2013-02-13 17:26:59 +01:00
|
|
|
/***
|
|
|
|
|
This file is part of PulseAudio.
|
|
|
|
|
|
|
|
|
|
Copyright 2004-2006 Lennart Poettering
|
|
|
|
|
Copyright 2006 Pierre Ossman <ossman@cendio.se> for Cendio AB
|
|
|
|
|
Copyright 2013 Peter Meerwald <pmeerw@pmeerw.net>
|
|
|
|
|
|
|
|
|
|
PulseAudio 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.
|
|
|
|
|
|
|
|
|
|
PulseAudio 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 Lesser General Public License
|
|
|
|
|
along with PulseAudio; if not, write to the Free Software
|
|
|
|
|
Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307
|
|
|
|
|
USA.
|
|
|
|
|
***/
|
|
|
|
|
|
|
|
|
|
#ifdef HAVE_CONFIG_H
|
|
|
|
|
#include <config.h>
|
|
|
|
|
#endif
|
|
|
|
|
|
|
|
|
|
#include <math.h>
|
|
|
|
|
|
|
|
|
|
#include <pulsecore/sample-util.h>
|
|
|
|
|
#include <pulsecore/macro.h>
|
|
|
|
|
#include <pulsecore/g711.h>
|
|
|
|
|
#include <pulsecore/endianmacros.h>
|
|
|
|
|
|
2014-04-18 09:59:32 +02:00
|
|
|
#include "cpu.h"
|
2013-02-13 17:26:59 +01:00
|
|
|
#include "mix.h"
|
|
|
|
|
|
|
|
|
|
#define VOLUME_PADDING 32
|
|
|
|
|
|
|
|
|
|
static void calc_linear_integer_volume(int32_t linear[], const pa_cvolume *volume) {
|
|
|
|
|
unsigned channel, nchannels, padding;
|
|
|
|
|
|
|
|
|
|
pa_assert(linear);
|
|
|
|
|
pa_assert(volume);
|
|
|
|
|
|
|
|
|
|
nchannels = volume->channels;
|
|
|
|
|
|
|
|
|
|
for (channel = 0; channel < nchannels; channel++)
|
|
|
|
|
linear[channel] = (int32_t) lrint(pa_sw_volume_to_linear(volume->values[channel]) * 0x10000);
|
|
|
|
|
|
|
|
|
|
for (padding = 0; padding < VOLUME_PADDING; padding++, channel++)
|
|
|
|
|
linear[channel] = linear[padding];
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static void calc_linear_float_volume(float linear[], const pa_cvolume *volume) {
|
|
|
|
|
unsigned channel, nchannels, padding;
|
|
|
|
|
|
|
|
|
|
pa_assert(linear);
|
|
|
|
|
pa_assert(volume);
|
|
|
|
|
|
|
|
|
|
nchannels = volume->channels;
|
|
|
|
|
|
|
|
|
|
for (channel = 0; channel < nchannels; channel++)
|
|
|
|
|
linear[channel] = (float) pa_sw_volume_to_linear(volume->values[channel]);
|
|
|
|
|
|
|
|
|
|
for (padding = 0; padding < VOLUME_PADDING; padding++, channel++)
|
|
|
|
|
linear[channel] = linear[padding];
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static void calc_linear_integer_stream_volumes(pa_mix_info streams[], unsigned nstreams, const pa_cvolume *volume, const pa_sample_spec *spec) {
|
|
|
|
|
unsigned k, channel;
|
|
|
|
|
float linear[PA_CHANNELS_MAX + VOLUME_PADDING];
|
|
|
|
|
|
|
|
|
|
pa_assert(streams);
|
|
|
|
|
pa_assert(spec);
|
|
|
|
|
pa_assert(volume);
|
|
|
|
|
|
|
|
|
|
calc_linear_float_volume(linear, volume);
|
|
|
|
|
|
|
|
|
|
for (k = 0; k < nstreams; k++) {
|
|
|
|
|
|
|
|
|
|
for (channel = 0; channel < spec->channels; channel++) {
|
|
|
|
|
pa_mix_info *m = streams + k;
|
|
|
|
|
m->linear[channel].i = (int32_t) lrint(pa_sw_volume_to_linear(m->volume.values[channel]) * linear[channel] * 0x10000);
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static void calc_linear_float_stream_volumes(pa_mix_info streams[], unsigned nstreams, const pa_cvolume *volume, const pa_sample_spec *spec) {
|
|
|
|
|
unsigned k, channel;
|
|
|
|
|
float linear[PA_CHANNELS_MAX + VOLUME_PADDING];
|
|
|
|
|
|
|
|
|
|
pa_assert(streams);
|
|
|
|
|
pa_assert(spec);
|
|
|
|
|
pa_assert(volume);
|
|
|
|
|
|
|
|
|
|
calc_linear_float_volume(linear, volume);
|
|
|
|
|
|
|
|
|
|
for (k = 0; k < nstreams; k++) {
|
|
|
|
|
|
|
|
|
|
for (channel = 0; channel < spec->channels; channel++) {
|
|
|
|
|
pa_mix_info *m = streams + k;
|
|
|
|
|
m->linear[channel].f = (float) (pa_sw_volume_to_linear(m->volume.values[channel]) * linear[channel]);
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2013-02-13 17:27:01 +01:00
|
|
|
typedef void (*pa_calc_stream_volumes_func_t) (pa_mix_info streams[], unsigned nstreams, const pa_cvolume *volume, const pa_sample_spec *spec);
|
|
|
|
|
|
|
|
|
|
static const pa_calc_stream_volumes_func_t calc_stream_volumes_table[] = {
|
|
|
|
|
[PA_SAMPLE_U8] = (pa_calc_stream_volumes_func_t) calc_linear_integer_stream_volumes,
|
|
|
|
|
[PA_SAMPLE_ALAW] = (pa_calc_stream_volumes_func_t) calc_linear_integer_stream_volumes,
|
|
|
|
|
[PA_SAMPLE_ULAW] = (pa_calc_stream_volumes_func_t) calc_linear_integer_stream_volumes,
|
|
|
|
|
[PA_SAMPLE_S16LE] = (pa_calc_stream_volumes_func_t) calc_linear_integer_stream_volumes,
|
|
|
|
|
[PA_SAMPLE_S16BE] = (pa_calc_stream_volumes_func_t) calc_linear_integer_stream_volumes,
|
|
|
|
|
[PA_SAMPLE_FLOAT32LE] = (pa_calc_stream_volumes_func_t) calc_linear_float_stream_volumes,
|
|
|
|
|
[PA_SAMPLE_FLOAT32BE] = (pa_calc_stream_volumes_func_t) calc_linear_float_stream_volumes,
|
|
|
|
|
[PA_SAMPLE_S32LE] = (pa_calc_stream_volumes_func_t) calc_linear_integer_stream_volumes,
|
|
|
|
|
[PA_SAMPLE_S32BE] = (pa_calc_stream_volumes_func_t) calc_linear_integer_stream_volumes,
|
|
|
|
|
[PA_SAMPLE_S24LE] = (pa_calc_stream_volumes_func_t) calc_linear_integer_stream_volumes,
|
|
|
|
|
[PA_SAMPLE_S24BE] = (pa_calc_stream_volumes_func_t) calc_linear_integer_stream_volumes,
|
|
|
|
|
[PA_SAMPLE_S24_32LE] = (pa_calc_stream_volumes_func_t) calc_linear_integer_stream_volumes,
|
|
|
|
|
[PA_SAMPLE_S24_32BE] = (pa_calc_stream_volumes_func_t) calc_linear_integer_stream_volumes
|
|
|
|
|
};
|
|
|
|
|
|
2013-02-13 17:27:07 +01:00
|
|
|
/* special case: mix 2 s16ne streams, 1 channel each */
|
2013-02-13 17:27:08 +01:00
|
|
|
static void pa_mix2_ch1_s16ne(pa_mix_info streams[], int16_t *data, unsigned length) {
|
2013-02-13 17:27:07 +01:00
|
|
|
const int16_t *ptr0 = streams[0].ptr;
|
|
|
|
|
const int16_t *ptr1 = streams[1].ptr;
|
|
|
|
|
|
|
|
|
|
const int32_t cv0 = streams[0].linear[0].i;
|
|
|
|
|
const int32_t cv1 = streams[1].linear[0].i;
|
|
|
|
|
|
2013-02-13 17:27:08 +01:00
|
|
|
length /= sizeof(int16_t);
|
|
|
|
|
|
|
|
|
|
for (; length > 0; length--) {
|
2013-02-13 17:27:07 +01:00
|
|
|
int32_t sum;
|
|
|
|
|
|
|
|
|
|
sum = pa_mult_s16_volume(*ptr0++, cv0);
|
|
|
|
|
sum += pa_mult_s16_volume(*ptr1++, cv1);
|
|
|
|
|
|
|
|
|
|
sum = PA_CLAMP_UNLIKELY(sum, -0x8000, 0x7FFF);
|
|
|
|
|
*data++ = sum;
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/* special case: mix 2 s16ne streams, 2 channels each */
|
2013-02-13 17:27:08 +01:00
|
|
|
static void pa_mix2_ch2_s16ne(pa_mix_info streams[], int16_t *data, unsigned length) {
|
2013-02-13 17:27:07 +01:00
|
|
|
const int16_t *ptr0 = streams[0].ptr;
|
|
|
|
|
const int16_t *ptr1 = streams[1].ptr;
|
|
|
|
|
|
2013-02-13 17:27:08 +01:00
|
|
|
length /= sizeof(int16_t) * 2;
|
|
|
|
|
|
|
|
|
|
for (; length > 0; length--) {
|
2013-02-13 17:27:07 +01:00
|
|
|
int32_t sum;
|
|
|
|
|
|
|
|
|
|
sum = pa_mult_s16_volume(*ptr0++, streams[0].linear[0].i);
|
|
|
|
|
sum += pa_mult_s16_volume(*ptr1++, streams[1].linear[0].i);
|
|
|
|
|
|
|
|
|
|
sum = PA_CLAMP_UNLIKELY(sum, -0x8000, 0x7FFF);
|
|
|
|
|
*data++ = sum;
|
|
|
|
|
|
|
|
|
|
sum = pa_mult_s16_volume(*ptr0++, streams[0].linear[1].i);
|
|
|
|
|
sum += pa_mult_s16_volume(*ptr1++, streams[1].linear[1].i);
|
|
|
|
|
|
|
|
|
|
sum = PA_CLAMP_UNLIKELY(sum, -0x8000, 0x7FFF);
|
|
|
|
|
*data++ = sum;
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/* special case: mix 2 s16ne streams */
|
2013-02-13 17:27:08 +01:00
|
|
|
static void pa_mix2_s16ne(pa_mix_info streams[], unsigned channels, int16_t *data, unsigned length) {
|
2013-02-13 17:27:07 +01:00
|
|
|
const int16_t *ptr0 = streams[0].ptr;
|
|
|
|
|
const int16_t *ptr1 = streams[1].ptr;
|
|
|
|
|
unsigned channel = 0;
|
|
|
|
|
|
2013-02-13 17:27:08 +01:00
|
|
|
length /= sizeof(int16_t);
|
|
|
|
|
|
|
|
|
|
for (; length > 0; length--) {
|
2013-02-13 17:27:07 +01:00
|
|
|
int32_t sum;
|
|
|
|
|
|
|
|
|
|
sum = pa_mult_s16_volume(*ptr0++, streams[0].linear[channel].i);
|
|
|
|
|
sum += pa_mult_s16_volume(*ptr1++, streams[1].linear[channel].i);
|
|
|
|
|
|
|
|
|
|
sum = PA_CLAMP_UNLIKELY(sum, -0x8000, 0x7FFF);
|
|
|
|
|
*data++ = sum;
|
|
|
|
|
|
|
|
|
|
if (PA_UNLIKELY(++channel >= channels))
|
|
|
|
|
channel = 0;
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/* special case: mix s16ne streams, 2 channels each */
|
2013-02-13 17:27:08 +01:00
|
|
|
static void pa_mix_ch2_s16ne(pa_mix_info streams[], unsigned nstreams, int16_t *data, unsigned length) {
|
|
|
|
|
|
|
|
|
|
length /= sizeof(int16_t) * 2;
|
|
|
|
|
|
|
|
|
|
for (; length > 0; length--) {
|
2013-02-13 17:27:07 +01:00
|
|
|
int32_t sum0 = 0, sum1 = 0;
|
|
|
|
|
unsigned i;
|
|
|
|
|
|
|
|
|
|
for (i = 0; i < nstreams; i++) {
|
|
|
|
|
pa_mix_info *m = streams + i;
|
|
|
|
|
int32_t cv0 = m->linear[0].i;
|
|
|
|
|
int32_t cv1 = m->linear[1].i;
|
|
|
|
|
|
|
|
|
|
sum0 += pa_mult_s16_volume(*((int16_t*) m->ptr), cv0);
|
|
|
|
|
m->ptr = (uint8_t*) m->ptr + sizeof(int16_t);
|
|
|
|
|
|
|
|
|
|
sum1 += pa_mult_s16_volume(*((int16_t*) m->ptr), cv1);
|
|
|
|
|
m->ptr = (uint8_t*) m->ptr + sizeof(int16_t);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
*data++ = PA_CLAMP_UNLIKELY(sum0, -0x8000, 0x7FFF);
|
|
|
|
|
*data++ = PA_CLAMP_UNLIKELY(sum1, -0x8000, 0x7FFF);
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2013-02-13 17:27:08 +01:00
|
|
|
static void pa_mix_generic_s16ne(pa_mix_info streams[], unsigned nstreams, unsigned channels, int16_t *data, unsigned length) {
|
2013-02-13 17:27:02 +01:00
|
|
|
unsigned channel = 0;
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:08 +01:00
|
|
|
length /= sizeof(int16_t);
|
|
|
|
|
|
|
|
|
|
for (; length > 0; length--) {
|
2013-02-13 17:27:02 +01:00
|
|
|
int32_t sum = 0;
|
|
|
|
|
unsigned i;
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:02 +01:00
|
|
|
for (i = 0; i < nstreams; i++) {
|
|
|
|
|
pa_mix_info *m = streams + i;
|
core: Refactor code to multiply s16 by volume
move code to function pa_mult_s16_volume() in sample-util.h
use 64 bit integers on 64 bit platforms (it's faster)
on i5, 2.5GHz (64-bit)
Running suite(s): Mult-s16
32 bit mult: 1272300 usec (avg: 12723, min = 12533, max = 18749, stddev = 620.48).
64 bit mult: 852241 usec (avg: 8522.41, min = 8420, max = 9148, stddev = 109.388).
100%: Checks: 1, Failures: 0, Errors: 0
on Pentium D, 3.4GHz (32-bit)
Running suite(s): Mult-s16
32 bit mult: 2228504 usec (avg: 22285, min = 18775, max = 29648, stddev = 3865.59).
64 bit mult: 5546861 usec (avg: 55468.6, min = 55028, max = 64924, stddev = 978.981).
100%: Checks: 1, Failures: 0, Errors: 0
on TI DM3730, Cortex-A8, 800MHz (32-bit)
Running suite(s): Mult-s16
32 bit mult: 23708900 usec (avg: 237089, min = 191864, max = 557312, stddev = 77503.6).
64 bit mult: 22190039 usec (avg: 221900, min = 177978, max = 480469, stddev = 68520.5).
100%: Checks: 1, Failures: 0, Errors: 0
there is a test program called mult-s16-test which checks that the functions compute the
same results, and compares runtime
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
2013-02-13 17:27:05 +01:00
|
|
|
int32_t cv = m->linear[channel].i;
|
2013-02-13 17:26:59 +01:00
|
|
|
|
core: Refactor code to multiply s16 by volume
move code to function pa_mult_s16_volume() in sample-util.h
use 64 bit integers on 64 bit platforms (it's faster)
on i5, 2.5GHz (64-bit)
Running suite(s): Mult-s16
32 bit mult: 1272300 usec (avg: 12723, min = 12533, max = 18749, stddev = 620.48).
64 bit mult: 852241 usec (avg: 8522.41, min = 8420, max = 9148, stddev = 109.388).
100%: Checks: 1, Failures: 0, Errors: 0
on Pentium D, 3.4GHz (32-bit)
Running suite(s): Mult-s16
32 bit mult: 2228504 usec (avg: 22285, min = 18775, max = 29648, stddev = 3865.59).
64 bit mult: 5546861 usec (avg: 55468.6, min = 55028, max = 64924, stddev = 978.981).
100%: Checks: 1, Failures: 0, Errors: 0
on TI DM3730, Cortex-A8, 800MHz (32-bit)
Running suite(s): Mult-s16
32 bit mult: 23708900 usec (avg: 237089, min = 191864, max = 557312, stddev = 77503.6).
64 bit mult: 22190039 usec (avg: 221900, min = 177978, max = 480469, stddev = 68520.5).
100%: Checks: 1, Failures: 0, Errors: 0
there is a test program called mult-s16-test which checks that the functions compute the
same results, and compares runtime
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
2013-02-13 17:27:05 +01:00
|
|
|
if (PA_LIKELY(cv > 0))
|
2013-02-13 17:27:07 +01:00
|
|
|
sum += pa_mult_s16_volume(*((int16_t*) m->ptr), cv);
|
2013-02-13 17:27:02 +01:00
|
|
|
m->ptr = (uint8_t*) m->ptr + sizeof(int16_t);
|
|
|
|
|
}
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:02 +01:00
|
|
|
sum = PA_CLAMP_UNLIKELY(sum, -0x8000, 0x7FFF);
|
2013-02-13 17:27:08 +01:00
|
|
|
*data++ = sum;
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:02 +01:00
|
|
|
if (PA_UNLIKELY(++channel >= channels))
|
|
|
|
|
channel = 0;
|
|
|
|
|
}
|
|
|
|
|
}
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:08 +01:00
|
|
|
static void pa_mix_s16ne_c(pa_mix_info streams[], unsigned nstreams, unsigned channels, int16_t *data, unsigned length) {
|
2013-02-13 17:27:07 +01:00
|
|
|
if (nstreams == 2 && channels == 1)
|
2013-02-13 17:27:08 +01:00
|
|
|
pa_mix2_ch1_s16ne(streams, data, length);
|
2013-02-13 17:27:07 +01:00
|
|
|
else if (nstreams == 2 && channels == 2)
|
2013-02-13 17:27:08 +01:00
|
|
|
pa_mix2_ch2_s16ne(streams, data, length);
|
2013-02-13 17:27:07 +01:00
|
|
|
else if (nstreams == 2)
|
2013-02-13 17:27:08 +01:00
|
|
|
pa_mix2_s16ne(streams, channels, data, length);
|
2013-02-13 17:27:07 +01:00
|
|
|
else if (channels == 2)
|
2013-04-09 14:52:54 +03:00
|
|
|
pa_mix_ch2_s16ne(streams, nstreams, data, length);
|
2013-02-13 17:27:07 +01:00
|
|
|
else
|
2013-02-13 17:27:08 +01:00
|
|
|
pa_mix_generic_s16ne(streams, nstreams, channels, data, length);
|
2013-02-13 17:27:07 +01:00
|
|
|
}
|
|
|
|
|
|
2013-02-13 17:27:08 +01:00
|
|
|
static void pa_mix_s16re_c(pa_mix_info streams[], unsigned nstreams, unsigned channels, int16_t *data, unsigned length) {
|
2013-02-13 17:27:02 +01:00
|
|
|
unsigned channel = 0;
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:08 +01:00
|
|
|
length /= sizeof(int16_t);
|
|
|
|
|
|
|
|
|
|
for (; length > 0; length--, data++) {
|
2013-02-13 17:27:02 +01:00
|
|
|
int32_t sum = 0;
|
|
|
|
|
unsigned i;
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:02 +01:00
|
|
|
for (i = 0; i < nstreams; i++) {
|
|
|
|
|
pa_mix_info *m = streams + i;
|
core: Refactor code to multiply s16 by volume
move code to function pa_mult_s16_volume() in sample-util.h
use 64 bit integers on 64 bit platforms (it's faster)
on i5, 2.5GHz (64-bit)
Running suite(s): Mult-s16
32 bit mult: 1272300 usec (avg: 12723, min = 12533, max = 18749, stddev = 620.48).
64 bit mult: 852241 usec (avg: 8522.41, min = 8420, max = 9148, stddev = 109.388).
100%: Checks: 1, Failures: 0, Errors: 0
on Pentium D, 3.4GHz (32-bit)
Running suite(s): Mult-s16
32 bit mult: 2228504 usec (avg: 22285, min = 18775, max = 29648, stddev = 3865.59).
64 bit mult: 5546861 usec (avg: 55468.6, min = 55028, max = 64924, stddev = 978.981).
100%: Checks: 1, Failures: 0, Errors: 0
on TI DM3730, Cortex-A8, 800MHz (32-bit)
Running suite(s): Mult-s16
32 bit mult: 23708900 usec (avg: 237089, min = 191864, max = 557312, stddev = 77503.6).
64 bit mult: 22190039 usec (avg: 221900, min = 177978, max = 480469, stddev = 68520.5).
100%: Checks: 1, Failures: 0, Errors: 0
there is a test program called mult-s16-test which checks that the functions compute the
same results, and compares runtime
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
2013-02-13 17:27:05 +01:00
|
|
|
int32_t cv = m->linear[channel].i;
|
2013-02-13 17:26:59 +01:00
|
|
|
|
core: Refactor code to multiply s16 by volume
move code to function pa_mult_s16_volume() in sample-util.h
use 64 bit integers on 64 bit platforms (it's faster)
on i5, 2.5GHz (64-bit)
Running suite(s): Mult-s16
32 bit mult: 1272300 usec (avg: 12723, min = 12533, max = 18749, stddev = 620.48).
64 bit mult: 852241 usec (avg: 8522.41, min = 8420, max = 9148, stddev = 109.388).
100%: Checks: 1, Failures: 0, Errors: 0
on Pentium D, 3.4GHz (32-bit)
Running suite(s): Mult-s16
32 bit mult: 2228504 usec (avg: 22285, min = 18775, max = 29648, stddev = 3865.59).
64 bit mult: 5546861 usec (avg: 55468.6, min = 55028, max = 64924, stddev = 978.981).
100%: Checks: 1, Failures: 0, Errors: 0
on TI DM3730, Cortex-A8, 800MHz (32-bit)
Running suite(s): Mult-s16
32 bit mult: 23708900 usec (avg: 237089, min = 191864, max = 557312, stddev = 77503.6).
64 bit mult: 22190039 usec (avg: 221900, min = 177978, max = 480469, stddev = 68520.5).
100%: Checks: 1, Failures: 0, Errors: 0
there is a test program called mult-s16-test which checks that the functions compute the
same results, and compares runtime
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
2013-02-13 17:27:05 +01:00
|
|
|
if (PA_LIKELY(cv > 0))
|
|
|
|
|
sum += pa_mult_s16_volume(PA_INT16_SWAP(*((int16_t*) m->ptr)), cv);
|
2013-02-13 17:27:02 +01:00
|
|
|
m->ptr = (uint8_t*) m->ptr + sizeof(int16_t);
|
2013-02-13 17:26:59 +01:00
|
|
|
}
|
|
|
|
|
|
2013-02-13 17:27:02 +01:00
|
|
|
sum = PA_CLAMP_UNLIKELY(sum, -0x8000, 0x7FFF);
|
2013-02-13 17:27:08 +01:00
|
|
|
*data = PA_INT16_SWAP((int16_t) sum);
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:02 +01:00
|
|
|
if (PA_UNLIKELY(++channel >= channels))
|
|
|
|
|
channel = 0;
|
|
|
|
|
}
|
|
|
|
|
}
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:08 +01:00
|
|
|
static void pa_mix_s32ne_c(pa_mix_info streams[], unsigned nstreams, unsigned channels, int32_t *data, unsigned length) {
|
2013-02-13 17:27:02 +01:00
|
|
|
unsigned channel = 0;
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:08 +01:00
|
|
|
length /= sizeof(int32_t);
|
|
|
|
|
|
|
|
|
|
for (; length > 0; length--, data++) {
|
2013-02-13 17:27:02 +01:00
|
|
|
int64_t sum = 0;
|
|
|
|
|
unsigned i;
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:02 +01:00
|
|
|
for (i = 0; i < nstreams; i++) {
|
|
|
|
|
pa_mix_info *m = streams + i;
|
|
|
|
|
int32_t cv = m->linear[channel].i;
|
|
|
|
|
int64_t v;
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:02 +01:00
|
|
|
if (PA_LIKELY(cv > 0)) {
|
|
|
|
|
v = *((int32_t*) m->ptr);
|
|
|
|
|
v = (v * cv) >> 16;
|
|
|
|
|
sum += v;
|
2013-02-13 17:26:59 +01:00
|
|
|
}
|
2013-02-13 17:27:02 +01:00
|
|
|
m->ptr = (uint8_t*) m->ptr + sizeof(int32_t);
|
2013-02-13 17:26:59 +01:00
|
|
|
}
|
|
|
|
|
|
2013-02-13 17:27:02 +01:00
|
|
|
sum = PA_CLAMP_UNLIKELY(sum, -0x80000000LL, 0x7FFFFFFFLL);
|
2013-02-13 17:27:08 +01:00
|
|
|
*data = (int32_t) sum;
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:02 +01:00
|
|
|
if (PA_UNLIKELY(++channel >= channels))
|
|
|
|
|
channel = 0;
|
|
|
|
|
}
|
|
|
|
|
}
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:08 +01:00
|
|
|
static void pa_mix_s32re_c(pa_mix_info streams[], unsigned nstreams, unsigned channels, int32_t *data, unsigned length) {
|
2013-02-13 17:27:02 +01:00
|
|
|
unsigned channel = 0;
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:08 +01:00
|
|
|
length /= sizeof(int32_t);
|
|
|
|
|
|
|
|
|
|
for (; length > 0; length--, data++) {
|
2013-02-13 17:27:02 +01:00
|
|
|
int64_t sum = 0;
|
|
|
|
|
unsigned i;
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:02 +01:00
|
|
|
for (i = 0; i < nstreams; i++) {
|
|
|
|
|
pa_mix_info *m = streams + i;
|
|
|
|
|
int32_t cv = m->linear[channel].i;
|
|
|
|
|
int64_t v;
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:02 +01:00
|
|
|
if (PA_LIKELY(cv > 0)) {
|
|
|
|
|
v = PA_INT32_SWAP(*((int32_t*) m->ptr));
|
|
|
|
|
v = (v * cv) >> 16;
|
|
|
|
|
sum += v;
|
2013-02-13 17:26:59 +01:00
|
|
|
}
|
2013-02-13 17:27:02 +01:00
|
|
|
m->ptr = (uint8_t*) m->ptr + sizeof(int32_t);
|
2013-02-13 17:26:59 +01:00
|
|
|
}
|
|
|
|
|
|
2013-02-13 17:27:02 +01:00
|
|
|
sum = PA_CLAMP_UNLIKELY(sum, -0x80000000LL, 0x7FFFFFFFLL);
|
2013-02-13 17:27:08 +01:00
|
|
|
*data = PA_INT32_SWAP((int32_t) sum);
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:02 +01:00
|
|
|
if (PA_UNLIKELY(++channel >= channels))
|
|
|
|
|
channel = 0;
|
|
|
|
|
}
|
|
|
|
|
}
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:08 +01:00
|
|
|
static void pa_mix_s24ne_c(pa_mix_info streams[], unsigned nstreams, unsigned channels, uint8_t *data, unsigned length) {
|
2013-02-13 17:27:02 +01:00
|
|
|
unsigned channel = 0;
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:08 +01:00
|
|
|
for (; length > 0; length -= 3, data += 3) {
|
2013-02-13 17:27:02 +01:00
|
|
|
int64_t sum = 0;
|
|
|
|
|
unsigned i;
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:02 +01:00
|
|
|
for (i = 0; i < nstreams; i++) {
|
|
|
|
|
pa_mix_info *m = streams + i;
|
|
|
|
|
int32_t cv = m->linear[channel].i;
|
|
|
|
|
int64_t v;
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:02 +01:00
|
|
|
if (PA_LIKELY(cv > 0)) {
|
|
|
|
|
v = (int32_t) (PA_READ24NE(m->ptr) << 8);
|
|
|
|
|
v = (v * cv) >> 16;
|
|
|
|
|
sum += v;
|
2013-02-13 17:26:59 +01:00
|
|
|
}
|
2013-02-13 17:27:02 +01:00
|
|
|
m->ptr = (uint8_t*) m->ptr + 3;
|
2013-02-13 17:26:59 +01:00
|
|
|
}
|
|
|
|
|
|
2013-02-13 17:27:02 +01:00
|
|
|
sum = PA_CLAMP_UNLIKELY(sum, -0x80000000LL, 0x7FFFFFFFLL);
|
|
|
|
|
PA_WRITE24NE(data, ((uint32_t) sum) >> 8);
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:02 +01:00
|
|
|
if (PA_UNLIKELY(++channel >= channels))
|
|
|
|
|
channel = 0;
|
|
|
|
|
}
|
|
|
|
|
}
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:08 +01:00
|
|
|
static void pa_mix_s24re_c(pa_mix_info streams[], unsigned nstreams, unsigned channels, uint8_t *data, unsigned length) {
|
2013-02-13 17:27:02 +01:00
|
|
|
unsigned channel = 0;
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:08 +01:00
|
|
|
for (; length > 0; length -= 3, data += 3) {
|
2013-02-13 17:27:02 +01:00
|
|
|
int64_t sum = 0;
|
|
|
|
|
unsigned i;
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:02 +01:00
|
|
|
for (i = 0; i < nstreams; i++) {
|
|
|
|
|
pa_mix_info *m = streams + i;
|
|
|
|
|
int32_t cv = m->linear[channel].i;
|
|
|
|
|
int64_t v;
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:02 +01:00
|
|
|
if (PA_LIKELY(cv > 0)) {
|
|
|
|
|
v = (int32_t) (PA_READ24RE(m->ptr) << 8);
|
|
|
|
|
v = (v * cv) >> 16;
|
|
|
|
|
sum += v;
|
2013-02-13 17:26:59 +01:00
|
|
|
}
|
2013-02-13 17:27:02 +01:00
|
|
|
m->ptr = (uint8_t*) m->ptr + 3;
|
2013-02-13 17:26:59 +01:00
|
|
|
}
|
|
|
|
|
|
2013-02-13 17:27:02 +01:00
|
|
|
sum = PA_CLAMP_UNLIKELY(sum, -0x80000000LL, 0x7FFFFFFFLL);
|
|
|
|
|
PA_WRITE24RE(data, ((uint32_t) sum) >> 8);
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:02 +01:00
|
|
|
if (PA_UNLIKELY(++channel >= channels))
|
|
|
|
|
channel = 0;
|
|
|
|
|
}
|
|
|
|
|
}
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:08 +01:00
|
|
|
static void pa_mix_s24_32ne_c(pa_mix_info streams[], unsigned nstreams, unsigned channels, uint32_t *data, unsigned length) {
|
2013-02-13 17:27:02 +01:00
|
|
|
unsigned channel = 0;
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:08 +01:00
|
|
|
length /= sizeof(uint32_t);
|
|
|
|
|
|
|
|
|
|
for (; length > 0; length--, data++) {
|
2013-02-13 17:27:02 +01:00
|
|
|
int64_t sum = 0;
|
|
|
|
|
unsigned i;
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:02 +01:00
|
|
|
for (i = 0; i < nstreams; i++) {
|
|
|
|
|
pa_mix_info *m = streams + i;
|
|
|
|
|
int32_t cv = m->linear[channel].i;
|
|
|
|
|
int64_t v;
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:02 +01:00
|
|
|
if (PA_LIKELY(cv > 0)) {
|
|
|
|
|
v = (int32_t) (*((uint32_t*)m->ptr) << 8);
|
|
|
|
|
v = (v * cv) >> 16;
|
|
|
|
|
sum += v;
|
2013-02-13 17:26:59 +01:00
|
|
|
}
|
2013-02-13 17:27:02 +01:00
|
|
|
m->ptr = (uint8_t*) m->ptr + sizeof(int32_t);
|
2013-02-13 17:26:59 +01:00
|
|
|
}
|
|
|
|
|
|
2013-02-13 17:27:02 +01:00
|
|
|
sum = PA_CLAMP_UNLIKELY(sum, -0x80000000LL, 0x7FFFFFFFLL);
|
2013-02-13 17:27:08 +01:00
|
|
|
*data = ((uint32_t) (int32_t) sum) >> 8;
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:02 +01:00
|
|
|
if (PA_UNLIKELY(++channel >= channels))
|
|
|
|
|
channel = 0;
|
|
|
|
|
}
|
|
|
|
|
}
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:08 +01:00
|
|
|
static void pa_mix_s24_32re_c(pa_mix_info streams[], unsigned nstreams, unsigned channels, uint32_t *data, unsigned length) {
|
2013-02-13 17:27:02 +01:00
|
|
|
unsigned channel = 0;
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:08 +01:00
|
|
|
length /= sizeof(uint32_t);
|
|
|
|
|
|
|
|
|
|
for (; length > 0; length--, data++) {
|
2013-02-13 17:27:02 +01:00
|
|
|
int64_t sum = 0;
|
|
|
|
|
unsigned i;
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:02 +01:00
|
|
|
for (i = 0; i < nstreams; i++) {
|
|
|
|
|
pa_mix_info *m = streams + i;
|
|
|
|
|
int32_t cv = m->linear[channel].i;
|
|
|
|
|
int64_t v;
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:02 +01:00
|
|
|
if (PA_LIKELY(cv > 0)) {
|
|
|
|
|
v = (int32_t) (PA_UINT32_SWAP(*((uint32_t*) m->ptr)) << 8);
|
|
|
|
|
v = (v * cv) >> 16;
|
|
|
|
|
sum += v;
|
2013-02-13 17:26:59 +01:00
|
|
|
}
|
2014-08-04 14:42:00 +02:00
|
|
|
m->ptr = (uint8_t*) m->ptr + sizeof(int32_t);
|
2013-02-13 17:26:59 +01:00
|
|
|
}
|
|
|
|
|
|
2013-02-13 17:27:02 +01:00
|
|
|
sum = PA_CLAMP_UNLIKELY(sum, -0x80000000LL, 0x7FFFFFFFLL);
|
2013-02-13 17:27:08 +01:00
|
|
|
*data = PA_INT32_SWAP(((uint32_t) (int32_t) sum) >> 8);
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:02 +01:00
|
|
|
if (PA_UNLIKELY(++channel >= channels))
|
|
|
|
|
channel = 0;
|
|
|
|
|
}
|
|
|
|
|
}
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:08 +01:00
|
|
|
static void pa_mix_u8_c(pa_mix_info streams[], unsigned nstreams, unsigned channels, uint8_t *data, unsigned length) {
|
2013-02-13 17:27:02 +01:00
|
|
|
unsigned channel = 0;
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:08 +01:00
|
|
|
length /= sizeof(uint8_t);
|
|
|
|
|
|
|
|
|
|
for (; length > 0; length--, data++) {
|
2013-02-13 17:27:02 +01:00
|
|
|
int32_t sum = 0;
|
|
|
|
|
unsigned i;
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:02 +01:00
|
|
|
for (i = 0; i < nstreams; i++) {
|
|
|
|
|
pa_mix_info *m = streams + i;
|
|
|
|
|
int32_t v, cv = m->linear[channel].i;
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:02 +01:00
|
|
|
if (PA_LIKELY(cv > 0)) {
|
|
|
|
|
v = (int32_t) *((uint8_t*) m->ptr) - 0x80;
|
|
|
|
|
v = (v * cv) >> 16;
|
|
|
|
|
sum += v;
|
2013-02-13 17:26:59 +01:00
|
|
|
}
|
2013-02-13 17:27:02 +01:00
|
|
|
m->ptr = (uint8_t*) m->ptr + 1;
|
2013-02-13 17:26:59 +01:00
|
|
|
}
|
|
|
|
|
|
2013-02-13 17:27:02 +01:00
|
|
|
sum = PA_CLAMP_UNLIKELY(sum, -0x80, 0x7F);
|
2013-02-13 17:27:08 +01:00
|
|
|
*data = (uint8_t) (sum + 0x80);
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:02 +01:00
|
|
|
if (PA_UNLIKELY(++channel >= channels))
|
|
|
|
|
channel = 0;
|
|
|
|
|
}
|
|
|
|
|
}
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:08 +01:00
|
|
|
static void pa_mix_ulaw_c(pa_mix_info streams[], unsigned nstreams, unsigned channels, uint8_t *data, unsigned length) {
|
2013-02-13 17:27:02 +01:00
|
|
|
unsigned channel = 0;
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:08 +01:00
|
|
|
length /= sizeof(uint8_t);
|
|
|
|
|
|
|
|
|
|
for (; length > 0; length--, data++) {
|
2013-02-13 17:27:02 +01:00
|
|
|
int32_t sum = 0;
|
|
|
|
|
unsigned i;
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:02 +01:00
|
|
|
for (i = 0; i < nstreams; i++) {
|
|
|
|
|
pa_mix_info *m = streams + i;
|
core: Refactor code to multiply s16 by volume
move code to function pa_mult_s16_volume() in sample-util.h
use 64 bit integers on 64 bit platforms (it's faster)
on i5, 2.5GHz (64-bit)
Running suite(s): Mult-s16
32 bit mult: 1272300 usec (avg: 12723, min = 12533, max = 18749, stddev = 620.48).
64 bit mult: 852241 usec (avg: 8522.41, min = 8420, max = 9148, stddev = 109.388).
100%: Checks: 1, Failures: 0, Errors: 0
on Pentium D, 3.4GHz (32-bit)
Running suite(s): Mult-s16
32 bit mult: 2228504 usec (avg: 22285, min = 18775, max = 29648, stddev = 3865.59).
64 bit mult: 5546861 usec (avg: 55468.6, min = 55028, max = 64924, stddev = 978.981).
100%: Checks: 1, Failures: 0, Errors: 0
on TI DM3730, Cortex-A8, 800MHz (32-bit)
Running suite(s): Mult-s16
32 bit mult: 23708900 usec (avg: 237089, min = 191864, max = 557312, stddev = 77503.6).
64 bit mult: 22190039 usec (avg: 221900, min = 177978, max = 480469, stddev = 68520.5).
100%: Checks: 1, Failures: 0, Errors: 0
there is a test program called mult-s16-test which checks that the functions compute the
same results, and compares runtime
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
2013-02-13 17:27:05 +01:00
|
|
|
int32_t cv = m->linear[channel].i;
|
2013-02-13 17:26:59 +01:00
|
|
|
|
core: Refactor code to multiply s16 by volume
move code to function pa_mult_s16_volume() in sample-util.h
use 64 bit integers on 64 bit platforms (it's faster)
on i5, 2.5GHz (64-bit)
Running suite(s): Mult-s16
32 bit mult: 1272300 usec (avg: 12723, min = 12533, max = 18749, stddev = 620.48).
64 bit mult: 852241 usec (avg: 8522.41, min = 8420, max = 9148, stddev = 109.388).
100%: Checks: 1, Failures: 0, Errors: 0
on Pentium D, 3.4GHz (32-bit)
Running suite(s): Mult-s16
32 bit mult: 2228504 usec (avg: 22285, min = 18775, max = 29648, stddev = 3865.59).
64 bit mult: 5546861 usec (avg: 55468.6, min = 55028, max = 64924, stddev = 978.981).
100%: Checks: 1, Failures: 0, Errors: 0
on TI DM3730, Cortex-A8, 800MHz (32-bit)
Running suite(s): Mult-s16
32 bit mult: 23708900 usec (avg: 237089, min = 191864, max = 557312, stddev = 77503.6).
64 bit mult: 22190039 usec (avg: 221900, min = 177978, max = 480469, stddev = 68520.5).
100%: Checks: 1, Failures: 0, Errors: 0
there is a test program called mult-s16-test which checks that the functions compute the
same results, and compares runtime
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
2013-02-13 17:27:05 +01:00
|
|
|
if (PA_LIKELY(cv > 0))
|
|
|
|
|
sum += pa_mult_s16_volume(st_ulaw2linear16(*((uint8_t*) m->ptr)), cv);
|
2013-02-13 17:27:02 +01:00
|
|
|
m->ptr = (uint8_t*) m->ptr + 1;
|
2013-02-13 17:26:59 +01:00
|
|
|
}
|
|
|
|
|
|
2013-02-13 17:27:02 +01:00
|
|
|
sum = PA_CLAMP_UNLIKELY(sum, -0x8000, 0x7FFF);
|
2013-02-13 17:27:08 +01:00
|
|
|
*data = (uint8_t) st_14linear2ulaw((int16_t) sum >> 2);
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:02 +01:00
|
|
|
if (PA_UNLIKELY(++channel >= channels))
|
|
|
|
|
channel = 0;
|
|
|
|
|
}
|
|
|
|
|
}
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:08 +01:00
|
|
|
static void pa_mix_alaw_c(pa_mix_info streams[], unsigned nstreams, unsigned channels, uint8_t *data, unsigned length) {
|
2013-02-13 17:27:02 +01:00
|
|
|
unsigned channel = 0;
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:08 +01:00
|
|
|
length /= sizeof(uint8_t);
|
|
|
|
|
|
|
|
|
|
for (; length > 0; length--, data++) {
|
2013-02-13 17:27:02 +01:00
|
|
|
int32_t sum = 0;
|
|
|
|
|
unsigned i;
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:02 +01:00
|
|
|
for (i = 0; i < nstreams; i++) {
|
|
|
|
|
pa_mix_info *m = streams + i;
|
core: Refactor code to multiply s16 by volume
move code to function pa_mult_s16_volume() in sample-util.h
use 64 bit integers on 64 bit platforms (it's faster)
on i5, 2.5GHz (64-bit)
Running suite(s): Mult-s16
32 bit mult: 1272300 usec (avg: 12723, min = 12533, max = 18749, stddev = 620.48).
64 bit mult: 852241 usec (avg: 8522.41, min = 8420, max = 9148, stddev = 109.388).
100%: Checks: 1, Failures: 0, Errors: 0
on Pentium D, 3.4GHz (32-bit)
Running suite(s): Mult-s16
32 bit mult: 2228504 usec (avg: 22285, min = 18775, max = 29648, stddev = 3865.59).
64 bit mult: 5546861 usec (avg: 55468.6, min = 55028, max = 64924, stddev = 978.981).
100%: Checks: 1, Failures: 0, Errors: 0
on TI DM3730, Cortex-A8, 800MHz (32-bit)
Running suite(s): Mult-s16
32 bit mult: 23708900 usec (avg: 237089, min = 191864, max = 557312, stddev = 77503.6).
64 bit mult: 22190039 usec (avg: 221900, min = 177978, max = 480469, stddev = 68520.5).
100%: Checks: 1, Failures: 0, Errors: 0
there is a test program called mult-s16-test which checks that the functions compute the
same results, and compares runtime
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
2013-02-13 17:27:05 +01:00
|
|
|
int32_t cv = m->linear[channel].i;
|
2013-02-13 17:26:59 +01:00
|
|
|
|
core: Refactor code to multiply s16 by volume
move code to function pa_mult_s16_volume() in sample-util.h
use 64 bit integers on 64 bit platforms (it's faster)
on i5, 2.5GHz (64-bit)
Running suite(s): Mult-s16
32 bit mult: 1272300 usec (avg: 12723, min = 12533, max = 18749, stddev = 620.48).
64 bit mult: 852241 usec (avg: 8522.41, min = 8420, max = 9148, stddev = 109.388).
100%: Checks: 1, Failures: 0, Errors: 0
on Pentium D, 3.4GHz (32-bit)
Running suite(s): Mult-s16
32 bit mult: 2228504 usec (avg: 22285, min = 18775, max = 29648, stddev = 3865.59).
64 bit mult: 5546861 usec (avg: 55468.6, min = 55028, max = 64924, stddev = 978.981).
100%: Checks: 1, Failures: 0, Errors: 0
on TI DM3730, Cortex-A8, 800MHz (32-bit)
Running suite(s): Mult-s16
32 bit mult: 23708900 usec (avg: 237089, min = 191864, max = 557312, stddev = 77503.6).
64 bit mult: 22190039 usec (avg: 221900, min = 177978, max = 480469, stddev = 68520.5).
100%: Checks: 1, Failures: 0, Errors: 0
there is a test program called mult-s16-test which checks that the functions compute the
same results, and compares runtime
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
2013-02-13 17:27:05 +01:00
|
|
|
if (PA_LIKELY(cv > 0))
|
|
|
|
|
sum += pa_mult_s16_volume(st_alaw2linear16(*((uint8_t*) m->ptr)), cv);
|
2013-02-13 17:27:02 +01:00
|
|
|
m->ptr = (uint8_t*) m->ptr + 1;
|
2013-02-13 17:26:59 +01:00
|
|
|
}
|
|
|
|
|
|
2013-02-13 17:27:02 +01:00
|
|
|
sum = PA_CLAMP_UNLIKELY(sum, -0x8000, 0x7FFF);
|
2013-02-13 17:27:08 +01:00
|
|
|
*data = (uint8_t) st_13linear2alaw((int16_t) sum >> 3);
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:02 +01:00
|
|
|
if (PA_UNLIKELY(++channel >= channels))
|
|
|
|
|
channel = 0;
|
|
|
|
|
}
|
|
|
|
|
}
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:08 +01:00
|
|
|
static void pa_mix_float32ne_c(pa_mix_info streams[], unsigned nstreams, unsigned channels, float *data, unsigned length) {
|
2013-02-13 17:27:02 +01:00
|
|
|
unsigned channel = 0;
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:08 +01:00
|
|
|
length /= sizeof(float);
|
|
|
|
|
|
|
|
|
|
for (; length > 0; length--, data++) {
|
2013-02-13 17:27:02 +01:00
|
|
|
float sum = 0;
|
|
|
|
|
unsigned i;
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:02 +01:00
|
|
|
for (i = 0; i < nstreams; i++) {
|
|
|
|
|
pa_mix_info *m = streams + i;
|
|
|
|
|
float v, cv = m->linear[channel].f;
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:02 +01:00
|
|
|
if (PA_LIKELY(cv > 0)) {
|
|
|
|
|
v = *((float*) m->ptr);
|
|
|
|
|
v *= cv;
|
|
|
|
|
sum += v;
|
2013-02-13 17:26:59 +01:00
|
|
|
}
|
2013-02-13 17:27:02 +01:00
|
|
|
m->ptr = (uint8_t*) m->ptr + sizeof(float);
|
2013-02-13 17:26:59 +01:00
|
|
|
}
|
|
|
|
|
|
2013-02-13 17:27:08 +01:00
|
|
|
*data = sum;
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:02 +01:00
|
|
|
if (PA_UNLIKELY(++channel >= channels))
|
|
|
|
|
channel = 0;
|
|
|
|
|
}
|
|
|
|
|
}
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:08 +01:00
|
|
|
static void pa_mix_float32re_c(pa_mix_info streams[], unsigned nstreams, unsigned channels, float *data, unsigned length) {
|
2013-02-13 17:27:02 +01:00
|
|
|
unsigned channel = 0;
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:08 +01:00
|
|
|
length /= sizeof(float);
|
|
|
|
|
|
|
|
|
|
for (; length > 0; length--, data++) {
|
2013-02-13 17:27:02 +01:00
|
|
|
float sum = 0;
|
|
|
|
|
unsigned i;
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:02 +01:00
|
|
|
for (i = 0; i < nstreams; i++) {
|
|
|
|
|
pa_mix_info *m = streams + i;
|
endianmacros: Replace borked PA_FLOAT32_SWAP() with PA_READ_FLOAT32RE() / PA_WRITE_FLOAT32RE()
building PA with -O0 leads to test failure in mix-test on i386
issue reported by Felipe, see
http://lists.freedesktop.org/archives/pulseaudio-discuss/2014-August/021406.html
the problem is the value 0xbeffbd7f: when byte-swapped it becomes 0x7fbdffbe and according
to IEEE-754 represents a signalling NaN (starting with s111 1111 10, see http://en.wikipedia.org/wiki/NaN)
when this value is assigned to a floating point register, it becomes 0x7ffdffbe, representing
a quiet NaN (starting with s111 1111 11) -- a signalling NaN is turned into a quiet NaN!
so PA_FLOAT32_SWAP(PA_FLOAT32_SWAP(x)) != x for certain values, uhuh!
the following test code can be used; due to volatile, it will always demonstrate the issue;
without volatile, it depends on the optimization level (i386, 32-bit, gcc 4.9):
// snip
static inline float PA_FLOAT32_SWAP(float x) {
union {
float f;
uint32_t u;
} t;
t.f = x;
t.u = bswap_32(t.u);
return t.f;
}
int main() {
unsigned x = 0xbeffbd7f;
volatile float f = PA_FLOAT32_SWAP(*(float *)&x);
printf("%08x %08x %08x %f\n", 0xbeffbd7f, *(unsigned *)&f, bswap_32(*(unsigned *)&f), f);
}
// snip
the problem goes away with optimization when no temporary floating point registers are used
the proposed solution is to avoid passing swapped floating point data in a
float; this is done with new functions PA_READ_FLOAT32RE() and PA_WRITE_FLOAT32RE()
which use uint32_t to dereference a pointer and byte-swap the data, hence no temporary
float variable is used
also delete PA_FLOAT32_TO_LE()/_BE(), not used
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
Reported-by: Felipe Sateler <fsateler@debian.org>
2014-09-02 23:53:09 +02:00
|
|
|
float cv = m->linear[channel].f;
|
2013-02-13 17:26:59 +01:00
|
|
|
|
endianmacros: Replace borked PA_FLOAT32_SWAP() with PA_READ_FLOAT32RE() / PA_WRITE_FLOAT32RE()
building PA with -O0 leads to test failure in mix-test on i386
issue reported by Felipe, see
http://lists.freedesktop.org/archives/pulseaudio-discuss/2014-August/021406.html
the problem is the value 0xbeffbd7f: when byte-swapped it becomes 0x7fbdffbe and according
to IEEE-754 represents a signalling NaN (starting with s111 1111 10, see http://en.wikipedia.org/wiki/NaN)
when this value is assigned to a floating point register, it becomes 0x7ffdffbe, representing
a quiet NaN (starting with s111 1111 11) -- a signalling NaN is turned into a quiet NaN!
so PA_FLOAT32_SWAP(PA_FLOAT32_SWAP(x)) != x for certain values, uhuh!
the following test code can be used; due to volatile, it will always demonstrate the issue;
without volatile, it depends on the optimization level (i386, 32-bit, gcc 4.9):
// snip
static inline float PA_FLOAT32_SWAP(float x) {
union {
float f;
uint32_t u;
} t;
t.f = x;
t.u = bswap_32(t.u);
return t.f;
}
int main() {
unsigned x = 0xbeffbd7f;
volatile float f = PA_FLOAT32_SWAP(*(float *)&x);
printf("%08x %08x %08x %f\n", 0xbeffbd7f, *(unsigned *)&f, bswap_32(*(unsigned *)&f), f);
}
// snip
the problem goes away with optimization when no temporary floating point registers are used
the proposed solution is to avoid passing swapped floating point data in a
float; this is done with new functions PA_READ_FLOAT32RE() and PA_WRITE_FLOAT32RE()
which use uint32_t to dereference a pointer and byte-swap the data, hence no temporary
float variable is used
also delete PA_FLOAT32_TO_LE()/_BE(), not used
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
Reported-by: Felipe Sateler <fsateler@debian.org>
2014-09-02 23:53:09 +02:00
|
|
|
if (PA_LIKELY(cv > 0))
|
|
|
|
|
sum += PA_READ_FLOAT32RE(m->ptr) * cv;
|
2013-02-13 17:27:02 +01:00
|
|
|
m->ptr = (uint8_t*) m->ptr + sizeof(float);
|
2013-02-13 17:26:59 +01:00
|
|
|
}
|
|
|
|
|
|
endianmacros: Replace borked PA_FLOAT32_SWAP() with PA_READ_FLOAT32RE() / PA_WRITE_FLOAT32RE()
building PA with -O0 leads to test failure in mix-test on i386
issue reported by Felipe, see
http://lists.freedesktop.org/archives/pulseaudio-discuss/2014-August/021406.html
the problem is the value 0xbeffbd7f: when byte-swapped it becomes 0x7fbdffbe and according
to IEEE-754 represents a signalling NaN (starting with s111 1111 10, see http://en.wikipedia.org/wiki/NaN)
when this value is assigned to a floating point register, it becomes 0x7ffdffbe, representing
a quiet NaN (starting with s111 1111 11) -- a signalling NaN is turned into a quiet NaN!
so PA_FLOAT32_SWAP(PA_FLOAT32_SWAP(x)) != x for certain values, uhuh!
the following test code can be used; due to volatile, it will always demonstrate the issue;
without volatile, it depends on the optimization level (i386, 32-bit, gcc 4.9):
// snip
static inline float PA_FLOAT32_SWAP(float x) {
union {
float f;
uint32_t u;
} t;
t.f = x;
t.u = bswap_32(t.u);
return t.f;
}
int main() {
unsigned x = 0xbeffbd7f;
volatile float f = PA_FLOAT32_SWAP(*(float *)&x);
printf("%08x %08x %08x %f\n", 0xbeffbd7f, *(unsigned *)&f, bswap_32(*(unsigned *)&f), f);
}
// snip
the problem goes away with optimization when no temporary floating point registers are used
the proposed solution is to avoid passing swapped floating point data in a
float; this is done with new functions PA_READ_FLOAT32RE() and PA_WRITE_FLOAT32RE()
which use uint32_t to dereference a pointer and byte-swap the data, hence no temporary
float variable is used
also delete PA_FLOAT32_TO_LE()/_BE(), not used
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
Reported-by: Felipe Sateler <fsateler@debian.org>
2014-09-02 23:53:09 +02:00
|
|
|
PA_WRITE_FLOAT32RE(data, sum);
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:02 +01:00
|
|
|
if (PA_UNLIKELY(++channel >= channels))
|
|
|
|
|
channel = 0;
|
|
|
|
|
}
|
|
|
|
|
}
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:02 +01:00
|
|
|
static pa_do_mix_func_t do_mix_table[] = {
|
|
|
|
|
[PA_SAMPLE_U8] = (pa_do_mix_func_t) pa_mix_u8_c,
|
|
|
|
|
[PA_SAMPLE_ALAW] = (pa_do_mix_func_t) pa_mix_alaw_c,
|
|
|
|
|
[PA_SAMPLE_ULAW] = (pa_do_mix_func_t) pa_mix_ulaw_c,
|
|
|
|
|
[PA_SAMPLE_S16NE] = (pa_do_mix_func_t) pa_mix_s16ne_c,
|
|
|
|
|
[PA_SAMPLE_S16RE] = (pa_do_mix_func_t) pa_mix_s16re_c,
|
|
|
|
|
[PA_SAMPLE_FLOAT32NE] = (pa_do_mix_func_t) pa_mix_float32ne_c,
|
|
|
|
|
[PA_SAMPLE_FLOAT32RE] = (pa_do_mix_func_t) pa_mix_float32re_c,
|
|
|
|
|
[PA_SAMPLE_S32NE] = (pa_do_mix_func_t) pa_mix_s32ne_c,
|
|
|
|
|
[PA_SAMPLE_S32RE] = (pa_do_mix_func_t) pa_mix_s32re_c,
|
|
|
|
|
[PA_SAMPLE_S24NE] = (pa_do_mix_func_t) pa_mix_s24ne_c,
|
|
|
|
|
[PA_SAMPLE_S24RE] = (pa_do_mix_func_t) pa_mix_s24re_c,
|
|
|
|
|
[PA_SAMPLE_S24_32NE] = (pa_do_mix_func_t) pa_mix_s24_32ne_c,
|
|
|
|
|
[PA_SAMPLE_S24_32RE] = (pa_do_mix_func_t) pa_mix_s24_32re_c
|
|
|
|
|
};
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2014-04-18 09:59:32 +02:00
|
|
|
void pa_mix_func_init(const pa_cpu_info *cpu_info) {
|
|
|
|
|
if (cpu_info->force_generic_code)
|
|
|
|
|
do_mix_table[PA_SAMPLE_S16NE] = (pa_do_mix_func_t) pa_mix_generic_s16ne;
|
|
|
|
|
else
|
|
|
|
|
do_mix_table[PA_SAMPLE_S16NE] = (pa_do_mix_func_t) pa_mix_s16ne_c;
|
|
|
|
|
}
|
|
|
|
|
|
2013-02-13 17:27:02 +01:00
|
|
|
size_t pa_mix(
|
|
|
|
|
pa_mix_info streams[],
|
|
|
|
|
unsigned nstreams,
|
|
|
|
|
void *data,
|
|
|
|
|
size_t length,
|
|
|
|
|
const pa_sample_spec *spec,
|
|
|
|
|
const pa_cvolume *volume,
|
2013-06-27 19:28:09 +02:00
|
|
|
bool mute) {
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:02 +01:00
|
|
|
pa_cvolume full_volume;
|
|
|
|
|
unsigned k;
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:02 +01:00
|
|
|
pa_assert(streams);
|
|
|
|
|
pa_assert(data);
|
|
|
|
|
pa_assert(length);
|
|
|
|
|
pa_assert(spec);
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:02 +01:00
|
|
|
if (!volume)
|
|
|
|
|
volume = pa_cvolume_reset(&full_volume, spec->channels);
|
2013-02-13 17:26:59 +01:00
|
|
|
|
2013-02-13 17:27:02 +01:00
|
|
|
if (mute || pa_cvolume_is_muted(volume) || nstreams <= 0) {
|
|
|
|
|
pa_silence_memory(data, length, spec);
|
|
|
|
|
return length;
|
2013-02-13 17:26:59 +01:00
|
|
|
}
|
|
|
|
|
|
2013-02-13 17:27:04 +01:00
|
|
|
for (k = 0; k < nstreams; k++) {
|
2013-02-13 17:27:02 +01:00
|
|
|
streams[k].ptr = pa_memblock_acquire_chunk(&streams[k].chunk);
|
2013-02-13 17:27:04 +01:00
|
|
|
if (length > streams[k].chunk.length)
|
|
|
|
|
length = streams[k].chunk.length;
|
|
|
|
|
}
|
2013-02-13 17:27:02 +01:00
|
|
|
|
|
|
|
|
calc_stream_volumes_table[spec->format](streams, nstreams, volume, spec);
|
2013-02-13 17:27:08 +01:00
|
|
|
do_mix_table[spec->format](streams, nstreams, spec->channels, data, length);
|
2013-02-13 17:27:02 +01:00
|
|
|
|
2013-02-13 17:26:59 +01:00
|
|
|
for (k = 0; k < nstreams; k++)
|
|
|
|
|
pa_memblock_release(streams[k].chunk.memblock);
|
|
|
|
|
|
|
|
|
|
return length;
|
|
|
|
|
}
|
|
|
|
|
|
2013-02-13 17:27:03 +01:00
|
|
|
pa_do_mix_func_t pa_get_mix_func(pa_sample_format_t f) {
|
2013-12-04 09:50:09 +02:00
|
|
|
pa_assert(pa_sample_format_valid(f));
|
2013-02-13 17:27:03 +01:00
|
|
|
|
|
|
|
|
return do_mix_table[f];
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
void pa_set_mix_func(pa_sample_format_t f, pa_do_mix_func_t func) {
|
2013-12-04 09:50:09 +02:00
|
|
|
pa_assert(pa_sample_format_valid(f));
|
2013-02-13 17:27:03 +01:00
|
|
|
|
|
|
|
|
do_mix_table[f] = func;
|
|
|
|
|
}
|
|
|
|
|
|
2013-02-13 17:26:59 +01:00
|
|
|
typedef union {
|
|
|
|
|
float f;
|
|
|
|
|
uint32_t i;
|
|
|
|
|
} volume_val;
|
|
|
|
|
|
|
|
|
|
typedef void (*pa_calc_volume_func_t) (void *volumes, const pa_cvolume *volume);
|
|
|
|
|
|
|
|
|
|
static const pa_calc_volume_func_t calc_volume_table[] = {
|
|
|
|
|
[PA_SAMPLE_U8] = (pa_calc_volume_func_t) calc_linear_integer_volume,
|
|
|
|
|
[PA_SAMPLE_ALAW] = (pa_calc_volume_func_t) calc_linear_integer_volume,
|
|
|
|
|
[PA_SAMPLE_ULAW] = (pa_calc_volume_func_t) calc_linear_integer_volume,
|
|
|
|
|
[PA_SAMPLE_S16LE] = (pa_calc_volume_func_t) calc_linear_integer_volume,
|
|
|
|
|
[PA_SAMPLE_S16BE] = (pa_calc_volume_func_t) calc_linear_integer_volume,
|
|
|
|
|
[PA_SAMPLE_FLOAT32LE] = (pa_calc_volume_func_t) calc_linear_float_volume,
|
|
|
|
|
[PA_SAMPLE_FLOAT32BE] = (pa_calc_volume_func_t) calc_linear_float_volume,
|
|
|
|
|
[PA_SAMPLE_S32LE] = (pa_calc_volume_func_t) calc_linear_integer_volume,
|
|
|
|
|
[PA_SAMPLE_S32BE] = (pa_calc_volume_func_t) calc_linear_integer_volume,
|
|
|
|
|
[PA_SAMPLE_S24LE] = (pa_calc_volume_func_t) calc_linear_integer_volume,
|
|
|
|
|
[PA_SAMPLE_S24BE] = (pa_calc_volume_func_t) calc_linear_integer_volume,
|
|
|
|
|
[PA_SAMPLE_S24_32LE] = (pa_calc_volume_func_t) calc_linear_integer_volume,
|
|
|
|
|
[PA_SAMPLE_S24_32BE] = (pa_calc_volume_func_t) calc_linear_integer_volume
|
|
|
|
|
};
|
|
|
|
|
|
|
|
|
|
void pa_volume_memchunk(
|
|
|
|
|
pa_memchunk*c,
|
|
|
|
|
const pa_sample_spec *spec,
|
|
|
|
|
const pa_cvolume *volume) {
|
|
|
|
|
|
|
|
|
|
void *ptr;
|
|
|
|
|
volume_val linear[PA_CHANNELS_MAX + VOLUME_PADDING];
|
|
|
|
|
pa_do_volume_func_t do_volume;
|
|
|
|
|
|
|
|
|
|
pa_assert(c);
|
|
|
|
|
pa_assert(spec);
|
|
|
|
|
pa_assert(pa_sample_spec_valid(spec));
|
|
|
|
|
pa_assert(pa_frame_aligned(c->length, spec));
|
|
|
|
|
pa_assert(volume);
|
|
|
|
|
|
|
|
|
|
if (pa_memblock_is_silence(c->memblock))
|
|
|
|
|
return;
|
|
|
|
|
|
|
|
|
|
if (pa_cvolume_channels_equal_to(volume, PA_VOLUME_NORM))
|
|
|
|
|
return;
|
|
|
|
|
|
|
|
|
|
if (pa_cvolume_channels_equal_to(volume, PA_VOLUME_MUTED)) {
|
|
|
|
|
pa_silence_memchunk(c, spec);
|
|
|
|
|
return;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
do_volume = pa_get_volume_func(spec->format);
|
|
|
|
|
pa_assert(do_volume);
|
|
|
|
|
|
|
|
|
|
calc_volume_table[spec->format] ((void *)linear, volume);
|
|
|
|
|
|
|
|
|
|
ptr = pa_memblock_acquire_chunk(c);
|
|
|
|
|
|
|
|
|
|
do_volume(ptr, (void *)linear, spec->channels, c->length);
|
|
|
|
|
|
|
|
|
|
pa_memblock_release(c->memblock);
|
|
|
|
|
}
|