tag:blogger.com,1999:blog-5248884563614062837.post3282017415728315537..comments2024-01-19T18:37:59.981+05:30Comments on CS-Tech-Era: Two Dimensional (2D) Image Convolution in CUDA by Shared & Constant Memory: An Optimized wayYogesh Desaihttp://www.blogger.com/profile/05717523093051221408noreply@blogger.comBlogger2125tag:blogger.com,1999:blog-5248884563614062837.post-72273873768821811532017-05-05T08:07:16.894+05:302017-05-05T08:07:16.894+05:30Well, maybe I've got that wrong;
"Concur...Well, maybe I've got that wrong;<br /><br />"Concurrent host execution is facilitated through asynchronous library functions that return control to the host thread before the device completes the requested task."<br /><br />http://docs.nvidia.com/cuda/cuda-c-programming-guide/index.html#asynchronous-concurrent-executionJupiterhttps://www.blogger.com/profile/13008508862847561845noreply@blogger.comtag:blogger.com,1999:blog-5248884563614062837.post-79489248365549839902017-05-05T07:56:57.477+05:302017-05-05T07:56:57.477+05:30I don't think that second synchthreads is nece...I don't think that second synchthreads is necessary. It's not like the kernel is going to return before all the threads are done.Jupiterhttps://www.blogger.com/profile/13008508862847561845noreply@blogger.com