Files
mozjpeg/jsimd_none.c
DRC d729f4da9c ARM NEON SIMD support for YCC-to-RGB565 conversion, and optimizations to the existing YCC-to-RGB conversion code:
-----

aee36252be.patch

From aee36252be20054afce371a92406fc66ba6627b5 Mon Sep 17 00:00:00 2001
From: Siarhei Siamashka <siarhei.siamashka@gmail.com>
Date: Wed, 13 Aug 2014 03:50:22 +0300
Subject: [PATCH] ARM: Faster NEON yuv->rgb conversion for Krait and Cortex-A15

The older code was developed and tested only on ARM Cortex-A8 and ARM Cortex-A9.
Tuning it for newer ARM processors can introduce some speed-up (up to 20%).

The performance of the inner loop (conversion of 8 pixels) improves from
~27 cycles down to ~22 cycles on Qualcomm Krait 300, and from ~20 cycles
down to ~18 cycles on ARM Cortex-A15.

The performance remains exactly the same on ARM Cortex-A7 (~58 cycles),
ARM Cortex-A8 (~25 cycles) and ARM Cortex-A9 (~30 cycles) processors.

Also use larger indentation in the source code for separating two independent
instruction streams.

-----

a5efdbf22c.patch

From a5efdbf22ce9c1acd4b14a353cec863c2c57557e Mon Sep 17 00:00:00 2001
From: Siarhei Siamashka <siarhei.siamashka@gmail.com>
Date: Wed, 13 Aug 2014 07:23:09 +0300
Subject: [PATCH] ARM: NEON optimized yuv->rgb565 conversion

The performance of the inner loop (conversion of 8 pixels):
* ARM Cortex-A7:  ~55 cycles
* ARM Cortex-A8:  ~28 cycles
* ARM Cortex-A9:  ~32 cycles
* ARM Cortex-A15: ~20 cycles
* Qualcomm Krait: ~24 cycles

Based on the Linaro rgb565 patch from
    https://sourceforge.net/p/libjpeg-turbo/patches/24/
but implements better instructions scheduling.


git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@1385 632fc199-4ca6-4c93-a231-07263d6284db
2014-08-23 15:47:51 +00:00

390 lines
7.0 KiB
C

/*
* jsimd_none.c
*
* Copyright 2009 Pierre Ossman <ossman@cendio.se> for Cendio AB
* Copyright 2009-2011, 2014 D. R. Commander
*
* Based on the x86 SIMD extension for IJG JPEG library,
* Copyright (C) 1999-2006, MIYASAKA Masaru.
* For conditions of distribution and use, see copyright notice in jsimdext.inc
*
* This file contains stubs for when there is no SIMD support available.
*/
#define JPEG_INTERNALS
#include "jinclude.h"
#include "jpeglib.h"
#include "jsimd.h"
#include "jdct.h"
#include "jsimddct.h"
GLOBAL(int)
jsimd_can_rgb_ycc (void)
{
return 0;
}
GLOBAL(int)
jsimd_can_rgb_gray (void)
{
return 0;
}
GLOBAL(int)
jsimd_can_ycc_rgb (void)
{
return 0;
}
GLOBAL(int)
jsimd_can_ycc_rgb565 (void)
{
return 0;
}
GLOBAL(int)
jsimd_c_can_null_convert (void)
{
return 0;
}
GLOBAL(void)
jsimd_rgb_ycc_convert (j_compress_ptr cinfo,
JSAMPARRAY input_buf, JSAMPIMAGE output_buf,
JDIMENSION output_row, int num_rows)
{
}
GLOBAL(void)
jsimd_rgb_gray_convert (j_compress_ptr cinfo,
JSAMPARRAY input_buf, JSAMPIMAGE output_buf,
JDIMENSION output_row, int num_rows)
{
}
GLOBAL(void)
jsimd_ycc_rgb_convert (j_decompress_ptr cinfo,
JSAMPIMAGE input_buf, JDIMENSION input_row,
JSAMPARRAY output_buf, int num_rows)
{
}
GLOBAL(void)
jsimd_ycc_rgb565_convert (j_decompress_ptr cinfo,
JSAMPIMAGE input_buf, JDIMENSION input_row,
JSAMPARRAY output_buf, int num_rows)
{
}
GLOBAL(void)
jsimd_c_null_convert (j_compress_ptr cinfo,
JSAMPARRAY input_buf, JSAMPIMAGE output_buf,
JDIMENSION output_row, int num_rows)
{
}
GLOBAL(int)
jsimd_can_h2v2_downsample (void)
{
return 0;
}
GLOBAL(int)
jsimd_can_h2v1_downsample (void)
{
return 0;
}
GLOBAL(int)
jsimd_can_h2v2_smooth_downsample (void)
{
return 0;
}
GLOBAL(void)
jsimd_h2v2_downsample (j_compress_ptr cinfo, jpeg_component_info * compptr,
JSAMPARRAY input_data, JSAMPARRAY output_data)
{
}
GLOBAL(void)
jsimd_h2v2_smooth_downsample (j_compress_ptr cinfo, jpeg_component_info * compptr,
JSAMPARRAY input_data, JSAMPARRAY output_data)
{
}
GLOBAL(void)
jsimd_h2v1_downsample (j_compress_ptr cinfo, jpeg_component_info * compptr,
JSAMPARRAY input_data, JSAMPARRAY output_data)
{
}
GLOBAL(int)
jsimd_can_h2v2_upsample (void)
{
return 0;
}
GLOBAL(int)
jsimd_can_h2v1_upsample (void)
{
return 0;
}
GLOBAL(int)
jsimd_can_int_upsample (void)
{
return 0;
}
GLOBAL(void)
jsimd_int_upsample (j_decompress_ptr cinfo, jpeg_component_info * compptr,
JSAMPARRAY input_data, JSAMPARRAY * output_data_ptr)
{
}
GLOBAL(void)
jsimd_h2v2_upsample (j_decompress_ptr cinfo,
jpeg_component_info * compptr,
JSAMPARRAY input_data,
JSAMPARRAY * output_data_ptr)
{
}
GLOBAL(void)
jsimd_h2v1_upsample (j_decompress_ptr cinfo,
jpeg_component_info * compptr,
JSAMPARRAY input_data,
JSAMPARRAY * output_data_ptr)
{
}
GLOBAL(int)
jsimd_can_h2v2_fancy_upsample (void)
{
return 0;
}
GLOBAL(int)
jsimd_can_h2v1_fancy_upsample (void)
{
return 0;
}
GLOBAL(void)
jsimd_h2v2_fancy_upsample (j_decompress_ptr cinfo,
jpeg_component_info * compptr,
JSAMPARRAY input_data,
JSAMPARRAY * output_data_ptr)
{
}
GLOBAL(void)
jsimd_h2v1_fancy_upsample (j_decompress_ptr cinfo,
jpeg_component_info * compptr,
JSAMPARRAY input_data,
JSAMPARRAY * output_data_ptr)
{
}
GLOBAL(int)
jsimd_can_h2v2_merged_upsample (void)
{
return 0;
}
GLOBAL(int)
jsimd_can_h2v1_merged_upsample (void)
{
return 0;
}
GLOBAL(void)
jsimd_h2v2_merged_upsample (j_decompress_ptr cinfo,
JSAMPIMAGE input_buf,
JDIMENSION in_row_group_ctr,
JSAMPARRAY output_buf)
{
}
GLOBAL(void)
jsimd_h2v1_merged_upsample (j_decompress_ptr cinfo,
JSAMPIMAGE input_buf,
JDIMENSION in_row_group_ctr,
JSAMPARRAY output_buf)
{
}
GLOBAL(int)
jsimd_can_convsamp (void)
{
return 0;
}
GLOBAL(int)
jsimd_can_convsamp_float (void)
{
return 0;
}
GLOBAL(void)
jsimd_convsamp (JSAMPARRAY sample_data, JDIMENSION start_col,
DCTELEM * workspace)
{
}
GLOBAL(void)
jsimd_convsamp_float (JSAMPARRAY sample_data, JDIMENSION start_col,
FAST_FLOAT * workspace)
{
}
GLOBAL(int)
jsimd_can_fdct_islow (void)
{
return 0;
}
GLOBAL(int)
jsimd_can_fdct_ifast (void)
{
return 0;
}
GLOBAL(int)
jsimd_can_fdct_float (void)
{
return 0;
}
GLOBAL(void)
jsimd_fdct_islow (DCTELEM * data)
{
}
GLOBAL(void)
jsimd_fdct_ifast (DCTELEM * data)
{
}
GLOBAL(void)
jsimd_fdct_float (FAST_FLOAT * data)
{
}
GLOBAL(int)
jsimd_can_quantize (void)
{
return 0;
}
GLOBAL(int)
jsimd_can_quantize_float (void)
{
return 0;
}
GLOBAL(void)
jsimd_quantize (JCOEFPTR coef_block, DCTELEM * divisors,
DCTELEM * workspace)
{
}
GLOBAL(void)
jsimd_quantize_float (JCOEFPTR coef_block, FAST_FLOAT * divisors,
FAST_FLOAT * workspace)
{
}
GLOBAL(int)
jsimd_can_idct_2x2 (void)
{
return 0;
}
GLOBAL(int)
jsimd_can_idct_4x4 (void)
{
return 0;
}
GLOBAL(int)
jsimd_can_idct_6x6 (void)
{
return 0;
}
GLOBAL(int)
jsimd_can_idct_12x12 (void)
{
return 0;
}
GLOBAL(void)
jsimd_idct_2x2 (j_decompress_ptr cinfo, jpeg_component_info * compptr,
JCOEFPTR coef_block, JSAMPARRAY output_buf,
JDIMENSION output_col)
{
}
GLOBAL(void)
jsimd_idct_4x4 (j_decompress_ptr cinfo, jpeg_component_info * compptr,
JCOEFPTR coef_block, JSAMPARRAY output_buf,
JDIMENSION output_col)
{
}
GLOBAL(void)
jsimd_idct_6x6 (j_decompress_ptr cinfo, jpeg_component_info * compptr,
JCOEFPTR coef_block, JSAMPARRAY output_buf,
JDIMENSION output_col)
{
}
GLOBAL(void)
jsimd_idct_12x12 (j_decompress_ptr cinfo, jpeg_component_info * compptr,
JCOEFPTR coef_block, JSAMPARRAY output_buf,
JDIMENSION output_col)
{
}
GLOBAL(int)
jsimd_can_idct_islow (void)
{
return 0;
}
GLOBAL(int)
jsimd_can_idct_ifast (void)
{
return 0;
}
GLOBAL(int)
jsimd_can_idct_float (void)
{
return 0;
}
GLOBAL(void)
jsimd_idct_islow (j_decompress_ptr cinfo, jpeg_component_info * compptr,
JCOEFPTR coef_block, JSAMPARRAY output_buf,
JDIMENSION output_col)
{
}
GLOBAL(void)
jsimd_idct_ifast (j_decompress_ptr cinfo, jpeg_component_info * compptr,
JCOEFPTR coef_block, JSAMPARRAY output_buf,
JDIMENSION output_col)
{
}
GLOBAL(void)
jsimd_idct_float (j_decompress_ptr cinfo, jpeg_component_info * compptr,
JCOEFPTR coef_block, JSAMPARRAY output_buf,
JDIMENSION output_col)
{
}