From 04bbfe10348ce1010c0367c50ce76dda14d7c793 Mon Sep 17 00:00:00 2001 From: Kostya Shishkov Date: Wed, 8 Apr 2020 13:56:16 +0200 Subject: [PATCH] intel263: use codec-specific DSP functions --- nihav-indeo/src/codecs/intel263.rs | 298 ++++++++++++++++++++++++++++- 1 file changed, 295 insertions(+), 3 deletions(-) diff --git a/nihav-indeo/src/codecs/intel263.rs b/nihav-indeo/src/codecs/intel263.rs index b2e58cb..4a78cfb 100644 --- a/nihav-indeo/src/codecs/intel263.rs +++ b/nihav-indeo/src/codecs/intel263.rs @@ -4,10 +4,11 @@ use nihav_core::formats; use nihav_core::frame::*; use nihav_core::codecs::*; use nihav_codec_support::codecs::{MV, ZIGZAG}; +use nihav_codec_support::codecs::blockdsp; use nihav_codec_support::codecs::h263::*; +use nihav_codec_support::codecs::h263::code::{H263_INTERP_FUNCS, H263_INTERP_AVG_FUNCS}; use nihav_codec_support::codecs::h263::decoder::*; use nihav_codec_support::codecs::h263::data::*; -use nihav_codec_support::codecs::h263::code::H263BlockDSP; #[allow(dead_code)] struct Tables { @@ -19,11 +20,14 @@ struct Tables { mv_cb: Codebook, } +#[derive(Default)] +struct I263BlockDSP {} + struct Intel263Decoder { info: NACodecInfoRef, dec: H263BaseDecoder, tables: Tables, - bdsp: H263BlockDSP, + bdsp: I263BlockDSP, lastframe: Option, lastpts: Option, } @@ -37,6 +41,294 @@ struct Intel263BR<'a> { is_ipb: bool, } +const W1: i32 = 2841; +const W2: i32 = 2676; +const W3: i32 = 2408; +const W5: i32 = 1609; +const W6: i32 = 1108; +const W7: i32 = 565; +const W8: i32 = 181; + +const ROW_SHIFT: u8 = 8; +const COL_SHIFT: u8 = 14; + +#[allow(clippy::erasing_op)] +fn idct_row(row: &mut [i16]) { + let in0 = ((i32::from(row[0])) << 11) + (1 << (ROW_SHIFT - 1)); + let in1 = (i32::from(row[4])) << 11; + let in2 = i32::from(row[6]); + let in3 = i32::from(row[2]); + let in4 = i32::from(row[1]); + let in5 = i32::from(row[7]); + let in6 = i32::from(row[5]); + let in7 = i32::from(row[3]); + + let tmp = W7 * (in4 + in5); + let a4 = tmp + (W1 - W7) * in4; + let a5 = tmp - (W1 + W7) * in5; + + let tmp = W3 * (in6 + in7); + let a6 = tmp - (W3 - W5) * in6; + let a7 = tmp - (W3 + W5) * in7; + + let tmp = in0 + in1; + + let a0 = in0 - in1; + let t1 = W6 * (in2 + in3); + let a2 = t1 - (W2 + W6) * in2; + let a3 = t1 + (W2 - W6) * in3; + let b1 = a4 + a6; + + let b4 = a4 - a6; + let t2 = a5 - a7; + let b6 = a5 + a7; + let b7 = tmp + a3; + let b5 = tmp - a3; + let b3 = a0 + a2; + let b0 = a0 - a2; + let b2 = (W8 * (b4 + t2) + 128) >> 8; + let b4 = (W8 * (b4 - t2) + 128) >> 8; + + row[0] = ((b7 + b1) >> ROW_SHIFT) as i16; + row[7] = ((b7 - b1) >> ROW_SHIFT) as i16; + row[1] = ((b3 + b2) >> ROW_SHIFT) as i16; + row[6] = ((b3 - b2) >> ROW_SHIFT) as i16; + row[2] = ((b0 + b4) >> ROW_SHIFT) as i16; + row[5] = ((b0 - b4) >> ROW_SHIFT) as i16; + row[3] = ((b5 + b6) >> ROW_SHIFT) as i16; + row[4] = ((b5 - b6) >> ROW_SHIFT) as i16; +} + +#[allow(clippy::erasing_op)] +fn idct_col(blk: &mut [i16; 64], off: usize) { + let in0 = ((i32::from(blk[off + 0*8])) << 8) + (1 << (COL_SHIFT - 1)); + let in1 = (i32::from(blk[off + 4*8])) << 8; + let in2 = i32::from(blk[off + 6*8]); + let in3 = i32::from(blk[off + 2*8]); + let in4 = i32::from(blk[off + 1*8]); + let in5 = i32::from(blk[off + 7*8]); + let in6 = i32::from(blk[off + 5*8]); + let in7 = i32::from(blk[off + 3*8]); + + let tmp = W7 * (in4 + in5); + let a4 = (tmp + (W1 - W7) * in4) >> 3; + let a5 = (tmp - (W1 + W7) * in5) >> 3; + + let tmp = W3 * (in6 + in7); + let a6 = (tmp - (W3 - W5) * in6) >> 3; + let a7 = (tmp - (W3 + W5) * in7) >> 3; + + let tmp = in0 + in1; + + let a0 = in0 - in1; + let t1 = W6 * (in2 + in3); + let a2 = (t1 - (W2 + W6) * in2) >> 3; + let a3 = (t1 + (W2 - W6) * in3) >> 3; + let b1 = a4 + a6; + + let b4 = a4 - a6; + let t2 = a5 - a7; + let b6 = a5 + a7; + let b7 = tmp + a3; + let b5 = tmp - a3; + let b3 = a0 + a2; + let b0 = a0 - a2; + let b2 = (W8 * (b4 + t2) + 128) >> 8; + let b4 = (W8 * (b4 - t2) + 128) >> 8; + + blk[off + 0*8] = ((b7 + b1) >> COL_SHIFT) as i16; + blk[off + 7*8] = ((b7 - b1) >> COL_SHIFT) as i16; + blk[off + 1*8] = ((b3 + b2) >> COL_SHIFT) as i16; + blk[off + 6*8] = ((b3 - b2) >> COL_SHIFT) as i16; + blk[off + 2*8] = ((b0 + b4) >> COL_SHIFT) as i16; + blk[off + 5*8] = ((b0 - b4) >> COL_SHIFT) as i16; + blk[off + 3*8] = ((b5 + b6) >> COL_SHIFT) as i16; + blk[off + 4*8] = ((b5 - b6) >> COL_SHIFT) as i16; +} + +const FILTER_STRENGTH: [u8; 32] = [ + 1, 1, 2, 2, 3, 3, 4, 4, 4, 5, 5, 5, 6, 6, 7, 7, + 7, 8, 8, 8, 9, 9, 9, 10, 10, 10, 11, 11, 11, 12, 12, 12 +]; + +#[allow(clippy::erasing_op)] +fn deblock_hor(buf: &mut NAVideoBuffer, comp: usize, strength: u8, off: usize) { + let stride = buf.get_stride(comp); + let dptr = buf.get_data_mut().unwrap(); + let buf = dptr.as_mut_slice(); + for x in 0..8 { + let a = buf[off - 2 * stride + x] as i16; + let b = buf[off - 1 * stride + x] as i16; + let c = buf[off + 0 * stride + x] as i16; + let d = buf[off + 1 * stride + x] as i16; + let diff = (3 * (a - d) + 8 * (c - b)) / 16; + if (diff != 0) && (diff > -24) && (diff < 24) { + let d1a = (diff.abs() - 2 * (diff.abs() - (strength as i16)).max(0)).max(0); + let d1 = if diff < 0 { -d1a } else { d1a }; + + buf[off - 1 * stride + x] = (b + d1).max(0).min(255) as u8; + buf[off + 0 * stride + x] = (c - d1).max(0).min(255) as u8; + } + } +} + +fn deblock_ver(buf: &mut NAVideoBuffer, comp: usize, strength: u8, off: usize) { + let stride = buf.get_stride(comp); + let dptr = buf.get_data_mut().unwrap(); + let buf = dptr.as_mut_slice(); + for y in 0..8 { + let a = buf[off - 2 + y * stride] as i16; + let b = buf[off - 1 + y * stride] as i16; + let c = buf[off + 0 + y * stride] as i16; + let d = buf[off + 1 + y * stride] as i16; + let diff = (3 * (a - d) + 8 * (c - b)) / 16; + if (diff != 0) && (diff > -24) && (diff < 24) { + let d1a = (diff.abs() - 2 * (diff.abs() - (strength as i16)).max(0)).max(0); + let d1 = if diff < 0 { -d1a } else { d1a }; + + buf[off - 1 + y * stride] = (b + d1).max(0).min(255) as u8; + buf[off + y * stride] = (c - d1).max(0).min(255) as u8; + } + } +} + +impl BlockDSP for I263BlockDSP { + fn idct(&self, blk: &mut [i16; 64]) { + for i in 0..8 { idct_row(&mut blk[i*8..(i+1)*8]); } + for i in 0..8 { idct_col(blk, i); } + } + fn copy_blocks(&self, dst: &mut NAVideoBuffer, src: NAVideoBufferRef, xpos: usize, ypos: usize, mv: MV) { + let mode = ((mv.x & 1) + (mv.y & 1) * 2) as usize; + let cmode = (if (mv.x & 3) != 0 { 1 } else { 0 }) + (if (mv.y & 3) != 0 { 2 } else { 0 }); + + let mut dst = NASimpleVideoFrame::from_video_buf(dst).unwrap(); + + blockdsp::copy_block(&mut dst, src.clone(), 0, xpos, ypos, mv.x >> 1, mv.y >> 1, 16, 16, 0, 1, mode, H263_INTERP_FUNCS); + blockdsp::copy_block(&mut dst, src.clone(), 1, xpos >> 1, ypos >> 1, mv.x >> 2, mv.y >> 2, 8, 8, 0, 1, cmode, H263_INTERP_FUNCS); + blockdsp::copy_block(&mut dst, src.clone(), 2, xpos >> 1, ypos >> 1, mv.x >> 2, mv.y >> 2, 8, 8, 0, 1, cmode, H263_INTERP_FUNCS); + } + fn copy_blocks8x8(&self, dst: &mut NAVideoBuffer, src: NAVideoBufferRef, xpos: usize, ypos: usize, mvs: &[MV; 4]) { + let mut dst = NASimpleVideoFrame::from_video_buf(dst).unwrap(); + + for i in 0..4 { + let xadd = (i & 1) * 8; + let yadd = (i & 2) * 4; + let mode = ((mvs[i].x & 1) + (mvs[i].y & 1) * 2) as usize; + + blockdsp::copy_block(&mut dst, src.clone(), 0, xpos + xadd, ypos + yadd, mvs[i].x >> 1, mvs[i].y >> 1, 8, 8, 0, 1, mode, H263_INTERP_FUNCS); + } + + let sum_mv = mvs[0] + mvs[1] + mvs[2] + mvs[3]; + let cmx = (sum_mv.x >> 3) + H263_CHROMA_ROUND[(sum_mv.x & 0xF) as usize]; + let cmy = (sum_mv.y >> 3) + H263_CHROMA_ROUND[(sum_mv.y & 0xF) as usize]; + let mode = ((cmx & 1) + (cmy & 1) * 2) as usize; + for plane in 1..3 { + blockdsp::copy_block(&mut dst, src.clone(), plane, xpos >> 1, ypos >> 1, cmx >> 1, cmy >> 1, 8, 8, 0, 1, mode, H263_INTERP_FUNCS); + } + } + fn avg_blocks(&self, dst: &mut NAVideoBuffer, src: NAVideoBufferRef, xpos: usize, ypos: usize, mv: MV) { + let mode = ((mv.x & 1) + (mv.y & 1) * 2) as usize; + let cmode = (if (mv.x & 3) != 0 { 1 } else { 0 }) + (if (mv.y & 3) != 0 { 2 } else { 0 }); + + let mut dst = NASimpleVideoFrame::from_video_buf(dst).unwrap(); + + blockdsp::copy_block(&mut dst, src.clone(), 0, xpos, ypos, mv.x >> 1, mv.y >> 1, 16, 16, 0, 1, mode, H263_INTERP_AVG_FUNCS); + blockdsp::copy_block(&mut dst, src.clone(), 1, xpos >> 1, ypos >> 1, mv.x >> 2, mv.y >> 2, 8, 8, 0, 1, cmode, H263_INTERP_AVG_FUNCS); + blockdsp::copy_block(&mut dst, src.clone(), 2, xpos >> 1, ypos >> 1, mv.x >> 2, mv.y >> 2, 8, 8, 0, 1, cmode, H263_INTERP_AVG_FUNCS); + } + fn avg_blocks8x8(&self, dst: &mut NAVideoBuffer, src: NAVideoBufferRef, xpos: usize, ypos: usize, mvs: &[MV; 4]) { + let mut dst = NASimpleVideoFrame::from_video_buf(dst).unwrap(); + + for i in 0..4 { + let xadd = (i & 1) * 8; + let yadd = (i & 2) * 4; + let mode = ((mvs[i].x & 1) + (mvs[i].y & 1) * 2) as usize; + + blockdsp::copy_block(&mut dst, src.clone(), 0, xpos + xadd, ypos + yadd, mvs[i].x >> 1, mvs[i].y >> 1, 8, 8, 0, 1, mode, H263_INTERP_AVG_FUNCS); + } + + let sum_mv = mvs[0] + mvs[1] + mvs[2] + mvs[3]; + let cmx = (sum_mv.x >> 3) + H263_CHROMA_ROUND[(sum_mv.x & 0xF) as usize]; + let cmy = (sum_mv.y >> 3) + H263_CHROMA_ROUND[(sum_mv.y & 0xF) as usize]; + let mode = ((cmx & 1) + (cmy & 1) * 2) as usize; + for plane in 1..3 { + blockdsp::copy_block(&mut dst, src.clone(), plane, xpos >> 1, ypos >> 1, cmx >> 1, cmy >> 1, 8, 8, 0, 1, mode, H263_INTERP_AVG_FUNCS); + } + } + fn filter_row(&self, buf: &mut NAVideoBuffer, mb_y: usize, mb_w: usize, cbpi: &CBPInfo) { + let stride = buf.get_stride(0); + let mut off = buf.get_offset(0) + mb_y * 16 * stride; + for mb_x in 0..mb_w { + let coff = off; + let coded0 = cbpi.is_coded(mb_x, 0); + let coded1 = cbpi.is_coded(mb_x, 1); + let q = cbpi.get_q(mb_w + mb_x); + let str = if q < 32 { FILTER_STRENGTH[q as usize] } else { 0 }; + if mb_y != 0 { + if coded0 && cbpi.is_coded_top(mb_x, 0) { deblock_hor(buf, 0, str, coff); } + if coded1 && cbpi.is_coded_top(mb_x, 1) { deblock_hor(buf, 0, str, coff + 8); } + } + let coff = off + 8 * stride; + if cbpi.is_coded(mb_x, 2) && coded0 { deblock_hor(buf, 0, q, coff); } + if cbpi.is_coded(mb_x, 3) && coded1 { deblock_hor(buf, 0, q, coff + 8); } + off += 16; + } + let mut leftt = false; + let mut leftc = false; + let mut off = buf.get_offset(0) + mb_y * 16 * stride; + for mb_x in 0..mb_w { + let ctop0 = cbpi.is_coded_top(mb_x, 0); + let ctop1 = cbpi.is_coded_top(mb_x, 0); + let ccur0 = cbpi.is_coded(mb_x, 0); + let ccur1 = cbpi.is_coded(mb_x, 1); + let q = cbpi.get_q(mb_w + mb_x); + let str = if q < 32 { FILTER_STRENGTH[q as usize] } else { 0 }; + if mb_y != 0 { + let coff = off - 8 * stride; + let qtop = cbpi.get_q(mb_x); + let strtop = if qtop < 32 { FILTER_STRENGTH[qtop as usize] } else { 0 }; + if leftt && ctop0 { deblock_ver(buf, 0, strtop, coff); } + if ctop0 && ctop1 { deblock_ver(buf, 0, strtop, coff + 8); } + } + if leftc && ccur0 { deblock_ver(buf, 0, str, off); } + if ccur0 && ccur1 { deblock_ver(buf, 0, str, off + 8); } + leftt = ctop1; + leftc = ccur1; + off += 16; + } + let strideu = buf.get_stride(1); + let stridev = buf.get_stride(2); + let offu = buf.get_offset(1) + mb_y * 8 * strideu; + let offv = buf.get_offset(2) + mb_y * 8 * stridev; + if mb_y != 0 { + for mb_x in 0..mb_w { + let ctu = cbpi.is_coded_top(mb_x, 4); + let ccu = cbpi.is_coded(mb_x, 4); + let ctv = cbpi.is_coded_top(mb_x, 5); + let ccv = cbpi.is_coded(mb_x, 5); + let q = cbpi.get_q(mb_w + mb_x); + let str = if q < 32 { FILTER_STRENGTH[q as usize] } else { 0 }; + if ctu && ccu { deblock_hor(buf, 1, str, offu + mb_x * 8); } + if ctv && ccv { deblock_hor(buf, 2, str, offv + mb_x * 8); } + } + let mut leftu = false; + let mut leftv = false; + let offu = buf.get_offset(1) + (mb_y - 1) * 8 * strideu; + let offv = buf.get_offset(2) + (mb_y - 1) * 8 * stridev; + for mb_x in 0..mb_w { + let ctu = cbpi.is_coded_top(mb_x, 4); + let ctv = cbpi.is_coded_top(mb_x, 5); + let qt = cbpi.get_q(mb_x); + let strt = if qt < 32 { FILTER_STRENGTH[qt as usize] } else { 0 }; + if leftu && ctu { deblock_ver(buf, 1, strt, offu + mb_x * 8); } + if leftv && ctv { deblock_ver(buf, 2, strt, offv + mb_x * 8); } + leftu = ctu; + leftv = ctv; + } + } + } +} + fn check_marker<'a>(br: &mut BitReader<'a>) -> DecoderResult<()> { let mark = br.read(1)?; validate!(mark == 1); @@ -369,7 +661,7 @@ impl Intel263Decoder { info: NACodecInfo::new_dummy(), dec: H263BaseDecoder::new(true), tables, - bdsp: H263BlockDSP::new(), + bdsp: I263BlockDSP::default(), lastframe: None, lastpts: None, } -- 2.30.2