Reworking YCgCo -> RGB

README.md

@@ -87,6 +87,7 @@ The following planar formats are used for HDR content, with either **10-bit** or
 ### Format Breakdown:
 
 - **I**: Represents the color space (as defined above), and these formats use **3 planes**: Y (luminance), U (chrominance), and V (chrominance).
+- **ICgC**: Represents the color space (as defined above), and these formats use **3 planes**: Y (luminance), Cg (chrominance), and Co (chrominance).
 - **P**: A **biplanar format** (similar to **NV12**), but with **16-bit precision**, where the valid bits are stored in the high bits. This format has:
     - A **Y plane**
     - A **UV plane** (with U and V interleaved)

app/benches/yuv8/main.rs

@@ -36,10 +36,10 @@ use yuv_sys::{
 use yuvutils_rs::{
     gbr_to_rgba, rgb_to_gbr, rgb_to_yuv400, rgb_to_yuv420, rgb_to_yuv422, rgb_to_yuv444,
     rgb_to_yuv_nv12, rgb_to_yuv_nv16, rgba_to_yuv420, rgba_to_yuv422, rgba_to_yuv444,
-    rgba_to_yuv_nv12, yuv400_to_rgba, yuv420_to_rgb, yuv420_to_rgba, yuv422_to_rgba,
-    yuv444_to_rgba, yuv_nv12_to_rgb, yuv_nv12_to_rgba, yuv_nv16_to_rgb, YuvBiPlanarImageMut,
-    YuvChromaSubsampling, YuvConversionMode, YuvGrayImageMut, YuvPlanarImageMut, YuvRange,
-    YuvStandardMatrix,
+    rgba_to_yuv_nv12, ycgco420_to_rgba, ycgco444_to_rgba, yuv400_to_rgba, yuv420_to_rgb,
+    yuv420_to_rgba, yuv422_to_rgba, yuv444_to_rgba, yuv_nv12_to_rgb, yuv_nv12_to_rgba,
+    yuv_nv16_to_rgb, YuvBiPlanarImageMut, YuvChromaSubsampling, YuvConversionMode, YuvGrayImageMut,
+    YuvPlanarImageMut, YuvRange, YuvStandardMatrix,
 };
 
 pub fn criterion_benchmark(c: &mut Criterion) {
@@ -675,6 +675,19 @@ pub fn criterion_benchmark(c: &mut Criterion) {
         })
     });
 
+    c.bench_function("yuvutils YCgCo 4:2:0 -> RGBA", |b| {
+        let mut rgb_bytes = vec![0u8; dimensions.0 as usize * 4 * dimensions.1 as usize];
+        b.iter(|| {
+            ycgco420_to_rgba(
+                &fixed_planar,
+                &mut rgb_bytes,
+                dimensions.0 * 4u32,
+                YuvRange::Full,
+            )
+            .unwrap();
+        })
+    });
+
     c.bench_function("yuvutils YUV 4:2:0 -> RGBA", |b| {
         let mut rgb_bytes = vec![0u8; dimensions.0 as usize * 4 * dimensions.1 as usize];
         b.iter(|| {
@@ -769,6 +782,19 @@ pub fn criterion_benchmark(c: &mut Criterion) {
 
     let fixed_planar444 = planar_image444.to_fixed();
 
+    c.bench_function("yuvutils YCgCo 4:4:4 -> RGBA", |b| {
+        let mut rgb_bytes = vec![0u8; dimensions.0 as usize * 4 * dimensions.1 as usize];
+        b.iter(|| {
+            ycgco444_to_rgba(
+                &fixed_planar444,
+                &mut rgb_bytes,
+                dimensions.0 * 4u32,
+                YuvRange::Full,
+            )
+            .unwrap();
+        })
+    });
+
     c.bench_function("yuvutils YUV 4:4:4 -> RGBA", |b| {
         let mut rgb_bytes = vec![0u8; dimensions.0 as usize * 4 * dimensions.1 as usize];
         b.iter(|| {

src/avx2/avx2_utils.rs

@@ -1191,20 +1191,6 @@ pub(crate) unsafe fn _mm256_expand_rgb_to_rgba(
     (v0, v1, v2, v3)
 }
 
-#[inline(always)]
-pub(crate) unsafe fn sse_interleave_even(x: __m128i) -> __m128i {
-    #[rustfmt::skip]
-    let shuffle = _mm_setr_epi8(0, 0, 2, 2, 4, 4, 6, 6,
-                                     8, 8, 10, 10, 12, 12, 14, 14);
-    _mm_shuffle_epi8(x, shuffle)
-}
-
-#[inline(always)]
-pub(crate) unsafe fn sse_interleave_odd(x: __m128i) -> __m128i {
-    let shuffle = _mm_setr_epi8(1, 1, 3, 3, 5, 5, 7, 7, 9, 9, 11, 11, 13, 13, 15, 15);
-    _mm_shuffle_epi8(x, shuffle)
-}
-
 #[inline(always)]
 pub(crate) unsafe fn _avx_from_msb_epi16<const BIT_DEPTH: usize>(a: __m128i) -> __m128i {
     if BIT_DEPTH == 10 {

src/avx2/mod.rs

@@ -59,7 +59,6 @@ mod rgba_to_yuv_fast420;
 mod shuffle;
 mod y_to_rgba;
 mod y_to_rgba_alpha;
-mod ycgco_to_rgba_alpha;
 mod yuv_nv_to_rgba;
 mod yuv_nv_to_rgba420;
 #[cfg(feature = "professional_mode")]
@@ -119,7 +118,6 @@ pub(crate) use rgba_to_yuv_fast420::avx2_rgba_to_yuv_dot_rgba420;
 pub(crate) use shuffle::{ShuffleConverterAvx2, ShuffleQTableConverterAvx2};
 pub(crate) use y_to_rgba::avx2_y_to_rgba_row;
 pub(crate) use y_to_rgba_alpha::avx2_y_to_rgba_alpha_row;
-pub(crate) use ycgco_to_rgba_alpha::avx2_ycgco_to_rgba_alpha;
 pub(crate) use yuv_nv_to_rgba::avx2_yuv_nv_to_rgba_row;
 pub(crate) use yuv_nv_to_rgba420::avx2_yuv_nv_to_rgba_row420;
 #[cfg(feature = "professional_mode")]

src/avx512bw/mod.rs

@@ -42,7 +42,6 @@ mod rgba_to_yuv_fast;
 #[cfg(feature = "fast_mode")]
 mod rgba_to_yuv_fast420;
 mod y_to_rgb;
-mod ycgco_to_rgba_alpha;
 mod yuv_nv_to_rgba;
 mod yuv_nv_to_rgba420;
 mod yuv_nv_to_rgba422;
@@ -70,7 +69,6 @@ pub(crate) use rgba_to_yuv_fast420::{
     avx512_rgba_to_yuv_dot_rgba420, avx512_rgba_to_yuv_dot_rgba420_vbmi,
 };
 pub(crate) use y_to_rgb::avx512_y_to_rgb_row;
-pub(crate) use ycgco_to_rgba_alpha::avx512_ycgco_to_rgba_alpha;
 pub(crate) use yuv_nv_to_rgba::avx512_yuv_nv_to_rgba;
 pub(crate) use yuv_nv_to_rgba420::avx512_yuv_nv_to_rgba420;
 pub(crate) use yuv_nv_to_rgba422::avx512_yuv_nv_to_rgba422;