Skip to content

[PyTorch] Change representation of SizesAndStrides #47508

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.

Sign up for GitHub

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

Jump to bottom
Closed
wants to merge 22 commits into from
Closed

[PyTorch] Change representation of SizesAndStrides #47508

wants to merge 22 commits into from

Conversation

swolchok
Copy link
Contributor

@swolchok swolchok commented Nov 6, 2020 •
edited
Loading

Stack from ghstack:

This moves SizesAndStrides to a specialized representation
that is 5 words smaller in the common case of tensor rank 5 or less.

Differential Revision: D24772023

NOTE FOR REVIEWERS: This PR has internal Facebook specific changes or comments, please review them on Phabricator!

@swolchok
[PyTorch] Change representation of SizesAndStrides
802178d 
This moves SizesAndStrides to a specialized representation
that is 5 words smaller in the common case of tensor rank 5 or less.

Differential Revision: [D24772023](https://our.internmc.facebook.com/intern/diff/D24772023/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D24772023/)!

[ghstack-poisoned]
@swolchok swolchok mentioned this pull request Nov 6, 2020
Closed
swolchok added a commit that referenced this pull request Nov 6, 2020
@swolchok
[PyTorch] Change representation of SizesAndStrides
0dc7d21 
This moves SizesAndStrides to a specialized representation
that is 5 words smaller in the common case of tensor rank 5 or less.

Differential Revision: [D24772023](https://our.internmc.facebook.com/intern/diff/D24772023/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D24772023/)!

ghstack-source-id: 116029079
Pull Request resolved: #47508
@dr-ci
Copy link

dr-ci bot commented Nov 6, 2020 •
edited by facebook-github-bot
Loading

💊 CI failures summary and remediations

As of commit 6974421 (more details on the Dr. CI page):

  • 3/3 failures possibly* introduced in this PR
    • 2/3 non-CircleCI failure(s)

🕵️ 1 new failure recognized by patterns

The following CI failures do not appear to be due to upstream breakages:

See CircleCI build pytorch_xla_linux_bionic_py3_6_clang9_build (1/1)

Step: "Build" (full log | diagnosis details | 🔁 rerun)

Jan 04 22:42:06 torch_xla/csrc/tensor_impl.cpp:143:16: error: no member named 'push_back' in 'c10::ArrayRef' 
Jan 04 22:42:06       strides_.push_back(stride);
Jan 04 22:42:06       ^~~~~~~~
Jan 04 22:42:06       strides
Jan 04 22:42:06 /var/lib/jenkins/workspace/c10/core/TensorImpl.h:387:23: note: 'strides' declared here
Jan 04 22:42:06   virtual IntArrayRef strides() const;
Jan 04 22:42:06                       ^
Jan 04 22:42:06 torch_xla/csrc/tensor_impl.cpp:143:7: error: reference to non-static member function must be called; did you mean to call it with no arguments?
Jan 04 22:42:06       strides_.push_back(stride);
Jan 04 22:42:06       ^~~~~~~~
Jan 04 22:42:06               ()
Jan 04 22:42:06 torch_xla/csrc/tensor_impl.cpp:143:16: error: no member named 'push_back' in 'c10::ArrayRef<long>'
Jan 04 22:42:06       strides_.push_back(stride);
Jan 04 22:42:06       ~~~~~~~~ ^
Jan 04 22:42:08 12 errors generated.
Jan 04 22:42:08 clang-9 -Wsign-compare -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes -fPIC -I/var/lib/jenkins/workspace/xla -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-tensorflow -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-bin -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-tensorflow/external/protobuf_archive/src -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-tensorflow/external/com_google_protobuf/src -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-tensorflow/external/eigen_archive -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-tensorflow/external/com_google_absl -I/var/lib/jenkins/workspace -I/var/lib/jenkins/workspace/torch/csrc -I/var/lib/jenkins/workspace/torch/lib/tmp_install/include -I/opt/conda/lib/python3.6/site-packages/torch/include -I/opt/conda/lib/python3.6/site-packages/torch/include/torch/csrc/api/include -I/opt/conda/lib/python3.6/site-packages/torch/include/TH -I/opt/conda/lib/python3.6/site-packages/torch/include/THC -I/opt/conda/include/python3.6m -c torch_xla/csrc/python_util.cpp -o build/temp.linux-x86_64-3.6/torch_xla/csrc/python_util.o -std=c++14 -Wno-sign-compare -Wno-deprecated-declarations -Wno-return-type -Wno-macro-redefined -Wno-return-std-move -DNDEBUG -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE="_clang" -DPYBIND11_STDLIB="_libstdcpp" -DPYBIND11_BUILD_ABI="_cxxabi1002" -DTORCH_EXTENSION_NAME=_XLAC -D_GLIBCXX_USE_CXX11_ABI=1
Jan 04 22:42:10 clang-9 -Wsign-compare -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes -fPIC -I/var/lib/jenkins/workspace/xla -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-tensorflow -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-bin -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-tensorflow/external/protobuf_archive/src -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-tensorflow/external/com_google_protobuf/src -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-tensorflow/external/eigen_archive -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-tensorflow/external/com_google_absl -I/var/lib/jenkins/workspace -I/var/lib/jenkins/workspace/torch/csrc -I/var/lib/jenkins/workspace/torch/lib/tmp_install/include -I/opt/conda/lib/python3.6/site-packages/torch/include -I/opt/conda/lib/python3.6/site-packages/torch/include/torch/csrc/api/include -I/opt/conda/lib/python3.6/site-packages/torch/include/TH -I/opt/conda/lib/python3.6/site-packages/torch/include/THC -I/opt/conda/include/python3.6m -c torch_xla/csrc/device.cpp -o build/temp.linux-x86_64-3.6/torch_xla/csrc/device.o -std=c++14 -Wno-sign-compare -Wno-deprecated-declarations -Wno-return-type -Wno-macro-redefined -Wno-return-std-move -DNDEBUG -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE="_clang" -DPYBIND11_STDLIB="_libstdcpp" -DPYBIND11_BUILD_ABI="_cxxabi1002" -DTORCH_EXTENSION_NAME=_XLAC -D_GLIBCXX_USE_CXX11_ABI=1
Jan 04 22:42:11 clang-9 -Wsign-compare -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes -fPIC -I/var/lib/jenkins/workspace/xla -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-tensorflow -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-bin -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-tensorflow/external/protobuf_archive/src -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-tensorflow/external/com_google_protobuf/src -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-tensorflow/external/eigen_archive -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-tensorflow/external/com_google_absl -I/var/lib/jenkins/workspace -I/var/lib/jenkins/workspace/torch/csrc -I/var/lib/jenkins/workspace/torch/lib/tmp_install/include -I/opt/conda/lib/python3.6/site-packages/torch/include -I/opt/conda/lib/python3.6/site-packages/torch/include/torch/csrc/api/include -I/opt/conda/lib/python3.6/site-packages/torch/include/TH -I/opt/conda/lib/python3.6/site-packages/torch/include/THC -I/opt/conda/include/python3.6m -c torch_xla/csrc/init_python_bindings.cpp -o build/temp.linux-x86_64-3.6/torch_xla/csrc/init_python_bindings.o -std=c++14 -Wno-sign-compare -Wno-deprecated-declarations -Wno-return-type -Wno-macro-redefined -Wno-return-std-move -DNDEBUG -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE="_clang" -DPYBIND11_STDLIB="_libstdcpp" -DPYBIND11_BUILD_ABI="_cxxabi1002" -DTORCH_EXTENSION_NAME=_XLAC -D_GLIBCXX_USE_CXX11_ABI=1
Jan 04 22:42:14 clang-9 -Wsign-compare -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes -fPIC -I/var/lib/jenkins/workspace/xla -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-tensorflow -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-bin -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-tensorflow/external/protobuf_archive/src -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-tensorflow/external/com_google_protobuf/src -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-tensorflow/external/eigen_archive -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-tensorflow/external/com_google_absl -I/var/lib/jenkins/workspace -I/var/lib/jenkins/workspace/torch/csrc -I/var/lib/jenkins/workspace/torch/lib/tmp_install/include -I/opt/conda/lib/python3.6/site-packages/torch/include -I/opt/conda/lib/python3.6/site-packages/torch/include/torch/csrc/api/include -I/opt/conda/lib/python3.6/site-packages/torch/include/TH -I/opt/conda/lib/python3.6/site-packages/torch/include/THC -I/opt/conda/include/python3.6m -c torch_xla/csrc/debug_util.cpp -o build/temp.linux-x86_64-3.6/torch_xla/csrc/debug_util.o -std=c++14 -Wno-sign-compare -Wno-deprecated-declarations -Wno-return-type -Wno-macro-redefined -Wno-return-std-move -DNDEBUG -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE="_clang" -DPYBIND11_STDLIB="_libstdcpp" -DPYBIND11_BUILD_ABI="_cxxabi1002" -DTORCH_EXTENSION_NAME=_XLAC -D_GLIBCXX_USE_CXX11_ABI=1
Jan 04 22:42:16 clang-9 -Wsign-compare -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes -fPIC -I/var/lib/jenkins/workspace/xla -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-tensorflow -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-bin -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-tensorflow/external/protobuf_archive/src -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-tensorflow/external/com_google_protobuf/src -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-tensorflow/external/eigen_archive -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-tensorflow/external/com_google_absl -I/var/lib/jenkins/workspace -I/var/lib/jenkins/workspace/torch/csrc -I/var/lib/jenkins/workspace/torch/lib/tmp_install/include -I/opt/conda/lib/python3.6/site-packages/torch/include -I/opt/conda/lib/python3.6/site-packages/torch/include/torch/csrc/api/include -I/opt/conda/lib/python3.6/site-packages/torch/include/TH -I/opt/conda/lib/python3.6/site-packages/torch/include/THC -I/opt/conda/include/python3.6m -c torch_xla/csrc/aten_xla_type_default.cpp -o build/temp.linux-x86_64-3.6/torch_xla/csrc/aten_xla_type_default.o -std=c++14 -Wno-sign-compare -Wno-deprecated-declarations -Wno-return-type -Wno-macro-redefined -Wno-return-std-move -DNDEBUG -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE="_clang" -DPYBIND11_STDLIB="_libstdcpp" -DPYBIND11_BUILD_ABI="_cxxabi1002" -DTORCH_EXTENSION_NAME=_XLAC -D_GLIBCXX_USE_CXX11_ABI=1
Jan 04 22:42:27 clang-9 -Wsign-compare -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes -fPIC -I/var/lib/jenkins/workspace/xla -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-tensorflow -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-bin -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-tensorflow/external/protobuf_archive/src -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-tensorflow/external/com_google_protobuf/src -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-tensorflow/external/eigen_archive -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-tensorflow/external/com_google_absl -I/var/lib/jenkins/workspace -I/var/lib/jenkins/workspace/torch/csrc -I/var/lib/jenkins/workspace/torch/lib/tmp_install/include -I/opt/conda/lib/python3.6/site-packages/torch/include -I/opt/conda/lib/python3.6/site-packages/torch/include/torch/csrc/api/include -I/opt/conda/lib/python3.6/site-packages/torch/include/TH -I/opt/conda/lib/python3.6/site-packages/torch/include/THC -I/opt/conda/include/python3.6m -c torch_xla/csrc/batch_norm.cpp -o build/temp.linux-x86_64-3.6/torch_xla/csrc/batch_norm.o -std=c++14 -Wno-sign-compare -Wno-deprecated-declarations -Wno-return-type -Wno-macro-redefined -Wno-return-std-move -DNDEBUG -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE="_clang" -DPYBIND11_STDLIB="_libstdcpp" -DPYBIND11_BUILD_ABI="_cxxabi1002" -DTORCH_EXTENSION_NAME=_XLAC -D_GLIBCXX_USE_CXX11_ABI=1
Jan 04 22:42:33 clang-9 -Wsign-compare -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes -fPIC -I/var/lib/jenkins/workspace/xla -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-tensorflow -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-bin -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-tensorflow/external/protobuf_archive/src -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-tensorflow/external/com_google_protobuf/src -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-tensorflow/external/eigen_archive -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-tensorflow/external/com_google_absl -I/var/lib/jenkins/workspace -I/var/lib/jenkins/workspace/torch/csrc -I/var/lib/jenkins/workspace/torch/lib/tmp_install/include -I/opt/conda/lib/python3.6/site-packages/torch/include -I/opt/conda/lib/python3.6/site-packages/torch/include/torch/csrc/api/include -I/opt/conda/lib/python3.6/site-packages/torch/include/TH -I/opt/conda/lib/python3.6/site-packages/torch/include/THC -I/opt/conda/include/python3.6m -c torch_xla/csrc/layout_manager.cpp -o build/temp.linux-x86_64-3.6/torch_xla/csrc/layout_manager.o -std=c++14 -Wno-sign-compare -Wno-deprecated-declarations -Wno-return-type -Wno-macro-redefined -Wno-return-std-move -DNDEBUG -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE="_clang" -DPYBIND11_STDLIB="_libstdcpp" -DPYBIND11_BUILD_ABI="_cxxabi1002" -DTORCH_EXTENSION_NAME=_XLAC -D_GLIBCXX_USE_CXX11_ABI=1

ci.pytorch.org: 1 failed

This comment was automatically generated by Dr. CI (expand for details).Follow this link to opt-out of these comments for your Pull Requests.

Please report bugs/suggestions to the (internal) Dr. CI Users group.

This comment has been revised 171 times.

@ezyang ezyang self-requested a review November 6, 2020 22:16
@swolchok
add more test coverage and fix resize bug on "[PyTorch] Change repres…
73cc2b0 
…entation of SizesAndStrides"

This moves SizesAndStrides to a specialized representation
that is 5 words smaller in the common case of tensor rank 5 or less.

Differential Revision: [D24772023](https://our.internmc.facebook.com/intern/diff/D24772023/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D24772023/)!

[ghstack-poisoned]
@swolchok
rebase on "[PyTorch] Change representation of SizesAndStrides"
1023d40 
This moves SizesAndStrides to a specialized representation
that is 5 words smaller in the common case of tensor rank 5 or less.

Differential Revision: [D24772023](https://our.internmc.facebook.com/intern/diff/D24772023/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D24772023/)!

[ghstack-poisoned]
swolchok added a commit that referenced this pull request Nov 9, 2020
@swolchok
[PyTorch] Change representation of SizesAndStrides
262da8e 
Pull Request resolved: #47508

This moves SizesAndStrides to a specialized representation
that is 5 words smaller in the common case of tensor rank 5 or less.
ghstack-source-id: 116190714

Differential Revision: [D24772023](https://our.internmc.facebook.com/intern/diff/D24772023/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D24772023/)!
@ezyang
Copy link
Contributor

ezyang commented Nov 10, 2020

I know the point of this change isn't really to affect instruction count, but I'm curious to know which way instruction count on the microbenchmarks goes this way. I'm sort of expecting count to go up but maybe I will be surprised.

ezyang
ezyang reviewed Nov 10, 2020
View reviewed changes
ezyang
ezyang reviewed Nov 10, 2020
View reviewed changes
@ezyang ezyang requested review from smessmer and robieta November 10, 2020 19:42
ezyang
ezyang reviewed Nov 10, 2020
View reviewed changes
ezyang
ezyang reviewed Nov 10, 2020
View reviewed changes
ezyang
ezyang reviewed Nov 10, 2020
View reviewed changes
ezyang
ezyang reviewed Nov 10, 2020
View reviewed changes
@swolchok
fix move-assign. delete unused method. vector-less representation com…
0cb96c9 
…ing in next update on "[PyTorch] Change representation of SizesAndStrides"

This moves SizesAndStrides to a specialized representation
that is 5 words smaller in the common case of tensor rank 5 or less.

Differential Revision: [D24772023](https://our.internmc.facebook.com/intern/diff/D24772023/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D24772023/)!

[ghstack-poisoned]
swolchok added a commit that referenced this pull request Nov 10, 2020
@swolchok
[PyTorch] Change representation of SizesAndStrides
0e60d4e 
Pull Request resolved: #47508

This moves SizesAndStrides to a specialized representation
that is 5 words smaller in the common case of tensor rank 5 or less.
ghstack-source-id: 116353586

Differential Revision: [D24772023](https://our.internmc.facebook.com/intern/diff/D24772023/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D24772023/)!
smessmer
smessmer reviewed Nov 10, 2020
View reviewed changes
Copy link
Contributor

@smessmer smessmer left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Not adding comments since you're planning a major rewrite. Let me know when that is finished so I can take a look.

@robieta
Copy link

robieta commented Nov 11, 2020

Sorry, #47507 (comment) was meant for this PR.

@swolchok
change to @ezyang's suggested representation. haven't gotten to do a …
9eddb4b 
…cleanup yet on "[PyTorch] Change representation of SizesAndStrides"

This moves SizesAndStrides to a specialized representation
that is 5 words smaller in the common case of tensor rank 5 or less.

Differential Revision: [D24772023](https://our.internmc.facebook.com/intern/diff/D24772023/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D24772023/)!

[ghstack-poisoned]
@swolchok
remove debug-only features on "[PyTorch] Change representation of Siz…
9b93702 
…esAndStrides"

This moves SizesAndStrides to a specialized representation
that is 5 words smaller in the common case of tensor rank 5 or less.

Differential Revision: [D24772023](https://our.internmc.facebook.com/intern/diff/D24772023/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D24772023/)!

[ghstack-poisoned]
swolchok added a commit that referenced this pull request Nov 11, 2020
@swolchok
[PyTorch] Change representation of SizesAndStrides
4c99d9a 
Pull Request resolved: #47508

This moves SizesAndStrides to a specialized representation
that is 5 words smaller in the common case of tensor rank 5 or less.
ghstack-source-id: 116368231

Differential Revision: [D24772023](https://our.internmc.facebook.com/intern/diff/D24772023/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D24772023/)!
@swolchok
fix typo on "[PyTorch] Change representation of SizesAndStrides"
9932aad 
This moves SizesAndStrides to a specialized representation
that is 5 words smaller in the common case of tensor rank 5 or less.

Differential Revision: [D24772023](https://our.internmc.facebook.com/intern/diff/D24772023/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D24772023/)!

[ghstack-poisoned]
swolchok added a commit that referenced this pull request Nov 11, 2020
@swolchok
[PyTorch] Change representation of SizesAndStrides
ab0a0d0 
Pull Request resolved: #47508

This moves SizesAndStrides to a specialized representation
that is 5 words smaller in the common case of tensor rank 5 or less.
ghstack-source-id: 116385725

Differential Revision: [D24772023](https://our.internmc.facebook.com/intern/diff/D24772023/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D24772023/)!
@swolchok
Update on "[PyTorch] Change representation of SizesAndStrides"
3632cd2 
This moves SizesAndStrides to a specialized representation
that is 5 words smaller in the common case of tensor rank 5 or less.

Differential Revision: [D24772023](https://our.internmc.facebook.com/intern/diff/D24772023/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D24772023/)!

[ghstack-poisoned]
@swolchok
remove more noexcept specs on "[PyTorch] Change representation of Siz…
1eecc5e 
…esAndStrides"

This moves SizesAndStrides to a specialized representation
that is 5 words smaller in the common case of tensor rank 5 or less.

Differential Revision: [D24772023](https://our.internmc.facebook.com/intern/diff/D24772023/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D24772023/)!

[ghstack-poisoned]
swolchok added a commit that referenced this pull request Nov 30, 2020
@swolchok
[PyTorch] Change representation of SizesAndStrides
4ac43c3 
Pull Request resolved: #47508

This moves SizesAndStrides to a specialized representation
that is 5 words smaller in the common case of tensor rank 5 or less.
ghstack-source-id: 117505381

Differential Revision: [D24772023](https://our.internmc.facebook.com/intern/diff/D24772023/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D24772023/)!
@bhosmer bhosmer self-requested a review December 1, 2020 01:16
@swolchok
Update on "[PyTorch] Change representation of SizesAndStrides"
1b4da8a 
This moves SizesAndStrides to a specialized representation
that is 5 words smaller in the common case of tensor rank 5 or less.

Differential Revision: [D24772023](https://our.internmc.facebook.com/intern/diff/D24772023/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D24772023/)!

[ghstack-poisoned]
swolchok added a commit that referenced this pull request Dec 1, 2020
@swolchok
[PyTorch] Change representation of SizesAndStrides
da911d8 
Pull Request resolved: #47508

This moves SizesAndStrides to a specialized representation
that is 5 words smaller in the common case of tensor rank 5 or less.
ghstack-source-id: 117593407

Differential Revision: [D24772023](https://our.internmc.facebook.com/intern/diff/D24772023/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D24772023/)!
@swolchok
fix 2 typos in bottom diff on "[PyTorch] Change representation of Siz…
5b207de 
…esAndStrides"

This moves SizesAndStrides to a specialized representation
that is 5 words smaller in the common case of tensor rank 5 or less.

Differential Revision: [D24772023](https://our.internmc.facebook.com/intern/diff/D24772023/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D24772023/)!

[ghstack-poisoned]
@swolchok
remove freeOutOfLineStorage so that operator= can be properly noexcep…
51d3870 
…t on "[PyTorch] Change representation of SizesAndStrides"

This moves SizesAndStrides to a specialized representation
that is 5 words smaller in the common case of tensor rank 5 or less.

Differential Revision: [D24772023](https://our.internmc.facebook.com/intern/diff/D24772023/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D24772023/)!

[ghstack-poisoned]
swolchok added a commit that referenced this pull request Dec 5, 2020
@swolchok
[PyTorch] Change representation of SizesAndStrides
a834689 
Pull Request resolved: #47508

This moves SizesAndStrides to a specialized representation
that is 5 words smaller in the common case of tensor rank 5 or less.
ghstack-source-id: 117939947

Differential Revision: [D24772023](https://our.internmc.facebook.com/intern/diff/D24772023/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D24772023/)!
@swolchok
Update on "[PyTorch] Change representation of SizesAndStrides"
63f6657 
This moves SizesAndStrides to a specialized representation
that is 5 words smaller in the common case of tensor rank 5 or less.

Differential Revision: [D24772023](https://our.internmc.facebook.com/intern/diff/D24772023/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D24772023/)!

[ghstack-poisoned]
swolchok added a commit that referenced this pull request Dec 7, 2020
@swolchok
[PyTorch] Change representation of SizesAndStrides
3c4c549 
Pull Request resolved: #47508

This moves SizesAndStrides to a specialized representation
that is 5 words smaller in the common case of tensor rank 5 or less.
ghstack-source-id: 118022738

Differential Revision: [D24772023](https://our.internmc.facebook.com/intern/diff/D24772023/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D24772023/)!
@swolchok
rebase on "[PyTorch] Change representation of SizesAndStrides"
994f9bb 
This moves SizesAndStrides to a specialized representation
that is 5 words smaller in the common case of tensor rank 5 or less.

Differential Revision: [D24772023](https://our.internmc.facebook.com/intern/diff/D24772023/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D24772023/)!

[ghstack-poisoned]
swolchok added a commit that referenced this pull request Dec 9, 2020
@swolchok
[PyTorch] Change representation of SizesAndStrides
ef5c068 
Pull Request resolved: #47508

This moves SizesAndStrides to a specialized representation
that is 5 words smaller in the common case of tensor rank 5 or less.
ghstack-source-id: 118146841

Differential Revision: [D24772023](https://our.internmc.facebook.com/intern/diff/D24772023/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D24772023/)!
@swolchok
add definition of static constexpr as required for C++14 on "[PyTorch…
d20f703 
…] Change representation of SizesAndStrides"

This moves SizesAndStrides to a specialized representation
that is 5 words smaller in the common case of tensor rank 5 or less.

Differential Revision: [D24772023](https://our.internmc.facebook.com/intern/diff/D24772023/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D24772023/)!

[ghstack-poisoned]
swolchok added a commit that referenced this pull request Dec 9, 2020
@swolchok
[PyTorch] Change representation of SizesAndStrides
729bb51 
Pull Request resolved: #47508

This moves SizesAndStrides to a specialized representation
that is 5 words smaller in the common case of tensor rank 5 or less.
ghstack-source-id: 118223017

Differential Revision: [D24772023](https://our.internmc.facebook.com/intern/diff/D24772023/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D24772023/)!
@swolchok
Update on "[PyTorch] Change representation of SizesAndStrides"
44e1296 
This moves SizesAndStrides to a specialized representation
that is 5 words smaller in the common case of tensor rank 5 or less.

Differential Revision: [D24772023](https://our.internmc.facebook.com/intern/diff/D24772023/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D24772023/)!

[ghstack-poisoned]
swolchok added a commit that referenced this pull request Dec 14, 2020
@swolchok
[PyTorch] Change representation of SizesAndStrides
f0cabe0 
Pull Request resolved: #47508

This moves SizesAndStrides to a specialized representation
that is 5 words smaller in the common case of tensor rank 5 or less.
ghstack-source-id: 118554470

Differential Revision: [D24772023](https://our.internmc.facebook.com/intern/diff/D24772023/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D24772023/)!
@swolchok
Update on "[PyTorch] Change representation of SizesAndStrides"
91c27db 
This moves SizesAndStrides to a specialized representation
that is 5 words smaller in the common case of tensor rank 5 or less.

Differential Revision: [D24772023](https://our.internmc.facebook.com/intern/diff/D24772023/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D24772023/)!

[ghstack-poisoned]
swolchok added a commit that referenced this pull request Dec 15, 2020
@swolchok
[PyTorch] Change representation of SizesAndStrides
595cda4 
Pull Request resolved: #47508

This moves SizesAndStrides to a specialized representation
that is 5 words smaller in the common case of tensor rank 5 or less.
ghstack-source-id: 118633676

Differential Revision: [D24772023](https://our.internmc.facebook.com/intern/diff/D24772023/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D24772023/)!
@swolchok
Update on "[PyTorch] Change representation of SizesAndStrides"
6974421 
This moves SizesAndStrides to a specialized representation
that is 5 words smaller in the common case of tensor rank 5 or less.

Differential Revision: [D24772023](https://our.internmc.facebook.com/intern/diff/D24772023/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D24772023/)!

[ghstack-poisoned]
swolchok added a commit that referenced this pull request Jan 4, 2021
@swolchok
[PyTorch] Change representation of SizesAndStrides
d0657bc 
Pull Request resolved: #47508

This moves SizesAndStrides to a specialized representation
that is 5 words smaller in the common case of tensor rank 5 or less.
ghstack-source-id: 119313560

Differential Revision: [D24772023](https://our.internmc.facebook.com/intern/diff/D24772023/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D24772023/)!
@facebook-github-bot
Copy link
Contributor

This pull request has been merged in b73c018.

@facebook-github-bot facebook-github-bot deleted the gh/swolchok/10/head branch January 11, 2021 15:17
hwangdeyu pushed a commit to hwangdeyu/pytorch that referenced this pull request Jan 14, 2021
@swolchok
[PyTorch] Change representation of SizesAndStrides (pytorch#47508)
20e9dd8 
Summary:
Pull Request resolved: pytorch#47508

This moves SizesAndStrides to a specialized representation
that is 5 words smaller in the common case of tensor rank 5 or less.
ghstack-source-id: 119313560

Test Plan:
SizesAndStridesTest added in previous diff passes under
ASAN + UBSAN.

Run framework overhead benchmarks. Looks more or less neutral.

Reviewed By: ezyang

Differential Revision: D24772023

fbshipit-source-id: 0a75fd6c2daabb0769e2f803e80e2d6831871316
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@smessmer smessmer smessmer left review comments

@ezyang ezyang ezyang approved these changes

@bhosmer bhosmer bhosmer approved these changes

@robieta robieta Awaiting requested review from robieta

Assignees
No one assigned
Labels
cla signed Merged
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
6 participants
@swolchok @ezyang @robieta @facebook-github-bot @bhosmer @smessmer