X-Git-Url: https://git.sesse.net/?a=blobdiff_plain;f=libavfilter%2Fvf_gblur.c;h=6ef261478fae455cfaec5fcc3bfd566dc3ba6ceb;hb=db99b32a1bfc35077933a068a09a1a8dc695644e;hp=f77a3fffc383a456e7202f1b5ccec66b931145a7;hpb=c7488f746154b5dcd70f8a3bef9a9fa5c42ac595;p=ffmpeg diff --git a/libavfilter/vf_gblur.c b/libavfilter/vf_gblur.c index f77a3fffc38..6ef261478fa 100644 --- a/libavfilter/vf_gblur.c +++ b/libavfilter/vf_gblur.c @@ -30,30 +30,10 @@ #include "libavutil/pixdesc.h" #include "avfilter.h" #include "formats.h" +#include "gblur.h" #include "internal.h" #include "video.h" -typedef struct GBlurContext { - const AVClass *class; - - float sigma; - float sigmaV; - int steps; - int planes; - - int depth; - int planewidth[4]; - int planeheight[4]; - float *buffer; - float boundaryscale; - float boundaryscaleV; - float postscale; - float postscaleV; - float nu; - float nuV; - int nb_planes; -} GBlurContext; - #define OFFSET(x) offsetof(GBlurContext, x) #define FLAGS AV_OPT_FLAG_VIDEO_PARAM|AV_OPT_FLAG_FILTERING_PARAM @@ -72,6 +52,28 @@ typedef struct ThreadData { int width; } ThreadData; +static void horiz_slice_c(float *buffer, int width, int height, int steps, + float nu, float bscale) +{ + int step, x, y; + float *ptr; + for (y = 0; y < height; y++) { + for (step = 0; step < steps; step++) { + ptr = buffer + width * y; + ptr[0] *= bscale; + + /* Filter rightwards */ + for (x = 1; x < width; x++) + ptr[x] += nu * ptr[x - 1]; + ptr[x = width - 1] *= bscale; + + /* Filter leftwards */ + for (; x > 0; x--) + ptr[x - 1] += nu * ptr[x]; + } + } +} + static int filter_horizontally(AVFilterContext *ctx, void *arg, int jobnr, int nb_jobs) { GBlurContext *s = ctx->priv; @@ -84,28 +86,45 @@ static int filter_horizontally(AVFilterContext *ctx, void *arg, int jobnr, int n const int steps = s->steps; const float nu = s->nu; float *buffer = s->buffer; - int y, x, step; - float *ptr; - /* Filter horizontally along each row */ - for (y = slice_start; y < slice_end; y++) { - for (step = 0; step < steps; step++) { - ptr = buffer + width * y; - ptr[0] *= boundaryscale; + s->horiz_slice(buffer + width * slice_start, width, slice_end - slice_start, + steps, nu, boundaryscale); + emms_c(); + return 0; +} - /* Filter rightwards */ - for (x = 1; x < width; x++) - ptr[x] += nu * ptr[x - 1]; +static void do_vertical_columns(float *buffer, int width, int height, + int column_begin, int column_end, int steps, + float nu, float boundaryscale, int column_step) +{ + const int numpixels = width * height; + int i, x, k, step; + float *ptr; + for (x = column_begin; x < column_end;) { + for (step = 0; step < steps; step++) { + ptr = buffer + x; + for (k = 0; k < column_step; k++) { + ptr[k] *= boundaryscale; + } + /* Filter downwards */ + for (i = width; i < numpixels; i += width) { + for (k = 0; k < column_step; k++) { + ptr[i + k] += nu * ptr[i - width + k]; + } + } + i = numpixels - width; - ptr[x = width - 1] *= boundaryscale; + for (k = 0; k < column_step; k++) + ptr[i + k] *= boundaryscale; - /* Filter leftwards */ - for (; x > 0; x--) - ptr[x - 1] += nu * ptr[x]; + /* Filter upwards */ + for (; i > 0; i -= width) { + for (k = 0; k < column_step; k++) + ptr[i - width + k] += nu * ptr[i + k]; + } } + x += column_step; } - - return 0; } static int filter_vertically(AVFilterContext *ctx, void *arg, int jobnr, int nb_jobs) @@ -117,31 +136,19 @@ static int filter_vertically(AVFilterContext *ctx, void *arg, int jobnr, int nb_ const int slice_start = (width * jobnr ) / nb_jobs; const int slice_end = (width * (jobnr+1)) / nb_jobs; const float boundaryscale = s->boundaryscaleV; - const int numpixels = width * height; const int steps = s->steps; const float nu = s->nuV; float *buffer = s->buffer; - int i, x, step; - float *ptr; - - /* Filter vertically along each column */ - for (x = slice_start; x < slice_end; x++) { - for (step = 0; step < steps; step++) { - ptr = buffer + x; - ptr[0] *= boundaryscale; - - /* Filter downwards */ - for (i = width; i < numpixels; i += width) - ptr[i] += nu * ptr[i - width]; + int aligned_end; - ptr[i = numpixels - width] *= boundaryscale; - - /* Filter upwards */ - for (; i > 0; i -= width) - ptr[i - width] += nu * ptr[i]; - } - } + aligned_end = slice_start + (((slice_end - slice_start) >> 3) << 3); + /* Filter vertically along columns (process 8 columns in each step) */ + do_vertical_columns(buffer, width, height, slice_start, aligned_end, + steps, nu, boundaryscale, 8); + /* Filter un-aligned columns one by one */ + do_vertical_columns(buffer, width, height, aligned_end, slice_end, + steps, nu, boundaryscale, 1); return 0; } @@ -209,6 +216,13 @@ static int query_formats(AVFilterContext *ctx) return ff_set_common_formats(ctx, ff_make_format_list(pix_fmts)); } +void ff_gblur_init(GBlurContext *s) +{ + s->horiz_slice = horiz_slice_c; + if (ARCH_X86_64) + ff_gblur_init_x86(s); +} + static int config_input(AVFilterLink *inlink) { const AVPixFmtDescriptor *desc = av_pix_fmt_desc_get(inlink->format); @@ -229,6 +243,7 @@ static int config_input(AVFilterLink *inlink) if (s->sigmaV < 0) { s->sigmaV = s->sigma; } + ff_gblur_init(s); return 0; } @@ -329,6 +344,22 @@ static int filter_frame(AVFilterLink *inlink, AVFrame *in) return ff_filter_frame(outlink, out); } +static int process_command(AVFilterContext *ctx, const char *cmd, const char *args, + char *res, int res_len, int flags) +{ + GBlurContext *s = ctx->priv; + int ret = 0; + + if ( !strcmp(cmd, "sigma") || !strcmp(cmd, "sigmaV") + || !strcmp(cmd, "steps") || !strcmp(cmd, "planes")) { + av_opt_set(s, cmd, args, 0); + } else { + ret = AVERROR(ENOSYS); + } + + return ret; +} + static av_cold void uninit(AVFilterContext *ctx) { GBlurContext *s = ctx->priv; @@ -364,4 +395,5 @@ AVFilter ff_vf_gblur = { .inputs = gblur_inputs, .outputs = gblur_outputs, .flags = AVFILTER_FLAG_SUPPORT_TIMELINE_GENERIC | AVFILTER_FLAG_SLICE_THREADS, + .process_command = process_command, };