Landing: 50e5bb1da91a
Project / Subsystem
gcc / gcc/i386
Date
2026-07-19
Author
Dipesh Sharma
Commit
50e5bb1da91a1e2025a1d040eddc95d2dbaa9148
Source
github
Perf win
Yes
Breaking
No
All attributes
- project
- gcc
- subsystem
- gcc/i386
- patch_id
- —
- commit_hash
- 50e5bb1da91a1e2025a1d040eddc95d2dbaa9148
- source_type
- github
- headline
- Add AVX10/v2 FP32 to FP8 Single-Source RTO converts.
- tldr
- Enables FP32 to FP8 conversions using AVX10/v2 intrinsics with Single-Source Round-To-Odd (RTO) behavior.
- author
- Dipesh Sharma
- outcome
- committed
- performance_win
- true
- breaking_change
- false
- series_id
- —
- series_parts
- []
- tags
-
- • x86
- • AVX
- • AI
- • floating-point
- discussion_id_link
- —
- bugzilla_pr
- —
- date
- 2026-07-19T00:00:00.000Z
This commit introduces AVX10/v2 intrinsics for converting FP32 (single-precision float) to FP8 (BF8/HF8) formats, specifically supporting Single-Source Round-To-Odd (RTO) behavior. It adds support for 128, 256, and 512-bit vectors, including masked operations. New intrinsic definitions, built-in descriptions, and machine code patterns are included, along with test cases, to enable these specialized conversions for AI and HPC workloads on compatible x86 hardware.