KernelGen - GPU Technology Conference
KernelGen - GPU Technology Conference
KernelGen - GPU Technology Conference
- No tags were found...
Create successful ePaper yourself
Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.
<strong>KernelGen</strong> execution workflow.Scenario 1: main kernel approached a point where another <strong>GPU</strong> kernel can be launched (for parallelloop). In this case, another kernel is launched via host callback (could be replaced withdynamic parallelism on Kepler sm_35). All <strong>GPU</strong> data is shared (.local data is disabled).CPUversionswitchcompilemain kernelcallback: requested loopkernel JIT-compile&launch<strong>GPU</strong>stream #2(many threads)<strong>GPU</strong>stream #1(single thread)launchmain kernel44 / 75