llvm icon indicating copy to clipboard operation
llvm copied to clipboard

Intel staging area for llvm.org contribution. Home for Intel LLVM-based projects.

Results 962 llvm issues
Sort by recently updated
recently updated
newest added

As part of the change, also start using "inline namespace _V1" to allow possible future ABI-affecting changes.

Two changes were made: Driver passes an absolute path to an original source file through -main-file-name if integration footer is enabled. Clang CodeGen uses the file specified through -main-file-name to...

This PR creates the `local_accessor` class by aliasing the accessor class with `target::local`. The motivation behind this is that `target::local` has been deprecated in favour of `local_accessor` in SYCL2020. The...

The addition of the new OpenMP offloading driver path, which unbundles archives later in the toolchain is interfering with the device linking behaviors involving AMD GPU targets for SYCL offload...

This in non-functional changes, will only make the document more readable. Signed-off by: Yuan, Alex

1) remove unused c-tor of SYCLMemObjT: Details: include/CL/sycl/detail/sycl_mem_obj_t.hpp:79 2) remove unused getBufSizeForContext accepting cl_mem: Details: include/CL/sycl/detail/sycl_mem_obj_t.hpp:299

Removed from L0 plugin: 1) make_context(const std::vector &DeviceList, pi_native_handle NativeHandle) - Details: [source/backend/level_zero.cpp:65](https://github.com/intel/llvm/blob/784cfa13046e4874e1b8173e44231987bc7db2f8/sycl/source/backend/level_zero.cpp#L64) 2) make_queue(const context &Context, pi_native_handle NativeHandle) - Details: [source/backend/level_zero.cpp:93](https://github.com/intel/llvm/blob/784cfa13046e4874e1b8173e44231987bc7db2f8/sycl/source/backend/level_zero.cpp#L101)

The purpose of this PR is to fix an issue that leads to runtime exception when unnamed ESIMD and non ESIMD kernels are present.

This PR allows CXX stdlib funcs to be used for NVPTX backend. See https://github.com/intel/llvm/discussions/6379 llvm-test-suite test: https://github.com/intel/llvm-test-suite/pull/1112 It also adds the compiler flag `"-fbundle-no-offload-arch"`, which allows device code bundles to...