Structs vs. Arrays in OpenCL

Intel® Quartus® Prime Software

Intel® Quartus® Prime Design Software, Design Entry, Synthesis, Simulation, Verification, Timing Analysis, System Design (Platform Designer, formerly Qsys)

Structs vs. Arrays in OpenCL

1,200 Views

Hi Everyone,

I'm new to OpenCL and trying to learn from the official Altera OpenCL design examples. Here is one question that might be naive, but I just could not find an answer (tried to search for a similar thread too):

In the FFT(1D) design example (https://www.altera.com/support/support-resources/design-examples/design-software/opencl/fft-1d.html), a struct of 8 float2 members is defined in fft_8.cl:

typedef struct {

float2 i0;

float2 i1;

float2 i2;

float2 i3;

float2 i4;

float2 i5;

float2 i6;

float2 i7;

} float2x8;

So what is the point of using the struct (which forces related code to be "hardcoded"), instead of using an array float2[8] here? Does it increase the efficiency of compilation / execution? It would be great if the source of that info could be provided as well.

Thanks a lot in advance, and have a great weekend!

-Tian

Link Copied

0 Replies

Community support is provided during standard business hours (Monday to Friday 7AM - 5PM PST). Other contact methods are available here.

Intel does not verify all solutions, including but not limited to any file transfers that may appear in this community. Accordingly, Intel disclaims all express and implied warranties, including without limitation, the implied warranties of merchantability, fitness for a particular purpose, and non-infringement, as well as any warranty arising from course of performance, course of dealing, or usage in trade.

For more complete information about compiler optimizations, see our Optimization Notice.