Merge "Cosmetic changes in handle_inter_mode_" into experimental
This commit is contained in:
Коммит
c0102fd88b
|
@ -2229,7 +2229,6 @@ static int64_t handle_inter_mode(VP9_COMP *cpi, MACROBLOCK *x,
|
||||||
|
|
||||||
|
|
||||||
if (pred_exists) {
|
if (pred_exists) {
|
||||||
// FIXME(rbultje): mb code still predicts into xd->predictor
|
|
||||||
for (i = 0; i < bh * MI_SIZE; ++i)
|
for (i = 0; i < bh * MI_SIZE; ++i)
|
||||||
vpx_memcpy(xd->plane[0].dst.buf + i * xd->plane[0].dst.stride,
|
vpx_memcpy(xd->plane[0].dst.buf + i * xd->plane[0].dst.stride,
|
||||||
tmp_ybuf + i * bw * MI_SIZE,
|
tmp_ybuf + i * bw * MI_SIZE,
|
||||||
|
@ -2264,17 +2263,11 @@ static int64_t handle_inter_mode(VP9_COMP *cpi, MACROBLOCK *x,
|
||||||
if (threshold < x->encode_breakout)
|
if (threshold < x->encode_breakout)
|
||||||
threshold = x->encode_breakout;
|
threshold = x->encode_breakout;
|
||||||
|
|
||||||
if (bsize != BLOCK_SIZE_MB16X16) {
|
var = cpi->fn_ptr[block_size].vf(x->plane[0].src.buf,
|
||||||
var = cpi->fn_ptr[block_size].vf(x->plane[0].src.buf,
|
x->plane[0].src.stride,
|
||||||
x->plane[0].src.stride,
|
xd->plane[0].dst.buf,
|
||||||
xd->plane[0].dst.buf,
|
xd->plane[0].dst.stride,
|
||||||
xd->plane[0].dst.stride,
|
&sse);
|
||||||
&sse);
|
|
||||||
} else {
|
|
||||||
var = vp9_variance16x16(x->plane[0].src.buf, x->plane[0].src.stride,
|
|
||||||
xd->plane[0].dst.buf, xd->plane[0].dst.stride,
|
|
||||||
&sse);
|
|
||||||
}
|
|
||||||
|
|
||||||
if ((int)sse < threshold) {
|
if ((int)sse < threshold) {
|
||||||
unsigned int q2dc = xd->plane[0].dequant[0];
|
unsigned int q2dc = xd->plane[0].dequant[0];
|
||||||
|
@ -2284,29 +2277,16 @@ static int64_t handle_inter_mode(VP9_COMP *cpi, MACROBLOCK *x,
|
||||||
(sse / 2 > var && sse - var < 64)) {
|
(sse / 2 > var && sse - var < 64)) {
|
||||||
// Check u and v to make sure skip is ok
|
// Check u and v to make sure skip is ok
|
||||||
int sse2;
|
int sse2;
|
||||||
|
unsigned int sse2u, sse2v;
|
||||||
if (bsize != BLOCK_SIZE_MB16X16) {
|
var = cpi->fn_ptr[uv_block_size].vf(x->plane[1].src.buf,
|
||||||
unsigned int sse2u, sse2v;
|
x->plane[1].src.stride,
|
||||||
// FIXME(rbultje): mb predictors predict into xd->predictor
|
xd->plane[1].dst.buf,
|
||||||
var = cpi->fn_ptr[uv_block_size].vf(x->plane[1].src.buf,
|
xd->plane[1].dst.stride, &sse2u);
|
||||||
x->plane[1].src.stride,
|
var = cpi->fn_ptr[uv_block_size].vf(x->plane[2].src.buf,
|
||||||
xd->plane[1].dst.buf,
|
x->plane[1].src.stride,
|
||||||
xd->plane[1].dst.stride, &sse2u);
|
xd->plane[2].dst.buf,
|
||||||
var = cpi->fn_ptr[uv_block_size].vf(x->plane[2].src.buf,
|
xd->plane[1].dst.stride, &sse2v);
|
||||||
x->plane[1].src.stride,
|
sse2 = sse2u + sse2v;
|
||||||
xd->plane[2].dst.buf,
|
|
||||||
xd->plane[1].dst.stride, &sse2v);
|
|
||||||
sse2 = sse2u + sse2v;
|
|
||||||
} else {
|
|
||||||
unsigned int sse2u, sse2v;
|
|
||||||
var = vp9_variance8x8(x->plane[1].src.buf, x->plane[1].src.stride,
|
|
||||||
xd->plane[1].dst.buf, xd->plane[1].dst.stride,
|
|
||||||
&sse2u);
|
|
||||||
var = vp9_variance8x8(x->plane[2].src.buf, x->plane[1].src.stride,
|
|
||||||
xd->plane[2].dst.buf, xd->plane[1].dst.stride,
|
|
||||||
&sse2v);
|
|
||||||
sse2 = sse2u + sse2v;
|
|
||||||
}
|
|
||||||
|
|
||||||
if (sse2 * 2 < threshold) {
|
if (sse2 * 2 < threshold) {
|
||||||
x->skip = 1;
|
x->skip = 1;
|
||||||
|
|
Загрузка…
Ссылка в новой задаче