Skip to content

[pull] master from fraunhoferhhi:master#163

Merged
pull[bot] merged 6 commits into
chenm001:masterfrom
fraunhoferhhi:master
Jun 5, 2026
Merged

[pull] master from fraunhoferhhi:master#163
pull[bot] merged 6 commits into
chenm001:masterfrom
fraunhoferhhi:master

Conversation

@pull

@pull pull Bot commented Jun 5, 2026

Copy link
Copy Markdown

See Commits and Changes for more details.


Created by pull[bot] (v2.0.0-alpha.4)

Can you help keep this open source service alive? 💖 Please sponsor : )

micro-arm and others added 6 commits June 4, 2026 15:20
Add Neon implementation of `SampleAdaptiveOffset::offsetBlock` covering
five sub-cases:
- SAO_TYPE_EO_0
- SAO_TYPE_EO_90
- SAO_TYPE_EO_135
- SAO_TYPE_EO_45
- SAO_TYPE_BO

Wire `initSampleAdaptiveOffsetARM` to install the Neon kernel in the
`offsetBlock` function pointer.

Make `offsetBlock_core` public so the Neon path can use it as a
fallback.

This implementation improves performance by 51% compared with the
existing SIMDe implementation when benchmarked on a Neoverse V2 using
LLVM 22.
Add unit tests for PelBufferOps::reco functions.
Add Neon implementation for PelBufferOps::reco with widths that are
multiples of eight.

This new implementation improves performance by approximately 25%
compared to the SIMDe version when benchmarked on a Neoverse V2 with
LLVM 22.
Arm: Add Neon implementation of SampleAdaptiveOffset::offsetBlock
Add unit tests for PelBufferOps::reco functions
Arm: Add Neon implementation for reco
@pull pull Bot locked and limited conversation to collaborators Jun 5, 2026
@pull pull Bot added the ⤵️ pull label Jun 5, 2026
@pull pull Bot merged commit fb50713 into chenm001:master Jun 5, 2026
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants