[PATCH] D159375: [RISCV] Lower fixed vectors extract_vector_elt through stack at high LMUL

[Philip Reames via Phabricator via llvm-commits]

reames added a comment.

In D159375#4635530 <https://reviews.llvm.org/D159375#4635530>, @craig.topper wrote:

> Just to make sure I understand. For a full explode, we still handle the lower LMUL2 portion of the vector using slides, but the upper portion will use vector store plus scalar loads?

That's correct.  This isn't the goal per se, it just happens to fall out of the existing code structure and be "good enough".


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D159375/new/

https://reviews.llvm.org/D159375