[llvm-branch-commits] [llvm] [RISCV] Support i64 accumulator for Zvdot4a8i partial reductions (PR #215517)

[Craig Topper via llvm-branch-commits]

topperc wrote:

> lowerPARTIAL_REDUCE_MLA lowers the i64 case via an i32 vdot4a*
followed by a widening partial.reduce.add, which the generic
expander turns into extract_subvector + add.

This isn't what the patch does. It directly emits the extract_subvector+add

https://github.com/llvm/llvm-project/pull/215517