Update CPU (x86, aarch64) ops microbenchmark#172159
Update CPU (x86, aarch64) ops microbenchmark#172159huydhn wants to merge 12 commits intopytorch:mainfrom
Conversation
Signed-off-by: Huy Do <[email protected]>
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/172159
Note: Links to docs will display an error until the docs builds have been completed. ❗ 1 Active SEVsThere are 1 currently active SEVs. If your PR is affected, please view them below: ❌ 2 New FailuresAs of commit 64b2c82 with merge base 8e5f493 ( NEW FAILURES - The following jobs have failed:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
Signed-off-by: Huy Do <[email protected]>
Signed-off-by: Huy Do <[email protected]>
Signed-off-by: Huy Do <[email protected]>
Signed-off-by: Huy Do <[email protected]>
Signed-off-by: Huy Do <[email protected]>
Signed-off-by: Huy Do <[email protected]>
Signed-off-by: Huy Do <[email protected]>
Signed-off-by: Huy Do <[email protected]>
Signed-off-by: Huy Do <[email protected]>
Signed-off-by: Huy Do <[email protected]>
Signed-off-by: Huy Do <[email protected]>
|
@claude, can you review this PR? |
|
Claude encountered an error —— View job I'll analyze this and get back to you. |
|
Looks like this PR hasn't been updated in a while so we're going to go ahead and mark this as |
Fix the 2 issues on https://github.com/pytorch/pytorch/actions/runs/20689246569/job/59395211465
FixFixed by [BE] Remove libstdcxx conda-forge hack from Docker builds #172260GLIBCXX_3.4.32on aarch64. This library from the conda env is busted somehow, but the systemlibstdc++is goodUSE_FBGEMMon aarch64 after [aarch64][caffe2] Fix FBGEMM detection on aarch64 #169379, andqengineis wrongly set tox86there.check_perf_csvscript where it fails to handle duplicated entries (same op)I also try to refresh the baseline CSVs, I guess I will just take the output CSV as baseline this time. There are too many entries otherwise
cc @albanD