On 2025-07-12 12:44 +0200, Niklas Haas wrote:
> From: Niklas Haas <g...@haasn.dev>
> 
> Provides a generic fast path for any operation list that can be decomposed
> into a series of memcpy and memset operations.
> 
> 25% faster than the x86 backend for yuv444p -> yuva444p
> 33% faster than the x86 backend for gray -> yuvj444p
> ---
>  libswscale/Makefile     |   1 +
>  libswscale/ops.c        |   2 +
>  libswscale/ops_memcpy.c | 132 ++++++++++++++++++++++++++++++++++++++++
>  3 files changed, 135 insertions(+)
>  create mode 100644 libswscale/ops_memcpy.c
> 
> diff --git a/libswscale/Makefile b/libswscale/Makefile
> index d3cb7bc555..0fed799542 100644
> --- a/libswscale/Makefile
> +++ b/libswscale/Makefile
> @@ -18,6 +18,7 @@ OBJS = alphablend.o                                     \
>         ops.o                                            \
>         ops_backend.o                                    \
>         ops_chain.o                                      \
> +       ops_memcpy.o                                     \
>         ops_optimizer.o                                  \
>         options.o                                        \
>         output.o                                         \
> diff --git a/libswscale/ops.c b/libswscale/ops.c
> index 2d8f9e6084..76423f631e 100644
> --- a/libswscale/ops.c
> +++ b/libswscale/ops.c
> @@ -28,8 +28,10 @@
>  #include "ops_internal.h"
>  
>  extern SwsOpBackend backend_c;
> +extern SwsOpBackend backend_murder;
>  
>  const SwsOpBackend * const ff_sws_op_backends[] = {
> +    &backend_murder,
>      &backend_c,
>      NULL
>  };
> diff --git a/libswscale/ops_memcpy.c b/libswscale/ops_memcpy.c
> new file mode 100644
> index 0000000000..ef4784faa4
> --- /dev/null
> +++ b/libswscale/ops_memcpy.c
[...]
> +
> +SwsOpBackend backend_murder = {
> +    .name    = "memcpy",
> +    .compile = compile,
> +};

Is this intentional?
While everything is name `memcpy` the variables are namend `_murder`?


  Alexander
_______________________________________________
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".

Reply via email to