Merge pull request #18222 from stonea/gpu_streamBlockSizeAndIdxComputation
Get streamPrototype GPU test to operate on vector
Get streamPrototype GPU test to operate on vector. Specifically by: (1) having the kernel launch use a block size equal to the difference between the upper and lower bounds of the loop that we have designated to be extracted into a GPU kernel, and (2) adding IR to compute the index to operate on in the extracted kernel.
For (1) see generateBlockSizeComputation().
For (2) see generateIndexComputation().
[Reviewed by @e-kayrakli]
Modified Files: M compiler/optimizations/deadCodeElimination.cpp