llvm
llvm copied to clipboard
Intel staging area for llvm.org contribution. Home for Intel LLVM-based projects.
Currently the SYCL library and its constituents build both a release build and a debug build (/MDd) on Windows, no matter what the build type is. This PR changes the...
Introduce initial_threshold, maximum_size, read_only, and zero_init runtime properties conforming to the SYCL properties extension, and apply these properties to the creation of memory pools. This PR is a draft until...
This PR deprecates the usage of `-fsycl-dump-device-code` option, making this option backend-agnostic, extending its functionality beyond just SYCL to include OpenMP and other offloading models in the future, with the...
A selection of OneDNN benchmarks added to the benchmark suite
First of multiple PRs to remove FPGA support from DPC++ (See #16929). This PR removes support for FPGAs from the sycl E2E tests. Note that instead of lit quitting with...
This is a list of entry points and information queries that a run of the following compute benchmarks runs without Offload: * api_overhead_benchmark * memory_benchmark * miscellaneous_benchmark * multithread_benchmark *...
This patch experiments with moving standard library math built-ins from libdevice into headers. This is based on the way clang handles this for CUDA and HIP. In these languages you...