highperfocused/ffmpeg
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

lavfi/nlmeans: random code shuffling to help compiler

Browse Source 
This makes nlmeans_slice() slightly faster at least on GCC 7.3.
This commit is contained in:
Clément Bœsch
2018-05-06 00:06:19 +02:00
parent 0683ad709b
commit f1248b7795
1 changed files with 4 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
libavfilter/vf_nlmeans.c
View File

@@ -368,7 +368,6 @@ static int nlmeans_slice(AVFilterContext *ctx, void *arg, int jobnr, int nb_jobs
int x, y; int x, y;
NLMeansContext *s = ctx->priv; NLMeansContext *s = ctx->priv;
const struct thread_data *td = arg; const struct thread_data *td = arg;
const uint8_t *src = td->src;
const int src_linesize = td->src_linesize; const int src_linesize = td->src_linesize;
const int process_h = td->endy - td->starty; const int process_h = td->endy - td->starty;
const int slice_start = (process_h * jobnr ) / nb_jobs; const int slice_start = (process_h * jobnr ) / nb_jobs;
@@ -377,14 +376,15 @@ static int nlmeans_slice(AVFilterContext *ctx, void *arg, int jobnr, int nb_jobs
const int endy = td->starty + slice_end; const int endy = td->starty + slice_end;
for (y = starty; y < endy; y++) { for (y = starty; y < endy; y++) {
const uint8_t *src = td->src + y*src_linesize;
struct weighted_avg *wa = s->wa + y*s->wa_linesize;
for (x = td->startx; x < td->endx; x++) { for (x = td->startx; x < td->endx; x++) {
const int patch_diff_sq = get_integral_patch_value(td->ii_start, s->ii_lz_32, x, y, td->p); const int patch_diff_sq = get_integral_patch_value(td->ii_start, s->ii_lz_32, x, y, td->p);
if (patch_diff_sq < s->max_meaningful_diff) { if (patch_diff_sq < s->max_meaningful_diff) {
struct weighted_avg *wa = &s->wa[y*s->wa_linesize + x];
const int weight_lut_idx = patch_diff_sq * s->pdiff_lut_scale; const int weight_lut_idx = patch_diff_sq * s->pdiff_lut_scale;
const double weight = s->weight_lut[weight_lut_idx]; // exp(-patch_diff_sq * s->pdiff_scale) const double weight = s->weight_lut[weight_lut_idx]; // exp(-patch_diff_sq * s->pdiff_scale)
wa->total_weight += weight; wa[x].total_weight += weight;
wa->sum += weight * src[y*src_linesize + x]; wa[x].sum += weight * src[x];
} }
} }
} }