Landing: 122a07880842
Project / Subsystem
gcc / gcc/tree-vect
Date
2026-02-25
Author
Richard Biener
Commit
122a0788084209b20372d8f9b6d440fb6443752d
Source
github
Perf win
No
Breaking
No
All attributes
- project
- gcc
- subsystem
- gcc/tree-vect
- patch_id
- —
- commit_hash
- 122a0788084209b20372d8f9b6d440fb6443752d
- source_type
- github
- headline
- Improve VIEW_CONVERT_EXPR Handling in vect_get_operand_map
- tldr
- The compiler now handles VIEW_CONVERT_EXPR nodes more cleanly during SLP vectorization by integrating the logic into `vect_get_operand_map`.
- author
- Richard Biener
- outcome
- committed
- performance_win
- false
- breaking_change
- false
- series_id
- —
- series_parts
- []
- tags
-
- • optimization
- • vectorization
- • SLP
- discussion_id_link
- —
- bugzilla_pr
- —
- date
- 2026-02-25T00:00:00.000Z
This commit improves the handling of VIEW_CONVERT_EXPR nodes within the vect_get_operand_map function. By integrating the handling of these expressions directly into vect_get_operand_map, the code becomes more streamlined, and some minor improvements are applied, which may lead to more robust and efficient vectorization.