TMA 4.8 Release #181
Merged
TMA 4.8 Release #181
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Contributor
calebbiggers
commented
May 10, 2024
calebbiggers
commented
- Updated TMA metrics to 4.8
- Added client platform metrics
- Updated TMA metrics to 4.8 - Added client platform metrics
captain5050
reviewed
May 10, 2024
| "MetricName": "tma_frontend_bound", | ||
| "ScaleUnit": "100%" | ||
| "ScaleUnit": "100%", | ||
| "Threshold": "tma_frontend_bound > 15" |
There was a problem hiding this comment.
Here is the existing perf version:
{
"BriefDescription": "This category represents fraction of slots where the processor's Fronte
nd undersupplies its Backend",
"MetricExpr": "IDQ_UOPS_NOT_DELIVERED.CORE / tma_info_thread_slots",
"MetricGroup": "PGO;TmaL1;TopdownL1;tma_L1_group",
"MetricName": "tma_frontend_bound",
"MetricThreshold": "tma_frontend_bound > 0.15",
"MetricgroupNoGroup": "TopdownL1",
"PublicDescription": "This category represents fraction of slots where the processor's Frontend undersupplies its Backend. Frontend denotes the first part of the processor core responsible to fetch operations that are executed later on by the Backend part. Within the Frontend; a branch predictor predicts the next address to fetch; cache-lines are fetched from the memory subsystem; parsed into instructions; and lastly decoded into micro-operations (uops). Ideally the Frontend can issue Pipeline_Width uops every cycle to the Backend. Frontend Bound denotes unutilized issue-slots when there is no Backend stall; i.e. bubbles where Frontend delivered no uops while Backend could have accepted them. For example; stalls due to instruction-cache misses would be categorized under Frontend Bound. Sample with: FRONTEND_RETIRED.LATENCY_GE_4_PS",
"ScaleUnit": "100%"
},
- It looks like Threshold should be MetricThreshold.
slotsrather thantma_info_slots- In the description
Sample with: FRONTEND_RETIRED.LATENCY_GE_4_PSisn't present.
There was a problem hiding this comment.
I've renamed the "Threshold" field to "MetricThreshold"
captain5050
reviewed
May 10, 2024
| "MetricExpr": "( ( BR_MISP_RETIRED.ALL_BRANCHES / ( BR_MISP_RETIRED.ALL_BRANCHES + MACHINE_CLEARS.COUNT ) ) * ( ( UOPS_ISSUED.ANY - ( UOPS_RETIRED.RETIRE_SLOTS ) + ( 4 ) * ( ( INT_MISC.RECOVERY_CYCLES_ANY / 2 ) if #SMT_on else INT_MISC.RECOVERY_CYCLES ) ) / ( ( 4 ) * ( ( CPU_CLK_UNHALTED.THREAD_ANY / 2 ) if #SMT_on else ( CPU_CLK_UNHALTED.THREAD ) ) ) ) )", | ||
| "MetricGroup": "BadSpec;BrMispredicts;TmaL2;TopdownL2;tma_L2_group;tma_bad_speculation_group", | ||
| "MetricExpr": "( BR_MISP_RETIRED.ALL_BRANCHES / ( BR_MISP_RETIRED.ALL_BRANCHES + MACHINE_CLEARS.COUNT ) ) * tma_bad_speculation", | ||
| "MetricGroup": "BadSpec;BrMispredicts;BvMP;TmaL2;TopdownL2;tma_L2_group;tma_bad_speculation_group;Slots", |
There was a problem hiding this comment.
In the existing generated from a spreadsheet perf version the MetricGroup here is:
"MetricGroup": "BadSpec;BrMispredicts;TmaL2;TopdownL2;tma_L2_group;tma_bad_speculation_group;tma_issueBM"
tma_issueBM in particular is missing. The issue groups come from the threshold column.
There was a problem hiding this comment.
I've added the issues to the "MetricGroup" field
- Renamed "Threshold" field to "MetricThreshold" - Added issues to "MetricGroup" field - Edited incorrect event names - Updated EMR metrics
weilinwa
reviewed
May 13, 2024
| "MetricName": "tma_dtlb_load", | ||
| "ScaleUnit": "100%", | ||
| "Threshold": "tma_dtlb_load > 10 && tma_l1_bound > 10 && tma_memory_bound > 20 && tma_backend_bound > 20" | ||
| "MetricThreshold": "tma_dtlb_load > 10 && tma_l1_bound > 10 && tma_memory_bound > 20 && tma_backend_bound > 20" |
There was a problem hiding this comment.
@calebbiggers, current perf JSON uses & instead of &&.
captain5050
reviewed
May 15, 2024
| "Threshold": "tma_frontend_bound > 15" | ||
| }, | ||
| { | ||
| "BriefDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend latency issues. For example; instruction-cache misses; iTLB misses or fetch stalls after a branch misprediction are categorized under Frontend Latency. In such cases; the Frontend eventually delivers no uops for some period.", |
There was a problem hiding this comment.
The existing definition has "Sample with":
"PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend latency issues. For example; instruction-cache misses; iTLB misses or fetch stalls after a branch misprediction are categorized under Frontend Latency. In such cases; the Frontend eventually delivers no uops for some period. Sample with: RS_EVENTS.EMPTY_END",
Which is pulled from the "Locate-with" column here:
https://github.com/intel/perfmon/blob/main/scripts/create_perf_json.py#L836
Could we add this to avoid a regression?
| "Threshold": "tma_dsb_switches > 5 && tma_fetch_latency > 10 && tma_frontend_bound > 15" | ||
| }, | ||
| { | ||
| "BriefDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend bandwidth issues. For example; inefficiencies at the instruction decoders; or restrictions for caching in the DSB (decoded uops cache) are categorized under Fetch Bandwidth. In such cases; the Frontend typically delivers suboptimal amount of uops to the Backend.", |
There was a problem hiding this comment.
This existing description has "Related metrics":
"PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend bandwidth issues. For example; inefficiencies at the instruction decoders; or restrictions for caching in the DSB (decoded uops cache) are categorized under Fetch Bandwidth. In such cases; the Frontend typically delivers suboptimal amount of uops to the Backend. Related metrics: tma_dsb_switches, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb, tma_lcp"
That has come from the issues in the threshold column:
https://github.com/intel/perfmon/blob/main/scripts/create_perf_json.py#L1321
Could we add this to avoid a regression?
1perrytaylor
approved these changes
May 15, 2024
kshiprab
approved these changes
May 21, 2024
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.