[Mlir-commits] [mlir] [mlir][gpu] Lower the remaining subgroup_mma_elementwise ops to NVVM (PR #221288)

[Hung-Kuan Tseng]

Tim096 wrote:

@fabianmcg @grypp @kuhar this fixes an abort in `convert-gpu-to-nvvm` that `convert-vector-to-gpu` output reaches with a `subf` or `negf` epilogue; @simpel01 the `truncf` case from #182499 hits the same `llvm_unreachable` on the NVVM side, and the PTX ISA rules out a lowering for it, so it now fails to legalize instead. Would you mind taking a look when you have a moment? Thanks.


https://github.com/llvm/llvm-project/pull/221288