Landing: 45c9868f1b20
Project / Subsystem
gcc / gcc/x86
Date
2026-08-25
Author
Roger Sayle
Commit
45c9868f1b20481f4eebdcce8fffc7d61724095b
Source
github
Perf win
Yes
Breaking
No
All attributes
- project
- gcc
- subsystem
- gcc/x86
- patch_id
- —
- commit_hash
- 45c9868f1b20481f4eebdcce8fffc7d61724095b
- source_type
- github
- headline
- x86 SSE vector initialization improved for V4SF type.
- tldr
- Optimized x86 SSE code generation for V4SF vector initialization, fixing a performance regression.
- author
- Roger Sayle
- outcome
- committed
- performance_win
- true
- breaking_change
- false
- series_id
- —
- series_parts
- []
- tags
-
- • x86
- • sse
- • optimization
- • compiler
- discussion_id_link
- —
- bugzilla_pr
- —
- date
- 2026-08-25T00:00:00.000Z
This commit addresses a performance regression in x86 SSE vector initialization for the V4SF type. The previous update inadvertently caused issues with implicit zero extension when a value was already in a register. The fix introduces improved register preferencing and utilizes idioms like pslldq/psrldq for more efficient initialization, reducing inter-unit moves and matching LLVM’s approach.