[GHA] Uplift Linux GPU RT version to 24.35.30872.22

[bb-sycl]

Scheduled drivers uplift

[sarnex]

@jsji Any idea what's going on here? Note this is not dev-igc.

[jsji]

@jsji Any idea what's going on here? Note this is not dev-igc.

https://github.com/intel/llvm/actions/runs/11006256785/job/30592910147?pr=15481

sudo -E bash devops/scripts/install_drivers.sh llvm/devops/dependencies.json --all
shell: sh -e {0}
env:
GITHUB_TOKEN: ***
sudo: a terminal is required to read the password; either use the -S option to read from standard input or configure an askpass helper
sudo: a password is required

looks like we failed to install the driver due to sudo password? Any recent changes related to the sudo/password or docker image?

[sarnex]

let me msg you

[sarnex]

The problem is actually that we're incorrectly trying to install intel drivers on the NVidia runner, we didn't do that before but it was an unintended side effect to a recent CI change. Udit is working on it.

[uditagarwal97]

The problem is actually that we're incorrectly trying to install intel drivers on the NVidia runner, we didn't do that before but it was an unintended side effect to a recent CI change. Udit is working on it.

PR for fix: #15528

[sarnex]

@AllanZyne Any ideas about the many asan failures on DG2 here? Note this is using a new GPU driver, newer than the one in normal CI.

********************
Failed Tests (39):
  SYCL :: AddressSanitizer/bad-free/bad-free-host.cpp
  SYCL :: AddressSanitizer/bad-free/bad-free-minus1.cpp
  SYCL :: AddressSanitizer/bad-free/bad-free-plus1.cpp
  SYCL :: AddressSanitizer/common/demangle-kernel-name.cpp
  SYCL :: AddressSanitizer/common/kernel-debug.cpp
  SYCL :: AddressSanitizer/common/option-redzone-size.cpp
  SYCL :: AddressSanitizer/common/options-invalid.cpp
  SYCL :: AddressSanitizer/double-free/double-free.cpp
  SYCL :: AddressSanitizer/misaligned/misalign-int.cpp
  SYCL :: AddressSanitizer/misaligned/misalign-long.cpp
  SYCL :: AddressSanitizer/misaligned/misalign-short.cpp
  SYCL :: AddressSanitizer/multiple-reports/multiple_kernels.cpp
  SYCL :: AddressSanitizer/multiple-reports/one_kernel.cpp
  SYCL :: AddressSanitizer/out-of-bounds/DeviceGlobal/device_global.cpp
  SYCL :: AddressSanitizer/out-of-bounds/DeviceGlobal/device_global_image_scope.cpp
  SYCL :: AddressSanitizer/out-of-bounds/DeviceGlobal/device_global_image_scope_unaligned.cpp
  SYCL :: AddressSanitizer/out-of-bounds/DeviceGlobal/multi_device_images.cpp
  SYCL :: AddressSanitizer/out-of-bounds/USM/large_group_size.cpp
  SYCL :: AddressSanitizer/out-of-bounds/USM/parallel_for_char.cpp
  SYCL :: AddressSanitizer/out-of-bounds/USM/parallel_for_func.cpp
  SYCL :: AddressSanitizer/out-of-bounds/USM/parallel_for_int.cpp
  SYCL :: AddressSanitizer/out-of-bounds/USM/parallel_for_short.cpp
  SYCL :: AddressSanitizer/out-of-bounds/USM/parallel_no_local_size.cpp
  SYCL :: AddressSanitizer/out-of-bounds/USM/unaligned_shadow_memory.cpp
  SYCL :: AddressSanitizer/out-of-bounds/buffer/buffer.cpp
  SYCL :: AddressSanitizer/out-of-bounds/buffer/buffer_2d.cpp
  SYCL :: AddressSanitizer/out-of-bounds/buffer/buffer_3d.cpp
  SYCL :: AddressSanitizer/out-of-bounds/buffer/buffer_copy_fill.cpp
  SYCL :: AddressSanitizer/out-of-bounds/buffer/subbuffer.cpp
  SYCL :: AddressSanitizer/out-of-bounds/local/group_local_memory.cpp
  SYCL :: AddressSanitizer/out-of-bounds/local/local_accessor_basic.cpp
  SYCL :: AddressSanitizer/out-of-bounds/local/local_accessor_function.cpp
  SYCL :: AddressSanitizer/out-of-bounds/local/local_accessor_multiargs.cpp
  SYCL :: AddressSanitizer/out-of-bounds/local/multiple_source.cpp
  SYCL :: AddressSanitizer/out-of-bounds/private/multiple_private.cpp
  SYCL :: AddressSanitizer/out-of-bounds/private/single_private.cpp
  SYCL :: AddressSanitizer/use-after-free/quarantine-free.cpp
  SYCL :: AddressSanitizer/use-after-free/quarantine-no-free.cpp
  SYCL :: AddressSanitizer/use-after-free/use-after-free.cpp

[AllanZyne]

@sarnex, I can't reproduce this.

wget https://github.com/intel/intel-graphics-compiler/releases/download/igc-1.0.17537.20/intel-igc-core_1.0.17537.20_amd64.deb
wget https://github.com/intel/intel-graphics-compiler/releases/download/igc-1.0.17537.20/intel-igc-opencl_1.0.17537.20_amd64.deb
wget https://github.com/intel/compute-runtime/releases/download/24.35.30872.22/intel-level-zero-gpu-dbgsym_1.3.30872.22_amd64.ddeb
wget https://github.com/intel/compute-runtime/releases/download/24.35.30872.22/intel-level-zero-gpu-legacy1-dbgsym_1.3.30872.22_amd64.ddeb
wget https://github.com/intel/compute-runtime/releases/download/24.35.30872.22/intel-level-zero-gpu-legacy1_1.3.30872.22_amd64.deb
wget https://github.com/intel/compute-runtime/releases/download/24.35.30872.22/intel-level-zero-gpu_1.3.30872.22_amd64.deb
wget https://github.com/intel/compute-runtime/releases/download/24.35.30872.22/intel-opencl-icd-dbgsym_24.35.30872.22_amd64.ddeb
wget https://github.com/intel/compute-runtime/releases/download/24.35.30872.22/intel-opencl-icd-legacy1-dbgsym_24.35.30872.22_amd64.ddeb
wget https://github.com/intel/compute-runtime/releases/download/24.35.30872.22/intel-opencl-icd-legacy1_24.35.30872.22_amd64.deb
wget https://github.com/intel/compute-runtime/releases/download/24.35.30872.22/intel-opencl-icd_24.35.30872.22_amd64.deb
wget https://github.com/intel/compute-runtime/releases/download/24.35.30872.22/libigdgmm12_22.5.0_amd64.deb
wget https://github.com/oneapi-src/level-zero/releases/download/v1.17.44/level-zero-devel_1.17.44+u22.04_amd64.deb
wget https://github.com/oneapi-src/level-zero/releases/download/v1.17.44/level-zero_1.17.44+u22.04_amd64.deb

I installed the above packages, but igc failed to build device code (without enabling sanitizer).

Compilation from IR - skipping loading of FCL
Error! Incompatible interface in IGC: GT_SYS_INFO
Error! IGC initialization failure. Error code = -6
Build failed with error code: -6
Command was: ocloc -file /tmp/parallel_for_char-pvc-6694e1-76748f.spv -output_no_suffix -spirv_input -device pvc -device_options pvc -ze-intel-enable-auto-large-GRF-mode

Do you know any reasons? Thank you!

[AllanZyne]

Besides, according to CI log, "UR_RESULT_ERROR_UNSUPPORTED_FEATURE" means we use some APIs that level zero doesn't support.

Indeed, we use some L0 APIs that other e2e tests may unlikely use (e.g., VirtualMemory related APIs, Global Variable related APIs). Maybe the released gfx package contained some issues on these APIs.

[sarnex]

I can't repro it locally either, let me try running again.

[sarnex]

@AllanZyne Sorry, I finally investigated this and found the problem. It's only failing for the opencl:gpu run, level_zero:gpu works. It fails on urVirtualMemReserve, and it seems that is unimplemented in UR for the opencl adapter based on here.

So probably we should disable all these tests on opencl:gpu. Does that seem reasonable to you?

Thanks

[AllanZyne]

So probably we should disable all these tests on opencl:gpu. Does that seem reasonable to you?

Yes, it makes sense. We don't support opencl gpu.

[sarnex]

@AllanZyne PR to disable ocl gpu here #15620