[x265] [PATCH 0/3] AArch64: Clean up and optimize Neon primitives

[chen]

Thank for the patches and improve algorithm on calc_energy_8x8, it looks good to me




Regards,
Chen




At 2025-05-01 02:16:58, "Li Zhang" <li.zhang2 at arm.com> wrote:
>Hello,
>
>This patch series cleans up and optimizes the SATD/SA8D primitives using
>Neon intrinsics, removes slower assembly versions, simplifies
>psyCost_pp_neon, and removes redundant __APPLE__ guards for better
>cross-platform behavior.
>
>Many thanks,
>Li
>
>Li Zhang (3):
>  AArch64: Clean up satd/sa8d functions
>  AArch64: Simplify SBD and HBD psyCost_pp_neon
>  AArch64: Remove redundant __APPLE__ guards
>
> source/common/aarch64/asm-primitives.cpp |   82 +-
> source/common/aarch64/mem-neon.h         |    6 +-
> source/common/aarch64/pixel-prim.cpp     | 1551 ++++++++++++----------
> source/common/aarch64/pixel-util-sve.S   |  258 ----
> source/common/aarch64/pixel-util.S       | 1039 ---------------
> 5 files changed, 840 insertions(+), 2096 deletions(-)
>
>--
>2.39.5 (Apple Git-154)
>
>_______________________________________________
>x265-devel mailing list
>x265-devel at videolan.org
>https://mailman.videolan.org/listinfo/x265-devel
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/x265-devel/attachments/20250501/e7d17a0d/attachment.htm>