Landing: 53c4e6602ac7

Project / Subsystem

gcc / risc-v

Date

2026-07-19

Author

Pan Li

Commit

53c4e6602ac7bb23cc6f873e59578840f15575db

Source

github

Perf win

Yes

Breaking

No

All attributes

project
gcc
subsystem
risc-v
patch_id
commit_hash
53c4e6602ac7bb23cc6f873e59578840f15575db
source_type
github
headline
RISC-V: Enable register overlap for vfwcvt.f.f.v vector instruction.
tldr
RISC-V backend now allows register overlap for the vfwcvt.f.f.v vector instruction.
author
Pan Li
outcome
committed
performance_win
true
breaking_change
false
series_id
series_parts
[]
tags
  • risc-v
  • rvv
  • vector
  • optimization
discussion_id_link
bugzilla_pr
date
2026-07-19T00:00:00.000Z

This commit updates the RISC-V backend to permit register overlap for the vfwcvt.f.f.v vector instruction. Similar to how vsext.vf8 and overlay.vf8 handle register reuse, this change allows the compiler to more efficiently schedule vector operations. By leveraging the Wvr constraint in the machine description, the backend can now better utilize available registers when generating code for this specific widen-converting floating-point vector instruction.