This is an archive of the discontinued LLVM Phabricator instance.

[RISCV] Implement codegen patterns for RVP ALU operations
Needs ReviewPublic

Authored by Jim on May 13 2021, 10:14 PM.

Details

Summary

Add codegen patterns for vector add, sub, and, or and xor operations.

Diff Detail

Event Timeline

Jim created this revision.May 13 2021, 10:14 PM
Jim requested review of this revision.May 13 2021, 10:14 PM
Herald added a project: Restricted Project. · View Herald TranscriptMay 13 2021, 10:14 PM
Jim updated this revision to Diff 349825.Jun 4 2021, 5:03 AM

Add support for vector and, or and xor operations and
implement getPreferredVectorAction hook to widen v4i8 and v2i16 to v8i8 and v4i16 for RV64P.

Jim retitled this revision from [RISCV] Implement codegen patterns for add8, add16, sub8 and sub16 to [RISCV] Implement codegen patterns for RVP ALU operations.Jun 4 2021, 5:07 AM
Jim edited the summary of this revision. (Show Details)
Jim updated this revision to Diff 350222.Jun 7 2021, 3:07 AM

Add codegen patterns for logic operation with immediate.

Jim updated this revision to Diff 352116.Jun 15 2021, 6:25 AM

Move tests out of rvp directory.