llvm
llvm copied to clipboard
Intel staging area for llvm.org contribution. Home for Intel LLVM-based projects.
The `atomic_memory_order_seq_cst.cpp` test only checks `info::device::max_mem_alloc_size` to limit the maximum allocation. It does not check the actual available on device. This lead to allocating exceeding amount of memory that I...
### Describe the bug Both #14273 and #14808 contain functional changes, but no tests to cover them which is a violation of our [contribution guildelines](https://github.com/intel/llvm/blob/sycl/sycl/doc/developer/ContributeToDPCPP.md#tests-development) ### To reproduce _No response_...
I'm working on adding GPU reset support for AMD/NVIDIA to hopefully improve runner stability. In order to run the commands to reset the GPU, we need `sudo`. We also [need](https://github.com/intel/llvm/blob/sycl/.github/workflows/sycl-linux-run-tests.yml#L160)...
### Describe the bug https://github.com/intel/llvm/actions/runs/10181110052/job/28161806914 You can't see the problem from the log, but I manually verified it's these two tests from logging into the runners ### To reproduce _No...
This commit makes it possible to copy-construct device_global variables if they do not have the device_image_scope property. The restriction on device_image_scope is due to static construction not being allowed in...
**Is your feature request related to a problem? Please describe** Compiling SYCL code with options like `-fprofile-instr-generate` `-fcoverage-mapping` produces coverage information that cannot be used to produce coverage reports. For...
Replaced __spirv_SubgroupShuffleINTEL, __spirv_SubgroupShuffleXorINTEL, __spirv_SubgroupShuffleDownINTEL and __spirv_SubgroupShuffleUpINT with __spirv_GroupNonUniformShuffle... generic versions.
### Describe the bug The ESIMD/mask_expand_load.cpp test has been observed to fail on unrelated changes (https://github.com/intel/llvm/pull/13817) in Gen12 Windows pre-commit CI: ``` ******************** TEST 'SYCL :: ESIMD/mask_expand_load.cpp' FAILED ******************** Exit...
Add vectorized_ternary and vectorized_with_pred. Update relu and vectorized_binary. Signed-off-by: Tang, Jiajun [email protected]
When using -fsycl-host-compiler, be sure that we only use the available PATH settings to find the compiler given (if given without a full PATH qualifier). This allows for only known...