unified-runtime
unified-runtime copied to clipboard
Now UR already implemented API "urProgramGetGlobalVariablePointer", so we can use it to query the size of device globals and remove "__AsanDeviceGlobalCount". SYCLOS Part: https://github.com/intel/llvm/pull/13678
This change moves all legacy queue code to a seprate class: ur_queue_handle_legacy_t_. and uses pointer to this class as a parameter in all internal functions (instead of using ur_queue_handle_t). ur_queue_handle_t...
This showed up in the UR CTS, it was returning a vector instead of the expected C array.
Reports `cl_khr_fp16` as supported on Native CPU. Corresponding UR PR: https://github.com/intel/llvm/pull/13829
This PR is just to make sure there errors are reported accordingly if new fails to allocate the requested space when creating UR event object from a native event. inte/llvm...
This PR changes the returned error code for exiting the kernel launch entry point in CUDA when exceeding the maximum available registers for execution on the SM. Previously we were...
Change unordered_map to vector as number of entries is expected to be low (1 in common case). Also, do not store mapping between subdevices and the kernel handles. Instead, just...
These were removed from the spec a while ago so this is dead code.
This PR: * Adds CTS tests for different image types. * Adds CTS tests for different image formats * Defines the primary and optional supported image formats. * Deletes redundant...