Display kernel stats and framework op stats in GB-25 CI#2316
Open
gbaraldi wants to merge 1 commit intoEnzymeAD:mainfrom
Open
Display kernel stats and framework op stats in GB-25 CI#2316gbaraldi wants to merge 1 commit intoEnzymeAD:mainfrom
gbaraldi wants to merge 1 commit intoEnzymeAD:mainfrom
Conversation
gbaraldi
force-pushed
the
gb25-pm-counters
branch
from
5293c00 to
23af8c8
Compare
Enhance the "Display profile results" step to show per-kernel hardware counter data from the profiled loop2 run: - Kernel stats: register count and occupancy per kernel - Framework op stats: memory bandwidth, operational intensity, and bottleneck classification per op These APIs are available on Reactant main (merged in #2703/#2705). When PM counters are collected (via GB-25 add-pm-counters branch), the framework op stats will include measured_memory_bw and operational_intensity from CUPTI PM sampling. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
gbaraldi
force-pushed
the
gb25-pm-counters
branch
from
23af8c8 to
b25d5fd
Compare
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
Enhance the GB-25 CI "Display profile results" step to show per-kernel profiling data from the xplane trace.
Changes
mst3reactant_commit matrix entry (branch no longer exists)These Reactant APIs (
get_kernel_stats,get_framework_op_stats) are already on main (merged in EnzymeAD/Reactant.jl#2703 and #2705). When PM counters are enabled in the GB-25with_profilercall (PRONTOLab/GB-25#267), the framework op stats will include measured memory bandwidth and operational intensity from CUPTI PM sampling.Example output (from local H100 run)
🤖 Generated with Claude Code