Page MenuHomePhabricator

[RISCV] Optimize vp.load with an all-ones mask
ClosedPublic

Authored by frasercrmck on Tue, Nov 2, 7:54 AM.

Details

Summary

Similar to D110206, this patch optimizes unmasked vp.load intrinsics to
avoid the need of a vmset instruction to set the mask. It does so by
selecting a riscv_vle intrinsic rather than a riscv_vle_mask intrinsic.

Diff Detail