Adam Paszke 4051ac2a2f [Mosaic GPU] Only call kernel initializer from inside a custom call ...

XLA:GPU custom call design is far from ideal, as there's apparently no way to figure
out the CUDA context that will be used to run an HLO module before the custom call is
first called. So, we can't preload the kernel onto the GPU, or else we'll get invalid
handle errors due to the load and launch happening in different CUDA contexts...

Also fix up build_wheel.py to match the rename of the runtime lib.

PiperOrigin-RevId: 629401858

2024-04-30 07:10:05 -07:00

..

_mlir_libs

[Mosaic GPU] Only call kernel initializer from inside a custom call

2024-04-30 07:10:05 -07:00

BUILD.bazel

Update GPU and NVGPU MLIR bindings to match upstream MLIR changes

2024-04-25 11:41:19 +00:00