mirror of
https://github.com/FFmpeg/FFmpeg.git
synced 2026-11-06 13:28:17 +00:00
Devin Heitmueller
821902ddf6
Compare packed input and all three output planes across the NEON block sizes, scalar-tail widths, and a 3840-pixel benchmark case.
104 lines
3.8 KiB
C
104 lines
3.8 KiB
C
/*
|
|
* Copyright (c) 2026 Devin Heitmueller <dheitmueller@ltnglobal.com>
|
|
*
|
|
* This file is part of FFmpeg.
|
|
*
|
|
* FFmpeg is free software; you can redistribute it and/or modify
|
|
* it under the terms of the GNU General Public License as published by
|
|
* the Free Software Foundation; either version 2 of the License, or
|
|
* (at your option) any later version.
|
|
*
|
|
* FFmpeg is distributed in the hope that it will be useful,
|
|
* but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|
* GNU General Public License for more details.
|
|
*
|
|
* You should have received a copy of the GNU General Public License along
|
|
* with FFmpeg; if not, write to the Free Software Foundation, Inc.,
|
|
* 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
|
|
*/
|
|
|
|
#include <string.h>
|
|
|
|
#include "libavutil/mem_internal.h"
|
|
|
|
#include "checkasm.h"
|
|
#include "libavcodec/bitpacked_dec_init.h"
|
|
|
|
#define MAX_WIDTH 3840
|
|
|
|
static void randomize_buffers(uint8_t *src0, uint8_t *src1, int len)
|
|
{
|
|
for (int i = 0; i < len; i++) {
|
|
uint8_t value = rnd();
|
|
src0[i] = value;
|
|
src1[i] = value;
|
|
}
|
|
}
|
|
|
|
void checkasm_check_bitpackeddec(void)
|
|
{
|
|
static const int widths[] = { 8, 16, 24, 32, 40, 48, 56, MAX_WIDTH };
|
|
static const int tail_widths[] = { 2, 6, 10, 14, 16 };
|
|
struct BitpackedContext h;
|
|
|
|
ff_bitpackeddec_init(&h);
|
|
|
|
if (check_func(h.unpack_yuv422p10, "bitpacked_unpack_yuv422p10")) {
|
|
LOCAL_ALIGNED_16(uint8_t, src0, [MAX_WIDTH * 5 / 2]);
|
|
LOCAL_ALIGNED_16(uint8_t, src1, [MAX_WIDTH * 5 / 2]);
|
|
LOCAL_ALIGNED_16(uint16_t, y0, [MAX_WIDTH]);
|
|
LOCAL_ALIGNED_16(uint16_t, y1, [MAX_WIDTH]);
|
|
LOCAL_ALIGNED_16(uint16_t, u0, [MAX_WIDTH / 2]);
|
|
LOCAL_ALIGNED_16(uint16_t, u1, [MAX_WIDTH / 2]);
|
|
LOCAL_ALIGNED_16(uint16_t, v0, [MAX_WIDTH / 2]);
|
|
LOCAL_ALIGNED_16(uint16_t, v1, [MAX_WIDTH / 2]);
|
|
declare_func(void, const uint8_t *src, uint16_t *y, uint16_t *u,
|
|
uint16_t *v, int width);
|
|
|
|
for (int i = 0; i < FF_ARRAY_ELEMS(widths); i++) {
|
|
const int width = widths[i];
|
|
const int src_size = width * 5 / 2;
|
|
|
|
randomize_buffers(src0, src1, src_size);
|
|
memset(y0, 0, width * sizeof(*y0));
|
|
memset(y1, 0, width * sizeof(*y1));
|
|
memset(u0, 0, width / 2 * sizeof(*u0));
|
|
memset(u1, 0, width / 2 * sizeof(*u1));
|
|
memset(v0, 0, width / 2 * sizeof(*v0));
|
|
memset(v1, 0, width / 2 * sizeof(*v1));
|
|
|
|
call_ref(src0, y0, u0, v0, width);
|
|
call_new(src1, y1, u1, v1, width);
|
|
if (memcmp(src0, src1, src_size) ||
|
|
memcmp(y0, y1, width * sizeof(*y0)) ||
|
|
memcmp(u0, u1, width / 2 * sizeof(*u0)) ||
|
|
memcmp(v0, v1, width / 2 * sizeof(*v0)))
|
|
fail();
|
|
}
|
|
|
|
for (int i = 0; i < FF_ARRAY_ELEMS(tail_widths); i++) {
|
|
const int width = tail_widths[i];
|
|
const int bulk = width & ~7;
|
|
const int src_size = width * 5 / 2;
|
|
|
|
randomize_buffers(src0, src1, src_size);
|
|
bitpacked_unpack_yuv422p10_c(src0, y0, u0, v0, width);
|
|
if (bulk)
|
|
h.unpack_yuv422p10(src1, y1, u1, v1, bulk);
|
|
bitpacked_unpack_yuv422p10_c(src1 + bulk * 5 / 2,
|
|
y1 + bulk, u1 + bulk / 2, v1 + bulk / 2,
|
|
width - bulk);
|
|
if (memcmp(src0, src1, src_size) ||
|
|
memcmp(y0, y1, width * sizeof(*y0)) ||
|
|
memcmp(u0, u1, width / 2 * sizeof(*u0)) ||
|
|
memcmp(v0, v1, width / 2 * sizeof(*v0)))
|
|
fail();
|
|
}
|
|
|
|
randomize_buffers(src0, src1, MAX_WIDTH * 5 / 2);
|
|
bench_new(src1, y1, u1, v1, MAX_WIDTH);
|
|
}
|
|
report("bitpacked_unpack_yuv422p10");
|
|
}
|