add int8 packed gemm support on CPU device

[mingfeima]

Stack from ghstack (oldest at bottom):

• -> add int8 packed gemm support on CPU device #118056

cc @jgong5 @XiaobingSuper @sanchitintel @ashokei @jingxu10 @voznesenskym @penguinwu @EikanWang @Guobing-Chen @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @chenyang78 @kadeng @chauhang @aakhundov

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/118056

• 📄 Preview Python docs built from this PR
• 📄 Preview C++ docs built from this PR
• ❓ Need help or want to give feedback on the CI? Visit the bot commands wiki or our office hours

Note: Links to docs will display an error until the docs builds have been completed.

✅ No Failures

As of commit bb3b044 with merge base b9087f8 ():
💚 Looks good so far! There are no failures yet. 💚

This comment was automatically generated by Dr. CI and updates every 15 minutes.

[pytorchmergebot]

Merge started

Your 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

Advanced Debugging

Check the merge workflow status
here

[izaitsevfb]

@pytorchbot revert -m "breaks internal builds" -c ghfirst

Hi @mingfeima, I have to revert this change temporarily, as it breaks a lot of meta-internal builds.

ld.lld: error: undefined symbol: at::native::DispatchStub<void (*)(at::Tensor const&, at::Tensor const&, at::Tensor const&, at::Tensor const&), at::native::int8pack_mm_stub>::DEFAULT
>>> referenced by new:0 (third-party/llvm-project/17.x-android/llvm-project/libcxx/include/new:0)
>>>               LinearAlgebra.cpp.pic.o:(at::native::_weight_int8pack_mm_cpu(at::Tensor const&, at::Tensor const&, at::Tensor const&)) in archive buck-out/v2/gen/fbsource/79732733b3759079/xplat/assistant/model_benchmark_tool/cpp_binary/__opsAndroid__/libopsAndroid.pic.a
clang: error: linker command failed with exit code 1 (use -v to see invocation)

ld.lld: error: undefined symbol: at::native::DispatchStub<void (*)(at::Tensor const&, at::Tensor const&, at::Tensor const&, at::Tensor const&), at::native::int8pack_mm_stub>::DEFAULT
>>> referenced by DispatchStub.h:137 (fbsource/ATen/native/DispatchStub.h:137)
>>>               __objects__/aten/src/ATen/native/LinearAlgebra.cpp.o:(at::native::_weight_int8pack_mm_cpu(at::Tensor const&, at::Tensor const&, at::Tensor const&)) in archive buck-out/v2/gen/fbsource/d0dc59ca80ab55db/xplat/caffe2/__pt_ops_full_et__/libpt_ops_full_et.a
clang-15: error: linker command failed with exit code 1 (use -v to see invocation)

cc @mikekgfb

[pytorchmergebot]

@pytorchbot successfully started a revert job. Check the current status here.
Questions? Feedback? Please reach out to the PyTorch DevX Team

[pytorchmergebot]

@mingfeima your PR has been successfully reverted.

[pytorch-bot]

Warning: Unknown label ciflow/android.
Currently recognized labels are

• ciflow/binaries
• ciflow/binaries_conda
• ciflow/binaries_libtorch
• ciflow/binaries_wheel
• ciflow/inductor
• ciflow/inductor-perf-compare
• ciflow/mps
• ciflow/nightly
• ciflow/periodic
• ciflow/rocm
• ciflow/slow
• ciflow/trunk
• ciflow/unstable
• ciflow/xpu

Please add the new label to .github/pytorch-probot.yml

[mingfeima]

@izaitsevfb I forgot to update build_variables.bzl when adding a new file in aten/src/ATen/native/cpu, the internal build failure should be triggered by this. Just fix it.

[mingfeima]

@mikekgfb should be safe to go this time :)

[mingfeima]

@pytorchbot merge

[pytorchmergebot]

Merge started

Your 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

Advanced Debugging

Check the merge workflow status
here