Windows install

[fablevi]

Describe the bug

Please create a better installer for windows, right now its pretty hard. I don't know if it 'cus the md file is wrong, or i miss something, but its hard.

Environment details

Intel ARC A770, win11, python3.10 and 3.13

[pbchekin]

@fablevi thank you for reporting this. We definitely need to improve build experience for Windows. Will nightly wheels work for your situation?

While we are working on this, could you please add more details on the issue: what didn't work, what is wrong with the instructions.
Thanks!

[fablevi]

Hey @pbchekin thank for reply :)

Soo this is the code where i was stuck.

$env:CMAKE_SHARED_LINKER_FLAGS = "/FORCE:MULTIPLE"
$env:CMAKE_MODULE_LINKER_FLAGS = "/FORCE:MULTIPLE"
$env:CMAKE_EXE_LINKER_FLAGS = "/FORCE:MULTIPLE"
bash -c "./scripts/install-pytorch.sh --source"

The install torch srcipt somewhere was stuck. At the an it sad, the code try 100 times but can not install.

I realy realy hope that you can somehow fix it.
If you can add like an argument that is specify how many times it try to install it would be great.

Or if you can make a wheel and an autobuild script for this this would be good.

If you need anymore error i try to inform you about it, but the logs was too much.

[fablevi]

Small update for windows users:
pip install c:\Users\fable\Documents\triton-3.3.0+gitfff17732-cp311-cp311-linux_x86_64.whl
ERROR: triton-3.3.0+gitfff17732-cp311-cp311-linux_x86_64.whl is not a supported wheel on this platform.

I donwloaded from the nightly link above.

[fablevi]

Another small updated:

Invoke-BatchFile "C:\Program Files (x86)\Intel\oneAPI\setvars.bat"
:: initializing oneAPI environment...
Initializing Visual Studio command-line environment...
Visual Studio version 17.14.0 environment configured.
"C:\Program Files\Microsoft Visual Studio\2022\Community"
Visual Studio command-line environment initialized for: 'x64'
: advisor -- latest
: compiler -- latest
: dal -- latest
: debugger -- latest
: dev-utilities -- latest
: dnnl -- latest
: dpcpp-ct -- latest
: dpl -- latest
: ipp -- latest
: ippcp -- latest
: mkl -- latest
: ocloc -- latest
: pti -- latest
: tbb -- latest
: umf -- latest
: vtune -- latest
:: oneAPI environment initialized ::

$env:CMAKE_SHARED_LINKER_FLAGS = "/FORCE:MULTIPLE"
$env:CMAKE_MODULE_LINKER_FLAGS = "/FORCE:MULTIPLE"
$env:CMAKE_EXE_LINKER_FLAGS = "/FORCE:MULTIPLE"
bash -c "./scripts/install-pytorch.sh --source"

||
||
||
v

--
-- ******** Summary ********
-- General:
-- CMake version : 3.31.6
-- CMake command : D:/git/vllm/src/intel-xpu-backend-for-triton/venv/Lib/site-packages/cmake/data/bin/cmake.exe
-- System : Windows
-- C++ compiler : C:/Program Files/Microsoft Visual Studio/2022/Community/VC/Tools/MSVC/14.44.35207/bin/Hostx64/x64/cl.exe
-- C++ compiler id : MSVC
-- C++ compiler version : 19.44.35207.1
-- Using ccache if found : OFF
-- CXX flags : /DWIN32 /D_WINDOWS /GR /EHsc /Zc:__cplusplus /bigobj /FS /utf-8 -DUSE_PTHREADPOOL -DNDEBUG -DUSE_KINETO -DLIBKINETO_NOCUPTI -DLIBKINETO_NOROCTRACER -DLIBKINETO_NOXPUPTI=ON -DUSE_FBGEMM -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE /wd4624 /wd4068 /wd4067 /wd4267 /wd4661 /wd4717 /wd4244 /wd4804 /wd4273 -DUSE_XPU
-- Shared LD flags : /FORCE:MULTIPLE /ignore:4049 /ignore:4217 /ignore:4099
-- Static LD flags : /machine:x64 /ignore:4049 /ignore:4217 /ignore:4099
-- Module LD flags : /FORCE:MULTIPLE /ignore:4049 /ignore:4217 /ignore:4099
-- Build type : Release
-- Compile definitions : ONNX_ML=1;ONNXIFI_ENABLE_EXT=1;ONNX_NAMESPACE=onnx_torch;_CRT_SECURE_NO_DEPRECATE=1;IDEEP_USE_MKL;USE_EXTERNAL_MZCRC;MINIZ_DISABLE_ZIP_READER_CRC32_CHECKS;EXPORT_AOTI_FUNCTIONS;WIN32_LEAN_AND_MEAN;_UCRT_LEGACY_INFINITY;NOMINMAX;USE_MIMALLOC
-- CMAKE_PREFIX_PATH : D:\git\vllm\src\intel-xpu-backend-for-triton\venv\Lib\site-packages;C:\Program Files (x86)\Intel\oneAPI\tbb\latest\env..;C:\Program Files (x86)\Intel\oneAPI\pti\latest\env..\lib\cmake\pti;C:\Program Files (x86)\Intel\oneAPI\mkl\latest\lib\cmake;C:\Program Files (x86)\Intel\oneAPI\ipp\latest\lib\cmake\ipp;C:\Program Files (x86)\Intel\oneAPI\dpl\latest\lib\cmake\oneDPL;C:\Program Files (x86)\Intel\oneAPI\dnnl\latest\env..\lib\cmake;C:\Program Files (x86)\Intel\oneAPI\dal\latest;C:\Program Files (x86)\Intel\oneAPI\compiler\latest;
-- CMAKE_INSTALL_PREFIX : D:/git/vllm/src/intel-xpu-backend-for-triton/.scripts_cache/pytorch/torch
-- USE_GOLD_LINKER : OFF

-- TORCH_VERSION : 2.8.0
-- BUILD_STATIC_RUNTIME_BENCHMARK: OFF
-- BUILD_BINARY : OFF
-- BUILD_CUSTOM_PROTOBUF : ON
-- Link local protobuf : ON
-- BUILD_PYTHON : True
-- Python version : 3.11.9
-- Python executable : D:\git\vllm\src\intel-xpu-backend-for-triton\venv\Scripts\python.exe
-- Python library : C:/Program Files/WindowsApps/PythonSoftwareFoundation.Python.3.11_3.11.2544.0_x64__qbz5n2kfra8p0/libs/python311.lib
-- Python includes : C:/Program Files/WindowsApps/PythonSoftwareFoundation.Python.3.11_3.11.2544.0_x64__qbz5n2kfra8p0/Include
-- Python site-package : D:\git\vllm\src\intel-xpu-backend-for-triton\venv\Lib\site-packages
-- BUILD_SHARED_LIBS : ON
-- CAFFE2_USE_MSVC_STATIC_RUNTIME : OFF
-- BUILD_TEST : True
-- BUILD_JNI : OFF
-- BUILD_MOBILE_AUTOGRAD : OFF
-- BUILD_LITE_INTERPRETER: OFF
-- INTERN_BUILD_MOBILE :
-- TRACING_BASED : OFF
-- USE_BLAS : 1
-- BLAS : mkl
-- BLAS_HAS_SBGEMM :
-- USE_LAPACK : 1
-- LAPACK : mkl
-- USE_ASAN : OFF
-- USE_TSAN : OFF
-- USE_CPP_CODE_COVERAGE : OFF
-- USE_CUDA : OFF
-- USE_XPU : ON
-- SYCL compiler version : 20250101
-- SYCL include path : C:/Program Files (x86)/Intel/oneAPI/compiler/latest/include;C:/Program Files (x86)/Intel/oneAPI/compiler/latest/include/sycl
-- SYCL library : C:/Program Files (x86)/Intel/oneAPI/compiler/latest/lib/sycl.lib
-- USE_ROCM : OFF
-- BUILD_NVFUSER :
-- USE_EIGEN_FOR_BLAS :
-- USE_FBGEMM : ON
-- USE_FAKELOWP : OFF
-- USE_KINETO : ON
-- USE_GFLAGS : OFF
-- USE_GLOG : OFF
-- USE_LITE_PROTO : OFF
-- USE_PYTORCH_METAL : OFF
-- USE_PYTORCH_METAL_EXPORT : OFF
-- USE_MPS : OFF
-- CAN_COMPILE_METAL :
-- USE_MKL : ON
-- USE_STATIC_MKL : 1
-- USE_MKLDNN : ON
-- USE_MKLDNN_ACL : OFF
-- USE_MKLDNN_CBLAS : OFF
-- USE_UCC : OFF
-- USE_ITT : ON
-- USE_XCCL : OFF
-- USE_NCCL : OFF
-- USE_NNPACK : OFF
-- USE_NUMPY : ON
-- USE_OBSERVERS : ON
-- USE_OPENCL : OFF
-- USE_OPENMP : ON
-- USE_MIMALLOC : ON
-- USE_MIMALLOC_ON_MKL : OFF
-- USE_VULKAN : OFF
-- USE_PROF : OFF
-- USE_PYTORCH_QNNPACK : OFF
-- USE_XNNPACK : ON
-- USE_DISTRIBUTED : OFF
-- Public Dependencies : caffe2::mkl
-- Private Dependencies : Threads::Threads;pthreadpool;cpuinfo;XNNPACK;microkernels-prod;fbgemm;ittnotify;fp16;caffe2::openmp;fmt::fmt-header-only;kineto
-- Public CUDA Deps. :
-- Private CUDA Deps. :
-- USE_COREML_DELEGATE : OFF
-- BUILD_LAZY_TS_BACKEND : ON
-- USE_ROCM_KERNEL_ASSERT : OFF
-- Performing Test HAS_WMISSING_PROTOTYPES
-- Performing Test HAS_WMISSING_PROTOTYPES - Failed
-- Performing Test HAS_WERROR_MISSING_PROTOTYPES
-- Performing Test HAS_WERROR_MISSING_PROTOTYPES - Failed
-- Configuring incomplete, errors occurred!