[FFmpeg-devel] [PATCH] avfilter: Added siti filter

[Anton Khirnov]

Quoting Paul B Mahol (2022-02-15 10:10:17)
> On Tue, Feb 15, 2022 at 9:55 AM Anton Khirnov <anton at khirnov.net> wrote:
> 
> > Quoting Thilo Borgmann (2022-02-12 11:55:39)
> > > Am 31.01.22 um 12:55 schrieb James Almer:
> > > +static int config_input(AVFilterLink *inlink)
> > > +{
> > > +    // Video input data avilable
> > > +    AVFilterContext *ctx = inlink->dst;
> > > +    SiTiContext *s = ctx->priv;
> > > +    int max_pixsteps[4];
> > > +    size_t pixel_sz;
> > > +    size_t data_sz;
> > > +    size_t gradient_sz;
> > > +    size_t motion_sz;
> > > +
> > > +    const AVPixFmtDescriptor *desc =
> > av_pix_fmt_desc_get(inlink->format);
> > > +    av_image_fill_max_pixsteps(max_pixsteps, NULL, desc);
> > > +
> > > +    s->pixel_depth = max_pixsteps[0];
> > > +    s->width = inlink->w;
> > > +    s->height = inlink->h;
> > > +    pixel_sz = s->pixel_depth == 1 ? sizeof(uint8_t) : sizeof(uint16_t);
> > > +    data_sz = s->width * pixel_sz * s->height;
> > > +
> > > +    s->prev_frame = av_malloc(data_sz);
> > > +
> > > +    gradient_sz = (s->width - 2) * sizeof(double) * (s->height - 2);
> > > +    s->gradient_matrix = (double*)av_malloc(gradient_sz);
> > > +
> > > +    motion_sz = s->width * sizeof(double) * s->height;
> > > +    s->motion_matrix = (double*)av_malloc(motion_sz);
> >
> > useless casts
> >
> > > +
> > > +    if (!s->prev_frame || ! s->gradient_matrix || !s->motion_matrix) {
> > > +        av_freep(&s->prev_frame);
> > > +        av_freep(&s->gradient_matrix);
> > > +        av_freep(&s->motion_matrix);
> >
> > You don't need to free them on failure, that will be done in uninit. But
> > you should free them at the beginning of this function, because
> > config_input can be called multiple times.
> >
> 
> In which scenarios? When its called, uninit is also called before.
> Was thinking to add better support for frame w/h changes when midstream
> filtering....

Actually I'm not sure it currently will be called more than once, but
IIRC the code was supposed to assume it, precisely for future parameter
change support.

-- 
Anton Khirnov