cosmetic: remove code duplication

The ARM and the X86 implementation of the conversion functions share
the same, non cpu specific implementation in separate files.

This commit removes the code duplication by putting the generic
implementation into a convert_base.c, similar to to convolve_base.c

Change-Id: Ic8d8534a343e27cde79ddc85be4998ebd0cb6e5c
This commit is contained in:
Philipp Maier 2017-03-16 14:50:25 +01:00 committed by Tom Tsou
parent 7e07cf2346
commit fe9769833f
5 changed files with 49 additions and 35 deletions

View File

@ -28,19 +28,6 @@
void neon_convert_ps_si16_4n(short *, const float *, const float *, int);
void neon_convert_si16_ps_4n(float *, const short *, int);
#ifndef HAVE_NEON
static void convert_si16_ps(float *out, const short *in, int len)
{
for (int i = 0; i < len; i++)
out[i] = in[i];
}
static void convert_ps_si16(short *out, const float *in, float scale, int len)
{
for (int i = 0; i < len; i++)
out[i] = in[i] * scale;
}
#else
/* 4*N 16-bit signed integer conversion with remainder */
static void neon_convert_si16_ps(float *out,
const short *in,
@ -79,7 +66,7 @@ void convert_float_short(short *out, const float *in, float scale, int len)
else
neon_convert_ps_si16_4n(out, in, q, len >> 2);
#else
convert_ps_si16(out, in, scale, len);
base_convert_float_short(out, in, scale, len);
#endif
}
@ -91,6 +78,6 @@ void convert_short_float(float *out, const short *in, int len)
else
neon_convert_si16_ps_4n(out, in, len >> 2);
#else
convert_si16_ps(out, in, len);
base_convert_short_float(out, in, len);
#endif
}

View File

@ -2,7 +2,14 @@
#define _CONVERT_H_
void convert_float_short(short *out, const float *in, float scale, int len);
void convert_short_float(float *out, const short *in, int len);
void base_convert_float_short(short *out, const float *in,
float scale, int len);
void base_convert_short_float(float *out, const short *in, int len);
void convert_init(void);
#endif /* _CONVERT_H_ */

View File

@ -0,0 +1,34 @@
/*
* Conversion
* Copyright (C) 2012, 2013 Thomas Tsou <tom@tsou.cc>
*
* This library is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public
* License as published by the Free Software Foundation; either
* version 2.1 of the License, or (at your option) any later version.
*
* This library 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
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
* License along with this library; if not, write to the Free Software
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
*/
#include "convert.h"
void base_convert_float_short(short *out, const float *in,
float scale, int len)
{
for (int i = 0; i < len; i++)
out[i] = in[i] * scale;
}
void base_convert_short_float(float *out, const short *in, int len)
{
for (int i = 0; i < len; i++)
out[i] = in[i];
}

View File

@ -5,6 +5,7 @@ noinst_LTLIBRARIES = libarch.la
libarch_la_SOURCES = \
../common/convolve_base.c \
../common/convert_base.c \
convert.c \
convolve.c
endif

View File

@ -170,28 +170,13 @@ static void _sse_convert_scale_ps_si16_16n(short *restrict out,
}
#endif
__attribute__((optimize("no-tree-vectorize")))
static void convert_scale_ps_si16(short *out, const float *in,
float scale, int len)
{
for (int i = 0; i < len; i++)
out[i] = in[i] * scale;
}
__attribute__((optimize("no-tree-vectorize")))
static void convert_si16_ps(float *out, const short *in, int len)
{
for (int i = 0; i < len; i++)
out[i] = in[i];
}
void convert_init(void)
{
c.convert_scale_ps_si16_16n = convert_scale_ps_si16;
c.convert_scale_ps_si16_8n = convert_scale_ps_si16;
c.convert_scale_ps_si16 = convert_scale_ps_si16;
c.convert_si16_ps_16n = convert_si16_ps;
c.convert_si16_ps = convert_si16_ps;
c.convert_scale_ps_si16_16n = base_convert_float_short;
c.convert_scale_ps_si16_8n = base_convert_float_short;
c.convert_scale_ps_si16 = base_convert_float_short;
c.convert_si16_ps_16n = base_convert_short_float;
c.convert_si16_ps = base_convert_short_float;
#ifdef HAVE_SSE4_1
if (__builtin_cpu_supports("sse4.1")) {