fix(health): scope TransferBench CPU endpoints on sparse NUMA [AIMVT-314] - #382
Open
speriaswamy-amd wants to merge 2 commits into
Open
fix(health): scope TransferBench CPU endpoints on sparse NUMA [AIMVT-314]#382speriaswamy-amd wants to merge 2 commits into
speriaswamy-amd wants to merge 2 commits into
Conversation
…314] Venice reports 38 NUMA nodes while only two have CPUs, so p2p/scaling abort before GPU work and parse_tb_p2p_bw raises AttributeError. Inject NUM_CPU_DEVICES inside the sudo wrapper and fail closed when averages are missing. Co-authored-by: Cursor <cursoragent@cursor.com>
cijohnson
reviewed
Sep 2, 2026
Replace grep -l | wc -l with an explicit cpulist walk. Keep Mems_allowed_list when the count is 0 (empty glob / no populated nodes). Co-authored-by: Cursor <cursoragent@cursor.com>
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
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
p2p/scalingenumerated all of them, aborted at CPU 2 (cannot allocate memory due to process memory policy/cpuset), andparse_tb_p2p_bw()crashed withAttributeErroron missing UniDir averages.env_varsnever reached thesudo bash -cwrapper, and health JSON had noNUM_CPU_DEVICESknob. This does not switch topodp2p(that would drop CPU↔GPU coverage).sudo bash -c. Auto-detect populated CPU NUMA count (auto_num_cpu_devices, default true) or settransferbench.num_cpu_devices/transferbench.env. Harden the p2p parser tofail_test()with a NUMA hint instead of crashing; match the cpuset abort inscan_test_results().Test plan
test_transferbench_health.py(env prefix, NUMA count parsing, abort fixture vs scoped success)NUM_CPU_DEVICES=2;test_transfer_bench_p2pPASSED in 65s (4 GPUs, UniDir GPU→GPU avg ~190 GB/s). No NUMA abort, no parser crash.Made with Cursor