It is possible that removal would be faster, though I haven't benchmarked that. I think that Kodi only uses this function for UI textures, so nothing performance-intensive. My main point was that the current MMX code without EMMS is basically broken (it breaks fmod() function and other code using x87 FPU). So, either my patch or yours will fix my problem. ________________________________ From: Ramiro Polla <ramiro.po...@gmail.com> Sent: Monday, July 1, 2024 1:46 PM To: FFmpeg development discussions and patches <ffmpeg-devel@ffmpeg.org> Cc: k3x-de...@outlook.com <k3x-de...@outlook.com> Subject: Re: [FFmpeg-devel] [PATCH v2] libswscale/x86/yuv2rgb: Add missing EMMS
On Wed, Jun 26, 2024 at 8:55 PM Ramiro Polla <ramiro.po...@gmail.com> wrote: > On Wed, Jun 26, 2024 at 8:03 PM Mario Hros <k3x-de...@outlook.com> wrote: > > Previous rewrite from inline assembly into nasm (commit e934194) missed the > > required EMMS instruction to bring the x87 FPU back into usable state. > > This needs to be done for 8-byte MMX or Extended MMX only. > > Sorry I didn't catch this thread earlier. I sent a patch to outright > remove the mmx/mmxext code (thread "swscale/yuv2rgb/x86: remove > mmx/mmxext yuv2rgb functions"): > https://lists.ffmpeg.org/pipermail/ffmpeg-devel/2024-June/329785.html > > The C code should be faster in most cases, or have very similar > performance to the mmx/mmxext code. Is this not the case for you? Mario, ping? If there are no more comments I'll go ahead and push the patch that removes mmx/mmxext. _______________________________________________ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-devel To unsubscribe, visit link above, or email ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".
