From 2a7d7bef6280f012e8ed86c4a118da422bf653df Mon Sep 17 00:00:00 2001 From: "Ronald S. Bultje" Date: Thu, 30 Apr 2015 15:33:36 +0200 Subject: [PATCH] vp9: fix mask_edges and filter_plane_rows/cols() for 440. --- libavcodec/vp9.c | 155 +++++++++++++++++++++-------------------------- 1 file changed, 70 insertions(+), 85 deletions(-) diff --git a/libavcodec/vp9.c b/libavcodec/vp9.c index dbedf334ff..71ed4e64d8 100644 --- a/libavcodec/vp9.c +++ b/libavcodec/vp9.c @@ -2853,11 +2853,14 @@ static void inter_recon(AVCodecContext *ctx) } } -static av_always_inline void mask_edges(uint8_t (*mask)[8][4], int is_uv, +static av_always_inline void mask_edges(uint8_t (*mask)[8][4], int ss_h, int ss_v, int row_and_7, int col_and_7, int w, int h, int col_end, int row_end, enum TxfmMode tx, int skip_inter) { + static const unsigned wide_filter_col_mask[2] = { 0x11, 0x01 }; + static const unsigned wide_filter_row_mask[2] = { 0x03, 0x07 }; + // FIXME I'm pretty sure all loops can be replaced by a single LUT if // we make VP9Filter.mask uint64_t (i.e. row/col all single variable) // and make the LUT 5-indexed (bl, bp, is_uv, tx and row/col), and then @@ -2868,14 +2871,14 @@ static av_always_inline void mask_edges(uint8_t (*mask)[8][4], int is_uv, // a time, and we only use the topleft block's mode information to set // things like block strength. Thus, for any block size smaller than // 16x16, ignore the odd portion of the block. - if (tx == TX_4X4 && is_uv) { - if (h == 1) { + if (tx == TX_4X4 && (ss_v | ss_h)) { + if (h == ss_v) { if (row_and_7 & 1) return; if (!row_end) h += 1; } - if (w == 1) { + if (w == ss_h) { if (col_and_7 & 1) return; if (!col_end) @@ -2885,58 +2888,46 @@ static av_always_inline void mask_edges(uint8_t (*mask)[8][4], int is_uv, if (tx == TX_4X4 && !skip_inter) { int t = 1 << col_and_7, m_col = (t << w) - t, y; - int m_col_odd = (t << (w - 1)) - t; - // on 32-px edges, use the 8-px wide loopfilter; else, use 4-px wide - if (is_uv) { - int m_row_8 = m_col & 0x01, m_row_4 = m_col - m_row_8; + int m_row_8 = m_col & wide_filter_col_mask[ss_h], m_row_4 = m_col - m_row_8; - for (y = row_and_7; y < h + row_and_7; y++) { - int col_mask_id = 2 - !(y & 7); + for (y = row_and_7; y < h + row_and_7; y++) { + int col_mask_id = 2 - !(y & wide_filter_row_mask[ss_v]); - mask[0][y][1] |= m_row_8; - mask[0][y][2] |= m_row_4; - // for odd lines, if the odd col is not being filtered, - // skip odd row also: - // .---. <-- a - // | | - // |___| <-- b - // ^ ^ - // c d - // - // if a/c are even row/col and b/d are odd, and d is skipped, - // e.g. right edge of size-66x66.webm, then skip b also (bug) - if ((col_end & 1) && (y & 1)) { - mask[1][y][col_mask_id] |= m_col_odd; - } else { - mask[1][y][col_mask_id] |= m_col; - } + mask[0][y][1] |= m_row_8; + mask[0][y][2] |= m_row_4; + // for odd lines, if the odd col is not being filtered, + // skip odd row also: + // .---. <-- a + // | | + // |___| <-- b + // ^ ^ + // c d + // + // if a/c are even row/col and b/d are odd, and d is skipped, + // e.g. right edge of size-66x66.webm, then skip b also (bug) + if ((ss_h & ss_v) && (col_end & 1) && (y & 1)) { + mask[1][y][col_mask_id] |= (t << (w - 1)) - t; + } else { + mask[1][y][col_mask_id] |= m_col; } - } else { - int m_row_8 = m_col & 0x11, m_row_4 = m_col - m_row_8; - - for (y = row_and_7; y < h + row_and_7; y++) { - int col_mask_id = 2 - !(y & 3); - - mask[0][y][1] |= m_row_8; // row edge - mask[0][y][2] |= m_row_4; - mask[1][y][col_mask_id] |= m_col; // col edge + if (!ss_h) mask[0][y][3] |= m_col; + if (!ss_v) mask[1][y][3] |= m_col; - } } } else { int y, t = 1 << col_and_7, m_col = (t << w) - t; if (!skip_inter) { int mask_id = (tx == TX_8X8); - int l2 = tx + is_uv - 1, step1d = 1 << l2; static const unsigned masks[4] = { 0xff, 0x55, 0x11, 0x01 }; + int l2 = tx + ss_h - 1, step1d; int m_row = m_col & masks[l2]; // at odd UV col/row edges tx16/tx32 loopfilter edges, force // 8wd loopfilter to prevent going off the visible edge. - if (is_uv && tx > TX_8X8 && (w ^ (w - 1)) == 1) { + if (ss_h && tx > TX_8X8 && (w ^ (w - 1)) == 1) { int m_row_16 = ((t << (w - 1)) - t) & masks[l2]; int m_row_8 = m_row - m_row_16; @@ -2949,7 +2940,9 @@ static av_always_inline void mask_edges(uint8_t (*mask)[8][4], int is_uv, mask[0][y][mask_id] |= m_row; } - if (is_uv && tx > TX_8X8 && (h ^ (h - 1)) == 1) { + l2 = tx + ss_v - 1; + step1d = 1 << l2; + if (ss_v && tx > TX_8X8 && (h ^ (h - 1)) == 1) { for (y = row_and_7; y < h + row_and_7 - 1; y += step1d) mask[1][y][0] |= m_col; if (y - row_and_7 == h - 1) @@ -2961,27 +2954,19 @@ static av_always_inline void mask_edges(uint8_t (*mask)[8][4], int is_uv, } else if (tx != TX_4X4) { int mask_id; - mask_id = (tx == TX_8X8) || (is_uv && h == 1); + mask_id = (tx == TX_8X8) || (h == ss_v); mask[1][row_and_7][mask_id] |= m_col; - mask_id = (tx == TX_8X8) || (is_uv && w == 1); + mask_id = (tx == TX_8X8) || (w == ss_h); for (y = row_and_7; y < h + row_and_7; y++) mask[0][y][mask_id] |= t; - } else if (is_uv) { - int t8 = t & 0x01, t4 = t - t8; - - for (y = row_and_7; y < h + row_and_7; y++) { - mask[0][y][2] |= t4; - mask[0][y][1] |= t8; - } - mask[1][row_and_7][2 - !(row_and_7 & 7)] |= m_col; } else { - int t8 = t & 0x11, t4 = t - t8; + int t8 = t & wide_filter_col_mask[ss_h], t4 = t - t8; for (y = row_and_7; y < h + row_and_7; y++) { mask[0][y][2] |= t4; mask[0][y][1] |= t8; } - mask[1][row_and_7][2 - !(row_and_7 & 3)] |= m_col; + mask[1][row_and_7][2 - !(row_and_7 & wide_filter_row_mask[ss_v])] |= m_col; } } } @@ -3131,9 +3116,9 @@ static void decode_b(AVCodecContext *ctx, int row, int col, int skip_inter = !b->intra && b->skip, col7 = s->col7, row7 = s->row7; setctx_2d(&lflvl->level[row7 * 8 + col7], w4, h4, 8, lvl); - mask_edges(lflvl->mask[0], 0, row7, col7, x_end, y_end, 0, 0, b->tx, skip_inter); + mask_edges(lflvl->mask[0], 0, 0, row7, col7, x_end, y_end, 0, 0, b->tx, skip_inter); if (s->ss_h || s->ss_v) - mask_edges(lflvl->mask[1], 1, row7, col7, x_end, y_end, + mask_edges(lflvl->mask[1], s->ss_h, s->ss_v, row7, col7, x_end, y_end, s->cols & 1 && col + w4 >= s->cols ? s->cols & 7 : 0, s->rows & 1 && row + h4 >= s->rows ? s->rows & 7 : 0, b->uvtx, skip_inter); @@ -3286,20 +3271,20 @@ static void decode_sb_mem(AVCodecContext *ctx, int row, int col, struct VP9Filte } } -static av_always_inline void filter_plane_cols(VP9Context *s, int col, int ss, +static av_always_inline void filter_plane_cols(VP9Context *s, int col, int ss_h, int ss_v, uint8_t *lvl, uint8_t (*mask)[4], uint8_t *dst, ptrdiff_t ls) { int y, x; // filter edges between columns (e.g. block1 | block2) - for (y = 0; y < 8; y += 2 << ss, dst += 16 * ls, lvl += 16 << ss) { - uint8_t *ptr = dst, *l = lvl, *hmask1 = mask[y], *hmask2 = mask[y + (1 << ss)]; + for (y = 0; y < 8; y += 2 << ss_v, dst += 16 * ls, lvl += 16 << ss_v) { + uint8_t *ptr = dst, *l = lvl, *hmask1 = mask[y], *hmask2 = mask[y + 1 + ss_v]; unsigned hm1 = hmask1[0] | hmask1[1] | hmask1[2], hm13 = hmask1[3]; unsigned hm2 = hmask2[1] | hmask2[2], hm23 = hmask2[3]; unsigned hm = hm1 | hm2 | hm13 | hm23; - for (x = 1; hm & ~(x - 1); x <<= 1, ptr += 8 >> ss) { + for (x = 1; hm & ~(x - 1); x <<= 1, ptr += 8 >> ss_h) { if (col || x > 1) { if (hm1 & x) { int L = *l, H = L >> 4; @@ -3307,13 +3292,13 @@ static av_always_inline void filter_plane_cols(VP9Context *s, int col, int ss, if (hmask1[0] & x) { if (hmask2[0] & x) { - av_assert2(l[8 << ss] == L); + av_assert2(l[8 << ss_v] == L); s->dsp.loop_filter_16[0](ptr, ls, E, I, H); } else { s->dsp.loop_filter_8[2][0](ptr, ls, E, I, H); } } else if (hm2 & x) { - L = l[8 << ss]; + L = l[8 << ss_v]; H |= (L >> 4) << 8; E |= s->filter.mblim_lut[L] << 8; I |= s->filter.lim_lut[L] << 8; @@ -3325,14 +3310,14 @@ static av_always_inline void filter_plane_cols(VP9Context *s, int col, int ss, [0](ptr, ls, E, I, H); } } else if (hm2 & x) { - int L = l[8 << ss], H = L >> 4; + int L = l[8 << ss_v], H = L >> 4; int E = s->filter.mblim_lut[L], I = s->filter.lim_lut[L]; s->dsp.loop_filter_8[!!(hmask2[1] & x)] [0](ptr + 8 * ls, ls, E, I, H); } } - if (ss) { + if (ss_h) { if (x & 0xAA) l += 2; } else { @@ -3341,7 +3326,7 @@ static av_always_inline void filter_plane_cols(VP9Context *s, int col, int ss, int E = s->filter.mblim_lut[L], I = s->filter.lim_lut[L]; if (hm23 & x) { - L = l[8]; + L = l[8 << ss_v]; H |= (L >> 4) << 8; E |= s->filter.mblim_lut[L] << 8; I |= s->filter.lim_lut[L] << 8; @@ -3350,7 +3335,7 @@ static av_always_inline void filter_plane_cols(VP9Context *s, int col, int ss, s->dsp.loop_filter_8[0][0](ptr + 4, ls, E, I, H); } } else if (hm23 & x) { - int L = l[8], H = L >> 4; + int L = l[8 << ss_v], H = L >> 4; int E = s->filter.mblim_lut[L], I = s->filter.lim_lut[L]; s->dsp.loop_filter_8[0][0](ptr + 8 * ls + 4, ls, E, I, H); @@ -3361,7 +3346,7 @@ static av_always_inline void filter_plane_cols(VP9Context *s, int col, int ss, } } -static av_always_inline void filter_plane_rows(VP9Context *s, int row, int ss, +static av_always_inline void filter_plane_rows(VP9Context *s, int row, int ss_h, int ss_v, uint8_t *lvl, uint8_t (*mask)[4], uint8_t *dst, ptrdiff_t ls) { @@ -3370,50 +3355,50 @@ static av_always_inline void filter_plane_rows(VP9Context *s, int row, int ss, // block1 // filter edges between rows (e.g. ------) // block2 - for (y = 0; y < 8; y++, dst += 8 * ls >> ss) { + for (y = 0; y < 8; y++, dst += 8 * ls >> ss_v) { uint8_t *ptr = dst, *l = lvl, *vmask = mask[y]; unsigned vm = vmask[0] | vmask[1] | vmask[2], vm3 = vmask[3]; - for (x = 1; vm & ~(x - 1); x <<= (2 << ss), ptr += 16, l += 2 << ss) { + for (x = 1; vm & ~(x - 1); x <<= (2 << ss_h), ptr += 16, l += 2 << ss_h) { if (row || y) { if (vm & x) { int L = *l, H = L >> 4; int E = s->filter.mblim_lut[L], I = s->filter.lim_lut[L]; if (vmask[0] & x) { - if (vmask[0] & (x << (1 + ss))) { - av_assert2(l[1 + ss] == L); + if (vmask[0] & (x << (1 + ss_h))) { + av_assert2(l[1 + ss_h] == L); s->dsp.loop_filter_16[1](ptr, ls, E, I, H); } else { s->dsp.loop_filter_8[2][1](ptr, ls, E, I, H); } - } else if (vm & (x << (1 + ss))) { - L = l[1 + ss]; + } else if (vm & (x << (1 + ss_h))) { + L = l[1 + ss_h]; H |= (L >> 4) << 8; E |= s->filter.mblim_lut[L] << 8; I |= s->filter.lim_lut[L] << 8; s->dsp.loop_filter_mix2[!!(vmask[1] & x)] - [!!(vmask[1] & (x << (1 + ss)))] + [!!(vmask[1] & (x << (1 + ss_h)))] [1](ptr, ls, E, I, H); } else { s->dsp.loop_filter_8[!!(vmask[1] & x)] [1](ptr, ls, E, I, H); } - } else if (vm & (x << (1 + ss))) { - int L = l[1 + ss], H = L >> 4; + } else if (vm & (x << (1 + ss_h))) { + int L = l[1 + ss_h], H = L >> 4; int E = s->filter.mblim_lut[L], I = s->filter.lim_lut[L]; - s->dsp.loop_filter_8[!!(vmask[1] & (x << (1 + ss)))] + s->dsp.loop_filter_8[!!(vmask[1] & (x << (1 + ss_h)))] [1](ptr + 8, ls, E, I, H); } } - if (!ss) { + if (!ss_v) { if (vm3 & x) { int L = *l, H = L >> 4; int E = s->filter.mblim_lut[L], I = s->filter.lim_lut[L]; - if (vm3 & (x << 1)) { - L = l[1]; + if (vm3 & (x << (1 + ss_h))) { + L = l[1 + ss_h]; H |= (L >> 4) << 8; E |= s->filter.mblim_lut[L] << 8; I |= s->filter.lim_lut[L] << 8; @@ -3421,15 +3406,15 @@ static av_always_inline void filter_plane_rows(VP9Context *s, int row, int ss, } else { s->dsp.loop_filter_8[0][1](ptr + ls * 4, ls, E, I, H); } - } else if (vm3 & (x << 1)) { - int L = l[1], H = L >> 4; + } else if (vm3 & (x << (1 + ss_h))) { + int L = l[1 + ss_h], H = L >> 4; int E = s->filter.mblim_lut[L], I = s->filter.lim_lut[L]; s->dsp.loop_filter_8[0][1](ptr + ls * 4 + 8, ls, E, I, H); } } } - if (ss) { + if (ss_v) { if (y & 1) lvl += 16; } else { @@ -3454,13 +3439,13 @@ static void loopfilter_sb(AVCodecContext *ctx, struct VP9Filter *lflvl, // 8 pixel blocks, and we won't always do that (we want at least 16px // to use SSE2 optimizations, perhaps 32 for AVX2) - filter_plane_cols(s, col, 0, lflvl->level, lflvl->mask[0][0], dst, ls_y); - filter_plane_rows(s, row, 0, lflvl->level, lflvl->mask[0][1], dst, ls_y); + filter_plane_cols(s, col, 0, 0, lflvl->level, lflvl->mask[0][0], dst, ls_y); + filter_plane_rows(s, row, 0, 0, lflvl->level, lflvl->mask[0][1], dst, ls_y); for (p = 0; p < 2; p++) { dst = f->data[1 + p] + uvoff; - filter_plane_cols(s, col, s->ss_h, lflvl->level, uv_masks[0], dst, ls_uv); - filter_plane_rows(s, row, s->ss_v, lflvl->level, uv_masks[1], dst, ls_uv); + filter_plane_cols(s, col, s->ss_h, s->ss_v, lflvl->level, uv_masks[0], dst, ls_uv); + filter_plane_rows(s, row, s->ss_h, s->ss_v, lflvl->level, uv_masks[1], dst, ls_uv); } }