[build] modernize build-frontend: python setup.py develop/install -> [uv ]pip install --no-build-isolation [-e ]. (#156027)

Modernize the development installation:

```bash
# python setup.py develop
python -m pip install --no-build-isolation -e .

# python setup.py install
python -m pip install --no-build-isolation .
```

Now, the `python setup.py develop` is a wrapper around `python -m pip install -e .` since `setuptools>=80.0`:

- pypa/setuptools#4955

`python setup.py install` is deprecated and will emit a warning during run. The warning will become an error on October 31, 2025.

- 9c4d383631/setuptools/command/install.py (L58-L67)

> ```python
> SetuptoolsDeprecationWarning.emit(
>     "setup.py install is deprecated.",
>     """
>     Please avoid running ``setup.py`` directly.
>     Instead, use pypa/build, pypa/installer or other
>     standards-based tools.
>     """,
>     see_url="https://blog.ganssle.io/articles/2021/10/setup-py-deprecated.html",
>     due_date=(2025, 10, 31),
> )
> ```

- pypa/setuptools#3849

Additional Resource:

- [Why you shouldn't invoke setup.py directly](https://blog.ganssle.io/articles/2021/10/setup-py-deprecated.html)

Pull Request resolved: https://github.com/pytorch/pytorch/pull/156027
Approved by: https://github.com/ezyang

.ci/manywheel/build_libtorch.sh

@@ -104,7 +104,7 @@ if [[ "$DESIRED_CUDA" == *"rocm"* ]]; then
     export ROCclr_DIR=/opt/rocm/rocclr/lib/cmake/rocclr
 fi
 
-echo "Calling setup.py install at $(date)"
+echo "Calling 'python -m pip install .' at $(date)"
 
 if [[ $LIBTORCH_VARIANT = *"static"* ]]; then
     STATIC_CMAKE_FLAG="-DTORCH_STATIC=1"
@@ -120,7 +120,7 @@ fi
         # TODO: Remove this flag once https://github.com/pytorch/pytorch/issues/55952 is closed
         CFLAGS='-Wno-deprecated-declarations' \
         BUILD_LIBTORCH_CPU_WITH_DEBUG=1 \
-        python setup.py install
+        python -m pip install --no-build-isolation -v .
 
     mkdir -p libtorch/{lib,bin,include,share}
 

.ci/pytorch/test.sh

@@ -436,11 +436,11 @@ test_inductor_aoti() {
     python3 tools/amd_build/build_amd.py
   fi
   if [[ "$BUILD_ENVIRONMENT" == *sm86* ]]; then
-    BUILD_COMMAND=(TORCH_CUDA_ARCH_LIST=8.6 USE_FLASH_ATTENTION=OFF python setup.py develop)
+    BUILD_COMMAND=(TORCH_CUDA_ARCH_LIST=8.6 USE_FLASH_ATTENTION=OFF python -m pip install --no-build-isolation -v -e .)
     # TODO: Replace me completely, as one should not use conda libstdc++, nor need special path to TORCH_LIB
     TEST_ENVS=(CPP_TESTS_DIR="${BUILD_BIN_DIR}" LD_LIBRARY_PATH="/opt/conda/envs/py_3.10/lib:${TORCH_LIB_DIR}:${LD_LIBRARY_PATH}")
   else
-    BUILD_COMMAND=(python setup.py develop)
+    BUILD_COMMAND=(python -m pip install --no-build-isolation -v -e .)
     TEST_ENVS=(CPP_TESTS_DIR="${BUILD_BIN_DIR}" LD_LIBRARY_PATH="${TORCH_LIB_DIR}")
   fi
 
@@ -1579,7 +1579,7 @@ test_operator_benchmark() {
   test_inductor_set_cpu_affinity
 
   cd benchmarks/operator_benchmark/pt_extension
-  python setup.py install
+  python -m pip install .
 
   cd "${TEST_DIR}"/benchmarks/operator_benchmark
   $TASKSET python -m benchmark_all_test --device "$1" --tag-filter "$2" \

.devcontainer/README.md

@@ -61,8 +61,8 @@ You are now all set to start developing with PyTorch in a DevContainer environme
 ## Step 8: Build PyTorch
 
 To build pytorch from source, simply run:
-   ```
+   ```bash
-   python setup.py develop
+   python -m pip install --no-build-isolation -v -e .
    ```
 
 The process involves compiling thousands of files, and would take a long time. Fortunately, the compiled objects can be useful for your next build. When you modify some files, you only need to compile the changed files the next time.

.github/scripts/td_llm_indexer.sh

@@ -6,7 +6,7 @@ set -euxo pipefail
 cd llm-target-determinator
 pip install -q -r requirements.txt
 cd ../codellama
-pip install -e .
+pip install --no-build-isolation -v -e .
 pip install numpy==1.26.0
 
 # Run indexer

CONTRIBUTING.md

@@ -88,20 +88,19 @@ source venv/bin/activate  # or `& .\venv\Scripts\Activate.ps1` on Windows
 
 * If you want to have no-op incremental rebuilds (which are fast), see [Make no-op build fast](#make-no-op-build-fast) below.
 
-* When installing with `python setup.py develop` (in contrast to `python setup.py install`) Python runtime will use
+* When installing with `python -m pip install -e .` (in contrast to `python -m pip install .`) Python runtime will use
   the current local source-tree when importing `torch` package. (This is done by creating [`.egg-link`](https://wiki.python.org/moin/PythonPackagingTerminology#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
+  However, you would need to reinstall if you modify Python interface (`.pyi`, `.pyi.in`) or non-Python files (`.cpp`, `.cc`, `.cu`, `.h`, ...).
-   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,
+  One way to avoid running `python -m pip install -e .` 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:
   ```bash
-   pushd torch/lib; sh -c "ln -sf ../../build/lib/libtorch_cpu.* ."; popd
+  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),
+  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.
+  would be sufficient to make change visible in `torch` package.
 
 
   To reinstall, first uninstall all existing PyTorch installs. You may need to run `pip
@@ -115,9 +114,9 @@ source venv/bin/activate  # or `& .\venv\Scripts\Activate.ps1` on Windows
   pip uninstall torch
   ```
 
-  Next run `python setup.py clean`. After that, you can install in `develop` mode again.
+  Next run `python setup.py clean`. After that, you can install in editable mode again.
 
-* If you run into errors when running `python setup.py develop`, here are some debugging steps:
+* If you run into errors when running `python -m pip install -e .`, 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
@@ -130,13 +129,20 @@ source venv/bin/activate  # or `& .\venv\Scripts\Activate.ps1` on Windows
       git clean -xdf
       python setup.py clean
       git submodule update --init --recursive
-      python setup.py develop
+      python -m pip install -r requirements.txt
+      python -m pip install --no-build-isolation -v -e .
       ```
-  4. The main step within `python setup.py develop` is running `make` from the `build` directory. If you want to
+  4. The main step within `python -m pip install -e .` is running `cmake --build build` from the `build` directory. If you want to
     experiment with some environment variables, you can pass them into the command:
       ```bash
-      ENV_KEY1=ENV_VAL1[, ENV_KEY2=ENV_VAL2]* python setup.py develop
+      ENV_KEY1=ENV_VAL1[, ENV_KEY2=ENV_VAL2]* CMAKE_FRESH=1 python -m pip install --no-build-isolation -v -e .
       ```
+  5. Try installing PyTorch without build isolation by adding `--no-build-isolation` to the `pip install` command.
+  This will use the current environment's packages instead of creating a new isolated environment for the build.
+      ```bash
+      python -m pip install --no-build-isolation -v -e .
+      ```
+
 
 * If you run into issue running `git submodule update --init --recursive`. Please try the following:
   - If you encounter an error such as
@@ -639,9 +645,9 @@ 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 `python -m pip install -e .` call to compile PyTorch with `DEBUG=1`.
-your operating system it may also be necessary to run `py-spy` with root
+Depending on your operating system it may also be necessary to run `py-spy` with
-privileges.
+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
@@ -649,7 +655,7 @@ details.
 
 ## Managing multiple build trees
 
-One downside to using `python setup.py develop` is that your development
+One downside to using `python -m pip install -e .` 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).
@@ -663,7 +669,7 @@ specific build of PyTorch. To set one up:
 python -m venv pytorch-myfeature
 source pytorch-myfeature/bin/activate  # or `& .\pytorch-myfeature\Scripts\Activate.ps1` on Windows
 # if you run python now, torch will NOT be installed
-python setup.py develop
+python -m pip install --no-build-isolation -v -e .
 ```
 
 ## C++ development tips
@@ -701,7 +707,9 @@ variables `DEBUG`, `USE_DISTRIBUTED`, `USE_MKLDNN`, `USE_CUDA`, `USE_FLASH_ATTEN
 For example:
 
 ```bash
-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
+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 -m pip install --no-build-isolation -v -e .
 ```
 
 For subsequent builds (i.e., when `build/CMakeCache.txt` exists), the build
@@ -711,7 +719,7 @@ options.
 
 ### Code completion and IDE support
 
-When using `python setup.py develop`, PyTorch will generate
+When using `python -m pip install -e .`, 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
@@ -772,7 +780,7 @@ If not, you can define these variables on the command line before invoking `setu
 export CMAKE_C_COMPILER_LAUNCHER=ccache
 export CMAKE_CXX_COMPILER_LAUNCHER=ccache
 export CMAKE_CUDA_COMPILER_LAUNCHER=ccache
-python setup.py develop
+python -m pip install --no-build-isolation -v -e .
 ```
 
 #### Use a faster linker
@@ -785,7 +793,7 @@ If you are editing a single file and rebuilding in a tight loop, the time spent
 Starting with CMake 3.29, you can specify the linker type using the [`CMAKE_LINKER_TYPE`](https://cmake.org/cmake/help/latest/variable/CMAKE_LINKER_TYPE.html) variable. For example, with `mold` installed:
 
 ```sh
-CMAKE_LINKER_TYPE=MOLD python setup.py develop
+CMAKE_LINKER_TYPE=MOLD python -m pip install --no-build-isolation -v -e .
 ```
 
 #### Use pre-compiled headers
@@ -797,7 +805,7 @@ setting `USE_PRECOMPILED_HEADERS=1` either on first setup, or in the
 `CMakeCache.txt` file.
 
 ```sh
-USE_PRECOMPILED_HEADERS=1 python setup.py develop
+USE_PRECOMPILED_HEADERS=1 python -m pip install --no-build-isolation -v -e .
 ```
 
 This adds a build step where the compiler takes `<ATen/ATen.h>` and essentially
@@ -820,7 +828,7 @@ A compiler-wrapper to fix this is provided in `tools/nvcc_fix_deps.py`. You can
 this as a compiler launcher, similar to `ccache`
 ```bash
 export CMAKE_CUDA_COMPILER_LAUNCHER="python;`pwd`/tools/nvcc_fix_deps.py;ccache"
-python setup.py develop
+python -m pip install --no-build-isolation -v -e .
 ```
 
 ### Rebuild few files with debug information
@@ -1171,7 +1179,7 @@ build_with_asan()
   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
+  python -m pip install --no-build-isolation -v -e .
 }
 
 run_with_asan()

Dockerfile

@@ -57,7 +57,7 @@ RUN --mount=type=cache,target=/opt/ccache \
     export eval ${CMAKE_VARS} && \
     TORCH_CUDA_ARCH_LIST="7.0 7.2 7.5 8.0 8.6 8.7 8.9 9.0 9.0a" TORCH_NVCC_FLAGS="-Xfatbin -compress-all" \
     CMAKE_PREFIX_PATH="$(dirname $(which conda))/../" \
-    python setup.py install
+    python -m pip install --no-build-isolation -v .
 
 FROM conda as conda-installs
 ARG PYTHON_VERSION=3.11

README.md

@@ -228,6 +228,7 @@ If you want to disable Intel GPU support, export the environment variable `USE_X
 Other potentially useful environment variables may be found in `setup.py`.
 
 #### Get the PyTorch Source
+
 ```bash
 git clone https://github.com/pytorch/pytorch
 cd pytorch
@@ -279,24 +280,29 @@ conda install -c conda-forge libuv=1.39
 ```
 
 #### Install PyTorch
+
 **On Linux**
 
 If you're compiling for AMD ROCm then first run this command:
+
 ```bash
 # Only run this if you're compiling for ROCm
 python tools/amd_build/build_amd.py
 ```
 
 Install PyTorch
+
 ```bash
 export CMAKE_PREFIX_PATH="${CONDA_PREFIX:-'$(dirname $(which conda))/../'}:${CMAKE_PREFIX_PATH}"
-python setup.py develop
+python -m pip install -r requirements.txt
+python -m pip install --no-build-isolation -v -e .
 ```
 
 **On macOS**
 
 ```bash
-python3 setup.py develop
+python -m pip install -r requirements.txt
+python -m pip install --no-build-isolation -v -e .
 ```
 
 **On Windows**
@@ -308,7 +314,7 @@ If you want to build legacy python code, please refer to [Building on legacy cod
 In this mode PyTorch computations will run on your CPU, not your GPU.
 
 ```cmd
-python setup.py develop
+python -m pip install --no-build-isolation -v -e .
 ```
 
 Note on OpenMP: The desired OpenMP implementation is Intel OpenMP (iomp). In order to link against iomp, you'll need to manually download the library and set up the building environment by tweaking `CMAKE_INCLUDE_PATH` and `LIB`. The instruction [here](https://github.com/pytorch/pytorch/blob/main/docs/source/notes/windows.rst#building-from-source) is an example for setting up both MKL and Intel OpenMP. Without these configurations for CMake, Microsoft Visual C OpenMP runtime (vcomp) will be used.
@@ -329,7 +335,6 @@ Additional libraries such as
 
 You can refer to the [build_pytorch.bat](https://github.com/pytorch/pytorch/blob/main/.ci/pytorch/win-test-helpers/build_pytorch.bat) script for some other environment variables configurations
 
-
 ```cmd
 cmd
 
@@ -349,8 +354,7 @@ for /f "usebackq tokens=*" %i in (`"%ProgramFiles(x86)%\Microsoft Visual Studio\
 :: [Optional] If you want to override the CUDA host compiler
 set CUDAHOSTCXX=C:\Program Files (x86)\Microsoft Visual Studio\2019\Community\VC\Tools\MSVC\14.27.29110\bin\HostX64\x64\cl.exe
 
-python setup.py develop
+python -m pip install --no-build-isolation -v -e .
-
 ```
 
 **Intel GPU builds**
@@ -372,7 +376,7 @@ if defined CMAKE_PREFIX_PATH (
     set "CMAKE_PREFIX_PATH=%CONDA_PREFIX%\Library"
 )
 
-python setup.py develop
+python -m pip install --no-build-isolation -v -e .
 ```
 
 ##### Adjust Build Options (Optional)
@@ -382,6 +386,7 @@ the following. For example, adjusting the pre-detected directories for CuDNN or
 with such a step.
 
 On Linux
+
 ```bash
 export CMAKE_PREFIX_PATH="${CONDA_PREFIX:-'$(dirname $(which conda))/../'}:${CMAKE_PREFIX_PATH}"
 CMAKE_ONLY=1 python setup.py build
@@ -389,6 +394,7 @@ ccmake build  # or cmake-gui build
 ```
 
 On macOS
+
 ```bash
 export CMAKE_PREFIX_PATH="${CONDA_PREFIX:-'$(dirname $(which conda))/../'}:${CMAKE_PREFIX_PATH}"
 MACOSX_DEPLOYMENT_TARGET=10.9 CC=clang CXX=clang++ CMAKE_ONLY=1 python setup.py build

benchmarks/README.md

@@ -8,12 +8,12 @@ It also provides mechanisms to compare PyTorch with other frameworks.
 Make sure you're on a machine with CUDA, torchvision, and pytorch installed. Install in the following order:
 ```
 # Install torchvision. It comes with the pytorch stable release binary
-pip3 install torch torchvision
+python -m pip install torch torchvision
 
 # Install the latest pytorch master from source.
 # It should supersede the installation from the release binary.
 cd $PYTORCH_HOME
-python setup.py build develop
+python -m pip install --no-build-isolation -v -e .
 
 # Check the pytorch installation version
 python -c "import torch; print(torch.__version__)"

benchmarks/functional_autograd_benchmark/README.md

@@ -17,8 +17,8 @@ export DEBUG=0
 export OMP_NUM_THREADS=10
 
 # Compile pytorch with the base revision
-git checkout master
+git checkout main
-python setup.py develop
+python -m pip install --no-build-isolation -v -e .
 
 # Install dependencies:
 # Scipy is required by detr
@@ -32,7 +32,7 @@ python functional_autograd_benchmark.py --output before.txt
 # Compile pytorch with your change
 popd
 git checkout your_feature_branch
-python setup.py develop
+python -m pip install --no-build-isolation -v -e .
 
 # Run the benchmark for the new version
 pushd benchmarks/functional_autograd_benchmark

benchmarks/operator_benchmark/README.md

@@ -20,7 +20,7 @@ Key Features:
 The instruction below installs a cpp\_extension for PyTorch and it is required to run the benchmark suite.
 ```bash
 cd pt_extension
-python setup.py install
+python -m pip install .
 ```
 
 ## How to run the benchmarks:

benchmarks/sparse/test_csr.sh

@@ -11,7 +11,7 @@ export USE_MKL=1
 CMAKE_ONLY=1 python setup.py build
 ccmake build  # or cmake-gui build
 
-python setup.py install
+python -m pip install --no-build-isolation -v .
 
 cd benchmarks
 echo "!! SPARSE SPMM TIME BENCHMARK!! " >> $OUTFILE
@@ -28,7 +28,7 @@ echo "----- USE_MKL=0 ------" >> $OUTFILE
 rm -rf build
 
 export USE_MKL=0
-python setup.py install
+python -m pip install --no-build-isolation -v .
 
 cd benchmarks
 for dim0 in 1000 5000 10000; do

scripts/onnx/install-develop.sh

@@ -15,4 +15,4 @@ pip install --no-use-pep517 -e "$tp2_dir/onnx"
 # Install caffe2 and pytorch
 pip install -r "$top_dir/caffe2/requirements.txt"
 pip install -r "$top_dir/requirements.txt"
-python setup.py develop
+python -m pip install --no-build-isolation -v -e .

scripts/onnx/install.sh

@@ -35,4 +35,4 @@ _pip_install -b "$BUILD_DIR/onnx" "file://$tp2_dir/onnx#egg=onnx"
 # Install caffe2 and pytorch
 pip install -r "$top_dir/caffe2/requirements.txt"
 pip install -r "$top_dir/requirements.txt"
-python setup.py install
+python -m pip install --no-build-isolation -v .

setup.py

@@ -263,6 +263,7 @@ import json
 import shutil
 import subprocess
 import sysconfig
+import textwrap
 import time
 from collections import defaultdict
 from pathlib import Path
@@ -601,7 +602,7 @@ def build_deps() -> None:
         report(
             'Finished running cmake. Run "ccmake build" or '
             '"cmake-gui build" to adjust build options and '
-            '"python setup.py install" to build.'
+            '"python -m pip install --no-build-isolation -v ." to build.'
         )
         sys.exit()
 
@@ -1207,24 +1208,25 @@ def configure_extension_build() -> tuple[
 
 # post run, warnings, printed at the end to make them more visible
 build_update_message = """
-    It is no longer necessary to use the 'build' or 'rebuild' targets
+It is no longer necessary to use the 'build' or 'rebuild' targets
 
-    To install:
+To install:
-      $ python setup.py install
+  $ python -m pip install --no-build-isolation -v .
-    To develop locally:
+To develop locally:
-      $ python setup.py develop
+  $ python -m pip install --no-build-isolation -v -e .
-    To force cmake to re-generate native build files (off by default):
+To force cmake to re-generate native build files (off by default):
-      $ CMAKE_FRESH=1 python setup.py develop
+  $ CMAKE_FRESH=1 python -m pip install --no-build-isolation -v -e .
-"""
+""".strip()
 
 
 def print_box(msg: str) -> None:
-    lines = msg.split("\n")
+    msg = textwrap.dedent(msg).strip()
-    size = max(len(l) + 1 for l in lines)
+    lines = ["", *msg.split("\n"), ""]
-    print("-" * (size + 2))
+    max_width = max(len(l) for l in lines)
-    for l in lines:
+    print("+" + "-" * (max_width + 4) + "+", file=sys.stderr, flush=True)
-        print("|{}{}|".format(l, " " * (size - len(l))))
+    for line in lines:
-    print("-" * (size + 2))
+        print(f"|  {line:<{max_width}s}  |", file=sys.stderr, flush=True)
+    print("+" + "-" * (max_width + 4) + "+", file=sys.stderr, flush=True)
 
 
 def main() -> None:

test/cpp/jit/README.md

@@ -36,7 +36,7 @@ The following commands assume you are in PyTorch root.
 
 ```bash
 # ... Build PyTorch from source, e.g.
-python setup.py develop
+python -m pip install --no-build-isolation -v -e .
 # (re)build just the binary
 ninja -C build bin/test_jit
 # run tests

test/cpp_extensions/open_registration_extension/README.md

@@ -4,8 +4,8 @@ This folder contains a self-contained example of a PyTorch out-of-tree backend l
 
 ## How to use
 
-Install as standalone with `python setup.py develop` (or install) from this folder.
+Install as standalone with `python -m pip install -e .` (or `python -m pip install .`)
-You can run test via `python {PYTORCH_ROOT_PATH}/test/test_openreg.py`.
+from this folder. You can run test via `python {PYTORCH_ROOT_PATH}/test/test_openreg.py`.
 
 ## Design principles
 

test/run_test.py

@@ -667,6 +667,7 @@ def install_cpp_extensions(cpp_extensions_test_dir, env=os.environ):
         shutil.rmtree(cpp_extensions_test_build_dir)
 
     # Build the test cpp extensions modules
+    # FIXME: change setup.py command to pip command
     cmd = [sys.executable, "setup.py", "install", "--root", "./install"]
     return_code = shell(cmd, cwd=cpp_extensions_test_dir, env=env)
     if return_code != 0:

test/test_cpp_extensions_aot.py

@@ -148,7 +148,7 @@ class TestCppExtensionAOT(common.TestCase):
 
     @unittest.skipIf(IS_WINDOWS, "Not available on Windows")
     def test_no_python_abi_suffix_sets_the_correct_library_name(self):
-        # For this test, run_test.py will call `python setup.py install` in the
+        # For this test, run_test.py will call `python -m pip install .` in the
         # cpp_extensions/no_python_abi_suffix_test folder, where the
         # `BuildExtension` class has a `no_python_abi_suffix` option set to
         # `True`. This *should* mean that on Python 3, the produced shared

tools/build_with_debinfo.py

@@ -95,7 +95,8 @@ def main() -> None:
         sys.exit(-95)
     if not is_devel_setup():
         print(
-            "Not a devel setup of PyTorch, please run `python3 setup.py develop --user` first"
+            "Not a devel setup of PyTorch, "
+            "please run `python -m pip install --no-build-isolation -v -e .` first"
         )
         sys.exit(-1)
     if not has_build_ninja():

torch/__init__.py

@@ -1019,10 +1019,10 @@ except ImportError:
                     of the PyTorch repository rather than the C extensions which
                     are expected in the `torch._C` namespace. This can occur when
                     using the `install` workflow. e.g.
-                        $ python setup.py install && python -c "import torch"
+                        $ python -m pip install --no-build-isolation -v . && python -c "import torch"
 
                     This error can generally be solved using the `develop` workflow
-                        $ python setup.py develop && python -c "import torch"  # This should succeed
+                        $ python -m pip install --no-build-isolation -v -e . && python -c "import torch"  # This should succeed
                     or by running Python from a different directory.
                 """
             ).strip()

torch/sparse/_triton_ops_meta.py

@@ -60,7 +60,7 @@ for a particular GPU can be computed by simply running this script in
 the pytorch development tree::
 
   cd /path/to/pytorch
-  python setup.py develop
+  python -m pip install --no-build-isolation -v -e .
   python torch/sparse/_triton_ops_meta.py
 
 This will compute the optimal kernel parameters for the GPU device