aboutsummaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
authorCarl Eugen Hoyos <ceffmpeg@gmail.com>2018-06-18 11:53:56 +0200
committerCarl Eugen Hoyos <ceffmpeg@gmail.com>2018-06-18 11:53:56 +0200
commita707a0b657da7784a1832c3d0b02c0cc5444f6a9 (patch)
tree78e08d3a05b848b0c03e611f3119676306e4f692
parentb41b6b323417ce1e79bf818f91d60526697954f1 (diff)
downloadffmpeg-a707a0b657da7784a1832c3d0b02c0cc5444f6a9.tar.gz
Revert "avcodec/vc1: add Simple and Main profile to vc1_put_signed_blocks_clamped"
This reverts commit 2065317db6dc3a219f8ed2de427fe7e19e02eb68. I committed a wrong version of this patch. Also reverts the follow-up commits 77a3dfb3 and 8331e591.
-rw-r--r--libavcodec/vc1.h1
-rw-r--r--libavcodec/vc1_block.c133
-rw-r--r--libavcodec/vc1_loopfilter.c30
3 files changed, 97 insertions, 67 deletions
diff --git a/libavcodec/vc1.h b/libavcodec/vc1.h
index 69f6ca9e4d..1d283f8589 100644
--- a/libavcodec/vc1.h
+++ b/libavcodec/vc1.h
@@ -422,6 +422,7 @@ void ff_vc1_init_transposed_scantables(VC1Context *v);
int ff_vc1_decode_end(AVCodecContext *avctx);
void ff_vc1_decode_blocks(VC1Context *v);
+void ff_vc1_loop_filter_iblk(VC1Context *v, int pq);
void ff_vc1_i_overlap_filter(VC1Context *v);
void ff_vc1_p_overlap_filter(VC1Context *v);
void ff_vc1_i_loop_filter(VC1Context *v);
diff --git a/libavcodec/vc1_block.c b/libavcodec/vc1_block.c
index b5454b2e60..0160095c6c 100644
--- a/libavcodec/vc1_block.c
+++ b/libavcodec/vc1_block.c
@@ -66,7 +66,7 @@ static inline void init_block_index(VC1Context *v)
/** @} */ //Bitplane group
-static void vc1_put_blocks_clamped(VC1Context *v, int put_signed)
+static void vc1_put_signed_blocks_clamped(VC1Context *v)
{
MpegEncContext *s = &v->s;
uint8_t *dest;
@@ -85,14 +85,9 @@ static void vc1_put_blocks_clamped(VC1Context *v, int put_signed)
if (i > 3 ? v->mb_type[0][s->block_index[i] - s->block_wrap[i] - 1] :
v->mb_type[0][s->block_index[i] - 2 * s->block_wrap[i] - 2]) {
dest = s->dest[0] + ((i & 2) - 4) * 4 * s->linesize + ((i & 1) - 2) * 8;
- if (put_signed)
- s->idsp.put_signed_pixels_clamped(v->block[v->topleft_blk_idx][i],
- i > 3 ? s->dest[i - 3] - 8 * s->uvlinesize - 8 : dest,
- i > 3 ? s->uvlinesize : s->linesize);
- else
- s->idsp.put_pixels_clamped(v->block[v->topleft_blk_idx][i],
- i > 3 ? s->dest[i - 3] - 8 * s->uvlinesize - 8 : dest,
- i > 3 ? s->uvlinesize : s->linesize);
+ s->idsp.put_signed_pixels_clamped(v->block[v->topleft_blk_idx][i],
+ i > 3 ? s->dest[i - 3] - 8 * s->uvlinesize - 8 : dest,
+ i > 3 ? s->uvlinesize : s->linesize);
}
}
}
@@ -101,14 +96,9 @@ static void vc1_put_blocks_clamped(VC1Context *v, int put_signed)
if (i > 3 ? v->mb_type[0][s->block_index[i] - s->block_wrap[i]] :
v->mb_type[0][s->block_index[i] - 2 * s->block_wrap[i]]) {
dest = s->dest[0] + ((i & 2) - 4) * 4 * s->linesize + (i & 1) * 8;
- if (put_signed)
- s->idsp.put_signed_pixels_clamped(v->block[v->top_blk_idx][i],
- i > 3 ? s->dest[i - 3] - 8 * s->uvlinesize : dest,
- i > 3 ? s->uvlinesize : s->linesize);
- else
- s->idsp.put_pixels_clamped(v->block[v->top_blk_idx][i],
- i > 3 ? s->dest[i - 3] - 8 * s->uvlinesize : dest,
- i > 3 ? s->uvlinesize : s->linesize);
+ s->idsp.put_signed_pixels_clamped(v->block[v->top_blk_idx][i],
+ i > 3 ? s->dest[i - 3] - 8 * s->uvlinesize : dest,
+ i > 3 ? s->uvlinesize : s->linesize);
}
}
}
@@ -124,14 +114,9 @@ static void vc1_put_blocks_clamped(VC1Context *v, int put_signed)
dest = s->dest[0] + ((i & 2) >> 1) * s->linesize + ((i & 1) - 2) * 8;
else
dest = s->dest[0] + (i & 2) * 4 * s->linesize + ((i & 1) - 2) * 8;
- if (put_signed)
- s->idsp.put_signed_pixels_clamped(v->block[v->left_blk_idx][i],
- i > 3 ? s->dest[i - 3] - 8 : dest,
- i > 3 ? s->uvlinesize : s->linesize << fieldtx);
- else
- s->idsp.put_pixels_clamped(v->block[v->left_blk_idx][i],
- i > 3 ? s->dest[i - 3] - 8 : dest,
- i > 3 ? s->uvlinesize : s->linesize << fieldtx);
+ s->idsp.put_signed_pixels_clamped(v->block[v->left_blk_idx][i],
+ i > 3 ? s->dest[i - 3] - 8 : dest,
+ i > 3 ? s->uvlinesize : s->linesize << fieldtx);
}
}
}
@@ -144,14 +129,9 @@ static void vc1_put_blocks_clamped(VC1Context *v, int put_signed)
dest = s->dest[0] + ((i & 2) >> 1) * s->linesize + (i & 1) * 8;
else
dest = s->dest[0] + (i & 2) * 4 * s->linesize + (i & 1) * 8;
- if (put_signed)
- s->idsp.put_signed_pixels_clamped(v->block[v->cur_blk_idx][i],
- i > 3 ? s->dest[i - 3] : dest,
- i > 3 ? s->uvlinesize : s->linesize << fieldtx);
- else
- s->idsp.put_pixels_clamped(v->block[v->cur_blk_idx][i],
- i > 3 ? s->dest[i - 3] : dest,
- i > 3 ? s->uvlinesize : s->linesize << fieldtx);
+ s->idsp.put_signed_pixels_clamped(v->block[v->cur_blk_idx][i],
+ i > 3 ? s->dest[i - 3] : dest,
+ i > 3 ? s->uvlinesize : s->linesize << fieldtx);
}
}
}
@@ -1489,7 +1469,7 @@ static int vc1_decode_p_mb(VC1Context *v)
end:
if (v->overlap && v->pq >= 9)
ff_vc1_p_overlap_filter(v);
- vc1_put_blocks_clamped(v, 1);
+ vc1_put_signed_blocks_clamped(v);
v->cbp[s->mb_x] = block_cbp;
v->ttblk[s->mb_x] = block_tt;
@@ -1700,7 +1680,7 @@ static int vc1_decode_p_mb_intfr(VC1Context *v)
}
if (v->overlap && v->pq >= 9)
ff_vc1_p_overlap_filter(v);
- vc1_put_blocks_clamped(v, 1);
+ vc1_put_signed_blocks_clamped(v);
v->cbp[s->mb_x] = block_cbp;
v->ttblk[s->mb_x] = block_tt;
@@ -1820,7 +1800,7 @@ static int vc1_decode_p_mb_intfi(VC1Context *v)
}
if (v->overlap && v->pq >= 9)
ff_vc1_p_overlap_filter(v);
- vc1_put_blocks_clamped(v, 1);
+ vc1_put_signed_blocks_clamped(v);
v->cbp[s->mb_x] = block_cbp;
v->ttblk[s->mb_x] = block_tt;
@@ -2537,28 +2517,30 @@ static void vc1_decode_i_blocks(VC1Context *v)
s->mb_x = s->mb_y = 0;
s->mb_intra = 1;
s->first_slice_line = 1;
- for (s->mb_y = s->start_mb_y; s->mb_y < s->end_mb_y; s->mb_y++) {
+ for (s->mb_y = 0; s->mb_y < s->end_mb_y; s->mb_y++) {
s->mb_x = 0;
init_block_index(v);
for (; s->mb_x < v->end_mb_x; s->mb_x++) {
- int16_t (*block)[64] = v->block[v->cur_blk_idx];
+ uint8_t *dst[6];
ff_update_block_index(s);
- s->bdsp.clear_blocks(block[0]);
+ dst[0] = s->dest[0];
+ dst[1] = dst[0] + 8;
+ dst[2] = s->dest[0] + s->linesize * 8;
+ dst[3] = dst[2] + 8;
+ dst[4] = s->dest[1];
+ dst[5] = s->dest[2];
+ s->bdsp.clear_blocks(s->block[0]);
mb_pos = s->mb_x + s->mb_y * s->mb_width;
s->current_picture.mb_type[mb_pos] = MB_TYPE_INTRA;
s->current_picture.qscale_table[mb_pos] = v->pq;
- for (int i = 0; i < 4; i++) {
- s->current_picture.motion_val[1][s->block_index[i]][0] = 0;
- s->current_picture.motion_val[1][s->block_index[i]][1] = 0;
- }
+ s->current_picture.motion_val[1][s->block_index[0]][0] = 0;
+ s->current_picture.motion_val[1][s->block_index[0]][1] = 0;
// do actual MB decoding and displaying
cbp = get_vlc2(&v->s.gb, ff_msmp4_mb_i_vlc.table, MB_INTRA_VLC_BITS, 2);
v->s.ac_pred = get_bits1(&v->s.gb);
for (k = 0; k < 6; k++) {
- v->mb_type[0][s->block_index[k]] = 1;
-
val = ((cbp >> (5 - k)) & 1);
if (k < 4) {
@@ -2568,30 +2550,52 @@ static void vc1_decode_i_blocks(VC1Context *v)
}
cbp |= val << (5 - k);
- vc1_decode_i_block(v, block[k], k, val, (k < 4) ? v->codingset : v->codingset2);
+ vc1_decode_i_block(v, s->block[k], k, val, (k < 4) ? v->codingset : v->codingset2);
if (CONFIG_GRAY && k > 3 && (s->avctx->flags & AV_CODEC_FLAG_GRAY))
continue;
- v->vc1dsp.vc1_inv_trans_8x8(block[k]);
- }
-
- if (v->overlap && v->pq >= 9) {
- ff_vc1_i_overlap_filter(v);
- if (v->rangeredfrm)
- for (k = 0; k < 6; k++)
+ v->vc1dsp.vc1_inv_trans_8x8(s->block[k]);
+ if (v->pq >= 9 && v->overlap) {
+ if (v->rangeredfrm)
for (j = 0; j < 64; j++)
- block[k][j] <<= 1;
- vc1_put_blocks_clamped(v, 1);
- } else {
- if (v->rangeredfrm)
- for (k = 0; k < 6; k++)
+ s->block[k][j] <<= 1;
+ s->idsp.put_signed_pixels_clamped(s->block[k], dst[k],
+ k & 4 ? s->uvlinesize
+ : s->linesize);
+ } else {
+ if (v->rangeredfrm)
for (j = 0; j < 64; j++)
- block[k][j] = (block[k][j] - 64) << 1;
- vc1_put_blocks_clamped(v, 0);
+ s->block[k][j] = (s->block[k][j] - 64) << 1;
+ s->idsp.put_pixels_clamped(s->block[k], dst[k],
+ k & 4 ? s->uvlinesize
+ : s->linesize);
+ }
}
+ if (v->pq >= 9 && v->overlap) {
+ if (s->mb_x) {
+ v->vc1dsp.vc1_h_overlap(s->dest[0], s->linesize);
+ v->vc1dsp.vc1_h_overlap(s->dest[0] + 8 * s->linesize, s->linesize);
+ if (!CONFIG_GRAY || !(s->avctx->flags & AV_CODEC_FLAG_GRAY)) {
+ v->vc1dsp.vc1_h_overlap(s->dest[1], s->uvlinesize);
+ v->vc1dsp.vc1_h_overlap(s->dest[2], s->uvlinesize);
+ }
+ }
+ v->vc1dsp.vc1_h_overlap(s->dest[0] + 8, s->linesize);
+ v->vc1dsp.vc1_h_overlap(s->dest[0] + 8 * s->linesize + 8, s->linesize);
+ if (!s->first_slice_line) {
+ v->vc1dsp.vc1_v_overlap(s->dest[0], s->linesize);
+ v->vc1dsp.vc1_v_overlap(s->dest[0] + 8, s->linesize);
+ if (!CONFIG_GRAY || !(s->avctx->flags & AV_CODEC_FLAG_GRAY)) {
+ v->vc1dsp.vc1_v_overlap(s->dest[1], s->uvlinesize);
+ v->vc1dsp.vc1_v_overlap(s->dest[2], s->uvlinesize);
+ }
+ }
+ v->vc1dsp.vc1_v_overlap(s->dest[0] + 8 * s->linesize, s->linesize);
+ v->vc1dsp.vc1_v_overlap(s->dest[0] + 8 * s->linesize + 8, s->linesize);
+ }
if (v->s.loop_filter)
- ff_vc1_i_loop_filter(v);
+ ff_vc1_loop_filter_iblk(v, v->pq);
if (get_bits_count(&s->gb) > v->bits) {
ff_er_add_slice(&s->er, 0, 0, s->mb_x, s->mb_y, ER_MB_ERROR);
@@ -2599,11 +2603,6 @@ static void vc1_decode_i_blocks(VC1Context *v)
get_bits_count(&s->gb), v->bits);
return;
}
-
- v->topleft_blk_idx = (v->topleft_blk_idx + 1) % (v->end_mb_x + 2);
- v->top_blk_idx = (v->top_blk_idx + 1) % (v->end_mb_x + 2);
- v->left_blk_idx = (v->left_blk_idx + 1) % (v->end_mb_x + 2);
- v->cur_blk_idx = (v->cur_blk_idx + 1) % (v->end_mb_x + 2);
}
if (!v->s.loop_filter)
ff_mpeg_draw_horiz_band(s, s->mb_y * 16, 16);
@@ -2728,7 +2727,7 @@ static void vc1_decode_i_blocks_adv(VC1Context *v)
if (v->overlap && v->condover != CONDOVER_NONE)
ff_vc1_i_overlap_filter(v);
- vc1_put_blocks_clamped(v, 1);
+ vc1_put_signed_blocks_clamped(v);
if (v->s.loop_filter)
ff_vc1_i_loop_filter(v);
diff --git a/libavcodec/vc1_loopfilter.c b/libavcodec/vc1_loopfilter.c
index 5d5630db67..cea7dae7f8 100644
--- a/libavcodec/vc1_loopfilter.c
+++ b/libavcodec/vc1_loopfilter.c
@@ -31,6 +31,36 @@
#include "vc1.h"
#include "vc1dsp.h"
+void ff_vc1_loop_filter_iblk(VC1Context *v, int pq)
+{
+ MpegEncContext *s = &v->s;
+ int j;
+ if (!s->first_slice_line) {
+ v->vc1dsp.vc1_v_loop_filter16(s->dest[0], s->linesize, pq);
+ if (s->mb_x)
+ v->vc1dsp.vc1_h_loop_filter16(s->dest[0] - 16 * s->linesize, s->linesize, pq);
+ v->vc1dsp.vc1_h_loop_filter16(s->dest[0] - 16 * s->linesize + 8, s->linesize, pq);
+ if (!CONFIG_GRAY || !(s->avctx->flags & AV_CODEC_FLAG_GRAY))
+ for (j = 0; j < 2; j++) {
+ v->vc1dsp.vc1_v_loop_filter8(s->dest[j + 1], s->uvlinesize, pq);
+ if (s->mb_x)
+ v->vc1dsp.vc1_h_loop_filter8(s->dest[j + 1] - 8 * s->uvlinesize, s->uvlinesize, pq);
+ }
+ }
+ v->vc1dsp.vc1_v_loop_filter16(s->dest[0] + 8 * s->linesize, s->linesize, pq);
+
+ if (s->mb_y == s->end_mb_y - 1) {
+ if (s->mb_x) {
+ v->vc1dsp.vc1_h_loop_filter16(s->dest[0], s->linesize, pq);
+ if (!CONFIG_GRAY || !(s->avctx->flags & AV_CODEC_FLAG_GRAY)) {
+ v->vc1dsp.vc1_h_loop_filter8(s->dest[1], s->uvlinesize, pq);
+ v->vc1dsp.vc1_h_loop_filter8(s->dest[2], s->uvlinesize, pq);
+ }
+ }
+ v->vc1dsp.vc1_h_loop_filter16(s->dest[0] + 8, s->linesize, pq);
+ }
+}
+
static av_always_inline void vc1_h_overlap_filter(VC1Context *v, int16_t (*left_block)[64],
int16_t (*right_block)[64], int block_num)
{