Thank you for your interest in contributing to PyTorch! If you're a new contributor, please first take a read through our Contributing Guide, specifically the Submitting a Change section that walks through the process of contributing a change to PyTorch.

The rest of this document (CONTRIBUTING.md) covers some of the more technical aspects of contributing to PyTorch.

• Developing PyTorch
  • Setup the development environment
  • Tips and Debugging
• Nightly Checkout & Pull
• Codebase structure
• Unit testing
  • Python Unit Testing
  • Better local unit tests with pytest
  • Local linting
    • Running mypy
  • C++ Unit Testing
  • Run Specific CI Jobs
• Merging your Change
• Writing documentation
  • Docstring type formatting
  • Building documentation
    • Tips
    • Building C++ Documentation
  • Previewing changes locally
  • Previewing documentation on PRs
  • Adding documentation tests
• Profiling with py-spy
• Managing multiple build trees
• C++ development tips
  • Build only what you need
  • Code completion and IDE support
  • Make no-op build fast
    • Use Ninja
    • Use CCache
    • Use a faster linker
    • Use pre-compiled headers
    • Workaround for header dependency bug in nvcc
  • Rebuild few files with debug information
  • C++ frontend development tips
  • GDB integration
  • C++ stacktraces
• CUDA development tips
• Windows development tips
  • Known MSVC (and MSVC with NVCC) bugs
  • Building on legacy code and CUDA
• Pre-commit tidy/linting hook
• Building PyTorch with ASAN
  • Getting ccache to work
  • Why this stuff with LD_PRELOAD and LIBASAN_RT?
  • Why LD_PRELOAD in the build function?
  • Why no leak detection?
• Caffe2 notes
• CI failure tips
  • Which commit is used in CI?
• Dev Infra Office Hours

Follow the instructions for installing PyTorch from source. If you get stuck when developing PyTorch on your machine, check out the tips and debugging section below for common solutions.

First, you need to fork the PyTorch project on GitHub and follow the instructions at Connecting to GitHub with SSH to setup your SSH authentication credentials.

Then clone the PyTorch project and setup the development environment:

git clone git@github.com:<USERNAME>/pytorch.git
cd pytorch
git remote add upstream git@github.com:pytorch/pytorch.git

make setup-env  # or make setup-env-cuda for pre-built CUDA binaries
conda activate pytorch-deps

• If you want to have no-op incremental rebuilds (which are fast), see Make no-op build fast below.

• When installing with python setup.py develop (in contrast to python setup.py install) Python runtime will use the current local source-tree when importing torch package. (This is done by creating .egg-link file in site-packages folder) This way you do not need to repeatedly install after modifying Python files (.py). However, you would need to reinstall if you modify Python interface (.pyi, .pyi.in) or non-Python files (.cpp, .cc, .cu, .h, ...).

  One way to avoid running python setup.py develop every time one makes a change to C++/CUDA/ObjectiveC files on Linux/Mac, is to create a symbolic link from build folder to torch/lib, for example, by issuing following:

   pushd torch/lib; sh -c "ln -sf ../../build/lib/libtorch_cpu.* ."; popd

  Afterwards rebuilding a library (for example to rebuild libtorch_cpu.so issue ninja torch_cpu from build folder), would be sufficient to make change visible in torch package.

  To reinstall, first uninstall all existing PyTorch installs. You may need to run pip uninstall torch multiple times. You'll know torch is fully uninstalled when you see WARNING: Skipping torch as it is not installed. (You should only have to pip uninstall a few times, but you can always uninstall with timeout or in a loop if you're feeling lazy.)

  conda uninstall pytorch -y
  yes | pip uninstall torch

  Next run python setup.py clean. After that, you can install in develop mode again.

• If you run into errors when running python setup.py develop, here are some debugging steps:

  1. Run printf '#include <stdio.h>\nint main() { printf("Hello World");}'|clang -x c -; ./a.out to make sure your CMake works and can compile this simple Hello World program without errors.
  2. Nuke your build directory. The setup.py script compiles binaries into the build folder and caches many details along the way, which saves time the next time you build. If you're running into issues, you can always rm -rf build from the toplevel pytorch directory and start over.
  3. If you have made edits to the PyTorch repo, commit any change you'd like to keep and clean the repo with the following commands (note that clean really removes all untracked files and changes.):

     git submodule deinit -f .
     git clean -xdf
     python setup.py clean
     git submodule update --init --recursive # very important to sync the submodules
     python setup.py develop                 # then try running the command again

  4. The main step within python setup.py develop is running make from the build directory. If you want to experiment with some environment variables, you can pass them into the command:

     ENV_KEY1=ENV_VAL1[, ENV_KEY2=ENV_VAL2]* python setup.py develop

• If you run into issue running git submodule update --init --recursive. Please try the following:

  • If you encounter an error such as

    error: Submodule 'third_party/pybind11' could not be updated
    

    check whether your Git local or global config file contains any submodule.* settings. If yes, remove them and try again. (please reference this doc for more info).

  • If you encounter an error such as

    fatal: unable to access 'https://github.com/pybind11/pybind11.git': could not load PEM client certificate ...
    

    this is likely that you are using HTTP proxying and the certificate expired. To check if the certificate is valid, run git config --global --list and search for config like http.proxysslcert=<cert_file>. Then check certificate valid date by running

    openssl x509 -noout -in <cert_file> -dates

  • If you encounter an error that some third_party modules are not checked out correctly, such as

    Could not find .../pytorch/third_party/pybind11/CMakeLists.txt
    

    remove any submodule.* settings in your local git config (.git/config of your pytorch repo) and try again.

• If you're a Windows contributor, please check out Best Practices.

• For help with any part of the contributing process, please don’t hesitate to utilize our Zoom office hours! See details here

The tools/nightly.py script is provided to ease pure Python development of PyTorch. This uses conda and git to check out the nightly development version of PyTorch and installs pre-built binaries into the current repository. This is like a development or editable install, but without needing the ability to compile any C++ code.

You can use this script to check out a new nightly branch with the following:

./tools/nightly.py checkout -b my-nightly-branch
conda activate pytorch-deps

Or if you would like to re-use an existing conda environment, you can pass in the regular environment parameters (--name or --prefix):

./tools/nightly.py checkout -b my-nightly-branch -n my-env
conda activate my-env

To install the nightly binaries built with CUDA, you can pass in the flag --cuda:

./tools/nightly.py checkout -b my-nightly-branch --cuda
conda activate pytorch-deps

You can also use this tool to pull the nightly commits into the current branch:

./tools/nightly.py pull -n my-env
conda activate my-env

Pulling will reinstall the PyTorch dependencies as well as the nightly binaries into the repo directory.

• c10 - Core library files that work everywhere, both server and mobile. We are slowly moving pieces from ATen/core here. This library is intended only to contain essential functionality, and appropriate to use in settings where binary size matters. (But you'll have a lot of missing functionality if you try to use it directly.)
• aten - C++ tensor library for PyTorch (no autograd support)
  • src - README
    • ATen
      • core - Core functionality of ATen. This is migrating to top-level c10 folder.
      • native - Modern implementations of operators. If you want to write a new operator, here is where it should go. Most CPU operators go in the top level directory, except for operators which need to be compiled specially; see cpu below.
        • cpu - Not actually CPU implementations of operators, but specifically implementations which are compiled with processor-specific instructions, like AVX. See the README for more details.
        • cuda - CUDA implementations of operators.
        • sparse - CPU and CUDA implementations of COO sparse tensor operations
        • mkl mkldnn miopen cudnn
          • implementations of operators which simply bind to some backend library.
        • quantized - Quantized tensor (i.e. QTensor) operation implementations. README contains details including how to implement native quantized operations.
• torch - The actual PyTorch library. Everything that is not in csrc is a Python module, following the PyTorch Python frontend module structure.
  • csrc - C++ files composing the PyTorch library. Files in this directory tree are a mix of Python binding code, and C++ heavy lifting. Consult setup.py for the canonical list of Python binding files; conventionally, they are often prefixed with python_. README
    • jit - Compiler and frontend for TorchScript JIT frontend. README
    • autograd - Implementation of reverse-mode automatic differentiation. README
    • api - The PyTorch C++ frontend.
    • distributed - Distributed training support for PyTorch.
• tools - Code generation scripts for the PyTorch library. See README of this directory for more details.
• test - Python unit tests for PyTorch Python frontend.
  • test_torch.py - Basic tests for PyTorch functionality.
  • test_autograd.py - Tests for non-NN automatic differentiation support.
  • test_nn.py - Tests for NN operators and their automatic differentiation.
  • test_jit.py - Tests for the JIT compiler and TorchScript.
  • ...
  • cpp - C++ unit tests for PyTorch C++ frontend.
    • api - README
    • jit - README
    • tensorexpr - README
  • expect - Automatically generated "expect" files which are used to compare against expected output.
  • onnx - Tests for ONNX export functionality, using both PyTorch and Caffe2.
• caffe2 - The Caffe2 library.
  • core - Core files of Caffe2, e.g., tensor, workspace, blobs, etc.
  • operators - Operators of Caffe2.
  • python - Python bindings to Caffe2.
  • ...
• .circleci - CircleCI configuration management. README

Prerequisites: The following packages should be installed with either conda or pip:

• expecttest and hypothesis - required to run tests
• mypy - recommended for linting
• pytest - recommended to run tests more selectively

All PyTorch test suites are located in the test folder and start with test_. Run the entire test suite with

python test/run_test.py

or run individual test suites using the command python test/FILENAME.py, where FILENAME represents the file containing the test suite you wish to run.

For example, to run all the TorchScript JIT tests (located at test/test_jit.py), you would run:

python test/test_jit.py

You can narrow down what you're testing even further by specifying the name of an individual test with TESTCLASSNAME.TESTNAME. Here, TESTNAME is the name of the test you want to run, and TESTCLASSNAME is the name of the class in which it is defined.

Going off the above example, let's say you want to run test_Sequential, which is defined as part of the TestJit class in test/test_jit.py. Your command would be:

python test/test_jit.py TestJit.test_Sequential

Weird note: In our CI (Continuous Integration) jobs, we actually run the tests from the test folder and not the root of the repo, since there are various dependencies we set up for CI that expects the tests to be run from the test folder. As such, there may be some inconsistencies between local testing and CI testing--if you observe an inconsistency, please file an issue.

We don't officially support pytest, but it works well with our unittest tests and offers a number of useful features for local developing. Install it via pip install pytest.

If you want to just run tests that contain a specific substring, you can use the -k flag:

pytest test/test_nn.py -k Loss -v

The above is an example of testing a change to all Loss functions: this command runs tests such as TestNN.test_BCELoss and TestNN.test_MSELoss and can be useful to save keystrokes.

Install all prerequisites by running

make setup-lint

You can now run the same linting steps that are used in CI locally via make:

make lint

Learn more about the linter on the lintrunner wiki page

mypy is an optional static type checker for Python. We have multiple mypy configs for the PyTorch codebase that are automatically validated against whenever the linter is run.

See Guide for adding type annotations to PyTorch for more information on how to set up mypy and tackle type annotation tasks.

PyTorch offers a series of tests located in the test/cpp folder. These tests are written in C++ and use the Google Test testing framework. After compiling PyTorch from source, the test runner binaries will be written to the build/bin folder. The command to run one of these tests is ./build/bin/FILENAME --gtest_filter=TESTSUITE.TESTNAME, where TESTNAME is the name of the test you'd like to run and TESTSUITE is the suite that test is defined in.

For example, if you wanted to run the test MayContainAlias, which is part of the test suite ContainerAliasingTest in the file test/cpp/jit/test_alias_analysis.cpp, the command would be:

./build/bin/test_jit --gtest_filter=ContainerAliasingTest.MayContainAlias

You can generate a commit that limits the CI to only run a specific job by using tools/testing/explicit_ci_jobs.py like so:

# --job: specify one or more times to filter to a specific job + its dependencies
# --filter-gha: specify github actions workflows to keep
# --make-commit: commit CI changes to git with a message explaining the change
python tools/testing/explicit_ci_jobs.py --job binary_linux_manywheel_3_6m_cpu_devtoolset7_nightly_test --filter-gha '*generated*gcc5.4*' --make-commit

# Make your changes

ghstack submit

NB: It is not recommended to use this workflow unless you are also using ghstack. It creates a large commit that is of very low signal to reviewers.

If you know the right people or team that should approve your PR (and you have the required permissions to do so), add them to the Reviewers list.

If not, leave the Reviewers section empty. Our triage squad will review your PR, add a module label, and assign it to the appropriate reviewer in a couple business days. The reviewer will then look at your PR and respond.

Occasionally, things might fall through the cracks (sorry!). In case your PR either doesn't get assigned to a reviewer or doesn't get any response from the reviewer for 4 business days, please leave comment on the PR (mentioning the reviewer if one has been assigned). That'll get it nudged back onto people's radar.

If that still doesn't help, come see us during our office hours

Once your PR is approved, you can merge it in by entering a comment with the content @pytorchmergebot merge (what's this bot?)

So you want to write some documentation and don't know where to start? PyTorch has two main types of documentation:

• User facing documentation: These are the docs that you see over at our docs website.
• Developer facing documentation: Developer facing documentation is spread around our READMEs in our codebase and in the PyTorch Developer Wiki. If you're interested in adding new developer docs, please read this page on the wiki on our best practices for where to put it.

The rest of this section is about user-facing documentation.

PyTorch uses Google style for formatting docstrings. Each line inside a docstrings block must be limited to 80 characters so that it fits into Jupyter documentation popups.

In addition to the standard Google Style docstring formatting rules, the following guidelines should be followed for docstring types (docstring types are the type information contained in the round brackets after the variable name):

• The "Callable", "Any", "Iterable", "Iterator", "Generator" types should have their first letter capitalized.

• The "list" and "tuple" types should be completely lowercase.

• Types should not be made plural. For example: tuple of int should be used instead of tuple of ints.

• The only acceptable delimiter words for types are or and of. No other non-type words should be used other than optional.

• The word optional should only be used after the types, and it is only used if the user does not have to specify a value for the variable. Default values are listed after the variable description. Example:

  my_var (int, optional): Variable description. Default: 1
  

• Basic Python types should match their type name so that the Intersphinx extension can correctly identify them. For example:

  • Use str instead of string.
  • Use bool instead of boolean.
  • Use dict instead of dictionary.

• Square brackets should be used for the dictionary type. For example:

  my_var (dict[str, int]): Variable description.
  

• If a variable has two different possible types, then the word or should be used without a comma. Otherwise variables with 3 or more types should use commas to separate the types. Example:

  x (type1 or type2): Variable description.
  y (type1, type2, or type3): Variable description.
  

To build the documentation:

1. Build and install PyTorch

2. Install the prerequisites

cd docs
pip install -r requirements.txt
# `katex` must also be available in your PATH.
# You can either install katex globally if you have properly configured npm:
# npm install -g katex
# Or if you prefer an uncontaminated global executable environment or do not want to go through the node configuration:
# npm install katex && export PATH="$PATH:$(pwd)/node_modules/.bin"

Note: if you installed nodejs with a different package manager (e.g., conda) then npm will probably install a version of katex that is not compatible with your version of nodejs and doc builds will fail. A combination of versions that is known to work is node@6.13.1 and katex@0.13.18. To install the latter with npm you can run npm install -g katex@0.13.18

Note that if you are a Facebook employee using a devserver, yarn may be more convenient to install katex:

yarn global add katex

If a specific version is required you can use for example yarn global add katex@0.13.18.

3. Generate the documentation HTML files. The generated files will be in docs/build/html.

make html

The .rst source files live in docs/source. Some of the .rst files pull in docstrings from PyTorch Python code (for example, via the autofunction or autoclass directives). To vastly shorten doc build times, it is helpful to remove the files you are not working on, only keeping the base index.rst file and the files you are editing. The Sphinx build will produce missing file warnings but will still complete. For example, to work on jit.rst:

cd docs/source
find . -type f | grep rst | grep -v index | grep -v jit | xargs rm

# Make your changes, build the docs, etc.

# Don't commit the deletions!
git add index.rst jit.rst
...

For C++ documentation (https://pytorch.org/cppdocs), we use Doxygen and then convert it to Sphinx via Breathe and Exhale. Check the Doxygen reference for more information on the documentation syntax.

We run Doxygen in CI (Travis) to verify that you do not use invalid Doxygen commands. To run this check locally, run ./check-doxygen.sh from inside docs/cpp/source.

To build the documentation, follow the same steps as above, but run them from docs/cpp instead of docs.

To view HTML files locally, you can open the files in your web browser. For example, navigate to file:///your_pytorch_folder/docs/build/html/index.html in a web browser.

If you are developing on a remote machine, you can set up an SSH tunnel so that you can access the HTTP server on the remote machine from your local machine. To map remote port 8000 to local port 8000, use either of the following commands.

# For SSH
ssh my_machine -L 8000:my_machine:8000

# For Eternal Terminal
et my_machine -t="8000:8000"

Then navigate to localhost:8000 in your web browser.

Tip: You can start a lightweight HTTP server on the remote machine with:

python -m http.server 8000 <path_to_html_output>

Alternatively, you can run rsync on your local machine to copy the files from your remote machine:

mkdir -p build cpp/build
rsync -az me@my_machine:/path/to/pytorch/docs/build/html build
rsync -az me@my_machine:/path/to/pytorch/docs/cpp/build/html cpp/build

PyTorch will host documentation previews at https://docs-preview.pytorch.org/pytorch/pytorch/<pr number>/index.html once the pytorch_python_doc_build GitHub Actions job has completed on your PR. You can visit that page directly or find its link in the automated Dr. CI comment on your PR.

It is easy for code snippets in docstrings and .rst files to get out of date. The docs build includes the Sphinx Doctest Extension, which can run code in documentation as a unit test. To use the extension, use the .. testcode:: directive in your .rst and docstrings.

To manually run these tests, follow steps 1 and 2 above, then run:

cd docs
make doctest

Evaluating the performance impact of code changes in PyTorch can be complicated, particularly if code changes happen in compiled code. One simple way to profile both Python and C++ code in PyTorch is to use py-spy, a sampling profiler for Python that has the ability to profile native code and Python code in the same session.

py-spy can be installed via pip:

pip install py-spy

To use py-spy, first write a Python test script that exercises the functionality you would like to profile. For example, this script profiles torch.add:

import torch

t1 = torch.tensor([[1, 1], [1, 1.]])
t2 = torch.tensor([[0, 0], [0, 0.]])

for _ in range(1000000):
    torch.add(t1, t2)

Since the torch.add operation happens in microseconds, we repeat it a large number of times to get good statistics. The most straightforward way to use py-spy with such a script is to generate a flame graph:

py-spy record -o profile.svg --native -- python test_tensor_tensor_add.py

This will output a file named profile.svg containing a flame graph you can view in a web browser or SVG viewer. Individual stack frame entries in the graph can be selected interactively with your mouse to zoom in on a particular part of the program execution timeline. The --native command-line option tells py-spy to record stack frame entries for PyTorch C++ code. To get line numbers for C++ code it may be necessary to compile PyTorch in debug mode by prepending your setup.py develop call to compile PyTorch with DEBUG=1. Depending on your operating system it may also be necessary to run py-spy with root privileges.

py-spy can also work in an htop-like "live profiling" mode and can be tweaked to adjust the stack sampling rate, see the py-spy readme for more details.

One downside to using python setup.py develop is that your development version of PyTorch will be installed globally on your account (e.g., if you run import torch anywhere else, the development version will be used.

If you want to manage multiple builds of PyTorch, you can make use of conda environments to maintain separate Python package environments, each of which can be tied to a specific build of PyTorch. To set one up:

conda create -n pytorch-myfeature
source activate pytorch-myfeature
# if you run python now, torch will NOT be installed
python setup.py develop

If you are working on the C++ code, there are a few important things that you will want to keep in mind:

1. How to rebuild only the code you are working on.
2. How to make rebuilds in the absence of changes go faster.

python setup.py build will build everything by default, but sometimes you are only interested in a specific component.

• Working on a test binary? Run (cd build && ninja bin/test_binary_name) to rebuild only that test binary (without rerunning cmake). (Replace ninja with make if you don't have ninja installed).

On the initial build, you can also speed things up with the environment variables DEBUG, USE_DISTRIBUTED, USE_MKLDNN, USE_CUDA, USE_FLASH_ATTENTION, USE_MEM_EFF_ATTENTION, BUILD_TEST, USE_FBGEMM, USE_NNPACK and USE_QNNPACK.

• DEBUG=1 will enable debug builds (-g -O0)
• REL_WITH_DEB_INFO=1 will enable debug symbols with optimizations (-g -O3)
• USE_DISTRIBUTED=0 will disable distributed (c10d, gloo, mpi, etc.) build.
• USE_MKLDNN=0 will disable using MKL-DNN.
• USE_CUDA=0 will disable compiling CUDA (in case you are developing on something not CUDA related), to save compile time.
• BUILD_TEST=0 will disable building C++ test binaries.
• USE_FBGEMM=0 will disable using FBGEMM (quantized 8-bit server operators).
• USE_NNPACK=0 will disable compiling with NNPACK.
• USE_QNNPACK=0 will disable QNNPACK build (quantized 8-bit operators).
• USE_XNNPACK=0 will disable compiling with XNNPACK.
• USE_FLASH_ATTENTION=0 and USE_MEM_EFF_ATTENTION=0 will disable compiling flash attention and memory efficient kernels respectively

For example:

DEBUG=1 USE_DISTRIBUTED=0 USE_MKLDNN=0 USE_CUDA=0 BUILD_TEST=0 USE_FBGEMM=0 USE_NNPACK=0 USE_QNNPACK=0 USE_XNNPACK=0 python setup.py develop

For subsequent builds (i.e., when build/CMakeCache.txt exists), the build options passed for the first time will persist; please run ccmake build/, run cmake-gui build/, or directly edit build/CMakeCache.txt to adapt build options.

When using python setup.py develop, PyTorch will generate a compile_commands.json file that can be used by many editors to provide command completion and error highlighting for PyTorch's C++ code. You need to pip install ninja to generate accurate information for the code in torch/csrc. More information at:

• https://sarcasm.github.io/notes/dev/compilation-database.html

By default, cmake will use its Makefile generator to generate your build system. You can get faster builds if you install the ninja build system with pip install ninja. If PyTorch was already built, you will need to run python setup.py clean once after installing ninja for builds to succeed.

Note: Make sure to use a machine with a larger number of CPU cores, this will significantly reduce your build times.

Even when dependencies are tracked with file modification, there are many situations where files get rebuilt when a previous compilation was exactly the same. Using ccache in a situation like this is a real time-saver.

Before building pytorch, install ccache from your package manager of choice:

conda install ccache -c conda-forge
sudo apt install ccache
sudo yum install ccache
brew install ccache

You may also find the default cache size in ccache is too small to be useful. The cache sizes can be increased from the command line:

# config: cache dir is ~/.ccache, conf file ~/.ccache/ccache.conf
# max size of cache
ccache -M 25Gi  # -M 0 for unlimited
# unlimited number of files
ccache -F 0

To check this is working, do two clean builds of pytorch in a row. The second build should be substantially and noticeably faster than the first build. If this doesn't seem to be the case, check the CMAKE_<LANG>_COMPILER_LAUNCHER rules in build/CMakeCache.txt, where <LANG> is C, CXX and CUDA. Each of these 3 variables should contain ccache, e.g.

//CXX compiler launcher
CMAKE_CXX_COMPILER_LAUNCHER:STRING=/usr/bin/ccache

If not, you can define these variables on the command line before invoking setup.py.

export CMAKE_C_COMPILER_LAUNCHER=ccache
export CMAKE_CXX_COMPILER_LAUNCHER=ccache
export CMAKE_CUDA_COMPILER_LAUNCHER=ccache
python setup.py develop

If you are editing a single file and rebuilding in a tight loop, the time spent linking will dominate. The system linker available in most Linux distributions (GNU ld) is quite slow. Use a faster linker, like lld.

People on Mac, follow this guide instead.

The easiest way to use lld this is download the latest LLVM binaries and run:

ln -s /path/to/downloaded/ld.lld /usr/local/bin/ld

Sometimes there's no way of getting around rebuilding lots of files, for example editing native_functions.yaml usually means 1000+ files being rebuilt. If you're using CMake newer than 3.16, you can enable pre-compiled headers by setting USE_PRECOMPILED_HEADERS=1 either on first setup, or in the CMakeCache.txt file.

USE_PRECOMPILED_HEADERS=1 python setup.py develop

This adds a build step where the compiler takes <ATen/ATen.h> and essentially dumps its internal AST to a file so the compiler can avoid repeating itself for every .cpp file.

One caveat is that when enabled, this header gets included in every file by default. Which may change what code is legal, for example:

• internal functions can never alias existing names in <ATen/ATen.h>
• names in <ATen/ATen.h> will work even if you don't explicitly include it.

If re-building without modifying any files results in several CUDA files being re-compiled, you may be running into an nvcc bug where header dependencies are not converted to absolute paths before reporting it to the build system. This makes ninja think one of the header files has been deleted, so it runs the build again.

A compiler-wrapper to fix this is provided in tools/nvcc_fix_deps.py. You can use this as a compiler launcher, similar to ccache

export CMAKE_CUDA_COMPILER_LAUNCHER="python;`pwd`/tools/nvcc_fix_deps.py;ccache"
python setup.py develop

While debugging a problem one often had to maintain a debug build in a separate folder. But often only a few files needs to be rebuild with debug info to get a symbolicated backtrace or enable source debugging One can easily solve this with the help of tools/build_with_debinfo.py

For example, suppose one wants to debug what is going on while tensor index is selected, which can be achieved by setting a breakpoint at applySelect function:

% lldb -o "b applySelect" -o "process launch" -- python3 -c "import torch;print(torch.rand(5)[3])"
(lldb) target create "python"
Current executable set to '/usr/bin/python3' (arm64).
(lldb) settings set -- target.run-args  "-c" "import torch;print(torch.rand(5)[3])"
(lldb) b applySelect
Breakpoint 1: no locations (pending).
WARNING:  Unable to resolve breakpoint to any actual locations.
(lldb) process launch
2 locations added to breakpoint 1
Process 87729 stopped
* thread #1, queue = 'com.apple.main-thread', stop reason = breakpoint 1.1
    frame #0: 0x00000001023d55a8 libtorch_python.dylib`at::indexing::impl::applySelect(at::Tensor const&, long long, c10::SymInt, long long, c10::Device const&, std::__1::optional<c10::ArrayRef<c10::SymInt>> const&)
libtorch_python.dylib`at::indexing::impl::applySelect:
->  0x1023d55a8 <+0>:  sub    sp, sp, #0xd0
    0x1023d55ac <+4>:  stp    x24, x23, [sp, #0x90]
    0x1023d55b0 <+8>:  stp    x22, x21, [sp, #0xa0]
    0x1023d55b4 <+12>: stp    x20, x19, [sp, #0xb0]
Target 0: (python) stopped.
Process 87729 launched: '/usr/bin/python' (arm64)

Which is not very informative, but can be easily remedied by rebuilding python_variable_indexing.cpp with debug information

% ./tools/build_with_debinfo.py torch/csrc/autograd/python_variable_indexing.cpp
[1 / 2] Building caffe2/torch/CMakeFiles/torch_python.dir/csrc/autograd/python_variable_indexing.cpp.o
[2 / 2] Building lib/libtorch_python.dylib

And afterwards:

% lldb -o "b applySelect" -o "process launch" -- python3 -c "import torch;print(torch.rand(5)[3])"
(lldb) target create "python"
Current executable set to '/usr/bin/python3' (arm64).
(lldb) settings set -- target.run-args  "-c" "import torch;print(torch.rand(5)[3])"
(lldb) b applySelect
Breakpoint 1: no locations (pending).
WARNING:  Unable to resolve breakpoint to any actual locations.
(lldb) process launch
2 locations added to breakpoint 1
Process 87741 stopped
* thread #1, queue = 'com.apple.main-thread', stop reason = breakpoint 1.1
    frame #0: 0x00000001024e2628 libtorch_python.dylib`at::indexing::impl::applySelect(self=0x00000001004ee8a8, dim=0, index=(data_ = 3), real_dim=0, (null)=0x000000016fdfe535, self_sizes= Has Value=true ) at TensorIndexing.h:239:7
   236         const at::Device& /*self_device*/,
   237         const c10::optional<SymIntArrayRef>& self_sizes) {
   238       // See NOTE [nested tensor size for indexing]
-> 239       if (self_sizes.has_value()) {
   240         auto maybe_index = index.maybe_as_int();
   241         if (maybe_index.has_value()) {
   242           TORCH_CHECK_INDEX(
Target 0: (python) stopped.
Process 87741 launched: '/usr/bin/python3' (arm64)

Which is much more useful, isn't it?

We have very extensive tests in the test/cpp/api folder. The tests are a great way to see how certain components are intended to be used. When compiling PyTorch from source, the test runner binary will be written to build/bin/test_api. The tests use the GoogleTest framework, which you can read up about to learn how to configure the test runner. When submitting a new feature, we care very much that you write appropriate tests. Please follow the lead of the other tests to see how to write a new test case.

If you are debugging pytorch inside GDB, you might be interested in pytorch-gdb. This script introduces some pytorch-specific commands which you can use from the GDB prompt. In particular, torch-tensor-repr prints a human-readable repr of an at::Tensor object. Example of usage:

$ gdb python
GNU gdb (GDB) 9.2
[...]
(gdb) # insert a breakpoint when we call .neg()
(gdb) break at::Tensor::neg
Function "at::Tensor::neg" not defined.
Make breakpoint pending on future shared library load? (y or [n]) y
Breakpoint 1 (at::Tensor::neg) pending.

(gdb) run
[...]
>>> import torch
>>> t = torch.tensor([1, 2, 3, 4], dtype=torch.float64)
>>> t
tensor([1., 2., 3., 4.], dtype=torch.float64)
>>> t.neg()

Thread 1 "python" hit Breakpoint 1, at::Tensor::neg (this=0x7ffb118a9c88) at aten/src/ATen/core/TensorBody.h:3295
3295    inline at::Tensor Tensor::neg() const {
(gdb) # the default repr of 'this' is not very useful
(gdb) p this
$1 = (const at::Tensor * const) 0x7ffb118a9c88
(gdb) p *this
$2 = {impl_ = {target_ = 0x55629b5cd330}}
(gdb) torch-tensor-repr *this
Python-level repr of *this:
tensor([1., 2., 3., 4.], dtype=torch.float64)

GDB tries to automatically load pytorch-gdb thanks to the .gdbinit at the root of the pytorch repo. However, auto-loadings is disabled by default, because of security reasons:

$ gdb
warning: File "/path/to/pytorch/.gdbinit" auto-loading has been declined by your `auto-load safe-path' set to "$debugdir:$datadir/auto-load".
To enable execution of this file add
        add-auto-load-safe-path /path/to/pytorch/.gdbinit
line to your configuration file "/home/YOUR-USERNAME/.gdbinit".
To completely disable this security protection add
        set auto-load safe-path /
line to your configuration file "/home/YOUR-USERNAME/.gdbinit".
For more information about this security protection see the
"Auto-loading safe path" section in the GDB manual.  E.g., run from the shell:
        info "(gdb)Auto-loading safe path"
(gdb)

As gdb itself suggests, the best way to enable auto-loading of pytorch-gdb is to add the following line to your ~/.gdbinit (i.e., the .gdbinit file which is in your home directory, not /path/to/pytorch/.gdbinit):

add-auto-load-safe-path /path/to/pytorch/.gdbinit

Set TORCH_SHOW_CPP_STACKTRACES=1 to get the C++ stacktrace when an error occurs in Python.

If you are working on the CUDA code, here are some useful CUDA debugging tips:

1. CUDA_DEVICE_DEBUG=1 will enable CUDA device function debug symbols (-g -G). This will be particularly helpful in debugging device code. However, it will slow down the build process for about 50% (compared to only DEBUG=1), so use wisely.
2. cuda-gdb and cuda-memcheck are your best CUDA debugging friends. Unlikegdb, cuda-gdb can display actual values in a CUDA tensor (rather than all zeros).
3. CUDA supports a lot of C++11/14 features such as, std::numeric_limits, std::nextafter, std::tuple etc. in device code. Many of such features are possible because of the --expt-relaxed-constexpr nvcc flag. There is a known issue that ROCm errors out on device code, which uses such stl functions.
4. A good performance metric for a CUDA kernel is the Effective Memory Bandwidth. It is useful for you to measure this metric whenever you are writing/optimizing a CUDA kernel. Following script shows how we can measure the effective bandwidth of CUDA uniform_ kernel.

   import torch
   from torch.utils.benchmark import Timer
   size = 128*512
   nrep = 100
   nbytes_read_write = 4 # this is number of bytes read + written by a kernel. Change this to fit your kernel.
   
   for i in range(10):
       a=torch.empty(size).cuda().uniform_()
       torch.cuda.synchronize()
       out = a.uniform_()
       torch.cuda.synchronize()
       t = Timer(stmt="a.uniform_()", globals=globals())
       res = t.blocked_autorange()
       timec = res.median
       print("uniform, size, elements", size, "forward", timec, "bandwidth (GB/s)", size*(nbytes_read_write)*1e-9/timec)
       size *=2

See more cuda development tips here

For building from source on Windows, consult our documentation on it.

Occasionally, you will write a patch which works on Linux, but fails CI on Windows. There are a few aspects in which MSVC (the Windows compiler toolchain we use) is stricter than Linux, which are worth keeping in mind when fixing these problems.

1. Symbols are NOT exported by default on Windows; instead, you have to explicitly mark a symbol as exported/imported in a header file with __declspec(dllexport) / __declspec(dllimport). We have codified this pattern into a set of macros which follow the convention *_API, e.g., TORCH_API inside Caffe2, Aten and Torch. (Every separate shared library needs a unique macro name, because symbol visibility is on a per shared library basis. See c10/macros/Macros.h for more details.)

   The upshot is if you see an "unresolved external" error in your Windows build, this is probably because you forgot to mark a function with *_API. However, there is one important counterexample to this principle: if you want a templated function to be instantiated at the call site, do NOT mark it with *_API (if you do mark it, you'll have to explicitly instantiate all of the specializations used by the call sites.)

2. If you link against a library, this does not make its dependencies transitively visible. You must explicitly specify a link dependency against every library whose symbols you use. (This is different from Linux where in most environments, transitive dependencies can be used to fulfill unresolved symbols.)

3. If you have a Windows box (we have a few on EC2 which you can request access to) and you want to run the build, the easiest way is to just run .ci/pytorch/win-build.sh. If you need to rebuild, run REBUILD=1 .ci/pytorch/win-build.sh (this will avoid blowing away your Conda environment.)

Even if you don't know anything about MSVC, you can use cmake to build simple programs on Windows; this can be helpful if you want to learn more about some peculiar linking behavior by reproducing it on a small example. Here's a simple example cmake file that defines two dynamic libraries, one linking with the other:

project(myproject CXX)
set(CMAKE_CXX_STANDARD 14)
add_library(foo SHARED foo.cpp)
add_library(bar SHARED bar.cpp)
# NB: don't forget to __declspec(dllexport) at least one symbol from foo,
# otherwise foo.lib will not be created.
target_link_libraries(bar PUBLIC foo)

You can build it with:

mkdir build
cd build
cmake ..
cmake --build .

The PyTorch codebase sometimes likes to use exciting C++ features, and these exciting features lead to exciting bugs in Windows compilers. To add insult to injury, the error messages will often not tell you which line of code actually induced the erroring template instantiation.

We've found the most effective way to debug these problems is to carefully read over diffs, keeping in mind known bugs in MSVC/NVCC. Here are a few well known pitfalls and workarounds:

• This is not actually a bug per se, but in general, code generated by MSVC is more sensitive to memory errors; you may have written some code that does a use-after-free or stack overflows; on Linux the code might work, but on Windows your program will crash. ASAN may not catch all of these problems: stay vigilant to the possibility that your crash is due to a real memory problem.

• (NVCC) c10::optional does not work when used from device code. Don't use it from kernels. Upstream issue: akrzemi1/Optional#58 and our local issue #10329.

• constexpr generally works less well on MSVC.

  • The idiom static_assert(f() == f()) to test if f is constexpr does not work; you'll get "error C2131: expression did not evaluate to a constant". Don't use these asserts on Windows. (Example: c10/util/intrusive_ptr.h)

• (NVCC) Code you access inside a static_assert will eagerly be evaluated as if it were device code, and so you might get an error that the code is "not accessible".

class A {
  static A singleton_;
  static constexpr inline A* singleton() {
    return &singleton_;
  }
};
static_assert(std::is_same(A*, decltype(A::singleton()))::value, "hmm");

• The compiler will run out of heap space if you attempt to compile files that are too large. Splitting such files into separate files helps. (Example: THTensorMath, THTensorMoreMath, THTensorEvenMoreMath.)

• MSVC's preprocessor (but not the standard compiler) has a bug where it incorrectly tokenizes raw string literals, ending when it sees a ". This causes preprocessor tokens inside the literal like an#endif to be incorrectly treated as preprocessor directives. See https://godbolt.org/z/eVTIJq as an example.

• Either MSVC or the Windows headers have a PURE macro defined and will replace any occurrences of the PURE token in code with an empty string. This is why we have AliasAnalysisKind::PURE_FUNCTION and not AliasAnalysisKind::PURE. The same is likely true for other identifiers that we just didn't try to use yet.

CUDA, MSVC, and PyTorch versions are interdependent; please install matching versions from this table:

Note: There's a compilation issue in several Visual Studio 2019 versions since 16.7.1, so please make sure your Visual Studio 2019 version is not in 16.7.1 ~ 16.7.5

We use clang-tidy to perform additional formatting and semantic checking of code. We provide a pre-commit git hook for performing these checks, before a commit is created:

ln -s ../../tools/git-pre-commit .git/hooks/pre-commit

If you have already committed files and CI reports flake8 errors, you can run the check locally in your PR branch with:

flake8 $(git diff --name-only $(git merge-base --fork-point main))

You'll need to install an appropriately configured flake8; see Lint as you type for documentation on how to do this.

Fix the code so that no errors are reported when you re-run the above check again, and then commit the fix.

ASAN is very useful for debugging memory errors in C++. We run it in CI, but here's how to get the same thing to run on your local machine.

First, install LLVM 8. The easiest way is to get prebuilt binaries and extract them to folder (later called $LLVM_ROOT).

Then set up the appropriate scripts. You can put this in your .bashrc:

LLVM_ROOT=<wherever your llvm install is>
PYTORCH_ROOT=<wherever your pytorch checkout is>

LIBASAN_RT="$LLVM_ROOT/lib/clang/8.0.0/lib/linux/libclang_rt.asan-x86_64.so"
build_with_asan()
{
  LD_PRELOAD=${LIBASAN_RT} \
  CC="$LLVM_ROOT/bin/clang" \
  CXX="$LLVM_ROOT/bin/clang++" \
  LDSHARED="clang --shared" \
  LDFLAGS="-stdlib=libstdc++" \
  CFLAGS="-fsanitize=address -fno-sanitize-recover=all -shared-libasan -pthread" \
  CXX_FLAGS="-pthread" \
  USE_CUDA=0 USE_OPENMP=0 USE_DISTRIBUTED=0 DEBUG=1 \
  python setup.py develop
}

run_with_asan()
{
  LD_PRELOAD=${LIBASAN_RT} $@
}

# you can look at build-asan.sh to find the latest options the CI uses
export ASAN_OPTIONS=detect_leaks=0:symbolize=1:strict_init_order=true
export UBSAN_OPTIONS=print_stacktrace=1:suppressions=$PYTORCH_ROOT/ubsan.supp
export ASAN_SYMBOLIZER_PATH=$LLVM_ROOT/bin/llvm-symbolizer

Then you can use the scripts like:

suo-devfair ~/pytorch ❯ build_with_asan
suo-devfair ~/pytorch ❯ run_with_asan python test/test_jit.py

The scripts above specify the clang and clang++ binaries directly, which bypasses ccache. Here's how to get ccache to work:

1. Make sure the ccache symlinks for clang and clang++ are set up (see CONTRIBUTING.md)
2. Make sure $LLVM_ROOT/bin is available on your $PATH.
3. Change the CC and CXX variables in build_with_asan() to point directly to clang and clang++.

The “standard” workflow for ASAN assumes you have a standalone binary:

1. Recompile your binary with -fsanitize=address.
2. Run the binary, and ASAN will report whatever errors it find.

Unfortunately, PyTorch is a distributed as a shared library that is loaded by a third-party executable (Python). It’s too much of a hassle to recompile all of Python every time we want to use ASAN. Luckily, the ASAN folks have a workaround for cases like this:

1. Recompile your library with -fsanitize=address -shared-libasan. The extra -shared-libasan tells the compiler to ask for the shared ASAN runtime library.
2. Use LD_PRELOAD to tell the dynamic linker to load the ASAN runtime library before anything else.

More information can be found here.

We need LD_PRELOAD because there is a cmake check that ensures that a simple program builds and runs. If we are building with ASAN as a shared library, we need to LD_PRELOAD the runtime library, otherwise there will dynamic linker errors and the check will fail.

We don’t actually need either of these if we fix the cmake checks.

Python leaks a lot of memory. Possibly we could configure a suppression file, but we haven’t gotten around to it.

In 2018, we merged Caffe2 into the PyTorch source repository. While the steady state aspiration is that Caffe2 and PyTorch share code freely, in the meantime there will be some separation.

There are a few "unusual" directories which, for historical reasons, are Caffe2/PyTorch specific. Here they are:

• CMakeLists.txt, Makefile, binaries, cmake, conda, modules, scripts are Caffe2-specific. Don't put PyTorch code in them without extra coordination.

• mypy*, requirements.txt, setup.py, test, tools are PyTorch-specific. Don't put Caffe2 code in them without extra coordination.

Once you submit a PR or push a new commit to a branch that is in an active PR, CI jobs will be run automatically. Some of these may fail and you will need to find out why, by looking at the logs.

Fairly often, a CI failure might be unrelated to your changes. You can confirm by going to our HUD and seeing if the CI job is failing upstream already. In this case, you can usually ignore the failure. See the following subsection for more details.

Some failures might be related to specific hardware or environment configurations. In this case, if you're a Meta employee, you can ssh into the job's session to perform manual debugging following the instructions in our CI wiki.

For CI run on main, this repository is checked out for a given main commit, and CI is run on that commit (there isn't really any other choice).

For PRs, however, it's a bit more complicated. Consider this commit graph, where main is at commit A, and the branch for PR #42 (just a placeholder) is at commit B:

       o---o---B (refs/pull/42/head)
      /         \
     /           C (refs/pull/42/merge)
    /           /
---o---o---o---A (merge-destination) - usually main

There are two possible choices for which commit to use:

1. Checkout commit B, the head of the PR (manually committed by the PR author).
2. Checkout commit C, the hypothetical result of what would happen if the PR were merged into its destination (usually main).

For all practical purposes, most people can think of the commit being used as commit B (choice 1).

However, if workflow files (which govern CI behavior) were modified (either by your PR or since dev branch were created ) there's a nuance to know about: The workflow files themselves get taken from checkpoint C, the merger of your PR and the main branch. But only the workflow files get taken from that merged checkpoint. Everything else (tests, code, etc) all get taken directly from your PR's commit (commit B). Please note, this scenario would never affect PRs authored by ghstack as they would not automatically ingest the updates from default branch.

Dev Infra Office Hours are hosted every Friday to answer any questions regarding developer experience, Green HUD, and CI.