why use clEnqueueWriteBuffer()?

Fu_J_Intel — Tue, 07 Jun 2016 22:22:55 GMT

Hi Robert,

in your example modulatemain.cpp, I wonder why you use clEnqueueWriteBuffer(). I looks to me clEnqueueWriteBuffer() causes extra mem copy and thus slow down computing.

thanks,

Jeffrey

Robert_I_Intel — Tue, 07 Jun 2016 22:34:58 GMT

Hi Jeffrey,

Yes, you actually don't want to use it on our hardware. The right thing to do is described here: https://software.intel.com/en-us/articles/getting-the-most-from-opencl-12-how-to-increase-performance-by-minimizing-buffer-copies-on-intel-processor-graphics

Modulate samples don't show good host side practices: the device side is OK.

Sorry for the confusion!

topic in OpenCL* for CPU

why use clEnqueueWriteBuffer()?