virtualenv/FFmpeg
1
0
Fork 0
mirror of https://github.com/FFmpeg/FFmpeg.git synced 2024-12-23 12:43:46 +02:00
Code Issues Releases Activity

lavfi/nlmeans: make compute_safe_ssd_integral_image_c faster

Browse Source 
before:  ssd_integral_image_c: 49204.6
after:   ssd_integral_image_c: 44272.8

Unrolling by 4 made the biggest difference on odroid-c2 (aarch64);
unrolling by 2 or 8 both raised 46k cycles vs 44k for 4.

Additionally, this is a much better reference when writing SIMD (SIMD
vectorization will just target 16 instead of 4).
This commit is contained in:
Clément Bœsch 2018-05-06 12:34:54 +02:00
parent f679711c1b
commit 43d16aef63
1 changed files with 16 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

25
libavfilter/vf_nlmeans.c
View File

@ -146,10 +146,6 @@ static inline int get_integral_patch_value(const uint32_t *ii, int ii_lz_32, int
* function, we do not need any clipping here. * function, we do not need any clipping here.
* *
* The line above dst and the column to its left are always readable. * The line above dst and the column to its left are always readable.
*
* This C version computes the SSD integral image using a scalar accumulator,
* while for SIMD implementation it is likely more interesting to use the
* two-loops algorithm variant.
*/ */
static void compute_safe_ssd_integral_image_c(uint32_t *dst, ptrdiff_t dst_linesize_32, static void compute_safe_ssd_integral_image_c(uint32_t *dst, ptrdiff_t dst_linesize_32,
const uint8_t *s1, ptrdiff_t linesize1, const uint8_t *s1, ptrdiff_t linesize1,
@ -157,21 +153,32 @@ static void compute_safe_ssd_integral_image_c(uint32_t *dst, ptrdiff_t dst_lines
int w, int h) int w, int h)
{ {
int x, y; int x, y;
const uint32_t *dst_top = dst - dst_linesize_32;
/* SIMD-friendly assumptions allowed here */ /* SIMD-friendly assumptions allowed here */
av_assert2(!(w & 0xf) && w >= 16 && h >= 1); av_assert2(!(w & 0xf) && w >= 16 && h >= 1);
for (y = 0; y < h; y++) { for (y = 0; y < h; y++) {
uint32_t acc = dst[-1] - dst[-dst_linesize_32 - 1]; for (x = 0; x < w; x += 4) {
const int d0 = s1[x ] - s2[x ];
const int d1 = s1[x + 1] - s2[x + 1];
const int d2 = s1[x + 2] - s2[x + 2];
const int d3 = s1[x + 3] - s2[x + 3];
for (x = 0; x < w; x++) { dst[x ] = dst_top[x ] - dst_top[x - 1] + d0*d0;
const int d = s1[x] - s2[x]; dst[x + 1] = dst_top[x + 1] - dst_top[x ] + d1*d1;
acc += d * d; dst[x + 2] = dst_top[x + 2] - dst_top[x + 1] + d2*d2;
dst[x] = dst[-dst_linesize_32 + x] + acc; dst[x + 3] = dst_top[x + 3] - dst_top[x + 2] + d3*d3;
dst[x ] += dst[x - 1];
dst[x + 1] += dst[x ];
dst[x + 2] += dst[x + 1];
dst[x + 3] += dst[x + 2];
} }
s1 += linesize1; s1 += linesize1;
s2 += linesize2; s2 += linesize2;
dst += dst_linesize_32; dst += dst_linesize_32;
dst_top += dst_linesize_32;
} }
} }