Searched refs:offload_n (Results 1 – 12 of 12) sorted by relevance
224 In this case, we can leverage tf::cudaFlow::offload_until or tf::cudaFlow::offload_n248 cf.offload_n(M);262 At the last line of the %cudaFlow closure, we call <tt>cf.offload_n(M)</tt> to ask the executor to 281 …K | M | CPU Sequential | CPU Parallel | GPU (conditional taksing) | GPU (using offload_n) |293 We can see that using the built-in predicate, tf::cudaFlow::offload_n,
372 void offload_n(size_t N);703 inline void syclFlow::offload_n(size_t n) { in offload_n() function in tf::syclFlow
49 void offload_n() { in offload_n() function87 cf.offload_n(times+1); in offload_n()127 offload_n<tf::cudaFlow>();131 offload_n<tf::cudaFlowCapturer>();179 cf.offload_n(times); in join()
103 cf.offload_n(9); in standalone()1419 cf.offload_n(100); in __anonf99ceec03302()
1026 void offload_n(size_t n);1235 inline void cudaFlowCapturer::offload_n(size_t n) { in offload_n() function in tf::cudaFlowCapturer
367 void offload_n(size_t N);1647 inline void cudaFlow::offload_n(size_t n) { in offload_n() function in tf::cudaFlow
72 flow.offload_n(3);
409 cf.offload_n(M); in gpu_predicate()
175 cf.offload_n(10); // offload the cudaFlow capturer and run it 10 times
224 sf.offload_n(10); // offload the syclFlow and run it 10 times
252 cf.offload_n(10); // offload the cudaFlow and run it 10 times
1421 inline void cudaFlow::offload_n(size_t n) { in offload_n() function in tf::cudaFlow