Re: [RFC] Remove double-int.
Project / Subsystem
gcc / gcc
Date
2026-09-08
Proposer
Richard Biener <richard.guenther@gmail.com>
Source type
public_inbox
Consensus
Proposed
Sentiment
—/10
Technical tradeoffs
- • Removal of `double-int` simplifies code but may impact performance.
- • Adoption of `fixed_wide_int` avoids heap allocations but needs performance validation.
All attributes
- project
- gcc
- subsystem
- gcc
- patch_id
- —
- discussion_id
- CAFiYyc2OeQMg23k72G791ec3i56uQa=tkHPbeHCFKwOWDWMQxg@mail.gmail.com
- source_type
- public_inbox
- title
- Re: [RFC] Remove double-int.
- headline
- Remove double-int.
- tldr
- Discusses removing double-int in favor of wide-int, considering potential efficiency losses.
- proposer
- Richard Biener <richard.guenther@gmail.com>
- consensus
- Proposed
- outcome
- proposed
- sentiment_score
- —
- technical_tradeoffs
-
- • Removal of `double-int` simplifies code but may impact performance.
- • Adoption of `fixed_wide_int` avoids heap allocations but needs performance validation.
- series_id
- —
- series_role
- standalone
- series_parts
- []
- tags
-
- • gcc
- • type system
- • optimization
- url
- https://inbox.sourceware.org/gcc/CAFiYyc2OeQMg23k72G791ec3i56uQa=tkHPbeHCFKwOWDWMQxg@mail.gmail.com
- bugzilla_url
- —
- date
- 2026-09-08T00:00:00.000Z
Re: [RFC] Remove double-int.
This discussion revolves around removing the double-int C++ type in favor of fixed_wide_int. The proposer supports the removal, contingent on a thorough evaluation of potential performance degradations in terms of size and speed. The potential impact on various compiler components is also noted.