What is the most effective way to copy subsets of arrays between local and global memory?

[WWP00]

For example, suppose we have a local memory array:

float local[10];

And a much larger, global memory array. Would we copy like:

int memStart = 50;

for (int i = 0; i < 10; ++i)

         local[i] = globalMem[memStart + i];

Or should we use #pragma unroll for this copy, to avoid making the loop take one clock cycle per copy? Or is there some other recommended way to move array data between local and global memory?

[KennyTan_Altera]

Hi, You may found information here: https://stackoverflow.com/questions/20616844/how-to-transfer-array-data-in-vhdl I would suggest you post it to there as it involve IEEE std. Thanks

[WWP00]

Sorry, I meant in Intel's OpenCL for FPGAs, not in RTL.

[KennyTan_Altera]

You can refer to https://www.intel.com/content/dam/www/programmable/us/en/pdfs/literature/hb/opencl-sdk/archives/aocl-best-practices-guide-15.1.pdf page 1-47.

Transferring Loop-Carried Dependency to Local Memory 

For more queries, can you put this to the correct category fall under open cl? https://forums.intel.com/s/topic/0TO0P0000001AUUWA2/intel-high-level-design

Thanks