colconv 0.1.0

SIMD-dispatched color-conversion kernels covering the FFmpeg AVPixelFormat space, with a Sink-based API so consumers pick which derived outputs (RGB / Luma / HSV / custom) they want without paying for the ones they don't.
Documentation
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
//! SSE4.1 1-bit-per-pixel unpack kernels (Monoblack / Monowhite).
//!
//! # Bit-mask broadcast pattern (16 px / iter, 2 bytes / iter)
//!
//! Each byte covers 8 pixels (MSB first):
//! 1. `_mm_set1_epi8(byte)` — broadcast to 16 lanes.
//! 2. AND with bit-position mask `[0x80,0x40,...,0x01]` repeated twice (16 bytes).
//! 3. `_mm_cmpeq_epi8(and, zero)` → 0x00 where bit was set, 0xFF where clear.
//! 4. Negate: `_mm_xor_si128(cmp, all_ones)` gives 0xFF for set bit (Monoblack).
//! 5. For Monowhite (`INVERT=true`): skip the negate — the `cmpeq` zero result is 0xFF.
//!
//! After unpacking 16 pixels, the Y u8x16 vector is ready for output:
//! - RGB: write via `write_rgb_16` (3-channel interleave).
//! - RGBA: write via `write_rgba_16` (4-channel interleave, α=0xFF).
//! - Luma u8: store with `_mm_storeu_si128`.
//! - u16 outputs: process 8 px / iter, zero-extend the low 8 bytes to u16x8 with
//!   `_mm_unpacklo_epi8(y, zero)`, then store via `write_rgb_u16_8`
//!   / `write_rgba_u16_8` (shared SSSE3/SSE2 interleave helpers).
//!
//! Tail (remaining pixels after last full 16-px block) falls back to scalar.

#![cfg_attr(not(feature = "std"), allow(dead_code))]

use core::arch::x86_64::*;

use crate::row::{
  arch::x86_common::{write_rgb_16, write_rgb_u16_8, write_rgba_16, write_rgba_u16_8},
  scalar::mono1bit as scalar,
};

/// Unpack 2 input bytes into a u8x16 luma vector (16 pixels).
/// For INVERT=false (Monoblack): bit=1 → lane=0xFF, bit=0 → lane=0x00.
/// For INVERT=true (Monowhite): bit=0 → lane=0xFF, bit=1 → lane=0x00.
#[inline]
#[target_feature(enable = "sse4.1")]
unsafe fn unpack_2bytes_sse41<const INVERT: bool>(b0: u8, b1: u8) -> __m128i {
  // Bit-position mask: lane i of each 8-lane half tests bit (7 - i).
  let mask = _mm_set_epi8(
    0x01u8 as i8,
    0x02u8 as i8,
    0x04u8 as i8,
    0x08u8 as i8,
    0x10u8 as i8,
    0x20u8 as i8,
    0x40u8 as i8,
    0x80u8 as i8,
    0x01u8 as i8,
    0x02u8 as i8,
    0x04u8 as i8,
    0x08u8 as i8,
    0x10u8 as i8,
    0x20u8 as i8,
    0x40u8 as i8,
    0x80u8 as i8,
  );
  // Build the 16-byte broadcast vector: low 8 lanes = b0, high 8 lanes = b1.
  let bcast = _mm_set_epi8(
    b1 as i8, b1 as i8, b1 as i8, b1 as i8, b1 as i8, b1 as i8, b1 as i8, b1 as i8, b0 as i8,
    b0 as i8, b0 as i8, b0 as i8, b0 as i8, b0 as i8, b0 as i8, b0 as i8,
  );
  let anded = _mm_and_si128(bcast, mask);
  let zero = _mm_setzero_si128();
  // cmpeq: 0xFF where (anded == 0), i.e., where bit was NOT set.
  let cmp = _mm_cmpeq_epi8(anded, zero);
  if INVERT {
    // Monowhite: bit=0 (not set) → white (0xFF) — already what cmp gives.
    cmp
  } else {
    // Monoblack: bit=1 (set) → white (0xFF) → negate the cmpeq.
    let all_ones = _mm_set1_epi8(-1i8);
    _mm_xor_si128(cmp, all_ones)
  }
}

/// Zero-extend a u8x8 (low 8 bytes of a __m128i) to u16x8.
/// White (0xFF) maps to 0x00FF, matching Gray8's `with_luma_u16` contract.
/// Returns a full __m128i with 8 u16 values.
#[inline]
#[target_feature(enable = "sse4.1")]
unsafe fn expand_y_to_u16x8_sse41(y_low8: __m128i) -> __m128i {
  // _mm_unpacklo_epi8(y, zero): interleave low 8 bytes with 0x00
  // → [y0, 0, y1, 0, ..., y7, 0] as a u8x16, i.e. u16x8 of y (zero-extended).
  let zero = _mm_setzero_si128();
  _mm_unpacklo_epi8(y_low8, zero)
}

// ---- mono1bit → RGB u8 -------------------------------------------------------

/// SSE4.1 `mono1bit_to_rgb_row<INVERT>`: unpack 1-bpp → packed RGB u8.
///
/// Block size: 16 px / iter (2 input bytes). Tail: scalar.
///
/// # Safety
/// SSE4.1 must be available. `data.len() >= width.div_ceil(8)`.
/// `out.len() >= width * 3`.
#[allow(dead_code)]
#[inline]
#[target_feature(enable = "sse4.1")]
pub(crate) unsafe fn mono1bit_to_rgb_row<const INVERT: bool>(
  data: &[u8],
  out: &mut [u8],
  width: usize,
) {
  debug_assert!(data.len() >= width.div_ceil(8));
  debug_assert!(out.len() >= width * 3);
  let mut x = 0usize;
  let mut byte_idx = 0usize;
  unsafe {
    while x + 16 <= width {
      let y = unpack_2bytes_sse41::<INVERT>(data[byte_idx], data[byte_idx + 1]);
      write_rgb_16(y, y, y, out.as_mut_ptr().add(x * 3));
      x += 16;
      byte_idx += 2;
    }
  }
  if x < width {
    if INVERT {
      scalar::monowhite_to_rgb_row(&data[byte_idx..], &mut out[x * 3..width * 3], width - x);
    } else {
      scalar::monoblack_to_rgb_row(&data[byte_idx..], &mut out[x * 3..width * 3], width - x);
    }
  }
}

// ---- mono1bit → RGBA u8 -----------------------------------------------------

/// SSE4.1 `mono1bit_to_rgba_row<INVERT>`: unpack 1-bpp → packed RGBA u8, α=0xFF.
///
/// # Safety
/// SSE4.1 must be available. `out.len() >= width * 4`.
#[allow(dead_code)]
#[inline]
#[target_feature(enable = "sse4.1")]
pub(crate) unsafe fn mono1bit_to_rgba_row<const INVERT: bool>(
  data: &[u8],
  out: &mut [u8],
  width: usize,
) {
  debug_assert!(data.len() >= width.div_ceil(8));
  debug_assert!(out.len() >= width * 4);
  let mut x = 0usize;
  let mut byte_idx = 0usize;
  unsafe {
    let alpha = _mm_set1_epi8(-1i8); // 0xFF
    while x + 16 <= width {
      let y = unpack_2bytes_sse41::<INVERT>(data[byte_idx], data[byte_idx + 1]);
      write_rgba_16(y, y, y, alpha, out.as_mut_ptr().add(x * 4));
      x += 16;
      byte_idx += 2;
    }
  }
  if x < width {
    if INVERT {
      scalar::monowhite_to_rgba_row(&data[byte_idx..], &mut out[x * 4..width * 4], width - x);
    } else {
      scalar::monoblack_to_rgba_row(&data[byte_idx..], &mut out[x * 4..width * 4], width - x);
    }
  }
}

// ---- mono1bit → Luma u8 -----------------------------------------------------

/// SSE4.1 `mono1bit_to_luma_row<INVERT>`: unpack 1-bpp → luma u8.
///
/// # Safety
/// SSE4.1 must be available. `out.len() >= width`.
#[allow(dead_code)]
#[inline]
#[target_feature(enable = "sse4.1")]
pub(crate) unsafe fn mono1bit_to_luma_row<const INVERT: bool>(
  data: &[u8],
  out: &mut [u8],
  width: usize,
) {
  debug_assert!(data.len() >= width.div_ceil(8));
  debug_assert!(out.len() >= width);
  let mut x = 0usize;
  let mut byte_idx = 0usize;
  unsafe {
    while x + 16 <= width {
      let y = unpack_2bytes_sse41::<INVERT>(data[byte_idx], data[byte_idx + 1]);
      _mm_storeu_si128(out.as_mut_ptr().add(x).cast(), y);
      x += 16;
      byte_idx += 2;
    }
  }
  if x < width {
    if INVERT {
      scalar::monowhite_to_luma_row(&data[byte_idx..], &mut out[x..width], width - x);
    } else {
      scalar::monoblack_to_luma_row(&data[byte_idx..], &mut out[x..width], width - x);
    }
  }
}

// ---- mono1bit → RGB u16 -----------------------------------------------------

/// SSE4.1 `mono1bit_to_rgb_u16_row<INVERT>`: unpack 1-bpp → RGB u16.
///
/// Block size: 8 px / iter (1 input byte). Tail: scalar.
///
/// # Safety
/// SSE4.1 must be available. `out.len() >= width * 3`.
#[allow(dead_code)]
#[inline]
#[target_feature(enable = "sse4.1")]
pub(crate) unsafe fn mono1bit_to_rgb_u16_row<const INVERT: bool>(
  data: &[u8],
  out: &mut [u16],
  width: usize,
) {
  debug_assert!(data.len() >= width.div_ceil(8));
  debug_assert!(out.len() >= width * 3);
  let mut x = 0usize;
  let mut byte_idx = 0usize;
  unsafe {
    while x + 8 <= width {
      // Process 1 byte = 8 pixels.
      let y8_128 = unpack_2bytes_sse41::<INVERT>(data[byte_idx], 0);
      // Extract low 8 bytes (our 8 pixels) and expand to u16x8.
      let y16 = expand_y_to_u16x8_sse41(y8_128);
      // Write 8 pixels × 3 channels = 24 u16 = 48 bytes via the shared
      // SSSE3 shuffle-based interleave helper (y broadcast to R, G, B).
      write_rgb_u16_8(y16, y16, y16, out.as_mut_ptr().add(x * 3));
      x += 8;
      byte_idx += 1;
    }
  }
  if x < width {
    if INVERT {
      scalar::monowhite_to_rgb_u16_row(&data[byte_idx..], &mut out[x * 3..width * 3], width - x);
    } else {
      scalar::monoblack_to_rgb_u16_row(&data[byte_idx..], &mut out[x * 3..width * 3], width - x);
    }
  }
}

// ---- mono1bit → RGBA u16 ----------------------------------------------------

/// SSE4.1 `mono1bit_to_rgba_u16_row<INVERT>`: unpack 1-bpp → RGBA u16, α=0xFFFF.
///
/// # Safety
/// SSE4.1 must be available. `out.len() >= width * 4`.
#[allow(dead_code)]
#[inline]
#[target_feature(enable = "sse4.1")]
pub(crate) unsafe fn mono1bit_to_rgba_u16_row<const INVERT: bool>(
  data: &[u8],
  out: &mut [u16],
  width: usize,
) {
  debug_assert!(data.len() >= width.div_ceil(8));
  debug_assert!(out.len() >= width * 4);
  let mut x = 0usize;
  let mut byte_idx = 0usize;
  unsafe {
    while x + 8 <= width {
      let y8_128 = unpack_2bytes_sse41::<INVERT>(data[byte_idx], 0);
      let y16 = expand_y_to_u16x8_sse41(y8_128);
      // α=0x00FF for all 8 pixels (zero-extend of 0xFF u8). Cast to i16 since __m128i is signed.
      let alpha = _mm_set1_epi16(0x00FFu16 as i16);
      // Write 8 pixels × 4 channels = 32 u16 = 64 bytes via the shared
      // SSE2 unpack-based interleave helper (y broadcast to R, G, B).
      write_rgba_u16_8(y16, y16, y16, alpha, out.as_mut_ptr().add(x * 4));
      x += 8;
      byte_idx += 1;
    }
  }
  if x < width {
    if INVERT {
      scalar::monowhite_to_rgba_u16_row(&data[byte_idx..], &mut out[x * 4..width * 4], width - x);
    } else {
      scalar::monoblack_to_rgba_u16_row(&data[byte_idx..], &mut out[x * 4..width * 4], width - x);
    }
  }
}

// ---- mono1bit → Luma u16 ----------------------------------------------------

/// SSE4.1 `mono1bit_to_luma_u16_row<INVERT>`: unpack 1-bpp → luma u16.
///
/// # Safety
/// SSE4.1 must be available. `out.len() >= width`.
#[allow(dead_code)]
#[inline]
#[target_feature(enable = "sse4.1")]
pub(crate) unsafe fn mono1bit_to_luma_u16_row<const INVERT: bool>(
  data: &[u8],
  out: &mut [u16],
  width: usize,
) {
  debug_assert!(data.len() >= width.div_ceil(8));
  debug_assert!(out.len() >= width);
  let mut x = 0usize;
  let mut byte_idx = 0usize;
  unsafe {
    while x + 8 <= width {
      let y8_128 = unpack_2bytes_sse41::<INVERT>(data[byte_idx], 0);
      let y16 = expand_y_to_u16x8_sse41(y8_128);
      _mm_storeu_si128(out.as_mut_ptr().add(x).cast(), y16);
      x += 8;
      byte_idx += 1;
    }
  }
  if x < width {
    if INVERT {
      scalar::monowhite_to_luma_u16_row(&data[byte_idx..], &mut out[x..width], width - x);
    } else {
      scalar::monoblack_to_luma_u16_row(&data[byte_idx..], &mut out[x..width], width - x);
    }
  }
}

// ---- mono1bit → HSV ---------------------------------------------------------

/// SSE4.1 `mono1bit_to_hsv_row<INVERT>`: H=0, S=0, V=Y.
///
/// # Safety
/// SSE4.1 must be available. All output slices >= width.
#[allow(dead_code)]
#[inline]
#[target_feature(enable = "sse4.1")]
pub(crate) unsafe fn mono1bit_to_hsv_row<const INVERT: bool>(
  data: &[u8],
  h: &mut [u8],
  s: &mut [u8],
  v: &mut [u8],
  width: usize,
) {
  debug_assert!(data.len() >= width.div_ceil(8));
  debug_assert!(h.len() >= width);
  debug_assert!(s.len() >= width);
  debug_assert!(v.len() >= width);
  let mut x = 0usize;
  let mut byte_idx = 0usize;
  unsafe {
    let zero = _mm_setzero_si128();
    while x + 16 <= width {
      let y = unpack_2bytes_sse41::<INVERT>(data[byte_idx], data[byte_idx + 1]);
      _mm_storeu_si128(h.as_mut_ptr().add(x).cast(), zero);
      _mm_storeu_si128(s.as_mut_ptr().add(x).cast(), zero);
      _mm_storeu_si128(v.as_mut_ptr().add(x).cast(), y);
      x += 16;
      byte_idx += 2;
    }
  }
  if x < width {
    if INVERT {
      scalar::monowhite_to_hsv_row(
        &data[byte_idx..],
        &mut h[x..width],
        &mut s[x..width],
        &mut v[x..width],
        width - x,
      );
    } else {
      scalar::monoblack_to_hsv_row(
        &data[byte_idx..],
        &mut h[x..width],
        &mut s[x..width],
        &mut v[x..width],
        width - x,
      );
    }
  }
}

// ---- Public wrappers --------------------------------------------------------

/// Monoblack → RGB u8 (SSE4.1).
#[allow(dead_code)]
#[inline]
#[target_feature(enable = "sse4.1")]
pub(crate) unsafe fn monoblack_to_rgb_row(data: &[u8], out: &mut [u8], width: usize) {
  unsafe { mono1bit_to_rgb_row::<false>(data, out, width) }
}

/// Monoblack → RGBA u8 (SSE4.1).
#[allow(dead_code)]
#[inline]
#[target_feature(enable = "sse4.1")]
pub(crate) unsafe fn monoblack_to_rgba_row(data: &[u8], out: &mut [u8], width: usize) {
  unsafe { mono1bit_to_rgba_row::<false>(data, out, width) }
}

/// Monoblack → RGB u16 (SSE4.1).
#[allow(dead_code)]
#[inline]
#[target_feature(enable = "sse4.1")]
pub(crate) unsafe fn monoblack_to_rgb_u16_row(data: &[u8], out: &mut [u16], width: usize) {
  unsafe { mono1bit_to_rgb_u16_row::<false>(data, out, width) }
}

/// Monoblack → RGBA u16 (SSE4.1).
#[allow(dead_code)]
#[inline]
#[target_feature(enable = "sse4.1")]
pub(crate) unsafe fn monoblack_to_rgba_u16_row(data: &[u8], out: &mut [u16], width: usize) {
  unsafe { mono1bit_to_rgba_u16_row::<false>(data, out, width) }
}

/// Monoblack → Luma u8 (SSE4.1).
#[allow(dead_code)]
#[inline]
#[target_feature(enable = "sse4.1")]
pub(crate) unsafe fn monoblack_to_luma_row(data: &[u8], out: &mut [u8], width: usize) {
  unsafe { mono1bit_to_luma_row::<false>(data, out, width) }
}

/// Monoblack → Luma u16 (SSE4.1).
#[allow(dead_code)]
#[inline]
#[target_feature(enable = "sse4.1")]
pub(crate) unsafe fn monoblack_to_luma_u16_row(data: &[u8], out: &mut [u16], width: usize) {
  unsafe { mono1bit_to_luma_u16_row::<false>(data, out, width) }
}

/// Monoblack → HSV (SSE4.1).
#[allow(dead_code)]
#[inline]
#[target_feature(enable = "sse4.1")]
pub(crate) unsafe fn monoblack_to_hsv_row(
  data: &[u8],
  h: &mut [u8],
  s: &mut [u8],
  v: &mut [u8],
  width: usize,
) {
  unsafe { mono1bit_to_hsv_row::<false>(data, h, s, v, width) }
}

/// Monowhite → RGB u8 (SSE4.1).
#[allow(dead_code)]
#[inline]
#[target_feature(enable = "sse4.1")]
pub(crate) unsafe fn monowhite_to_rgb_row(data: &[u8], out: &mut [u8], width: usize) {
  unsafe { mono1bit_to_rgb_row::<true>(data, out, width) }
}

/// Monowhite → RGBA u8 (SSE4.1).
#[allow(dead_code)]
#[inline]
#[target_feature(enable = "sse4.1")]
pub(crate) unsafe fn monowhite_to_rgba_row(data: &[u8], out: &mut [u8], width: usize) {
  unsafe { mono1bit_to_rgba_row::<true>(data, out, width) }
}

/// Monowhite → RGB u16 (SSE4.1).
#[allow(dead_code)]
#[inline]
#[target_feature(enable = "sse4.1")]
pub(crate) unsafe fn monowhite_to_rgb_u16_row(data: &[u8], out: &mut [u16], width: usize) {
  unsafe { mono1bit_to_rgb_u16_row::<true>(data, out, width) }
}

/// Monowhite → RGBA u16 (SSE4.1).
#[allow(dead_code)]
#[inline]
#[target_feature(enable = "sse4.1")]
pub(crate) unsafe fn monowhite_to_rgba_u16_row(data: &[u8], out: &mut [u16], width: usize) {
  unsafe { mono1bit_to_rgba_u16_row::<true>(data, out, width) }
}

/// Monowhite → Luma u8 (SSE4.1).
#[allow(dead_code)]
#[inline]
#[target_feature(enable = "sse4.1")]
pub(crate) unsafe fn monowhite_to_luma_row(data: &[u8], out: &mut [u8], width: usize) {
  unsafe { mono1bit_to_luma_row::<true>(data, out, width) }
}

/// Monowhite → Luma u16 (SSE4.1).
#[allow(dead_code)]
#[inline]
#[target_feature(enable = "sse4.1")]
pub(crate) unsafe fn monowhite_to_luma_u16_row(data: &[u8], out: &mut [u16], width: usize) {
  unsafe { mono1bit_to_luma_u16_row::<true>(data, out, width) }
}

/// Monowhite → HSV (SSE4.1).
#[allow(dead_code)]
#[inline]
#[target_feature(enable = "sse4.1")]
pub(crate) unsafe fn monowhite_to_hsv_row(
  data: &[u8],
  h: &mut [u8],
  s: &mut [u8],
  v: &mut [u8],
  width: usize,
) {
  unsafe { mono1bit_to_hsv_row::<true>(data, h, s, v, width) }
}