Commit 8e732a31 authored by Artem Belevich's avatar Artem Belevich Committed by TensorFlower Gardener
Browse files

Prefer cubin over PTX when we launch CUDA kernels. 

Native GPU code, if we have it, should be preferred over JIT compilation of PTX.

PiperOrigin-RevId: 174110646
parent 2ccf3aba
Show whitespace changes
Inline Side-by-side
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Please to comment