Forum Discussion

Altera_Forum's avatar
Altera_Forum
Icon for Honored Contributor rankHonored Contributor
10 years ago

Bad Performance of Host to OpenCL Memory Transfers

I'm working with the Socrates II SoC Cyclone V board and am just getting started. After trying a few basic examples I noticed very low performance for an integer division test. It was almost as slow ...
Altera_Forum's avatar
Altera_Forum
Icon for Honored Contributor rankHonored Contributor
10 years ago

I did some more tests, here are the results:

Memory read test kernel (Read array of integers and compute XOR): ~2.1GB/s

Memory write test kernel (Write consecutive integers into an array): ~2.0GB/s

Host memory test (memcpy): ~1.5GB/s

To clarify:

The extremely low performance is caused by the clEnqueueReadBuffer and clEnqueueWriteBuffer calls. These calls block even though I pass CL_FALSE to the blocking_write/blocking_read parameter. CPU usage is also high during these operations. Is this simply an ALTERA OpenCL bug?