Bandwidth Efficiency values in the Altera Dynamic Profiler for OpenCL

[Altera_Forum]

Hi everyone, 

 can somebody explain me the values of bandwidth efficiency in the Altera Dynamic Profiler for OpenCL? Like in the screens attached to this post, in a kernel that I'm working on I get an efficiency value of 1.5%. When, in reality, I read at ~6510 MB/s out of 12800 MB/s from each bank. What am I missing here? :confused:  Thanks, Nicola.

[Altera_Forum]

It depends on the memory access unit size you use and the number of memory load/store instructions.

[Altera_Forum]

ah thanks wzk6_3_8! 

 I also found this:   --- Quote Start ---  The overall efficiency is the percentage of total bytes, acquired from the global memory system, that the kernel program uses.  --- Quote End ---    So I guess I'm getting low efficiency values because the .aocx file contains multiple lead/store instructions.  Thanks, Nicola.

[Altera_Forum]

--- Quote Start ---  ah thanks wzk6_3_8!  I also found this:   So I guess I'm getting low efficiency values because the .aocx file contains multiple lead/store instructions.  Thanks, Nicola.  --- Quote End ---    Yes, the load/store instructions share the memory bandwidth.