summaryrefslogtreecommitdiff
path: root/src
diff options
context:
space:
mode:
Diffstat (limited to 'src')
-rw-r--r--src/audio_out/audio_alsa_out.c193
1 files changed, 83 insertions, 110 deletions
diff --git a/src/audio_out/audio_alsa_out.c b/src/audio_out/audio_alsa_out.c
index 23632c96d..e76a0f730 100644
--- a/src/audio_out/audio_alsa_out.c
+++ b/src/audio_out/audio_alsa_out.c
@@ -24,7 +24,7 @@
* (c) 2001 James Courtier-Dutton <James@superbug.demon.co.uk>
*
*
- * $Id: audio_alsa_out.c,v 1.12 2001/07/18 21:38:16 f1rmb Exp $
+ * $Id: audio_alsa_out.c,v 1.13 2001/08/01 18:17:40 joachim_koenig Exp $
*/
#ifdef HAVE_CONFIG_H
@@ -98,15 +98,14 @@ typedef struct alsa_functions_s {
* For this plugin, we will use frames instead of bytes for everything.
* The term sample is also equil to frames
*/
- snd_pcm_sframes_t frames_in_buffer; /* number of frames writen to audio hardware */
- uint32_t last_vpts; /* vpts at which last written package ends */
+ uint32_t bytes_in_buffer; /* number of bytes writen to audio hardware */
- uint32_t sync_vpts; /* this syncpoint is used as a starting point */
- snd_pcm_sframes_t sync_frames_in_buffer; /* for vpts <-> samplecount assoc */
int audio_step; /* pts per 32 768 frames (frame = #bytes/2(16 bits)/channels) */
/* frames = pts * rate / pts_per_second */
/* pts = frame * pts_per_second / rate */
+ int32_t bytes_per_kpts; /* bytes per 1024/90000 sec */
+
snd_pcm_sframes_t pts_per_second; /* pts per second */
@@ -120,6 +119,7 @@ typedef struct alsa_functions_s {
} alsa_functions_t;
void write_pause_burst(alsa_functions_t *,int );
+void write_burst(alsa_functions_t *,u_char *, size_t );
static snd_output_t *jcd_out;
/*
@@ -160,11 +160,8 @@ static int ao_open(ao_functions_t *this_gen, uint32_t bits, uint32_t rate, int m
snd_pcm_close (this->audio_fd);
}
this->input_sample_rate = rate;
- this->frames_in_buffer = 0;
- this->last_vpts = 0;
+ this->bytes_in_buffer = 0;
this->output_rate_correction = 0;
- this->sync_vpts = 0;
- this->sync_frames_in_buffer = 0;
this->audio_started = 0;
this->open_mode = mode;
this->last_audio_vpts = 0;
@@ -297,9 +294,10 @@ static int ao_open(ao_functions_t *this_gen, uint32_t bits, uint32_t rate, int m
this->output_sample_rate = this->input_sample_rate;
this->sample_rate_factor = (double) this->output_sample_rate / (double) this->input_sample_rate;
- this->audio_step = (double) 90000 * (double) 32768
+ this->audio_step = (uint32_t) 90000 * (uint32_t) 32768
/ this->input_sample_rate;
- this->pts_per_second = 90000;
+ this->bytes_per_kpts = this->output_sample_rate * this->num_channels * 2 * 1024 / 90000;
+
this->metronom->set_audio_rate(this->metronom, this->audio_step);
/*
* audio buffer size handling
@@ -307,6 +305,8 @@ static int ao_open(ao_functions_t *this_gen, uint32_t bits, uint32_t rate, int m
/* Copy current parameters into swparams */
snd_pcm_sw_params_current(this->audio_fd, swparams);
tmp=snd_pcm_sw_params_set_xfer_align(this->audio_fd, swparams, 4);
+ tmp=snd_pcm_sw_params_set_avail_min(this->audio_fd, swparams, 1);
+ tmp=snd_pcm_sw_params_set_start_threshold(this->audio_fd, swparams, 1);
/* Install swparams into current parameters */
snd_pcm_sw_params(this->audio_fd, swparams);
@@ -315,7 +315,7 @@ static int ao_open(ao_functions_t *this_gen, uint32_t bits, uint32_t rate, int m
snd_pcm_sw_params_dump(swparams, jcd_out);
- write_pause_burst(this,0);
+ // write_pause_burst(this,0);
return 1;
@@ -325,13 +325,13 @@ __close:
return -1;
}
-static uint32_t ao_get_current_vpts (alsa_functions_t *this)
+static uint32_t ao_get_current_pos (alsa_functions_t *this)
{
snd_pcm_sframes_t pos ;
snd_pcm_status_t *pcm_stat;
snd_pcm_sframes_t delay;
int err;
- uint32_t vpts ;
+
snd_pcm_status_alloca(&pcm_stat);
snd_pcm_status(this->audio_fd, pcm_stat);
/* Dump ALSA info to stderr */
@@ -341,49 +341,51 @@ static uint32_t ao_get_current_vpts (alsa_functions_t *this)
if(err < 0) {
//Hide error report
error("snd_pcm_delay() failed");
- return 0;
+ return this->bytes_in_buffer;
}
- /* Correction factor, bigger -, sound earlier
- * bigger +, sound later
- * current setting for SB Live
- */
- pos = this->frames_in_buffer - delay + 1500;
+ pos = delay * 2 * this->num_channels;
} else {
- pos=0;
+ pos=this->bytes_in_buffer;
}
- vpts = ((double)pos * (double)this->pts_per_second / (double)this->input_sample_rate);
- return vpts;
+ return pos;
}
-static void ao_fill_gap (alsa_functions_t *this, uint32_t pts_len)
-{
- snd_pcm_sframes_t res;
- int num_frames;
+static void ao_fill_gap (alsa_functions_t *this, uint32_t pts_len) {
+
+ int num_bytes ;
if (pts_len > MAX_GAP)
pts_len = MAX_GAP;
+ num_bytes = pts_len * this->bytes_per_kpts / 1024;
+ num_bytes = (num_bytes / (2*this->num_channels)) * (2*this->num_channels);
- num_frames = (double)pts_len * (double)this->input_sample_rate / (double)this->pts_per_second;
- num_frames = (num_frames / 4) * 4;
- this->frames_in_buffer += num_frames;
- while (num_frames>0) {
- if (num_frames>2048) {
- res=snd_pcm_writei(this->audio_fd, this->zero_space, 2048 );
- num_frames -= 2048;
+ if(this->open_mode == AO_CAP_MODE_AC3) {
+ write_pause_burst(this,0);
+printf("audio_alsa_out: SPDIF write pause\n");
+ return;
+ }
+
+ printf ("audio_alsa_out: inserting %d 0-bytes to fill a gap of %d pts\n",num_bytes, pts_len);
+
+ this->bytes_in_buffer += num_bytes;
+
+ while (num_bytes>0) {
+ if (num_bytes>8192) {
+ write_burst(this, (unsigned char *)this->zero_space, 8192 / (2 * this->num_channels));
+ num_bytes -= 8192;
} else {
- res=snd_pcm_writei(this->audio_fd, this->zero_space, num_frames );
- num_frames = 0;
+ write_burst(this, (unsigned char *)this->zero_space, num_bytes / (2 * this->num_channels));
+ num_bytes = 0;
}
}
- this->last_vpts += pts_len;
}
void xrun(alsa_functions_t *this)
{
- snd_pcm_status_t *status;
- int res;
+ snd_pcm_status_t *status;
+ int res;
snd_pcm_status_alloca(&status);
if ((res = snd_pcm_status(this->audio_fd, status))<0) {
@@ -395,10 +397,9 @@ void xrun(alsa_functions_t *this)
gettimeofday(&now, 0);
snd_pcm_status_get_trigger_tstamp(status, &tstamp);
timersub(&now, &tstamp, &diff);
- fprintf(stderr, "xrun!!! (at least %.3f ms long)\n",
-diff.tv_sec * 1000 + diff.tv_usec / 1000.0);
+ fprintf(stderr, "xrun!!! (at least %.3f ms long)\n", diff.tv_sec * 1000 + diff.tv_usec / 1000.0);
- if ((res = snd_pcm_prepare(this->audio_fd))<0) {
+ if ((res = snd_pcm_prepare(this->audio_fd))<0) {
printf("xrun: prepare error: %s", snd_strerror(res));
return;
}
@@ -407,9 +408,10 @@ diff.tv_sec * 1000 + diff.tv_usec / 1000.0);
}
void write_burst(alsa_functions_t *this,u_char *data, size_t count)
- {
+{
ssize_t r;
- while(count > 0) {
+
+ while( count > 0) {
r = snd_pcm_writei(this->audio_fd, data, count);
if (r == -EAGAIN || (r >=0 && r < count)) {
snd_pcm_wait(this->audio_fd, 1000);
@@ -418,9 +420,10 @@ void write_burst(alsa_functions_t *this,u_char *data, size_t count)
}
if (r > 0) {
count -= r;
- data += r * 4;
+ data += r * 2 * this->num_channels;
}
}
+
}
void write_pause_burst(alsa_functions_t *this,int error)
@@ -458,9 +461,9 @@ static int ao_write_audio_data(ao_functions_t *this_gen,
alsa_functions_t *this = (alsa_functions_t *) this_gen;
uint32_t vpts,
- audio_vpts,
- master_vpts;
- int32_t diff, gap;
+ pos,
+ buffer_vpts;
+ int32_t gap;
int bDropPackage;
uint16_t sample_buffer[8192];
int num_output_samples;
@@ -468,10 +471,15 @@ static int ao_write_audio_data(ao_functions_t *this_gen,
if (this->audio_fd == NULL) {
error("Nothing open");
- return;
+ return 1;
}
- vpts = this->metronom->got_audio_samples (this->metronom, pts_, num_samples);
+ if(this->open_mode == AO_CAP_MODE_AC3)
+ num_samples = 1536; /* FIXME */
+
+ vpts = this->metronom->got_audio_samples (this->metronom, pts_, num_samples);
+
+
if (vpts<this->last_audio_vpts) {
/* reject this */
@@ -479,15 +487,24 @@ static int ao_write_audio_data(ao_functions_t *this_gen,
return 1;
}
this->last_audio_vpts = vpts;
+ bDropPackage = 0;
+
+ pos = ao_get_current_pos (this) ;
+
+// if (pos>this->bytes_in_buffer) /* buffer ran dry */
+// this->bytes_in_buffer = pos;
+
+//printf("samples %d pos %d bytes_in_buffer %d ",num_samples,pos,this->bytes_in_buffer);
+ buffer_vpts = this->metronom->get_current_time (this->metronom);
+//printf("buffer_vpts %d vpts %d ",buffer_vpts,vpts);
+ buffer_vpts += (this->bytes_in_buffer - pos) * 1024 / this->bytes_per_kpts;
+ this->bytes_in_buffer -= (this->bytes_in_buffer - pos);
+
+ gap = vpts - buffer_vpts;
+
+//printf("gap %d \n",gap);
- /*
- * check if these samples "fit" in the audio output buffer
- * or do we have an audio "gap" here?
- */
- gap = vpts - this->last_vpts ;
- bDropPackage = 0;
-#if 0
if (gap>GAP_TOLERANCE) {
ao_fill_gap (this, gap);
@@ -498,50 +515,21 @@ static int ao_write_audio_data(ao_functions_t *this_gen,
} else if (gap<-GAP_TOLERANCE) {
bDropPackage = 1;
+//printf("audio_alsa_out: drop package\n");
}
-#endif
- /*
- * sync on master clock
- */
- audio_vpts = ao_get_current_vpts (this) ;
- master_vpts = this->metronom->get_current_time (this->metronom);
- diff = audio_vpts - master_vpts;
- /*
- * method 1 : resampling
- */
- if (abs(diff)>5000) {
- if (diff>5000) {
- error("Fill Gap");
- if ((this->open_mode & AO_CAP_MODE_AC3) == 0)
- ao_fill_gap (this,diff);
- } else if (diff<-5000) {
- error("Drop");
- bDropPackage = 1;
- }
- } else if (abs(diff)>1000) {
- this->output_rate_correction = diff/10 ;
- error("diff = %d => rate correction : %d", diff, this->output_rate_correction);
- if ( this->output_rate_correction < -500)
- this->output_rate_correction = -500;
- else if ( this->output_rate_correction > 500)
- this->output_rate_correction = 500;
- }
- /*
- * method 2: adjust master clock
- */
- if (abs(diff)>MAX_MASTER_CLOCK_DIV) {
- error ("master clock adjust time %d -> %d (diff: %d)", master_vpts, audio_vpts, diff);
- this->metronom->adjust_clock (this->metronom, audio_vpts);
- }
+
/*
* resample and output samples
*/
+
+ if(this->open_mode == AO_CAP_MODE_AC3) bDropPackage=0;
+
+
if (!bDropPackage) {
if ((this->open_mode & AO_CAP_MODE_AC3) == 0) {
- /* Multiples of xfer_align eg:- 4 */
- num_output_samples = ((num_samples * (this->output_sample_rate + this->output_rate_correction) / this->input_sample_rate / 4) * 4)+4;
+ num_output_samples = num_samples * (this->output_sample_rate) / this->input_sample_rate;
audio_out_resample_stereo (output_samples, num_samples,
sample_buffer, num_output_samples);
do {
@@ -566,29 +554,14 @@ static int ao_write_audio_data(ao_functions_t *this_gen,
* A Frame is one sample for all channels, so here a Stereo 16 bits frame is 4 bytes.
*/
-// res=snd_pcm_writei(this->audio_fd, sample_buffer, num_output_samples);
- write_burst(this, sample_buffer, num_output_samples);
- res = num_output_samples;
-
- if(res != num_output_samples) error("BUFFER MAYBE FULL!!!!!!!!!!!!");
- if (res < 0)
- error("writei returned error: %s", snd_strerror(res));
- /*
- * remember vpts
- */
- this->sync_vpts = vpts;
- this->sync_frames_in_buffer = this->frames_in_buffer;
+ write_burst(this, (unsigned char *)sample_buffer, num_output_samples);
/*
* step values
*/
- this->frames_in_buffer += num_samples ;
+ this->bytes_in_buffer += num_samples * 2 * this->num_channels ;
this->audio_started = 1;
- } else {
- this->sync_vpts = vpts;
}
- this->last_vpts = vpts + num_samples * this->pts_per_second / this->input_sample_rate ;
-
return 1;
}
generated by cgit v1.2.3 (git 2.25.1) at 2025-04-22 19:00:34 +0000