generic-library/vpx
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

RTCD: add variance functions

Browse Source 
This commit continues the process of converting to the new RTCD
system.

Change-Id: Ie5c1aa480637e98dc3918fb562ff45c37a66c538
This commit is contained in:
John Koleszar
2012-01-12 16:55:44 -08:00
parent f103dcefaf
commit 83a91e789c
20 changed files with 388 additions and 1207 deletions
Download Patch File Download Diff File Expand all files Collapse all files

134
vp8/encoder/onyx_if.c
View File

@@ -69,7 +69,7 @@ extern void vp8_yv12_copy_src_frame_func_neon(YV12_BUFFER_CONFIG *src_ybc, YV12_
int vp8_estimate_entropy_savings(VP8_COMP *cpi);
int vp8_calc_ss_err(YV12_BUFFER_CONFIG *source, YV12_BUFFER_CONFIG *dest, const vp8_variance_rtcd_vtable_t *rtcd);
int vp8_calc_ss_err(YV12_BUFFER_CONFIG *source, YV12_BUFFER_CONFIG *dest);
extern void vp8_temporal_filter_prepare_c(VP8_COMP *cpi, int distance);
@@ -85,8 +85,7 @@ extern double vp8_calc_ssim
YV12_BUFFER_CONFIG *source,
YV12_BUFFER_CONFIG *dest,
int lumamask,
double *weight,
const vp8_variance_rtcd_vtable_t *rtcd
double *weight
);
@@ -96,8 +95,7 @@ extern double vp8_calc_ssimg
YV12_BUFFER_CONFIG *dest,
double *ssim_y,
double *ssim_u,
double *ssim_v,
const vp8_variance_rtcd_vtable_t *rtcd
double *ssim_v
);
@@ -1947,62 +1945,62 @@ struct VP8_COMP* vp8_create_compressor(VP8_CONFIG *oxcf)
vp8cx_create_encoder_threads(cpi);
#endif
cpi->fn_ptr[BLOCK_16X16].sdf = VARIANCE_INVOKE(&cpi->rtcd.variance, sad16x16);
cpi->fn_ptr[BLOCK_16X16].vf = VARIANCE_INVOKE(&cpi->rtcd.variance, var16x16);
cpi->fn_ptr[BLOCK_16X16].svf = VARIANCE_INVOKE(&cpi->rtcd.variance, subpixvar16x16);
cpi->fn_ptr[BLOCK_16X16].svf_halfpix_h = VARIANCE_INVOKE(&cpi->rtcd.variance, halfpixvar16x16_h);
cpi->fn_ptr[BLOCK_16X16].svf_halfpix_v = VARIANCE_INVOKE(&cpi->rtcd.variance, halfpixvar16x16_v);
cpi->fn_ptr[BLOCK_16X16].svf_halfpix_hv = VARIANCE_INVOKE(&cpi->rtcd.variance, halfpixvar16x16_hv);
cpi->fn_ptr[BLOCK_16X16].sdx3f = VARIANCE_INVOKE(&cpi->rtcd.variance, sad16x16x3);
cpi->fn_ptr[BLOCK_16X16].sdx8f = VARIANCE_INVOKE(&cpi->rtcd.variance, sad16x16x8);
cpi->fn_ptr[BLOCK_16X16].sdx4df = VARIANCE_INVOKE(&cpi->rtcd.variance, sad16x16x4d);
cpi->fn_ptr[BLOCK_16X16].sdf = vp8_sad16x16;
cpi->fn_ptr[BLOCK_16X16].vf = vp8_variance16x16;
cpi->fn_ptr[BLOCK_16X16].svf = vp8_sub_pixel_variance16x16;
cpi->fn_ptr[BLOCK_16X16].svf_halfpix_h = vp8_variance_halfpixvar16x16_h;
cpi->fn_ptr[BLOCK_16X16].svf_halfpix_v = vp8_variance_halfpixvar16x16_v;
cpi->fn_ptr[BLOCK_16X16].svf_halfpix_hv = vp8_variance_halfpixvar16x16_hv;
cpi->fn_ptr[BLOCK_16X16].sdx3f = vp8_sad16x16x3;
cpi->fn_ptr[BLOCK_16X16].sdx8f = vp8_sad16x16x8;
cpi->fn_ptr[BLOCK_16X16].sdx4df = vp8_sad16x16x4d;
cpi->fn_ptr[BLOCK_16X8].sdf = VARIANCE_INVOKE(&cpi->rtcd.variance, sad16x8);
cpi->fn_ptr[BLOCK_16X8].vf = VARIANCE_INVOKE(&cpi->rtcd.variance, var16x8);
cpi->fn_ptr[BLOCK_16X8].svf = VARIANCE_INVOKE(&cpi->rtcd.variance, subpixvar16x8);
cpi->fn_ptr[BLOCK_16X8].sdf = vp8_sad16x8;
cpi->fn_ptr[BLOCK_16X8].vf = vp8_variance16x8;
cpi->fn_ptr[BLOCK_16X8].svf = vp8_sub_pixel_variance16x8;
cpi->fn_ptr[BLOCK_16X8].svf_halfpix_h = NULL;
cpi->fn_ptr[BLOCK_16X8].svf_halfpix_v = NULL;
cpi->fn_ptr[BLOCK_16X8].svf_halfpix_hv = NULL;
cpi->fn_ptr[BLOCK_16X8].sdx3f = VARIANCE_INVOKE(&cpi->rtcd.variance, sad16x8x3);
cpi->fn_ptr[BLOCK_16X8].sdx8f = VARIANCE_INVOKE(&cpi->rtcd.variance, sad16x8x8);
cpi->fn_ptr[BLOCK_16X8].sdx4df = VARIANCE_INVOKE(&cpi->rtcd.variance, sad16x8x4d);
cpi->fn_ptr[BLOCK_16X8].sdx3f = vp8_sad16x8x3;
cpi->fn_ptr[BLOCK_16X8].sdx8f = vp8_sad16x8x8;
cpi->fn_ptr[BLOCK_16X8].sdx4df = vp8_sad16x8x4d;
cpi->fn_ptr[BLOCK_8X16].sdf = VARIANCE_INVOKE(&cpi->rtcd.variance, sad8x16);
cpi->fn_ptr[BLOCK_8X16].vf = VARIANCE_INVOKE(&cpi->rtcd.variance, var8x16);
cpi->fn_ptr[BLOCK_8X16].svf = VARIANCE_INVOKE(&cpi->rtcd.variance, subpixvar8x16);
cpi->fn_ptr[BLOCK_8X16].sdf = vp8_sad8x16;
cpi->fn_ptr[BLOCK_8X16].vf = vp8_variance8x16;
cpi->fn_ptr[BLOCK_8X16].svf = vp8_sub_pixel_variance8x16;
cpi->fn_ptr[BLOCK_8X16].svf_halfpix_h = NULL;
cpi->fn_ptr[BLOCK_8X16].svf_halfpix_v = NULL;
cpi->fn_ptr[BLOCK_8X16].svf_halfpix_hv = NULL;
cpi->fn_ptr[BLOCK_8X16].sdx3f = VARIANCE_INVOKE(&cpi->rtcd.variance, sad8x16x3);
cpi->fn_ptr[BLOCK_8X16].sdx8f = VARIANCE_INVOKE(&cpi->rtcd.variance, sad8x16x8);
cpi->fn_ptr[BLOCK_8X16].sdx4df = VARIANCE_INVOKE(&cpi->rtcd.variance, sad8x16x4d);
cpi->fn_ptr[BLOCK_8X16].sdx3f = vp8_sad8x16x3;
cpi->fn_ptr[BLOCK_8X16].sdx8f = vp8_sad8x16x8;
cpi->fn_ptr[BLOCK_8X16].sdx4df = vp8_sad8x16x4d;
cpi->fn_ptr[BLOCK_8X8].sdf = VARIANCE_INVOKE(&cpi->rtcd.variance, sad8x8);
cpi->fn_ptr[BLOCK_8X8].vf = VARIANCE_INVOKE(&cpi->rtcd.variance, var8x8);
cpi->fn_ptr[BLOCK_8X8].svf = VARIANCE_INVOKE(&cpi->rtcd.variance, subpixvar8x8);
cpi->fn_ptr[BLOCK_8X8].sdf = vp8_sad8x8;
cpi->fn_ptr[BLOCK_8X8].vf = vp8_variance8x8;
cpi->fn_ptr[BLOCK_8X8].svf = vp8_sub_pixel_variance8x8;
cpi->fn_ptr[BLOCK_8X8].svf_halfpix_h = NULL;
cpi->fn_ptr[BLOCK_8X8].svf_halfpix_v = NULL;
cpi->fn_ptr[BLOCK_8X8].svf_halfpix_hv = NULL;
cpi->fn_ptr[BLOCK_8X8].sdx3f = VARIANCE_INVOKE(&cpi->rtcd.variance, sad8x8x3);
cpi->fn_ptr[BLOCK_8X8].sdx8f = VARIANCE_INVOKE(&cpi->rtcd.variance, sad8x8x8);
cpi->fn_ptr[BLOCK_8X8].sdx4df = VARIANCE_INVOKE(&cpi->rtcd.variance, sad8x8x4d);
cpi->fn_ptr[BLOCK_8X8].sdx3f = vp8_sad8x8x3;
cpi->fn_ptr[BLOCK_8X8].sdx8f = vp8_sad8x8x8;
cpi->fn_ptr[BLOCK_8X8].sdx4df = vp8_sad8x8x4d;
cpi->fn_ptr[BLOCK_4X4].sdf = VARIANCE_INVOKE(&cpi->rtcd.variance, sad4x4);
cpi->fn_ptr[BLOCK_4X4].vf = VARIANCE_INVOKE(&cpi->rtcd.variance, var4x4);
cpi->fn_ptr[BLOCK_4X4].svf = VARIANCE_INVOKE(&cpi->rtcd.variance, subpixvar4x4);
cpi->fn_ptr[BLOCK_4X4].sdf = vp8_sad4x4;
cpi->fn_ptr[BLOCK_4X4].vf = vp8_variance4x4;
cpi->fn_ptr[BLOCK_4X4].svf = vp8_sub_pixel_variance4x4;
cpi->fn_ptr[BLOCK_4X4].svf_halfpix_h = NULL;
cpi->fn_ptr[BLOCK_4X4].svf_halfpix_v = NULL;
cpi->fn_ptr[BLOCK_4X4].svf_halfpix_hv = NULL;
cpi->fn_ptr[BLOCK_4X4].sdx3f = VARIANCE_INVOKE(&cpi->rtcd.variance, sad4x4x3);
cpi->fn_ptr[BLOCK_4X4].sdx8f = VARIANCE_INVOKE(&cpi->rtcd.variance, sad4x4x8);
cpi->fn_ptr[BLOCK_4X4].sdx4df = VARIANCE_INVOKE(&cpi->rtcd.variance, sad4x4x4d);
cpi->fn_ptr[BLOCK_4X4].sdx3f = vp8_sad4x4x3;
cpi->fn_ptr[BLOCK_4X4].sdx8f = vp8_sad4x4x8;
cpi->fn_ptr[BLOCK_4X4].sdx4df = vp8_sad4x4x4d;
#if ARCH_X86 || ARCH_X86_64
cpi->fn_ptr[BLOCK_16X16].copymem = VARIANCE_INVOKE(&cpi->rtcd.variance, copy32xn);
cpi->fn_ptr[BLOCK_16X8].copymem = VARIANCE_INVOKE(&cpi->rtcd.variance, copy32xn);
cpi->fn_ptr[BLOCK_8X16].copymem = VARIANCE_INVOKE(&cpi->rtcd.variance, copy32xn);
cpi->fn_ptr[BLOCK_8X8].copymem = VARIANCE_INVOKE(&cpi->rtcd.variance, copy32xn);
cpi->fn_ptr[BLOCK_4X4].copymem = VARIANCE_INVOKE(&cpi->rtcd.variance, copy32xn);
cpi->fn_ptr[BLOCK_16X16].copymem = vp8_copy32xn;
cpi->fn_ptr[BLOCK_16X8].copymem = vp8_copy32xn;
cpi->fn_ptr[BLOCK_8X16].copymem = vp8_copy32xn;
cpi->fn_ptr[BLOCK_8X8].copymem = vp8_copy32xn;
cpi->fn_ptr[BLOCK_4X4].copymem = vp8_copy32xn;
#endif
cpi->full_search_sad = SEARCH_INVOKE(&cpi->rtcd.search, full_search);
@@ -2334,8 +2332,7 @@ void vp8_remove_compressor(VP8_COMP **ptr)
static uint64_t calc_plane_error(unsigned char *orig, int orig_stride,
unsigned char *recon, int recon_stride,
unsigned int cols, unsigned int rows,
vp8_variance_rtcd_vtable_t *rtcd)
unsigned int cols, unsigned int rows)
{
unsigned int row, col;
uint64_t total_sse = 0;
@@ -2347,7 +2344,7 @@ static uint64_t calc_plane_error(unsigned char *orig, int orig_stride,
{
unsigned int sse;
VARIANCE_INVOKE(rtcd, mse16x16)(orig + col, orig_stride,
vp8_mse16x16(orig + col, orig_stride,
recon + col, recon_stride,
&sse);
total_sse += sse;
@@ -2408,8 +2405,7 @@ static void generate_psnr_packet(VP8_COMP *cpi)
pkt.kind = VPX_CODEC_PSNR_PKT;
sse = calc_plane_error(orig->y_buffer, orig->y_stride,
recon->y_buffer, recon->y_stride,
width, height,
IF_RTCD(&cpi->rtcd.variance));
width, height);
pkt.data.psnr.sse[0] = sse;
pkt.data.psnr.sse[1] = sse;
pkt.data.psnr.samples[0] = width * height;
@@ -2420,8 +2416,7 @@ static void generate_psnr_packet(VP8_COMP *cpi)
sse = calc_plane_error(orig->u_buffer, orig->uv_stride,
recon->u_buffer, recon->uv_stride,
width, height,
IF_RTCD(&cpi->rtcd.variance));
width, height);
pkt.data.psnr.sse[0] += sse;
pkt.data.psnr.sse[2] = sse;
pkt.data.psnr.samples[0] += width * height;
@@ -2429,8 +2424,7 @@ static void generate_psnr_packet(VP8_COMP *cpi)
sse = calc_plane_error(orig->v_buffer, orig->uv_stride,
recon->v_buffer, recon->uv_stride,
width, height,
IF_RTCD(&cpi->rtcd.variance));
width, height);
pkt.data.psnr.sse[0] += sse;
pkt.data.psnr.sse[3] = sse;
pkt.data.psnr.samples[0] += width * height;
@@ -3819,8 +3813,7 @@ static void encode_frame_to_data_rate
{
int last_q = Q;
int kf_err = vp8_calc_ss_err(cpi->Source,
&cm->yv12_fb[cm->new_fb_idx],
IF_RTCD(&cpi->rtcd.variance));
&cm->yv12_fb[cm->new_fb_idx]);
// The key frame is not good enough
if ( kf_err > ((cpi->ambient_err * 7) >> 3) )
@@ -4016,8 +4009,7 @@ static void encode_frame_to_data_rate
if ( cpi->next_key_frame_forced && (cpi->twopass.frames_to_key == 0) )
{
cpi->ambient_err = vp8_calc_ss_err(cpi->Source,
&cm->yv12_fb[cm->new_fb_idx],
IF_RTCD(&cpi->rtcd.variance));
&cm->yv12_fb[cm->new_fb_idx]);
}
/* This frame's MVs are saved and will be used in next frame's MV predictor.
@@ -4961,16 +4953,13 @@ int vp8_get_compressed_data(VP8_COMP *cpi, unsigned int *frame_flags, unsigned l
int64_t sq_error, sq_error2;
ye = calc_plane_error(orig->y_buffer, orig->y_stride,
recon->y_buffer, recon->y_stride, orig->y_width, orig->y_height,
IF_RTCD(&cpi->rtcd.variance));
recon->y_buffer, recon->y_stride, orig->y_width, orig->y_height);
ue = calc_plane_error(orig->u_buffer, orig->uv_stride,
recon->u_buffer, recon->uv_stride, orig->uv_width, orig->uv_height,
IF_RTCD(&cpi->rtcd.variance));
recon->u_buffer, recon->uv_stride, orig->uv_width, orig->uv_height);
ve = calc_plane_error(orig->v_buffer, orig->uv_stride,
recon->v_buffer, recon->uv_stride, orig->uv_width, orig->uv_height,
IF_RTCD(&cpi->rtcd.variance));
recon->v_buffer, recon->uv_stride, orig->uv_width, orig->uv_height);
sq_error = ye + ue + ve;
@@ -4985,20 +4974,17 @@ int vp8_get_compressed_data(VP8_COMP *cpi, unsigned int *frame_flags, unsigned l
double frame_psnr2, frame_ssim2 = 0;
double weight = 0;
vp8_deblock(cm->frame_to_show, &cm->post_proc_buffer, cm->filter_level * 10 / 6, 1, 0, IF_RTCD(&cm->rtcd.postproc));
vp8_deblock(cm->frame_to_show, &cm->post_proc_buffer, cm->filter_level * 10 / 6, 1, 0);
vp8_clear_system_state();
ye = calc_plane_error(orig->y_buffer, orig->y_stride,
pp->y_buffer, pp->y_stride, orig->y_width, orig->y_height,
IF_RTCD(&cpi->rtcd.variance));
pp->y_buffer, pp->y_stride, orig->y_width, orig->y_height);
ue = calc_plane_error(orig->u_buffer, orig->uv_stride,
pp->u_buffer, pp->uv_stride, orig->uv_width, orig->uv_height,
IF_RTCD(&cpi->rtcd.variance));
pp->u_buffer, pp->uv_stride, orig->uv_width, orig->uv_height);
ve = calc_plane_error(orig->v_buffer, orig->uv_stride,
pp->v_buffer, pp->uv_stride, orig->uv_width, orig->uv_height,
IF_RTCD(&cpi->rtcd.variance));
pp->v_buffer, pp->uv_stride, orig->uv_width, orig->uv_height);
sq_error2 = ye + ue + ve;
@@ -5011,8 +4997,7 @@ int vp8_get_compressed_data(VP8_COMP *cpi, unsigned int *frame_flags, unsigned l
cpi->totalp += frame_psnr2;
frame_ssim2 = vp8_calc_ssim(cpi->Source,
&cm->post_proc_buffer, 1, &weight,
IF_RTCD(&cpi->rtcd.variance));
&cm->post_proc_buffer, 1, &weight);
cpi->summed_quality += frame_ssim2 * weight;
cpi->summed_weights += weight;
@@ -5042,7 +5027,7 @@ int vp8_get_compressed_data(VP8_COMP *cpi, unsigned int *frame_flags, unsigned l
{
double y, u, v, frame_all;
frame_all = vp8_calc_ssimg(cpi->Source, cm->frame_to_show,
&y, &u, &v, IF_RTCD(&cpi->rtcd.variance));
&y, &u, &v);
if (cpi->oxcf.number_of_layers > 1)
{
@@ -5222,14 +5207,13 @@ int vp8_set_internal_size(VP8_COMP *cpi, VPX_SCALING horiz_mode, VPX_SCALING ver
int vp8_calc_ss_err(YV12_BUFFER_CONFIG *source, YV12_BUFFER_CONFIG *dest, const vp8_variance_rtcd_vtable_t *rtcd)
int vp8_calc_ss_err(YV12_BUFFER_CONFIG *source, YV12_BUFFER_CONFIG *dest)
{
int i, j;
int Total = 0;
unsigned char *src = source->y_buffer;
unsigned char *dst = dest->y_buffer;
(void)rtcd;
// Loop through the Y plane raw and reconstruction data summing (square differences)
for (i = 0; i < source->y_height; i += 16)
@@ -5237,7 +5221,7 @@ int vp8_calc_ss_err(YV12_BUFFER_CONFIG *source, YV12_BUFFER_CONFIG *dest, const
for (j = 0; j < source->y_width; j += 16)
{
unsigned int sse;
Total += VARIANCE_INVOKE(rtcd, mse16x16)(src + j, source->y_stride, dst + j, dest->y_stride, &sse);
Total += vp8_mse16x16(src + j, source->y_stride, dst + j, dest->y_stride, &sse);
}
src += 16 * source->y_stride;