llvm
llvm copied to clipboard
intel
Intel staging area for llvm.org contribution. Home for Intel LLVM-based projects.
Median does not respond very fast to changes in performance, and thus is not a suitable metric to be used for regression checking. This PR implements an option to use...
Addresses https://github.com/intel/llvm/issues/17891: Given recent influx of issues related to windows with debug flags enabled (e.g. https://github.com/intel/llvm/pull/16802 and https://github.com/intel/llvm/pull/18400), this PR runs E2E tests on a debug build on windows to...
Update UR adapter due to changes in https://github.com/llvm/llvm-project/pull/143901
Initial device-side code generation support for functions declared with the `sycl_kernel_entry_point` was recently added to upstream Clang, but did not take effect due to downstream customizations. This change adjusts the...
The runtime tries to build a kernel again if compilation fails. But if UR returns a memory error the attempt counter was not compared against the maximum number of attempts,...
The SYCLVirtualFunctionsAnalysis pass was crashing (calling `llvm_unreachable`) when it encountered a non-constant non-global use of an indirectly callable function. Instead of that just ignore any user we don't know how...
Currently for command buffers we set submission time equal to start time if the recorded submit time is greater than the start time. This is needed because the submission time...
AsanInterceptor::prepareLaunch has this call, which aligns with msan and tsan.
urDeviceGetInfo is now able to retrieve the max memory bandwidth.
