV.CORE]", "CounterMask": "6", "EventCode": "0x9c", "EventName": "IDQ_UOPS_NOT_DELIVERED.CYCLES_0_UOPS_DELIV.CORE", - "PublicDescription": "Counts the number of cycles when no uops were delivered by the Instruction Decode Queue (IDQ) to the back-end of the pipeline when there was no back-end stalls. This event counts for one SMT thread in a given cycle.", + "PublicDescription": "Counts the number of cycles when no uops were delivered by the Instruction Decode Queue (IDQ) to the back-end of the pipeline when there was no back-end stalls. This event counts for one SMT thread in a given cycle. [This event is alias to IDQ_BUBBLES.CYCLES_0_UOPS_DELIV.CORE]", "SampleAfterValue": "1000003", "UMask": "0x1", "Unit": "cpu_core" }, { - "BriefDescription": "Cycles when optimal number of uops was delivered to the back-end when the back-end is not stalled", + "BriefDescription": "Cycles when optimal number of uops was delivered to the back-end when the back-end is not stalled [This event is alias to IDQ_BUBBLES.CYCLES_FE_WAS_OK]", "CounterMask": "1", "EventCode": "0x9c", "EventName": "IDQ_UOPS_NOT_DELIVERED.CYCLES_FE_WAS_OK", "Invert": "1", - "PublicDescription": "Counts the number of cycles when the optimal number of uops were delivered by the Instruction Decode Queue (IDQ) to the back-end of the pipeline when there was no back-end stalls. This event counts for one SMT thread in a given cycle.", + "PublicDescription": "Counts the number of cycles when the optimal number of uops were delivered by the Instruction Decode Queue (IDQ) to the back-end of the pipeline when there was no back-end stalls. This event counts for one SMT thread in a given cycle. [This event is alias to IDQ_BUBBLES.CYCLES_FE_WAS_OK]", "SampleAfterValue": "1000003", "UMask": "0x1", "Unit": "cpu_core"tools/perf/pmu-events/arch/x86/alderlake/memory.json +2
-2@@ -248,7 +248,7 @@ "Unit": "cpu_core" }, { - "BriefDescription": "Counts demand data reads that were not supplied by the L3 cache.", + "BriefDescription": "Counts demand data reads that were not supplied by the L3 cache. [L3_MISS_LOCAL is alias to L3_MISS]", "EventCode": "0xB7", "EventName": "OCR.DEMAND_DATA_RD.L3_MISS_LOCAL", "MSRIndex": "0x1a6,0x1a7",@@ -278,7 +278,7 @@ "Unit": "cpu_core" }, { - "BriefDescription": "Counts demand reads for ownership (RFO) and software prefetches for exclusive ownership (PREFETCHW) that were not supplied by the L3 cache.", + "BriefDescription": "Counts demand reads for ownership (RFO) and software prefetches for exclusive ownership (PREFETCHW) that were not supplied by the L3 cache. [L3_MISS_LOCAL is alias to L3_MISS]", "EventCode": "0xB7", "EventName": "OCR.DEMAND_RFO.L3_MISS_LOCAL", "MSRIndex": "0x1a6,0x1a7",tools/perf/pmu-events/arch/x86/alderlake/pipeline.json +19
-1@@ -238,6 +238,15 @@ "UMask": "0x8", "Unit": "cpu_core" }, + { + "BriefDescription": "Counts the number of near taken branch instructions retired.", + "EventCode": "0xc4", + "EventName": "BR_INST_RETIRED.NEAR_TAKEN", + "PEBS": "1", + "SampleAfterValue": "200003", + "UMask": "0xc0", + "Unit": "cpu_atom" + }, { "BriefDescription": "Taken branch instructions retired.", "EventCode": "0xc4",@@ -411,6 +420,15 @@ "UMask": "0x7e", "Unit": "cpu_atom" }, + { + "BriefDescription": "Counts the number of mispredicted near taken branch instructions retired.", + "EventCode": "0xc5", + "EventName": "BR_MISP_RETIRED.NEAR_TAKEN", + "PEBS": "1", + "SampleAfterValue": "200003", + "UMask": "0x80", + "Unit": "cpu_atom" + }, { "BriefDescription": "Number of near branch instructions retired that were mispredicted and taken.", "EventCode": "0xc5",@@ -842,7 +860,7 @@ "Unit": "cpu_core" }, { - "BriefDescription": "INT_MISC.UNKNOWN_BRANCH_CYCLES", + "BriefDescription": "Bubble cycles of BAClear (Unknown Branch).", "EventCode": "0xad", "EventName": "INT_MISC.UNKNOWN_BRANCH_CYCLES", "MSRIndex": "0x3F7",tools/perf/pmu-events/arch/x86/alderlake/uncore-interconnect.json +2
-0@@ -25,6 +25,7 @@ }, { "BriefDescription": "This event is deprecated. Refer to new event UNC_ARB_REQ_TRK_REQUEST.DRD", + "Deprecated": "1", "EventCode": "0x81", "EventName": "UNC_ARB_DAT_REQUESTS.RD", "PerPkg": "1",@@ -33,6 +34,7 @@ }, { "BriefDescription": "This event is deprecated. Refer to new event UNC_ARB_DAT_OCCUPANCY.ALL", + "Deprecated": "1", "EventCode": "0x85", "EventName": "UNC_ARB_IFA_OCCUPANCY.ALL", "PerPkg": "1",tools/perf/pmu-events/arch/x86/alderlaken/memory.json +2
-2@@ -59,7 +59,7 @@ "UMask": "0x1" }, { - "BriefDescription": "Counts demand data reads that were not supplied by the L3 cache.", + "BriefDescription": "Counts demand data reads that were not supplied by the L3 cache. [L3_MISS_LOCAL is alias to L3_MISS]", "EventCode": "0xB7", "EventName": "OCR.DEMAND_DATA_RD.L3_MISS_LOCAL", "MSRIndex": "0x1a6,0x1a7",@@ -77,7 +77,7 @@ "UMask": "0x1" }, { - "BriefDescription": "Counts demand reads for ownership (RFO) and software prefetches for exclusive ownership (PREFETCHW) that were not supplied by the L3 cache.", + "BriefDescription": "Counts demand reads for ownership (RFO) and software prefetches for exclusive ownership (PREFETCHW) that were not supplied by the L3 cache. [L3_MISS_LOCAL is alias to L3_MISS]", "EventCode": "0xB7", "EventName": "OCR.DEMAND_RFO.L3_MISS_LOCAL", "MSRIndex": "0x1a6,0x1a7",tools/perf/pmu-events/arch/x86/alderlaken/pipeline.json +16
-0@@ -90,6 +90,14 @@ "SampleAfterValue": "200003", "UMask": "0xf7" }, + { + "BriefDescription": "Counts the number of near taken branch instructions retired.", + "EventCode": "0xc4", + "EventName": "BR_INST_RETIRED.NEAR_TAKEN", + "PEBS": "1", + "SampleAfterValue": "200003", + "UMask": "0xc0" + }, { "BriefDescription": "This event is deprecated. Refer to new event BR_INST_RETIRED.INDIRECT", "Deprecated": "1",@@ -183,6 +191,14 @@ "SampleAfterValue": "200003", "UMask": "0x7e" }, + { + "BriefDescription": "Counts the number of mispredicted near taken branch instructions retired.", + "EventCode": "0xc5", + "EventName": "BR_MISP_RETIRED.NEAR_TAKEN", + "PEBS": "1", + "SampleAfterValue": "200003", + "UMask": "0x80" + }, { "BriefDescription": "This event is deprecated. Refer to new event BR_MISP_RETIRED.INDIRECT", "Deprecated": "1",tools/perf/pmu-events/arch/x86/alderlaken/uncore-interconnect.json +66
-0@@ -7,6 +7,56 @@ "UMask": "0x1", "Unit": "ARB" }, + { + "BriefDescription": "Each cycle counts number of any coherent request at memory controller that were issued by any core.", + "EventCode": "0x85", + "EventName": "UNC_ARB_DAT_OCCUPANCY.ALL", + "PerPkg": "1", + "UMask": "0x1", + "Unit": "ARB" + }, + { + "BriefDescription": "Each cycle counts number of coherent reads pending on data return from memory controller that were issued by any core.", + "EventCode": "0x85", + "EventName": "UNC_ARB_DAT_OCCUPANCY.RD", + "PerPkg": "1", + "UMask": "0x2", + "Unit": "ARB" + }, + { + "BriefDescription": "This event is deprecated. Refer to new event UNC_ARB_REQ_TRK_REQUEST.DRD", + "Deprecated": "1", + "EventCode": "0x81", + "EventName": "UNC_ARB_DAT_REQUESTS.RD", + "PerPkg": "1", + "UMask": "0x2", + "Unit": "ARB" + }, + { + "BriefDescription": "This event is deprecated. Refer to new event UNC_ARB_DAT_OCCUPANCY.ALL", + "Deprecated": "1", + "EventCode": "0x85", + "EventName": "UNC_ARB_IFA_OCCUPANCY.ALL", + "PerPkg": "1", + "UMask": "0x1", + "Unit": "ARB" + }, + { + "BriefDescription": "Each cycle count number of 'valid' coherent Data Read entries . Such entry is defined as valid when it is allocated till deallocation. Doesn't include prefetches [This event is alias to UNC_ARB_TRK_OCCUPANCY.RD]", + "EventCode": "0x80", + "EventName": "UNC_ARB_REQ_TRK_OCCUPANCY.DRD", + "PerPkg": "1", + "UMask": "0x2", + "Unit": "ARB" + }, + { + "BriefDescription": "Number of all coherent Data Read entries. Doesn't include prefetches [This event is alias to UNC_ARB_TRK_REQUESTS.RD]", + "EventCode": "0x81", + "EventName": "UNC_ARB_REQ_TRK_REQUEST.DRD", + "PerPkg": "1", + "UMask": "0x2", + "Unit": "ARB" + }, { "BriefDescription": "Each cycle counts number of all outgoing valid entries in ReqTrk. Such entry is defined as valid from its allocation in ReqTrk till deallocation. Accounts for Coherent and non-coherent traffic.", "EventCode": "0x80",@@ -15,6 +65,14 @@ "UMask": "0x1", "Unit": "ARB" }, + { + "BriefDescription": "Each cycle count number of 'valid' coherent Data Read entries . Such entry is defined as valid when it is allocated till deallocation. Doesn't include prefetches [This event is alias to UNC_ARB_REQ_TRK_OCCUPANCY.DRD]", + "EventCode": "0x80", + "EventName": "UNC_ARB_TRK_OCCUPANCY.RD", + "PerPkg": "1", + "UMask": "0x2", + "Unit": "ARB" + }, { "BriefDescription": "Counts the number of coherent and in-coherent requests initiated by IA cores, processor graphic units, or LLC.", "EventCode": "0x81",@@ -22,5 +80,13 @@ "PerPkg": "1", "UMask": "0x1", "Unit": "ARB" + }, + { + "BriefDescription": "Number of all coherent Data Read entries. Doesn't include prefetches [This event is alias to UNC_ARB_REQ_TRK_REQUEST.DRD]", + "EventCode": "0x81", + "EventName": "UNC_ARB_TRK_REQUESTS.RD", + "PerPkg": "1", + "UMask": "0x2", + "Unit": "ARB" } ]tools/perf/pmu-events/arch/x86/mapfile.csv +2
-2@@ -1,6 +1,6 @@ Family-model,Version,Filename,EventType -GenuineIntel-6-(97|9A|B7|BA|BF),v1.21,alderlake,core -GenuineIntel-6-BE,v1.21,alderlaken,core +GenuineIntel-6-(97|9A|B7|BA|BF),v1.23,alderlake,core +GenuineIntel-6-BE,v1.23,alderlaken,core GenuineIntel-6-(1C|26|27|35|36),v4,bonnell,core GenuineIntel-6-(3D|47),v28,broadwell,core GenuineIntel-6-56,v11,broadwellde,core
Quoted 9 lines · Kan
Quoted 5 lines · Ian Rogers
Update alderlake and alderlaken events from v1.21 to v1.23 adding the changes from:
https://github.com/intel/perfmon/commit/8df4db9433a2aab59dbbac1a70281032d1af7734
https://github.com/intel/perfmon/commit/846bd247c6e04acc572ca56c992e9e65852bbe63
Quoted 8 lines
Quoted 4 lines
The
tsx_cycles_per_elisionmetric is updated from PR:
https://github.com/intel/perfmon/pull/116Signed-off-by Ian Rogers <[email protected]>Thanks Ian. The whole patch series looks good to me.
Reviewed-by Kan Liang <[email protected]>
Applied to perf-tools-next, thanks!