OrgACRN
/
sof
mirror of https://github.com/thesofproject/sof.git
1
0
Fork 0
Code Issues Projects Releases Wiki Activity

Audio: EQFIR: Optimize source and sink buffers use in generic C version 

This patch replaces the audio stream read/write frag based
access to source and sink by block processing based on
audio_stream_bytes_without_wrap() bytes count.

In a test with forced generic C for xtensa build processing load in
original was 324 MCPS, in optimized 309 MCPS, saving is 15 MCPS. The
load is nearly same for all formats s16/s24/s32. The base load was
very high in test due to a very long used FIR filter. The MCPS saving
should be the same for all stereo 48k streams.

Signed-off-by: Seppo Ingalsuo <seppo.ingalsuo@linux.intel.com>
This commit is contained in:
Seppo Ingalsuo 2022-02-09 18:09:44 +02:00 committed by Liam Girdwood
parent d0057a7def
commit 2b6001e065
2 changed files with 72 additions and 41 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

109
src/audio/eq_fir/eq_fir_generic.c
View File

@ -21,23 +21,32 @@ void eq_fir_s16(struct fir_state_32x16 fir[], const struct audio_stream *source,
struct audio_stream *sink, int frames, int nch)
{
struct fir_state_32x16 *filter;
int16_t *x;
int16_t *y;
int32_t z;
int idx;
int ch;
int i;
int16_t *x0, *y0;
int16_t *x = source->r_ptr;
int16_t *y = sink->w_ptr;
int nmax, n, i, j;
int remaining_samples = frames * nch;
for (ch = 0; ch < nch; ch++) {
filter = &fir[ch];
idx = ch;
for (i = 0; i < frames; i++) {
x = audio_stream_read_frag_s16(source, idx);
y = audio_stream_write_frag_s16(sink, idx);
z = fir_32x16(filter, *x << 16);
*y = sat_int16(Q_SHIFT_RND(z, 31, 15));
idx += nch;
while (remaining_samples) {
nmax = EQ_FIR_BYTES_TO_S16_SAMPLES(audio_stream_bytes_without_wrap(source, x));
n = MIN(remaining_samples, nmax);
nmax = EQ_FIR_BYTES_TO_S16_SAMPLES(audio_stream_bytes_without_wrap(sink, y));
n = MIN(n, nmax);
for (j = 0; j < nch; j++) {
x0 = x + j;
y0 = y + j;
filter = &fir[j];
for (i = 0; i < n; i += nch) {
z = fir_32x16(filter, *x0 << 16);
*y0 = sat_int16(Q_SHIFT_RND(z, 31, 15));
x0 += nch;
y0 += nch;
}
}
remaining_samples -= n;
x = audio_stream_wrap(source, x + n);
y = audio_stream_wrap(sink, y + n);
}
}
#endif /* CONFIG_FORMAT_S16LE */
@ -47,23 +56,32 @@ void eq_fir_s24(struct fir_state_32x16 fir[], const struct audio_stream *source,
struct audio_stream *sink, int frames, int nch)
{
struct fir_state_32x16 *filter;
int32_t *x;
int32_t *y;
int32_t z;
int idx;
int ch;
int i;
int32_t *x0, *y0;
int32_t *x = source->r_ptr;
int32_t *y = sink->w_ptr;
int nmax, n, i, j;
int remaining_samples = frames * nch;
for (ch = 0; ch < nch; ch++) {
filter = &fir[ch];
idx = ch;
for (i = 0; i < frames; i++) {
x = audio_stream_read_frag_s32(source, idx);
y = audio_stream_write_frag_s32(sink, idx);
z = fir_32x16(filter, *x << 8);
*y = sat_int24(Q_SHIFT_RND(z, 31, 23));
idx += nch;
while (remaining_samples) {
nmax = EQ_FIR_BYTES_TO_S32_SAMPLES(audio_stream_bytes_without_wrap(source, x));
n = MIN(remaining_samples, nmax);
nmax = EQ_FIR_BYTES_TO_S32_SAMPLES(audio_stream_bytes_without_wrap(sink, y));
n = MIN(n, nmax);
for (j = 0; j < nch; j++) {
x0 = x + j;
y0 = y + j;
filter = &fir[j];
for (i = 0; i < n; i += nch) {
z = fir_32x16(filter, *x0 << 8);
*y0 = sat_int24(Q_SHIFT_RND(z, 31, 23));
x0 += nch;
y0 += nch;
}
}
remaining_samples -= n;
x = audio_stream_wrap(source, x + n);
y = audio_stream_wrap(sink, y + n);
}
}
#endif /* CONFIG_FORMAT_S24LE */
@ -73,21 +91,30 @@ void eq_fir_s32(struct fir_state_32x16 fir[], const struct audio_stream *source,
struct audio_stream *sink, int frames, int nch)
{
struct fir_state_32x16 *filter;
int32_t *x;
int32_t *y;
int idx;
int ch;
int i;
int32_t *x0, *y0;
int32_t *x = source->r_ptr;
int32_t *y = sink->w_ptr;
int nmax, n, i, j;
int remaining_samples = frames * nch;
for (ch = 0; ch < nch; ch++) {
filter = &fir[ch];
idx = ch;
for (i = 0; i < frames; i++) {
x = audio_stream_read_frag_s32(source, idx);
y = audio_stream_write_frag_s32(sink, idx);
*y = fir_32x16(filter, *x);
idx += nch;
while (remaining_samples) {
nmax = EQ_FIR_BYTES_TO_S32_SAMPLES(audio_stream_bytes_without_wrap(source, x));
n = MIN(remaining_samples, nmax);
nmax = EQ_FIR_BYTES_TO_S32_SAMPLES(audio_stream_bytes_without_wrap(sink, y));
n = MIN(n, nmax);
for (j = 0; j < nch; j++) {
x0 = x + j;
y0 = y + j;
filter = &fir[j];
for (i = 0; i < n; i += nch) {
*y0 = fir_32x16(filter, *x0);
x0 += nch;
y0 += nch;
}
}
remaining_samples -= n;
x = audio_stream_wrap(source, x + n);
y = audio_stream_wrap(sink, y + n);
}
}
#endif /* CONFIG_FORMAT_S32LE */

4
src/include/sof/audio/eq_fir/eq_fir.h
View File

@ -23,6 +23,10 @@
#include <user/fir.h>
#include <stdint.h>
/** \brief Macros to convert without division bytes count to samples count */
#define EQ_FIR_BYTES_TO_S16_SAMPLES(b) ((b) >> 1)
#define EQ_FIR_BYTES_TO_S32_SAMPLES(b) ((b) >> 2)
#if CONFIG_FORMAT_S16LE
void eq_fir_s16(struct fir_state_32x16 *fir, const struct audio_stream *source,
struct audio_stream *sink, int frames, int nch);