Question

Anyone knows the instruction cache size for Nvidia GPU (both Fermi and Kepler)? I have tried to search a lot, but not too much such info is provided by Nvidia. There's one paper describing how they benchmark Nvidia GT200 GPU. And there's also one work describe the data cache in Fermi GPU, but it doesn't analyze the instruction cache.

No correct solution

OTHER TIPS

I can not access to any of the links that you provided. Please read this paper which performs micro benchmarking on GT200 GPU and reports the size of instruction caches (http://www.stuffedcow.net/files/gpuarch-ispass2010.pdf)

Licensed under: CC-BY-SA with attribution
Not affiliated with StackOverflow
scroll top