No XPU devices found while running on Docker

[v3vishal]

Hi there, I'm trying to get the intel-extension-for-tensorflow working on my Core Ultra 5 125H Arc Graphics, and I have made a docker container as in the docs. However, while running env_check.py, I get this as my output

__file__:     //env_check.py
Check Python
         Python 3.10.12 is Supported.
Check Python Passed

Check OS
        OS ubuntu:22.04 is Supported
Check OS Passed

Check Tensorflow
2024-11-17 07:52:27.501076: I tensorflow/core/util/port.cc:113] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.
2024-11-17 07:52:27.504476: I external/local_tsl/tsl/cuda/cudart_stub.cc:31] Could not find cuda drivers on your machine, GPU will not be used. 
2024-11-17 07:52:27.553677: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered
2024-11-17 07:52:27.553752: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered
2024-11-17 07:52:27.556099: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered
2024-11-17 07:52:27.567004: I external/local_tsl/tsl/cuda/cudart_stub.cc:31] Could not find cuda drivers on your machine, GPU will not be used. 
2024-11-17 07:52:27.567281: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
To enable the following instructions: AVX2 AVX_VNNI FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
2024-11-17 07:52:29.058960: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT
2024-11-17 07:52:30.818112: W external/local_tsl/tsl/lib/monitoring/collection_registry.cc:81] Trying to register 2 metrics with the same name: /tensorflow/core/bfc_allocator_delay. The old value will be erased in order to register a new one. Please check if you link the metric more than once, or if the name is already used by other metrics.
2024-11-17 07:52:30.818389: W external/local_tsl/tsl/lib/monitoring/collection_registry.cc:81] Trying to register 2 metrics with the same name: /xla/service/gpu/compiled_programs_count. The old value will be erased in order to register a new one. Please check if you link the metric more than once, or if the name is already used by other metrics.
2024-11-17 07:52:30.821138: W external/local_tsl/tsl/lib/monitoring/collection_registry.cc:81] Trying to register 2 metrics with the same name: /jax/pjrt/pjrt_executable_executions. The old value will be erased in order to register a new one. Please check if you link the metric more than once, or if the name is already used by other metrics.
2024-11-17 07:52:30.821247: W external/local_tsl/tsl/lib/monitoring/collection_registry.cc:81] Trying to register 2 metrics with the same name: /jax/pjrt/pjrt_executable_execution_time_usecs. The old value will be erased in order to register a new one. Please check if you link the metric more than once, or if the name is already used by other metrics.
2024-11-17 07:52:31.173258: I itex/core/wrapper/itex_gpu_wrapper.cc:38] Intel Extension for Tensorflow* GPU backend is loaded.
2024-11-17 07:52:31.174278: I external/local_xla/xla/pjrt/pjrt_api.cc:67] PJRT_Api is set for device type xpu
2024-11-17 07:52:31.174314: I external/local_xla/xla/pjrt/pjrt_api.cc:72] PJRT plugin for XPU has PJRT API version 0.33. The framework PJRT API version is 0.34.
2024-11-17 07:52:31.206038: E external/intel_xla/xla/stream_executor/sycl/sycl_gpu_runtime.cc:178] Can not found any devices.
2024-11-17 07:52:31.206283: E itex/core/kernels/xpu_kernel.cc:60] Failed precondition: No visible XPU devices. To check runtime environment on your host, please run itex/tools/python/env_check.py.
If you need help, create an issue at https://github.com/intel/intel-extension-for-tensorflow/issues
2024-11-17 07:52:31.255435: E itex/core/devices/gpu/itex_gpu_runtime.cc:174] Can not found any devices. To check runtime environment on your host, please run itex/tools/python/env_check.py.
If you need help, create an issue at https://github.com/intel/intel-extension-for-tensorflow/issues
        Tensorflow 2.15.1 is installed.
Check Tensorflow Passed

Check Intel GPU Driver
Package: intel-level-zero-gpu
Status: install ok installed
Priority: optional
Section: libs
Installed-Size: 28239
Maintainer: Intel Graphics Team <[email protected]>
Architecture: amd64
Source: intel-compute-runtime
Version: 1.3.27642.50-803~22.04
Depends: libc6 (>= 2.34), libgcc-s1 (>= 3.4), libigdgmm12 (>= 22.3.15), libstdc++6 (>= 12), libigc1 (>= 1.0.12812), libigdfcl1 (>= 1.0.12812), libnl-3-200, libnl-route-3-200
Description: Intel(R) Graphics Compute Runtime for oneAPI Level Zero.
 Level Zero is the primary low-level interface for language and runtime
 libraries. Level Zero offers fine-grain control over accelerators
 capabilities, delivering a simplified and low-latency interface to
 hardware, and efficiently exposing hardware capabilities to applications.
Homepage: https://github.com/oneapi-src/level-zero
Original-Maintainer: Debian OpenCL Maintainers <[email protected]>
Package: intel-opencl-icd
Status: install ok installed
Priority: optional
Section: libs
Installed-Size: 23865
Maintainer: Intel Graphics Team <[email protected]>
Architecture: amd64
Source: intel-compute-runtime
Version: 23.43.27642.50-803~22.04
Replaces: intel-opencl
Provides: opencl-icd
Depends: libc6 (>= 2.34), libgcc-s1 (>= 3.4), libigdgmm12 (>= 22.3.15), libstdc++6 (>= 12), ocl-icd-libopencl1, libigc1 (>= 1.0.12812), libigdfcl1 (>= 1.0.12812)
Recommends: intel-igc-cm (>= 1.0.100)
Breaks: intel-opencl
Conffiles:
 /etc/OpenCL/vendors/intel.icd d0a34d0b4f75385c56ee357bb1b8e2d0
Description: Intel graphics compute runtime for OpenCL
 The Intel(R) Graphics Compute Runtime for OpenCL(TM) is a open source
 project to converge Intel's development efforts on OpenCL(TM) compute
 stacks supporting the GEN graphics hardware architecture.
 .
 Supported platforms:
 - Intel Core Processors with Gen8 GPU (Broadwell) - OpenCL 2.1
 - Intel Core Processors with Gen9 GPU (Skylake, Kaby Lake, Coffee Lake) - OpenCL 2.1
 - Intel Atom Processors with Gen9 GPU (Apollo Lake, Gemini Lake) - OpenCL 1.2
 - Intel Core Processors with Gen11 GPU (Ice Lake) - OpenCL 2.1
 - Intel Core Processors with Gen12 graphics devices (formerly Tiger Lake) - OpenCL 2.1
Homepage: https://github.com/intel/compute-runtime
Original-Maintainer: Debian OpenCL Maintainers <[email protected]>
Package: level-zero
Status: install ok installed
Priority: optional
Section: libs
Installed-Size: 1049
Maintainer: Intel Graphics Team <[email protected]>
Architecture: amd64
Source: level-zero-loader
Version: 1.14.0-744~22.04
Depends: libc6 (>= 2.34), libgcc-s1 (>= 3.3.1), libstdc++6 (>= 11)
Description: Intel(R) Graphics Compute Runtime for oneAPI Level Zero.
 Level Zero is the primary low-level interface for language and runtime
 libraries. Level Zero offers fine-grain control over accelerators
 capabilities, delivering a simplified and low-latency interface to
 hardware, and efficiently exposing hardware capabilities to applications.
 .
 This package provides the loader for oneAPI Level Zero compute runtimes.
Homepage: https://github.com/oneapi-src/level-zero
Package: libigc1
Status: install ok installed
Priority: optional
Section: libs
Installed-Size: 86364
Maintainer: Intel Graphics Team <[email protected]>
Architecture: amd64
Source: intel-graphics-compiler
Version: 1.0.15468.29-803~22.04
Depends: libc6 (>= 2.34), libgcc-s1 (>= 3.4), libstdc++6 (>= 12), zlib1g (>= 1:1.2.2)
Description: Intel graphics compiler for OpenCL -- core libs
 The Intel(R) Graphics Compiler for OpenCL(TM) is an llvm based compiler
 for OpenCL(TM) targeting Intel Gen graphics hardware architecture.
 .
 This package includes the core libraries.
Homepage: https://github.com/intel/intel-graphics-compiler
Original-Maintainer: Debian OpenCL team <[email protected]>
Package: libigdfcl1
Status: install ok installed
Priority: optional
Section: libs
Installed-Size: 116046
Maintainer: Intel Graphics Team <[email protected]>
Architecture: amd64
Source: intel-graphics-compiler
Version: 1.0.15468.29-803~22.04
Depends: libc6 (>= 2.34), libgcc-s1 (>= 3.4), libstdc++6 (>= 11), zlib1g (>= 1:1.2.0), libz3-4 (>= 4.7.1)
Description: Intel graphics compiler for OpenCL -- OpenCL library
 The Intel(R) Graphics Compiler for OpenCL(TM) is an llvm based compiler
 for OpenCL(TM) targeting Intel Gen graphics hardware architecture.
 .
 This package includes the library for OpenCL.
Homepage: https://github.com/intel/intel-graphics-compiler
Original-Maintainer: Debian OpenCL team <[email protected]>
Package: libigdgmm12
Status: install ok installed
Priority: optional
Section: libs
Installed-Size: 648
Maintainer: Intel Graphics Team <[email protected]>
Architecture: amd64
Multi-Arch: same
Source: intel-gmmlib
Version: 22.3.15-803~22.04
Replaces: libigdgmm11
Depends: libc6 (>= 2.34), libgcc-s1 (>= 3.3.1), libstdc++6 (>= 4.1.1)
Description: Intel Graphics Memory Management Library -- shared library
 The Intel Graphics Memory Management Library provides device specific
 and buffer management for the Intel Graphics Compute Runtime for
 OpenCL and the Intel Media Driver for VAAPI.
 .
 This library is only useful for Broadwell and newer CPUs.
 .
 This package includes the shared library.
Homepage: https://github.com/intel/gmmlib
Original-Maintainer: Debian Multimedia Maintainers <[email protected]>
Check Intel GPU Driver Passsed

Check OneAPI
       223:     find library=libsycl.so.7 [0]; searching
       223:       trying file=/usr/local/lib/python3.10/dist-packages/tensorflow-plugins/../intel_extension_for_tensorflow/libsycl.so.7
       223:       trying file=/opt/intel/oneapi/redist/lib/libsycl.so.7
       223:     calling init: /opt/intel/oneapi/redist/lib/libsycl.so.7
       223:     calling fini: /opt/intel/oneapi/redist/lib/libsycl.so.7 [0]
        Intel(R) OneAPI DPC++/C++ Compiler is Installed.
Recommended dpcpp version is 2024.2.1-1079
       223:     find library=libmkl_sycl_blas.so.4 [0]; searching
       223:       trying file=/usr/local/lib/python3.10/dist-packages/tensorflow-plugins/../intel_extension_for_tensorflow/libmkl_sycl_blas.so.4
       223:       trying file=/opt/intel/oneapi/redist/lib/libmkl_sycl_blas.so.4
       223:     calling init: /opt/intel/oneapi/redist/lib/libmkl_sycl_blas.so.4
       223:     calling fini: /opt/intel/oneapi/redist/lib/libmkl_sycl_blas.so.4 [0]
       223:     find library=libmkl_sycl_lapack.so.4 [0]; searching
       223:       trying file=/usr/local/lib/python3.10/dist-packages/tensorflow-plugins/../intel_extension_for_tensorflow/libmkl_sycl_lapack.so.4
       223:       trying file=/opt/intel/oneapi/redist/lib/libmkl_sycl_lapack.so.4
       223:     calling init: /opt/intel/oneapi/redist/lib/libmkl_sycl_lapack.so.4
       223:     calling fini: /opt/intel/oneapi/redist/lib/libmkl_sycl_lapack.so.4 [0]
       223:     find library=libmkl_sycl_dft.so.4 [0]; searching
       223:       trying file=/usr/local/lib/python3.10/dist-packages/tensorflow-plugins/../intel_extension_for_tensorflow/libmkl_sycl_dft.so.4 
       223:       trying file=/opt/intel/oneapi/redist/lib/libmkl_sycl_dft.so.4
       223:     calling init: /opt/intel/oneapi/redist/lib/libmkl_sycl_dft.so.4
       223:     calling fini: /opt/intel/oneapi/redist/lib/libmkl_sycl_dft.so.4 [0]
        Intel(R) OneAPI Math Kernel Library is Installed.
Recommended onemkl version is 2024.2.1-103
Check OneAPI Passed

Check Tensorflow Requirements

Check Intel(R) Extension for TensorFlow* Requirements Passed

I'm unable to get XPU devices detected, but /dev/dri shows card0 which means my card is attached

I'm on a time crunch right now, so any help would be immensely appreciated.

I've tried WSL, and docker is the only one working with TensorFlow and oneAPI

Thanks in advance!

[srinarayan-srikanthan]

Hi @v3vishal , did you build the container or pull it from dockerhub?

[v3vishal]

Pulled it from docker hub.

[srinarayan-srikanthan]

Okay, can you try steps here to make sure we set up the container the rite way : https://github.com/intel/ai-containers/tree/main/preset#run-on-gpu