unified-runtime
unified-runtime copied to clipboard
oneapi-src
As we discussed before, we intercepted buffer with usm for short-term solution. For long-term, we will draft a new ocl extension to allow user to create buffer with fixed address....
When configured without match files, the CTS was trying to run a command and its arguments as a single command, as it was represented by a single string variable. It...
I just went to change one line in a file, then found out the whole file changed since it was committed differently to my local system, and we don't do...
Introduces an extendable extension for kernel launch attributes. Currently only three are supported on the cuda backend only: - UR_EXP_LAUNCH_ATTRIBUTE_ID_CLUSTER_DIMENSION - UR_EXP_LAUNCH_ATTRIBUTE_ID_COOPERATIVE - UR_EXP_LAUNCH_ATTRIBUTE_ID_IGNORE (does nothing, but useful for testing)....
Defines `urKernelSuggestMaxCooperativeGroupCountExp` and `urEnqueueCooperativeKernelLaunchExp` to enable cooperative kernels with more than one work group.
This tests that the `urEnqueueEventsWaitWithBarrier` function functions as a barrier. That is, it blocks until all events prior to it in the queue are completed. Since this test may fail...
The interop import API has been adapted to cater to multiple external memory and semaphore handle types. This enables memory and semaphores to be imported by single functions irregardless of...
A set of new device info enums was [added to PI](https://github.com/intel/llvm/pull/13389/files#diff-952d2515bc7a8c91d37420f519ff726c0e6ce71f5a81bfd98e5c6fcfcaa3ada0R440) recently to support the [forward progress extension](https://github.com/intel/llvm/blob/sycl/sycl/doc/extensions/proposed/sycl_ext_oneapi_forward_progress.asciidoc). These do not need adapter implementations, but they will be removed as...
