Commit b565a1d5 authored by Artem Belevich's avatar Artem Belevich Committed by TensorFlower Gardener
Browse files

GPU JIT improvements. 

* Use ptxas to compile generated PTX.
* Run PTX compilations in parallel.
* Cache results of PTX compilation.

PiperOrigin-RevId: 174921332
parent 8c88be0d
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