I have seen that the AMD APP SDK samples work on a machine that has only an Intel processor. How can this happen? How does the compiler target a different machine architecture? I don't need a set of Intel compilers to run code on an Intel processor?
I think that if we need to run the OpenCL application on specific hardware, I should (re) compile it using the device specifier compiler.
Where is my understanding wrong?
Firstly, OpenCL is built to work with processors and GPUs. You can compile and run the same source code on any type of device. However, it is very likely that the CPU code will not be optimal for the GPU and vice versa.
AMD H/W 7% - 14% x86/x64. AMD AMD Intel. AMD . , Intel , AMD, . .
OpenCL AMD APP SDK , AMD Intel AMD. Intel OpenCL SDK AMD x86 H/W.
OpenCL , , OpenCL , . , , , , .
OpenCL , .
SDK "" , , , , .
- , OpenCL, , Intel.
Intel , iCore C2D, SDK iCore, , , Intel , OpenCL , SDK .
, , , .
, AMD SDK x86 OpenCL. , Intel, . , C- x86: Intel AMD ( , x86).
- , , , user827992, , , AMD Intel. Intel OpenCL.
, (, ) . AMD SDK , NVIDIA, . , -, , , , , , .