-
Notifications
You must be signed in to change notification settings - Fork 21.4k
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
[Quant] make x86 the default quantization backend (qengine) #91235
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/91235
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit 1bdd152: This comment was automatically generated by Dr. CI and updates every 15 minutes. |
f529bbe
to
6f1024e
Compare
6f1024e
to
755cad9
Compare
755cad9
to
1bdd152
Compare
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.
PR description does not match with the actual implementation - i.e. it makes x86 a default backend for all platforms (not just x86, but ARM as well)
If this quantization backend supposed to work on all CPUs, let's rename it to something else.
If not, please guard the check with #if defined(__x86_64__) || defined(WHATEVER_IT_SHOULD_BE_FOR_WINDOWS)
Thanks @malfet. X86 backend is only enabled when fbgemm is enabled. Looks like fbgemm only supports x86_64 platforms, right? In that case, x86 backend is enabled on x86_64 platforms only, and additional guard is not needed. |
Good point, looks good to me then. |
@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 |
Summary
Make x86 the default quantization backend (qengine) for X86 CPU platforms.
X86 is a unified quantization backend combining goodness of fbgemm and onednn. For more details please see #83888
Test plan
python test/test_quantization.py
cc @jerryzh168 @jianyuh @raghuramank100 @jamesr66a @vkuzo @jgong5 @leslie-fang-intel @mingfeima @XiaobingSuper @sanchitintel @ashokei @jingxu10