perf vendor events amd: Add Zen 5 events and metrics
v2 cover perf vendor events amd: Add Zen 5 events and metrics
Add events and metrics taken from the Performance Monitor Counters for AMD Family 1Ah Model 00h-0Fh Processors document available at
https://bugzilla.kernel.org/attachment.cgi?id=305974
Previous versions can be found at:
v1: https://lore.kernel.org/all/[email protected]/
Changes in v2:
- Categorize events based on their prefixes and rename the JSON files accordingly.
- Change metric scale units based on previous suggestions.
- Describe pipeline utilization metrics as percentages instead of fractions.
- Rename some pipeline utilization metrics ending with suffixes like "_bandwidth" or "_latency" to correctly denote that these are not actually reporting bandwidth or latency.
- Change branch misprediction ratio metric to misprediction rate.
- Remove some duplicate UMC metrics.
Sandipan Das (4):
perf vendor events amd: Add Zen 5 core events
perf vendor events amd: Add Zen 5 uncore events
perf vendor events amd: Add Zen 5 metrics
perf vendor events amd: Add Zen 5 mapping
.../arch/x86/amdzen5/branch-prediction.json | 93 ++ .../pmu-events/arch/x86/amdzen5/decode.json | 115 +++ .../arch/x86/amdzen5/execution.json | 174 ++++ .../arch/x86/amdzen5/floating-point.json | 812 ++++++++++++++++++ .../arch/x86/amdzen5/inst-cache.json | 72 ++ .../pmu-events/arch/x86/amdzen5/l2-cache.json | 266 ++++++ .../pmu-events/arch/x86/amdzen5/l3-cache.json | 177 ++++ .../arch/x86/amdzen5/load-store.json | 451 ++++++++++ .../arch/x86/amdzen5/memory-controller.json | 101 +++ .../pmu-events/arch/x86/amdzen5/pipeline.json | 99 +++ .../arch/x86/amdzen5/recommended.json | 345 ++++++++ tools/perf/pmu-events/arch/x86/mapfile.csv | 1 + 12 files changed, 2706 insertions(+) create mode 100644 tools/perf/pmu-events/arch/x86/amdzen5/branch-prediction.json create mode 100644 tools/perf/pmu-events/arch/x86/amdzen5/decode.json create mode 100644 tools/perf/pmu-events/arch/x86/amdzen5/execution.json create mode 100644 tools/perf/pmu-events/arch/x86/amdzen5/floating-point.json create mode 100644 tools/perf/pmu-events/arch/x86/amdzen5/inst-cache.json create mode 100644 tools/perf/pmu-events/arch/x86/amdzen5/l2-cache.json create mode 100644 tools/perf/pmu-events/arch/x86/amdzen5/l3-cache.json create mode 100644 tools/perf/pmu-events/arch/x86/amdzen5/load-store.json create mode 100644 tools/perf/pmu-events/arch/x86/amdzen5/memory-controller.json create mode 100644 tools/perf/pmu-events/arch/x86/amdzen5/pipeline.json create mode 100644 tools/perf/pmu-events/arch/x86/amdzen5/recommended.json
Signature · 1 line
2.34.1
PATCH v2 1/4 perf vendor events amd: Add Zen 5 core events
Add core events taken from Section 1.4 "Core Performance Monitor Counters" of the Performance Monitor Counters for AMD Family 1Ah Model 00h-0Fh Processors document available at the link below. This constitutes events which capture information on op dispatch, execution and retirement, branch prediction, L1 and L2 cache activity, TLB activity, etc.
--- .../arch/x86/amdzen5/branch-prediction.json | 93 ++ .../pmu-events/arch/x86/amdzen5/decode.json | 115 +++ .../arch/x86/amdzen5/execution.json | 174 ++++ .../arch/x86/amdzen5/floating-point.json | 812 ++++++++++++++++++ .../arch/x86/amdzen5/inst-cache.json | 72 ++ .../pmu-events/arch/x86/amdzen5/l2-cache.json | 266 ++++++ .../arch/x86/amdzen5/load-store.json | 451 ++++++++++ 7 files changed, 1983 insertions(+) create mode 100644 tools/perf/pmu-events/arch/x86/amdzen5/branch-prediction.json create mode 100644 tools/perf/pmu-events/arch/x86/amdzen5/decode.json create mode 100644 tools/perf/pmu-events/arch/x86/amdzen5/execution.json create mode 100644 tools/perf/pmu-events/arch/x86/amdzen5/floating-point.json create mode 100644 tools/perf/pmu-events/arch/x86/amdzen5/inst-cache.json create mode 100644 tools/perf/pmu-events/arch/x86/amdzen5/l2-cache.json create mode 100644 tools/perf/pmu-events/arch/x86/amdzen5/load-store.json
tools/perf/pmu-events/arch/x86/amdzen5/branch-prediction.json new file +93 -0
@@ -0,0 +1,93 @@ +[ + { + "EventName": "bp_l1_tlb_miss_l2_tlb_hit", + "EventCode": "0x84", + "BriefDescription": "Instruction fetches that miss in the L1 ITLB but hit in the L2 ITLB." + }, + { + "EventName": "bp_l1_tlb_miss_l2_tlb_miss.if4k", + "EventCode": "0x85", + "BriefDescription": "Instruction fetches that miss in both the L1 and L2 ITLBs (page-table walks are requested) for 4k pages.", + "UMask": "0x01" + }, + { + "EventName": "bp_l1_tlb_miss_l2_tlb_miss.if2m", + "EventCode": "0x85", + "BriefDescription": "Instruction fetches that miss in both the L1 and L2 ITLBs (page-table walks are requested) for 2M pages.", + "UMask": "0x02" + }, + { + "EventName": "bp_l1_tlb_miss_l2_tlb_miss.if1g", + "EventCode": "0x85", + "BriefDescription": "Instruction fetches that miss in both the L1 and L2 ITLBs (page-table walks are requested) for 1G pages.", + "UMask": "0x04" + }, + { + "EventName": "bp_l1_tlb_miss_l2_tlb_miss.coalesced_4k", + "EventCode": "0x85", + "BriefDescription": "Instruction fetches that miss in both the L1 and L2 ITLBs (page-table walks are requested) for coalesced pages. A coalesced page is a 16k page created from four adjacent 4k pages.", + "UMask": "0x08" + }, + { + "EventName": "bp_l1_tlb_miss_l2_tlb_miss.all", + "EventCode": "0x85", + "BriefDescription": "Instruction fetches that miss in both the L1 and L2 ITLBs (page-table walks are requested) for all page sizes.", + "UMask": "0x0f" + }, + { + "EventName": "bp_l2_btb_correct", + "EventCode": "0x8b", + "BriefDescription": "L2 branch prediction overrides existing prediction (speculative)." + }, + { + "EventName": "bp_dyn_ind_pred", + "EventCode": "0x8e", + "BriefDescription": "Dynamic indirect predictions (branch used the indirect predictor to make a prediction)." + }, + { + "EventName": "bp_de_redirect", + "EventCode": "0x91", + "BriefDescription": "Number of times an early redirect is sent to branch predictor. This happens when either the decoder or dispatch logic is able to detect that the branch predictor needs to be redirected." + }, + { + "EventName": "bp_l1_tlb_fetch_hit.if4k", + "EventCode": "0x94", + "BriefDescription": "Instruction fetches that hit in the L1 ITLB for 4k or coalesced pages. A coalesced page is a 16k page created from four adjacent 4k pages.", + "UMask": "0x01" + }, + { + "EventName": "bp_l1_tlb_fetch_hit.if2m", + "EventCode": "0x94", + "BriefDescription": "Instruction fetches that hit in the L1 ITLB for 2M pages.", + "UMask": "0x02" + }, + { + "EventName": "bp_l1_tlb_fetch_hit.if1g", + "EventCode": "0x94", + "BriefDescription": "Instruction fetches that hit in the L1 ITLB for 1G pages.", + "UMask": "0x04" + }, + { + "EventName": "bp_l1_tlb_fetch_hit.all", + "EventCode": "0x94", + "BriefDescription": "Instruction fetches that hit in the L1 ITLB for all page sizes.", + "UMask": "0x07" + }, + { + "EventName": "bp_redirects.resync", + "EventCode": "0x9f", + "BriefDescription": "Redirects of the branch predictor caused by resyncs.", + "UMask": "0x01" + }, + { + "EventName": "bp_redirects.ex_redir", + "EventCode": "0x9f", + "BriefDescription": "Redirects of the branch predictor caused by mispredicts.", + "UMask": "0x02" + }, + { + "EventName": "bp_redirects.all", + "EventCode": "0x9f", + "BriefDescription": "Redirects of the branch predictor." + } +]
tools/perf/pmu-events/arch/x86/amdzen5/decode.json new file +115 -0
115 lines not on this page · raw message
tools/perf/pmu-events/arch/x86/amdzen5/execution.json new file +174 -0
174 lines not on this page · raw message
tools/perf/pmu-events/arch/x86/amdzen5/floating-point.json new file +812 -0
812 lines not on this page · raw message
tools/perf/pmu-events/arch/x86/amdzen5/inst-cache.json new file +72 -0
72 lines not on this page · raw message
tools/perf/pmu-events/arch/x86/amdzen5/l2-cache.json new file +266 -0
266 lines not on this page · raw message
tools/perf/pmu-events/arch/x86/amdzen5/load-store.json new file +451 -0
451 lines not on this page · raw message
Signature · 1 line
2.34.1
PATCH v2 2/4 perf vendor events amd: Add Zen 5 uncore events
In reply to Sandipan DasAdd uncore events taken from Section 1.5 "L3 Cache Performance Monitor Counters" and Section 2 "UMC Performance Monitors" of the Performance Monitor Counters for AMD Family 1Ah Model 00h-0Fh Processors document available at the link below. This constitutes events which capture L3 cache and UMC command activity.
--- .../pmu-events/arch/x86/amdzen5/l3-cache.json | 177 ++++++++++++++++++ .../arch/x86/amdzen5/memory-controller.json | 101 ++++++++++ 2 files changed, 278 insertions(+) create mode 100644 tools/perf/pmu-events/arch/x86/amdzen5/l3-cache.json create mode 100644 tools/perf/pmu-events/arch/x86/amdzen5/memory-controller.json
tools/perf/pmu-events/arch/x86/amdzen5/l3-cache.json new file +177 -0
@@ -0,0 +1,177 @@ +[ + { + "EventName": "l3_lookup_state.l3_miss", + "EventCode": "0x04", + "BriefDescription": "L3 cache misses.", + "UMask": "0x01", + "Unit": "L3PMC" + }, + { + "EventName": "l3_lookup_state.l3_hit", + "EventCode": "0x04", + "BriefDescription": "L3 cache hits.", + "UMask": "0xfe", + "Unit": "L3PMC" + }, + { + "EventName": "l3_lookup_state.all_coherent_accesses_to_l3", + "EventCode": "0x04", + "BriefDescription": "L3 cache requests for all coherent accesses.", + "UMask": "0xff", + "Unit": "L3PMC" + }, + { + "EventName": "l3_xi_sampled_latency.dram_near", + "EventCode": "0xac", + "BriefDescription": "Average sampled latency when data is sourced from DRAM in the same NUMA node.", + "UMask": "0x01", + "EnAllCores": "0x1", + "EnAllSlices": "0x1", + "SliceId": "0x3", + "ThreadMask": "0x3", + "Unit": "L3PMC" + }, + { + "EventName": "l3_xi_sampled_latency.dram_far", + "EventCode": "0xac", + "BriefDescription": "Average sampled latency when data is sourced from DRAM in a different NUMA node.", + "UMask": "0x02", + "EnAllCores": "0x1", + "EnAllSlices": "0x1", + "SliceId": "0x3", + "ThreadMask": "0x3", + "Unit": "L3PMC" + }, + { + "EventName": "l3_xi_sampled_latency.near_cache", + "EventCode": "0xac", + "BriefDescription": "Average sampled latency when data is sourced from another CCX's cache when the address was in the same NUMA node.", + "UMask": "0x04", + "EnAllCores": "0x1", + "EnAllSlices": "0x1", + "SliceId": "0x3", + "ThreadMask": "0x3", + "Unit": "L3PMC" + }, + { + "EventName": "l3_xi_sampled_latency.far_cache", + "EventCode": "0xac", + "BriefDescription": "Average sampled latency when data is sourced from another CCX's cache when the address was in a different NUMA node.", + "UMask": "0x08", + "EnAllCores": "0x1", + "EnAllSlices": "0x1", + "SliceId": "0x3", + "ThreadMask": "0x3", + "Unit": "L3PMC" + }, + { + "EventName": "l3_xi_sampled_latency.ext_near", + "EventCode": "0xac", + "BriefDescription": "Average sampled latency when data is sourced from extension memory (CXL) in the same NUMA node.", + "UMask": "0x10", + "EnAllCores": "0x1", + "EnAllSlices": "0x1", + "SliceId": "0x3", + "ThreadMask": "0x3", + "Unit": "L3PMC" + }, + { + "EventName": "l3_xi_sampled_latency.ext_far", + "EventCode": "0xac", + "BriefDescription": "Average sampled latency when data is sourced from extension memory (CXL) in a different NUMA node.", + "UMask": "0x20", + "EnAllCores": "0x1", + "EnAllSlices": "0x1", + "SliceId": "0x3", + "ThreadMask": "0x3", + "Unit": "L3PMC" + }, + { + "EventName": "l3_xi_sampled_latency.all", + "EventCode": "0xac", + "BriefDescription": "Average sampled latency from all data sources.", + "UMask": "0x3f", + "EnAllCores": "0x1", + "EnAllSlices": "0x1", + "SliceId": "0x3", + "ThreadMask": "0x3", + "Unit": "L3PMC" + }, + { + "EventName": "l3_xi_sampled_latency_requests.dram_near", + "EventCode": "0xad", + "BriefDescription": "L3 cache fill requests sourced from DRAM in the same NUMA node.", + "UMask": "0x01", + "EnAllCores": "0x1", + "EnAllSlices": "0x1", + "SliceId": "0x3", + "ThreadMask": "0x3", + "Unit": "L3PMC" + }, + { + "EventName": "l3_xi_sampled_latency_requests.dram_far", + "EventCode": "0xad", + "BriefDescription": "L3 cache fill requests sourced from DRAM in a different NUMA node.", + "UMask": "0x02", + "EnAllCores": "0x1", + "EnAllSlices": "0x1", + "SliceId": "0x3", + "ThreadMask": "0x3", + "Unit": "L3PMC" + }, + { + "EventName": "l3_xi_sampled_latency_requests.near_cache", + "EventCode": "0xad", + "BriefDescription": "L3 cache fill requests sourced from another CCX's cache when the address was in the same NUMA node.", + "UMask": "0x04", + "EnAllCores": "0x1", + "EnAllSlices": "0x1", + "SliceId": "0x3", + "ThreadMask": "0x3", + "Unit": "L3PMC" + }, + { + "EventName": "l3_xi_sampled_latency_requests.far_cache", + "EventCode": "0xad", + "BriefDescription": "L3 cache fill requests sourced from another CCX's cache when the address was in a different NUMA node.", + "UMask": "0x08", + "EnAllCores": "0x1", + "EnAllSlices": "0x1", + "SliceId": "0x3", + "ThreadMask": "0x3", + "Unit": "L3PMC" + }, + { + "EventName": "l3_xi_sampled_latency_requests.ext_near", + "EventCode": "0xad", + "BriefDescription": "L3 cache fill requests sourced from extension memory (CXL) in the same NUMA node.", + "UMask": "0x10", + "EnAllCores": "0x1", + "EnAllSlices": "0x1", + "SliceId": "0x3", + "ThreadMask": "0x3", + "Unit": "L3PMC" + }, + { + "EventName": "l3_xi_sampled_latency_requests.ext_far", + "EventCode": "0xad", + "BriefDescription": "L3 cache fill requests sourced from extension memory (CXL) in a different NUMA node.", + "UMask": "0x20", + "EnAllCores": "0x1", + "EnAllSlices": "0x1", + "SliceId": "0x3", + "ThreadMask": "0x3", + "Unit": "L3PMC" + }, + { + "EventName": "l3_xi_sampled_latency_requests.all", + "EventCode": "0xad", + "BriefDescription": "L3 cache fill requests sourced from all data sources.", + "UMask": "0x3f", + "EnAllCores": "0x1", + "EnAllSlices": "0x1", + "SliceId": "0x3", + "ThreadMask": "0x3", + "Unit": "L3PMC" + } +]
tools/perf/pmu-events/arch/x86/amdzen5/memory-controller.json new file +101 -0
101 lines not on this page · raw message
Signature · 1 line
2.34.1
PATCH v2 3/4 perf vendor events amd: Add Zen 5 metrics
In reply to Sandipan DasAdd metrics taken from Section 1.2 "Performance Measurement" of the Performance Monitor Counters for AMD Family 1Ah Model 00h-0Fh Processors document available at the link below.
The recommended metrics are sourced from Table 1 "Guidance for Common Performance Statistics with Complex Event Selects".
The pipeline utilization metrics are sourced from Table 2 "Guidance for Pipeline Utilization Analysis Statistics". These are useful for finding performance bottlenecks by analyzing activity at different stages of the pipeline. There are metric groups available for Level 1 and Level 2 analysis.
--- .../pmu-events/arch/x86/amdzen5/pipeline.json | 99 +++++ .../arch/x86/amdzen5/recommended.json | 345 ++++++++++++++++++ 2 files changed, 444 insertions(+) create mode 100644 tools/perf/pmu-events/arch/x86/amdzen5/pipeline.json create mode 100644 tools/perf/pmu-events/arch/x86/amdzen5/recommended.json
tools/perf/pmu-events/arch/x86/amdzen5/pipeline.json new file +99 -0
@@ -0,0 +1,99 @@ +[ + { + "MetricName": "total_dispatch_slots", + "BriefDescription": "Total dispatch slots (up to 8 instructions can be dispatched in each cycle).", + "MetricExpr": "8 * ls_not_halted_cyc", + "ScaleUnit": "1slots" + }, + { + "MetricName": "frontend_bound", + "BriefDescription": "Percentage of dispatch slots that remained unused because the frontend did not supply enough instructions/ops.", + "MetricExpr": "d_ratio(de_no_dispatch_per_slot.no_ops_from_frontend, total_dispatch_slots)", + "MetricGroup": "PipelineL1", + "ScaleUnit": "100%slots" + }, + { + "MetricName": "bad_speculation", + "BriefDescription": "Percentage of dispatched ops that did not retire.", + "MetricExpr": "d_ratio(de_src_op_disp.all - ex_ret_ops, total_dispatch_slots)", + "MetricGroup": "PipelineL1", + "ScaleUnit": "100%ops" + }, + { + "MetricName": "backend_bound", + "BriefDescription": "Percentage of dispatch slots that remained unused because of backend stalls.", + "MetricExpr": "d_ratio(de_no_dispatch_per_slot.backend_stalls, total_dispatch_slots)", + "MetricGroup": "PipelineL1", + "ScaleUnit": "100%slots" + }, + { + "MetricName": "smt_contention", + "BriefDescription": "Percentage of dispatch slots that remained unused because the other thread was selected.", + "MetricExpr": "d_ratio(de_no_dispatch_per_slot.smt_contention, total_dispatch_slots)", + "MetricGroup": "PipelineL1", + "ScaleUnit": "100%slots" + }, + { + "MetricName": "retiring", + "BriefDescription": "Percentage of dispatch slots used by ops that retired.", + "MetricExpr": "d_ratio(ex_ret_ops, total_dispatch_slots)", + "MetricGroup": "PipelineL1", + "ScaleUnit": "100%slots" + }, + { + "MetricName": "frontend_bound_by_latency", + "BriefDescription": "Percentage of dispatch slots that remained unused because of a latency bottleneck in the frontend (such as instruction cache or TLB misses).", + "MetricExpr": "d_ratio((8 * cpu@de_no_dispatch_per_slot.no_ops_from_frontend\\,cmask\\=0x8@), total_dispatch_slots)", + "MetricGroup": "PipelineL2;frontend_bound_group", + "ScaleUnit": "100%slots" + }, + { + "MetricName": "frontend_bound_by_bandwidth", + "BriefDescription": "Percentage of dispatch slots that remained unused because of a bandwidth bottleneck in the frontend (such as decode or op cache fetch bandwidth).", + "MetricExpr": "d_ratio(de_no_dispatch_per_slot.no_ops_from_frontend - (8 * cpu@de_no_dispatch_per_slot.no_ops_from_frontend\\,cmask\\=0x8@), total_dispatch_slots)", + "MetricGroup": "PipelineL2;frontend_bound_group", + "ScaleUnit": "100%slots" + }, + { + "MetricName": "bad_speculation_from_mispredicts", + "BriefDescription": "Percentage of dispatched ops that were flushed due to branch mispredicts.", + "MetricExpr": "d_ratio(bad_speculation * ex_ret_brn_misp, ex_ret_brn_misp + bp_redirects.resync)", + "MetricGroup": "PipelineL2;bad_speculation_group", + "ScaleUnit": "100%ops" + }, + { + "MetricName": "bad_speculation_from_pipeline_restarts", + "BriefDescription": "Percentage of dispatched ops that were flushed due to pipeline restarts (resyncs).", + "MetricExpr": "d_ratio(bad_speculation * bp_redirects.resync, ex_ret_brn_misp + bp_redirects.resync)", + "MetricGroup": "PipelineL2;bad_speculation_group", + "ScaleUnit": "100%ops" + }, + { + "MetricName": "backend_bound_by_memory", + "BriefDescription": "Percentage of dispatch slots that remained unused because of stalls due to the memory subsystem.", + "MetricExpr": "backend_bound * d_ratio(ex_no_retire.load_not_complete, ex_no_retire.not_complete)", + "MetricGroup": "PipelineL2;backend_bound_group", + "ScaleUnit": "100%slots" + }, + { + "MetricName": "backend_bound_by_cpu", + "BriefDescription": "Percentage of dispatch slots that remained unused because of stalls not related to the memory subsystem.", + "MetricExpr": "backend_bound * (1 - d_ratio(ex_no_retire.load_not_complete, ex_no_retire.not_complete))", + "MetricGroup": "PipelineL2;backend_bound_group", + "ScaleUnit": "100%slots" + }, + { + "MetricName": "retiring_from_fastpath", + "BriefDescription": "Percentage of dispatch slots used by fastpath ops that retired.", + "MetricExpr": "retiring * (1 - d_ratio(ex_ret_ucode_ops, ex_ret_ops))", + "MetricGroup": "PipelineL2;retiring_group", + "ScaleUnit": "100%slots" + }, + { + "MetricName": "retiring_from_microcode", + "BriefDescription": "Percentage of dispatch slots used by microcode ops that retired.", + "MetricExpr": "retiring * d_ratio(ex_ret_ucode_ops, ex_ret_ops)", + "MetricGroup": "PipelineL2;retiring_group", + "ScaleUnit": "100%slots" + } +]
tools/perf/pmu-events/arch/x86/amdzen5/recommended.json new file +345 -0
345 lines not on this page · raw message
Signature · 1 line
2.34.1
PATCH v2 4/4 perf vendor events amd: Add Zen 5 mapping
In reply to Sandipan DasAdd a regular expression in the map file so that appropriate JSON event files are used for AMD Zen 5 processors belonging to Family 1Ah.
--- tools/perf/pmu-events/arch/x86/mapfile.csv | 1 + 1 file changed, 1 insertion(+)
tools/perf/pmu-events/arch/x86/mapfile.csv +1 -0
@@ -40,3 +40,4 @@ AuthenticAMD-23-([12][0-9A-F]|[0-9A-F]),v2,amdzen1,core AuthenticAMD-23-[[:xdigit:]]+,v1,amdzen2,core AuthenticAMD-25-([245][[:xdigit:]]|[[:xdigit:]]),v1,amdzen3,core AuthenticAMD-25-[[:xdigit:]]+,v1,amdzen4,core +AuthenticAMD-26-[[:xdigit:]]+,v1,amdzen5,core
Signature · 1 line
2.34.1
Quoted 6 lines · Ian Rogers
On Fri, May 3, 2024 at 12:16 AM Sandipan Das <[1][email protected]> wrote:
Add events and metrics taken from the Performance Monitor Counters for
AMD Family 1Ah Model 00h-0Fh Processors document available at [2]https://bugzilla.kernel.org/attachment.cgi?id=305974
<SNIP>
Quoted 8 lines
Sandipan Das (4):
perf vendor events amd: Add Zen 5 core events perf vendor events amd: Add Zen 5 uncore events perf vendor events amd: Add Zen 5 metrics perf vendor events amd: Add Zen 5 mappingSeries:
Reviewed-by Ian Rogers <[4][email protected]>
Thanks, applied to perf-tools-next,
- Arnaldo