GPUOcelot
|
#include <cal.h>
CAL function information
Slow mode
Is this the max num active wavefronts per SIMD
LDS size available to the program
LDS size used by the program
Maximum number of scratch regs needed
number of GPRs available to the program
number of GPRs used by the program
Number of shared GPRs including ones used by SC
Number of shared GPRs
Flattend umber of threads per group
x dimension of numThreadPerGroup
y dimension of numThreadPerGroup
z dimension of numThreadPerGroup
Number of wavefronts per SIMD
Need to set up buffer for info on number of thread groups?
stack size availabe to the program
stack size use by the program
Total number of thread groups
number of threads per wavefront.