-
Notifications
You must be signed in to change notification settings - Fork 434
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
Make AVX Support Optional (#1698) #1795
Conversation
✅ Deploy Preview for pytorch-fbgemm-docs canceled.
|
This pull request was exported from Phabricator. Differential Revision: D45139063 |
Summary: Pull Request resolved: pytorch#1795 - Add `FindAVX.cmake` from the PyTorch project to detect AVX compilation flags and conditionally add them to FBGEMM_GPU targets Pull Request resolved: pytorch#1698 Test Plan: On a devarm machine: - Set up the environment according to [the building instruction](https://github.com/pytorch/FBGEMM/blob/main/fbgemm_gpu/docs/BuildInstructions.md). For convenience we can source `.github/scripts/setup_env.bash` in the FBGEMM repo and run the functions in it. - Build and install CPU-only variant of fbgemm_gpu: # Assuming the CWD is the FBGEMM repo directory and setup_env.bash is sourced cd fbgemm_gpu build_fbgemm_gpu_install <conda-env-name> cpu Example log of a a successful building and installing process is like this: P717846260 - Run tests: `run_fbgemm_gpu_tests mycondaenv cpu` under `fbgemm_gpu/tests` directory Differential Revision: D45139063 Pulled By: excelle08 fbshipit-source-id: 35a1b605ee8503c5eb85783ca4b55030d296abc8
2ae3789
to
48f6b01
Compare
This pull request was exported from Phabricator. Differential Revision: D45139063 |
Summary: Pull Request resolved: pytorch#1795 - Add `FindAVX.cmake` from the PyTorch project to detect AVX compilation flags and conditionally add them to FBGEMM_GPU targets Pull Request resolved: pytorch#1698 Test Plan: On a devarm machine: - Set up the environment according to [the building instruction](https://github.com/pytorch/FBGEMM/blob/main/fbgemm_gpu/docs/BuildInstructions.md). For convenience we can source `.github/scripts/setup_env.bash` in the FBGEMM repo and run the functions in it. - Build and install CPU-only variant of fbgemm_gpu: # Assuming the CWD is the FBGEMM repo directory and setup_env.bash is sourced cd fbgemm_gpu build_fbgemm_gpu_install <conda-env-name> cpu Example log of a a successful building and installing process is like this: P717846260 - Run tests: `run_fbgemm_gpu_tests mycondaenv cpu` under `fbgemm_gpu/tests` directory Reviewed By: q10 Differential Revision: D45139063 Pulled By: excelle08 fbshipit-source-id: 561ccf4ef5eb7778ea6a3443b2f604216fb4542b
48f6b01
to
2b0994e
Compare
This pull request was exported from Phabricator. Differential Revision: D45139063 |
Summary: Pull Request resolved: pytorch#1795 - Add `FindAVX.cmake` from the PyTorch project to detect AVX compilation flags and conditionally add them to FBGEMM_GPU targets Pull Request resolved: pytorch#1698 Test Plan: On a devarm machine: - Set up the environment according to [the building instruction](https://github.com/pytorch/FBGEMM/blob/main/fbgemm_gpu/docs/BuildInstructions.md). For convenience we can source `.github/scripts/setup_env.bash` in the FBGEMM repo and run the functions in it. - Build and install CPU-only variant of fbgemm_gpu: # Assuming the CWD is the FBGEMM repo directory and setup_env.bash is sourced cd fbgemm_gpu build_fbgemm_gpu_install <conda-env-name> cpu Example log of a a successful building and installing process is like this: P717846260 - Run tests: `run_fbgemm_gpu_tests mycondaenv cpu` under `fbgemm_gpu/tests` directory Reviewed By: q10 Differential Revision: D45139063 Pulled By: excelle08 fbshipit-source-id: 86b5cc55c33861f09265746dda7427359a4db3e0
2b0994e
to
1faf2ba
Compare
This pull request was exported from Phabricator. Differential Revision: D45139063 |
Summary: Pull Request resolved: pytorch#1795 - Add `FindAVX.cmake` from the PyTorch project to detect AVX compilation flags and conditionally add them to FBGEMM_GPU targets Pull Request resolved: pytorch#1698 Test Plan: On a devarm machine: - Set up the environment according to [the building instruction](https://github.com/pytorch/FBGEMM/blob/main/fbgemm_gpu/docs/BuildInstructions.md). For convenience we can source `.github/scripts/setup_env.bash` in the FBGEMM repo and run the functions in it. - Build and install CPU-only variant of fbgemm_gpu: # Assuming the CWD is the FBGEMM repo directory and setup_env.bash is sourced cd fbgemm_gpu build_fbgemm_gpu_install <conda-env-name> cpu Example log of a a successful building and installing process is like this: P717846260 - Run tests: `run_fbgemm_gpu_tests mycondaenv cpu` under `fbgemm_gpu/tests` directory Reviewed By: q10 Differential Revision: D45139063 Pulled By: excelle08 fbshipit-source-id: d9f6660d84d6ed6add3742ecb58ee18c37d2591a
1faf2ba
to
dd825f4
Compare
This pull request was exported from Phabricator. Differential Revision: D45139063 |
Summary: Pull Request resolved: pytorch#1795 - Add `FindAVX.cmake` from the PyTorch project to detect AVX compilation flags and conditionally add them to FBGEMM_GPU targets Pull Request resolved: pytorch#1698 Test Plan: On a devarm machine: - Set up the environment according to [the building instruction](https://github.com/pytorch/FBGEMM/blob/main/fbgemm_gpu/docs/BuildInstructions.md). For convenience we can source `.github/scripts/setup_env.bash` in the FBGEMM repo and run the functions in it. - Build and install CPU-only variant of fbgemm_gpu: # Assuming the CWD is the FBGEMM repo directory and setup_env.bash is sourced cd fbgemm_gpu build_fbgemm_gpu_install <conda-env-name> cpu Example log of a a successful building and installing process is like this: P717846260 - Run tests: `run_fbgemm_gpu_tests mycondaenv cpu` under `fbgemm_gpu/tests` directory Reviewed By: q10 Differential Revision: D45139063 Pulled By: excelle08 fbshipit-source-id: fa1c1065888630dc1e9b8ec84e4493be3565596e
dd825f4
to
6438361
Compare
This pull request was exported from Phabricator. Differential Revision: D45139063 |
Summary: Pull Request resolved: pytorch#1795 - Add `FindAVX.cmake` from the PyTorch project to detect AVX compilation flags and conditionally add them to FBGEMM_GPU targets Pull Request resolved: pytorch#1698 Test Plan: On a devarm machine: - Set up the environment according to [the building instruction](https://github.com/pytorch/FBGEMM/blob/main/fbgemm_gpu/docs/BuildInstructions.md). For convenience we can source `.github/scripts/setup_env.bash` in the FBGEMM repo and run the functions in it. - Build and install CPU-only variant of fbgemm_gpu: # Assuming the CWD is the FBGEMM repo directory and setup_env.bash is sourced cd fbgemm_gpu build_fbgemm_gpu_install <conda-env-name> cpu Example log of a a successful building and installing process is like this: P717846260 - Run tests: `run_fbgemm_gpu_tests mycondaenv cpu` under `fbgemm_gpu/tests` directory Reviewed By: q10 Differential Revision: D45139063 Pulled By: excelle08 fbshipit-source-id: fe841aa6321881c7e7d76da8400f3d6eb0354b2e
6438361
to
cb09c9b
Compare
This pull request was exported from Phabricator. Differential Revision: D45139063 |
@excelle08 merged this pull request in 323afc5. |
Summary:
FindAVX.cmake
from the PyTorch project to detect AVX compilation flags and conditionally add them to FBGEMM_GPU targetsTest Plan:
On a devarm machine:
.github/scripts/setup_env.bash
in the FBGEMM repo and run the functions in it.# Assuming the CWD is the FBGEMM repo directory and setup_env.bash is sourced
cd fbgemm_gpu
build_fbgemm_gpu_install cpu
Example log of a a successful building and installing process is like this: P717846260
run_fbgemm_gpu_tests mycondaenv cpu
underfbgemm_gpu/tests
directoryDifferential Revision: D45139063
Pulled By: excelle08