llvm
llvm copied to clipboard
Intel staging area for llvm.org contribution. Home for Intel LLVM-based projects.
Update the KernelArgsTest.KernelCopy unit test to use urEnqueueKernelLaunchWithArgsExp instead of urKernelSetArgValue.
There are long-standing known failures on the test for this urMemBufferCreate flag - https://github.com/intel/llvm/blob/sycl/unified-runtime/test/conformance/memory/urMemBufferCreate.cpp#L122 Essentially it looks like it only works on CPU devices. Relates to https://github.com/intel/llvm/issues/9789 but the workaround...
### Describe the bug https://github.com/intel/llvm/pull/20230 ### To reproduce checkout llvmspirv_pulldown branch and run the test on Intel GPU
### Describe the bug First identified in: https://github.com/intel/llvm/actions/runs/19515665783/job/55869141453?pr=20690. The newly added test [host_task_in_order_dependency.cpp](https://github.com/intel/llvm/pull/20690/files#diff-9fdab9fc208b022659edfd64da445e98e5f8dc01bbf4f6fe96a38f2201fec467) leaks L0 objects. This issue has only shown up on gen12 in testing and the test is...
This PR is created for testing SYCL E2E test that are failing with new offloading model, and will not be merged.
Level Zero adapter segfaults if older intel-level-zero-gpu is installed but not device is available
### Describe the bug Level Zero adapter segfaults if older intel-level-zero-gpu is installed but not device is available. ### To reproduce Command output omitted for readability where not relevant: ```...
duplicateNodes() is quite hot in graph.finalize() code path.
While resulting in extra restrictions to ensure not ABI-violations when passing data between host and device, that makes distribution of SYCL applications using SYCL RTC much easier, so we're willing...