Reland of #153153, which was incidentally closed.
Update the minimum CMake version to 3.27 because of it provides more CUDA targets such as CUDA::nvperf_host so that it is possible to remove some of our forked CUDA modules. See https://github.com/pytorch/pytorch/pull/153783.
It's also possible to facilitate future third-party updates such as FBGEMM (its current shipped version requires 3.21).
Pull Request resolved: https://github.com/pytorch/pytorch/pull/154783
Approved by: https://github.com/ezyang
### Changes
- Detect NVSHMEM install location via `sysconfig.get_path("purelib")`, which typically resolves to `<conda_env>/lib/python/site-packages`, and NVSHMEM include and lib live under `nvidia/nvshmem`
- Added link dir via `target_link_directories`
- Removed direct dependency on mlx5
- Added preload rule (following other other NVIDIA libs)
### Plan of Record
1. End user experience: link against NVSHMEM dynamically (NVSHMEM lib size is 100M, similar to NCCL, thus we'd like users to `pip install nvshmem` than torch carrying the bits)
2. Developer experience: at compile time, prefers wheel dependency than using Git submodule
General rule: submodule for small lib that torch can statically link with
If user pip install a lib, our CI build process should do the same, rather than building from Git submodule (just for its header, for example)
3. Keep `USE_NVSHMEM` to gate non-Linux platforms, like Windows, Mac
4. At configuration time, we should be able to detect whether nvshmem is available, if not, we don't build `NVSHMEMSymmetricMemory` at all.
For now, we have symbol dependency on two particular libs from NVSHMEM:
- libnvshmem_host.so: contains host side APIs;
- libnvshmem_device.a: contains device-side global variables AND device function impls.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/153010
Approved by: https://github.com/ngimel, https://github.com/fduwjj, https://github.com/Skylion007
Changes by apply order:
1. Replace all `".."` and `os.pardir` usage with `os.path.dirname(...)`.
2. Replace nested `os.path.dirname(os.path.dirname(...))` call with `str(Path(...).parent.parent)`.
3. Reorder `.absolute()` ~/ `.resolve()`~ and `.parent`: always resolve the path first.
`.parent{...}.absolute()` -> `.absolute().parent{...}`
4. Replace chained `.parent x N` with `.parents[${N - 1}]`: the code is easier to read (see 5.)
`.parent.parent.parent.parent` -> `.parents[3]`
5. ~Replace `.parents[${N - 1}]` with `.parents[${N} - 1]`: the code is easier to read and does not introduce any runtime overhead.~
~`.parents[3]` -> `.parents[4 - 1]`~
6. ~Replace `.parents[2 - 1]` with `.parent.parent`: because the code is shorter and easier to read.~
Pull Request resolved: https://github.com/pytorch/pytorch/pull/129374
Approved by: https://github.com/justinchuby, https://github.com/malfet
Changes by apply order:
1. Replace all `".."` and `os.pardir` usage with `os.path.dirname(...)`.
2. Replace nested `os.path.dirname(os.path.dirname(...))` call with `str(Path(...).parent.parent)`.
3. Reorder `.absolute()` ~/ `.resolve()`~ and `.parent`: always resolve the path first.
`.parent{...}.absolute()` -> `.absolute().parent{...}`
4. Replace chained `.parent x N` with `.parents[${N - 1}]`: the code is easier to read (see 5.)
`.parent.parent.parent.parent` -> `.parents[3]`
5. ~Replace `.parents[${N - 1}]` with `.parents[${N} - 1]`: the code is easier to read and does not introduce any runtime overhead.~
~`.parents[3]` -> `.parents[4 - 1]`~
6. ~Replace `.parents[2 - 1]` with `.parent.parent`: because the code is shorter and easier to read.~
Pull Request resolved: https://github.com/pytorch/pytorch/pull/129374
Approved by: https://github.com/justinchuby, https://github.com/malfet
Changes by apply order:
1. Replace all `".."` and `os.pardir` usage with `os.path.dirname(...)`.
2. Replace nested `os.path.dirname(os.path.dirname(...))` call with `str(Path(...).parent.parent)`.
3. Reorder `.absolute()` ~/ `.resolve()`~ and `.parent`: always resolve the path first.
`.parent{...}.absolute()` -> `.absolute().parent{...}`
4. Replace chained `.parent x N` with `.parents[${N - 1}]`: the code is easier to read (see 5.)
`.parent.parent.parent.parent` -> `.parents[3]`
5. ~Replace `.parents[${N - 1}]` with `.parents[${N} - 1]`: the code is easier to read and does not introduce any runtime overhead.~
~`.parents[3]` -> `.parents[4 - 1]`~
6. ~Replace `.parents[2 - 1]` with `.parent.parent`: because the code is shorter and easier to read.~
Pull Request resolved: https://github.com/pytorch/pytorch/pull/129374
Approved by: https://github.com/justinchuby, https://github.com/malfet
Changes by apply order:
1. Replace all `".."` and `os.pardir` usage with `os.path.dirname(...)`.
2. Replace nested `os.path.dirname(os.path.dirname(...))` call with `str(Path(...).parent.parent)`.
3. Reorder `.absolute()` ~/ `.resolve()`~ and `.parent`: always resolve the path first.
`.parent{...}.absolute()` -> `.absolute().parent{...}`
4. Replace chained `.parent x N` with `.parents[${N - 1}]`: the code is easier to read (see 5.)
`.parent.parent.parent.parent` -> `.parents[3]`
5. ~Replace `.parents[${N - 1}]` with `.parents[${N} - 1]`: the code is easier to read and does not introduce any runtime overhead.~
~`.parents[3]` -> `.parents[4 - 1]`~
6. ~Replace `.parents[2 - 1]` with `.parent.parent`: because the code is shorter and easier to read.~
Pull Request resolved: https://github.com/pytorch/pytorch/pull/129374
Approved by: https://github.com/justinchuby, https://github.com/malfet
As FindPythonInterp and FindPythonLibs has been deprecated since cmake-3.12
Replace `PYTHON_EXECUTABLE` with `Python_EXECUTABLE` everywhere (CMake variable names are case-sensitive)
This makes PyTorch buildable with python3 binary shipped with XCode on MacOS
TODO: Get rid of `FindNumpy` as its part of Python package
Pull Request resolved: https://github.com/pytorch/pytorch/pull/124613
Approved by: https://github.com/cyyever, https://github.com/Skylion007
The warning complains that `TORCH_CUDA_ARCH_LIST` is set on the environment
instead of being defined as a build variable, which is fixed by the change to
`tools/setup_helpers/cmake.py`.
However, I still see the warning even with this fix because
```cmake
if((NOT EXISTS ${TORCH_CUDA_ARCH_LIST}) ...
```
is actually checking whether a file exists called "7.5" (or whatever arch is
being requested). Instead we want to check if the variable is defined.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/104680
Approved by: https://github.com/albanD
If `CMAKE_GENERATOR=Visual Studio 16 2019` then the build will fail if `USE_NINJA=False` not set.
This PR changes that if CMAKE_GENERATOR is set an not equal to ninja then it won't use Ninja.
This is just for easier setting another generator.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/98605
Approved by: https://github.com/kit1980
Summary: Currently, the model tracer build is broken because of 2 reasons:
1. A few source files are missing, resulting in missing link time symbols
2. The `TRACING_BASED` flag isn't passed correctly from the command line (specified as an evnironment variable) as a CMake flag
Both these issues were fixed.
Test Plan: Ran this command: `USE_CUDA=0 TRACING_BASED=1 python setup.py develop --cmake`
and saw that the tracer binary was built at `build/bin/model_tracer` - also ran it to ensure that it can generate a YAML file.
Differential Revision: [D39391270](https://our.internmc.facebook.com/intern/diff/D39391270)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/84755
Approved by: https://github.com/cccclai
To fix#78540 I committed #78983 which is reverted due to internal CI failure. Then I comitted #79215 which was only fixing the failure but didn't have the full feature of #78983. This PR is another try.
This PR adds script to dump all operators from test models and automatically write into `lightweight_dispatch_ops.yaml`. This way we don't have to manually update the yaml file.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/80791
Approved by: https://github.com/raziel
With ufmt in place https://github.com/pytorch/pytorch/pull/81157, we can now use it to gradually format all files. I'm breaking this down into multiple smaller batches to avoid too many merge conflicts later on.
This batch (as copied from the current BLACK linter config):
* `tools/**/*.py`
Upcoming batchs:
* `torchgen/**/*.py`
* `torch/package/**/*.py`
* `torch/onnx/**/*.py`
* `torch/_refs/**/*.py`
* `torch/_prims/**/*.py`
* `torch/_meta_registrations.py`
* `torch/_decomp/**/*.py`
* `test/onnx/**/*.py`
Once they are all formatted, BLACK linter will be removed.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/81285
Approved by: https://github.com/suo
This PR introduces selective build to lightweight dispatch CI job. By doing so we can't run the `test_lite_intepreter_runtime` test suite anymore because it requires some other operators.
From now on, if we are adding a new unit test in `test_codegen_unboxing`, we will have to export the operators for the unit test model and add them into `lightweight_dispatch_ops.yaml`. This can be automated by introducing tracing based selective build, but that's for next PR to do.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/78983
Approved by: https://github.com/kit1980
Allows to choose the BLAS backend with Eigen. Previously this was a CMake option only and the env variable was ignored.
Related to f1f3c8b0fa
The claimed options BLAS=BLIS WITH_BLAS=blis are misleading: When BLAS=BLIS is set the WITH_BLAS option does not matter at all, it would only matter for BLAS=Eigen hence this issue went undetected so far.
Supersedes #59220
Pull Request resolved: https://github.com/pytorch/pytorch/pull/78037
Approved by: https://github.com/adamjstewart, https://github.com/janeyx99
Summary:
RFC: https://github.com/pytorch/rfcs/pull/40
This PR (re)introduces python codegen for unboxing wrappers. Given an entry of `native_functions.yaml` the codegen should be able to generate the corresponding C++ code to convert ivalues from the stack to their proper types. To trigger the codegen, run
```
tools/jit/gen_unboxing.py -d cg/torch/share/ATen
```
Merged changes on CI test. In https://github.com/pytorch/pytorch/issues/71782 I added an e2e test for static dispatch + codegen unboxing. The test exports a mobile model of mobilenetv2, load and run it on a new binary for lite interpreter: `test/mobile/custom_build/lite_predictor.cpp`.
## Lite predictor build specifics
1. Codegen: `gen.py` generates `RegisterCPU.cpp` and `RegisterSchema.cpp`. Now with this PR, once `static_dispatch` mode is enabled, `gen.py` will not generate `TORCH_LIBRARY` API calls in those cpp files, hence avoids interaction with the dispatcher. Once `USE_LIGHTWEIGHT_DISPATCH` is turned on, `cmake/Codegen.cmake` calls `gen_unboxing.py` which generates `UnboxingFunctions.h`, `UnboxingFunctions_[0-4].cpp` and `RegisterCodegenUnboxedKernels_[0-4].cpp`.
2. Build: `USE_LIGHTWEIGHT_DISPATCH` adds generated sources into `all_cpu_cpp` in `aten/src/ATen/CMakeLists.txt`. All other files remain unchanged. In reality all the `Operators_[0-4].cpp` are not necessary but we can rely on linker to strip them off.
## Current CI job test coverage update
Created a new CI job `linux-xenial-py3-clang5-mobile-lightweight-dispatch-build` that enables the following build options:
* `USE_LIGHTWEIGHT_DISPATCH=1`
* `BUILD_LITE_INTERPRETER=1`
* `STATIC_DISPATCH_BACKEND=CPU`
This job triggers `test/mobile/lightweight_dispatch/build.sh` and builds `libtorch`. Then the script runs C++ tests written in `test_lightweight_dispatch.cpp` and `test_codegen_unboxing.cpp`. Recent commits added tests to cover as many C++ argument type as possible: in `build.sh` we installed PyTorch Python API so that we can export test models in `tests_setup.py`. Then we run C++ test binary to run these models on lightweight dispatch enabled runtime.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/69881
Reviewed By: iseeyuan
Differential Revision: D33692299
Pulled By: larryliu0820
fbshipit-source-id: 211e59f2364100703359b4a3d2ab48ca5155a023
(cherry picked from commit 58e1c9a25e3d1b5b656282cf3ac2f548d98d530b)
This patch enables Pytorch build from source with Ninja and
'Visual Studio 16 2019' CMake generator on Windows on Arm.
Tests:
- Build from source: 'python setup.py develop'.
- Run simple Pytorch example: passed
- python test\test_torch.py:
-- same results as on x64
-- Ran 1344 tests, failures=2
Pull Request resolved: https://github.com/pytorch/pytorch/pull/72424
Summary:
For Pytorch source build when using Ninja generator, it requires **CMake >=3.13**, Pytorch always checks **cmake3 >= 3.10** first, so when **3.13> cmake3 >= 3.10** and then PyTorch will use cmake3, there will report an error: ```Using the Ninja generator requires CMake version 3.13 or greater``` even the **CMake >=3.13** .
For example: for my centos machine, the system CMake3 is ```3.12```, and my conda env's CMake is ```3.19.6```, there will have a build error which PyTorch choose CMake 3, I can update CMake3 or create an alias or a symlink to solve this problem, but the more reasonable way is that ```_get_cmake_command ``` always return the newest CMake executable (unless explicitly overridden with a same CMAKE_PATH environment variable).
Pull Request resolved: https://github.com/pytorch/pytorch/pull/69355
Reviewed By: jbschlosser
Differential Revision: D33062274
Pulled By: malfet
fbshipit-source-id: c6c77ce1374e6090a498be227032af1e1a82d418
Summary:
`_mkdir_p` feels like a remnant of Python-2 era, add `exist_ok` argument and re-raise OSError to make it more human readable.
After the change attempt to build PyTorch in a folder that does not have write permissions will result in:
```
% python3.6 setup.py develop
Building wheel torch-1.10.0a0+git9509e8a
-- Building version 1.10.0a0+git9509e8a
Traceback (most recent call last):
File "/Users/nshulga/git/pytorch-worktree/tools/setup_helpers/cmake.py", line 21, in _mkdir_p
os.makedirs(d, exist_ok=True)
File "/opt/homebrew/Cellar/python36/3.6.2+_254.20170915/Frameworks/Python.framework/Versions/3.6/lib/python3.6/os.py", line 220, in makedirs
mkdir(name, mode)
PermissionError: [Errno 13] Permission denied: 'build'
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File "setup.py", line 895, in <module>
build_deps()
File "setup.py", line 370, in build_deps
cmake=cmake)
File "/Users/nshulga/git/pytorch-worktree/tools/build_pytorch_libs.py", line 63, in build_caffe2
rerun_cmake)
File "/Users/nshulga/git/pytorch-worktree/tools/setup_helpers/cmake.py", line 225, in generate
_mkdir_p(self.build_dir)
File "/Users/nshulga/git/pytorch-worktree/tools/setup_helpers/cmake.py", line 23, in _mkdir_p
raise RuntimeError(f"Failed to create folder {os.path.abspath(d)}: {e.strerror}") from e
RuntimeError: Failed to create folder /Users/nshulga/git/pytorch-worktree/build: Permission denied
```
Fixes https://github.com/pytorch/pytorch/issues/65920
Pull Request resolved: https://github.com/pytorch/pytorch/pull/66492
Reviewed By: seemethere
Differential Revision: D31578820
Pulled By: malfet
fbshipit-source-id: afe8240983100ac0a26cc540376b9dd71b1b53af
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/62445
PyTorch currently uses the old style of compiling CUDA in CMake which is just a
bunch of scripts in `FindCUDA.cmake`. Newer versions support CUDA natively as
a language just like C++ or C.
Test Plan: Imported from OSS
Reviewed By: ejguan
Differential Revision: D31503350
fbshipit-source-id: 2ee817edc9698531ae1b87eda3ad271ee459fd55
