-
Notifications
You must be signed in to change notification settings - Fork 25.6k
multiply int8/uint8 for AVX512 #116346
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
multiply int8/uint8 for AVX512 #116346
Conversation
Summary: multiply int8/uint8 for AVX512 Test Plan: sandcastle, github Differential Revision: D52393918
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/116346
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit e035c91 with merge base 4f4b931 ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
This pull request was exported from Phabricator. Differential Revision: D52393918 |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM
@pytorchbot merge |
Merge startedYour change will be merged once all checks pass (ETA 0-4 Hours). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
Sorry for jumping in late, but why no unit tests/benchmarks for this one? (Considering that AVX512 can sometimes be slower than AVX2) |
This is for users that already have picked AVX512 - if they don't, they end up in AVX2 land (upgrade as separate PR n#116235. I started writing unit tests, but found that we already have pre-existing unit tests for testing the various operations in vec.h and it's derived children. @malfet , please also check out additional discussion about Aten testing in fbcode on the internal diff. |
Summary: multiply int8/uint8 for AVX512
Test Plan: sandcastle, github
Differential Revision: D52393918
cc @jgong5 @mingfeima @XiaobingSuper @sanchitintel @ashokei @jingxu10