diff options
author | Ian Rogers <irogers@google.com> | 2022-02-01 04:58:47 +0300 |
---|---|---|
committer | Arnaldo Carvalho de Melo <acme@redhat.com> | 2022-02-15 03:16:38 +0300 |
commit | dd7415ce88d0ca4cc0cda52a6c93fba4d0e99128 (patch) | |
tree | 039c78e6a7394f56fb40b7b18af97aed0c0b53bf /tools/perf/pmu-events | |
parent | 4dd252721b288e469650db83b831e2236baf4671 (diff) | |
download | linux-dd7415ce88d0ca4cc0cda52a6c93fba4d0e99128.tar.xz |
perf vendor events: Update metrics for Icelake
Based on TMA_metrics-full.csv version 4.3 at 01.org:
https://download.01.org/perfmon/
Events are updated to version 1.12:
https://download.01.org/perfmon/ICL
Json files generated by the latest code at:
https://github.com/intel/event-converter-for-linux-perf
Tested:
Not tested on an Icelake, on a SkylakeX:
...
9: Parse perf pmu format : Ok
10: PMU events :
10.1: PMU event table sanity : Ok
10.2: PMU event map aliases : Ok
10.3: Parsing of PMU event table metrics : Ok
10.4: Parsing of PMU event table metrics with fake PMUs : Ok
...
Reviewed-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Ian Rogers <irogers@google.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Alexandre Torgue <alexandre.torgue@foss.st.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: James Clark <james.clark@arm.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: John Garry <john.garry@huawei.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Maxime Coquelin <mcoquelin.stm32@gmail.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Zhengjun Xing <zhengjun.xing@linux.intel.com>
Link: https://lore.kernel.org/r/20220201015858.1226914-16-irogers@google.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Diffstat (limited to 'tools/perf/pmu-events')
-rw-r--r-- | tools/perf/pmu-events/arch/x86/icelake/cache.json | 658 | ||||
-rw-r--r-- | tools/perf/pmu-events/arch/x86/icelake/floating-point.json | 69 | ||||
-rw-r--r-- | tools/perf/pmu-events/arch/x86/icelake/frontend.json | 449 | ||||
-rw-r--r-- | tools/perf/pmu-events/arch/x86/icelake/icl-metrics.json | 338 | ||||
-rw-r--r-- | tools/perf/pmu-events/arch/x86/icelake/memory.json | 591 | ||||
-rw-r--r-- | tools/perf/pmu-events/arch/x86/icelake/other.json | 622 | ||||
-rw-r--r-- | tools/perf/pmu-events/arch/x86/icelake/pipeline.json | 1067 | ||||
-rw-r--r-- | tools/perf/pmu-events/arch/x86/icelake/virtual-memory.json | 178 |
8 files changed, 2105 insertions, 1867 deletions
diff --git a/tools/perf/pmu-events/arch/x86/icelake/cache.json b/tools/perf/pmu-events/arch/x86/icelake/cache.json index 49fe78fb6538..96dcd387c70e 100644 --- a/tools/perf/pmu-events/arch/x86/icelake/cache.json +++ b/tools/perf/pmu-events/arch/x86/icelake/cache.json @@ -1,89 +1,78 @@ [ { - "BriefDescription": "L2 code requests", + "BriefDescription": "Counts the number of cache lines replaced in L1 data cache.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0x24", - "EventName": "L2_RQSTS.ALL_CODE_RD", + "EventCode": "0x51", + "EventName": "L1D.REPLACEMENT", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts the total number of L2 code requests.", - "SampleAfterValue": "200003", + "PublicDescription": "Counts L1D data line replacements including opportunistic replacements, and replacements that require stall-for-replace or block-for-replace.", + "SampleAfterValue": "100003", "Speculative": "1", - "UMask": "0xe4" - }, - { - "BriefDescription": "Retired load instructions whose data sources were L3 hit and cross-core snoop missed in on-pkg core cache.", - "CollectPEBSRecord": "2", - "Counter": "0,1,2,3", - "Data_LA": "1", - "EventCode": "0xd2", - "EventName": "MEM_LOAD_L3_HIT_RETIRED.XSNP_MISS", - "PEBS": "1", - "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts the retired load instructions whose data sources were L3 hit and cross-core snoop missed in on-pkg core cache.", - "SampleAfterValue": "20011", "UMask": "0x1" }, { - "BriefDescription": "Demand requests that miss L2 cache", + "BriefDescription": "Number of cycles a demand request has waited due to L1D Fill Buffer (FB) unavailability.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0x24", - "EventName": "L2_RQSTS.ALL_DEMAND_MISS", + "EventCode": "0x48", + "EventName": "L1D_PEND_MISS.FB_FULL", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts demand requests that miss L2 cache.", - "SampleAfterValue": "200003", + "PublicDescription": "Counts number of cycles a demand request has waited due to L1D Fill Buffer (FB) unavailablability. Demand requests include cacheable/uncacheable demand load, store, lock or SW prefetch accesses.", + "SampleAfterValue": "1000003", "Speculative": "1", - "UMask": "0x27" + "UMask": "0x2" }, { - "BriefDescription": "Demand RFO requests including regular RFOs, locks, ItoM", + "BriefDescription": "Number of phases a demand request has waited due to L1D Fill Buffer (FB) unavailablability.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0xb0", - "EventName": "OFFCORE_REQUESTS.DEMAND_RFO", + "CounterMask": "1", + "EdgeDetect": "1", + "EventCode": "0x48", + "EventName": "L1D_PEND_MISS.FB_FULL_PERIODS", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts the demand RFO (read for ownership) requests including regular RFOs, locks, ItoM.", - "SampleAfterValue": "100003", + "PublicDescription": "Counts number of phases a demand request has waited due to L1D Fill Buffer (FB) unavailablability. Demand requests include cacheable/uncacheable demand load, store, lock or SW prefetch accesses.", + "SampleAfterValue": "1000003", "Speculative": "1", - "UMask": "0x4" + "UMask": "0x2" }, { - "BriefDescription": "RFO requests that hit L2 cache", + "BriefDescription": "Number of cycles a demand request has waited due to L1D due to lack of L2 resources.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0x24", - "EventName": "L2_RQSTS.RFO_HIT", + "EventCode": "0x48", + "EventName": "L1D_PEND_MISS.L2_STALL", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts the RFO (Read-for-Ownership) requests that hit L2 cache.", - "SampleAfterValue": "200003", + "PublicDescription": "Counts number of cycles a demand request has waited due to L1D due to lack of L2 resources. Demand requests include cacheable/uncacheable demand load, store, lock or SW prefetch accesses.", + "SampleAfterValue": "1000003", "Speculative": "1", - "UMask": "0xc2" + "UMask": "0x4" }, { - "BriefDescription": "Number of completed demand load requests that missed the L1, but hit the FB(fill buffer), because a preceding miss to the same cacheline initiated the line to be brought into L1, but data is not yet ready in L1.", + "BriefDescription": "Number of L1D misses that are outstanding", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "Data_LA": "1", - "EventCode": "0xd1", - "EventName": "MEM_LOAD_RETIRED.FB_HIT", - "PEBS": "1", + "EventCode": "0x48", + "EventName": "L1D_PEND_MISS.PENDING", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts retired load instructions with at least one uop was load missed in L1 but hit FB (Fill Buffers) due to preceding miss to the same cache line with data not ready.", - "SampleAfterValue": "100007", - "UMask": "0x40" + "PublicDescription": "Counts number of L1D misses that are outstanding in each cycle, that is each cycle the number of Fill Buffers (FB) outstanding required by Demand Reads. FB either is held by demand loads, or it is held by non-demand loads and gets hit at least once by demand. The valid outstanding interval is defined until the FB deallocation by one of the following ways: from FB allocation, if FB is allocated by demand from the demand Hit FB, if it is allocated by hardware or software prefetch. Note: In the L1D, a Demand Read contains cacheable or noncacheable demand loads, including ones causing cache-line splits and reads due to page walks resulted from any request type.", + "SampleAfterValue": "1000003", + "Speculative": "1", + "UMask": "0x1" }, { - "BriefDescription": "Offcore outstanding cacheable Core Data Read transactions in SuperQueue (SQ), queue to uncore", + "BriefDescription": "Cycles with L1D load Misses outstanding.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0x60", - "EventName": "OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD", + "CounterMask": "1", + "EventCode": "0x48", + "EventName": "L1D_PEND_MISS.PENDING_CYCLES", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts the number of offcore outstanding cacheable Core Data Read transactions in the super queue every cycle. A transaction is considered to be in the Offcore outstanding state between L2 miss and transaction completion sent to requestor (SQ de-allocation). See corresponding Umask under OFFCORE_REQUESTS.", + "PublicDescription": "Counts duration of L1D miss outstanding in cycles.", "SampleAfterValue": "1000003", "Speculative": "1", - "UMask": "0x8" + "UMask": "0x1" }, { "BriefDescription": "L2 cache lines filling L2", @@ -98,127 +87,124 @@ "UMask": "0x1f" }, { - "BriefDescription": "Retired load instructions that split across a cacheline boundary.", + "BriefDescription": "Modified cache lines that are evicted by L2 cache when triggered by an L2 cache fill.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "Data_LA": "1", - "EventCode": "0xd0", - "EventName": "MEM_INST_RETIRED.SPLIT_LOADS", - "PEBS": "1", + "EventCode": "0xF2", + "EventName": "L2_LINES_OUT.NON_SILENT", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts retired load instructions that split across a cacheline boundary.", - "SampleAfterValue": "100003", - "UMask": "0x41" + "PublicDescription": "Counts the number of lines that are evicted by L2 cache when triggered by an L2 cache fill. Those lines are in Modified state. Modified lines are written back to L3", + "SampleAfterValue": "200003", + "Speculative": "1", + "UMask": "0x2" }, { - "BriefDescription": "Retired load instructions with L3 cache hits as data sources", + "BriefDescription": "Non-modified cache lines that are silently dropped by L2 cache when triggered by an L2 cache fill.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "Data_LA": "1", - "EventCode": "0xd1", - "EventName": "MEM_LOAD_RETIRED.L3_HIT", - "PEBS": "1", + "EventCode": "0xF2", + "EventName": "L2_LINES_OUT.SILENT", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts retired load instructions with at least one uop that hit in the L3 cache.", - "SampleAfterValue": "100021", - "UMask": "0x4" + "PublicDescription": "Counts the number of lines that are silently dropped by L2 cache when triggered by an L2 cache fill. These lines are typically in Shared or Exclusive state. A non-threaded event.", + "SampleAfterValue": "200003", + "Speculative": "1", + "UMask": "0x1" }, { - "BriefDescription": "Demand Data Read miss L2, no rejects", + "BriefDescription": "Cache lines that have been L2 hardware prefetched but not used by demand accesses", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0x24", - "EventName": "L2_RQSTS.DEMAND_DATA_RD_MISS", + "EventCode": "0xf2", + "EventName": "L2_LINES_OUT.USELESS_HWPF", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts the number of demand Data Read requests that miss L2 cache. Only not rejected loads are counted.", + "PublicDescription": "Counts the number of cache lines that have been prefetched by the L2 hardware prefetcher but not used by demand access when evicted from the L2 cache", "SampleAfterValue": "200003", "Speculative": "1", - "UMask": "0x21" + "UMask": "0x4" }, { - "BriefDescription": "L2 cache misses when fetching instructions", + "BriefDescription": "L2 code requests", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0x24", - "EventName": "L2_RQSTS.CODE_RD_MISS", + "EventName": "L2_RQSTS.ALL_CODE_RD", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts L2 cache misses when fetching instructions.", + "PublicDescription": "Counts the total number of L2 code requests.", "SampleAfterValue": "200003", "Speculative": "1", - "UMask": "0x24" + "UMask": "0xe4" }, { - "BriefDescription": "Number of cycles a demand request has waited due to L1D Fill Buffer (FB) unavailability.", + "BriefDescription": "Demand Data Read requests", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0x48", - "EventName": "L1D_PEND_MISS.FB_FULL", + "EventCode": "0x24", + "EventName": "L2_RQSTS.ALL_DEMAND_DATA_RD", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts number of cycles a demand request has waited due to L1D Fill Buffer (FB) unavailablability. Demand requests include cacheable/uncacheable demand load, store, lock or SW prefetch accesses.", - "SampleAfterValue": "1000003", + "PublicDescription": "Counts the number of demand Data Read requests (including requests from L1D hardware prefetchers). These loads may hit or miss L2 cache. Only non rejected loads are counted.", + "SampleAfterValue": "200003", "Speculative": "1", - "UMask": "0x2" + "UMask": "0xe1" }, { - "BriefDescription": "Counts the number of cache lines replaced in L1 data cache.", + "BriefDescription": "Demand requests that miss L2 cache", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0x51", - "EventName": "L1D.REPLACEMENT", + "EventCode": "0x24", + "EventName": "L2_RQSTS.ALL_DEMAND_MISS", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts L1D data line replacements including opportunistic replacements, and replacements that require stall-for-replace or block-for-replace.", - "SampleAfterValue": "100003", + "PublicDescription": "Counts demand requests that miss L2 cache.", + "SampleAfterValue": "200003", "Speculative": "1", - "UMask": "0x1" + "UMask": "0x27" }, { - "BriefDescription": "All retired load instructions.", + "BriefDescription": "Demand requests to L2 cache", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "Data_LA": "1", - "EventCode": "0xd0", - "EventName": "MEM_INST_RETIRED.ALL_LOADS", - "PEBS": "1", + "EventCode": "0x24", + "EventName": "L2_RQSTS.ALL_DEMAND_REFERENCES", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts all retired load instructions. This event accounts for SW prefetch instructions for loads.", - "SampleAfterValue": "1000003", - "UMask": "0x81" + "PublicDescription": "Counts demand requests to L2 cache.", + "SampleAfterValue": "200003", + "Speculative": "1", + "UMask": "0xe7" }, { - "BriefDescription": "L2 writebacks that access L2 cache", + "BriefDescription": "RFO requests to L2 cache", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0xF0", - "EventName": "L2_TRANS.L2_WB", + "EventCode": "0x24", + "EventName": "L2_RQSTS.ALL_RFO", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts L2 writebacks that access L2 cache.", + "PublicDescription": "Counts the total number of RFO (read for ownership) requests to L2 cache. L2 RFO requests include both L1D demand RFO misses as well as L1D RFO prefetches.", "SampleAfterValue": "200003", "Speculative": "1", - "UMask": "0x40" + "UMask": "0xe2" }, { - "BriefDescription": "Demand Data Read requests", + "BriefDescription": "L2 cache hits when fetching instructions, code reads.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0x24", - "EventName": "L2_RQSTS.ALL_DEMAND_DATA_RD", + "EventName": "L2_RQSTS.CODE_RD_HIT", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts the number of demand Data Read requests (including requests from L1D hardware prefetchers). These loads may hit or miss L2 cache. Only non rejected loads are counted.", + "PublicDescription": "Counts L2 cache hits when fetching instructions, code reads.", "SampleAfterValue": "200003", "Speculative": "1", - "UMask": "0xe1" + "UMask": "0xc4" }, { - "BriefDescription": "Demand Data Read transactions pending for off-core. Highly correlated.", + "BriefDescription": "L2 cache misses when fetching instructions", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0x60", - "EventName": "OFFCORE_REQUESTS_OUTSTANDING.DEMAND_DATA_RD", + "EventCode": "0x24", + "EventName": "L2_RQSTS.CODE_RD_MISS", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts the number of off-core outstanding Demand Data Read transactions every cycle. A transaction is considered to be in the Off-core outstanding state between L2 cache miss and data-return to the core.", - "SampleAfterValue": "1000003", + "PublicDescription": "Counts L2 cache misses when fetching instructions.", + "SampleAfterValue": "200003", "Speculative": "1", - "UMask": "0x1" + "UMask": "0x24" }, { "BriefDescription": "Demand Data Read requests that hit L2 cache", @@ -233,131 +219,101 @@ "UMask": "0xc1" }, { - "BriefDescription": "Cycles the superQ cannot take any more entries.", + "BriefDescription": "Demand Data Read miss L2, no rejects", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0xf4", - "EventName": "SQ_MISC.SQ_FULL", + "EventCode": "0x24", + "EventName": "L2_RQSTS.DEMAND_DATA_RD_MISS", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts the cycles for which the thread is active and the superQ cannot take any more entries.", - "SampleAfterValue": "100003", + "PublicDescription": "Counts the number of demand Data Read requests that miss L2 cache. Only not rejected loads are counted.", + "SampleAfterValue": "200003", "Speculative": "1", - "UMask": "0x4" + "UMask": "0x21" }, { - "BriefDescription": "Cycles with L1D load Misses outstanding.", + "BriefDescription": "RFO requests that hit L2 cache", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "CounterMask": "1", - "EventCode": "0x48", - "EventName": "L1D_PEND_MISS.PENDING_CYCLES", + "EventCode": "0x24", + "EventName": "L2_RQSTS.RFO_HIT", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts duration of L1D miss outstanding in cycles.", - "SampleAfterValue": "1000003", + "PublicDescription": "Counts the RFO (Read-for-Ownership) requests that hit L2 cache.", + "SampleAfterValue": "200003", "Speculative": "1", - "UMask": "0x1" + "UMask": "0xc2" }, { - "BriefDescription": "Demand Data Read requests sent to uncore", + "BriefDescription": "RFO requests that miss L2 cache", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0xb0", - "EventName": "OFFCORE_REQUESTS.DEMAND_DATA_RD", + "EventCode": "0x24", + "EventName": "L2_RQSTS.RFO_MISS", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts the Demand Data Read requests sent to uncore. Use it in conjunction with OFFCORE_REQUESTS_OUTSTANDING to determine average latency in the uncore.", - "SampleAfterValue": "100003", + "PublicDescription": "Counts the RFO (Read-for-Ownership) requests that miss L2 cache.", + "SampleAfterValue": "200003", "Speculative": "1", - "UMask": "0x1" - }, - { - "BriefDescription": "Retired load instructions with L1 cache hits as data sources", - "CollectPEBSRecord": "2", - "Counter": "0,1,2,3", - "Data_LA": "1", - "EventCode": "0xd1", - "EventName": "MEM_LOAD_RETIRED.L1_HIT", - "PEBS": "1", - "PEBScounters": "0,1,2,3", - "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.", - "SampleAfterValue": "1000003", - "UMask": "0x1" + "UMask": "0x22" }, { - "BriefDescription": "Cycles when offcore outstanding cacheable Core Data Read transactions are present in SuperQueue (SQ), queue to uncore.", + "BriefDescription": "SW prefetch requests that hit L2 cache.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "CounterMask": "1", - "EventCode": "0x60", - "EventName": "OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD", + "EventCode": "0x24", + "EventName": "L2_RQSTS.SWPF_HIT", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts cycles when offcore outstanding cacheable Core Data Read transactions are present in the super queue. A transaction is considered to be in the Offcore outstanding state between L2 miss and transaction completion sent to requestor (SQ de-allocation). See corresponding Umask under OFFCORE_REQUESTS.", - "SampleAfterValue": "1000003", + "PublicDescription": "Counts Software prefetch requests that hit the L2 cache. Accounts for PREFETCHNTA and PREFETCHT0/1/2 instructions when FB is not full.", + "SampleAfterValue": "200003", "Speculative": "1", - "UMask": "0x8" + "UMask": "0xc8" }, { - "BriefDescription": "Cycles with offcore outstanding demand rfo reads transactions in SuperQueue (SQ), queue to uncore.", + "BriefDescription": "SW prefetch requests that miss L2 cache.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "CounterMask": "1", - "EventCode": "0x60", - "EventName": "OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO", + "EventCode": "0x24", + "EventName": "L2_RQSTS.SWPF_MISS", "PEBScounters": "0,1,2,3", - "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.", - "SampleAfterValue": "1000003", + "PublicDescription": "Counts Software prefetch requests that miss the L2 cache. Accounts for PREFETCHNTA and PREFETCHT0/1/2 instructions when FB is not full.", + "SampleAfterValue": "200003", "Speculative": "1", - "UMask": "0x4" + "UMask": "0x28" }, { - "BriefDescription": "Number of cycles a demand request has waited due to L1D due to lack of L2 resources.", + "BriefDescription": "L2 writebacks that access L2 cache", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0x48", - "EventName": "L1D_PEND_MISS.L2_STALL", + "EventCode": "0xF0", + "EventName": "L2_TRANS.L2_WB", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts number of cycles a demand request has waited due to L1D due to lack of L2 resources. Demand requests include cacheable/uncacheable demand load, store, lock or SW prefetch accesses.", - "SampleAfterValue": "1000003", + "PublicDescription": "Counts L2 writebacks that access L2 cache.", + "SampleAfterValue": "200003", "Speculative": "1", - "UMask": "0x4" + "UMask": "0x40" }, { - "BriefDescription": "Retired load instructions with L2 cache hits as data sources", + "BriefDescription": "Core-originated cacheable requests that missed L3 (Except hardware prefetches to the L3)", "CollectPEBSRecord": "2", - "Counter": "0,1,2,3", - "Data_LA": "1", - "EventCode": "0xd1", - "EventName": "MEM_LOAD_RETIRED.L2_HIT", - "PEBS": "1", - "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts retired load instructions with L2 cache hits as data sources.", - "SampleAfterValue": "200003", - "UMask": "0x2" + "Counter": "0,1,2,3,4,5,6,7", + "EventCode": "0x2e", + "EventName": "LONGEST_LAT_CACHE.MISS", + "PEBScounters": "0,1,2,3,4,5,6,7", + "PublicDescription": "Counts core-originated cacheable requests that miss the L3 cache (Longest Latency cache). Requests include data and code reads, Reads-for-Ownership (RFOs), speculative accesses and hardware prefetches to the L1 and L2. It does not include hardware prefetches to the L3, and may not count other types of requests to the L3.", + "SampleAfterValue": "100003", + "Speculative": "1", + "UMask": "0x41" }, { - "BriefDescription": "Retired load instructions with locked access.", + "BriefDescription": "All retired load instructions.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd0", - "EventName": "MEM_INST_RETIRED.LOCK_LOADS", - "PEBS": "1", - "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts retired load instructions with locked access.", - "SampleAfterValue": "100007", - "UMask": "0x21" - }, - { - "BriefDescription": "Retired load instructions missed L3 cache as data sources", - "CollectPEBSRecord": "2", - "Counter": "0,1,2,3", - "Data_LA": "1", - "EventCode": "0xd1", - "EventName": "MEM_LOAD_RETIRED.L3_MISS", + "EventName": "MEM_INST_RETIRED.ALL_LOADS", "PEBS": "1", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts retired load instructions with at least one uop that missed in the L3 cache.", - "SampleAfterValue": "50021", - "UMask": "0x20" + "PublicDescription": "Counts all retired load instructions. This event accounts for SW prefetch instructions for loads.", + "SampleAfterValue": "1000003", + "UMask": "0x81" }, { "BriefDescription": "All retired store instructions.", @@ -374,102 +330,97 @@ "UMask": "0x82" }, { - "BriefDescription": "Demand requests to L2 cache", + "BriefDescription": "All retired memory instructions.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0x24", - "EventName": "L2_RQSTS.ALL_DEMAND_REFERENCES", + "Data_LA": "1", + "EventCode": "0xd0", + "EventName": "MEM_INST_RETIRED.ANY", + "L1_Hit_Indication": "1", + "PEBS": "1", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts demand requests to L2 cache.", - "SampleAfterValue": "200003", - "Speculative": "1", - "UMask": "0xe7" + "PublicDescription": "Counts all retired memory instructions - loads and stores.", + "SampleAfterValue": "1000003", + "UMask": "0x83" }, { - "BriefDescription": "L2 cache hits when fetching instructions, code reads.", + "BriefDescription": "Retired load instructions with locked access.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0x24", - "EventName": "L2_RQSTS.CODE_RD_HIT", + "Data_LA": "1", + "EventCode": "0xd0", + "EventName": "MEM_INST_RETIRED.LOCK_LOADS", + "PEBS": "1", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts L2 cache hits when fetching instructions, code reads.", - "SampleAfterValue": "200003", - "Speculative": "1", - "UMask": "0xc4" + "PublicDescription": "Counts retired load instructions with locked access.", + "SampleAfterValue": "100007", + "UMask": "0x21" }, { - "BriefDescription": "Demand and prefetch data reads", + "BriefDescription": "Retired load instructions that split across a cacheline boundary.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0xB0", - "EventName": "OFFCORE_REQUESTS.ALL_DATA_RD", + "Data_LA": "1", + "EventCode": "0xd0", + "EventName": "MEM_INST_RETIRED.SPLIT_LOADS", + "PEBS": "1", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts the demand and prefetch data reads. All Core Data Reads include cacheable 'Demands' and L2 prefetchers (not L3 prefetchers). Counting also covers reads due to page walks resulted from any request type.", - "SampleAfterValue": "100003", - "Speculative": "1", - "UMask": "0x8" - }, - { - "BriefDescription": "Core-originated cacheable demand requests missed L3", - "CollectPEBSRecord": "2", - "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0x2e", - "EventName": "LONGEST_LAT_CACHE.MISS", - "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts core-originated cacheable requests that miss the L3 cache (Longest Latency cache). Requests include data and code reads, Reads-for-Ownership (RFOs), speculative accesses and hardware prefetches from L1 and L2. It does not include all misses to the L3.", + "PublicDescription": "Counts retired load instructions that split across a cacheline boundary.", "SampleAfterValue": "100003", - "Speculative": "1", - "UMask": "0x1" + "UMask": "0x41" }, { - "BriefDescription": "SW prefetch requests that miss L2 cache.", + "BriefDescription": "Retired store instructions that split across a cacheline boundary.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0x24", - "EventName": "L2_RQSTS.SWPF_MISS", + "Data_LA": "1", + "EventCode": "0xd0", + "EventName": "MEM_INST_RETIRED.SPLIT_STORES", + "L1_Hit_Indication": "1", + "PEBS": "1", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts Software prefetch requests that miss the L2 cache. This event accounts for PREFETCHNTA and PREFETCHT0/1/2 instructions.", - "SampleAfterValue": "200003", - "Speculative": "1", - "UMask": "0x28" + "PublicDescription": "Counts retired store instructions that split across a cacheline boundary.", + "SampleAfterValue": "100003", + "UMask": "0x42" }, { - "BriefDescription": "Retired load instructions missed L1 cache as data sources", + "BriefDescription": "Retired load instructions that miss the STLB.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "Data_LA": "1", - "EventCode": "0xd1", - "EventName": "MEM_LOAD_RETIRED.L1_MISS", + "EventCode": "0xd0", + "EventName": "MEM_INST_RETIRED.STLB_MISS_LOADS", "PEBS": "1", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts retired load instructions with at least one uop that missed in the L1 cache.", - "SampleAfterValue": "200003", - "UMask": "0x8" + "PublicDescription": "Number of retired load instructions that (start a) miss in the 2nd-level TLB (STLB).", + "SampleAfterValue": "100003", + "UMask": "0x11" }, { - "BriefDescription": "Number of L1D misses that are outstanding", + "BriefDescription": "Retired store instructions that miss the STLB.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0x48", - "EventName": "L1D_PEND_MISS.PENDING", + "Data_LA": "1", + "EventCode": "0xd0", + "EventName": "MEM_INST_RETIRED.STLB_MISS_STORES", + "L1_Hit_Indication": "1", + "PEBS": "1", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts number of L1D misses that are outstanding in each cycle, that is each cycle the number of Fill Buffers (FB) outstanding required by Demand Reads. FB either is held by demand loads, or it is held by non-demand loads and gets hit at least once by demand. The valid outstanding interval is defined until the FB deallocation by one of the following ways: from FB allocation, if FB is allocated by demand from the demand Hit FB, if it is allocated by hardware or software prefetch. Note: In the L1D, a Demand Read contains cacheable or noncacheable demand loads, including ones causing cache-line splits and reads due to page walks resulted from any request type.", - "SampleAfterValue": "1000003", - "Speculative": "1", - "UMask": "0x1" + "PublicDescription": "Number of retired store instructions that (start a) miss in the 2nd-level TLB (STLB).", + "SampleAfterValue": "100003", + "UMask": "0x12" }, { - "BriefDescription": "Number of phases a demand request has waited due to L1D Fill Buffer (FB) unavailablability.", + "BriefDescription": "Retired load instructions whose data sources were L3 and cross-core snoop hits in on-pkg core cache", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "CounterMask": "1", - "EdgeDetect": "1", - "EventCode": "0x48", - "EventName": "L1D_PEND_MISS.FB_FULL_PERIODS", + "Data_LA": "1", + "EventCode": "0xd2", + "EventName": "MEM_LOAD_L3_HIT_RETIRED.XSNP_HIT", + "PEBS": "1", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts number of phases a demand request has waited due to L1D Fill Buffer (FB) unavailablability. Demand requests include cacheable/uncacheable demand load, store, lock or SW prefetch accesses.", - "SampleAfterValue": "1000003", - "Speculative": "1", + "PublicDescription": "Counts retired load instructions whose data sources were L3 and cross-core snoop hits in on-pkg core cache.", + "SampleAfterValue": "20011", "UMask": "0x2" }, { @@ -486,17 +437,17 @@ "UMask": "0x4" }, { - "BriefDescription": "Retired load instructions whose data sources were L3 and cross-core snoop hits in on-pkg core cache", + "BriefDescription": "Retired load instructions whose data sources were L3 hit and cross-core snoop missed in on-pkg core cache.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd2", - "EventName": "MEM_LOAD_L3_HIT_RETIRED.XSNP_HIT", + "EventName": "MEM_LOAD_L3_HIT_RETIRED.XSNP_MISS", "PEBS": "1", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts retired load instructions whose data sources were L3 and cross-core snoop hits in on-pkg core cache.", + "PublicDescription": "Counts the retired load instructions whose data sources were L3 hit and cross-core snoop missed in on-pkg core cache.", "SampleAfterValue": "20011", - "UMask": "0x2" + "UMask": "0x1" }, { "BriefDescription": "Retired load instructions whose data sources were hits in L3 without snoops required", @@ -512,30 +463,56 @@ "UMask": "0x8" }, { - "BriefDescription": "Retired store instructions that miss the STLB.", + "BriefDescription": "Number of completed demand load requests that missed the L1, but hit the FB(fill buffer), because a preceding miss to the same cacheline initiated the line to be brought into L1, but data is not yet ready in L1.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "Data_LA": "1", - "EventCode": "0xd0", - "EventName": "MEM_INST_RETIRED.STLB_MISS_STORES", - "L1_Hit_Indication": "1", + "EventCode": "0xd1", + "EventName": "MEM_LOAD_RETIRED.FB_HIT", "PEBS": "1", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts retired store instructions that true miss the STLB.", - "SampleAfterValue": "100003", - "UMask": "0x12" + "PublicDescription": "Counts retired load instructions with at least one uop was load missed in L1 but hit FB (Fill Buffers) due to preceding miss to the same cache line with data not ready.", + "SampleAfterValue": "100007", + "UMask": "0x40" }, { - "BriefDescription": "RFO requests to L2 cache", + "BriefDescription": "Retired load instructions with L1 cache hits as data sources", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0x24", - "EventName": "L2_RQSTS.ALL_RFO", + "Data_LA": "1", + "EventCode": "0xd1", + "EventName": "MEM_LOAD_RETIRED.L1_HIT", + "PEBS": "1", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts the total number of RFO (read for ownership) requests to L2 cache. L2 RFO requests include both L1D demand RFO misses as well as L1D RFO prefetches.", + "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.", + "SampleAfterValue": "1000003", + "UMask": "0x1" + }, + { + "BriefDescription": "Retired load instructions missed L1 cache as data sources", + "CollectPEBSRecord": "2", + "Counter": "0,1,2,3", + "Data_LA": "1", + "EventCode": "0xd1", + "EventName": "MEM_LOAD_RETIRED.L1_MISS", + "PEBS": "1", + "PEBScounters": "0,1,2,3", + "PublicDescription": "Counts retired load instructions with at least one uop that missed in the L1 cache.", "SampleAfterValue": "200003", - "Speculative": "1", - "UMask": "0xe2" + "UMask": "0x8" + }, + { + "BriefDescription": "Retired load instructions with L2 cache hits as data sources", + "CollectPEBSRecord": "2", + "Counter": "0,1,2,3", + "Data_LA": "1", + "EventCode": "0xd1", + "EventName": "MEM_LOAD_RETIRED.L2_HIT", + "PEBS": "1", + "PEBScounters": "0,1,2,3", + "PublicDescription": "Counts retired load instructions with L2 cache hits as data sources.", + "SampleAfterValue": "200003", + "UMask": "0x2" }, { "BriefDescription": "Retired load instructions missed L2 cache as data sources", @@ -551,113 +528,150 @@ "UMask": "0x10" }, { - "BriefDescription": "Store Read transactions pending for off-core. Highly correlated.", + "BriefDescription": "Retired load instructions with L3 cache hits as data sources", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0x60", - "EventName": "OFFCORE_REQUESTS_OUTSTANDING.DEMAND_RFO", + "Data_LA": "1", + "EventCode": "0xd1", + "EventName": "MEM_LOAD_RETIRED.L3_HIT", + "PEBS": "1", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts the number of off-core outstanding read-for-ownership (RFO) store transactions every cycle. An RFO transaction is considered to be in the Off-core outstanding state between L2 cache miss and transaction completion.", - "SampleAfterValue": "1000003", - "Speculative": "1", + "PublicDescription": "Counts retired load instructions with at least one uop that hit in the L3 cache.", + "SampleAfterValue": "100021", "UMask": "0x4" }, { - "BriefDescription": "Non-modified cache lines that are silently dropped by L2 cache when triggered by an L2 cache fill.", + "BriefDescription": "Retired load instructions missed L3 cache as data sources", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0xF2", - "EventName": "L2_LINES_OUT.SILENT", + "Data_LA": "1", + "EventCode": "0xd1", + "EventName": "MEM_LOAD_RETIRED.L3_MISS", + "PEBS": "1", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts the number of lines that are silently dropped by L2 cache when triggered by an L2 cache fill. These lines are typically in Shared or Exclusive state. A non-threaded event.", - "SampleAfterValue": "200003", + "PublicDescription": "Counts retired load instructions with at least one uop that missed in the L3 cache.", + "SampleAfterValue": "50021", + "UMask": "0x20" + }, + { + "BriefDescription": "Demand and prefetch data reads", + "CollectPEBSRecord": "2", + "Counter": "0,1,2,3", + "EventCode": "0xB0", + "EventName": "OFFCORE_REQUESTS.ALL_DATA_RD", + "PEBScounters": "0,1,2,3", + "PublicDescription": "Counts the demand and prefetch data reads. All Core Data Reads include cacheable 'Demands' and L2 prefetchers (not L3 prefetchers). Counting also covers reads due to page walks resulted from any request type.", + "SampleAfterValue": "100003", "Speculative": "1", - "UMask": "0x1" + "UMask": "0x8" }, { - "BriefDescription": "Retired store instructions that split across a cacheline boundary.", + "BriefDescription": "Counts memory transactions sent to the uncore.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "Data_LA": "1", - "EventCode": "0xd0", - "EventName": "MEM_INST_RETIRED.SPLIT_STORES", - "L1_Hit_Indication": "1", - "PEBS": "1", + "EventCode": "0xB0", + "EventName": "OFFCORE_REQUESTS.ALL_REQUESTS", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts retired store instructions that split across a cacheline boundary.", + "PublicDescription": "Counts memory transactions sent to the uncore including requests initiated by the core, all L3 prefetches, reads resulting from page walks, and snoop responses.", "SampleAfterValue": "100003", - "UMask": "0x42" + "Speculative": "1", + "UMask": "0x80" }, { - "BriefDescription": "SW prefetch requests that hit L2 cache.", + "BriefDescription": "Demand Data Read requests sent to uncore", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0x24", - "EventName": "L2_RQSTS.SWPF_HIT", + "EventCode": "0xb0", + "EventName": "OFFCORE_REQUESTS.DEMAND_DATA_RD", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts Software prefetch requests that hit the L2 cache. This event accounts for PREFETCHNTA and PREFETCHT0/1/2 instructions.", - "SampleAfterValue": "200003", + "PublicDescription": "Counts the Demand Data Read requests sent to uncore. Use it in conjunction with OFFCORE_REQUESTS_OUTSTANDING to determine average latency in the uncore.", + "SampleAfterValue": "100003", "Speculative": "1", - "UMask": "0xc8" + "UMask": "0x1" }, { - "BriefDescription": "Retired load instructions that miss the STLB.", + "BriefDescription": "Demand RFO requests including regular RFOs, locks, ItoM", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "Data_LA": "1", - "EventCode": "0xd0", - "EventName": "MEM_INST_RETIRED.STLB_MISS_LOADS", - "PEBS": "1", + "EventCode": "0xb0", + "EventName": "OFFCORE_REQUESTS.DEMAND_RFO", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts retired load instructions that true miss the STLB.", + "PublicDescription": "Counts the demand RFO (read for ownership) requests including regular RFOs, locks, ItoM.", "SampleAfterValue": "100003", - "UMask": "0x11" + "Speculative": "1", + "UMask": "0x4" }, { - "BriefDescription": "RFO requests that miss L2 cache", + "BriefDescription": "For every cycle, increments by the number of outstanding data read requests pending.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0x24", - "EventName": "L2_RQSTS.RFO_MISS", + "EventCode": "0x60", + "EventName": "OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts the RFO (Read-for-Ownership) requests that miss L2 cache.", - "SampleAfterValue": "200003", + "PublicDescription": "For every cycle, increments by the number of outstanding data read requests pending. Data read requests include cacheable demand reads and L2 prefetches, but do not include RFOs, code reads or prefetches to the L3. Reads due to page walks resulting from any request type will also be counted. Requests are considered outstanding from the time they miss the core's L2 cache until the transaction completion message is sent to the requestor.", + "SampleAfterValue": "1000003", "Speculative": "1", - "UMask": "0x22" + "UMask": "0x8" }, { - "BriefDescription": "Modified cache lines that are evicted by L2 cache when triggered by an L2 cache fill.", + "BriefDescription": "Cycles where at least 1 outstanding data read request is pending.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0xF2", - "EventName": "L2_LINES_OUT.NON_SILENT", + "CounterMask": "1", + "EventCode": "0x60", + "EventName": "OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts the number of lines that are evicted by L2 cache when triggered by an L2 cache fill. Those lines are in Modified state. Modified lines are written back to L3", - "SampleAfterValue": "200003", + "PublicDescription": "Cycles where at least 1 outstanding data read request is pending. Data read requests include cacheable demand reads and L2 prefetches, but do not include RFOs, code reads or prefetches to the L3. Reads due to page walks resulting from any request type will also be counted. Requests are considered outstanding from the time they miss the core's L2 cache until the transaction completion message is sent to the requestor.", + "SampleAfterValue": "1000003", "Speculative": "1", - "UMask": "0x2" + "UMask": "0x8" }, { - "BriefDescription": "Any memory transaction that reached the SQ.", + "BriefDescription": "Cycles where at least 1 outstanding Demand RFO request is pending.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0xB0", - "EventName": "OFFCORE_REQUESTS.ALL_REQUESTS", + "CounterMask": "1", + "EventCode": "0x60", + "EventName": "OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts memory transactions reached the super queue including requests initiated by the core, all L3 prefetches, page walks, etc..", - "SampleAfterValue": "100003", + "PublicDescription": "Cycles where at least 1 outstanding Demand RFO request is pending. RFOs are initiated by a core as part of a data store operation. Demand RFO requests include RFOs, locks, and ItoM transactions. Requests are considered outstanding from the time they miss the core's L2 cache until the transaction completion message is sent to the requestor.", + "SampleAfterValue": "1000003", "Speculative": "1", - "UMask": "0x80" + "UMask": "0x4" }, { - "BriefDescription": "Cache lines that have been L2 hardware prefetched but not used by demand accesses", + "BriefDescription": "For every cycle, increments by the number of outstanding demand data read requests pending.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0xf2", - "EventName": "L2_LINES_OUT.USELESS_HWPF", + "EventCode": "0x60", + "EventName": "OFFCORE_REQUESTS_OUTSTANDING.DEMAND_DATA_RD", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts the number of cache lines that have been prefetched by the L2 hardware prefetcher but not used by demand access when evicted from the L2 cache", - "SampleAfterValue": "200003", + "PublicDescription": "For every cycle, increments by the number of outstanding demand data read requests pending. Requests are considered outstanding from the time they miss the core's L2 cache until the transaction completion message is sent to the requestor.", + "SampleAfterValue": "1000003", + "Speculative": "1", + "UMask": "0x1" + }, + { + "BriefDescription": "Store Read transactions pending for off-core. Highly correlated.", + "CollectPEBSRecord": "2", + "Counter": "0,1,2,3", + "EventCode": "0x60", + "EventName": "OFFCORE_REQUESTS_OUTSTANDING.DEMAND_RFO", + "PEBScounters": "0,1,2,3", + "PublicDescription": "Counts the number of off-core outstanding read-for-ownership (RFO) store transactions every cycle. An RFO transaction is considered to be in the Off-core outstanding state between L2 cache miss and transaction completion.", + "SampleAfterValue": "1000003", + "Speculative": "1", + "UMask": "0x4" + }, + { + "BriefDescription": "Cycles the queue waiting for offcore responses is full.", + "CollectPEBSRecord": "2", + "Counter": "0,1,2,3", + "EventCode": "0xf4", + "EventName": "SQ_MISC.SQ_FULL", + "PEBScounters": "0,1,2,3", + "PublicDescription": "Counts the cycles for which the thread is active and the queue waiting for responses from the uncore cannot take any more entries.", + "SampleAfterValue": "100003", "Speculative": "1", "UMask": "0x4" } diff --git a/tools/perf/pmu-events/arch/x86/icelake/floating-point.json b/tools/perf/pmu-events/arch/x86/icelake/floating-point.json index 5391c4f6eca3..4347e2d0d090 100644 --- a/tools/perf/pmu-events/arch/x86/icelake/floating-point.json +++ b/tools/perf/pmu-events/arch/x86/icelake/floating-point.json @@ -1,95 +1,102 @@ [ { - "BriefDescription": "Counts number of SSE/AVX computational 512-bit packed double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 8 computation operations, one for each element. Applies to SSE* and AVX* packed double precision floating-point instructions: ADD SUB MUL DIV MIN MAX SQRT RSQRT14 RCP14 FM(N)ADD/SUB. FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element.", + "BriefDescription": "Counts all microcode FP assists.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xc7", - "EventName": "FP_ARITH_INST_RETIRED.512B_PACKED_DOUBLE", + "EventCode": "0xc1", + "EventName": "ASSISTS.FP", "PEBScounters": "0,1,2,3,4,5,6,7", + "PublicDescription": "Counts all microcode Floating Point assists.", "SampleAfterValue": "100003", - "UMask": "0x40" + "Speculative": "1", + "UMask": "0x2" }, { - "BriefDescription": "Number of SSE/AVX computational 128-bit packed single precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 4 computation operations, one for each element. Applies to SSE* and AVX* packed single precision floating-point instructions: ADD SUB MUL DIV MIN MAX RCP14 RSQRT14 SQRT DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element.", + "BriefDescription": "Counts number of SSE/AVX computational 128-bit packed double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 2 computation operations, one for each element. Applies to SSE* and AVX* packed double precision floating-point instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX SQRT DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc7", - "EventName": "FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE", + "EventName": "FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts number of SSE/AVX computational 128-bit packed single precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 4 computation operations, one for each element. Applies to SSE* and AVX* packed single precision floating-point instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX SQRT RSQRT RCP DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element.", + "PublicDescription": "Number of SSE/AVX computational 128-bit packed double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 2 computation operations, one for each element. Applies to SSE* and AVX* packed double precision floating-point instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX SQRT DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.", "SampleAfterValue": "100003", - "UMask": "0x8" + "UMask": "0x4" }, { - "BriefDescription": "Counts number of SSE/AVX computational 512-bit packed double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 16 computation operations, one for each element. Applies to SSE* and AVX* packed double precision floating-point instructions: ADD SUB MUL DIV MIN MAX SQRT RSQRT14 RCP14 FM(N)ADD/SUB. FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element.", + "BriefDescription": "Number of SSE/AVX computational 128-bit packed single precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 4 computation operations, one for each element. Applies to SSE* and AVX* packed single precision floating-point instructions: ADD SUB MUL DIV MIN MAX RCP14 RSQRT14 SQRT DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc7", - "EventName": "FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE", + "EventName": "FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE", "PEBScounters": "0,1,2,3,4,5,6,7", + "PublicDescription": "Number of SSE/AVX computational 128-bit packed single precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 4 computation operations, one for each element. Applies to SSE* and AVX* packed single precision floating-point instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX SQRT RSQRT RCP DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.", "SampleAfterValue": "100003", - "UMask": "0x80" + "UMask": "0x8" }, { - "BriefDescription": "Counts number of SSE/AVX computational scalar double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 1 computational operation. Applies to SSE* and AVX* scalar double precision floating-point instructions: ADD SUB MUL DIV MIN MAX SQRT FM(N)ADD/SUB. FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element.", + "BriefDescription": "Counts number of SSE/AVX computational 256-bit packed double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 4 computation operations, one for each element. Applies to SSE* and AVX* packed double precision floating-point instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX SQRT FM(N)ADD/SUB. FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc7", - "EventName": "FP_ARITH_INST_RETIRED.SCALAR_DOUBLE", + "EventName": "FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE", "PEBScounters": "0,1,2,3,4,5,6,7", + "PublicDescription": "Number of SSE/AVX computational 256-bit packed double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 4 computation operations, one for each element. Applies to SSE* and AVX* packed double precision floating-point instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX SQRT FM(N)ADD/SUB. FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.", "SampleAfterValue": "100003", - "UMask": "0x1" + "UMask": "0x10" }, { - "BriefDescription": "Counts number of SSE/AVX computational 128-bit packed double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 2 computation operations, one for each element. Applies to SSE* and AVX* packed double precision floating-point instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX SQRT DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element.", + "BriefDescription": "Counts number of SSE/AVX computational 256-bit packed single precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 8 computation operations, one for each element. Applies to SSE* and AVX* packed single precision floating-point instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX SQRT RSQRT RCP DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc7", - "EventName": "FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE", + "EventName": "FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE", "PEBScounters": "0,1,2,3,4,5,6,7", + "PublicDescription": "Number of SSE/AVX computational 256-bit packed single precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 8 computation operations, one for each element. Applies to SSE* and AVX* packed single precision floating-point instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX SQRT RSQRT RCP DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.", "SampleAfterValue": "100003", - "UMask": "0x4" + "UMask": "0x20" }, { - "BriefDescription": "Counts number of SSE/AVX computational 256-bit packed single precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 8 computation operations, one for each element. Applies to SSE* and AVX* packed single precision floating-point instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX SQRT RSQRT RCP DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element.", + "BriefDescription": "Counts number of SSE/AVX computational 512-bit packed double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 8 computation operations, one for each element. Applies to SSE* and AVX* packed double precision floating-point instructions: ADD SUB MUL DIV MIN MAX SQRT RSQRT14 RCP14 FM(N)ADD/SUB. FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc7", - "EventName": "FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE", + "EventName": "FP_ARITH_INST_RETIRED.512B_PACKED_DOUBLE", "PEBScounters": "0,1,2,3,4,5,6,7", + "PublicDescription": "Number of SSE/AVX computational 512-bit packed double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 8 computation operations, one for each element. Applies to SSE* and AVX* packed double precision floating-point instructions: ADD SUB MUL DIV MIN MAX SQRT RSQRT14 RCP14 FM(N)ADD/SUB. FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.", "SampleAfterValue": "100003", - "UMask": "0x20" + "UMask": "0x40" }, { - "BriefDescription": "Counts number of SSE/AVX computational scalar single precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 1 computational operation. Applies to SSE* and AVX* scalar single precision floating-point instructions: ADD SUB MUL DIV MIN MAX SQRT RSQRT RCP FM(N)ADD/SUB. FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element.", + "BriefDescription": "Counts number of SSE/AVX computational 512-bit packed single precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 16 computation operations, one for each element. Applies to SSE* and AVX* packed single precision floating-point instructions: ADD SUB MUL DIV MIN MAX SQRT RSQRT14 RCP14 FM(N)ADD/SUB. FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc7", - "EventName": "FP_ARITH_INST_RETIRED.SCALAR_SINGLE", + "EventName": "FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE", "PEBScounters": "0,1,2,3,4,5,6,7", + "PublicDescription": "Number of SSE/AVX computational 512-bit packed single precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 16 computation operations, one for each element. Applies to SSE* and AVX* packed single precision floating-point instructions: ADD SUB MUL DIV MIN MAX SQRT RSQRT14 RCP14 FM(N)ADD/SUB. FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.", "SampleAfterValue": "100003", - "UMask": "0x2" + "UMask": "0x80" }, { - "BriefDescription": "Counts number of SSE/AVX computational 256-bit packed double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 4 computation operations, one for each element. Applies to SSE* and AVX* packed double precision floating-point instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX SQRT FM(N)ADD/SUB. FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element.", + "BriefDescription": "Counts number of SSE/AVX computational scalar double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 1 computational operation. Applies to SSE* and AVX* scalar double precision floating-point instructions: ADD SUB MUL DIV MIN MAX SQRT FM(N)ADD/SUB. FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc7", - "EventName": "FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE", + "EventName": "FP_ARITH_INST_RETIRED.SCALAR_DOUBLE", "PEBScounters": "0,1,2,3,4,5,6,7", + "PublicDescription": "Number of SSE/AVX computational scalar double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 1 computational operation. Applies to SSE* and AVX* scalar double precision floating-point instructions: ADD SUB MUL DIV MIN MAX SQRT FM(N)ADD/SUB. FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.", "SampleAfterValue": "100003", - "UMask": "0x10" + "UMask": "0x1" }, { - "BriefDescription": "Counts all microcode FP assists.", + "BriefDescription": "Counts number of SSE/AVX computational scalar single precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 1 computational operation. Applies to SSE* and AVX* scalar single precision floating-point instructions: ADD SUB MUL DIV MIN MAX SQRT RSQRT RCP FM(N)ADD/SUB. FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xc1", - "EventName": "ASSISTS.FP", + "EventCode": "0xc7", + "EventName": "FP_ARITH_INST_RETIRED.SCALAR_SINGLE", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts all microcode Floating Point assists.", + "PublicDescription": "Number of SSE/AVX computational scalar single precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 1 computational operation. Applies to SSE* and AVX* scalar single precision floating-point instructions: ADD SUB MUL DIV MIN MAX SQRT RSQRT RCP FM(N)ADD/SUB. FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.", "SampleAfterValue": "100003", - "Speculative": "1", "UMask": "0x2" } ]
\ No newline at end of file diff --git a/tools/perf/pmu-events/arch/x86/icelake/frontend.json b/tools/perf/pmu-events/arch/x86/icelake/frontend.json index 4fa2a4186ee3..b510dd5d80da 100644 --- a/tools/perf/pmu-events/arch/x86/icelake/frontend.json +++ b/tools/perf/pmu-events/arch/x86/icelake/frontend.json @@ -12,13 +12,39 @@ "UMask": "0x1" }, { + "BriefDescription": "Decode Stream Buffer (DSB)-to-MITE transitions count.", + "CollectPEBSRecord": "2", + "Counter": "0,1,2,3", + "CounterMask": "1", + "EdgeDetect": "1", + "EventCode": "0xab", + "EventName": "DSB2MITE_SWITCHES.COUNT", + "PEBScounters": "0,1,2,3", + "PublicDescription": "Counts the number of Decode Stream Buffer (DSB a.k.a. Uop Cache)-to-MITE speculative transitions.", + "SampleAfterValue": "100003", + "Speculative": "1", + "UMask": "0x2" + }, + { + "BriefDescription": "DSB-to-MITE switch true penalty cycles.", + "CollectPEBSRecord": "2", + "Counter": "0,1,2,3", + "EventCode": "0xab", + "EventName": "DSB2MITE_SWITCHES.PENALTY_CYCLES", + "PEBScounters": "0,1,2,3", + "PublicDescription": "Decode Stream Buffer (DSB) is a Uop-cache that holds translations of previously fetched instructions that were decoded by the legacy x86 decode pipeline (MITE). This event counts fetch penalty cycles when a transition occurs from DSB to MITE.", + "SampleAfterValue": "100003", + "Speculative": "1", + "UMask": "0x2" + }, + { "BriefDescription": "Retired Instructions who experienced DSB miss.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc6", - "EventName": "FRONTEND_RETIRED.DSB_MISS", + "EventName": "FRONTEND_RETIRED.ANY_DSB_MISS", "MSRIndex": "0x3F7", - "MSRValue": "0x11", + "MSRValue": "0x1", "PEBS": "1", "PEBScounters": "0,1,2,3,4,5,6,7", "PublicDescription": "Counts retired Instructions that experienced DSB (Decode stream buffer i.e. the decoded instruction-cache) miss.", @@ -27,17 +53,19 @@ "UMask": "0x1" }, { - "BriefDescription": "Cycles MITE is delivering optimal number of Uops", + "BriefDescription": "Retired Instructions who experienced a critical DSB miss.", "CollectPEBSRecord": "2", - "Counter": "0,1,2,3", - "CounterMask": "5", - "EventCode": "0x79", - "EventName": "IDQ.MITE_CYCLES_OK", - "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts the number of cycles where optimal number of uops was delivered to the Instruction Decode Queue (IDQ) from the MITE (legacy decode pipeline) path. During these cycles uops are not being delivered from the Decode Stream Buffer (DSB).", - "SampleAfterValue": "2000003", - "Speculative": "1", - "UMask": "0x4" + "Counter": "0,1,2,3,4,5,6,7", + "EventCode": "0xc6", + "EventName": "FRONTEND_RETIRED.DSB_MISS", + "MSRIndex": "0x3F7", + "MSRValue": "0x11", + "PEBS": "1", + "PEBScounters": "0,1,2,3,4,5,6,7", + "PublicDescription": "Number of retired Instructions that experienced a critical DSB (Decode stream buffer i.e. the decoded instruction-cache) miss. Critical means stalls were exposed to the back-end as a result of the DSB miss.", + "SampleAfterValue": "100007", + "TakenAlone": "1", + "UMask": "0x1" }, { "BriefDescription": "Retired Instructions who experienced iTLB true miss.", @@ -55,156 +83,81 @@ "UMask": "0x1" }, { - "BriefDescription": "Cycles when no uops are not delivered by the IDQ when backend of the machine is not stalled", - "CollectPEBSRecord": "2", - "Counter": "0,1,2,3,4,5,6,7", - "CounterMask": "5", - "EventCode": "0x9c", - "EventName": "IDQ_UOPS_NOT_DELIVERED.CYCLES_0_UOPS_DELIV.CORE", - "PEBScounters": "0,1,2,3,4,5,6,7", - "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.", - "SampleAfterValue": "1000003", - "Speculative": "1", - "UMask": "0x1" - }, - { - "BriefDescription": "Cycles where a code fetch is stalled due to L1 instruction cache miss.", - "CollectPEBSRecord": "2", - "Counter": "0,1,2,3", - "EventCode": "0x80", - "EventName": "ICACHE_16B.IFDATA_STALL", - "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts cycles where a code line fetch is stalled due to an L1 instruction cache miss. The legacy decode pipeline works at a 16 Byte granularity.", - "SampleAfterValue": "500009", - "Speculative": "1", - "UMask": "0x4" - }, - { - "BriefDescription": "Retired instructions that are fetched after an interval where the front-end delivered no uops for a period of 256 cycles which was not interrupted by a back-end stall.", + "BriefDescription": "Retired Instructions who experienced Instruction L1 Cache true miss.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc6", - "EventName": "FRONTEND_RETIRED.LATENCY_GE_256", + "EventName": "FRONTEND_RETIRED.L1I_MISS", "MSRIndex": "0x3F7", - "MSRValue": "0x510006", + "MSRValue": "0x12", "PEBS": "1", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts retired instructions that are fetched after an interval where the front-end delivered no uops for a period of 256 cycles which was not interrupted by a back-end stall.", + "PublicDescription": "Counts retired Instructions who experienced Instruction L1 Cache true miss.", "SampleAfterValue": "100007", "TakenAlone": "1", "UMask": "0x1" }, { - "BriefDescription": "Cycles Decode Stream Buffer (DSB) is delivering any Uop", - "CollectPEBSRecord": "2", - "Counter": "0,1,2,3", - "CounterMask": "1", - "EventCode": "0x79", - "EventName": "IDQ.DSB_CYCLES_ANY", - "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts the number of cycles uops were delivered to Instruction Decode Queue (IDQ) from the Decode Stream Buffer (DSB) path.", - "SampleAfterValue": "2000003", - "Speculative": "1", - "UMask": "0x8" - }, - { - "BriefDescription": "Retired instructions that are fetched after an interval where the front-end had at least 1 bubble-slot for a period of 2 cycles which was not interrupted by a back-end stall.", + "BriefDescription": "Retired Instructions who experienced Instruction L2 Cache true miss.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc6", - "EventName": "FRONTEND_RETIRED.LATENCY_GE_2_BUBBLES_GE_1", + "EventName": "FRONTEND_RETIRED.L2_MISS", "MSRIndex": "0x3F7", - "MSRValue": "0x100206", + "MSRValue": "0x13", "PEBS": "1", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts retired instructions that are delivered to the back-end after the front-end had at least 1 bubble-slot for a period of 2 cycles. A bubble-slot is an empty issue-pipeline slot while there was no RAT stall.", + "PublicDescription": "Counts retired Instructions who experienced Instruction L2 Cache true miss.", "SampleAfterValue": "100007", "TakenAlone": "1", "UMask": "0x1" }, { - "BriefDescription": "DSB-to-MITE switch true penalty cycles.", - "CollectPEBSRecord": "2", - "Counter": "0,1,2,3", - "EventCode": "0xab", - "EventName": "DSB2MITE_SWITCHES.PENALTY_CYCLES", - "PEBScounters": "0,1,2,3", - "PublicDescription": "Decode Stream Buffer (DSB) is a Uop-cache that holds translations of previously fetched instructions that were decoded by the legacy x86 decode pipeline (MITE). This event counts fetch penalty cycles when a transition occurs from DSB to MITE.", - "SampleAfterValue": "100003", - "Speculative": "1", - "UMask": "0x2" - }, - { - "BriefDescription": "Retired Instructions who experienced STLB (2nd level TLB) true miss.", + "BriefDescription": "Retired instructions after front-end starvation of at least 1 cycle", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc6", - "EventName": "FRONTEND_RETIRED.STLB_MISS", + "EventName": "FRONTEND_RETIRED.LATENCY_GE_1", "MSRIndex": "0x3F7", - "MSRValue": "0x15", + "MSRValue": "0x500106", "PEBS": "1", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts retired Instructions that experienced STLB (2nd level TLB) true miss.", + "PublicDescription": "Retired instructions that are fetched after an interval where the front-end delivered no uops for a period of at least 1 cycle which was not interrupted by a back-end stall.", "SampleAfterValue": "100007", "TakenAlone": "1", "UMask": "0x1" }, { - "BriefDescription": "Uops delivered to Instruction Decode Queue (IDQ) from MITE path", - "CollectPEBSRecord": "2", - "Counter": "0,1,2,3", - "EventCode": "0x79", - "EventName": "IDQ.MITE_UOPS", - "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts the number of uops delivered to Instruction Decode Queue (IDQ) from the MITE path. This also means that uops are not being delivered from the Decode Stream Buffer (DSB).", - "SampleAfterValue": "2000003", - "Speculative": "1", - "UMask": "0x4" - }, - { - "BriefDescription": "Retired instructions that are fetched after an interval where the front-end delivered no uops for a period of 64 cycles which was not interrupted by a back-end stall.", + "BriefDescription": "Retired instructions that are fetched after an interval where the front-end delivered no uops for a period of 128 cycles which was not interrupted by a back-end stall.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc6", - "EventName": "FRONTEND_RETIRED.LATENCY_GE_64", + "EventName": "FRONTEND_RETIRED.LATENCY_GE_128", "MSRIndex": "0x3F7", - "MSRValue": "0x504006", + "MSRValue": "0x508006", "PEBS": "1", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts retired instructions that are fetched after an interval where the front-end delivered no uops for a period of 64 cycles which was not interrupted by a back-end stall.", + "PublicDescription": "Counts retired instructions that are fetched after an interval where the front-end delivered no uops for a period of 128 cycles which was not interrupted by a back-end stall.", "SampleAfterValue": "100007", "TakenAlone": "1", "UMask": "0x1" }, { - "BriefDescription": "Retired instructions that are fetched after an interval where the front-end delivered no uops for a period of 32 cycles which was not interrupted by a back-end stall.", + "BriefDescription": "Retired instructions that are fetched after an interval where the front-end delivered no uops for a period of 16 cycles which was not interrupted by a back-end stall.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc6", - "EventName": "FRONTEND_RETIRED.LATENCY_GE_32", + "EventName": "FRONTEND_RETIRED.LATENCY_GE_16", "MSRIndex": "0x3F7", - "MSRValue": "0x502006", + "MSRValue": "0x501006", "PEBS": "1", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts retired instructions that are delivered to the back-end after a front-end stall of at least 32 cycles. During this period the front-end delivered no uops.", + "PublicDescription": "Counts retired instructions that are delivered to the back-end after a front-end stall of at least 16 cycles. During this period the front-end delivered no uops.", "SampleAfterValue": "100007", "TakenAlone": "1", "UMask": "0x1" }, { - "BriefDescription": "Cycles MITE is delivering any Uop", - "CollectPEBSRecord": "2", - "Counter": "0,1,2,3", - "CounterMask": "1", - "EventCode": "0x79", - "EventName": "IDQ.MITE_CYCLES_ANY", - "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts the number of cycles uops were delivered to the Instruction Decode Queue (IDQ) from the MITE (legacy decode pipeline) path. During these cycles uops are not being delivered from the Decode Stream Buffer (DSB).", - "SampleAfterValue": "2000003", - "Speculative": "1", - "UMask": "0x4" - }, - { "BriefDescription": "Retired instructions after front-end starvation of at least 2 cycles", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", @@ -220,113 +173,91 @@ "UMask": "0x1" }, { - "BriefDescription": "Decode Stream Buffer (DSB)-to-MITE transitions count.", - "CollectPEBSRecord": "2", - "Counter": "0,1,2,3", - "CounterMask": "1", - "EdgeDetect": "1", - "EventCode": "0xab", - "EventName": "DSB2MITE_SWITCHES.COUNT", - "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts the number of Decode Stream Buffer (DSB a.k.a. Uop Cache)-to-MITE speculative transitions.", - "SampleAfterValue": "100003", - "Speculative": "1", - "UMask": "0x2" - }, - { - "BriefDescription": "Uops delivered to Instruction Decode Queue (IDQ) from the Decode Stream Buffer (DSB) path", - "CollectPEBSRecord": "2", - "Counter": "0,1,2,3", - "EventCode": "0x79", - "EventName": "IDQ.DSB_UOPS", - "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts the number of uops delivered to Instruction Decode Queue (IDQ) from the Decode Stream Buffer (DSB) path.", - "SampleAfterValue": "2000003", - "Speculative": "1", - "UMask": "0x8" - }, - { - "BriefDescription": "Retired Instructions who experienced Instruction L2 Cache true miss.", + "BriefDescription": "Retired instructions that are fetched after an interval where the front-end delivered no uops for a period of 256 cycles which was not interrupted by a back-end stall.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc6", - "EventName": "FRONTEND_RETIRED.L2_MISS", + "EventName": "FRONTEND_RETIRED.LATENCY_GE_256", "MSRIndex": "0x3F7", - "MSRValue": "0x13", + "MSRValue": "0x510006", "PEBS": "1", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts retired Instructions who experienced Instruction L2 Cache true miss.", + "PublicDescription": "Counts retired instructions that are fetched after an interval where the front-end delivered no uops for a period of 256 cycles which was not interrupted by a back-end stall.", "SampleAfterValue": "100007", "TakenAlone": "1", "UMask": "0x1" }, { - "BriefDescription": "Instruction fetch tag lookups that hit in the instruction cache (L1I). Counts at 64-byte cache-line granularity.", + "BriefDescription": "Retired instructions that are fetched after an interval where the front-end had at least 1 bubble-slot for a period of 2 cycles which was not interrupted by a back-end stall.", "CollectPEBSRecord": "2", - "Counter": "0,1,2,3", - "EventCode": "0x83", - "EventName": "ICACHE_64B.IFTAG_HIT", - "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts instruction fetch tag lookups that hit in the instruction cache (L1I). Counts at 64-byte cache-line granularity. Accounts for both cacheable and uncacheable accesses.", - "SampleAfterValue": "200003", - "Speculative": "1", + "Counter": "0,1,2,3,4,5,6,7", + "EventCode": "0xc6", + "EventName": "FRONTEND_RETIRED.LATENCY_GE_2_BUBBLES_GE_1", + "MSRIndex": "0x3F7", + "MSRValue": "0x100206", + "PEBS": "1", + "PEBScounters": "0,1,2,3,4,5,6,7", + "PublicDescription": "Counts retired instructions that are delivered to the back-end after the front-end had at least 1 bubble-slot for a period of 2 cycles. A bubble-slot is an empty issue-pipeline slot while there was no RAT stall.", + "SampleAfterValue": "100007", + "TakenAlone": "1", "UMask": "0x1" }, { - "BriefDescription": "Retired instructions that are fetched after an interval where the front-end delivered no uops for a period of 512 cycles which was not interrupted by a back-end stall.", + "BriefDescription": "Retired instructions that are fetched after an interval where the front-end delivered no uops for a period of 32 cycles which was not interrupted by a back-end stall.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc6", - "EventName": "FRONTEND_RETIRED.LATENCY_GE_512", + "EventName": "FRONTEND_RETIRED.LATENCY_GE_32", "MSRIndex": "0x3F7", - "MSRValue": "0x520006", + "MSRValue": "0x502006", "PEBS": "1", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts retired instructions that are fetched after an interval where the front-end delivered no uops for a period of 512 cycles which was not interrupted by a back-end stall.", + "PublicDescription": "Counts retired instructions that are delivered to the back-end after a front-end stall of at least 32 cycles. During this period the front-end delivered no uops.", "SampleAfterValue": "100007", "TakenAlone": "1", "UMask": "0x1" }, { - "BriefDescription": "Cycles when optimal number of uops was delivered to the back-end when the back-end is not stalled", + "BriefDescription": "Retired instructions that are fetched after an interval where the front-end delivered no uops for a period of 4 cycles which was not interrupted by a back-end stall.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "CounterMask": "1", - "EventCode": "0x9C", - "EventName": "IDQ_UOPS_NOT_DELIVERED.CYCLES_FE_WAS_OK", - "Invert": "1", + "EventCode": "0xc6", + "EventName": "FRONTEND_RETIRED.LATENCY_GE_4", + "MSRIndex": "0x3F7", + "MSRValue": "0x500406", + "PEBS": "1", "PEBScounters": "0,1,2,3,4,5,6,7", - "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.", - "SampleAfterValue": "1000003", - "Speculative": "1", + "PublicDescription": "Counts retired instructions that are fetched after an interval where the front-end delivered no uops for a period of 4 cycles which was not interrupted by a back-end stall.", + "SampleAfterValue": "100007", + "TakenAlone": "1", "UMask": "0x1" }, { - "BriefDescription": "Retired instructions that are fetched after an interval where the front-end delivered no uops for a period of 16 cycles which was not interrupted by a back-end stall.", + "BriefDescription": "Retired instructions that are fetched after an interval where the front-end delivered no uops for a period of 512 cycles which was not interrupted by a back-end stall.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc6", - "EventName": "FRONTEND_RETIRED.LATENCY_GE_16", + "EventName": "FRONTEND_RETIRED.LATENCY_GE_512", "MSRIndex": "0x3F7", - "MSRValue": "0x501006", + "MSRValue": "0x520006", "PEBS": "1", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts retired instructions that are delivered to the back-end after a front-end stall of at least 16 cycles. During this period the front-end delivered no uops.", + "PublicDescription": "Counts retired instructions that are fetched after an interval where the front-end delivered no uops for a period of 512 cycles which was not interrupted by a back-end stall.", "SampleAfterValue": "100007", "TakenAlone": "1", "UMask": "0x1" }, { - "BriefDescription": "Retired instructions that are fetched after an interval where the front-end delivered no uops for a period of 128 cycles which was not interrupted by a back-end stall.", + "BriefDescription": "Retired instructions that are fetched after an interval where the front-end delivered no uops for a period of 64 cycles which was not interrupted by a back-end stall.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc6", - "EventName": "FRONTEND_RETIRED.LATENCY_GE_128", + "EventName": "FRONTEND_RETIRED.LATENCY_GE_64", "MSRIndex": "0x3F7", - "MSRValue": "0x508006", + "MSRValue": "0x504006", "PEBS": "1", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts retired instructions that are fetched after an interval where the front-end delivered no uops for a period of 128 cycles which was not interrupted by a back-end stall.", + "PublicDescription": "Counts retired instructions that are fetched after an interval where the front-end delivered no uops for a period of 64 cycles which was not interrupted by a back-end stall.", "SampleAfterValue": "100007", "TakenAlone": "1", "UMask": "0x1" @@ -347,48 +278,55 @@ "UMask": "0x1" }, { - "BriefDescription": "Retired instructions after front-end starvation of at least 1 cycle", + "BriefDescription": "Retired Instructions who experienced STLB (2nd level TLB) true miss.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc6", - "EventName": "FRONTEND_RETIRED.LATENCY_GE_1", + "EventName": "FRONTEND_RETIRED.STLB_MISS", "MSRIndex": "0x3F7", - "MSRValue": "0x500106", + "MSRValue": "0x15", "PEBS": "1", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Retired instructions that are fetched after an interval where the front-end delivered no uops for a period of at least 1 cycle which was not interrupted by a back-end stall.", + "PublicDescription": "Counts retired Instructions that experienced STLB (2nd level TLB) true miss.", "SampleAfterValue": "100007", "TakenAlone": "1", "UMask": "0x1" }, { - "BriefDescription": "Retired instructions that are fetched after an interval where the front-end delivered no uops for a period of 4 cycles which was not interrupted by a back-end stall.", + "BriefDescription": "Cycles where a code fetch is stalled due to L1 instruction cache miss.", "CollectPEBSRecord": "2", - "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xc6", - "EventName": "FRONTEND_RETIRED.LATENCY_GE_4", - "MSRIndex": "0x3F7", - "MSRValue": "0x500406", - "PEBS": "1", - "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts retired instructions that are fetched after an interval where the front-end delivered no uops for a period of 4 cycles which was not interrupted by a back-end stall.", - "SampleAfterValue": "100007", - "TakenAlone": "1", + "Counter": "0,1,2,3", + "EventCode": "0x80", + "EventName": "ICACHE_16B.IFDATA_STALL", + "PEBScounters": "0,1,2,3", + "PublicDescription": "Counts cycles where a code line fetch is stalled due to an L1 instruction cache miss. The legacy decode pipeline works at a 16 Byte granularity.", + "SampleAfterValue": "500009", + "Speculative": "1", + "UMask": "0x4" + }, + { + "BriefDescription": "Instruction fetch tag lookups that hit in the instruction cache (L1I). Counts at 64-byte cache-line granularity.", + "CollectPEBSRecord": "2", + "Counter": "0,1,2,3", + "EventCode": "0x83", + "EventName": "ICACHE_64B.IFTAG_HIT", + "PEBScounters": "0,1,2,3", + "PublicDescription": "Counts instruction fetch tag lookups that hit in the instruction cache (L1I). Counts at 64-byte cache-line granularity. Accounts for both cacheable and uncacheable accesses.", + "SampleAfterValue": "200003", + "Speculative": "1", "UMask": "0x1" }, { - "BriefDescription": "Number of switches from DSB or MITE to the MS", + "BriefDescription": "Instruction fetch tag lookups that miss in the instruction cache (L1I). Counts at 64-byte cache-line granularity.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "CounterMask": "1", - "EdgeDetect": "1", - "EventCode": "0x79", - "EventName": "IDQ.MS_SWITCHES", + "EventCode": "0x83", + "EventName": "ICACHE_64B.IFTAG_MISS", "PEBScounters": "0,1,2,3", - "PublicDescription": "Number of switches from DSB (Decode Stream Buffer) or MITE (legacy decode pipeline) to the Microcode Sequencer.", - "SampleAfterValue": "100003", + "PublicDescription": "Counts instruction fetch tag lookups that miss in the instruction cache (L1I). Counts at 64-byte cache-line granularity. Accounts for both cacheable and uncacheable accesses.", + "SampleAfterValue": "200003", "Speculative": "1", - "UMask": "0x30" + "UMask": "0x2" }, { "BriefDescription": "Cycles where a code fetch is stalled due to L1 instruction cache tag miss.", @@ -403,28 +341,80 @@ "UMask": "0x4" }, { - "BriefDescription": "Uops delivered to IDQ while MS is busy", + "BriefDescription": "Cycles Decode Stream Buffer (DSB) is delivering any Uop", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", + "CounterMask": "1", "EventCode": "0x79", - "EventName": "IDQ.MS_UOPS", + "EventName": "IDQ.DSB_CYCLES_ANY", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts the total number of uops delivered by the Microcode Sequencer (MS). Any instruction over 4 uops will be delivered by the MS. Some instructions such as transcendentals may additionally generate uops from the MS.", - "SampleAfterValue": "100003", + "PublicDescription": "Counts the number of cycles uops were delivered to Instruction Decode Queue (IDQ) from the Decode Stream Buffer (DSB) path.", + "SampleAfterValue": "2000003", "Speculative": "1", - "UMask": "0x30" + "UMask": "0x8" }, { - "BriefDescription": "Instruction fetch tag lookups that miss in the instruction cache (L1I). Counts at 64-byte cache-line granularity.", + "BriefDescription": "Cycles DSB is delivering optimal number of Uops", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0x83", - "EventName": "ICACHE_64B.IFTAG_MISS", + "CounterMask": "5", + "EventCode": "0x79", + "EventName": "IDQ.DSB_CYCLES_OK", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts instruction fetch tag lookups that miss in the instruction cache (L1I). Counts at 64-byte cache-line granularity. Accounts for both cacheable and uncacheable accesses.", - "SampleAfterValue": "200003", + "PublicDescription": "Counts the number of cycles where optimal number of uops was delivered to the Instruction Decode Queue (IDQ) from the MITE (legacy decode pipeline) path. During these cycles uops are not being delivered from the Decode Stream Buffer (DSB).", + "SampleAfterValue": "2000003", "Speculative": "1", - "UMask": "0x2" + "UMask": "0x8" + }, + { + "BriefDescription": "Uops delivered to Instruction Decode Queue (IDQ) from the Decode Stream Buffer (DSB) path", + "CollectPEBSRecord": "2", + "Counter": "0,1,2,3", + "EventCode": "0x79", + "EventName": "IDQ.DSB_UOPS", + "PEBScounters": "0,1,2,3", + "PublicDescription": "Counts the number of uops delivered to Instruction Decode Queue (IDQ) from the Decode Stream Buffer (DSB) path.", + "SampleAfterValue": "2000003", + "Speculative": "1", + "UMask": "0x8" + }, + { + "BriefDescription": "Cycles MITE is delivering any Uop", + "CollectPEBSRecord": "2", + "Counter": "0,1,2,3", + "CounterMask": "1", + "EventCode": "0x79", + "EventName": "IDQ.MITE_CYCLES_ANY", + "PEBScounters": "0,1,2,3", + "PublicDescription": "Counts the number of cycles uops were delivered to the Instruction Decode Queue (IDQ) from the MITE (legacy decode pipeline) path. During these cycles uops are not being delivered from the Decode Stream Buffer (DSB).", + "SampleAfterValue": "2000003", + "Speculative": "1", + "UMask": "0x4" + }, + { + "BriefDescription": "Cycles MITE is delivering optimal number of Uops", + "CollectPEBSRecord": "2", + "Counter": "0,1,2,3", + "CounterMask": "5", + "EventCode": "0x79", + "EventName": "IDQ.MITE_CYCLES_OK", + "PEBScounters": "0,1,2,3", + "PublicDescription": "Counts the number of cycles where optimal number of uops was delivered to the Instruction Decode Queue (IDQ) from the MITE (legacy decode pipeline) path. During these cycles uops are not being delivered from the Decode Stream Buffer (DSB).", + "SampleAfterValue": "2000003", + "Speculative": "1", + "UMask": "0x4" + }, + { + "BriefDescription": "Uops delivered to Instruction Decode Queue (IDQ) from MITE path", + "CollectPEBSRecord": "2", + "Counter": "0,1,2,3", + "EventCode": "0x79", + "EventName": "IDQ.MITE_UOPS", + "PEBScounters": "0,1,2,3", + "PublicDescription": "Counts the number of uops delivered to Instruction Decode Queue (IDQ) from the MITE path. This also means that uops are not being delivered from the Decode Stream Buffer (DSB).", + "SampleAfterValue": "2000003", + "Speculative": "1", + "UMask": "0x4" }, { "BriefDescription": "Cycles when uops are being delivered to IDQ while MS is busy", @@ -440,32 +430,30 @@ "UMask": "0x30" }, { - "BriefDescription": "Retired Instructions who experienced Instruction L1 Cache true miss.", + "BriefDescription": "Number of switches from DSB or MITE to the MS", "CollectPEBSRecord": "2", - "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xc6", - "EventName": "FRONTEND_RETIRED.L1I_MISS", - "MSRIndex": "0x3F7", - "MSRValue": "0x12", - "PEBS": "1", - "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts retired Instructions who experienced Instruction L1 Cache true miss.", - "SampleAfterValue": "100007", - "TakenAlone": "1", - "UMask": "0x1" + "Counter": "0,1,2,3", + "CounterMask": "1", + "EdgeDetect": "1", + "EventCode": "0x79", + "EventName": "IDQ.MS_SWITCHES", + "PEBScounters": "0,1,2,3", + "PublicDescription": "Number of switches from DSB (Decode Stream Buffer) or MITE (legacy decode pipeline) to the Microcode Sequencer.", + "SampleAfterValue": "100003", + "Speculative": "1", + "UMask": "0x30" }, { - "BriefDescription": "Cycles DSB is delivering optimal number of Uops", + "BriefDescription": "Uops delivered to IDQ while MS is busy", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "CounterMask": "5", "EventCode": "0x79", - "EventName": "IDQ.DSB_CYCLES_OK", + "EventName": "IDQ.MS_UOPS", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts the number of cycles where optimal number of uops was delivered to the Instruction Decode Queue (IDQ) from the MITE (legacy decode pipeline) path. During these cycles uops are not being delivered from the Decode Stream Buffer (DSB).", - "SampleAfterValue": "2000003", + "PublicDescription": "Counts the total number of uops delivered by the Microcode Sequencer (MS). Any instruction over 4 uops will be delivered by the MS. Some instructions such as transcendentals may additionally generate uops from the MS.", + "SampleAfterValue": "100003", "Speculative": "1", - "UMask": "0x8" + "UMask": "0x30" }, { "BriefDescription": "Uops not delivered by IDQ when backend of the machine is not stalled", @@ -478,5 +466,32 @@ "SampleAfterValue": "1000003", "Speculative": "1", "UMask": "0x1" + }, + { + "BriefDescription": "Cycles when no uops are not delivered by the IDQ when backend of the machine is not stalled", + "CollectPEBSRecord": "2", + "Counter": "0,1,2,3,4,5,6,7", + "CounterMask": "5", + "EventCode": "0x9c", + "EventName": "IDQ_UOPS_NOT_DELIVERED.CYCLES_0_UOPS_DELIV.CORE", + "PEBScounters": "0,1,2,3,4,5,6,7", + "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.", + "SampleAfterValue": "1000003", + "Speculative": "1", + "UMask": "0x1" + }, + { + "BriefDescription": "Cycles when optimal number of uops was delivered to the back-end when the back-end is not stalled", + "CollectPEBSRecord": "2", + "Counter": "0,1,2,3,4,5,6,7", + "CounterMask": "1", + "EventCode": "0x9C", + "EventName": "IDQ_UOPS_NOT_DELIVERED.CYCLES_FE_WAS_OK", + "Invert": "1", + "PEBScounters": "0,1,2,3,4,5,6,7", + "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.", + "SampleAfterValue": "1000003", + "Speculative": "1", + "UMask": "0x1" } ]
\ No newline at end of file diff --git a/tools/perf/pmu-events/arch/x86/icelake/icl-metrics.json b/tools/perf/pmu-events/arch/x86/icelake/icl-metrics.json index 432e45ac6814..4af23c04dc18 100644 --- a/tools/perf/pmu-events/arch/x86/icelake/icl-metrics.json +++ b/tools/perf/pmu-events/arch/x86/icelake/icl-metrics.json @@ -1,272 +1,452 @@ [ { - "MetricExpr": "INST_RETIRED.ANY / CPU_CLK_UNHALTED.THREAD", + "BriefDescription": "Total pipeline cost of branch related instructions (used for program control-flow including function calls)", + "MetricExpr": "100 * (( BR_INST_RETIRED.COND + 3 * BR_INST_RETIRED.NEAR_CALL + (BR_INST_RETIRED.NEAR_TAKEN - BR_INST_RETIRED.COND_TAKEN - 2 * BR_INST_RETIRED.NEAR_CALL) ) / TOPDOWN.SLOTS)", + "MetricGroup": "Ret", + "MetricName": "Branching_Overhead" + }, + { + "BriefDescription": "Total pipeline cost of instruction fetch related bottlenecks by large code footprint programs (i-side cache; TLB and BTB misses)", + "MetricExpr": "100 * (( 5 * IDQ_UOPS_NOT_DELIVERED.CYCLES_0_UOPS_DELIV.CORE - INT_MISC.UOP_DROPPING ) / TOPDOWN.SLOTS) * ( (ICACHE_64B.IFTAG_STALL / CPU_CLK_UNHALTED.THREAD) + (ICACHE_16B.IFDATA_STALL / CPU_CLK_UNHALTED.THREAD) + (10 * BACLEARS.ANY / CPU_CLK_UNHALTED.THREAD) ) / #(( 5 * IDQ_UOPS_NOT_DELIVERED.CYCLES_0_UOPS_DELIV.CORE - INT_MISC.UOP_DROPPING ) / TOPDOWN.SLOTS)", + "MetricGroup": "BigFoot;Fed;Frontend;IcMiss;MemoryTLB", + "MetricName": "Big_Code" + }, + { "BriefDescription": "Instructions Per Cycle (per Logical Processor)", - "MetricGroup": "Summary", + "MetricExpr": "INST_RETIRED.ANY / CPU_CLK_UNHALTED.THREAD", + "MetricGroup": "Ret;Summary", "MetricName": "IPC" }, { - "MetricExpr": "UOPS_RETIRED.SLOTS / INST_RETIRED.ANY", "BriefDescription": "Uops Per Instruction", - "MetricGroup": "Pipeline;Retire", + "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / INST_RETIRED.ANY", + "MetricGroup": "Pipeline;Ret;Retire", "MetricName": "UPI" }, { - "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.NEAR_TAKEN", "BriefDescription": "Instruction per taken branch", - "MetricGroup": "Branches;FetchBW;PGO", - "MetricName": "IpTB" + "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / BR_INST_RETIRED.NEAR_TAKEN", + "MetricGroup": "Branches;Fed;FetchBW", + "MetricName": "UpTB" }, { - "MetricExpr": "1 / (INST_RETIRED.ANY / CPU_CLK_UNHALTED.THREAD)", "BriefDescription": "Cycles Per Instruction (per Logical Processor)", - "MetricGroup": "Pipeline", + "MetricExpr": "1 / (INST_RETIRED.ANY / CPU_CLK_UNHALTED.THREAD)", + "MetricGroup": "Pipeline;Mem", "MetricName": "CPI" }, { - "MetricExpr": "CPU_CLK_UNHALTED.THREAD", "BriefDescription": "Per-Logical Processor actual clocks when the Logical Processor is active.", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD", "MetricGroup": "Pipeline", "MetricName": "CLKS" }, { + "BriefDescription": "Total issue-pipeline slots (per-Physical Core till ICL; per-Logical Processor ICL onward)", + "MetricExpr": "TOPDOWN.SLOTS", + "MetricGroup": "TmaL1", + "MetricName": "SLOTS" + }, + { + "BriefDescription": "Fraction of Physical Core issue-slots utilized by this Logical Processor", + "MetricExpr": "TOPDOWN.SLOTS / ( TOPDOWN.SLOTS / 2 ) if #SMT_on else 1", + "MetricGroup": "SMT", + "MetricName": "Slots_Utilization" + }, + { + "BriefDescription": "The ratio of Executed- by Issued-Uops", + "MetricExpr": "UOPS_EXECUTED.THREAD / UOPS_ISSUED.ANY", + "MetricGroup": "Cor;Pipeline", + "MetricName": "Execute_per_Issue", + "PublicDescription": "The ratio of Executed- by Issued-Uops. Ratio > 1 suggests high rate of uop micro-fusions. Ratio < 1 suggest high rate of \"execute\" at rename stage." + }, + { + "BriefDescription": "Instructions Per Cycle across hyper-threads (per physical core)", "MetricExpr": "INST_RETIRED.ANY / CPU_CLK_UNHALTED.DISTRIBUTED", - "BriefDescription": "Instructions Per Cycle (per physical core)", - "MetricGroup": "SMT;TmaL1", + "MetricGroup": "Ret;SMT;TmaL1", "MetricName": "CoreIPC" }, { - "MetricExpr": "( 1 * ( FP_ARITH_INST_RETIRED.SCALAR_SINGLE + FP_ARITH_INST_RETIRED.SCALAR_DOUBLE ) + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * ( FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE ) + 8 * ( FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.512B_PACKED_DOUBLE ) + 16 * FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE ) / CPU_CLK_UNHALTED.DISTRIBUTED", "BriefDescription": "Floating Point Operations Per Cycle", - "MetricGroup": "Flops", + "MetricExpr": "( 1 * ( FP_ARITH_INST_RETIRED.SCALAR_SINGLE + FP_ARITH_INST_RETIRED.SCALAR_DOUBLE ) + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * ( FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE ) + 8 * ( FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.512B_PACKED_DOUBLE ) + 16 * FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE ) / CPU_CLK_UNHALTED.DISTRIBUTED", + "MetricGroup": "Ret;Flops", "MetricName": "FLOPc" }, { - "MetricExpr": "UOPS_EXECUTED.THREAD / ( UOPS_EXECUTED.CORE_CYCLES_GE_1 / 2 )", + "BriefDescription": "Actual per-core usage of the Floating Point execution units (regardless of the vector width)", + "MetricExpr": "( (FP_ARITH_INST_RETIRED.SCALAR_SINGLE + FP_ARITH_INST_RETIRED.SCALAR_DOUBLE) + (FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.512B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE) ) / ( 2 * CPU_CLK_UNHALTED.DISTRIBUTED )", + "MetricGroup": "Cor;Flops;HPC", + "MetricName": "FP_Arith_Utilization", + "PublicDescription": "Actual per-core usage of the Floating Point execution units (regardless of the vector width). Values > 1 are possible due to Fused-Multiply Add (FMA) counting." + }, + { "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is at least 1 uop executed)", - "MetricGroup": "Pipeline;PortsUtil", + "MetricExpr": "UOPS_EXECUTED.THREAD / (( UOPS_EXECUTED.CORE_CYCLES_GE_1 / 2 ) if #SMT_on else UOPS_EXECUTED.CORE_CYCLES_GE_1)", + "MetricGroup": "Backend;Cor;Pipeline;PortsUtil", "MetricName": "ILP" }, { - "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.ALL_BRANCHES", "BriefDescription": "Number of Instructions per non-speculative Branch Misprediction (JEClear)", - "MetricGroup": "BrMispredicts", + "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.ALL_BRANCHES", + "MetricGroup": "Bad;BadSpec;BrMispredicts", "MetricName": "IpMispredict" }, { - "MetricExpr": "CPU_CLK_UNHALTED.DISTRIBUTED", "BriefDescription": "Core actual clocks when any Logical Processor is active on the Physical Core", + "MetricExpr": "CPU_CLK_UNHALTED.DISTRIBUTED", "MetricGroup": "SMT", "MetricName": "CORE_CLKS" }, { - "MetricExpr": "INST_RETIRED.ANY / MEM_INST_RETIRED.ALL_LOADS", "BriefDescription": "Instructions per Load (lower number means higher occurrence rate)", + "MetricExpr": "INST_RETIRED.ANY / MEM_INST_RETIRED.ALL_LOADS", "MetricGroup": "InsType", "MetricName": "IpLoad" }, { - "MetricExpr": "INST_RETIRED.ANY / MEM_INST_RETIRED.ALL_STORES", "BriefDescription": "Instructions per Store (lower number means higher occurrence rate)", + "MetricExpr": "INST_RETIRED.ANY / MEM_INST_RETIRED.ALL_STORES", "MetricGroup": "InsType", "MetricName": "IpStore" }, { - "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.ALL_BRANCHES", "BriefDescription": "Instructions per Branch (lower number means higher occurrence rate)", - "MetricGroup": "Branches;InsType", + "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.ALL_BRANCHES", + "MetricGroup": "Branches;Fed;InsType", "MetricName": "IpBranch" }, { - "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.NEAR_CALL", "BriefDescription": "Instructions per (near) call (lower number means higher occurrence rate)", - "MetricGroup": "Branches", + "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.NEAR_CALL", + "MetricGroup": "Branches;Fed;PGO", "MetricName": "IpCall" }, { - "MetricExpr": "BR_INST_RETIRED.ALL_BRANCHES / BR_INST_RETIRED.NEAR_TAKEN", + "BriefDescription": "Instruction per taken branch", + "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.NEAR_TAKEN", + "MetricGroup": "Branches;Fed;FetchBW;Frontend;PGO", + "MetricName": "IpTB" + }, + { "BriefDescription": "Branch instructions per taken branch. ", - "MetricGroup": "Branches;PGO", + "MetricExpr": "BR_INST_RETIRED.ALL_BRANCHES / BR_INST_RETIRED.NEAR_TAKEN", + "MetricGroup": "Branches;Fed;PGO", "MetricName": "BpTkBranch" }, { - "MetricExpr": "INST_RETIRED.ANY / ( 1 * ( FP_ARITH_INST_RETIRED.SCALAR_SINGLE + FP_ARITH_INST_RETIRED.SCALAR_DOUBLE ) + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * ( FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE ) + 8 * ( FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.512B_PACKED_DOUBLE ) + 16 * FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE )", "BriefDescription": "Instructions per Floating Point (FP) Operation (lower number means higher occurrence rate)", - "MetricGroup": "Flops;FpArith;InsType", + "MetricExpr": "INST_RETIRED.ANY / ( 1 * ( FP_ARITH_INST_RETIRED.SCALAR_SINGLE + FP_ARITH_INST_RETIRED.SCALAR_DOUBLE ) + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * ( FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE ) + 8 * ( FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.512B_PACKED_DOUBLE ) + 16 * FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE )", + "MetricGroup": "Flops;InsType", "MetricName": "IpFLOP" }, { + "BriefDescription": "Instructions per FP Arithmetic instruction (lower number means higher occurrence rate)", + "MetricExpr": "INST_RETIRED.ANY / ( (FP_ARITH_INST_RETIRED.SCALAR_SINGLE + FP_ARITH_INST_RETIRED.SCALAR_DOUBLE) + (FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.512B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE) )", + "MetricGroup": "Flops;InsType", + "MetricName": "IpArith", + "PublicDescription": "Instructions per FP Arithmetic instruction (lower number means higher occurrence rate). May undercount due to FMA double counting. Approximated prior to BDW." + }, + { + "BriefDescription": "Instructions per FP Arithmetic Scalar Single-Precision instruction (lower number means higher occurrence rate)", + "MetricExpr": "INST_RETIRED.ANY / FP_ARITH_INST_RETIRED.SCALAR_SINGLE", + "MetricGroup": "Flops;FpScalar;InsType", + "MetricName": "IpArith_Scalar_SP", + "PublicDescription": "Instructions per FP Arithmetic Scalar Single-Precision instruction (lower number means higher occurrence rate). May undercount due to FMA double counting." + }, + { + "BriefDescription": "Instructions per FP Arithmetic Scalar Double-Precision instruction (lower number means higher occurrence rate)", + "MetricExpr": "INST_RETIRED.ANY / FP_ARITH_INST_RETIRED.SCALAR_DOUBLE", + "MetricGroup": "Flops;FpScalar;InsType", + "MetricName": "IpArith_Scalar_DP", + "PublicDescription": "Instructions per FP Arithmetic Scalar Double-Precision instruction (lower number means higher occurrence rate). May undercount due to FMA double counting." + }, + { + "BriefDescription": "Instructions per FP Arithmetic AVX/SSE 128-bit instruction (lower number means higher occurrence rate)", + "MetricExpr": "INST_RETIRED.ANY / ( FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE )", + "MetricGroup": "Flops;FpVector;InsType", + "MetricName": "IpArith_AVX128", + "PublicDescription": "Instructions per FP Arithmetic AVX/SSE 128-bit instruction (lower number means higher occurrence rate). May undercount due to FMA double counting." + }, + { + "BriefDescription": "Instructions per FP Arithmetic AVX* 256-bit instruction (lower number means higher occurrence rate)", + "MetricExpr": "INST_RETIRED.ANY / ( FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE )", + "MetricGroup": "Flops;FpVector;InsType", + "MetricName": "IpArith_AVX256", + "PublicDescription": "Instructions per FP Arithmetic AVX* 256-bit instruction (lower number means higher occurrence rate). May undercount due to FMA double counting." + }, + { + "BriefDescription": "Instructions per FP Arithmetic AVX 512-bit instruction (lower number means higher occurrence rate)", + "MetricExpr": "INST_RETIRED.ANY / ( FP_ARITH_INST_RETIRED.512B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE )", + "MetricGroup": "Flops;FpVector;InsType", + "MetricName": "IpArith_AVX512", + "PublicDescription": "Instructions per FP Arithmetic AVX 512-bit instruction (lower number means higher occurrence rate). May undercount due to FMA double counting." + }, + { + "BriefDescription": "Total number of retired Instructions, Sample with: INST_RETIRED.PREC_DIST", "MetricExpr": "INST_RETIRED.ANY", - "BriefDescription": "Total number of retired Instructions", "MetricGroup": "Summary;TmaL1", "MetricName": "Instructions" }, { - "MetricExpr": "LSD.UOPS / (IDQ.DSB_UOPS + LSD.UOPS + IDQ.MITE_UOPS + IDQ.MS_UOPS)", + "BriefDescription": "Average number of Uops issued by front-end when it issued something", + "MetricExpr": "UOPS_ISSUED.ANY / cpu@UOPS_ISSUED.ANY\\,cmask\\=1@", + "MetricGroup": "Fed;FetchBW", + "MetricName": "Fetch_UpC" + }, + { "BriefDescription": "Fraction of Uops delivered by the LSD (Loop Stream Detector; aka Loop Cache)", - "MetricGroup": "LSD", + "MetricExpr": "LSD.UOPS / (IDQ.DSB_UOPS + LSD.UOPS + IDQ.MITE_UOPS + IDQ.MS_UOPS)", + "MetricGroup": "Fed;LSD", "MetricName": "LSD_Coverage" }, { - "MetricExpr": "IDQ.DSB_UOPS / (IDQ.DSB_UOPS + LSD.UOPS + IDQ.MITE_UOPS + IDQ.MS_UOPS)", "BriefDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache)", - "MetricGroup": "DSB;FetchBW", + "MetricExpr": "IDQ.DSB_UOPS / (IDQ.DSB_UOPS + LSD.UOPS + IDQ.MITE_UOPS + IDQ.MS_UOPS)", + "MetricGroup": "DSB;Fed;FetchBW", "MetricName": "DSB_Coverage" }, { + "BriefDescription": "Number of Instructions per non-speculative DSB miss", + "MetricExpr": "INST_RETIRED.ANY / FRONTEND_RETIRED.ANY_DSB_MISS", + "MetricGroup": "DSBmiss;Fed", + "MetricName": "IpDSB_Miss_Ret" + }, + { + "BriefDescription": "Fraction of branches that are non-taken conditionals", + "MetricExpr": "BR_INST_RETIRED.COND_NTAKEN / BR_INST_RETIRED.ALL_BRANCHES", + "MetricGroup": "Bad;Branches;CodeGen;PGO", + "MetricName": "Cond_NT" + }, + { + "BriefDescription": "Fraction of branches that are taken conditionals", + "MetricExpr": "BR_INST_RETIRED.COND_TAKEN / BR_INST_RETIRED.ALL_BRANCHES", + "MetricGroup": "Bad;Branches;CodeGen;PGO", + "MetricName": "Cond_TK" + }, + { + "BriefDescription": "Fraction of branches that are CALL or RET", + "MetricExpr": "( BR_INST_RETIRED.NEAR_CALL + BR_INST_RETIRED.NEAR_RETURN ) / BR_INST_RETIRED.ALL_BRANCHES", + "MetricGroup": "Bad;Branches", + "MetricName": "CallRet" + }, + { + "BriefDescription": "Fraction of branches that are unconditional (direct or indirect) jumps", + "MetricExpr": "(BR_INST_RETIRED.NEAR_TAKEN - BR_INST_RETIRED.COND_TAKEN - 2 * BR_INST_RETIRED.NEAR_CALL) / BR_INST_RETIRED.ALL_BRANCHES", + "MetricGroup": "Bad;Branches", + "MetricName": "Jump" + }, + { + "BriefDescription": "Fraction of branches of other types (not individually covered by other metrics in Info.Branches group)", + "MetricExpr": "1 - ( (BR_INST_RETIRED.COND_NTAKEN / BR_INST_RETIRED.ALL_BRANCHES) + (BR_INST_RETIRED.COND_TAKEN / BR_INST_RETIRED.ALL_BRANCHES) + (( BR_INST_RETIRED.NEAR_CALL + BR_INST_RETIRED.NEAR_RETURN ) / BR_INST_RETIRED.ALL_BRANCHES) + ((BR_INST_RETIRED.NEAR_TAKEN - BR_INST_RETIRED.COND_TAKEN - 2 * BR_INST_RETIRED.NEAR_CALL) / BR_INST_RETIRED.ALL_BRANCHES) )", + "MetricGroup": "Bad;Branches", + "MetricName": "Other_Branches" + }, + { + "BriefDescription": "Actual Average Latency for L1 data-cache miss demand load instructions (in core cycles)", "MetricExpr": "L1D_PEND_MISS.PENDING / ( MEM_LOAD_RETIRED.L1_MISS + MEM_LOAD_RETIRED.FB_HIT )", - "BriefDescription": "Actual Average Latency for L1 data-cache miss demand loads (in core cycles)", - "MetricGroup": "MemoryBound;MemoryLat", - "MetricName": "Load_Miss_Real_Latency" + "MetricGroup": "Mem;MemoryBound;MemoryLat", + "MetricName": "Load_Miss_Real_Latency", + "PublicDescription": "Actual Average Latency for L1 data-cache miss demand load instructions (in core cycles). Latency may be overestimated for multi-load instructions - e.g. repeat strings." }, { - "MetricExpr": "L1D_PEND_MISS.PENDING / L1D_PEND_MISS.PENDING_CYCLES", "BriefDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss. Per-Logical Processor)", - "MetricGroup": "MemoryBound;MemoryBW", + "MetricExpr": "L1D_PEND_MISS.PENDING / L1D_PEND_MISS.PENDING_CYCLES", + "MetricGroup": "Mem;MemoryBound;MemoryBW", "MetricName": "MLP" }, { - "MetricConstraint": "NO_NMI_WATCHDOG", - "MetricExpr": "( ITLB_MISSES.WALK_PENDING + DTLB_LOAD_MISSES.WALK_PENDING + DTLB_STORE_MISSES.WALK_PENDING ) / ( 2 * CPU_CLK_UNHALTED.DISTRIBUTED )", - "BriefDescription": "Utilization of the core's Page Walker(s) serving STLB misses triggered by instruction/Load/Store accesses", - "MetricGroup": "MemoryTLB", - "MetricName": "Page_Walks_Utilization" - }, - { - "MetricExpr": "64 * L1D.REPLACEMENT / 1000000000 / duration_time", "BriefDescription": "Average data fill bandwidth to the L1 data cache [GB / sec]", - "MetricGroup": "MemoryBW", + "MetricExpr": "64 * L1D.REPLACEMENT / 1000000000 / duration_time", + "MetricGroup": "Mem;MemoryBW", "MetricName": "L1D_Cache_Fill_BW" }, { - "MetricExpr": "64 * L2_LINES_IN.ALL / 1000000000 / duration_time", "BriefDescription": "Average data fill bandwidth to the L2 cache [GB / sec]", - "MetricGroup": "MemoryBW", + "MetricExpr": "64 * L2_LINES_IN.ALL / 1000000000 / duration_time", + "MetricGroup": "Mem;MemoryBW", "MetricName": "L2_Cache_Fill_BW" }, { - "MetricExpr": "64 * LONGEST_LAT_CACHE.MISS / 1000000000 / duration_time", "BriefDescription": "Average per-core data fill bandwidth to the L3 cache [GB / sec]", - "MetricGroup": "MemoryBW", + "MetricExpr": "64 * LONGEST_LAT_CACHE.MISS / 1000000000 / duration_time", + "MetricGroup": "Mem;MemoryBW", "MetricName": "L3_Cache_Fill_BW" }, { - "MetricExpr": "64 * OFFCORE_REQUESTS.ALL_REQUESTS / 1000000000 / duration_time", "BriefDescription": "Average per-core data access bandwidth to the L3 cache [GB / sec]", - "MetricGroup": "MemoryBW;Offcore", + "MetricExpr": "64 * OFFCORE_REQUESTS.ALL_REQUESTS / 1000000000 / duration_time", + "MetricGroup": "Mem;MemoryBW;Offcore", "MetricName": "L3_Cache_Access_BW" }, { - "MetricExpr": "1000 * MEM_LOAD_RETIRED.L1_MISS / INST_RETIRED.ANY", "BriefDescription": "L1 cache true misses per kilo instruction for retired demand loads", - "MetricGroup": "CacheMisses", + "MetricExpr": "1000 * MEM_LOAD_RETIRED.L1_MISS / INST_RETIRED.ANY", + "MetricGroup": "Mem;CacheMisses", "MetricName": "L1MPKI" }, { - "MetricExpr": "1000 * MEM_LOAD_RETIRED.L2_MISS / INST_RETIRED.ANY", + "BriefDescription": "L1 cache true misses per kilo instruction for all demand loads (including speculative)", + "MetricExpr": "1000 * L2_RQSTS.ALL_DEMAND_DATA_RD / INST_RETIRED.ANY", + "MetricGroup": "Mem;CacheMisses", + "MetricName": "L1MPKI_Load" + }, + { "BriefDescription": "L2 cache true misses per kilo instruction for retired demand loads", - "MetricGroup": "CacheMisses", + "MetricExpr": "1000 * MEM_LOAD_RETIRED.L2_MISS / INST_RETIRED.ANY", + "MetricGroup": "Mem;Backend;CacheMisses", "MetricName": "L2MPKI" }, { - "MetricExpr": "1000 * ( ( OFFCORE_REQUESTS.ALL_DATA_RD - OFFCORE_REQUESTS.DEMAND_DATA_RD ) + L2_RQSTS.ALL_DEMAND_MISS + L2_RQSTS.SWPF_MISS ) / INST_RETIRED.ANY", "BriefDescription": "L2 cache misses per kilo instruction for all request types (including speculative)", - "MetricGroup": "CacheMisses;Offcore", + "MetricExpr": "1000 * ( ( OFFCORE_REQUESTS.ALL_DATA_RD - OFFCORE_REQUESTS.DEMAND_DATA_RD ) + L2_RQSTS.ALL_DEMAND_MISS + L2_RQSTS.SWPF_MISS ) / INST_RETIRED.ANY", + "MetricGroup": "Mem;CacheMisses;Offcore", "MetricName": "L2MPKI_All" }, { - "MetricExpr": "1000 * MEM_LOAD_RETIRED.L3_MISS / INST_RETIRED.ANY", + "BriefDescription": "L2 cache misses per kilo instruction for all demand loads (including speculative)", + "MetricExpr": "1000 * L2_RQSTS.DEMAND_DATA_RD_MISS / INST_RETIRED.ANY", + "MetricGroup": "Mem;CacheMisses", + "MetricName": "L2MPKI_Load" + }, + { + "BriefDescription": "L2 cache hits per kilo instruction for all demand loads (including speculative)", + "MetricExpr": "1000 * L2_RQSTS.DEMAND_DATA_RD_HIT / INST_RETIRED.ANY", + "MetricGroup": "Mem;CacheMisses", + "MetricName": "L2HPKI_Load" + }, + { "BriefDescription": "L3 cache true misses per kilo instruction for retired demand loads", - "MetricGroup": "CacheMisses", + "MetricExpr": "1000 * MEM_LOAD_RETIRED.L3_MISS / INST_RETIRED.ANY", + "MetricGroup": "Mem;CacheMisses", "MetricName": "L3MPKI" }, { - "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC / msr@tsc@", + "BriefDescription": "Fill Buffer (FB) true hits per kilo instructions for retired demand loads", + "MetricExpr": "1000 * MEM_LOAD_RETIRED.FB_HIT / INST_RETIRED.ANY", + "MetricGroup": "Mem;CacheMisses", + "MetricName": "FB_HPKI" + }, + { + "BriefDescription": "Utilization of the core's Page Walker(s) serving STLB misses triggered by instruction/Load/Store accesses", + "MetricConstraint": "NO_NMI_WATCHDOG", + "MetricExpr": "( ITLB_MISSES.WALK_PENDING + DTLB_LOAD_MISSES.WALK_PENDING + DTLB_STORE_MISSES.WALK_PENDING ) / ( 2 * CPU_CLK_UNHALTED.DISTRIBUTED )", + "MetricGroup": "Mem;MemoryTLB", + "MetricName": "Page_Walks_Utilization" + }, + { "BriefDescription": "Average CPU Utilization", + "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC / msr@tsc@", "MetricGroup": "HPC;Summary", "MetricName": "CPU_Utilization" }, { - "MetricExpr": "(CPU_CLK_UNHALTED.THREAD / CPU_CLK_UNHALTED.REF_TSC) * msr@tsc@ / 1000000000 / duration_time", "BriefDescription": "Measured Average Frequency for unhalted processors [GHz]", + "MetricExpr": "(CPU_CLK_UNHALTED.THREAD / CPU_CLK_UNHALTED.REF_TSC) * msr@tsc@ / 1000000000 / duration_time", "MetricGroup": "Summary;Power", "MetricName": "Average_Frequency" }, { - "MetricExpr": "( ( 1 * ( FP_ARITH_INST_RETIRED.SCALAR_SINGLE + FP_ARITH_INST_RETIRED.SCALAR_DOUBLE ) + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * ( FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE ) + 8 * ( FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.512B_PACKED_DOUBLE ) + 16 * FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE ) / 1000000000 ) / duration_time", "BriefDescription": "Giga Floating Point Operations Per Second", - "MetricGroup": "Flops;HPC", + "MetricExpr": "( ( 1 * ( FP_ARITH_INST_RETIRED.SCALAR_SINGLE + FP_ARITH_INST_RETIRED.SCALAR_DOUBLE ) + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * ( FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE ) + 8 * ( FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.512B_PACKED_DOUBLE ) + 16 * FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE ) / 1000000000 ) / duration_time", + "MetricGroup": "Cor;Flops;HPC", "MetricName": "GFLOPs" }, { - "MetricExpr": "CPU_CLK_UNHALTED.THREAD / CPU_CLK_UNHALTED.REF_TSC", "BriefDescription": "Average Frequency Utilization relative nominal frequency", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD / CPU_CLK_UNHALTED.REF_TSC", "MetricGroup": "Power", "MetricName": "Turbo_Utilization" }, { - "MetricExpr": "1 - CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / CPU_CLK_UNHALTED.REF_DISTRIBUTED", + "BriefDescription": "Fraction of Core cycles where the core was running with power-delivery for baseline license level 0", + "MetricExpr": "CORE_POWER.LVL0_TURBO_LICENSE / CPU_CLK_UNHALTED.DISTRIBUTED", + "MetricGroup": "Power", + "MetricName": "Power_License0_Utilization", + "PublicDescription": "Fraction of Core cycles where the core was running with power-delivery for baseline license level 0. This includes non-AVX codes, SSE, AVX 128-bit, and low-current AVX 256-bit codes." + }, + { + "BriefDescription": "Fraction of Core cycles where the core was running with power-delivery for license level 1", + "MetricExpr": "CORE_POWER.LVL1_TURBO_LICENSE / CPU_CLK_UNHALTED.DISTRIBUTED", + "MetricGroup": "Power", + "MetricName": "Power_License1_Utilization", + "PublicDescription": "Fraction of Core cycles where the core was running with power-delivery for license level 1. This includes high current AVX 256-bit instructions as well as low current AVX 512-bit instructions." + }, + { + "BriefDescription": "Fraction of Core cycles where the core was running with power-delivery for license level 2 (introduced in SKX)", + "MetricExpr": "CORE_POWER.LVL2_TURBO_LICENSE / CPU_CLK_UNHALTED.DISTRIBUTED", + "MetricGroup": "Power", + "MetricName": "Power_License2_Utilization", + "PublicDescription": "Fraction of Core cycles where the core was running with power-delivery for license level 2 (introduced in SKX). This includes high current AVX 512-bit instructions." + }, + { "BriefDescription": "Fraction of cycles where both hardware Logical Processors were active", + "MetricExpr": "1 - CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / CPU_CLK_UNHALTED.REF_DISTRIBUTED if #SMT_on else 0", "MetricGroup": "SMT", "MetricName": "SMT_2T_Utilization" }, { - "MetricExpr": "CPU_CLK_UNHALTED.THREAD:k / CPU_CLK_UNHALTED.THREAD", "BriefDescription": "Fraction of cycles spent in the Operating System (OS) Kernel mode", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / CPU_CLK_UNHALTED.THREAD", "MetricGroup": "OS", "MetricName": "Kernel_Utilization" }, { - "MetricExpr": "64 * ( arb@event\\=0x81\\,umask\\=0x1@ + arb@event\\=0x84\\,umask\\=0x1@ ) / 1000000 / duration_time / 1000", + "BriefDescription": "Cycles Per Instruction for the Operating System (OS) Kernel mode", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / INST_RETIRED.ANY_P:k", + "MetricGroup": "OS", + "MetricName": "Kernel_CPI" + }, + { "BriefDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]", - "MetricGroup": "HPC;MemoryBW;SoC", + "MetricExpr": "64 * ( arb@event\\=0x81\\,umask\\=0x1@ + arb@event\\=0x84\\,umask\\=0x1@ ) / 1000000 / duration_time / 1000", + "MetricGroup": "HPC;Mem;MemoryBW;SoC", "MetricName": "DRAM_BW_Use" }, { - "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.FAR_BRANCH:u", "BriefDescription": "Instructions per Far Branch ( Far Branches apply upon transition from application to operating system, handling interrupts, exceptions) [lower number means higher occurrence rate]", + "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.FAR_BRANCH:u", "MetricGroup": "Branches;OS", "MetricName": "IpFarBranch" }, { - "MetricExpr": "(cstate_core@c3\\-residency@ / msr@tsc@) * 100", "BriefDescription": "C3 residency percent per core", + "MetricExpr": "(cstate_core@c3\\-residency@ / msr@tsc@) * 100", "MetricGroup": "Power", "MetricName": "C3_Core_Residency" }, { - "MetricExpr": "(cstate_core@c6\\-residency@ / msr@tsc@) * 100", "BriefDescription": "C6 residency percent per core", + "MetricExpr": "(cstate_core@c6\\-residency@ / msr@tsc@) * 100", "MetricGroup": "Power", "MetricName": "C6_Core_Residency" }, { - "MetricExpr": "(cstate_core@c7\\-residency@ / msr@tsc@) * 100", "BriefDescription": "C7 residency percent per core", + "MetricExpr": "(cstate_core@c7\\-residency@ / msr@tsc@) * 100", "MetricGroup": "Power", "MetricName": "C7_Core_Residency" }, { - "MetricExpr": "(cstate_pkg@c2\\-residency@ / msr@tsc@) * 100", "BriefDescription": "C2 residency percent per package", + "MetricExpr": "(cstate_pkg@c2\\-residency@ / msr@tsc@) * 100", "MetricGroup": "Power", "MetricName": "C2_Pkg_Residency" }, { - "MetricExpr": "(cstate_pkg@c3\\-residency@ / msr@tsc@) * 100", "BriefDescription": "C3 residency percent per package", + "MetricExpr": "(cstate_pkg@c3\\-residency@ / msr@tsc@) * 100", "MetricGroup": "Power", "MetricName": "C3_Pkg_Residency" }, { - "MetricExpr": "(cstate_pkg@c6\\-residency@ / msr@tsc@) * 100", "BriefDescription": "C6 residency percent per package", + "MetricExpr": "(cstate_pkg@c6\\-residency@ / msr@tsc@) * 100", "MetricGroup": "Power", "MetricName": "C6_Pkg_Residency" }, { - "MetricExpr": "(cstate_pkg@c7\\-residency@ / msr@tsc@) * 100", "BriefDescription": "C7 residency percent per package", + "MetricExpr": "(cstate_pkg@c7\\-residency@ / msr@tsc@) * 100", "MetricGroup": "Power", "MetricName": "C7_Pkg_Residency" } diff --git a/tools/perf/pmu-events/arch/x86/icelake/memory.json b/tools/perf/pmu-events/arch/x86/icelake/memory.json index 3701bd93a462..f045e1f6a868 100644 --- a/tools/perf/pmu-events/arch/x86/icelake/memory.json +++ b/tools/perf/pmu-events/arch/x86/icelake/memory.json @@ -1,15 +1,27 @@ [ { - "BriefDescription": "Number of times a transactional abort was signaled due to a data conflict on a transactionally accessed address", + "BriefDescription": "Cycles while L3 cache miss demand load is outstanding.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0x54", - "EventName": "TX_MEM.ABORT_CONFLICT", + "CounterMask": "2", + "EventCode": "0xA3", + "EventName": "CYCLE_ACTIVITY.CYCLES_L3_MISS", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts the number of times a TSX line had a cache conflict.", - "SampleAfterValue": "100003", + "SampleAfterValue": "1000003", "Speculative": "1", - "UMask": "0x1" + "UMask": "0x2" + }, + { + "BriefDescription": "Execution stalls while L3 cache miss demand load is outstanding.", + "CollectPEBSRecord": "2", + "Counter": "0,1,2,3", + "CounterMask": "6", + "EventCode": "0xa3", + "EventName": "CYCLE_ACTIVITY.STALLS_L3_MISS", + "PEBScounters": "0,1,2,3", + "SampleAfterValue": "1000003", + "Speculative": "1", + "UMask": "0x6" }, { "BriefDescription": "Number of times an HLE execution aborted due to any reasons (multiple categories may count as one).", @@ -23,50 +35,15 @@ "UMask": "0x4" }, { - "BriefDescription": "Counts demand data reads that was not supplied by the L3 cache.", - "CollectPEBSRecord": "2", - "Counter": "0,1,2,3", - "EventCode": "0xB7, 0xBB", - "EventName": "OCR.DEMAND_DATA_RD.L3_MISS", - "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x3FFFC00001", - "Offcore": "1", - "PEBScounters": "0,1,2,3", - "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", - "SampleAfterValue": "100003", - "Speculative": "1", - "UMask": "0x1" - }, - { - "BriefDescription": "Counts randomly selected loads when the latency from first dispatch to completion is greater than 16 cycles.", + "BriefDescription": "Number of times an HLE execution aborted due to unfriendly events (such as interrupts).", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "Data_LA": "1", - "EventCode": "0xcd", - "EventName": "MEM_TRANS_RETIRED.LOAD_LATENCY_GT_16", - "MSRIndex": "0x3F6", - "MSRValue": "0x10", - "PEBS": "2", + "EventCode": "0xc8", + "EventName": "HLE_RETIRED.ABORTED_EVENTS", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts randomly selected loads when the latency from first dispatch to completion is greater than 16 cycles. Reported latency may be longer than just the memory latency.", - "SampleAfterValue": "20011", - "TakenAlone": "1", - "UMask": "0x1" - }, - { - "BriefDescription": "Counts hardware prefetch data reads (which bring data to L2) that was not supplied by the L3 cache.", - "CollectPEBSRecord": "2", - "Counter": "0,1,2,3", - "EventCode": "0xB7, 0xBB", - "EventName": "OCR.HWPF_L2_DATA_RD.L3_MISS", - "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x3FFFC00010", - "Offcore": "1", - "PEBScounters": "0,1,2,3", - "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", + "PublicDescription": "Counts the number of times an HLE execution aborted due to unfriendly events (such as interrupts).", "SampleAfterValue": "100003", - "Speculative": "1", - "UMask": "0x1" + "UMask": "0x80" }, { "BriefDescription": "Number of times an HLE execution aborted due to various memory events (e.g., read/write capacity and conflicts).", @@ -80,185 +57,231 @@ "UMask": "0x8" }, { - "BriefDescription": "Number of times an HLE transactional execution aborted due to an unsupported read alignment from the elision buffer.", + "BriefDescription": "Number of times an HLE execution aborted due to HLE-unfriendly instructions and certain unfriendly events (such as AD assists etc.).", "CollectPEBSRecord": "2", - "Counter": "0,1,2,3", - "EventCode": "0x54", - "EventName": "TX_MEM.ABORT_HLE_ELISION_BUFFER_UNSUPPORTED_ALIGNMENT", - "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts the number of times a TSX Abort was triggered due to attempting an unsupported alignment from Lock Buffer.", + "Counter": "0,1,2,3,4,5,6,7", + "EventCode": "0xc8", + "EventName": "HLE_RETIRED.ABORTED_UNFRIENDLY", + "PEBScounters": "0,1,2,3,4,5,6,7", + "PublicDescription": "Counts the number of times an HLE execution aborted due to HLE-unfriendly instructions and certain unfriendly events (such as AD assists etc.).", "SampleAfterValue": "100003", - "Speculative": "1", "UMask": "0x20" }, { - "BriefDescription": "Number of times an HLE transactional execution aborted due to NoAllocatedElisionBuffer being non-zero.", + "BriefDescription": "Number of times an HLE execution successfully committed", "CollectPEBSRecord": "2", - "Counter": "0,1,2,3", - "EventCode": "0x54", - "EventName": "TX_MEM.ABORT_HLE_ELISION_BUFFER_NOT_EMPTY", - "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts the number of times a TSX Abort was triggered due to commit but Lock Buffer not empty.", + "Counter": "0,1,2,3,4,5,6,7", + "EventCode": "0xc8", + "EventName": "HLE_RETIRED.COMMIT", + "PEBScounters": "0,1,2,3,4,5,6,7", + "PublicDescription": "Counts the number of times HLE commit succeeded.", "SampleAfterValue": "100003", - "Speculative": "1", - "UMask": "0x8" + "UMask": "0x2" }, { - "BriefDescription": "Number of times an instruction execution caused the transactional nest count supported to be exceeded", + "BriefDescription": "Number of times an HLE execution started.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0x5d", - "EventName": "TX_EXEC.MISC3", + "EventCode": "0xc8", + "EventName": "HLE_RETIRED.START", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts Unfriendly TSX abort triggered by a nest count that is too deep.", + "PublicDescription": "Counts the number of times we entered an HLE region. Does not count nested transactions.", "SampleAfterValue": "100003", - "Speculative": "1", - "UMask": "0x4" + "UMask": "0x1" }, { - "BriefDescription": "Counts the number of times a class of instructions that may cause a transactional abort was executed inside a transactional region", + "BriefDescription": "Number of machine clears due to memory ordering conflicts.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0x5d", - "EventName": "TX_EXEC.MISC2", + "EventCode": "0xc3", + "EventName": "MACHINE_CLEARS.MEMORY_ORDERING", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts Unfriendly TSX abort triggered by a vzeroupper instruction.", + "PublicDescription": "Counts the number of Machine Clears detected dye to memory ordering. Memory Ordering Machine Clears may apply when a memory read may not conform to the memory ordering rules of the x86 architecture", "SampleAfterValue": "100003", "Speculative": "1", "UMask": "0x2" }, { - "BriefDescription": "Cycles where data return is pending for a Demand Data Read request who miss L3 cache.", + "BriefDescription": "Counts randomly selected loads when the latency from first dispatch to completion is greater than 128 cycles.", "CollectPEBSRecord": "2", - "Counter": "0,1,2,3", - "CounterMask": "1", - "EventCode": "0x60", - "EventName": "OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_L3_MISS_DEMAND_DATA_RD", - "PEBScounters": "0,1,2,3", - "PublicDescription": "Cycles with at least 1 Demand Data Read requests who miss L3 cache in the superQ.", - "SampleAfterValue": "1000003", - "Speculative": "1", - "UMask": "0x10" + "Counter": "0,1,2,3,4,5,6,7", + "Data_LA": "1", + "EventCode": "0xcd", + "EventName": "MEM_TRANS_RETIRED.LOAD_LATENCY_GT_128", + "MSRIndex": "0x3F6", + "MSRValue": "0x80", + "PEBS": "2", + "PEBScounters": "0,1,2,3,4,5,6,7", + "PublicDescription": "Counts randomly selected loads when the latency from first dispatch to completion is greater than 128 cycles. Reported latency may be longer than just the memory latency.", + "SampleAfterValue": "1009", + "TakenAlone": "1", + "UMask": "0x1" }, { - "BriefDescription": "Counts demand reads for ownership (RFO) requests and software prefetches for exclusive ownership (PREFETCHW) that was not supplied by the L3 cache.", + "BriefDescription": "Counts randomly selected loads when the latency from first dispatch to completion is greater than 16 cycles.", "CollectPEBSRecord": "2", - "Counter": "0,1,2,3", - "EventCode": "0xB7, 0xBB", - "EventName": "OCR.DEMAND_RFO.L3_MISS", - "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x3FFFC00002", - "Offcore": "1", - "PEBScounters": "0,1,2,3", - "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", - "SampleAfterValue": "100003", - "Speculative": "1", + "Counter": "0,1,2,3,4,5,6,7", + "Data_LA": "1", + "EventCode": "0xcd", + "EventName": "MEM_TRANS_RETIRED.LOAD_LATENCY_GT_16", + "MSRIndex": "0x3F6", + "MSRValue": "0x10", + "PEBS": "2", + "PEBScounters": "0,1,2,3,4,5,6,7", + "PublicDescription": "Counts randomly selected loads when the latency from first dispatch to completion is greater than 16 cycles. Reported latency may be longer than just the memory latency.", + "SampleAfterValue": "20011", + "TakenAlone": "1", "UMask": "0x1" }, { - "BriefDescription": "Counts randomly selected loads when the latency from first dispatch to completion is greater than 512 cycles.", + "BriefDescription": "Counts randomly selected loads when the latency from first dispatch to completion is greater than 256 cycles.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", "Data_LA": "1", "EventCode": "0xcd", - "EventName": "MEM_TRANS_RETIRED.LOAD_LATENCY_GT_512", + "EventName": "MEM_TRANS_RETIRED.LOAD_LATENCY_GT_256", "MSRIndex": "0x3F6", - "MSRValue": "0x200", + "MSRValue": "0x100", "PEBS": "2", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts randomly selected loads when the latency from first dispatch to completion is greater than 512 cycles. Reported latency may be longer than just the memory latency.", - "SampleAfterValue": "101", + "PublicDescription": "Counts randomly selected loads when the latency from first dispatch to completion is greater than 256 cycles. Reported latency may be longer than just the memory latency.", + "SampleAfterValue": "503", "TakenAlone": "1", "UMask": "0x1" }, { - "BriefDescription": "Number of times an RTM execution successfully committed", + "BriefDescription": "Counts randomly selected loads when the latency from first dispatch to completion is greater than 32 cycles.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xc9", - "EventName": "RTM_RETIRED.COMMIT", + "Data_LA": "1", + "EventCode": "0xcd", + "EventName": "MEM_TRANS_RETIRED.LOAD_LATENCY_GT_32", + "MSRIndex": "0x3F6", + "MSRValue": "0x20", + "PEBS": "2", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts the number of times RTM commit succeeded.", - "SampleAfterValue": "100003", - "UMask": "0x2" + "PublicDescription": "Counts randomly selected loads when the latency from first dispatch to completion is greater than 32 cycles. Reported latency may be longer than just the memory latency.", + "SampleAfterValue": "100007", + "TakenAlone": "1", + "UMask": "0x1" }, { - "BriefDescription": "Speculatively counts the number of TSX aborts due to a data capacity limitation for transactional writes.", + "BriefDescription": "Counts randomly selected loads when the latency from first dispatch to completion is greater than 4 cycles.", "CollectPEBSRecord": "2", - "Counter": "0,1,2,3", - "EventCode": "0x54", - "EventName": "TX_MEM.ABORT_CAPACITY_WRITE", - "PEBScounters": "0,1,2,3", - "PublicDescription": "Speculatively counts the number of Transactional Synchronization Extensions (TSX) aborts due to a data capacity limitation for transactional writes.", + "Counter": "0,1,2,3,4,5,6,7", + "Data_LA": "1", + "EventCode": "0xcd", + "EventName": "MEM_TRANS_RETIRED.LOAD_LATENCY_GT_4", + "MSRIndex": "0x3F6", + "MSRValue": "0x4", + "PEBS": "2", + "PEBScounters": "0,1,2,3,4,5,6,7", + "PublicDescription": "Counts randomly selected loads when the latency from first dispatch to completion is greater than 4 cycles. Reported latency may be longer than just the memory latency.", "SampleAfterValue": "100003", - "Speculative": "1", - "UMask": "0x2" + "TakenAlone": "1", + "UMask": "0x1" }, { - "BriefDescription": "Number of times an HLE execution aborted due to unfriendly events (such as interrupts).", + "BriefDescription": "Counts randomly selected loads when the latency from first dispatch to completion is greater than 512 cycles.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xc8", - "EventName": "HLE_RETIRED.ABORTED_EVENTS", + "Data_LA": "1", + "EventCode": "0xcd", + "EventName": "MEM_TRANS_RETIRED.LOAD_LATENCY_GT_512", + "MSRIndex": "0x3F6", + "MSRValue": "0x200", + "PEBS": "2", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts the number of times an HLE execution aborted due to unfriendly events (such as interrupts).", - "SampleAfterValue": "100003", - "UMask": "0x80" + "PublicDescription": "Counts randomly selected loads when the latency from first dispatch to completion is greater than 512 cycles. Reported latency may be longer than just the memory latency.", + "SampleAfterValue": "101", + "TakenAlone": "1", + "UMask": "0x1" }, { - "BriefDescription": "Number of times an HLE execution successfully committed", + "BriefDescription": "Counts randomly selected loads when the latency from first dispatch to completion is greater than 64 cycles.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xc8", - "EventName": "HLE_RETIRED.COMMIT", + "Data_LA": "1", + "EventCode": "0xcd", + "EventName": "MEM_TRANS_RETIRED.LOAD_LATENCY_GT_64", + "MSRIndex": "0x3F6", + "MSRValue": "0x40", + "PEBS": "2", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts the number of times HLE commit succeeded.", - "SampleAfterValue": "100003", - "UMask": "0x2" + "PublicDescription": "Counts randomly selected loads when the latency from first dispatch to completion is greater than 64 cycles. Reported latency may be longer than just the memory latency.", + "SampleAfterValue": "2003", + "TakenAlone": "1", + "UMask": "0x1" }, { - "BriefDescription": "Number of times an RTM execution aborted due to incompatible memory type", + "BriefDescription": "Counts randomly selected loads when the latency from first dispatch to completion is greater than 8 cycles.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xc9", - "EventName": "RTM_RETIRED.ABORTED_MEMTYPE", + "Data_LA": "1", + "EventCode": "0xcd", + "EventName": "MEM_TRANS_RETIRED.LOAD_LATENCY_GT_8", + "MSRIndex": "0x3F6", + "MSRValue": "0x8", + "PEBS": "2", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts the number of times an RTM execution aborted due to incompatible memory type.", + "PublicDescription": "Counts randomly selected loads when the latency from first dispatch to completion is greater than 8 cycles. Reported latency may be longer than just the memory latency.", + "SampleAfterValue": "50021", + "TakenAlone": "1", + "UMask": "0x1" + }, + { + "BriefDescription": "Counts demand instruction fetches and L1 instruction cache prefetches that was not supplied by the L3 cache.", + "CollectPEBSRecord": "2", + "Counter": "0,1,2,3", + "EventCode": "0xB7, 0xBB", + "EventName": "OCR.DEMAND_CODE_RD.L3_MISS", + "MSRIndex": "0x1a6,0x1a7", + "MSRValue": "0x3FFFC00004", + "Offcore": "1", + "PEBScounters": "0,1,2,3", + "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", "SampleAfterValue": "100003", - "UMask": "0x40" + "Speculative": "1", + "UMask": "0x1" }, { - "BriefDescription": "Number of machine clears due to memory ordering conflicts.", + "BriefDescription": "Counts demand data reads that was not supplied by the L3 cache.", "CollectPEBSRecord": "2", - "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xc3", - "EventName": "MACHINE_CLEARS.MEMORY_ORDERING", - "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts the number of Machine Clears detected dye to memory ordering. Memory Ordering Machine Clears may apply when a memory read may not conform to the memory ordering rules of the x86 architecture", + "Counter": "0,1,2,3", + "EventCode": "0xB7, 0xBB", + "EventName": "OCR.DEMAND_DATA_RD.L3_MISS", + "MSRIndex": "0x1a6,0x1a7", + "MSRValue": "0x3FFFC00001", + "Offcore": "1", + "PEBScounters": "0,1,2,3", + "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", "SampleAfterValue": "100003", "Speculative": "1", - "UMask": "0x2" + "UMask": "0x1" }, { - "BriefDescription": "Number of times an HLE transactional execution aborted due to XRELEASE lock not satisfying the address and value requirements in the elision buffer", + "BriefDescription": "Counts demand reads for ownership (RFO) requests and software prefetches for exclusive ownership (PREFETCHW) that was not supplied by the L3 cache.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0x54", - "EventName": "TX_MEM.ABORT_HLE_ELISION_BUFFER_MISMATCH", + "EventCode": "0xB7, 0xBB", + "EventName": "OCR.DEMAND_RFO.L3_MISS", + "MSRIndex": "0x1a6,0x1a7", + "MSRValue": "0x3FFFC00002", + "Offcore": "1", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts the number of times a TSX Abort was triggered due to release/commit but data and address mismatch.", + "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", "SampleAfterValue": "100003", "Speculative": "1", - "UMask": "0x10" + "UMask": "0x1" }, { - "BriefDescription": "Counts streaming stores that was not supplied by the L3 cache.", + "BriefDescription": "Counts L1 data cache prefetch requests and software prefetches (except PREFETCHW) that was not supplied by the L3 cache.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.STREAMING_WR.L3_MISS", + "EventName": "OCR.HWPF_L1D_AND_SWPF.L3_MISS", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x3FFFC00800", + "MSRValue": "0x3FFFC00400", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -267,16 +290,34 @@ "UMask": "0x1" }, { - "BriefDescription": "Speculatively counts the number of TSX aborts due to a data capacity limitation for transactional reads", + "BriefDescription": "Counts hardware prefetch data reads (which bring data to L2) that was not supplied by the L3 cache.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0x54", - "EventName": "TX_MEM.ABORT_CAPACITY_READ", + "EventCode": "0xB7, 0xBB", + "EventName": "OCR.HWPF_L2_DATA_RD.L3_MISS", + "MSRIndex": "0x1a6,0x1a7", + "MSRValue": "0x3FFFC00010", + "Offcore": "1", "PEBScounters": "0,1,2,3", - "PublicDescription": "Speculatively counts the number of Transactional Synchronization Extensions (TSX) aborts due to a data capacity limitation for transactional reads", + "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", "SampleAfterValue": "100003", "Speculative": "1", - "UMask": "0x80" + "UMask": "0x1" + }, + { + "BriefDescription": "Counts hardware prefetch RFOs (which bring data to L2) that was not supplied by the L3 cache.", + "CollectPEBSRecord": "2", + "Counter": "0,1,2,3", + "EventCode": "0xB7, 0xBB", + "EventName": "OCR.HWPF_L2_RFO.L3_MISS", + "MSRIndex": "0x1a6,0x1a7", + "MSRValue": "0x3FFFC00020", + "Offcore": "1", + "PEBScounters": "0,1,2,3", + "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", + "SampleAfterValue": "100003", + "Speculative": "1", + "UMask": "0x1" }, { "BriefDescription": "Counts miscellaneous requests, such as I/O and un-cacheable accesses that was not supplied by the L3 cache.", @@ -294,13 +335,13 @@ "UMask": "0x1" }, { - "BriefDescription": "Counts hardware prefetch RFOs (which bring data to L2) that was not supplied by the L3 cache.", + "BriefDescription": "Counts streaming stores that was not supplied by the L3 cache.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.HWPF_L2_RFO.L3_MISS", + "EventName": "OCR.STREAMING_WR.L3_MISS", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x3FFFC00020", + "MSRValue": "0x3FFFC00800", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -309,39 +350,39 @@ "UMask": "0x1" }, { - "BriefDescription": "Demand Data Read requests who miss L3 cache", + "BriefDescription": "Counts demand data read requests that miss the L3 cache.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xb0", "EventName": "OFFCORE_REQUESTS.L3_MISS_DEMAND_DATA_RD", "PEBScounters": "0,1,2,3", - "PublicDescription": "Demand Data Read requests who miss L3 cache.", "SampleAfterValue": "100003", "Speculative": "1", "UMask": "0x10" }, { - "BriefDescription": "Cycles while L3 cache miss demand load is outstanding.", + "BriefDescription": "Cycles where at least one demand data read request known to have missed the L3 cache is pending.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "CounterMask": "2", - "EventCode": "0xA3", - "EventName": "CYCLE_ACTIVITY.CYCLES_L3_MISS", + "CounterMask": "1", + "EventCode": "0x60", + "EventName": "OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_L3_MISS_DEMAND_DATA_RD", "PEBScounters": "0,1,2,3", + "PublicDescription": "Cycles where at least one demand data read request known to have missed the L3 cache is pending. Note that this does not capture all elapsed cycles while requests are outstanding - only cycles from when the requests were known to have missed the L3 cache.", "SampleAfterValue": "1000003", "Speculative": "1", - "UMask": "0x2" + "UMask": "0x10" }, { - "BriefDescription": "Number of times an RTM execution aborted due to HLE-unfriendly instructions", + "BriefDescription": "Number of times an RTM execution aborted.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc9", - "EventName": "RTM_RETIRED.ABORTED_UNFRIENDLY", + "EventName": "RTM_RETIRED.ABORTED", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts the number of times an RTM execution aborted due to HLE-unfriendly instructions.", + "PublicDescription": "Counts the number of times RTM abort was triggered.", "SampleAfterValue": "100003", - "UMask": "0x20" + "UMask": "0x4" }, { "BriefDescription": "Number of times an RTM execution aborted due to none of the previous 4 categories (e.g. interrupt)", @@ -355,208 +396,154 @@ "UMask": "0x80" }, { - "BriefDescription": "Number of times an HLE execution started.", + "BriefDescription": "Number of times an RTM execution aborted due to various memory events (e.g. read/write capacity and conflicts)", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xc8", - "EventName": "HLE_RETIRED.START", + "EventCode": "0xc9", + "EventName": "RTM_RETIRED.ABORTED_MEM", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts the number of times we entered an HLE region. Does not count nested transactions.", + "PublicDescription": "Counts the number of times an RTM execution aborted due to various memory events (e.g. read/write capacity and conflicts).", "SampleAfterValue": "100003", - "UMask": "0x1" + "UMask": "0x8" }, { - "BriefDescription": "Counts randomly selected loads when the latency from first dispatch to completion is greater than 4 cycles.", + "BriefDescription": "Number of times an RTM execution aborted due to incompatible memory type", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "Data_LA": "1", - "EventCode": "0xcd", - "EventName": "MEM_TRANS_RETIRED.LOAD_LATENCY_GT_4", - "MSRIndex": "0x3F6", - "MSRValue": "0x4", - "PEBS": "2", + "EventCode": "0xc9", + "EventName": "RTM_RETIRED.ABORTED_MEMTYPE", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts randomly selected loads when the latency from first dispatch to completion is greater than 4 cycles. Reported latency may be longer than just the memory latency.", + "PublicDescription": "Counts the number of times an RTM execution aborted due to incompatible memory type.", "SampleAfterValue": "100003", - "TakenAlone": "1", - "UMask": "0x1" + "UMask": "0x40" }, { - "BriefDescription": "Counts randomly selected loads when the latency from first dispatch to completion is greater than 128 cycles.", + "BriefDescription": "Number of times an RTM execution aborted due to HLE-unfriendly instructions", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "Data_LA": "1", - "EventCode": "0xcd", - "EventName": "MEM_TRANS_RETIRED.LOAD_LATENCY_GT_128", - "MSRIndex": "0x3F6", - "MSRValue": "0x80", - "PEBS": "2", + "EventCode": "0xc9", + "EventName": "RTM_RETIRED.ABORTED_UNFRIENDLY", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts randomly selected loads when the latency from first dispatch to completion is greater than 128 cycles. Reported latency may be longer than just the memory latency.", - "SampleAfterValue": "1009", - "TakenAlone": "1", - "UMask": "0x1" - }, - { - "BriefDescription": "Number of times HLE lock could not be elided due to ElisionBufferAvailable being zero.", - "CollectPEBSRecord": "2", - "Counter": "0,1,2,3", - "EventCode": "0x54", - "EventName": "TX_MEM.HLE_ELISION_BUFFER_FULL", - "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts the number of times we could not allocate Lock Buffer.", + "PublicDescription": "Counts the number of times an RTM execution aborted due to HLE-unfriendly instructions.", "SampleAfterValue": "100003", - "Speculative": "1", - "UMask": "0x40" + "UMask": "0x20" }, { - "BriefDescription": "Counts randomly selected loads when the latency from first dispatch to completion is greater than 8 cycles.", + "BriefDescription": "Number of times an RTM execution successfully committed", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "Data_LA": "1", - "EventCode": "0xcd", - "EventName": "MEM_TRANS_RETIRED.LOAD_LATENCY_GT_8", - "MSRIndex": "0x3F6", - "MSRValue": "0x8", - "PEBS": "2", + "EventCode": "0xc9", + "EventName": "RTM_RETIRED.COMMIT", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts randomly selected loads when the latency from first dispatch to completion is greater than 8 cycles. Reported latency may be longer than just the memory latency.", - "SampleAfterValue": "50021", - "TakenAlone": "1", - "UMask": "0x1" + "PublicDescription": "Counts the number of times RTM commit succeeded.", + "SampleAfterValue": "100003", + "UMask": "0x2" }, { - "BriefDescription": "Counts randomly selected loads when the latency from first dispatch to completion is greater than 256 cycles.", + "BriefDescription": "Number of times an RTM execution started.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "Data_LA": "1", - "EventCode": "0xcd", - "EventName": "MEM_TRANS_RETIRED.LOAD_LATENCY_GT_256", - "MSRIndex": "0x3F6", - "MSRValue": "0x100", - "PEBS": "2", + "EventCode": "0xc9", + "EventName": "RTM_RETIRED.START", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts randomly selected loads when the latency from first dispatch to completion is greater than 256 cycles. Reported latency may be longer than just the memory latency.", - "SampleAfterValue": "503", - "TakenAlone": "1", + "PublicDescription": "Counts the number of times we entered an RTM region. Does not count nested transactions.", + "SampleAfterValue": "100003", "UMask": "0x1" }, { - "BriefDescription": "Execution stalls while L3 cache miss demand load is outstanding.", - "CollectPEBSRecord": "2", - "Counter": "0,1,2,3", - "CounterMask": "6", - "EventCode": "0xa3", - "EventName": "CYCLE_ACTIVITY.STALLS_L3_MISS", - "PEBScounters": "0,1,2,3", - "SampleAfterValue": "1000003", - "Speculative": "1", - "UMask": "0x6" - }, - { - "BriefDescription": "Counts randomly selected loads when the latency from first dispatch to completion is greater than 64 cycles.", + "BriefDescription": "Counts the number of times a class of instructions that may cause a transactional abort was executed inside a transactional region", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "Data_LA": "1", - "EventCode": "0xcd", - "EventName": "MEM_TRANS_RETIRED.LOAD_LATENCY_GT_64", - "MSRIndex": "0x3F6", - "MSRValue": "0x40", - "PEBS": "2", + "EventCode": "0x5d", + "EventName": "TX_EXEC.MISC2", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts randomly selected loads when the latency from first dispatch to completion is greater than 64 cycles. Reported latency may be longer than just the memory latency.", - "SampleAfterValue": "2003", - "TakenAlone": "1", - "UMask": "0x1" + "PublicDescription": "Counts Unfriendly TSX abort triggered by a vzeroupper instruction.", + "SampleAfterValue": "100003", + "Speculative": "1", + "UMask": "0x2" }, { - "BriefDescription": "Counts randomly selected loads when the latency from first dispatch to completion is greater than 32 cycles.", + "BriefDescription": "Number of times an instruction execution caused the transactional nest count supported to be exceeded", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "Data_LA": "1", - "EventCode": "0xcd", - "EventName": "MEM_TRANS_RETIRED.LOAD_LATENCY_GT_32", - "MSRIndex": "0x3F6", - "MSRValue": "0x20", - "PEBS": "2", + "EventCode": "0x5d", + "EventName": "TX_EXEC.MISC3", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts randomly selected loads when the latency from first dispatch to completion is greater than 32 cycles. Reported latency may be longer than just the memory latency.", - "SampleAfterValue": "100007", - "TakenAlone": "1", - "UMask": "0x1" + "PublicDescription": "Counts Unfriendly TSX abort triggered by a nest count that is too deep.", + "SampleAfterValue": "100003", + "Speculative": "1", + "UMask": "0x4" }, { - "BriefDescription": "Number of times an RTM execution aborted due to various memory events (e.g. read/write capacity and conflicts)", + "BriefDescription": "Speculatively counts the number of TSX aborts due to a data capacity limitation for transactional reads", "CollectPEBSRecord": "2", - "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xc9", - "EventName": "RTM_RETIRED.ABORTED_MEM", - "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts the number of times an RTM execution aborted due to various memory events (e.g. read/write capacity and conflicts).", + "Counter": "0,1,2,3", + "EventCode": "0x54", + "EventName": "TX_MEM.ABORT_CAPACITY_READ", + "PEBScounters": "0,1,2,3", + "PublicDescription": "Speculatively counts the number of Transactional Synchronization Extensions (TSX) aborts due to a data capacity limitation for transactional reads", "SampleAfterValue": "100003", - "UMask": "0x8" + "Speculative": "1", + "UMask": "0x80" }, { - "BriefDescription": "Counts L1 data cache prefetch requests and software prefetches (except PREFETCHW) that was not supplied by the L3 cache.", + "BriefDescription": "Speculatively counts the number of TSX aborts due to a data capacity limitation for transactional writes.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0xB7, 0xBB", - "EventName": "OCR.HWPF_L1D_AND_SWPF.L3_MISS", - "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x3FFFC00400", - "Offcore": "1", + "EventCode": "0x54", + "EventName": "TX_MEM.ABORT_CAPACITY_WRITE", "PEBScounters": "0,1,2,3", - "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", + "PublicDescription": "Speculatively counts the number of Transactional Synchronization Extensions (TSX) aborts due to a data capacity limitation for transactional writes.", "SampleAfterValue": "100003", "Speculative": "1", - "UMask": "0x1" + "UMask": "0x2" }, { - "BriefDescription": "Counts demand instruction fetches and L1 instruction cache prefetches that was not supplied by the L3 cache.", + "BriefDescription": "Number of times a transactional abort was signaled due to a data conflict on a transactionally accessed address", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0xB7, 0xBB", - "EventName": "OCR.DEMAND_CODE_RD.L3_MISS", - "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x3FFFC00004", - "Offcore": "1", + "EventCode": "0x54", + "EventName": "TX_MEM.ABORT_CONFLICT", "PEBScounters": "0,1,2,3", - "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", + "PublicDescription": "Counts the number of times a TSX line had a cache conflict.", "SampleAfterValue": "100003", "Speculative": "1", "UMask": "0x1" }, { - "BriefDescription": "Number of times an RTM execution aborted.", + "BriefDescription": "Number of times an HLE transactional execution aborted due to XRELEASE lock not satisfying the address and value requirements in the elision buffer", "CollectPEBSRecord": "2", - "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xc9", - "EventName": "RTM_RETIRED.ABORTED", - "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts the number of times RTM abort was triggered.", + "Counter": "0,1,2,3", + "EventCode": "0x54", + "EventName": "TX_MEM.ABORT_HLE_ELISION_BUFFER_MISMATCH", + "PEBScounters": "0,1,2,3", + "PublicDescription": "Counts the number of times a TSX Abort was triggered due to release/commit but data and address mismatch.", "SampleAfterValue": "100003", - "UMask": "0x4" + "Speculative": "1", + "UMask": "0x10" }, { - "BriefDescription": "Number of times an RTM execution started.", + "BriefDescription": "Number of times an HLE transactional execution aborted due to NoAllocatedElisionBuffer being non-zero.", "CollectPEBSRecord": "2", - "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xc9", - "EventName": "RTM_RETIRED.START", - "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts the number of times we entered an RTM region. Does not count nested transactions.", + "Counter": "0,1,2,3", + "EventCode": "0x54", + "EventName": "TX_MEM.ABORT_HLE_ELISION_BUFFER_NOT_EMPTY", + "PEBScounters": "0,1,2,3", + "PublicDescription": "Counts the number of times a TSX Abort was triggered due to commit but Lock Buffer not empty.", "SampleAfterValue": "100003", - "UMask": "0x1" + "Speculative": "1", + "UMask": "0x8" }, { - "BriefDescription": "Number of times an HLE execution aborted due to HLE-unfriendly instructions and certain unfriendly events (such as AD assists etc.).", + "BriefDescription": "Number of times an HLE transactional execution aborted due to an unsupported read alignment from the elision buffer.", "CollectPEBSRecord": "2", - "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xc8", - "EventName": "HLE_RETIRED.ABORTED_UNFRIENDLY", - "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts the number of times an HLE execution aborted due to HLE-unfriendly instructions and certain unfriendly events (such as AD assists etc.).", + "Counter": "0,1,2,3", + "EventCode": "0x54", + "EventName": "TX_MEM.ABORT_HLE_ELISION_BUFFER_UNSUPPORTED_ALIGNMENT", + "PEBScounters": "0,1,2,3", + "PublicDescription": "Counts the number of times a TSX Abort was triggered due to attempting an unsupported alignment from Lock Buffer.", "SampleAfterValue": "100003", + "Speculative": "1", "UMask": "0x20" }, { @@ -570,5 +557,17 @@ "SampleAfterValue": "100003", "Speculative": "1", "UMask": "0x4" + }, + { + "BriefDescription": "Number of times HLE lock could not be elided due to ElisionBufferAvailable being zero.", + "CollectPEBSRecord": "2", + "Counter": "0,1,2,3", + "EventCode": "0x54", + "EventName": "TX_MEM.HLE_ELISION_BUFFER_FULL", + "PEBScounters": "0,1,2,3", + "PublicDescription": "Counts the number of times we could not allocate Lock Buffer.", + "SampleAfterValue": "100003", + "Speculative": "1", + "UMask": "0x40" } ]
\ No newline at end of file diff --git a/tools/perf/pmu-events/arch/x86/icelake/other.json b/tools/perf/pmu-events/arch/x86/icelake/other.json index a806b00f8616..10e8582774ce 100644 --- a/tools/perf/pmu-events/arch/x86/icelake/other.json +++ b/tools/perf/pmu-events/arch/x86/icelake/other.json @@ -1,57 +1,60 @@ [ { - "BriefDescription": "Counts miscellaneous requests, such as I/O and un-cacheable accesses that hit a cacheline in the L3 where a snoop was not needed to satisfy the request.", + "BriefDescription": "Number of occurrences where a microcode assist is invoked by hardware.", + "CollectPEBSRecord": "2", + "Counter": "0,1,2,3,4,5,6,7", + "EventCode": "0xc1", + "EventName": "ASSISTS.ANY", + "PEBScounters": "0,1,2,3,4,5,6,7", + "PublicDescription": "Counts the number of occurrences where a microcode assist is invoked by hardware Examples include AD (page Access Dirty), FP and AVX related assists.", + "SampleAfterValue": "100003", + "Speculative": "1", + "UMask": "0x7" + }, + { + "BriefDescription": "Core cycles where the core was running in a manner where Turbo may be clipped to the Non-AVX turbo schedule.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0xB7, 0xBB", - "EventName": "OCR.OTHER.L3_HIT.SNOOP_NOT_NEEDED", - "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x01003C8000", - "Offcore": "1", + "EventCode": "0x28", + "EventName": "CORE_POWER.LVL0_TURBO_LICENSE", "PEBScounters": "0,1,2,3", - "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", - "SampleAfterValue": "100003", + "PublicDescription": "Counts Core cycles where the core was running with power-delivery for baseline license level 0. This includes non-AVX codes, SSE, AVX 128-bit, and low-current AVX 256-bit codes.", + "SampleAfterValue": "200003", "Speculative": "1", - "UMask": "0x1" + "UMask": "0x7" }, { - "BriefDescription": "Counts hardware prefetch RFOs (which bring data to L2) that DRAM supplied the request.", + "BriefDescription": "Core cycles where the core was running in a manner where Turbo may be clipped to the AVX2 turbo schedule.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0xB7, 0xBB", - "EventName": "OCR.HWPF_L2_RFO.DRAM", - "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x0184000020", - "Offcore": "1", + "EventCode": "0x28", + "EventName": "CORE_POWER.LVL1_TURBO_LICENSE", "PEBScounters": "0,1,2,3", - "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", - "SampleAfterValue": "100003", + "PublicDescription": "Counts Core cycles where the core was running with power-delivery for license level 1. This includes high current AVX 256-bit instructions as well as low current AVX 512-bit instructions.", + "SampleAfterValue": "200003", "Speculative": "1", - "UMask": "0x1" + "UMask": "0x18" }, { - "BriefDescription": "Counts miscellaneous requests, such as I/O and un-cacheable accesses that DRAM supplied the request.", + "BriefDescription": "Core cycles where the core was running in a manner where Turbo may be clipped to the AVX512 turbo schedule.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0xB7, 0xBB", - "EventName": "OCR.OTHER.LOCAL_DRAM", - "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x0184008000", - "Offcore": "1", + "EventCode": "0x28", + "EventName": "CORE_POWER.LVL2_TURBO_LICENSE", "PEBScounters": "0,1,2,3", - "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", - "SampleAfterValue": "100003", + "PublicDescription": "Core cycles where the core was running with power-delivery for license level 2 (introduced in Skylake Server microarchtecture). This includes high current AVX 512-bit instructions.", + "SampleAfterValue": "200003", "Speculative": "1", - "UMask": "0x1" + "UMask": "0x20" }, { - "BriefDescription": "Counts hardware prefetch data reads (which bring data to L2) that DRAM supplied the request.", + "BriefDescription": "Counts demand instruction fetches and L1 instruction cache prefetches that have any type of response.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.HWPF_L2_DATA_RD.LOCAL_DRAM", + "EventName": "OCR.DEMAND_CODE_RD.ANY_RESPONSE", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x0184000010", + "MSRValue": "0x10004", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -60,13 +63,13 @@ "UMask": "0x1" }, { - "BriefDescription": "Counts demand reads for ownership (RFO) requests and software prefetches for exclusive ownership (PREFETCHW) that hit a cacheline in the L3 where a snoop hit in another cores caches, data forwarding is required as the data is modified.", + "BriefDescription": "Counts demand instruction fetches and L1 instruction cache prefetches that DRAM supplied the request.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.DEMAND_RFO.L3_HIT.SNOOP_HITM", + "EventName": "OCR.DEMAND_CODE_RD.DRAM", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x10003C0002", + "MSRValue": "0x184000004", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -75,25 +78,28 @@ "UMask": "0x1" }, { - "BriefDescription": "Number of PREFETCHNTA instructions executed.", + "BriefDescription": "Counts demand instruction fetches and L1 instruction cache prefetches that hit a cacheline in the L3 where a snoop was sent or not.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0x32", - "EventName": "SW_PREFETCH_ACCESS.NTA", + "EventCode": "0xB7, 0xBB", + "EventName": "OCR.DEMAND_CODE_RD.L3_HIT.ANY", + "MSRIndex": "0x1a6,0x1a7", + "MSRValue": "0x3FC03C0004", + "Offcore": "1", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts the number of PREFETCHNTA instructions executed.", + "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", "SampleAfterValue": "100003", "Speculative": "1", "UMask": "0x1" }, { - "BriefDescription": "Counts demand reads for ownership (RFO) requests and software prefetches for exclusive ownership (PREFETCHW) that hit a cacheline in the L3 where a snoop was sent but no other cores had the data.", + "BriefDescription": "Counts demand instruction fetches and L1 instruction cache prefetches that hit a cacheline in the L3 where a snoop hit in another cores caches, data forwarding is required as the data is modified.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.DEMAND_RFO.L3_HIT.SNOOP_MISS", + "EventName": "OCR.DEMAND_CODE_RD.L3_HIT.SNOOP_HITM", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x02003C0002", + "MSRValue": "0x10003C0004", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -102,13 +108,13 @@ "UMask": "0x1" }, { - "BriefDescription": "Counts demand instruction fetches and L1 instruction cache prefetches that have any type of response.", + "BriefDescription": "Counts demand instruction fetches and L1 instruction cache prefetches that hit a cacheline in the L3 where a snoop hit in another core, data forwarding is not required.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.DEMAND_CODE_RD.ANY_RESPONSE", + "EventName": "OCR.DEMAND_CODE_RD.L3_HIT.SNOOP_HIT_NO_FWD", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x0000010004", + "MSRValue": "0x4003C0004", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -117,13 +123,13 @@ "UMask": "0x1" }, { - "BriefDescription": "Counts L1 data cache prefetch requests and software prefetches (except PREFETCHW) that hit a cacheline in the L3 where a snoop was sent or not.", + "BriefDescription": "Counts demand instruction fetches and L1 instruction cache prefetches that hit a cacheline in the L3 where a snoop was sent but no other cores had the data.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.HWPF_L1D_AND_SWPF.L3_HIT.ANY", + "EventName": "OCR.DEMAND_CODE_RD.L3_HIT.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x3FC03C0400", + "MSRValue": "0x2003C0004", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -132,13 +138,13 @@ "UMask": "0x1" }, { - "BriefDescription": "Counts miscellaneous requests, such as I/O and un-cacheable accesses that DRAM supplied the request.", + "BriefDescription": "Counts demand instruction fetches and L1 instruction cache prefetches that hit a cacheline in the L3 where a snoop was not needed to satisfy the request.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.OTHER.DRAM", + "EventName": "OCR.DEMAND_CODE_RD.L3_HIT.SNOOP_NOT_NEEDED", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x0184008000", + "MSRValue": "0x1003C0004", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -147,13 +153,13 @@ "UMask": "0x1" }, { - "BriefDescription": "Counts demand reads for ownership (RFO) requests and software prefetches for exclusive ownership (PREFETCHW) that have any type of response.", + "BriefDescription": "Counts demand instruction fetches and L1 instruction cache prefetches that hit a cacheline in the L3 where a snoop was sent.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.DEMAND_RFO.ANY_RESPONSE", + "EventName": "OCR.DEMAND_CODE_RD.L3_HIT.SNOOP_SENT", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x0000010002", + "MSRValue": "0x1E003C0004", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -162,13 +168,13 @@ "UMask": "0x1" }, { - "BriefDescription": "Counts miscellaneous requests, such as I/O and un-cacheable accesses that hit a cacheline in the L3 where a snoop was sent but no other cores had the data.", + "BriefDescription": "Counts demand instruction fetches and L1 instruction cache prefetches that DRAM supplied the request.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.OTHER.L3_HIT.SNOOP_MISS", + "EventName": "OCR.DEMAND_CODE_RD.LOCAL_DRAM", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x02003C8000", + "MSRValue": "0x184000004", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -177,13 +183,13 @@ "UMask": "0x1" }, { - "BriefDescription": "Counts demand reads for ownership (RFO) requests and software prefetches for exclusive ownership (PREFETCHW) that hit a cacheline in the L3 where a snoop was not needed to satisfy the request.", + "BriefDescription": "Counts demand data reads that have any type of response.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.DEMAND_RFO.L3_HIT.SNOOP_NOT_NEEDED", + "EventName": "OCR.DEMAND_DATA_RD.ANY_RESPONSE", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x01003C0002", + "MSRValue": "0x10001", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -192,25 +198,13 @@ "UMask": "0x1" }, { - "BriefDescription": "TMA slots wasted due to incorrect speculation by branch mispredictions", - "CollectPEBSRecord": "2", - "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xa4", - "EventName": "TOPDOWN.BR_MISPREDICT_SLOTS", - "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Number of TMA slots that were wasted due to incorrect speculation by branch mispredictions. This event estimates number of operations that were issued but not retired from the specualtive path as well as the out-of-order engine recovery past a branch misprediction.", - "SampleAfterValue": "10000003", - "Speculative": "1", - "UMask": "0x8" - }, - { - "BriefDescription": "Counts hardware prefetch data reads (which bring data to L2) that hit a cacheline in the L3 where a snoop was not needed to satisfy the request.", + "BriefDescription": "Counts demand data reads that DRAM supplied the request.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.HWPF_L2_DATA_RD.L3_HIT.SNOOP_NOT_NEEDED", + "EventName": "OCR.DEMAND_DATA_RD.DRAM", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x01003C0010", + "MSRValue": "0x184000001", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -219,13 +213,13 @@ "UMask": "0x1" }, { - "BriefDescription": "Counts streaming stores that have any type of response.", + "BriefDescription": "Counts demand data reads that hit a cacheline in the L3 where a snoop was sent or not.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.STREAMING_WR.ANY_RESPONSE", + "EventName": "OCR.DEMAND_DATA_RD.L3_HIT.ANY", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x0000010800", + "MSRValue": "0x3FC03C0001", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -234,13 +228,13 @@ "UMask": "0x1" }, { - "BriefDescription": "Counts streaming stores that DRAM supplied the request.", + "BriefDescription": "Counts demand data reads that hit a cacheline in the L3 where a snoop hit in another cores caches, data forwarding is required as the data is modified.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.STREAMING_WR.DRAM", + "EventName": "OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HITM", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x0184000800", + "MSRValue": "0x10003C0001", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -249,13 +243,13 @@ "UMask": "0x1" }, { - "BriefDescription": "Counts hardware prefetch RFOs (which bring data to L2) that have any type of response.", + "BriefDescription": "Counts demand data reads that hit a cacheline in the L3 where a snoop hit in another core, data forwarding is not required.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.HWPF_L2_RFO.ANY_RESPONSE", + "EventName": "OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HIT_NO_FWD", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x0000010020", + "MSRValue": "0x4003C0001", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -264,13 +258,13 @@ "UMask": "0x1" }, { - "BriefDescription": "Counts demand data reads that hit a cacheline in the L3 where a snoop hit in another cores caches, data forwarding is required as the data is modified.", + "BriefDescription": "Counts demand data reads that hit a cacheline in the L3 where a snoop was sent but no other cores had the data.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HITM", + "EventName": "OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x10003C0001", + "MSRValue": "0x2003C0001", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -279,13 +273,13 @@ "UMask": "0x1" }, { - "BriefDescription": "Counts hardware prefetch data reads (which bring data to L2) that DRAM supplied the request.", + "BriefDescription": "Counts demand data reads that hit a cacheline in the L3 where a snoop was not needed to satisfy the request.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.HWPF_L2_DATA_RD.DRAM", + "EventName": "OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_NOT_NEEDED", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x0184000010", + "MSRValue": "0x1003C0001", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -294,13 +288,13 @@ "UMask": "0x1" }, { - "BriefDescription": "Counts miscellaneous requests, such as I/O and un-cacheable accesses that hit a cacheline in the L3 where a snoop was sent.", + "BriefDescription": "Counts demand data reads that hit a cacheline in the L3 where a snoop was sent.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.OTHER.L3_HIT.SNOOP_SENT", + "EventName": "OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_SENT", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x1E003C8000", + "MSRValue": "0x1E003C0001", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -309,13 +303,13 @@ "UMask": "0x1" }, { - "BriefDescription": "Counts hardware prefetch data reads (which bring data to L2) that have any type of response.", + "BriefDescription": "Counts demand data reads that DRAM supplied the request.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.HWPF_L2_DATA_RD.ANY_RESPONSE", + "EventName": "OCR.DEMAND_DATA_RD.LOCAL_DRAM", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x0000010010", + "MSRValue": "0x184000001", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -324,13 +318,13 @@ "UMask": "0x1" }, { - "BriefDescription": "Counts hardware prefetch data reads (which bring data to L2) that hit a cacheline in the L3 where a snoop was sent or not.", + "BriefDescription": "Counts demand reads for ownership (RFO) requests and software prefetches for exclusive ownership (PREFETCHW) that have any type of response.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.HWPF_L2_DATA_RD.L3_HIT.ANY", + "EventName": "OCR.DEMAND_RFO.ANY_RESPONSE", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x3FC03C0010", + "MSRValue": "0x10002", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -339,25 +333,28 @@ "UMask": "0x1" }, { - "BriefDescription": "Core cycles where the core was running in a manner where Turbo may be clipped to the AVX2 turbo schedule.", + "BriefDescription": "Counts demand reads for ownership (RFO) requests and software prefetches for exclusive ownership (PREFETCHW) that DRAM supplied the request.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0x28", - "EventName": "CORE_POWER.LVL1_TURBO_LICENSE", + "EventCode": "0xB7, 0xBB", + "EventName": "OCR.DEMAND_RFO.DRAM", + "MSRIndex": "0x1a6,0x1a7", + "MSRValue": "0x184000002", + "Offcore": "1", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts Core cycles where the core was running with power-delivery for license level 1. This includes high current AVX 256-bit instructions as well as low current AVX 512-bit instructions.", - "SampleAfterValue": "200003", + "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", + "SampleAfterValue": "100003", "Speculative": "1", - "UMask": "0x18" + "UMask": "0x1" }, { - "BriefDescription": "Counts demand data reads that hit a cacheline in the L3 where a snoop was sent but no other cores had the data.", + "BriefDescription": "Counts demand reads for ownership (RFO) requests and software prefetches for exclusive ownership (PREFETCHW) that hit a cacheline in the L3 where a snoop was sent or not.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_MISS", + "EventName": "OCR.DEMAND_RFO.L3_HIT.ANY", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x02003C0001", + "MSRValue": "0x3FC03C0002", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -366,13 +363,13 @@ "UMask": "0x1" }, { - "BriefDescription": "Counts demand data reads that hit a cacheline in the L3 where a snoop was sent.", + "BriefDescription": "Counts demand reads for ownership (RFO) requests and software prefetches for exclusive ownership (PREFETCHW) that hit a cacheline in the L3 where a snoop hit in another cores caches, data forwarding is required as the data is modified.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_SENT", + "EventName": "OCR.DEMAND_RFO.L3_HIT.SNOOP_HITM", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x1E003C0001", + "MSRValue": "0x10003C0002", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -381,13 +378,13 @@ "UMask": "0x1" }, { - "BriefDescription": "Counts hardware prefetch data reads (which bring data to L2) that hit a cacheline in the L3 where a snoop hit in another cores caches, data forwarding is required as the data is modified.", + "BriefDescription": "Counts demand reads for ownership (RFO) requests and software prefetches for exclusive ownership (PREFETCHW) that hit a cacheline in the L3 where a snoop hit in another core, data forwarding is not required.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.HWPF_L2_DATA_RD.L3_HIT.SNOOP_HITM", + "EventName": "OCR.DEMAND_RFO.L3_HIT.SNOOP_HIT_NO_FWD", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x10003C0010", + "MSRValue": "0x4003C0002", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -396,13 +393,13 @@ "UMask": "0x1" }, { - "BriefDescription": "Counts demand data reads that DRAM supplied the request.", + "BriefDescription": "Counts demand reads for ownership (RFO) requests and software prefetches for exclusive ownership (PREFETCHW) that hit a cacheline in the L3 where a snoop was sent but no other cores had the data.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.DEMAND_DATA_RD.DRAM", + "EventName": "OCR.DEMAND_RFO.L3_HIT.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x0184000001", + "MSRValue": "0x2003C0002", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -411,13 +408,13 @@ "UMask": "0x1" }, { - "BriefDescription": "Counts streaming stores that DRAM supplied the request.", + "BriefDescription": "Counts demand reads for ownership (RFO) requests and software prefetches for exclusive ownership (PREFETCHW) that hit a cacheline in the L3 where a snoop was not needed to satisfy the request.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.STREAMING_WR.LOCAL_DRAM", + "EventName": "OCR.DEMAND_RFO.L3_HIT.SNOOP_NOT_NEEDED", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x0184000800", + "MSRValue": "0x1003C0002", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -426,13 +423,13 @@ "UMask": "0x1" }, { - "BriefDescription": "Counts demand instruction fetches and L1 instruction cache prefetches that DRAM supplied the request.", + "BriefDescription": "Counts demand reads for ownership (RFO) requests and software prefetches for exclusive ownership (PREFETCHW) that hit a cacheline in the L3 where a snoop was sent.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.DEMAND_CODE_RD.LOCAL_DRAM", + "EventName": "OCR.DEMAND_RFO.L3_HIT.SNOOP_SENT", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x0184000004", + "MSRValue": "0x1E003C0002", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -441,13 +438,13 @@ "UMask": "0x1" }, { - "BriefDescription": "Counts demand instruction fetches and L1 instruction cache prefetches that hit a cacheline in the L3 where a snoop was not needed to satisfy the request.", + "BriefDescription": "Counts demand reads for ownership (RFO) requests and software prefetches for exclusive ownership (PREFETCHW) that DRAM supplied the request.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.DEMAND_CODE_RD.L3_HIT.SNOOP_NOT_NEEDED", + "EventName": "OCR.DEMAND_RFO.LOCAL_DRAM", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x01003C0004", + "MSRValue": "0x184000002", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -456,13 +453,13 @@ "UMask": "0x1" }, { - "BriefDescription": "Counts hardware prefetch RFOs (which bring data to L2) that DRAM supplied the request.", + "BriefDescription": "Counts L1 data cache prefetch requests and software prefetches (except PREFETCHW) that have any type of response.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.HWPF_L2_RFO.LOCAL_DRAM", + "EventName": "OCR.HWPF_L1D_AND_SWPF.ANY_RESPONSE", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x0184000020", + "MSRValue": "0x10400", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -471,13 +468,13 @@ "UMask": "0x1" }, { - "BriefDescription": "Counts demand instruction fetches and L1 instruction cache prefetches that DRAM supplied the request.", + "BriefDescription": "Counts L1 data cache prefetch requests and software prefetches (except PREFETCHW) that DRAM supplied the request.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.DEMAND_CODE_RD.DRAM", + "EventName": "OCR.HWPF_L1D_AND_SWPF.DRAM", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x0184000004", + "MSRValue": "0x184000400", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -486,13 +483,13 @@ "UMask": "0x1" }, { - "BriefDescription": "Counts demand reads for ownership (RFO) requests and software prefetches for exclusive ownership (PREFETCHW) that hit a cacheline in the L3 where a snoop hit in another core, data forwarding is not required.", + "BriefDescription": "Counts L1 data cache prefetch requests and software prefetches (except PREFETCHW) that hit a cacheline in the L3 where a snoop was sent or not.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.DEMAND_RFO.L3_HIT.SNOOP_HIT_NO_FWD", + "EventName": "OCR.HWPF_L1D_AND_SWPF.L3_HIT.ANY", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x04003C0002", + "MSRValue": "0x3FC03C0400", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -507,7 +504,7 @@ "EventCode": "0xB7, 0xBB", "EventName": "OCR.HWPF_L1D_AND_SWPF.L3_HIT.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x02003C0400", + "MSRValue": "0x2003C0400", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -516,13 +513,13 @@ "UMask": "0x1" }, { - "BriefDescription": "Counts hardware prefetch RFOs (which bring data to L2) that hit a cacheline in the L3 where a snoop was sent.", + "BriefDescription": "Counts L1 data cache prefetch requests and software prefetches (except PREFETCHW) that hit a cacheline in the L3 where a snoop was not needed to satisfy the request.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.HWPF_L2_RFO.L3_HIT.SNOOP_SENT", + "EventName": "OCR.HWPF_L1D_AND_SWPF.L3_HIT.SNOOP_NOT_NEEDED", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x1E003C0020", + "MSRValue": "0x1003C0400", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -531,13 +528,13 @@ "UMask": "0x1" }, { - "BriefDescription": "Counts hardware prefetch data reads (which bring data to L2) that hit a cacheline in the L3 where a snoop hit in another core, data forwarding is not required.", + "BriefDescription": "Counts L1 data cache prefetch requests and software prefetches (except PREFETCHW) that DRAM supplied the request.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.HWPF_L2_DATA_RD.L3_HIT.SNOOP_HIT_NO_FWD", + "EventName": "OCR.HWPF_L1D_AND_SWPF.LOCAL_DRAM", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x04003C0010", + "MSRValue": "0x184000400", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -546,13 +543,13 @@ "UMask": "0x1" }, { - "BriefDescription": "Counts demand data reads that hit a cacheline in the L3 where a snoop hit in another core, data forwarding is not required.", + "BriefDescription": "Counts hardware prefetch data reads (which bring data to L2) that have any type of response.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HIT_NO_FWD", + "EventName": "OCR.HWPF_L2_DATA_RD.ANY_RESPONSE", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x04003C0001", + "MSRValue": "0x10010", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -561,13 +558,13 @@ "UMask": "0x1" }, { - "BriefDescription": "Counts hardware prefetch RFOs (which bring data to L2) that hit a cacheline in the L3 where a snoop was sent or not.", + "BriefDescription": "Counts hardware prefetch data reads (which bring data to L2) that DRAM supplied the request.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.HWPF_L2_RFO.L3_HIT.ANY", + "EventName": "OCR.HWPF_L2_DATA_RD.DRAM", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x3FC03C0020", + "MSRValue": "0x184000010", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -576,13 +573,13 @@ "UMask": "0x1" }, { - "BriefDescription": "Counts demand data reads that hit a cacheline in the L3 where a snoop was not needed to satisfy the request.", + "BriefDescription": "Counts hardware prefetch data reads (which bring data to L2) that hit a cacheline in the L3 where a snoop was sent or not.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_NOT_NEEDED", + "EventName": "OCR.HWPF_L2_DATA_RD.L3_HIT.ANY", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x01003C0001", + "MSRValue": "0x3FC03C0010", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -591,13 +588,13 @@ "UMask": "0x1" }, { - "BriefDescription": "Counts miscellaneous requests, such as I/O and un-cacheable accesses that hit a cacheline in the L3 where a snoop hit in another core, data forwarding is not required.", + "BriefDescription": "Counts hardware prefetch data reads (which bring data to L2) that hit a cacheline in the L3 where a snoop hit in another cores caches, data forwarding is required as the data is modified.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.OTHER.L3_HIT.SNOOP_HIT_NO_FWD", + "EventName": "OCR.HWPF_L2_DATA_RD.L3_HIT.SNOOP_HITM", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x04003C8000", + "MSRValue": "0x10003C0010", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -606,36 +603,28 @@ "UMask": "0x1" }, { - "BriefDescription": "TMA slots available for an unhalted logical processor. Fixed counter - architectural event", - "CollectPEBSRecord": "2", - "Counter": "35", - "EventName": "TOPDOWN.SLOTS", - "PEBScounters": "35", - "PublicDescription": "Number of available slots for an unhalted logical processor. The event increments by machine-width of the narrowest pipeline as employed by the Top-down Microarchitecture Analysis method (TMA). The count is distributed among unhalted logical processors (hyper-threads) who share the same physical core. Software can use this event as the denominator for the top-level metrics of the TMA method. This architectural event is counted on a designated fixed counter (Fixed Counter 3).", - "SampleAfterValue": "10000003", - "Speculative": "1", - "UMask": "0x4" - }, - { - "BriefDescription": "Number of PREFETCHT1 or PREFETCHT2 instructions executed.", + "BriefDescription": "Counts hardware prefetch data reads (which bring data to L2) that hit a cacheline in the L3 where a snoop hit in another core, data forwarding is not required.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0x32", - "EventName": "SW_PREFETCH_ACCESS.T1_T2", + "EventCode": "0xB7, 0xBB", + "EventName": "OCR.HWPF_L2_DATA_RD.L3_HIT.SNOOP_HIT_NO_FWD", + "MSRIndex": "0x1a6,0x1a7", + "MSRValue": "0x4003C0010", + "Offcore": "1", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts the number of PREFETCHT1 or PREFETCHT2 instructions executed.", + "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", "SampleAfterValue": "100003", "Speculative": "1", - "UMask": "0x4" + "UMask": "0x1" }, { - "BriefDescription": "Counts demand data reads that hit a cacheline in the L3 where a snoop was sent or not.", + "BriefDescription": "Counts hardware prefetch data reads (which bring data to L2) that hit a cacheline in the L3 where a snoop was sent but no other cores had the data.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.DEMAND_DATA_RD.L3_HIT.ANY", + "EventName": "OCR.HWPF_L2_DATA_RD.L3_HIT.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x3FC03C0001", + "MSRValue": "0x2003C0010", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -644,13 +633,13 @@ "UMask": "0x1" }, { - "BriefDescription": "Counts L1 data cache prefetch requests and software prefetches (except PREFETCHW) that DRAM supplied the request.", + "BriefDescription": "Counts hardware prefetch data reads (which bring data to L2) that hit a cacheline in the L3 where a snoop was not needed to satisfy the request.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.HWPF_L1D_AND_SWPF.LOCAL_DRAM", + "EventName": "OCR.HWPF_L2_DATA_RD.L3_HIT.SNOOP_NOT_NEEDED", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x0184000400", + "MSRValue": "0x1003C0010", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -659,25 +648,13 @@ "UMask": "0x1" }, { - "BriefDescription": "TMA slots where no uops were being issued due to lack of back-end resources.", - "CollectPEBSRecord": "2", - "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xa4", - "EventName": "TOPDOWN.BACKEND_BOUND_SLOTS", - "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts the number of Top-down Microarchitecture Analysis (TMA) method's slots where no micro-operations were being issued from front-end to back-end of the machine due to lack of back-end resources.", - "SampleAfterValue": "10000003", - "Speculative": "1", - "UMask": "0x2" - }, - { - "BriefDescription": "Counts hardware prefetch RFOs (which bring data to L2) that hit a cacheline in the L3 where a snoop was not needed to satisfy the request.", + "BriefDescription": "Counts hardware prefetch data reads (which bring data to L2) that hit a cacheline in the L3 where a snoop was sent.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.HWPF_L2_RFO.L3_HIT.SNOOP_NOT_NEEDED", + "EventName": "OCR.HWPF_L2_DATA_RD.L3_HIT.SNOOP_SENT", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x01003C0020", + "MSRValue": "0x1E003C0010", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -686,25 +663,28 @@ "UMask": "0x1" }, { - "BriefDescription": "Number of PREFETCHT0 instructions executed.", + "BriefDescription": "Counts hardware prefetch data reads (which bring data to L2) that DRAM supplied the request.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0x32", - "EventName": "SW_PREFETCH_ACCESS.T0", + "EventCode": "0xB7, 0xBB", + "EventName": "OCR.HWPF_L2_DATA_RD.LOCAL_DRAM", + "MSRIndex": "0x1a6,0x1a7", + "MSRValue": "0x184000010", + "Offcore": "1", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts the number of PREFETCHT0 instructions executed.", + "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", "SampleAfterValue": "100003", "Speculative": "1", - "UMask": "0x2" + "UMask": "0x1" }, { - "BriefDescription": "Counts demand instruction fetches and L1 instruction cache prefetches that hit a cacheline in the L3 where a snoop hit in another cores caches, data forwarding is required as the data is modified.", + "BriefDescription": "Counts hardware prefetch RFOs (which bring data to L2) that have any type of response.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.DEMAND_CODE_RD.L3_HIT.SNOOP_HITM", + "EventName": "OCR.HWPF_L2_RFO.ANY_RESPONSE", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x10003C0004", + "MSRValue": "0x10020", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -713,13 +693,13 @@ "UMask": "0x1" }, { - "BriefDescription": "Counts demand data reads that have any type of response.", + "BriefDescription": "Counts hardware prefetch RFOs (which bring data to L2) that DRAM supplied the request.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.DEMAND_DATA_RD.ANY_RESPONSE", + "EventName": "OCR.HWPF_L2_RFO.DRAM", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x0000010001", + "MSRValue": "0x184000020", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -728,13 +708,13 @@ "UMask": "0x1" }, { - "BriefDescription": "Counts demand instruction fetches and L1 instruction cache prefetches that hit a cacheline in the L3 where a snoop hit in another core, data forwarding is not required.", + "BriefDescription": "Counts hardware prefetch RFOs (which bring data to L2) that hit a cacheline in the L3 where a snoop was sent or not.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.DEMAND_CODE_RD.L3_HIT.SNOOP_HIT_NO_FWD", + "EventName": "OCR.HWPF_L2_RFO.L3_HIT.ANY", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x04003C0004", + "MSRValue": "0x3FC03C0020", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -758,13 +738,13 @@ "UMask": "0x1" }, { - "BriefDescription": "Counts miscellaneous requests, such as I/O and un-cacheable accesses that have any type of response.", + "BriefDescription": "Counts hardware prefetch RFOs (which bring data to L2) that hit a cacheline in the L3 where a snoop hit in another core, data forwarding is not required.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.OTHER.ANY_RESPONSE", + "EventName": "OCR.HWPF_L2_RFO.L3_HIT.SNOOP_HIT_NO_FWD", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x0000018000", + "MSRValue": "0x4003C0020", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -773,25 +753,28 @@ "UMask": "0x1" }, { - "BriefDescription": "Number of PREFETCHW instructions executed.", + "BriefDescription": "Counts hardware prefetch RFOs (which bring data to L2) that hit a cacheline in the L3 where a snoop was sent but no other cores had the data.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0x32", - "EventName": "SW_PREFETCH_ACCESS.PREFETCHW", + "EventCode": "0xB7, 0xBB", + "EventName": "OCR.HWPF_L2_RFO.L3_HIT.SNOOP_MISS", + "MSRIndex": "0x1a6,0x1a7", + "MSRValue": "0x2003C0020", + "Offcore": "1", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts the number of PREFETCHW instructions executed.", + "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", "SampleAfterValue": "100003", "Speculative": "1", - "UMask": "0x8" + "UMask": "0x1" }, { - "BriefDescription": "Counts hardware prefetch data reads (which bring data to L2) that hit a cacheline in the L3 where a snoop was sent.", + "BriefDescription": "Counts hardware prefetch RFOs (which bring data to L2) that hit a cacheline in the L3 where a snoop was not needed to satisfy the request.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.HWPF_L2_DATA_RD.L3_HIT.SNOOP_SENT", + "EventName": "OCR.HWPF_L2_RFO.L3_HIT.SNOOP_NOT_NEEDED", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x1E003C0010", + "MSRValue": "0x1003C0020", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -800,13 +783,13 @@ "UMask": "0x1" }, { - "BriefDescription": "Counts L1 data cache prefetch requests and software prefetches (except PREFETCHW) that DRAM supplied the request.", + "BriefDescription": "Counts hardware prefetch RFOs (which bring data to L2) that hit a cacheline in the L3 where a snoop was sent.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.HWPF_L1D_AND_SWPF.DRAM", + "EventName": "OCR.HWPF_L2_RFO.L3_HIT.SNOOP_SENT", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x0184000400", + "MSRValue": "0x1E003C0020", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -815,13 +798,13 @@ "UMask": "0x1" }, { - "BriefDescription": "Counts demand reads for ownership (RFO) requests and software prefetches for exclusive ownership (PREFETCHW) that DRAM supplied the request.", + "BriefDescription": "Counts hardware prefetch RFOs (which bring data to L2) that DRAM supplied the request.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.DEMAND_RFO.DRAM", + "EventName": "OCR.HWPF_L2_RFO.LOCAL_DRAM", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x0184000002", + "MSRValue": "0x184000020", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -830,25 +813,13 @@ "UMask": "0x1" }, { - "BriefDescription": "Number of occurrences where a microcode assist is invoked by hardware.", - "CollectPEBSRecord": "2", - "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xc1", - "EventName": "ASSISTS.ANY", - "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts the number of occurrences where a microcode assist is invoked by hardware Examples include AD (page Access Dirty), FP and AVX related assists.", - "SampleAfterValue": "100003", - "Speculative": "1", - "UMask": "0x7" - }, - { - "BriefDescription": "Counts hardware prefetch data reads (which bring data to L2) that hit a cacheline in the L3 where a snoop was sent but no other cores had the data.", + "BriefDescription": "Counts hardware prefetches to the L3 only that hit a cacheline in the L3 where a snoop was sent or not.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.HWPF_L2_DATA_RD.L3_HIT.SNOOP_MISS", + "EventName": "OCR.HWPF_L3.L3_HIT.ANY", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x02003C0010", + "MSRValue": "0x3FC03C2380", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -857,13 +828,13 @@ "UMask": "0x1" }, { - "BriefDescription": "Counts demand instruction fetches and L1 instruction cache prefetches that hit a cacheline in the L3 where a snoop was sent or not.", + "BriefDescription": "Counts miscellaneous requests, such as I/O and un-cacheable accesses that have any type of response.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.DEMAND_CODE_RD.L3_HIT.ANY", + "EventName": "OCR.OTHER.ANY_RESPONSE", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x3FC03C0004", + "MSRValue": "0x18000", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -872,25 +843,28 @@ "UMask": "0x1" }, { - "BriefDescription": "Core cycles where the core was running in a manner where Turbo may be clipped to the AVX512 turbo schedule.", + "BriefDescription": "Counts miscellaneous requests, such as I/O and un-cacheable accesses that DRAM supplied the request.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0x28", - "EventName": "CORE_POWER.LVL2_TURBO_LICENSE", + "EventCode": "0xB7, 0xBB", + "EventName": "OCR.OTHER.DRAM", + "MSRIndex": "0x1a6,0x1a7", + "MSRValue": "0x184008000", + "Offcore": "1", "PEBScounters": "0,1,2,3", - "PublicDescription": "Core cycles where the core was running with power-delivery for license level 2 (introduced in Skylake Server microarchtecture). This includes high current AVX 512-bit instructions.", - "SampleAfterValue": "200003", + "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", + "SampleAfterValue": "100003", "Speculative": "1", - "UMask": "0x20" + "UMask": "0x1" }, { - "BriefDescription": "Counts streaming stores that hit a cacheline in the L3 where a snoop was sent or not.", + "BriefDescription": "Counts miscellaneous requests, such as I/O and un-cacheable accesses that hit a cacheline in the L3 where a snoop hit in another core, data forwarding is not required.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.STREAMING_WR.L3_HIT.ANY", + "EventName": "OCR.OTHER.L3_HIT.SNOOP_HIT_NO_FWD", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x3FC03C0800", + "MSRValue": "0x4003C8000", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -899,13 +873,13 @@ "UMask": "0x1" }, { - "BriefDescription": "Counts hardware prefetch RFOs (which bring data to L2) that hit a cacheline in the L3 where a snoop was sent but no other cores had the data.", + "BriefDescription": "Counts miscellaneous requests, such as I/O and un-cacheable accesses that hit a cacheline in the L3 where a snoop was sent but no other cores had the data.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.HWPF_L2_RFO.L3_HIT.SNOOP_MISS", + "EventName": "OCR.OTHER.L3_HIT.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x02003C0020", + "MSRValue": "0x2003C8000", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -914,13 +888,13 @@ "UMask": "0x1" }, { - "BriefDescription": "Counts demand data reads that DRAM supplied the request.", + "BriefDescription": "Counts miscellaneous requests, such as I/O and un-cacheable accesses that hit a cacheline in the L3 where a snoop was not needed to satisfy the request.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.DEMAND_DATA_RD.LOCAL_DRAM", + "EventName": "OCR.OTHER.L3_HIT.SNOOP_NOT_NEEDED", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x0184000001", + "MSRValue": "0x1003C8000", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -929,13 +903,13 @@ "UMask": "0x1" }, { - "BriefDescription": "Counts demand instruction fetches and L1 instruction cache prefetches that hit a cacheline in the L3 where a snoop was sent.", + "BriefDescription": "Counts miscellaneous requests, such as I/O and un-cacheable accesses that hit a cacheline in the L3 where a snoop was sent.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.DEMAND_CODE_RD.L3_HIT.SNOOP_SENT", + "EventName": "OCR.OTHER.L3_HIT.SNOOP_SENT", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x1E003C0004", + "MSRValue": "0x1E003C8000", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -944,13 +918,13 @@ "UMask": "0x1" }, { - "BriefDescription": "Counts demand reads for ownership (RFO) requests and software prefetches for exclusive ownership (PREFETCHW) that hit a cacheline in the L3 where a snoop was sent or not.", + "BriefDescription": "Counts miscellaneous requests, such as I/O and un-cacheable accesses that DRAM supplied the request.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.DEMAND_RFO.L3_HIT.ANY", + "EventName": "OCR.OTHER.LOCAL_DRAM", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x3FC03C0002", + "MSRValue": "0x184008000", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -959,13 +933,13 @@ "UMask": "0x1" }, { - "BriefDescription": "Counts hardware prefetches to the L3 only that hit a cacheline in the L3 where a snoop was sent or not.", + "BriefDescription": "Counts streaming stores that have any type of response.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.HWPF_L3.L3_HIT.ANY", + "EventName": "OCR.STREAMING_WR.ANY_RESPONSE", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x3FC03C2380", + "MSRValue": "0x10800", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -974,25 +948,13 @@ "UMask": "0x1" }, { - "BriefDescription": "TMA slots available for an unhalted logical processor. General counter - architectural event", - "CollectPEBSRecord": "2", - "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xa4", - "EventName": "TOPDOWN.SLOTS_P", - "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts the number of available slots for an unhalted logical processor. The event increments by machine-width of the narrowest pipeline as employed by the Top-down Microarchitecture Analysis method. The count is distributed among unhalted logical processors (hyper-threads) who share the same physical core.", - "SampleAfterValue": "10000003", - "Speculative": "1", - "UMask": "0x1" - }, - { - "BriefDescription": "Counts L1 data cache prefetch requests and software prefetches (except PREFETCHW) that hit a cacheline in the L3 where a snoop was not needed to satisfy the request.", + "BriefDescription": "Counts streaming stores that DRAM supplied the request.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.HWPF_L1D_AND_SWPF.L3_HIT.SNOOP_NOT_NEEDED", + "EventName": "OCR.STREAMING_WR.DRAM", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x01003C0400", + "MSRValue": "0x184000800", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -1001,13 +963,13 @@ "UMask": "0x1" }, { - "BriefDescription": "Counts demand reads for ownership (RFO) requests and software prefetches for exclusive ownership (PREFETCHW) that hit a cacheline in the L3 where a snoop was sent.", + "BriefDescription": "Counts streaming stores that hit a cacheline in the L3 where a snoop was sent or not.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.DEMAND_RFO.L3_HIT.SNOOP_SENT", + "EventName": "OCR.STREAMING_WR.L3_HIT.ANY", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x1E003C0002", + "MSRValue": "0x3FC03C0800", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -1016,13 +978,13 @@ "UMask": "0x1" }, { - "BriefDescription": "Counts L1 data cache prefetch requests and software prefetches (except PREFETCHW) that have any type of response.", + "BriefDescription": "Counts streaming stores that DRAM supplied the request.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", - "EventName": "OCR.HWPF_L1D_AND_SWPF.ANY_RESPONSE", + "EventName": "OCR.STREAMING_WR.LOCAL_DRAM", "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x0000010400", + "MSRValue": "0x184000800", "Offcore": "1", "PEBScounters": "0,1,2,3", "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", @@ -1031,60 +993,98 @@ "UMask": "0x1" }, { - "BriefDescription": "Counts demand reads for ownership (RFO) requests and software prefetches for exclusive ownership (PREFETCHW) that DRAM supplied the request.", + "BriefDescription": "Number of PREFETCHNTA instructions executed.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0xB7, 0xBB", - "EventName": "OCR.DEMAND_RFO.LOCAL_DRAM", - "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x0184000002", - "Offcore": "1", + "EventCode": "0x32", + "EventName": "SW_PREFETCH_ACCESS.NTA", "PEBScounters": "0,1,2,3", - "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", + "PublicDescription": "Counts the number of PREFETCHNTA instructions executed.", "SampleAfterValue": "100003", "Speculative": "1", "UMask": "0x1" }, { - "BriefDescription": "Core cycles where the core was running in a manner where Turbo may be clipped to the Non-AVX turbo schedule.", + "BriefDescription": "Number of PREFETCHW instructions executed.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0x28", - "EventName": "CORE_POWER.LVL0_TURBO_LICENSE", + "EventCode": "0x32", + "EventName": "SW_PREFETCH_ACCESS.PREFETCHW", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts Core cycles where the core was running with power-delivery for baseline license level 0. This includes non-AVX codes, SSE, AVX 128-bit, and low-current AVX 256-bit codes.", - "SampleAfterValue": "200003", + "PublicDescription": "Counts the number of PREFETCHW instructions executed.", + "SampleAfterValue": "100003", "Speculative": "1", - "UMask": "0x7" + "UMask": "0x8" }, { - "BriefDescription": "Counts hardware prefetch RFOs (which bring data to L2) that hit a cacheline in the L3 where a snoop hit in another core, data forwarding is not required.", + "BriefDescription": "Number of PREFETCHT0 instructions executed.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0xB7, 0xBB", - "EventName": "OCR.HWPF_L2_RFO.L3_HIT.SNOOP_HIT_NO_FWD", - "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x04003C0020", - "Offcore": "1", + "EventCode": "0x32", + "EventName": "SW_PREFETCH_ACCESS.T0", "PEBScounters": "0,1,2,3", - "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", + "PublicDescription": "Counts the number of PREFETCHT0 instructions executed.", "SampleAfterValue": "100003", "Speculative": "1", - "UMask": "0x1" + "UMask": "0x2" }, { - "BriefDescription": "Counts demand instruction fetches and L1 instruction cache prefetches that hit a cacheline in the L3 where a snoop was sent but no other cores had the data.", + "BriefDescription": "Number of PREFETCHT1 or PREFETCHT2 instructions executed.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0xB7, 0xBB", - "EventName": "OCR.DEMAND_CODE_RD.L3_HIT.SNOOP_MISS", - "MSRIndex": "0x1a6,0x1a7", - "MSRValue": "0x02003C0004", - "Offcore": "1", + "EventCode": "0x32", + "EventName": "SW_PREFETCH_ACCESS.T1_T2", "PEBScounters": "0,1,2,3", - "PublicDescription": "Offcore response can be programmed only with a specific pair of event select and counter MSR, and with specific event codes and predefine mask bit value in a dedicated MSR to specify attributes of the offcore transaction.", + "PublicDescription": "Counts the number of PREFETCHT1 or PREFETCHT2 instructions executed.", "SampleAfterValue": "100003", "Speculative": "1", + "UMask": "0x4" + }, + { + "BriefDescription": "TMA slots where no uops were being issued due to lack of back-end resources.", + "CollectPEBSRecord": "2", + "Counter": "0,1,2,3,4,5,6,7", + "EventCode": "0xa4", + "EventName": "TOPDOWN.BACKEND_BOUND_SLOTS", + "PEBScounters": "0,1,2,3,4,5,6,7", + "PublicDescription": "Counts the number of Top-down Microarchitecture Analysis (TMA) method's slots where no micro-operations were being issued from front-end to back-end of the machine due to lack of back-end resources.", + "SampleAfterValue": "10000003", + "Speculative": "1", + "UMask": "0x2" + }, + { + "BriefDescription": "TMA slots wasted due to incorrect speculation by branch mispredictions", + "CollectPEBSRecord": "2", + "Counter": "0,1,2,3,4,5,6,7", + "EventCode": "0xa4", + "EventName": "TOPDOWN.BR_MISPREDICT_SLOTS", + "PEBScounters": "0,1,2,3,4,5,6,7", + "PublicDescription": "Number of TMA slots that were wasted due to incorrect speculation by branch mispredictions. This event estimates number of operations that were issued but not retired from the specualtive path as well as the out-of-order engine recovery past a branch misprediction.", + "SampleAfterValue": "10000003", + "Speculative": "1", + "UMask": "0x8" + }, + { + "BriefDescription": "TMA slots available for an unhalted logical processor. Fixed counter - architectural event", + "CollectPEBSRecord": "2", + "Counter": "Fixed counter 3", + "EventName": "TOPDOWN.SLOTS", + "PEBScounters": "35", + "PublicDescription": "Number of available slots for an unhalted logical processor. The event increments by machine-width of the narrowest pipeline as employed by the Top-down Microarchitecture Analysis method (TMA). The count is distributed among unhalted logical processors (hyper-threads) who share the same physical core. Software can use this event as the denominator for the top-level metrics of the TMA method. This architectural event is counted on a designated fixed counter (Fixed Counter 3).", + "SampleAfterValue": "10000003", + "Speculative": "1", + "UMask": "0x4" + }, + { + "BriefDescription": "TMA slots available for an unhalted logical processor. General counter - architectural event", + "CollectPEBSRecord": "2", + "Counter": "0,1,2,3,4,5,6,7", + "EventCode": "0xa4", + "EventName": "TOPDOWN.SLOTS_P", + "PEBScounters": "0,1,2,3,4,5,6,7", + "PublicDescription": "Counts the number of available slots for an unhalted logical processor. The event increments by machine-width of the narrowest pipeline as employed by the Top-down Microarchitecture Analysis method. The count is distributed among unhalted logical processors (hyper-threads) who share the same physical core.", + "SampleAfterValue": "10000003", + "Speculative": "1", "UMask": "0x1" } ]
\ No newline at end of file diff --git a/tools/perf/pmu-events/arch/x86/icelake/pipeline.json b/tools/perf/pmu-events/arch/x86/icelake/pipeline.json index 4f4ce309c2f8..2b305bdc8cfc 100644 --- a/tools/perf/pmu-events/arch/x86/icelake/pipeline.json +++ b/tools/perf/pmu-events/arch/x86/icelake/pipeline.json @@ -1,50 +1,39 @@ [ { - "BriefDescription": "Mispredicted indirect CALL instructions retired.", - "CollectPEBSRecord": "2", - "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xc5", - "EventName": "BR_MISP_RETIRED.INDIRECT_CALL", - "PEBS": "1", - "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts retired mispredicted indirect (near taken) CALL instructions, including both register and memory indirect.", - "SampleAfterValue": "50021", - "UMask": "0x2" - }, - { - "BriefDescription": "Number of uops executed on the core.", + "BriefDescription": "Cycles when divide unit is busy executing divide or square root operations.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xB1", - "EventName": "UOPS_EXECUTED.CORE", + "CounterMask": "1", + "EventCode": "0x14", + "EventName": "ARITH.DIVIDER_ACTIVE", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts the number of uops executed from any thread.", - "SampleAfterValue": "2000003", + "PublicDescription": "Counts cycles when divide unit is busy executing divide or square root operations. Accounts for integer and floating-point operations.", + "SampleAfterValue": "1000003", "Speculative": "1", - "UMask": "0x2" + "UMask": "0x9" }, { - "BriefDescription": "Number of uops executed on port 4 and 9", + "BriefDescription": "All branch instructions retired.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xa1", - "EventName": "UOPS_DISPATCHED.PORT_4_9", + "EventCode": "0xc4", + "EventName": "BR_INST_RETIRED.ALL_BRANCHES", + "PEBS": "1", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts, on the per-thread basis, cycles during which at least one uop is dispatched from the Reservation Station (RS) to ports 5 and 9.", - "SampleAfterValue": "2000003", - "Speculative": "1", - "UMask": "0x10" + "PublicDescription": "Counts all branch instructions retired.", + "SampleAfterValue": "400009" }, { - "BriefDescription": "Counts the number of uops to be executed per-thread each cycle.", + "BriefDescription": "Conditional branch instructions retired.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xb1", - "EventName": "UOPS_EXECUTED.THREAD", + "EventCode": "0xc4", + "EventName": "BR_INST_RETIRED.COND", + "PEBS": "1", "PEBScounters": "0,1,2,3,4,5,6,7", - "SampleAfterValue": "2000003", - "Speculative": "1", - "UMask": "0x1" + "PublicDescription": "Counts conditional branch instructions retired.", + "SampleAfterValue": "400009", + "UMask": "0x11" }, { "BriefDescription": "Not taken branch instructions retired.", @@ -59,66 +48,64 @@ "UMask": "0x10" }, { - "BriefDescription": "Uops inserted at issue-stage in order to preserve upper bits of vector registers.", + "BriefDescription": "Taken conditional branch instructions retired.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0x0e", - "EventName": "UOPS_ISSUED.VECTOR_WIDTH_MISMATCH", + "EventCode": "0xc4", + "EventName": "BR_INST_RETIRED.COND_TAKEN", + "PEBS": "1", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts the number of Blend Uops issued by the Resource Allocation Table (RAT) to the reservation station (RS) in order to preserve upper bits of vector registers. Starting with the Skylake microarchitecture, these Blend uops are needed since every Intel SSE instruction executed in Dirty Upper State needs to preserve bits 128-255 of the destination register. For more information, refer to Mixing Intel AVX and Intel SSE Code section of the Optimization Guide.", - "SampleAfterValue": "100003", - "Speculative": "1", - "UMask": "0x2" + "PublicDescription": "Counts taken conditional branch instructions retired.", + "SampleAfterValue": "400009", + "UMask": "0x1" }, { - "BriefDescription": "Counts number of cycles no uops were dispatched to be executed on this thread.", + "BriefDescription": "Far branch instructions retired.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "CounterMask": "1", - "EventCode": "0xB1", - "EventName": "UOPS_EXECUTED.STALL_CYCLES", - "Invert": "1", + "EventCode": "0xc4", + "EventName": "BR_INST_RETIRED.FAR_BRANCH", + "PEBS": "1", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts cycles during which no uops were dispatched from the Reservation Station (RS) per thread.", - "SampleAfterValue": "2000003", - "Speculative": "1", - "UMask": "0x1" + "PublicDescription": "Counts far branch instructions retired.", + "SampleAfterValue": "100007", + "UMask": "0x40" }, { - "BriefDescription": "All indirect branch instructions retired (excluding RETs. TSX aborts are considered indirect branch).", + "BriefDescription": "Indirect near branch instructions retired (excluding returns)", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc4", "EventName": "BR_INST_RETIRED.INDIRECT", "PEBS": "1", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts all indirect branch instructions retired (excluding RETs. TSX aborts is considered indirect branch).", + "PublicDescription": "Counts near indirect branch instructions retired excluding returns. TSX abort is an indirect branch.", "SampleAfterValue": "100003", "UMask": "0x80" }, { - "BriefDescription": "Cycles total of 4 uops are executed on all ports and Reservation Station was not empty.", + "BriefDescription": "Direct and indirect near call instructions retired.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xa6", - "EventName": "EXE_ACTIVITY.4_PORTS_UTIL", + "EventCode": "0xc4", + "EventName": "BR_INST_RETIRED.NEAR_CALL", + "PEBS": "1", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Cycles total of 4 uops are executed on all ports and Reservation Station (RS) was not empty.", - "SampleAfterValue": "2000003", - "Speculative": "1", - "UMask": "0x10" + "PublicDescription": "Counts both direct and indirect near call instructions retired.", + "SampleAfterValue": "100007", + "UMask": "0x2" }, { - "BriefDescription": "Number of uops executed on port 2 and 3", + "BriefDescription": "Return instructions retired.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xa1", - "EventName": "UOPS_DISPATCHED.PORT_2_3", + "EventCode": "0xc4", + "EventName": "BR_INST_RETIRED.NEAR_RETURN", + "PEBS": "1", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts, on the per-thread basis, cycles during which at least one uop is dispatched from the Reservation Station (RS) to ports 2 and 3.", - "SampleAfterValue": "2000003", - "Speculative": "1", - "UMask": "0x4" + "PublicDescription": "Counts return instructions retired.", + "SampleAfterValue": "100007", + "UMask": "0x8" }, { "BriefDescription": "Taken branch instructions retired.", @@ -133,188 +120,192 @@ "UMask": "0x20" }, { - "BriefDescription": "Counts the number of demand load dispatches that hit L1D fill buffer (FB) allocated for software prefetch.", + "BriefDescription": "All mispredicted branch instructions retired.", "CollectPEBSRecord": "2", - "Counter": "0,1,2,3", - "EventCode": "0x4c", - "EventName": "LOAD_HIT_PREFETCH.SWPF", - "PEBScounters": "0,1,2,3", - "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.", - "SampleAfterValue": "100003", - "Speculative": "1", - "UMask": "0x1" + "Counter": "0,1,2,3,4,5,6,7", + "EventCode": "0xc5", + "EventName": "BR_MISP_RETIRED.ALL_BRANCHES", + "PEBS": "1", + "PEBScounters": "0,1,2,3,4,5,6,7", + "PublicDescription": "Counts all the retired branch instructions that were mispredicted by the processor. A branch misprediction occurs when the processor incorrectly predicts the destination of the branch. When the misprediction is discovered at execution, all the instructions executed in the wrong (speculative) path must be discarded, and the processor must start fetching from the correct path.", + "SampleAfterValue": "50021" }, { - "BriefDescription": "Number of uops executed on port 1", + "BriefDescription": "Mispredicted conditional branch instructions retired.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xa1", - "EventName": "UOPS_DISPATCHED.PORT_1", + "EventCode": "0xc5", + "EventName": "BR_MISP_RETIRED.COND", + "PEBS": "1", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts, on the per-thread basis, cycles during which at least one uop is dispatched from the Reservation Station (RS) to port 1.", - "SampleAfterValue": "2000003", - "Speculative": "1", - "UMask": "0x2" + "PublicDescription": "Counts mispredicted conditional branch instructions retired.", + "SampleAfterValue": "50021", + "UMask": "0x11" }, { - "BriefDescription": "Number of Uops delivered by the LSD.", + "BriefDescription": "Mispredicted non-taken conditional branch instructions retired.", "CollectPEBSRecord": "2", - "Counter": "0,1,2,3", - "EventCode": "0xa8", - "EventName": "LSD.UOPS", - "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts the number of uops delivered to the back-end by the LSD(Loop Stream Detector).", - "SampleAfterValue": "2000003", - "Speculative": "1", - "UMask": "0x1" + "Counter": "0,1,2,3,4,5,6,7", + "EventCode": "0xc5", + "EventName": "BR_MISP_RETIRED.COND_NTAKEN", + "PEBS": "1", + "PEBScounters": "0,1,2,3,4,5,6,7", + "PublicDescription": "Counts the number of conditional branch instructions retired that were mispredicted and the branch direction was not taken.", + "SampleAfterValue": "50021", + "UMask": "0x10" }, { - "BriefDescription": "Number of uops executed on port 5", + "BriefDescription": "number of branch instructions retired that were mispredicted and taken. Non PEBS", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xa1", - "EventName": "UOPS_DISPATCHED.PORT_5", + "EventCode": "0xc5", + "EventName": "BR_MISP_RETIRED.COND_TAKEN", + "PEBS": "1", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts, on the per-thread basis, cycles during which at least one uop is dispatched from the Reservation Station (RS) to port 5.", - "SampleAfterValue": "2000003", - "Speculative": "1", - "UMask": "0x20" + "PublicDescription": "Counts taken conditional mispredicted branch instructions retired.", + "SampleAfterValue": "50021", + "UMask": "0x1" }, { - "BriefDescription": "Number of uops executed on port 6", + "BriefDescription": "All miss-predicted indirect branch instructions retired (excluding RETs. TSX aborts is considered indirect branch).", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xa1", - "EventName": "UOPS_DISPATCHED.PORT_6", + "EventCode": "0xc5", + "EventName": "BR_MISP_RETIRED.INDIRECT", + "PEBS": "1", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts, on the per-thread basis, cycles during which at least one uop is dispatched from the Reservation Station (RS) to port 6.", - "SampleAfterValue": "2000003", - "Speculative": "1", - "UMask": "0x40" + "PublicDescription": "Counts all miss-predicted indirect branch instructions retired (excluding RETs. TSX aborts is considered indirect branch).", + "SampleAfterValue": "50021", + "UMask": "0x80" }, { - "BriefDescription": "Cycles Uops delivered by the LSD, but didn't come from the decoder.", + "BriefDescription": "Mispredicted indirect CALL instructions retired.", "CollectPEBSRecord": "2", - "Counter": "0,1,2,3", - "CounterMask": "1", - "EventCode": "0xA8", - "EventName": "LSD.CYCLES_ACTIVE", - "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts the cycles when at least one uop is delivered by the LSD (Loop-stream detector).", - "SampleAfterValue": "2000003", - "Speculative": "1", - "UMask": "0x1" + "Counter": "0,1,2,3,4,5,6,7", + "EventCode": "0xc5", + "EventName": "BR_MISP_RETIRED.INDIRECT_CALL", + "PEBS": "1", + "PEBScounters": "0,1,2,3,4,5,6,7", + "PublicDescription": "Counts retired mispredicted indirect (near taken) CALL instructions, including both register and memory indirect.", + "SampleAfterValue": "50021", + "UMask": "0x2" }, { - "BriefDescription": "Core cycles the allocator was stalled due to recovery from earlier clear event for this thread", + "BriefDescription": "Number of near branch instructions retired that were mispredicted and taken.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0x0D", - "EventName": "INT_MISC.RECOVERY_CYCLES", + "EventCode": "0xc5", + "EventName": "BR_MISP_RETIRED.NEAR_TAKEN", + "PEBS": "1", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts core cycles when the Resource allocator was stalled due to recovery from an earlier branch misprediction or machine clear event.", - "SampleAfterValue": "500009", - "Speculative": "1", - "UMask": "0x1" + "PublicDescription": "Counts number of near branch instructions retired that were mispredicted and taken.", + "SampleAfterValue": "50021", + "UMask": "0x20" }, { - "BriefDescription": "Cycles where the Store Buffer was full and no loads caused an execution stall.", + "BriefDescription": "Cycle counts are evenly distributed between active threads in the Core.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "CounterMask": "2", - "EventCode": "0xA6", - "EventName": "EXE_ACTIVITY.BOUND_ON_STORES", + "EventCode": "0xec", + "EventName": "CPU_CLK_UNHALTED.DISTRIBUTED", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts cycles where the Store Buffer was full and no loads caused an execution stall.", - "SampleAfterValue": "1000003", + "PublicDescription": "This event distributes cycle counts between active hyperthreads, i.e., those in C0. A hyperthread becomes inactive when it executes the HLT or MWAIT instructions. If all other hyperthreads are inactive (or disabled or do not exist), all counts are attributed to this hyperthread. To obtain the full count when the Core is active, sum the counts from each hyperthread.", + "SampleAfterValue": "2000003", "Speculative": "1", - "UMask": "0x40" + "UMask": "0x2" }, { - "BriefDescription": "Core crystal clock cycles when the thread is unhalted.", + "BriefDescription": "Core crystal clock cycles when this thread is unhalted and the other thread is halted.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0x3C", - "EventName": "CPU_CLK_UNHALTED.REF_XCLK", + "EventName": "CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts core crystal clock cycles when the thread is unhalted.", + "PublicDescription": "Counts Core crystal clock cycles when current thread is unhalted and the other thread is halted.", "SampleAfterValue": "25003", "Speculative": "1", - "UMask": "0x1" + "UMask": "0x2" }, { - "BriefDescription": "Stalls caused by changing prefix length of the instruction.", + "BriefDescription": "Core crystal clock cycles. Cycle counts are evenly distributed between active threads in the Core.", "CollectPEBSRecord": "2", - "Counter": "0,1,2,3", - "EventCode": "0x87", - "EventName": "ILD_STALL.LCP", - "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts cycles that the Instruction Length decoder (ILD) stalls occurred due to dynamically changing prefix length of the decoded instruction (by operand size prefix instruction 0x66, address size prefix instruction 0x67 or REX.W for Intel64). Count is proportional to the number of prefixes in a 16B-line. This may result in a three-cycle penalty for each LCP (Length changing prefix) in a 16-byte chunk.", - "SampleAfterValue": "500009", + "Counter": "0,1,2,3,4,5,6,7", + "EventCode": "0x3c", + "EventName": "CPU_CLK_UNHALTED.REF_DISTRIBUTED", + "PEBScounters": "0,1,2,3,4,5,6,7", + "PublicDescription": "This event distributes Core crystal clock cycle counts between active hyperthreads, i.e., those in C0 sleep-state. A hyperthread becomes inactive when it executes the HLT or MWAIT instructions. If one thread is active in a core, all counts are attributed to this hyperthread. To obtain the full count when the Core is active, sum the counts from each hyperthread.", + "SampleAfterValue": "2000003", "Speculative": "1", - "UMask": "0x1" + "UMask": "0x8" }, { - "BriefDescription": "False dependencies in MOB due to partial compare on address.", + "BriefDescription": "Reference cycles when the core is not in halt state.", "CollectPEBSRecord": "2", - "Counter": "0,1,2,3", - "EventCode": "0x07", - "EventName": "LD_BLOCKS_PARTIAL.ADDRESS_ALIAS", - "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts the number of times a load got blocked due to false dependencies in MOB due to partial compare on address.", - "SampleAfterValue": "100003", + "Counter": "Fixed counter 2", + "EventName": "CPU_CLK_UNHALTED.REF_TSC", + "PEBScounters": "34", + "PublicDescription": "Counts the number of reference cycles when the core is not in a halt state. The core enters the halt state when it is running the HLT instruction or the MWAIT instruction. This event is not affected by core frequency changes (for example, P states, TM2 transitions) but has the same incrementing frequency as the time stamp counter. This event can approximate elapsed time while the core was not in a halt state. This event has a constant ratio with the CPU_CLK_UNHALTED.REF_XCLK event. It is counted on a dedicated fixed counter, leaving the eight programmable counters available for other events. Note: On all current platforms this event stops counting during 'throttling (TM)' states duty off periods the processor is 'halted'. The counter update is done at a lower clock rate then the core clock the overflow status bit for this counter may appear 'sticky'. After the counter has overflowed and software clears the overflow status bit and resets the counter to less than MAX. The reset value to the counter is not clocked immediately so the overflow status bit will flip 'high (1)' and generate another PMI (if enabled) after which the reset value gets clocked into the counter. Therefore, software will get the interrupt, read the overflow status bit '1 for bit 34 while the counter value is less than MAX. Software should ignore this case.", + "SampleAfterValue": "2000003", "Speculative": "1", - "UMask": "0x1" + "UMask": "0x3" }, { - "BriefDescription": "Cycles when Reservation Station (RS) is empty for the thread", + "BriefDescription": "Core crystal clock cycles when the thread is unhalted.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0x5e", - "EventName": "RS_EVENTS.EMPTY_CYCLES", + "EventCode": "0x3C", + "EventName": "CPU_CLK_UNHALTED.REF_XCLK", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts cycles during which the reservation station (RS) is empty for this logical processor. This is usually caused when the front-end pipeline runs into stravation periods (e.g. branch mispredictions or i-cache misses)", - "SampleAfterValue": "1000003", + "PublicDescription": "Counts core crystal clock cycles when the thread is unhalted.", + "SampleAfterValue": "25003", "Speculative": "1", "UMask": "0x1" }, { - "BriefDescription": "Loads blocked due to overlapping with a preceding store that cannot be forwarded.", + "BriefDescription": "Core cycles when the thread is not in halt state", "CollectPEBSRecord": "2", - "Counter": "0,1,2,3", - "EventCode": "0x03", - "EventName": "LD_BLOCKS.STORE_FORWARD", - "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts the number of times where store forwarding was prevented for a load operation. The most common case is a load blocked due to the address of memory access (partially) overlapping with a preceding uncompleted store. Note: See the table of not supported store forwards in the Optimization Guide.", - "SampleAfterValue": "100003", + "Counter": "Fixed counter 1", + "EventName": "CPU_CLK_UNHALTED.THREAD", + "PEBScounters": "33", + "PublicDescription": "Counts the number of core cycles while the thread is not in a halt state. The thread enters the halt state when it is running the HLT instruction. This event is a component in many key event ratios. The core frequency may change from time to time due to transitions associated with Enhanced Intel SpeedStep Technology or TM2. For this reason this event may have a changing ratio with regards to time. When the core frequency is constant, this event can approximate elapsed time while the core was not in the halt state. It is counted on a dedicated fixed counter, leaving the eight programmable counters available for other events.", + "SampleAfterValue": "2000003", "Speculative": "1", "UMask": "0x2" }, { - "BriefDescription": "Cycles without actually retired uops.", + "BriefDescription": "Thread cycles when thread is not in halt state", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "CounterMask": "1", - "EventCode": "0xc2", - "EventName": "UOPS_RETIRED.STALL_CYCLES", - "Invert": "1", + "EventCode": "0x3C", + "EventName": "CPU_CLK_UNHALTED.THREAD_P", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "This event counts cycles without actually retired uops.", + "PublicDescription": "This is an architectural event that counts the number of thread cycles while the thread is not in a halt state. The thread enters the halt state when it is running the HLT instruction. The core frequency may change from time to time due to power or thermal throttling. For this reason, this event may have a changing ratio with regards to wall clock time.", + "SampleAfterValue": "2000003", + "Speculative": "1" + }, + { + "BriefDescription": "Cycles while L1 cache miss demand load is outstanding.", + "CollectPEBSRecord": "2", + "Counter": "0,1,2,3", + "CounterMask": "8", + "EventCode": "0xA3", + "EventName": "CYCLE_ACTIVITY.CYCLES_L1D_MISS", + "PEBScounters": "0,1,2,3", "SampleAfterValue": "1000003", "Speculative": "1", - "UMask": "0x2" + "UMask": "0x8" }, { - "BriefDescription": "Far branch instructions retired.", + "BriefDescription": "Cycles while L2 cache miss demand load is outstanding.", "CollectPEBSRecord": "2", - "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xc4", - "EventName": "BR_INST_RETIRED.FAR_BRANCH", - "PEBS": "1", - "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts far branch instructions retired.", - "SampleAfterValue": "100007", - "UMask": "0x40" + "Counter": "0,1,2,3", + "CounterMask": "1", + "EventCode": "0xA3", + "EventName": "CYCLE_ACTIVITY.CYCLES_L2_MISS", + "PEBScounters": "0,1,2,3", + "SampleAfterValue": "1000003", + "Speculative": "1", + "UMask": "0x1" }, { "BriefDescription": "Cycles while memory subsystem has an outstanding load.", @@ -329,160 +320,169 @@ "UMask": "0x10" }, { - "BriefDescription": "Number of instructions retired. Fixed Counter - architectural event", + "BriefDescription": "Execution stalls while L1 cache miss demand load is outstanding.", "CollectPEBSRecord": "2", - "Counter": "32", - "EventName": "INST_RETIRED.ANY", - "PEBS": "1", - "PEBScounters": "32", - "PublicDescription": "Counts the number of X86 instructions retired - an Architectural PerfMon event. Counting continues during hardware interrupts, traps, and inside interrupt handlers. Notes: INST_RETIRED.ANY is counted by a designated fixed counter freeing up programmable counters to count other events. INST_RETIRED.ANY_P is counted by a programmable counter.", - "SampleAfterValue": "2000003", - "UMask": "0x1" + "Counter": "0,1,2,3", + "CounterMask": "12", + "EventCode": "0xA3", + "EventName": "CYCLE_ACTIVITY.STALLS_L1D_MISS", + "PEBScounters": "0,1,2,3", + "SampleAfterValue": "1000003", + "Speculative": "1", + "UMask": "0xc" }, { - "BriefDescription": "Counts cycles where the pipeline is stalled due to serializing operations.", + "BriefDescription": "Execution stalls while L2 cache miss demand load is outstanding.", "CollectPEBSRecord": "2", - "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xa2", - "EventName": "RESOURCE_STALLS.SCOREBOARD", - "PEBScounters": "0,1,2,3,4,5,6,7", - "SampleAfterValue": "100003", + "Counter": "0,1,2,3", + "CounterMask": "5", + "EventCode": "0xa3", + "EventName": "CYCLE_ACTIVITY.STALLS_L2_MISS", + "PEBScounters": "0,1,2,3", + "SampleAfterValue": "1000003", "Speculative": "1", - "UMask": "0x2" + "UMask": "0x5" }, { - "BriefDescription": "Increments whenever there is an update to the LBR array.", + "BriefDescription": "Execution stalls while memory subsystem has an outstanding load.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xcc", - "EventName": "MISC_RETIRED.LBR_INSERTS", + "CounterMask": "20", + "EventCode": "0xa3", + "EventName": "CYCLE_ACTIVITY.STALLS_MEM_ANY", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Increments when an entry is added to the Last Branch Record (LBR) array (or removed from the array in case of RETURNs in call stack mode). The event requires LBR enable via IA32_DEBUGCTL MSR and branch type selection via MSR_LBR_SELECT.", - "SampleAfterValue": "100003", - "UMask": "0x20" + "SampleAfterValue": "1000003", + "Speculative": "1", + "UMask": "0x14" }, { - "BriefDescription": "Number of instructions retired. General Counter - architectural event", + "BriefDescription": "Total execution stalls.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xc0", - "EventName": "INST_RETIRED.ANY_P", - "PEBS": "1", + "CounterMask": "4", + "EventCode": "0xa3", + "EventName": "CYCLE_ACTIVITY.STALLS_TOTAL", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts the number of X86 instructions retired - an Architectural PerfMon event. Counting continues during hardware interrupts, traps, and inside interrupt handlers. Notes: INST_RETIRED.ANY is counted by a designated fixed counter freeing up programmable counters to count other events. INST_RETIRED.ANY_P is counted by a programmable counter.", - "SampleAfterValue": "2000003" + "SampleAfterValue": "1000003", + "Speculative": "1", + "UMask": "0x4" }, { - "BriefDescription": "Counts the number of x87 uops dispatched.", + "BriefDescription": "Cycles total of 1 uop is executed on all ports and Reservation Station was not empty.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xB1", - "EventName": "UOPS_EXECUTED.X87", + "EventCode": "0xa6", + "EventName": "EXE_ACTIVITY.1_PORTS_UTIL", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts the number of x87 uops executed.", + "PublicDescription": "Counts cycles during which a total of 1 uop was executed on all ports and Reservation Station (RS) was not empty.", "SampleAfterValue": "2000003", "Speculative": "1", - "UMask": "0x10" + "UMask": "0x2" }, { - "BriefDescription": "Cycles at least 2 micro-op is executed from any thread on physical core.", + "BriefDescription": "Cycles total of 2 uops are executed on all ports and Reservation Station was not empty.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "CounterMask": "2", - "EventCode": "0xB1", - "EventName": "UOPS_EXECUTED.CORE_CYCLES_GE_2", + "EventCode": "0xa6", + "EventName": "EXE_ACTIVITY.2_PORTS_UTIL", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts cycles when at least 2 micro-ops are executed from any thread on physical core.", + "PublicDescription": "Counts cycles during which a total of 2 uops were executed on all ports and Reservation Station (RS) was not empty.", "SampleAfterValue": "2000003", "Speculative": "1", - "UMask": "0x2" + "UMask": "0x4" }, { - "BriefDescription": "Cycles stalled due to no store buffers available. (not including draining form sync).", + "BriefDescription": "Cycles total of 3 uops are executed on all ports and Reservation Station was not empty.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xa2", - "EventName": "RESOURCE_STALLS.SB", + "EventCode": "0xa6", + "EventName": "EXE_ACTIVITY.3_PORTS_UTIL", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts allocation stall cycles caused by the store buffer (SB) being full. This counts cycles that the pipeline back-end blocked uop delivery from the front-end.", - "SampleAfterValue": "100003", + "PublicDescription": "Cycles total of 3 uops are executed on all ports and Reservation Station (RS) was not empty.", + "SampleAfterValue": "2000003", "Speculative": "1", "UMask": "0x8" }, { - "BriefDescription": "The number of times that split load operations are temporarily blocked because all resources for handling the split accesses are in use.", + "BriefDescription": "Cycles total of 4 uops are executed on all ports and Reservation Station was not empty.", "CollectPEBSRecord": "2", - "Counter": "0,1,2,3", - "EventCode": "0x03", - "EventName": "LD_BLOCKS.NO_SR", - "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts the number of times that split load operations are temporarily blocked because all resources for handling the split accesses are in use.", - "SampleAfterValue": "100003", + "Counter": "0,1,2,3,4,5,6,7", + "EventCode": "0xa6", + "EventName": "EXE_ACTIVITY.4_PORTS_UTIL", + "PEBScounters": "0,1,2,3,4,5,6,7", + "PublicDescription": "Cycles total of 4 uops are executed on all ports and Reservation Station (RS) was not empty.", + "SampleAfterValue": "2000003", "Speculative": "1", - "UMask": "0x8" + "UMask": "0x10" }, { - "BriefDescription": "Number of machine clears (nukes) of any type.", + "BriefDescription": "Cycles where the Store Buffer was full and no loads caused an execution stall.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "CounterMask": "1", - "EdgeDetect": "1", - "EventCode": "0xc3", - "EventName": "MACHINE_CLEARS.COUNT", + "CounterMask": "2", + "EventCode": "0xA6", + "EventName": "EXE_ACTIVITY.BOUND_ON_STORES", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts the number of machine clears (nukes) of any type.", - "SampleAfterValue": "100003", + "PublicDescription": "Counts cycles where the Store Buffer was full and no loads caused an execution stall.", + "SampleAfterValue": "1000003", + "Speculative": "1", + "UMask": "0x40" + }, + { + "BriefDescription": "Stalls caused by changing prefix length of the instruction.", + "CollectPEBSRecord": "2", + "Counter": "0,1,2,3", + "EventCode": "0x87", + "EventName": "ILD_STALL.LCP", + "PEBScounters": "0,1,2,3", + "PublicDescription": "Counts cycles that the Instruction Length decoder (ILD) stalls occurred due to dynamically changing prefix length of the decoded instruction (by operand size prefix instruction 0x66, address size prefix instruction 0x67 or REX.W for Intel64). Count is proportional to the number of prefixes in a 16B-line. This may result in a three-cycle penalty for each LCP (Length changing prefix) in a 16-byte chunk.", + "SampleAfterValue": "500009", "Speculative": "1", "UMask": "0x1" }, { - "BriefDescription": "Number of near branch instructions retired that were mispredicted and taken.", + "BriefDescription": "Number of instructions retired. Fixed Counter - architectural event", "CollectPEBSRecord": "2", - "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xc5", - "EventName": "BR_MISP_RETIRED.NEAR_TAKEN", + "Counter": "Fixed counter 0", + "EventName": "INST_RETIRED.ANY", "PEBS": "1", - "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts number of near branch instructions retired that were mispredicted and taken.", - "SampleAfterValue": "50021", - "UMask": "0x20" + "PEBScounters": "32", + "PublicDescription": "Counts the number of instructions retired - an Architectural PerfMon event. Counting continues during hardware interrupts, traps, and inside interrupt handlers. Notes: INST_RETIRED.ANY is counted by a designated fixed counter freeing up programmable counters to count other events. INST_RETIRED.ANY_P is counted by a programmable counter.", + "SampleAfterValue": "2000003", + "UMask": "0x1" }, { - "BriefDescription": "Return instructions retired.", + "BriefDescription": "Number of instructions retired. General Counter - architectural event", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xc4", - "EventName": "BR_INST_RETIRED.NEAR_RETURN", + "EventCode": "0xc0", + "EventName": "INST_RETIRED.ANY_P", "PEBS": "1", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts return instructions retired.", - "SampleAfterValue": "100007", - "UMask": "0x8" + "PublicDescription": "Counts the number of instructions retired - an Architectural PerfMon event. Counting continues during hardware interrupts, traps, and inside interrupt handlers. Notes: INST_RETIRED.ANY is counted by a designated fixed counter freeing up programmable counters to count other events. INST_RETIRED.ANY_P is counted by a programmable counter.", + "SampleAfterValue": "2000003" }, { - "BriefDescription": "Cycles when divide unit is busy executing divide or square root operations.", + "BriefDescription": "Number of all retired NOP instructions.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "CounterMask": "1", - "EventCode": "0x14", - "EventName": "ARITH.DIVIDER_ACTIVE", + "EventCode": "0xc0", + "EventName": "INST_RETIRED.NOP", + "PEBS": "1", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts cycles when divide unit is busy executing divide or square root operations. Accounts for integer and floating-point operations.", - "SampleAfterValue": "1000003", - "Speculative": "1", - "UMask": "0x9" + "SampleAfterValue": "2000003", + "UMask": "0x2" }, { - "BriefDescription": "Cycles total of 1 uop is executed on all ports and Reservation Station was not empty.", + "BriefDescription": "Precise instruction retired event with a reduced effect of PEBS shadow in IP distribution", "CollectPEBSRecord": "2", - "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xa6", - "EventName": "EXE_ACTIVITY.1_PORTS_UTIL", - "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts cycles during which a total of 1 uop was executed on all ports and Reservation Station (RS) was not empty.", + "Counter": "Fixed counter 0", + "EventName": "INST_RETIRED.PREC_DIST", + "PEBS": "1", + "PEBScounters": "32", + "PublicDescription": "A version of INST_RETIRED that allows for a more unbiased distribution of samples across instructions retired. It utilizes the Precise Distribution of Instructions Retired (PDIR) feature to mitigate some bias in how retired instructions get sampled. Use on Fixed Counter 0.", "SampleAfterValue": "2000003", - "Speculative": "1", - "UMask": "0x2" + "UMask": "0x1" }, { "BriefDescription": "Cycles without actually retired instructions.", @@ -499,241 +499,221 @@ "UMask": "0x1" }, { - "BriefDescription": "Mispredicted non-taken conditional branch instructions retired.", + "BriefDescription": "Cycles the Backend cluster is recovering after a miss-speculation or a Store Buffer or Load Buffer drain stall.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xc5", - "EventName": "BR_MISP_RETIRED.COND_NTAKEN", - "PEBS": "1", + "CounterMask": "1", + "EventCode": "0x0D", + "EventName": "INT_MISC.ALL_RECOVERY_CYCLES", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts the number of conditional branch instructions retired that were mispredicted and the branch direction was not taken.", - "SampleAfterValue": "50021", - "UMask": "0x10" - }, - { - "BriefDescription": "Core cycles when the thread is not in halt state", - "CollectPEBSRecord": "2", - "Counter": "33", - "EventName": "CPU_CLK_UNHALTED.THREAD", - "PEBScounters": "33", - "PublicDescription": "Counts the number of core cycles while the thread is not in a halt state. The thread enters the halt state when it is running the HLT instruction. This event is a component in many key event ratios. The core frequency may change from time to time due to transitions associated with Enhanced Intel SpeedStep Technology or TM2. For this reason this event may have a changing ratio with regards to time. When the core frequency is constant, this event can approximate elapsed time while the core was not in the halt state. It is counted on a dedicated fixed counter, leaving the eight programmable counters available for other events.", + "PublicDescription": "Counts cycles the Backend cluster is recovering after a miss-speculation or a Store Buffer or Load Buffer drain stall.", "SampleAfterValue": "2000003", "Speculative": "1", - "UMask": "0x2" + "UMask": "0x3" }, { - "BriefDescription": "Taken conditional branch instructions retired.", + "BriefDescription": "Counts cycles after recovery from a branch misprediction or machine clear till the first uop is issued from the resteered path.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xc4", - "EventName": "BR_INST_RETIRED.COND_TAKEN", - "PEBS": "1", + "EventCode": "0x0d", + "EventName": "INT_MISC.CLEAR_RESTEER_CYCLES", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts taken conditional branch instructions retired.", - "SampleAfterValue": "400009", - "UMask": "0x1" + "PublicDescription": "Cycles after recovery from a branch misprediction or machine clear till the first uop is issued from the resteered path.", + "SampleAfterValue": "500009", + "Speculative": "1", + "UMask": "0x80" }, { - "BriefDescription": "Direct and indirect near call instructions retired.", + "BriefDescription": "Core cycles the allocator was stalled due to recovery from earlier clear event for this thread", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xc4", - "EventName": "BR_INST_RETIRED.NEAR_CALL", - "PEBS": "1", + "EventCode": "0x0D", + "EventName": "INT_MISC.RECOVERY_CYCLES", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts both direct and indirect near call instructions retired.", - "SampleAfterValue": "100007", - "UMask": "0x2" + "PublicDescription": "Counts core cycles when the Resource allocator was stalled due to recovery from an earlier branch misprediction or machine clear event.", + "SampleAfterValue": "500009", + "Speculative": "1", + "UMask": "0x1" }, { - "BriefDescription": "Cycles at least 4 micro-op is executed from any thread on physical core.", + "BriefDescription": "TMA slots where uops got dropped", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "CounterMask": "4", - "EventCode": "0xB1", - "EventName": "UOPS_EXECUTED.CORE_CYCLES_GE_4", + "EventCode": "0x0d", + "EventName": "INT_MISC.UOP_DROPPING", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts cycles when at least 4 micro-ops are executed from any thread on physical core.", - "SampleAfterValue": "2000003", + "PublicDescription": "Estimated number of Top-down Microarchitecture Analysis slots that got dropped due to non front-end reasons", + "SampleAfterValue": "1000003", "Speculative": "1", - "UMask": "0x2" - }, - { - "BriefDescription": "Precise instruction retired event with a reduced effect of PEBS shadow in IP distribution", - "CollectPEBSRecord": "2", - "Counter": "32", - "EventName": "INST_RETIRED.PREC_DIST", - "PEBS": "1", - "PEBScounters": "32", - "PublicDescription": "A version of INST_RETIRED that allows for a more unbiased distribution of samples across instructions retired. It utilizes the Precise Distribution of Instructions Retired (PDIR) feature to mitigate some bias in how retired instructions get sampled. Use on Fixed Counter 0.", - "SampleAfterValue": "2000003", - "UMask": "0x1" + "UMask": "0x10" }, { - "BriefDescription": "Total execution stalls.", + "BriefDescription": "The number of times that split load operations are temporarily blocked because all resources for handling the split accesses are in use.", "CollectPEBSRecord": "2", - "Counter": "0,1,2,3,4,5,6,7", - "CounterMask": "4", - "EventCode": "0xa3", - "EventName": "CYCLE_ACTIVITY.STALLS_TOTAL", - "PEBScounters": "0,1,2,3,4,5,6,7", - "SampleAfterValue": "1000003", + "Counter": "0,1,2,3", + "EventCode": "0x03", + "EventName": "LD_BLOCKS.NO_SR", + "PEBScounters": "0,1,2,3", + "PublicDescription": "Counts the number of times that split load operations are temporarily blocked because all resources for handling the split accesses are in use.", + "SampleAfterValue": "100003", "Speculative": "1", - "UMask": "0x4" + "UMask": "0x8" }, { - "BriefDescription": "Execution stalls while L1 cache miss demand load is outstanding.", + "BriefDescription": "Loads blocked due to overlapping with a preceding store that cannot be forwarded.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "CounterMask": "12", - "EventCode": "0xA3", - "EventName": "CYCLE_ACTIVITY.STALLS_L1D_MISS", + "EventCode": "0x03", + "EventName": "LD_BLOCKS.STORE_FORWARD", "PEBScounters": "0,1,2,3", - "SampleAfterValue": "1000003", + "PublicDescription": "Counts the number of times where store forwarding was prevented for a load operation. The most common case is a load blocked due to the address of memory access (partially) overlapping with a preceding uncompleted store. Note: See the table of not supported store forwards in the Optimization Guide.", + "SampleAfterValue": "100003", "Speculative": "1", - "UMask": "0xc" + "UMask": "0x2" }, { - "BriefDescription": "Number of retired PAUSE instructions. This event is not supported on first SKL and KBL products.", + "BriefDescription": "False dependencies due to partial compare on address.", "CollectPEBSRecord": "2", - "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xcc", - "EventName": "MISC_RETIRED.PAUSE_INST", - "PublicDescription": "Counts number of retired PAUSE instructions. This event is not supported on first SKL and KBL products.", + "Counter": "0,1,2,3", + "EventCode": "0x07", + "EventName": "LD_BLOCKS_PARTIAL.ADDRESS_ALIAS", + "PEBScounters": "0,1,2,3", + "PublicDescription": "Counts the number of times a load got blocked due to false dependencies due to partial compare on address.", "SampleAfterValue": "100003", - "UMask": "0x40" + "Speculative": "1", + "UMask": "0x1" }, { - "BriefDescription": "Self-modifying code (SMC) detected.", + "BriefDescription": "Counts the number of demand load dispatches that hit L1D fill buffer (FB) allocated for software prefetch.", "CollectPEBSRecord": "2", - "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xc3", - "EventName": "MACHINE_CLEARS.SMC", - "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts self-modifying code (SMC) detected, which causes a machine clear.", + "Counter": "0,1,2,3", + "EventCode": "0x4c", + "EventName": "LOAD_HIT_PREFETCH.SWPF", + "PEBScounters": "0,1,2,3", + "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.", "SampleAfterValue": "100003", "Speculative": "1", - "UMask": "0x4" + "UMask": "0x1" }, { - "BriefDescription": "Uops that RAT issues to RS", + "BriefDescription": "Cycles Uops delivered by the LSD, but didn't come from the decoder.", "CollectPEBSRecord": "2", - "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0x0e", - "EventName": "UOPS_ISSUED.ANY", - "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts the number of uops that the Resource Allocation Table (RAT) issues to the Reservation Station (RS).", + "Counter": "0,1,2,3", + "CounterMask": "1", + "EventCode": "0xA8", + "EventName": "LSD.CYCLES_ACTIVE", + "PEBScounters": "0,1,2,3", + "PublicDescription": "Counts the cycles when at least one uop is delivered by the LSD (Loop-stream detector).", "SampleAfterValue": "2000003", "Speculative": "1", "UMask": "0x1" }, { - "BriefDescription": "Execution stalls while L2 cache miss demand load is outstanding.", + "BriefDescription": "Cycles optimal number of Uops delivered by the LSD, but did not come from the decoder.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "CounterMask": "5", - "EventCode": "0xa3", - "EventName": "CYCLE_ACTIVITY.STALLS_L2_MISS", + "EventCode": "0xa8", + "EventName": "LSD.CYCLES_OK", "PEBScounters": "0,1,2,3", - "SampleAfterValue": "1000003", + "PublicDescription": "Counts the cycles when optimal number of uops is delivered by the LSD (Loop-stream detector).", + "SampleAfterValue": "2000003", "Speculative": "1", - "UMask": "0x5" + "UMask": "0x1" }, { - "BriefDescription": "Reference cycles when the core is not in halt state.", + "BriefDescription": "Number of Uops delivered by the LSD.", "CollectPEBSRecord": "2", - "Counter": "34", - "EventName": "CPU_CLK_UNHALTED.REF_TSC", - "PEBScounters": "34", - "PublicDescription": "Counts the number of reference cycles when the core is not in a halt state. The core enters the halt state when it is running the HLT instruction or the MWAIT instruction. This event is not affected by core frequency changes (for example, P states, TM2 transitions) but has the same incrementing frequency as the time stamp counter. This event can approximate elapsed time while the core was not in a halt state. This event has a constant ratio with the CPU_CLK_UNHALTED.REF_XCLK event. It is counted on a dedicated fixed counter, leaving the eight programmable counters available for other events. Note: On all current platforms this event stops counting during 'throttling (TM)' states duty off periods the processor is 'halted'. The counter update is done at a lower clock rate then the core clock the overflow status bit for this counter may appear 'sticky'. After the counter has overflowed and software clears the overflow status bit and resets the counter to less than MAX. The reset value to the counter is not clocked immediately so the overflow status bit will flip 'high (1)' and generate another PMI (if enabled) after which the reset value gets clocked into the counter. Therefore, software will get the interrupt, read the overflow status bit '1 for bit 34 while the counter value is less than MAX. Software should ignore this case.", + "Counter": "0,1,2,3", + "EventCode": "0xa8", + "EventName": "LSD.UOPS", + "PEBScounters": "0,1,2,3", + "PublicDescription": "Counts the number of uops delivered to the back-end by the LSD(Loop Stream Detector).", "SampleAfterValue": "2000003", "Speculative": "1", - "UMask": "0x3" + "UMask": "0x1" }, { - "BriefDescription": "Cycles the Backend cluster is recovering after a miss-speculation or a Store Buffer or Load Buffer drain stall.", + "BriefDescription": "Number of machine clears (nukes) of any type.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", "CounterMask": "1", - "EventCode": "0x0D", - "EventName": "INT_MISC.ALL_RECOVERY_CYCLES", + "EdgeDetect": "1", + "EventCode": "0xc3", + "EventName": "MACHINE_CLEARS.COUNT", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts cycles the Backend cluster is recovering after a miss-speculation or a Store Buffer or Load Buffer drain stall.", - "SampleAfterValue": "2000003", + "PublicDescription": "Counts the number of machine clears (nukes) of any type.", + "SampleAfterValue": "100003", "Speculative": "1", - "UMask": "0x3" + "UMask": "0x1" }, { - "BriefDescription": "Cycles total of 2 uops are executed on all ports and Reservation Station was not empty.", + "BriefDescription": "Self-modifying code (SMC) detected.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xa6", - "EventName": "EXE_ACTIVITY.2_PORTS_UTIL", + "EventCode": "0xc3", + "EventName": "MACHINE_CLEARS.SMC", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts cycles during which a total of 2 uops were executed on all ports and Reservation Station (RS) was not empty.", - "SampleAfterValue": "2000003", + "PublicDescription": "Counts self-modifying code (SMC) detected, which causes a machine clear.", + "SampleAfterValue": "100003", "Speculative": "1", "UMask": "0x4" }, { - "BriefDescription": "Cycles total of 3 uops are executed on all ports and Reservation Station was not empty.", + "BriefDescription": "Increments whenever there is an update to the LBR array.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xa6", - "EventName": "EXE_ACTIVITY.3_PORTS_UTIL", + "EventCode": "0xcc", + "EventName": "MISC_RETIRED.LBR_INSERTS", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Cycles total of 3 uops are executed on all ports and Reservation Station (RS) was not empty.", - "SampleAfterValue": "2000003", - "Speculative": "1", - "UMask": "0x8" + "PublicDescription": "Increments when an entry is added to the Last Branch Record (LBR) array (or removed from the array in case of RETURNs in call stack mode). The event requires LBR to be enabled properly.", + "SampleAfterValue": "100003", + "UMask": "0x20" }, { - "BriefDescription": "Cycles while L1 cache miss demand load is outstanding.", + "BriefDescription": "Number of retired PAUSE instructions. This event is not supported on first SKL and KBL products.", "CollectPEBSRecord": "2", - "Counter": "0,1,2,3", - "CounterMask": "8", - "EventCode": "0xA3", - "EventName": "CYCLE_ACTIVITY.CYCLES_L1D_MISS", - "PEBScounters": "0,1,2,3", - "SampleAfterValue": "1000003", - "Speculative": "1", - "UMask": "0x8" + "Counter": "0,1,2,3,4,5,6,7", + "EventCode": "0xcc", + "EventName": "MISC_RETIRED.PAUSE_INST", + "PublicDescription": "Counts number of retired PAUSE instructions. This event is not supported on first SKL and KBL products.", + "SampleAfterValue": "100003", + "UMask": "0x40" }, { - "BriefDescription": "Counts cycles after recovery from a branch misprediction or machine clear till the first uop is issued from the resteered path.", + "BriefDescription": "Cycles stalled due to no store buffers available. (not including draining form sync).", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0x0d", - "EventName": "INT_MISC.CLEAR_RESTEER_CYCLES", + "EventCode": "0xa2", + "EventName": "RESOURCE_STALLS.SB", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Cycles after recovery from a branch misprediction or machine clear till the first uop is issued from the resteered path.", - "SampleAfterValue": "500009", + "PublicDescription": "Counts allocation stall cycles caused by the store buffer (SB) being full. This counts cycles that the pipeline back-end blocked uop delivery from the front-end.", + "SampleAfterValue": "100003", "Speculative": "1", - "UMask": "0x80" + "UMask": "0x8" }, { - "BriefDescription": "Cycles with less than 10 actually retired uops.", + "BriefDescription": "Counts cycles where the pipeline is stalled due to serializing operations.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "CounterMask": "10", - "EventCode": "0xc2", - "EventName": "UOPS_RETIRED.TOTAL_CYCLES", - "Invert": "1", + "EventCode": "0xa2", + "EventName": "RESOURCE_STALLS.SCOREBOARD", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts the number of cycles using always true condition (uops_ret &lt; 16) applied to non PEBS uops retired event.", - "SampleAfterValue": "1000003", + "SampleAfterValue": "100003", + "Speculative": "1", "UMask": "0x2" }, { - "BriefDescription": "All branch instructions retired.", + "BriefDescription": "Cycles when Reservation Station (RS) is empty for the thread", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xc4", - "EventName": "BR_INST_RETIRED.ALL_BRANCHES", - "PEBS": "1", + "EventCode": "0x5e", + "EventName": "RS_EVENTS.EMPTY_CYCLES", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts all branch instructions retired.", - "SampleAfterValue": "400009" + "PublicDescription": "Counts cycles during which the reservation station (RS) is empty for this logical processor. This is usually caused when the front-end pipeline runs into stravation periods (e.g. branch mispredictions or i-cache misses)", + "SampleAfterValue": "1000003", + "Speculative": "1", + "UMask": "0x1" }, { "BriefDescription": "Counts end of periods where the Reservation Station (RS) was empty.", @@ -751,121 +731,174 @@ "UMask": "0x1" }, { - "BriefDescription": "Cycle counts are evenly distributed between active threads in the Core.", + "BriefDescription": "Number of uops decoded out of instructions exclusively fetched by decoder 0", + "CollectPEBSRecord": "2", + "Counter": "0,1,2,3", + "EventCode": "0x56", + "EventName": "UOPS_DECODED.DEC0", + "PEBScounters": "0,1,2,3", + "PublicDescription": "Uops exclusively fetched by decoder 0", + "SampleAfterValue": "1000003", + "Speculative": "1", + "UMask": "0x1" + }, + { + "BriefDescription": "Number of uops executed on port 0", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xec", - "EventName": "CPU_CLK_UNHALTED.DISTRIBUTED", + "EventCode": "0xa1", + "EventName": "UOPS_DISPATCHED.PORT_0", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "This event distributes cycle counts between active hyperthreads, i.e., those in C0. A hyperthread becomes inactive when it executes the HLT or MWAIT instructions. If all other hyperthreads are inactive (or disabled or do not exist), all counts are attributed to this hyperthread. To obtain the full count when the Core is active, sum the counts from each hyperthread.", + "PublicDescription": "Counts, on the per-thread basis, cycles during which at least one uop is dispatched from the Reservation Station (RS) to port 0.", "SampleAfterValue": "2000003", "Speculative": "1", - "UMask": "0x2" + "UMask": "0x1" }, { - "BriefDescription": "Core crystal clock cycles when this thread is unhalted and the other thread is halted.", + "BriefDescription": "Number of uops executed on port 1", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0x3C", - "EventName": "CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE", + "EventCode": "0xa1", + "EventName": "UOPS_DISPATCHED.PORT_1", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts Core crystal clock cycles when current thread is unhalted and the other thread is halted.", - "SampleAfterValue": "25003", + "PublicDescription": "Counts, on the per-thread basis, cycles during which at least one uop is dispatched from the Reservation Station (RS) to port 1.", + "SampleAfterValue": "2000003", "Speculative": "1", "UMask": "0x2" }, { - "BriefDescription": "Thread cycles when thread is not in halt state", + "BriefDescription": "Number of uops executed on port 2 and 3", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0x3C", - "EventName": "CPU_CLK_UNHALTED.THREAD_P", + "EventCode": "0xa1", + "EventName": "UOPS_DISPATCHED.PORT_2_3", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "This is an architectural event that counts the number of thread cycles while the thread is not in a halt state. The thread enters the halt state when it is running the HLT instruction. The core frequency may change from time to time due to power or thermal throttling. For this reason, this event may have a changing ratio with regards to wall clock time.", + "PublicDescription": "Counts, on the per-thread basis, cycles during which at least one uop is dispatched from the Reservation Station (RS) to ports 2 and 3.", "SampleAfterValue": "2000003", - "Speculative": "1" + "Speculative": "1", + "UMask": "0x4" }, { - "BriefDescription": "Mispredicted conditional branch instructions retired.", + "BriefDescription": "Number of uops executed on port 4 and 9", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xc5", - "EventName": "BR_MISP_RETIRED.COND", - "PEBS": "1", + "EventCode": "0xa1", + "EventName": "UOPS_DISPATCHED.PORT_4_9", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts mispredicted conditional branch instructions retired.", - "SampleAfterValue": "50021", - "UMask": "0x11" + "PublicDescription": "Counts, on the per-thread basis, cycles during which at least one uop is dispatched from the Reservation Station (RS) to ports 5 and 9.", + "SampleAfterValue": "2000003", + "Speculative": "1", + "UMask": "0x10" }, { - "BriefDescription": "Number of uops executed on port 0", + "BriefDescription": "Number of uops executed on port 5", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xa1", - "EventName": "UOPS_DISPATCHED.PORT_0", + "EventName": "UOPS_DISPATCHED.PORT_5", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts, on the per-thread basis, cycles during which at least one uop is dispatched from the Reservation Station (RS) to port 0.", + "PublicDescription": "Counts, on the per-thread basis, cycles during which at least one uop is dispatched from the Reservation Station (RS) to port 5.", "SampleAfterValue": "2000003", "Speculative": "1", - "UMask": "0x1" + "UMask": "0x20" }, { - "BriefDescription": "Conditional branch instructions retired.", + "BriefDescription": "Number of uops executed on port 6", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xc4", - "EventName": "BR_INST_RETIRED.COND", - "PEBS": "1", + "EventCode": "0xa1", + "EventName": "UOPS_DISPATCHED.PORT_6", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts conditional branch instructions retired.", - "SampleAfterValue": "400009", - "UMask": "0x11" + "PublicDescription": "Counts, on the per-thread basis, cycles during which at least one uop is dispatched from the Reservation Station (RS) to port 6.", + "SampleAfterValue": "2000003", + "Speculative": "1", + "UMask": "0x40" }, { - "BriefDescription": "Retirement slots used.", + "BriefDescription": "Number of uops executed on port 7 and 8", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xc2", - "EventName": "UOPS_RETIRED.SLOTS", + "EventCode": "0xa1", + "EventName": "UOPS_DISPATCHED.PORT_7_8", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts the retirement slots used each cycle.", + "PublicDescription": "Counts, on the per-thread basis, cycles during which at least one uop is dispatched from the Reservation Station (RS) to ports 7 and 8.", + "SampleAfterValue": "2000003", + "Speculative": "1", + "UMask": "0x80" + }, + { + "BriefDescription": "Number of uops executed on the core.", + "CollectPEBSRecord": "2", + "Counter": "0,1,2,3,4,5,6,7", + "EventCode": "0xB1", + "EventName": "UOPS_EXECUTED.CORE", + "PEBScounters": "0,1,2,3,4,5,6,7", + "PublicDescription": "Counts the number of uops executed from any thread.", "SampleAfterValue": "2000003", + "Speculative": "1", "UMask": "0x2" }, { - "BriefDescription": "Cycles optimal number of Uops delivered by the LSD, but did not come from the decoder.", + "BriefDescription": "Cycles at least 1 micro-op is executed from any thread on physical core.", "CollectPEBSRecord": "2", - "Counter": "0,1,2,3", - "CounterMask": "5", - "EventCode": "0xa8", - "EventName": "LSD.CYCLES_OK", - "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts the cycles when optimal number of uops is delivered by the LSD (Loop-stream detector).", + "Counter": "0,1,2,3,4,5,6,7", + "CounterMask": "1", + "EventCode": "0xB1", + "EventName": "UOPS_EXECUTED.CORE_CYCLES_GE_1", + "PEBScounters": "0,1,2,3,4,5,6,7", + "PublicDescription": "Counts cycles when at least 1 micro-op is executed from any thread on physical core.", "SampleAfterValue": "2000003", "Speculative": "1", - "UMask": "0x1" + "UMask": "0x2" }, { - "BriefDescription": "Core crystal clock cycles. Cycle counts are evenly distributed between active threads in the Core.", + "BriefDescription": "Cycles at least 2 micro-op is executed from any thread on physical core.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0x3c", - "EventName": "CPU_CLK_UNHALTED.REF_DISTRIBUTED", + "CounterMask": "2", + "EventCode": "0xB1", + "EventName": "UOPS_EXECUTED.CORE_CYCLES_GE_2", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "This event distributes Core crystal clock cycle counts between active hyperthreads, i.e., those in C0 sleep-state. A hyperthread becomes inactive when it executes the HLT or MWAIT instructions. If one thread is active in a core, all counts are attributed to this hyperthread. To obtain the full count when the Core is active, sum the counts from each hyperthread.", + "PublicDescription": "Counts cycles when at least 2 micro-ops are executed from any thread on physical core.", "SampleAfterValue": "2000003", "Speculative": "1", - "UMask": "0x8" + "UMask": "0x2" }, { - "BriefDescription": "Cycles where at least 3 uops were executed per-thread", + "BriefDescription": "Cycles at least 3 micro-op is executed from any thread on physical core.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", "CounterMask": "3", + "EventCode": "0xB1", + "EventName": "UOPS_EXECUTED.CORE_CYCLES_GE_3", + "PEBScounters": "0,1,2,3,4,5,6,7", + "PublicDescription": "Counts cycles when at least 3 micro-ops are executed from any thread on physical core.", + "SampleAfterValue": "2000003", + "Speculative": "1", + "UMask": "0x2" + }, + { + "BriefDescription": "Cycles at least 4 micro-op is executed from any thread on physical core.", + "CollectPEBSRecord": "2", + "Counter": "0,1,2,3,4,5,6,7", + "CounterMask": "4", + "EventCode": "0xB1", + "EventName": "UOPS_EXECUTED.CORE_CYCLES_GE_4", + "PEBScounters": "0,1,2,3,4,5,6,7", + "PublicDescription": "Counts cycles when at least 4 micro-ops are executed from any thread on physical core.", + "SampleAfterValue": "2000003", + "Speculative": "1", + "UMask": "0x2" + }, + { + "BriefDescription": "Cycles where at least 1 uop was executed per-thread", + "CollectPEBSRecord": "2", + "Counter": "0,1,2,3,4,5,6,7", + "CounterMask": "1", "EventCode": "0xb1", - "EventName": "UOPS_EXECUTED.CYCLES_GE_3", + "EventName": "UOPS_EXECUTED.CYCLES_GE_1", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Cycles where at least 3 uops were executed per-thread.", + "PublicDescription": "Cycles where at least 1 uop was executed per-thread.", "SampleAfterValue": "2000003", "Speculative": "1", "UMask": "0x1" @@ -884,14 +917,14 @@ "UMask": "0x1" }, { - "BriefDescription": "Cycles where at least 1 uop was executed per-thread", + "BriefDescription": "Cycles where at least 3 uops were executed per-thread", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "CounterMask": "1", + "CounterMask": "3", "EventCode": "0xb1", - "EventName": "UOPS_EXECUTED.CYCLES_GE_1", + "EventName": "UOPS_EXECUTED.CYCLES_GE_3", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Cycles where at least 1 uop was executed per-thread.", + "PublicDescription": "Cycles where at least 3 uops were executed per-thread.", "SampleAfterValue": "2000003", "Speculative": "1", "UMask": "0x1" @@ -910,126 +943,116 @@ "UMask": "0x1" }, { - "BriefDescription": "Cycles while L2 cache miss demand load is outstanding.", - "CollectPEBSRecord": "2", - "Counter": "0,1,2,3", - "CounterMask": "1", - "EventCode": "0xA3", - "EventName": "CYCLE_ACTIVITY.CYCLES_L2_MISS", - "PEBScounters": "0,1,2,3", - "SampleAfterValue": "1000003", - "Speculative": "1", - "UMask": "0x1" - }, - { - "BriefDescription": "Cycles when RAT does not issue Uops to RS for the thread", + "BriefDescription": "Counts number of cycles no uops were dispatched to be executed on this thread.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", "CounterMask": "1", - "EventCode": "0x0E", - "EventName": "UOPS_ISSUED.STALL_CYCLES", + "EventCode": "0xB1", + "EventName": "UOPS_EXECUTED.STALL_CYCLES", "Invert": "1", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts cycles during which the Resource Allocation Table (RAT) does not issue any Uops to the reservation station (RS) for the current thread.", - "SampleAfterValue": "1000003", + "PublicDescription": "Counts cycles during which no uops were dispatched from the Reservation Station (RS) per thread.", + "SampleAfterValue": "2000003", "Speculative": "1", "UMask": "0x1" }, { - "BriefDescription": "Cycles at least 3 micro-op is executed from any thread on physical core.", + "BriefDescription": "Counts the number of uops to be executed per-thread each cycle.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "CounterMask": "3", - "EventCode": "0xB1", - "EventName": "UOPS_EXECUTED.CORE_CYCLES_GE_3", + "EventCode": "0xb1", + "EventName": "UOPS_EXECUTED.THREAD", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts cycles when at least 3 micro-ops are executed from any thread on physical core.", "SampleAfterValue": "2000003", "Speculative": "1", - "UMask": "0x2" + "UMask": "0x1" }, { - "BriefDescription": "Cycles at least 1 micro-op is executed from any thread on physical core.", + "BriefDescription": "Counts the number of x87 uops dispatched.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "CounterMask": "1", "EventCode": "0xB1", - "EventName": "UOPS_EXECUTED.CORE_CYCLES_GE_1", + "EventName": "UOPS_EXECUTED.X87", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts cycles when at least 1 micro-op is executed from any thread on physical core.", + "PublicDescription": "Counts the number of x87 uops executed.", "SampleAfterValue": "2000003", "Speculative": "1", - "UMask": "0x2" + "UMask": "0x10" }, { - "BriefDescription": "All miss-predicted indirect branch instructions retired (excluding RETs. TSX aborts is considered indirect branch).", + "BriefDescription": "Uops that RAT issues to RS", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xc5", - "EventName": "BR_MISP_RETIRED.INDIRECT", - "PEBS": "1", + "EventCode": "0x0e", + "EventName": "UOPS_ISSUED.ANY", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts all miss-predicted indirect branch instructions retired (excluding RETs. TSX aborts is considered indirect branch).", - "SampleAfterValue": "50021", - "UMask": "0x80" + "PublicDescription": "Counts the number of uops that the Resource Allocation Table (RAT) issues to the Reservation Station (RS).", + "SampleAfterValue": "2000003", + "Speculative": "1", + "UMask": "0x1" }, { - "BriefDescription": "TMA slots where uops got dropped", + "BriefDescription": "Cycles when RAT does not issue Uops to RS for the thread", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0x0d", - "EventName": "INT_MISC.UOP_DROPPING", + "CounterMask": "1", + "EventCode": "0x0E", + "EventName": "UOPS_ISSUED.STALL_CYCLES", + "Invert": "1", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Estimated number of Top-down Microarchitecture Analysis slots that got dropped due to non front-end reasons", + "PublicDescription": "Counts cycles during which the Resource Allocation Table (RAT) does not issue any Uops to the reservation station (RS) for the current thread.", "SampleAfterValue": "1000003", "Speculative": "1", - "UMask": "0x10" + "UMask": "0x1" }, { - "BriefDescription": "Execution stalls while memory subsystem has an outstanding load.", + "BriefDescription": "Uops inserted at issue-stage in order to preserve upper bits of vector registers.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "CounterMask": "20", - "EventCode": "0xa3", - "EventName": "CYCLE_ACTIVITY.STALLS_MEM_ANY", + "EventCode": "0x0e", + "EventName": "UOPS_ISSUED.VECTOR_WIDTH_MISMATCH", "PEBScounters": "0,1,2,3,4,5,6,7", - "SampleAfterValue": "1000003", + "PublicDescription": "Counts the number of Blend Uops issued by the Resource Allocation Table (RAT) to the reservation station (RS) in order to preserve upper bits of vector registers. Starting with the Skylake microarchitecture, these Blend uops are needed since every Intel SSE instruction executed in Dirty Upper State needs to preserve bits 128-255 of the destination register. For more information, refer to Mixing Intel AVX and Intel SSE Code section of the Optimization Guide.", + "SampleAfterValue": "100003", "Speculative": "1", - "UMask": "0x14" + "UMask": "0x2" }, { - "BriefDescription": "Number of uops executed on port 7 and 8", + "BriefDescription": "Retirement slots used.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xa1", - "EventName": "UOPS_DISPATCHED.PORT_7_8", + "EventCode": "0xc2", + "EventName": "UOPS_RETIRED.SLOTS", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts, on the per-thread basis, cycles during which at least one uop is dispatched from the Reservation Station (RS) to ports 7 and 8.", + "PublicDescription": "Counts the retirement slots used each cycle.", "SampleAfterValue": "2000003", - "Speculative": "1", - "UMask": "0x80" + "UMask": "0x2" }, { - "BriefDescription": "number of branch instructions retired that were mispredicted and taken. Non PEBS", + "BriefDescription": "Cycles without actually retired uops.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xc5", - "EventName": "BR_MISP_RETIRED.COND_TAKEN", - "PEBS": "1", + "CounterMask": "1", + "EventCode": "0xc2", + "EventName": "UOPS_RETIRED.STALL_CYCLES", + "Invert": "1", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts taken conditional mispredicted branch instructions retired.", - "SampleAfterValue": "50021", - "UMask": "0x1" + "PublicDescription": "This event counts cycles without actually retired uops.", + "SampleAfterValue": "1000003", + "Speculative": "1", + "UMask": "0x2" }, { - "BriefDescription": "All mispredicted branch instructions retired.", + "BriefDescription": "Cycles with less than 10 actually retired uops.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3,4,5,6,7", - "EventCode": "0xc5", - "EventName": "BR_MISP_RETIRED.ALL_BRANCHES", - "PEBS": "1", + "CounterMask": "10", + "EventCode": "0xc2", + "EventName": "UOPS_RETIRED.TOTAL_CYCLES", + "Invert": "1", "PEBScounters": "0,1,2,3,4,5,6,7", - "PublicDescription": "Counts all the retired branch instructions that were mispredicted by the processor. A branch misprediction occurs when the processor incorrectly predicts the destination of the branch. When the misprediction is discovered at execution, all the instructions executed in the wrong (speculative) path must be discarded, and the processor must start fetching from the correct path.", - "SampleAfterValue": "50021" + "PublicDescription": "Counts the number of cycles using always true condition (uops_ret &lt; 16) applied to non PEBS uops retired event.", + "SampleAfterValue": "1000003", + "UMask": "0x2" } ]
\ No newline at end of file diff --git a/tools/perf/pmu-events/arch/x86/icelake/virtual-memory.json b/tools/perf/pmu-events/arch/x86/icelake/virtual-memory.json index f485f4664ea6..a006fd7f7b18 100644 --- a/tools/perf/pmu-events/arch/x86/icelake/virtual-memory.json +++ b/tools/perf/pmu-events/arch/x86/icelake/virtual-memory.json @@ -1,38 +1,14 @@ [ { - "BriefDescription": "DTLB flush attempts of the thread-specific entries", - "CollectPEBSRecord": "2", - "Counter": "0,1,2,3", - "EventCode": "0xBD", - "EventName": "TLB_FLUSH.DTLB_THREAD", - "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts the number of DTLB flush attempts of the thread-specific entries.", - "SampleAfterValue": "100007", - "Speculative": "1", - "UMask": "0x1" - }, - { - "BriefDescription": "Code miss in all TLB levels causes a page walk that completes. (All page sizes)", + "BriefDescription": "Loads that miss the DTLB and hit the STLB.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0x85", - "EventName": "ITLB_MISSES.WALK_COMPLETED", + "EventCode": "0x08", + "EventName": "DTLB_LOAD_MISSES.STLB_HIT", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts completed page walks (all page sizes) caused by a code fetch. This implies it missed in the ITLB (Instruction TLB) and further levels of TLB. The page walk can end with or without a fault.", + "PublicDescription": "Counts loads that miss the DTLB (Data TLB) and hit the STLB (Second level TLB).", "SampleAfterValue": "100003", "Speculative": "1", - "UMask": "0xe" - }, - { - "BriefDescription": "STLB flush attempts", - "CollectPEBSRecord": "2", - "Counter": "0,1,2,3", - "EventCode": "0xBD", - "EventName": "TLB_FLUSH.STLB_ANY", - "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts the number of any STLB flush attempts (such as entire, VPID, PCID, InvPage, CR3 write, etc.).", - "SampleAfterValue": "100007", - "Speculative": "1", "UMask": "0x20" }, { @@ -49,77 +25,77 @@ "UMask": "0x10" }, { - "BriefDescription": "Code miss in all TLB levels causes a page walk that completes. (4K)", + "BriefDescription": "Load miss in all TLB levels causes a page walk that completes. (All page sizes)", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0x85", - "EventName": "ITLB_MISSES.WALK_COMPLETED_4K", + "EventCode": "0x08", + "EventName": "DTLB_LOAD_MISSES.WALK_COMPLETED", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts completed page walks (4K page sizes) caused by a code fetch. This implies it missed in the ITLB (Instruction TLB) and further levels of TLB. The page walk can end with or without a fault.", + "PublicDescription": "Counts completed page walks (all page sizes) caused by demand data loads. This implies it missed in the DTLB and further levels of TLB. The page walk can end with or without a fault.", "SampleAfterValue": "100003", "Speculative": "1", - "UMask": "0x2" + "UMask": "0xe" }, { - "BriefDescription": "Page walks completed due to a demand data load to a 4K page.", + "BriefDescription": "Page walks completed due to a demand data load to a 2M/4M page.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0x08", - "EventName": "DTLB_LOAD_MISSES.WALK_COMPLETED_4K", + "EventName": "DTLB_LOAD_MISSES.WALK_COMPLETED_2M_4M", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts completed page walks (4K sizes) caused by demand data loads. This implies address translations missed in the DTLB and further levels of TLB. The page walk can end with or without a fault.", + "PublicDescription": "Counts completed page walks (2M/4M sizes) caused by demand data loads. This implies address translations missed in the DTLB and further levels of TLB. The page walk can end with or without a fault.", "SampleAfterValue": "100003", "Speculative": "1", - "UMask": "0x2" + "UMask": "0x4" }, { - "BriefDescription": "Code miss in all TLB levels causes a page walk that completes. (2M/4M)", + "BriefDescription": "Page walks completed due to a demand data load to a 4K page.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0x85", - "EventName": "ITLB_MISSES.WALK_COMPLETED_2M_4M", + "EventCode": "0x08", + "EventName": "DTLB_LOAD_MISSES.WALK_COMPLETED_4K", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts completed page walks (2M/4M page sizes) caused by a code fetch. This implies it missed in the ITLB (Instruction TLB) and further levels of TLB. The page walk can end with or without a fault.", + "PublicDescription": "Counts completed page walks (4K sizes) caused by demand data loads. This implies address translations missed in the DTLB and further levels of TLB. The page walk can end with or without a fault.", "SampleAfterValue": "100003", "Speculative": "1", - "UMask": "0x4" + "UMask": "0x2" }, { - "BriefDescription": "Cycles when at least one PMH is busy with a page walk for a store.", + "BriefDescription": "Number of page walks outstanding for a demand load in the PMH each cycle.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "CounterMask": "1", - "EventCode": "0x49", - "EventName": "DTLB_STORE_MISSES.WALK_ACTIVE", + "EventCode": "0x08", + "EventName": "DTLB_LOAD_MISSES.WALK_PENDING", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts cycles when at least one PMH (Page Miss Handler) is busy with a page walk for a store.", + "PublicDescription": "Counts the number of page walks outstanding for a demand load in the PMH (Page Miss Handler) each cycle.", "SampleAfterValue": "100003", "Speculative": "1", "UMask": "0x10" }, { - "BriefDescription": "Page walks completed due to a demand data store to a 2M/4M page.", + "BriefDescription": "Stores that miss the DTLB and hit the STLB.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0x49", - "EventName": "DTLB_STORE_MISSES.WALK_COMPLETED_2M_4M", + "EventName": "DTLB_STORE_MISSES.STLB_HIT", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts completed page walks (2M/4M sizes) caused by demand data stores. This implies address translations missed in the DTLB and further levels of TLB. The page walk can end with or without a fault.", + "PublicDescription": "Counts stores that miss the DTLB (Data TLB) and hit the STLB (2nd Level TLB).", "SampleAfterValue": "100003", "Speculative": "1", - "UMask": "0x4" + "UMask": "0x20" }, { - "BriefDescription": "Stores that miss the DTLB and hit the STLB.", + "BriefDescription": "Cycles when at least one PMH is busy with a page walk for a store.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", + "CounterMask": "1", "EventCode": "0x49", - "EventName": "DTLB_STORE_MISSES.STLB_HIT", + "EventName": "DTLB_STORE_MISSES.WALK_ACTIVE", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts stores that miss the DTLB (Data TLB) and hit the STLB (2nd Level TLB).", + "PublicDescription": "Counts cycles when at least one PMH (Page Miss Handler) is busy with a page walk for a store.", "SampleAfterValue": "100003", "Speculative": "1", - "UMask": "0x20" + "UMask": "0x10" }, { "BriefDescription": "Store misses in all TLB levels causes a page walk that completes. (All page sizes)", @@ -134,16 +110,16 @@ "UMask": "0xe" }, { - "BriefDescription": "Load miss in all TLB levels causes a page walk that completes. (All page sizes)", + "BriefDescription": "Page walks completed due to a demand data store to a 2M/4M page.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0x08", - "EventName": "DTLB_LOAD_MISSES.WALK_COMPLETED", + "EventCode": "0x49", + "EventName": "DTLB_STORE_MISSES.WALK_COMPLETED_2M_4M", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts completed page walks (all page sizes) caused by demand data loads. This implies it missed in the DTLB and further levels of TLB. The page walk can end with or without a fault.", + "PublicDescription": "Counts completed page walks (2M/4M sizes) caused by demand data stores. This implies address translations missed in the DTLB and further levels of TLB. The page walk can end with or without a fault.", "SampleAfterValue": "100003", "Speculative": "1", - "UMask": "0xe" + "UMask": "0x4" }, { "BriefDescription": "Page walks completed due to a demand data store to a 4K page.", @@ -158,6 +134,18 @@ "UMask": "0x2" }, { + "BriefDescription": "Number of page walks outstanding for a store in the PMH each cycle.", + "CollectPEBSRecord": "2", + "Counter": "0,1,2,3", + "EventCode": "0x49", + "EventName": "DTLB_STORE_MISSES.WALK_PENDING", + "PEBScounters": "0,1,2,3", + "PublicDescription": "Counts the number of page walks outstanding for a store in the PMH (Page Miss Handler) each cycle.", + "SampleAfterValue": "100003", + "Speculative": "1", + "UMask": "0x10" + }, + { "BriefDescription": "Instruction fetch requests that miss the ITLB and hit the STLB.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", @@ -170,76 +158,88 @@ "UMask": "0x20" }, { - "BriefDescription": "Page walks completed due to a demand data load to a 2M/4M page.", + "BriefDescription": "Cycles when at least one PMH is busy with a page walk for code (instruction fetch) request.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0x08", - "EventName": "DTLB_LOAD_MISSES.WALK_COMPLETED_2M_4M", + "CounterMask": "1", + "EventCode": "0x85", + "EventName": "ITLB_MISSES.WALK_ACTIVE", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts completed page walks (2M/4M sizes) caused by demand data loads. This implies address translations missed in the DTLB and further levels of TLB. The page walk can end with or without a fault.", + "PublicDescription": "Counts cycles when at least one PMH (Page Miss Handler) is busy with a page walk for a code (instruction fetch) request.", "SampleAfterValue": "100003", "Speculative": "1", - "UMask": "0x4" + "UMask": "0x10" }, { - "BriefDescription": "Number of page walks outstanding for an outstanding code request in the PMH each cycle.", + "BriefDescription": "Code miss in all TLB levels causes a page walk that completes. (All page sizes)", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", "EventCode": "0x85", - "EventName": "ITLB_MISSES.WALK_PENDING", + "EventName": "ITLB_MISSES.WALK_COMPLETED", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts the number of page walks outstanding for an outstanding code (instruction fetch) request in the PMH (Page Miss Handler) each cycle.", + "PublicDescription": "Counts completed page walks (all page sizes) caused by a code fetch. This implies it missed in the ITLB (Instruction TLB) and further levels of TLB. The page walk can end with or without a fault.", "SampleAfterValue": "100003", "Speculative": "1", - "UMask": "0x10" + "UMask": "0xe" }, { - "BriefDescription": "Cycles when at least one PMH is busy with a page walk for code (instruction fetch) request.", + "BriefDescription": "Code miss in all TLB levels causes a page walk that completes. (2M/4M)", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "CounterMask": "1", "EventCode": "0x85", - "EventName": "ITLB_MISSES.WALK_ACTIVE", + "EventName": "ITLB_MISSES.WALK_COMPLETED_2M_4M", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts cycles when at least one PMH (Page Miss Handler) is busy with a page walk for a code (instruction fetch) request.", + "PublicDescription": "Counts completed page walks (2M/4M page sizes) caused by a code fetch. This implies it missed in the ITLB (Instruction TLB) and further levels of TLB. The page walk can end with or without a fault.", "SampleAfterValue": "100003", "Speculative": "1", - "UMask": "0x10" + "UMask": "0x4" }, { - "BriefDescription": "Loads that miss the DTLB and hit the STLB.", + "BriefDescription": "Code miss in all TLB levels causes a page walk that completes. (4K)", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0x08", - "EventName": "DTLB_LOAD_MISSES.STLB_HIT", + "EventCode": "0x85", + "EventName": "ITLB_MISSES.WALK_COMPLETED_4K", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts loads that miss the DTLB (Data TLB) and hit the STLB (Second level TLB).", + "PublicDescription": "Counts completed page walks (4K page sizes) caused by a code fetch. This implies it missed in the ITLB (Instruction TLB) and further levels of TLB. The page walk can end with or without a fault.", "SampleAfterValue": "100003", "Speculative": "1", - "UMask": "0x20" + "UMask": "0x2" }, { - "BriefDescription": "Number of page walks outstanding for a demand load in the PMH each cycle.", + "BriefDescription": "Number of page walks outstanding for an outstanding code request in the PMH each cycle.", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0x08", - "EventName": "DTLB_LOAD_MISSES.WALK_PENDING", + "EventCode": "0x85", + "EventName": "ITLB_MISSES.WALK_PENDING", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts the number of page walks outstanding for a demand load in the PMH (Page Miss Handler) each cycle.", + "PublicDescription": "Counts the number of page walks outstanding for an outstanding code (instruction fetch) request in the PMH (Page Miss Handler) each cycle.", "SampleAfterValue": "100003", "Speculative": "1", "UMask": "0x10" }, { - "BriefDescription": "Number of page walks outstanding for a store in the PMH each cycle.", + "BriefDescription": "DTLB flush attempts of the thread-specific entries", "CollectPEBSRecord": "2", "Counter": "0,1,2,3", - "EventCode": "0x49", - "EventName": "DTLB_STORE_MISSES.WALK_PENDING", + "EventCode": "0xBD", + "EventName": "TLB_FLUSH.DTLB_THREAD", "PEBScounters": "0,1,2,3", - "PublicDescription": "Counts the number of page walks outstanding for a store in the PMH (Page Miss Handler) each cycle.", - "SampleAfterValue": "100003", + "PublicDescription": "Counts the number of DTLB flush attempts of the thread-specific entries.", + "SampleAfterValue": "100007", "Speculative": "1", - "UMask": "0x10" + "UMask": "0x1" + }, + { + "BriefDescription": "STLB flush attempts", + "CollectPEBSRecord": "2", + "Counter": "0,1,2,3", + "EventCode": "0xBD", + "EventName": "TLB_FLUSH.STLB_ANY", + "PEBScounters": "0,1,2,3", + "PublicDescription": "Counts the number of any STLB flush attempts (such as entire, VPID, PCID, InvPage, CR3 write, etc.).", + "SampleAfterValue": "100007", + "Speculative": "1", + "UMask": "0x20" } ]
\ No newline at end of file |