From 892ecb90cfce0eb7c47389b375c69416b4eae7c0 Mon Sep 17 00:00:00 2001 From: Kostya Shishkov Date: Sat, 30 May 2026 12:15:00 +0200 Subject: [PATCH] nihav_indeo: rewrite some bits of Indeo 4/5 using more modern Rust --- nihav-indeo/src/codecs/indeo4.rs | 99 +++----------------- nihav-indeo/src/codecs/indeo5.rs | 9 +- nihav-indeo/src/codecs/ivibr.rs | 75 ++++++---------- nihav-indeo/src/codecs/ividsp.rs | 149 ++++++++----------------------- 4 files changed, 77 insertions(+), 255 deletions(-) diff --git a/nihav-indeo/src/codecs/indeo4.rs b/nihav-indeo/src/codecs/indeo4.rs index c403313..6dd20ec 100644 --- a/nihav-indeo/src/codecs/indeo4.rs +++ b/nihav-indeo/src/codecs/indeo4.rs @@ -6,26 +6,13 @@ use nihav_codec_support::codecs::ZIGZAG; use super::ivi::*; use super::ivibr::*; -#[inline(always)] -fn mclip8(a: i32) -> u8 { - if (a as u16) > 255 { !(a >> 16) as u8 } - else { a as u8 } -} - struct Indeo4Parser { mb_cb: IVICodebook, blk_cb: IVICodebook, } fn calc_quant(glob_q: u32, qd: i16) -> u8 { - let q = (glob_q as i16) + qd; - if q < 0 { - 0 - } else if q > 31 { - 31 - } else { - q as u8 - } + ((glob_q as i16) + qd).clamp(0, 31) as u8 } impl Indeo4Parser { @@ -309,81 +296,23 @@ impl IndeoXParser for Indeo4Parser { } fn recombine_plane(&mut self, src: &[i16], sstride: usize, dst: &mut [u8], dstride: usize, w: usize, h: usize) { -/* let mut idx0 = 0; - let mut idx1 = w / 2; - let mut idx2 = (h / 2) * sstride; - let mut idx3 = idx2 + idx1; - let mut oidx0 = 0; - let mut oidx1 = dstride; - - for _ in 0..(h/2) { - for x in 0..(w/2) { - let p0 = src[idx0 + x]; - let p1 = src[idx1 + x]; - let p2 = src[idx2 + x]; - let p3 = src[idx3 + x]; + let (first, second) = src.split_at(sstride * (h / 2)); + for (dstrip, (line0, line1)) in dst.chunks_exact_mut(dstride * 2) + .zip(first.chunks_exact(sstride).zip(second.chunks_exact(sstride))) { + let (dline0, dline1) = dstrip.split_at_mut(dstride); + let (h00, h01) = line0.split_at(w / 2); + let (h10, h11) = line1.split_at(w / 2); + for ((dst0, dst1), ((&p0, &p1), (&p2, &p3))) in dline0.chunks_exact_mut(2) + .zip(dline1.chunks_exact_mut(2)) + .zip(h00.iter().zip(h01.iter()).zip(h10.iter().zip(h11.iter()))) { let s0 = p0 + p2; let d0 = p0 - p2; let s1 = p1 + p3; let d1 = p1 - p3; - dst[oidx0 + x * 2 + 0] = clip8(((s0 + s1 + 2) >> 2) + 128); - dst[oidx0 + x * 2 + 1] = clip8(((d0 + d1 + 2) >> 2) + 128); - dst[oidx1 + x * 2 + 0] = clip8(((s0 - s1 + 2) >> 2) + 128); - dst[oidx1 + x * 2 + 1] = clip8(((d0 - d1 + 2) >> 2) + 128); - } - idx0 += sstride; - idx1 += sstride; - idx2 += sstride; - idx3 += sstride; - oidx0 += dstride * 2; - oidx1 += dstride * 2; - }*/ - unsafe { - let hw = (w / 2) as isize; - let hh = (h / 2) as isize; - let mut band0 = src.as_ptr(); - let mut band1 = band0.offset(hw); - let mut band2 = band0.add((h / 2) * sstride); - let mut band3 = band2.offset(hw); - let mut dst0 = dst.as_mut_ptr(); - let mut dst1 = dst0.add(dstride); - for _ in 0..hh { - let mut b0_ptr = band0; - let mut b1_ptr = band1; - let mut b2_ptr = band2; - let mut b3_ptr = band3; - let mut d0_ptr = dst0; - let mut d1_ptr = dst1; - for _ in 0..hw { - let p0 = i32::from(*b0_ptr); - let p1 = i32::from(*b1_ptr); - let p2 = i32::from(*b2_ptr); - let p3 = i32::from(*b3_ptr); - let s0 = p0.wrapping_add(p2); - let s1 = p1.wrapping_add(p3); - let d0 = p0.wrapping_sub(p2); - let d1 = p1.wrapping_sub(p3); - let o0 = s0.wrapping_add(s1).wrapping_add(2); - let o1 = d0.wrapping_add(d1).wrapping_add(2); - let o2 = s0.wrapping_sub(s1).wrapping_add(2); - let o3 = d0.wrapping_sub(d1).wrapping_add(2); - *d0_ptr.offset(0) = mclip8((o0 >> 2).wrapping_add(128)); - *d0_ptr.offset(1) = mclip8((o1 >> 2).wrapping_add(128)); - *d1_ptr.offset(0) = mclip8((o2 >> 2).wrapping_add(128)); - *d1_ptr.offset(1) = mclip8((o3 >> 2).wrapping_add(128)); - b0_ptr = b0_ptr.offset(1); - b1_ptr = b1_ptr.offset(1); - b2_ptr = b2_ptr.offset(1); - b3_ptr = b3_ptr.offset(1); - d0_ptr = d0_ptr.offset(2); - d1_ptr = d1_ptr.offset(2); - } - band0 = band0.add(sstride); - band1 = band1.add(sstride); - band2 = band2.add(sstride); - band3 = band3.add(sstride); - dst0 = dst0.add(dstride * 2); - dst1 = dst1.add(dstride * 2); + dst0[0] = clip8(((s0 + s1 + 2) >> 2) + 128); + dst0[1] = clip8(((d0 + d1 + 2) >> 2) + 128); + dst1[0] = clip8(((s0 - s1 + 2) >> 2) + 128); + dst1[1] = clip8(((d0 - d1 + 2) >> 2) + 128); } } } diff --git a/nihav-indeo/src/codecs/indeo5.rs b/nihav-indeo/src/codecs/indeo5.rs index 4f0887d..add6826 100644 --- a/nihav-indeo/src/codecs/indeo5.rs +++ b/nihav-indeo/src/codecs/indeo5.rs @@ -7,14 +7,7 @@ use super::ivi::*; use super::ivibr::*; fn calc_quant(glob_q: u32, qd: i16) -> usize { - let qq = (glob_q as i16) + qd; - if qq < 0 { - 0 - } else if qq > 23 { - 23 - } else { - qq as usize - } + ((glob_q as i16) + qd).clamp(0, 23) as usize } struct Indeo5Parser { diff --git a/nihav-indeo/src/codecs/ivibr.rs b/nihav-indeo/src/codecs/ivibr.rs index 89cd4bf..f820f4a 100644 --- a/nihav-indeo/src/codecs/ivibr.rs +++ b/nihav-indeo/src/codecs/ivibr.rs @@ -1,6 +1,6 @@ use std::mem; use nihav_core::io::bitreader::*; -//use io::intcode::*; +//use nihav_core::io::intcode::*; use nihav_core::codecs::*; use nihav_core::frame::NABufferRef; use super::ivi::*; @@ -31,18 +31,6 @@ impl IVICodebook { pub const IVI_CB_ZERO: IVICodebook = IVICodebook { len: 0, bits: [0; 16], offs: [0; 16] }; -const IVI_REV0: [u32; 1] = [0]; -const IVI_REV1: [u32; 2] = [0, 1]; -const IVI_REV2: [u32; 4] = [0, 2, 1, 3]; -const IVI_REV3: [u32; 8] = [0, 4, 2, 6, 1, 5, 3, 7]; -const IVI_REV4: [u32; 16] = [ 0, 8, 4, 12, 2, 10, 6, 14, 1, 9, 5, 13, 3, 11, 7, 15]; -const IVI_REV5: [u32; 32] = [ 0, 16, 8, 24, 4, 20, 12, 28, 2, 18, 10, 26, 6, 22, 14, 30, 1, 17, 9, 25, 5, 21, 13, 29, 3, 19, 11, 27, 7, 23, 15, 31]; -const IVI_REV6: [u32; 64] = [ 0, 32, 16, 48, 8, 40, 24, 56, 4, 36, 20, 52, 12, 44, 28, 60, 2, 34, 18, 50, 10, 42, 26, 58, 6, 38, 22, 54, 14, 46, 30, 62, 1, 33, 17, 49, 9, 41, 25, 57, 5, 37, 21, 53, 13, 45, 29, 61, 3, 35, 19, 51, 11, 43, 27, 59, 7, 39, 23, 55, 15, 47, 31, 63]; -const IVI_REV7: [u32; 128] = [ 0, 64, 32, 96, 16, 80, 48, 112, 8, 72, 40, 104, 24, 88, 56, 120, 4, 68, 36, 100, 20, 84, 52, 116, 12, 76, 44, 108, 28, 92, 60, 124, 2, 66, 34, 98, 18, 82, 50, 114, 10, 74, 42, 106, 26, 90, 58, 122, 6, 70, 38, 102, 22, 86, 54, 118, 14, 78, 46, 110, 30, 94, 62, 126, 1, 65, 33, 97, 17, 81, 49, 113, 9, 73, 41, 105, 25, 89, 57, 121, 5, 69, 37, 101, 21, 85, 53, 117, 13, 77, 45, 109, 29, 93, 61, 125, 3, 67, 35, 99, 19, 83, 51, 115, 11, 75, 43, 107, 27, 91, 59, 123, 7, 71, 39, 103, 23, 87, 55, 119, 15, 79, 47, 111, 31, 95, 63, 127]; -const IVI_REV8: [u32; 256] = [ 0, 128, 64, 192, 32, 160, 96, 224, 16, 144, 80, 208, 48, 176, 112, 240, 8, 136, 72, 200, 40, 168, 104, 232, 24, 152, 88, 216, 56, 184, 120, 248, 4, 132, 68, 196, 36, 164, 100, 228, 20, 148, 84, 212, 52, 180, 116, 244, 12, 140, 76, 204, 44, 172, 108, 236, 28, 156, 92, 220, 60, 188, 124, 252, 2, 130, 66, 194, 34, 162, 98, 226, 18, 146, 82, 210, 50, 178, 114, 242, 10, 138, 74, 202, 42, 170, 106, 234, 26, 154, 90, 218, 58, 186, 122, 250, 6, 134, 70, 198, 38, 166, 102, 230, 22, 150, 86, 214, 54, 182, 118, 246, 14, 142, 78, 206, 46, 174, 110, 238, 30, 158, 94, 222, 62, 190, 126, 254, 1, 129, 65, 193, 33, 161, 97, 225, 17, 145, 81, 209, 49, 177, 113, 241, 9, 137, 73, 201, 41, 169, 105, 233, 25, 153, 89, 217, 57, 185, 121, 249, 5, 133, 69, 197, 37, 165, 101, 229, 21, 149, 85, 213, 53, 181, 117, 245, 13, 141, 77, 205, 45, 173, 109, 237, 29, 157, 93, 221, 61, 189, 125, 253, 3, 131, 67, 195, 35, 163, 99, 227, 19, 147, 83, 211, 51, 179, 115, 243, 11, 139, 75, 203, 43, 171, 107, 235, 27, 155, 91, 219, 59, 187, 123, 251, 7, 135, 71, 199, 39, 167, 103, 231, 23, 151, 87, 215, 55, 183, 119, 247, 15, 143, 79, 207, 47, 175, 111, 239, 31, 159, 95, 223, 63, 191, 127, 255]; - -const IVI_REVS: [&[u32]; 9] = [ &IVI_REV0, &IVI_REV1, &IVI_REV2, &IVI_REV3, &IVI_REV4, &IVI_REV5, &IVI_REV6, &IVI_REV7, &IVI_REV8]; - pub trait IVICodebookReader { fn read_ivi_codebook_desc(&mut self, mb_cb: bool, try_default: bool) -> DecoderResult; fn read_ivi_cb(&mut self, cb: &IVICodebook) -> BitReaderResult; @@ -98,11 +86,11 @@ impl<'a> IVICodebookReader for BitReader<'a> { let nbits = cb.bits[pfx]; let base = cb.offs[pfx]; let rval = self.read(nbits)?; - let add = IVI_REVS[nbits as usize][rval as usize]; + let add = reverse_bits(rval, nbits); Ok(base + add) } else { let nbits = cb.bits[0]; - Ok(IVI_REVS[nbits as usize][self.read(nbits)? as usize]) + Ok(reverse_bits(self.read(nbits)?, nbits)) } } #[inline(always)] @@ -173,7 +161,8 @@ fn read_trans_band_header(br: &mut BitReader, w: usize, h: usize, dst: &mut [i16 cb = cb.init(); br.align(); -let tile_start = br.tell(); + let tile_start = br.tell(); + let empty = br.read_bool()?; if !empty { br.read_bool()?; @@ -182,7 +171,8 @@ let tile_start = br.tell(); len = br.read(24)? as usize; } br.align(); -let tile_end = tile_start + len * 8; + + let tile_end = tile_start + len * 8; let first_val = br.read_bool()?; @@ -226,7 +216,6 @@ let tile_end = tile_start + len * 8; Ok(()) } -#[allow(clippy::collapsible_else_if)] fn decode_block8x8(br: &mut BitReader, blk_cb: &IVICodebook, rvmap: &RVMap, tables: &TxParams8x8, is_intra: bool, is_2d: bool, prev_dc: &mut i32, quant: u8, coeffs: &mut [i32; 64], transform: TrFunc) -> DecoderResult<()> { let mut idx: isize = -1; let quant_mat = if is_intra { tables.quant_intra } else { tables.quant_inter }; @@ -278,7 +267,6 @@ fn decode_block8x8(br: &mut BitReader, blk_cb: &IVICodebook, rvmap: &RVMap, tabl Ok(()) } -#[allow(clippy::collapsible_else_if)] fn decode_block4x4(br: &mut BitReader, blk_cb: &IVICodebook, rvmap: &RVMap, tables: &TxParams4x4, is_intra: bool, is_2d: bool, prev_dc: &mut i32, quant: u8, coeffs: &mut [i32; 64], transform: TrFunc) -> DecoderResult<()> { let mut idx: isize = -1; let quant_mat = if is_intra { tables.quant_intra } else { tables.quant_inter }; @@ -299,10 +287,11 @@ fn decode_block4x4(br: &mut BitReader, blk_cb: &IVICodebook, rvmap: &RVMap, tabl if v == 0 { val = 0; // should not happen but still... } else { + let vv = (v >> 1) as i32; if (v & 1) != 0 { - val = ((v >> 1) as i32) + 1; + val = vv + 1; } else { - val = -((v >> 1) as i32); + val = -vv; } } } @@ -330,25 +319,17 @@ fn decode_block4x4(br: &mut BitReader, blk_cb: &IVICodebook, rvmap: &RVMap, tabl } fn put_block(frame: &mut [i16], offs: usize, stride: usize, blk: &[i32], blk_size: usize) { - unsafe { - let mut dptr = frame.as_mut_ptr().add(offs); - for y in 0..blk_size { - for x in 0..blk_size { - *dptr.add(x) = blk[x + y * blk_size] as i16; - } - dptr = dptr.add(stride); + for (dline, row) in frame[offs..].chunks_mut(stride).zip(blk.chunks(blk_size)).take(blk_size) { + for (dst, &src) in dline.iter_mut().zip(row.iter()) { + *dst = src as i16; } } } fn add_block(frame: &mut [i16], offs: usize, stride: usize, blk: &[i32], blk_size: usize) { - unsafe { - let mut dptr = frame.as_mut_ptr().add(offs); - for y in 0..blk_size { - for x in 0..blk_size { - *dptr.add(x) = (*dptr.add(x)).wrapping_add(blk[x + y * blk_size] as i16); - } - dptr = dptr.add(stride); + for (dline, row) in frame[offs..].chunks_mut(stride).zip(blk.chunks(blk_size)).take(blk_size) { + for (dst, &src) in dline.iter_mut().zip(row.iter()) { + *dst = dst.wrapping_add(src as i16); } } } @@ -393,18 +374,15 @@ impl FrameData { fn fill_plane(&mut self, vb: &mut NAVideoBuffer, plane: usize) { let dplane = if (plane == 1) || (plane == 2) { plane ^ 3 } else { plane }; let (w, h) = vb.get_dimensions(dplane); - let mut didx = vb.get_offset(dplane); + let didx = vb.get_offset(dplane); let dstride = vb.get_stride(dplane); let dst = vb.get_data_mut().unwrap(); let src = &self.plane_buf[plane]; - let mut sidx = 0; let sstride = self.plane_stride[plane]; - for _ in 0..h { - for x in 0..w { - dst[didx + x] = clip8(src[sidx + x] + 128); + for (dline, sline) in dst[didx..].chunks_mut(dstride).zip(src.chunks_exact(sstride)).take(h) { + for (dst, &src) in dline.iter_mut().zip(sline.iter()).take(w) { + *dst = clip8(src + 128); } - didx += dstride; - sidx += sstride; } } } @@ -608,7 +586,7 @@ impl IVIDecoder { validate!(tile_end > br.tell()); validate!(tile_end <= br.tell() + (br.left() as usize)); { - let ref_tile: Option<&IVITile>; + let mut ref_tile: Option<&IVITile> = None; let mv_scale = if (plane_no == 0) && (band_no == 0) { 0 } else { (((self.bands[0].mb_size >> 3) as i8) - ((band.mb_size >> 3) as i8)) as u8 }; @@ -616,21 +594,18 @@ impl IVIDecoder { let tile = &mut cur_tiles[0]; if plane_no != 0 || band_no != 0 { let rtile = &ref_tiles[tile_no % self.band_tiles]; - if (tile.mb_w != rtile.mb_w) || (tile.mb_h != rtile.mb_h) { - ref_tile = None; - } else { + if (tile.mb_w == rtile.mb_w) && (tile.mb_h == rtile.mb_h) { ref_tile = Some(rtile); } - } else { - ref_tile = None; } dec.decode_mb_info(br, pic_hdr, &band, tile, ref_tile, mv_scale)?; } self.decode_tile(br, &band, tile_no, tr, tr_dc)?; br.align(); -let skip_part = tile_end - br.tell(); -br.skip(skip_part as u32)?; + + let skip_part = tile_end - br.tell(); + br.skip(skip_part as u32)?; } else { { let ref_tile: Option<&IVITile>; diff --git a/nihav-indeo/src/codecs/ividsp.rs b/nihav-indeo/src/codecs/ividsp.rs index 8a19391..36cad58 100644 --- a/nihav-indeo/src/codecs/ividsp.rs +++ b/nihav-indeo/src/codecs/ividsp.rs @@ -304,96 +304,34 @@ pub fn ivi_get_transform4x4_funcs(ttype: IVITransformType) -> (TrFunc, TrFuncDC) } pub fn ivi_mc_put(dst: &mut [i16], dstride: usize, src: &[i16], sstride: usize, mode: u8, w: usize, h: usize) { - let mut sidx = 0; - let mut didx = 0; if src.len() < w + h * sstride { return; } match mode { 0 => { - for _ in 0..h { - let dest = &mut dst[didx..didx+w]; - dest.copy_from_slice(&src[sidx..sidx+w]); - sidx += sstride; - didx += dstride; + for (dline, sline) in dst.chunks_mut(dstride).zip(src.chunks(sstride)).take(h) { + dline[..w].copy_from_slice(&sline[..w]); } }, 1 => { - /*for _ in 0..h { - for x in 0..w { - let val = (src[sidx + x] + src[sidx + x + 1]) >> 1; - dst[didx + x] = val; - } - sidx += sstride; - didx += dstride; - }*/ - unsafe { - let mut sptr = src.as_ptr(); - let mut dptr = dst.as_mut_ptr(); - for _ in 0..h { - let mut last = *sptr; - for x in 0..w { - let nv = *sptr.add(x + 1); - *dptr.add(x) = nv.wrapping_add(last) >> 1; - last = nv; - } - sptr = sptr.add(sstride); - dptr = dptr.add(dstride); + for (dline, sline) in dst.chunks_mut(dstride).zip(src.chunks(sstride)).take(h) { + for (dpix, pair) in dline.iter_mut().zip(sline.windows(2)).take(w) { + *dpix = (pair[0] + pair[1]) >> 1; } } }, 2 => { - /*for _ in 0..h { - for x in 0..w { - let val = (src[sidx + x] + src[sidx + x + sstride]) >> 1; - dst[didx + x] = val; - } - sidx += sstride; - didx += dstride; - }*/ - unsafe { - let mut sptr0 = src.as_ptr(); - let mut sptr1 = sptr0.add(sstride); - let mut dptr = dst.as_mut_ptr(); - for _ in 0..h { - for x in 0..w { - let a = *sptr0.add(x); - let b = *sptr1.add(x); - *dptr.add(x) = a.wrapping_add(b) >> 1; - } - sptr0 = sptr0.add(sstride); - sptr1 = sptr1.add(sstride); - dptr = dptr.add(sstride); + for (dline, (sline0, sline1)) in dst.chunks_mut(dstride) + .zip(src.chunks(sstride).zip(src.chunks(sstride).skip(1))).take(h) { + for (dpix, (&s0, &s1)) in dline.iter_mut().zip(sline0.iter().zip(sline1.iter())).take(w) { + *dpix = (s0 + s1) >> 1; } } }, 3 => { - /*for _ in 0..h { - for x in 0..w { - let val = (src[sidx + x + 0] + src[sidx + x + sstride + 0] + - src[sidx + x + 1] + src[sidx + x + sstride + 1]) >> 2; - dst[didx + x] = val; - } - sidx += sstride; - didx += dstride; - }*/ - unsafe { - let mut sptr0 = src.as_ptr(); - let mut sptr1 = sptr0.add(sstride); - let mut dptr = dst.as_mut_ptr(); - for _ in 0..h { - let mut la = *sptr0; - let mut lb = *sptr1; - for x in 0..w { - let a = *sptr0.add(x + 1); - let b = *sptr1.add(x + 1); - let aas = a.wrapping_add(la); - let bbs = b.wrapping_add(lb); - *dptr.add(x) = aas.wrapping_add(bbs) >> 2; - la = a; - lb = b; - } - sptr0 = sptr0.add(sstride); - sptr1 = sptr1.add(sstride); - dptr = dptr.add(dstride); + for (dline, (sline0, sline1)) in dst.chunks_mut(dstride) + .zip(src.chunks(sstride).zip(src.chunks(sstride).skip(1))).take(h) { + for (dpix, (pair0, pair1)) in dline.iter_mut() + .zip(sline0.windows(2).zip(sline1.windows(2))).take(w) { + *dpix = (pair0[0] + pair0[1] + pair1[0] + pair1[1]) >> 2; } } }, @@ -401,47 +339,36 @@ pub fn ivi_mc_put(dst: &mut [i16], dstride: usize, src: &[i16], sstride: usize, } } fn ivi_mc_add(dst: &mut [i16], dstride: usize, src: &[i16], sstride: usize, mode: u8, w: usize, h: usize) { - let mut sidx = 0; - let mut didx = 0; match mode { 0 => { - for _ in 0..h { - for x in 0..w { - dst[didx + x] += src[sidx + x]; + for (dline, sline) in dst.chunks_mut(dstride).zip(src.chunks(sstride)).take(h) { + for (dst, &src) in dline.iter_mut().zip(sline.iter()).take(w) { + *dst += src; } - sidx += sstride; - didx += dstride; } }, 1 => { - for _ in 0..h { - for x in 0..w { - let val = (src[sidx + x] + src[sidx + x + 1]) >> 1; - dst[didx + x] += val; + for (dline, sline) in dst.chunks_mut(dstride).zip(src.chunks(sstride)).take(h) { + for (dpix, pair) in dline.iter_mut().zip(sline.windows(2)).take(w) { + *dpix += (pair[0] + pair[1]) >> 1; } - sidx += sstride; - didx += dstride; } }, 2 => { - for _ in 0..h { - for x in 0..w { - let val = (src[sidx + x] + src[sidx + x + sstride]) >> 1; - dst[didx + x] += val; + for (dline, (sline0, sline1)) in dst.chunks_mut(dstride) + .zip(src.chunks(sstride).zip(src.chunks(sstride).skip(1))).take(h) { + for (dpix, (&s0, &s1)) in dline.iter_mut().zip(sline0.iter().zip(sline1.iter())).take(w) { + *dpix += (s0 + s1) >> 1; } - sidx += sstride; - didx += dstride; } }, 3 => { - for _ in 0..h { - for x in 0..w { - let val = (src[sidx + x + 0] + src[sidx + x + sstride + 0] + - src[sidx + x + 1] + src[sidx + x + sstride + 1]) >> 2; - dst[didx + x] += val; + for (dline, (sline0, sline1)) in dst.chunks_mut(dstride) + .zip(src.chunks(sstride).zip(src.chunks(sstride).skip(1))).take(h) { + for (dpix, (pair0, pair1)) in dline.iter_mut() + .zip(sline0.windows(2).zip(sline1.windows(2))).take(w) { + *dpix += (pair0[0] + pair0[1] + pair1[0] + pair1[1]) >> 2; } - sidx += sstride; - didx += dstride; } }, _ => {}, @@ -451,15 +378,13 @@ pub fn ivi_mc_avg(dst: &mut [i16], dstride: usize, src1: &[i16], sstride1: usize, mode1: u8, src2: &[i16], sstride2: usize, mode2: u8, w: usize, h: usize) { - let mut tidx = 0; - let tstride = 8; - let mut didx = 0; - let mut tmp: [i16; 64] = [0; 64]; - ivi_mc_add(&mut tmp, tstride, src1, sstride1, mode1, w, h); - ivi_mc_add(&mut tmp, tstride, src2, sstride2, mode2, w, h); - for _ in 0..h { - for x in 0..w { dst[didx + x] = tmp[tidx + x] >> 1; } - tidx += tstride; - didx += dstride; + const TSTRIDE: usize = 8; + let mut tmp: [i16; 8 * TSTRIDE] = [0; 8 * TSTRIDE]; + ivi_mc_add(&mut tmp, TSTRIDE, src1, sstride1, mode1, w, h); + ivi_mc_add(&mut tmp, TSTRIDE, src2, sstride2, mode2, w, h); + for (dline, row) in dst.chunks_mut(dstride).zip(tmp.chunks_exact(TSTRIDE)).take(h) { + for (dpix, &src) in dline.iter_mut().zip(row.iter()).take(w) { + *dpix = src >> 1; + } } } -- 2.39.5