[FFmpeg-devel] Re: [PATCH] avcodec/aarch64/vvc: optimised alf_classify function 8/10/12bit of vvc codec for aarch64

2025-08-31 Thread George Zaguri via ffmpeg-devel
Hello Martin, I opened PR as you requested: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20377 I fixed most of the comments, except the one about decrement operation instead of increment. I have done increment of loops in less conditions as you told me, but I have logic based on index of loop which

[FFmpeg-devel] Re: [PATCH] avcodec/aarch64/vvc: optimised alf_classify function 8/10/12bit of vvc codec for aarch64

2025-08-30 Thread George Zaguri via ffmpeg-devel
Hello Martin, I just found a time to have a look at your comments, mostly I fixed it, additionally I optimised code a bit deeper because on Mac platform it showed me reduction in performance (turned on auto-vectorisation by default) + excluded unused code. You mentioned submition of patch as a PR,