2019-03-04 11:38:24 +00:00
|
|
|
|
|
|
|
/* palette_neon_intrinsics.c - NEON optimised palette expansion functions
|
|
|
|
*
|
2019-04-26 08:37:13 +00:00
|
|
|
* Copyright (c) 2018-2019 Cosmin Truta
|
2019-03-04 11:38:24 +00:00
|
|
|
* Copyright (c) 2017-2018 Arm Holdings. All rights reserved.
|
|
|
|
* Written by Richard Townsend <Richard.Townsend@arm.com>, February 2017.
|
|
|
|
*
|
|
|
|
* This code is released under the libpng license.
|
|
|
|
* For conditions of distribution and use, see the disclaimer
|
|
|
|
* and license in png.h
|
|
|
|
*/
|
|
|
|
|
|
|
|
#include "../pngpriv.h"
|
|
|
|
|
|
|
|
#if PNG_ARM_NEON_IMPLEMENTATION == 1
|
|
|
|
|
|
|
|
#if defined(_MSC_VER) && defined(_M_ARM64)
|
|
|
|
# include <arm64_neon.h>
|
|
|
|
#else
|
|
|
|
# include <arm_neon.h>
|
|
|
|
#endif
|
|
|
|
|
2019-04-26 08:37:13 +00:00
|
|
|
/* Build an RGBA8 palette from the separate RGB and alpha palettes. */
|
2019-03-04 11:38:24 +00:00
|
|
|
void
|
2019-04-26 08:37:13 +00:00
|
|
|
png_riffle_palette_neon(png_structrp png_ptr)
|
2019-03-04 11:38:24 +00:00
|
|
|
{
|
|
|
|
png_const_colorp palette = png_ptr->palette;
|
|
|
|
png_bytep riffled_palette = png_ptr->riffled_palette;
|
|
|
|
png_const_bytep trans_alpha = png_ptr->trans_alpha;
|
|
|
|
int num_trans = png_ptr->num_trans;
|
|
|
|
int i;
|
|
|
|
|
2019-04-26 08:37:13 +00:00
|
|
|
png_debug(1, "in png_riffle_palette_neon");
|
|
|
|
|
2019-03-04 11:38:24 +00:00
|
|
|
/* Initially black, opaque. */
|
|
|
|
uint8x16x4_t w = {{
|
|
|
|
vdupq_n_u8(0x00),
|
|
|
|
vdupq_n_u8(0x00),
|
|
|
|
vdupq_n_u8(0x00),
|
|
|
|
vdupq_n_u8(0xff),
|
|
|
|
}};
|
|
|
|
|
2019-04-26 08:37:13 +00:00
|
|
|
/* First, riffle the RGB colours into an RGBA8 palette.
|
|
|
|
* The alpha component is set to opaque for now.
|
2019-03-04 11:38:24 +00:00
|
|
|
*/
|
2019-04-26 08:37:13 +00:00
|
|
|
for (i = 0; i < 256; i += 16)
|
2019-03-04 11:38:24 +00:00
|
|
|
{
|
|
|
|
uint8x16x3_t v = vld3q_u8((png_const_bytep)(palette + i));
|
|
|
|
w.val[0] = v.val[0];
|
|
|
|
w.val[1] = v.val[1];
|
|
|
|
w.val[2] = v.val[2];
|
|
|
|
vst4q_u8(riffled_palette + (i << 2), w);
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Fix up the missing transparency values. */
|
|
|
|
for (i = 0; i < num_trans; i++)
|
|
|
|
riffled_palette[(i << 2) + 3] = trans_alpha[i];
|
|
|
|
}
|
|
|
|
|
2019-04-26 08:37:13 +00:00
|
|
|
/* Expands a palettized row into RGBA8. */
|
2019-03-04 11:38:24 +00:00
|
|
|
int
|
2019-04-26 08:37:13 +00:00
|
|
|
png_do_expand_palette_rgba8_neon(png_structrp png_ptr, png_row_infop row_info,
|
2019-03-04 11:38:24 +00:00
|
|
|
png_const_bytep row, png_bytepp ssp, png_bytepp ddp)
|
|
|
|
{
|
|
|
|
png_uint_32 row_width = row_info->width;
|
|
|
|
const png_uint_32 *riffled_palette =
|
|
|
|
(const png_uint_32 *)png_ptr->riffled_palette;
|
|
|
|
const png_int_32 pixels_per_chunk = 4;
|
|
|
|
int i;
|
|
|
|
|
2019-04-26 08:37:13 +00:00
|
|
|
png_debug(1, "in png_do_expand_palette_rgba8_neon");
|
|
|
|
|
2019-03-04 11:38:24 +00:00
|
|
|
if (row_width < pixels_per_chunk)
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
/* This function originally gets the last byte of the output row.
|
|
|
|
* The NEON part writes forward from a given position, so we have
|
|
|
|
* to seek this back by 4 pixels x 4 bytes.
|
|
|
|
*/
|
|
|
|
*ddp = *ddp - ((pixels_per_chunk * sizeof(png_uint_32)) - 1);
|
|
|
|
|
|
|
|
for (i = 0; i < row_width; i += pixels_per_chunk)
|
|
|
|
{
|
|
|
|
uint32x4_t cur;
|
|
|
|
png_bytep sp = *ssp - i, dp = *ddp - (i << 2);
|
|
|
|
cur = vld1q_dup_u32 (riffled_palette + *(sp - 3));
|
|
|
|
cur = vld1q_lane_u32(riffled_palette + *(sp - 2), cur, 1);
|
|
|
|
cur = vld1q_lane_u32(riffled_palette + *(sp - 1), cur, 2);
|
|
|
|
cur = vld1q_lane_u32(riffled_palette + *(sp - 0), cur, 3);
|
|
|
|
vst1q_u32((void *)dp, cur);
|
|
|
|
}
|
|
|
|
if (i != row_width)
|
|
|
|
{
|
|
|
|
/* Remove the amount that wasn't processed. */
|
|
|
|
i -= pixels_per_chunk;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Decrement output pointers. */
|
|
|
|
*ssp = *ssp - i;
|
|
|
|
*ddp = *ddp - (i << 2);
|
|
|
|
return i;
|
|
|
|
}
|
|
|
|
|
2019-04-26 08:37:13 +00:00
|
|
|
/* Expands a palettized row into RGB8. */
|
2019-03-04 11:38:24 +00:00
|
|
|
int
|
2019-04-26 08:37:13 +00:00
|
|
|
png_do_expand_palette_rgb8_neon(png_structrp png_ptr, png_row_infop row_info,
|
2019-03-04 11:38:24 +00:00
|
|
|
png_const_bytep row, png_bytepp ssp, png_bytepp ddp)
|
|
|
|
{
|
|
|
|
png_uint_32 row_width = row_info->width;
|
|
|
|
png_const_bytep palette = (png_const_bytep)png_ptr->palette;
|
|
|
|
const png_uint_32 pixels_per_chunk = 8;
|
|
|
|
int i;
|
|
|
|
|
2019-04-26 08:37:13 +00:00
|
|
|
png_debug(1, "in png_do_expand_palette_rgb8_neon");
|
|
|
|
|
2019-03-04 11:38:24 +00:00
|
|
|
if (row_width <= pixels_per_chunk)
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
/* Seeking this back by 8 pixels x 3 bytes. */
|
|
|
|
*ddp = *ddp - ((pixels_per_chunk * sizeof(png_color)) - 1);
|
|
|
|
|
|
|
|
for (i = 0; i < row_width; i += pixels_per_chunk)
|
|
|
|
{
|
|
|
|
uint8x8x3_t cur;
|
|
|
|
png_bytep sp = *ssp - i, dp = *ddp - ((i << 1) + i);
|
|
|
|
cur = vld3_dup_u8(palette + sizeof(png_color) * (*(sp - 7)));
|
|
|
|
cur = vld3_lane_u8(palette + sizeof(png_color) * (*(sp - 6)), cur, 1);
|
|
|
|
cur = vld3_lane_u8(palette + sizeof(png_color) * (*(sp - 5)), cur, 2);
|
|
|
|
cur = vld3_lane_u8(palette + sizeof(png_color) * (*(sp - 4)), cur, 3);
|
|
|
|
cur = vld3_lane_u8(palette + sizeof(png_color) * (*(sp - 3)), cur, 4);
|
|
|
|
cur = vld3_lane_u8(palette + sizeof(png_color) * (*(sp - 2)), cur, 5);
|
|
|
|
cur = vld3_lane_u8(palette + sizeof(png_color) * (*(sp - 1)), cur, 6);
|
|
|
|
cur = vld3_lane_u8(palette + sizeof(png_color) * (*(sp - 0)), cur, 7);
|
|
|
|
vst3_u8((void *)dp, cur);
|
|
|
|
}
|
|
|
|
|
|
|
|
if (i != row_width)
|
|
|
|
{
|
|
|
|
/* Remove the amount that wasn't processed. */
|
|
|
|
i -= pixels_per_chunk;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Decrement output pointers. */
|
|
|
|
*ssp = *ssp - i;
|
|
|
|
*ddp = *ddp - ((i << 1) + i);
|
|
|
|
return i;
|
|
|
|
}
|
|
|
|
|
|
|
|
#endif /* PNG_ARM_NEON_IMPLEMENTATION */
|