This is an archive of the discontinued LLVM Phabricator instance.

[RISCV][NFC] Add some tail agnostic tests for nomask operations.
ClosedPublic

Authored by khchen on Feb 16 2022, 9:50 PM.

Details

Summary

Improve test coverage for tail agnostic nomask vslidedown/up, vmv.s.x
vfmv.s.f and vcompress.

Diff Detail

Event Timeline

khchen created this revision.Feb 16 2022, 9:50 PM
khchen requested review of this revision.Feb 16 2022, 9:50 PM
rogfer01 accepted this revision.Feb 16 2022, 11:20 PM

LGTM. Thanks @khchen!

llvm/test/CodeGen/RISCV/rvv/unmasked-ta.ll
1124

For those who care about RV32V I wonder if vslidedown.vi ..., 2 followed by two vslide1up.vx could be used here so we avoid the load and the merge (maybe it is not faster after all).

No need to fix it in this change, of course.

This revision is now accepted and ready to land.Feb 16 2022, 11:20 PM
craig.topper added inline comments.Feb 16 2022, 11:25 PM
llvm/test/CodeGen/RISCV/rvv/unmasked-ta.ll
1124

Do you need a slidedown? Couldn't you do two SEW=32 vslide1ups with a vl of 2?

This revision was landed with ongoing or failed builds.Feb 17 2022, 4:46 PM
This revision was automatically updated to reflect the committed changes.