Landing: 408162bbcde3
Project / Subsystem
gcc / tree-optimization
Date
2026-08-26
Author
Richard Biener
Commit
408162bbcde30c7d6672a20f96aa6dc7a1c5fa58
Source
github
Perf win
Yes
Breaking
No
All attributes
- project
- gcc
- subsystem
- tree-optimization
- patch_id
- —
- commit_hash
- 408162bbcde30c7d6672a20f96aa6dc7a1c5fa58
- source_type
- github
- headline
- tree-optimization: Avoid vect_get_num_copies on SLP inputs
- tldr
- Improves vectorization costing by using a new function for SLP inputs.
- author
- Richard Biener
- outcome
- committed
- performance_win
- true
- breaking_change
- false
- series_id
- —
- series_parts
- []
- tags
-
- • tree-optimization
- • vectorization
- • performance
- discussion_id_link
- —
- bugzilla_pr
- —
- date
- 2026-08-26T00:00:00.000Z
This commit enhances the vectorization process by avoiding the use of vect_get_num_copies on SLP (Sliding Window Vectorization) inputs during cost estimation. Instead, it introduces and uses vect_get_num_copies_for_invariant for costing, which accurately reflects the number of vector inputs generated by the transform phase and improves the precision of cost-based vectorization decisions.