Re:Intel proposes XeGPU dialect for LLVM MLIR
by ColonelPhantom on Monday December 18, 2023 @05:33AM
(#38680967)
It looks like offload is mostly about generic abstractions for accelerator devices. XeGPU on the other hand seems to be a way of encoding Xe-specific hardware features/instructions, so pretty different? I don't see how these are anything but orthogonal.
[ Parent
| on HN ]
There are no comments on this comment yet.