Is there an IL equivalent of the get_global(local)_size(0) command from OpenCL? Thanks! P.S. Tried to get .il files by exporting GPU_DUMP_DEVICE_KERNEL=3, but it didn't work. Any idea why?
There are similar IL registers, but they are not equivalent in all cases. The registers are vAbsTid, vThreadGrpId, and vTidInGrp an they can be found in the Intermediate Language spec.
Micah, I am aware of thread ID registers, but I would like to know in my kernel the total number of threads without using e.g. constant memory to provide that information to the kernel.
