Quantcast
Channel: Intel® C++ Compiler
Viewing all articles
Browse latest Browse all 1616

Computation offload design decisions

$
0
0

So, I just read the article about How to offload computation to Intel(R) Graphics Technology and now I am wondering why the support for offloading to GPU was not done through extending existing __declspec(cpu_dispatch(...)) and __declspec(cpu_specific(...))?

For example:
__declspec(cpu_dispatch(generic, core_2_duo_sse4_1, core_4th_gen_avx, gfx)

And:
__declspec(cpu_specific(gfx))

For me it would make more sense to build upon existing CPU dispatch functionality. Maybe I am missing something, but I don't see how I could integrate existing CPU dispatch/targeting with this new GPU offload feature?


Viewing all articles
Browse latest Browse all 1616

Trending Articles



<script src="https://jsc.adskeeper.com/r/s/rssing.com.1596347.js" async> </script>