pic-scale 0.7.8

/*
 * Copyright (c) Radzivon Bartoshyk. All rights reserved.
 *
 * Redistribution and use in source and binary forms, with or without modification,
 * are permitted provided that the following conditions are met:
 *
 * 1.  Redistributions of source code must retain the above copyright notice, this
 * list of conditions and the following disclaimer.
 *
 * 2.  Redistributions in binary form must reproduce the above copyright notice,
 * this list of conditions and the following disclaimer in the documentation
 * and/or other materials provided with the distribution.
 *
 * 3.  Neither the name of the copyright holder nor the names of its
 * contributors may be used to endorse or promote products derived from
 * this software without specific prior written permission.
 *
 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
 * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
 * DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE
 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
 * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
 * CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
 * OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
 * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 */

#[cfg(target_arch = "x86")]
use std::arch::x86::*;
#[cfg(target_arch = "x86_64")]
use std::arch::x86_64::*;

use crate::filter_weights::FilterWeights;
use crate::sse::{_mm_muladd_wide_epi16, shuffle};
use crate::support::{PRECISION, ROUNDING_CONST};

#[inline]
#[target_feature(enable = "sse4.1")]
fn s_accumulate_8_horiz(store: __m128i, ptr: &[u8], weights: (__m128i, __m128i)) -> __m128i {
    let pixel_colors = unsafe { _mm_loadu_si64(ptr.as_ptr()) };
    let px_16 = _mm_unpacklo_epi8(pixel_colors, _mm_setzero_si128());
    let px_lo = _mm_unpacklo_epi16(px_16, _mm_setzero_si128());
    let px_hi = _mm_unpackhi_epi16(px_16, _mm_setzero_si128());

    let store = _mm_muladd_wide_epi16(store, px_lo, weights.0);
    _mm_muladd_wide_epi16(store, px_hi, weights.1)
}

#[inline]
#[target_feature(enable = "sse4.1")]
fn s_accumulate_4_horiz(store: __m128i, ptr: &[u8], weights: __m128i) -> __m128i {
    let zeros = _mm_setzero_si128();
    let pixel_colors = _mm_unpacklo_epi16(
        _mm_unpacklo_epi8(unsafe { _mm_loadu_si32(ptr.as_ptr().cast()) }, zeros),
        zeros,
    );
    _mm_muladd_wide_epi16(store, pixel_colors, weights)
}

#[inline]
#[target_feature(enable = "sse4.1")]
fn s_accumulate_1_horiz(store: __m128i, ptr: &[u8], weight: __m128i) -> __m128i {
    let pixel_colors = _mm_setr_epi32(unsafe { *ptr.get_unchecked(0) } as i32, 0, 0, 0);
    _mm_muladd_wide_epi16(store, pixel_colors, weight)
}

pub(crate) fn convolve_horizontal_plane_sse_rows_4_u8(
    src: &[u8],
    src_stride: usize,
    dst: &mut [u8],
    dst_stride: usize,
    filter_weights: &FilterWeights<i16>,
    _: u32,
) {
    unsafe {
        convolve_horizontal_plane_sse_rows_4_u8_impl(
            src,
            src_stride,
            dst,
            dst_stride,
            filter_weights,
        );
    }
}

#[inline(always)]
fn _mm_hsum_reduce_epi32<const PRECISION: i32>(x: __m128i) -> u8 {
    unsafe {
        const FIRST_MASK: i32 = shuffle(1, 0, 3, 2);
        let hi64 = _mm_shuffle_epi32::<FIRST_MASK>(x);
        let sum64 = _mm_add_epi32(hi64, x);
        const SM: i32 = shuffle(1, 0, 3, 2);
        let hi32 = _mm_shufflelo_epi16::<SM>(sum64);
        let sum32 = _mm_srai_epi32::<PRECISION>(_mm_add_epi32(sum64, hi32));
        let v0 = _mm_packus_epi32(sum32, sum32);
        let v1 = _mm_packus_epi16(v0, v0);
        _mm_extract_epi8::<0>(v1) as u8
    }
}

#[target_feature(enable = "sse4.1")]
fn convolve_horizontal_plane_sse_rows_4_u8_impl(
    src: &[u8],
    src_stride: usize,
    dst: &mut [u8],
    dst_stride: usize,
    filter_weights: &FilterWeights<i16>,
) {
    unsafe {
        let zeros = _mm_setzero_si128();

        let (row0_ref, rest) = dst.split_at_mut(dst_stride);
        let (row1_ref, rest) = rest.split_at_mut(dst_stride);
        let (row2_ref, row3_ref) = rest.split_at_mut(dst_stride);

        let iter_row0 = row0_ref.iter_mut();
        let iter_row1 = row1_ref.iter_mut();
        let iter_row2 = row2_ref.iter_mut();
        let iter_row3 = row3_ref.iter_mut();

        for (((((chunk0, chunk1), chunk2), chunk3), &bounds), weights) in iter_row0
            .zip(iter_row1)
            .zip(iter_row2)
            .zip(iter_row3)
            .zip(filter_weights.bounds.iter())
            .zip(
                filter_weights
                    .weights
                    .chunks_exact(filter_weights.aligned_size),
            )
        {
            let mut jx = 0usize;
            let mut store0 = _mm_setr_epi32(ROUNDING_CONST, 0i32, 0i32, 0i32);
            let mut store1 = _mm_setr_epi32(ROUNDING_CONST, 0i32, 0i32, 0i32);
            let mut store2 = _mm_setr_epi32(ROUNDING_CONST, 0i32, 0i32, 0i32);
            let mut store3 = _mm_setr_epi32(ROUNDING_CONST, 0i32, 0i32, 0i32);

            let src0 = src;
            let src1 = src0.get_unchecked(src_stride..);
            let src2 = src1.get_unchecked(src_stride..);
            let src3 = src2.get_unchecked(src_stride..);

            while jx + 8 <= bounds.size {
                let w_ptr = weights.get_unchecked(jx..(jx + 8));
                let weights_i16 = _mm_loadu_si128(w_ptr.as_ptr() as *const __m128i);
                let weights = (
                    _mm_unpacklo_epi16(weights_i16, zeros),
                    _mm_unpackhi_epi16(weights_i16, zeros),
                );
                let bounds_start = bounds.start + jx;

                let src_ptr = src0.get_unchecked(bounds_start..);
                store0 = s_accumulate_8_horiz(store0, src_ptr, weights);

                let src_ptr1 = src1.get_unchecked(bounds_start..);
                store1 = s_accumulate_8_horiz(store1, src_ptr1, weights);

                let src_ptr2 = src2.get_unchecked(bounds_start..);
                store2 = s_accumulate_8_horiz(store2, src_ptr2, weights);

                let src_ptr3 = src3.get_unchecked(bounds_start..);
                store3 = s_accumulate_8_horiz(store3, src_ptr3, weights);

                jx += 8;
            }

            while jx + 4 <= bounds.size {
                let w_ptr = weights.get_unchecked(jx..(jx + 4));
                let weights = _mm_cvtepi16_epi32(_mm_loadu_si64(w_ptr.as_ptr() as *const u8));
                let bounds_start = bounds.start + jx;

                let src_ptr = src0.get_unchecked(bounds_start..);
                store0 = s_accumulate_4_horiz(store0, src_ptr, weights);

                let src_ptr1 = src1.get_unchecked(bounds_start..);
                store1 = s_accumulate_4_horiz(store1, src_ptr1, weights);

                let src_ptr2 = src2.get_unchecked(bounds_start..);
                store2 = s_accumulate_4_horiz(store2, src_ptr2, weights);

                let src_ptr3 = src3.get_unchecked(bounds_start..);
                store3 = s_accumulate_4_horiz(store3, src_ptr3, weights);

                jx += 4;
            }

            while jx < bounds.size {
                let w_ptr = weights.get_unchecked(jx..(jx + 1));
                let weight = _mm_setr_epi32(w_ptr.as_ptr().read_unaligned() as i32, 0, 0, 0);
                let bounds_start = bounds.start + jx;

                let src_ptr = src0.get_unchecked(bounds_start..);
                store0 = s_accumulate_1_horiz(store0, src_ptr, weight);

                let src_ptr1 = src1.get_unchecked(bounds_start..);
                store1 = s_accumulate_1_horiz(store1, src_ptr1, weight);

                let src_ptr2 = src2.get_unchecked(bounds_start..);
                store2 = s_accumulate_1_horiz(store2, src_ptr2, weight);

                let src_ptr3 = src3.get_unchecked(bounds_start..);
                store3 = s_accumulate_1_horiz(store3, src_ptr3, weight);

                jx += 1;
            }

            let value0 = _mm_hsum_reduce_epi32::<PRECISION>(store0);
            *chunk0 = value0;

            let value1 = _mm_hsum_reduce_epi32::<PRECISION>(store1);
            *chunk1 = value1;

            let value2 = _mm_hsum_reduce_epi32::<PRECISION>(store2);
            *chunk2 = value2;

            let value3 = _mm_hsum_reduce_epi32::<PRECISION>(store3);
            *chunk3 = value3;
        }
    }
}

pub(crate) fn convolve_horizontal_plane_sse_row(
    src: &[u8],
    dst: &mut [u8],
    filter_weights: &FilterWeights<i16>,
    _: u32,
) {
    unsafe {
        convolve_horizontal_plane_sse_row_impl(src, dst, filter_weights);
    }
}

#[target_feature(enable = "sse4.1")]
fn convolve_horizontal_plane_sse_row_impl(
    src: &[u8],
    dst: &mut [u8],
    filter_weights: &FilterWeights<i16>,
) {
    unsafe {
        let zeros = _mm_setzero_si128();

        for ((dst, bounds), weights) in dst.iter_mut().zip(filter_weights.bounds.iter()).zip(
            filter_weights
                .weights
                .chunks_exact(filter_weights.aligned_size),
        ) {
            let mut jx = 0usize;
            let mut store = _mm_setr_epi32(ROUNDING_CONST, 0i32, 0i32, 0i32);

            while jx + 8 <= bounds.size {
                let w_ptr = weights.get_unchecked(jx..(jx + 8));
                let weights_i16 = _mm_loadu_si128(w_ptr.as_ptr() as *const __m128i);
                let weights = (
                    _mm_unpacklo_epi16(weights_i16, zeros),
                    _mm_unpackhi_epi16(weights_i16, zeros),
                );
                let bounds_start = bounds.start + jx;

                let src_ptr = src.get_unchecked(bounds_start..);
                store = s_accumulate_8_horiz(store, src_ptr, weights);

                jx += 8;
            }

            while jx + 4 <= bounds.size {
                let w_ptr = weights.get_unchecked(jx..(jx + 4));
                let weights = _mm_cvtepi16_epi32(_mm_loadu_si64(w_ptr.as_ptr() as *const u8));
                let bounds_start = bounds.start + jx;

                let src_ptr = src.get_unchecked(bounds_start..);
                store = s_accumulate_4_horiz(store, src_ptr, weights);

                jx += 4;
            }

            while jx < bounds.size {
                let w_ptr = weights.get_unchecked(jx..(jx + 1));
                let weight = _mm_setr_epi32(w_ptr.as_ptr().read_unaligned() as i32, 0, 0, 0);
                let bounds_start = bounds.start + jx;
                let src_ptr = src.get_unchecked(bounds_start..);
                store = s_accumulate_1_horiz(store, src_ptr, weight);
                jx += 1;
            }

            let value = _mm_hsum_reduce_epi32::<PRECISION>(store);
            *dst = value;
        }
    }
}