Correct level of indirection used in KernelSetArgPointer calls.
Also attempt to clarify the wording around this a bit.
Addresses #558
LLVM testing https://github.com/intel/llvm/pull/12270
I was able to properly test this on more than just cuda this morning and I can confirm this will need a small patch for the cuda and opencl adapters when it gets into intel-llvm. Maybe merging this should wait until the adapters move in here and they can be patched simultaneously?
Maybe merging this should wait until the adapters move in here and they can be patched simultaneously?
That makes sense, convert this to draft so we don't merge it unintentionally.
Codecov Report
Attention: 5 lines in your changes are missing coverage. Please review.
Comparison is base (
fb83446) 15.72% compared to head (1c7d0c7) 15.72%.
| Files | Patch % | Lines |
|---|---|---|
| test/conformance/kernel/urKernelSetArgPointer.cpp | 0.00% | 5 Missing :warning: |
:exclamation: Your organization needs to install the Codecov GitHub app to enable full functionality.
Additional details and impacted files
@@ Coverage Diff @@
## main #805 +/- ##
=======================================
Coverage 15.72% 15.72%
=======================================
Files 223 223
Lines 31475 31475
Branches 3557 3557
=======================================
Hits 4951 4951
Misses 26473 26473
Partials 51 51
:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.
@oneapi-src/unified-runtime-native-cpu-write @mmoadeli would be great to get your reviews on this.
@oneapi-src/unified-runtime-cuda-write @oneapi-src/unified-runtime-hip-write would appriciate a review on this, the CUDA/HIP changes are very small.
I've removed ready to merge since the CUDA checks aren't passing, please add it again once they go green.