virtualenv/FFmpeg
1
0
Fork 0
mirror of https://github.com/FFmpeg/FFmpeg.git synced 2025-08-10 06:10:52 +02:00
Code Issues Releases Activity

avfilter/vf_mix: improve speed even more by avoiding using filter private context

Browse Source
This commit is contained in:
Paul B Mahol
2023-05-19 19:12:37 +02:00
parent c789a2324a
commit 613f716450
1 changed files with 8 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

13
libavfilter/vf_mix.c
View File

@@ -161,27 +161,27 @@ typedef struct ThreadData {
continue; \ continue; \
} \ } \
\ \
for (int i = 0; i < s->nb_inputs; i++) \ for (int i = 0; i < nb_inputs; i++) \
linesize[i] = in[i]->linesize[p]; \ linesize[i] = in[i]->linesize[p]; \
\ \
for (int i = 0; i < s->nb_inputs; i++) \ for (int i = 0; i < nb_inputs; i++) \
srcf[i] = in[i]->data[p] + slice_start * linesize[i]; \ srcf[i] = in[i]->data[p] + slice_start * linesize[i]; \
\ \
for (int y = slice_start; y < slice_end; y++) { \ for (int y = slice_start; y < slice_end; y++) { \
for (int x = 0; x < width; x++) { \ for (int x = 0; x < width; x++) { \
float val = 0.f; \ float val = 0.f; \
\ \
for (int i = 0; i < s->nb_inputs; i++) { \ for (int i = 0; i < nb_inputs; i++) { \
float src = *(type *)(srcf[i] + x * sizeof(type)); \ float src = *(type *)(srcf[i] + x * sizeof(type)); \
\ \
val += src * weights[i]; \ val += src * weights[i]; \
} \ } \
\ \
dst[x] = clip(fun(val * s->wfactor), 0, s->max); \ dst[x] = clip(fun(val * wfactor), 0, max); \
} \ } \
\ \
dst += dst_linesize; \ dst += dst_linesize; \
for (int i = 0; i < s->nb_inputs; i++) \ for (int i = 0; i < nb_inputs; i++) \
srcf[i] += linesize[i]; \ srcf[i] += linesize[i]; \
} \ } \
} }
@@ -200,6 +200,9 @@ static int mix_frames(AVFilterContext *ctx, void *arg, int jobnr, int nb_jobs)
const float *weights = s->weights; const float *weights = s->weights;
uint8_t **srcf = s->data + jobnr * s->nb_inputs; uint8_t **srcf = s->data + jobnr * s->nb_inputs;
int *linesize = s->linesize + jobnr * s->nb_inputs; int *linesize = s->linesize + jobnr * s->nb_inputs;
const int nb_inputs = s->nb_inputs;
const float wfactor = s->wfactor;
const int max = s->max;
if (s->depth <= 8) { if (s->depth <= 8) {
MIX_SLICE(uint8_t, lrintf, CLIP8) MIX_SLICE(uint8_t, lrintf, CLIP8)