nils/shairport-sync
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Move to using pthread_cancel in place of pthread_kill -- due to a problem in CYGWIN, but the move makes the code simpler and more rugged. Tidy up the code a bit and remove a clobbering warning.

Browse Source
This commit is contained in:
Mike Brady
2018-05-17 10:29:59 +01:00
parent 7d0eb37958
commit 8a5a29e2c2
4 changed files with 67 additions and 64 deletions
Download Patch File Download Diff File Expand all files Collapse all files
audio_ao.c
+1 -1
View File
@@ -121,7 +121,7 @@ static void deinit(void) {
static void start(__attribute__((unused)) int sample_rate,
__attribute__((unused)) int sample_format) {}
static void play(void * buf, int samples) { ao_play(dev, buf, samples * 4); }
static void play(void *buf, int samples) { ao_play(dev, buf, samples * 4); }
static void stop(void) {}
mdns.c
+8 -8
View File
@@ -106,14 +106,14 @@ void mdns_unregister(void) {
void *mdns_dacp_monitor(char *dacp_id) {
void *reply = NULL;
if ((dacp_id!=NULL) && (*dacp_id!='\0')) {
if ((config.mdns) && (config.mdns->mdns_dacp_monitor)) {
reply = config.mdns->mdns_dacp_monitor(dacp_id);
if (reply == NULL) {
debug(1, "Error starting a DACP monitor.");
}
} else
debug(3, "Can't start a DACP monitor -- none registered.");
if ((dacp_id != NULL) && (*dacp_id != '\0')) {
if ((config.mdns) && (config.mdns->mdns_dacp_monitor)) {
reply = config.mdns->mdns_dacp_monitor(dacp_id);
if (reply == NULL) {
debug(1, "Error starting a DACP monitor.");
}
} else
debug(3, "Can't start a DACP monitor -- none registered.");
}
return reply;
}
player.c
+3 -2
View File
@@ -545,8 +545,9 @@ void player_put_packet(seq_t seqno, uint32_t actual_timestamp, int64_t timestamp
if (config.disable_resend_requests == 0) {
rtp_request_resend(next, 1, conn);
if ((back_step + k + resend_interval) >= sd)
debug(2, "Last-ditch (#%d) resend request for packet %u in range %u to %u. Looking back %d packets.", j, next,
conn->ab_read, conn->ab_write,back_step + k);
debug(2, "Last-ditch (#%d) resend request for packet %u in range %u to %u. "
"Looking back %d packets.",
j, next, conn->ab_read, conn->ab_write, back_step + k);
conn->resend_requests++;
}
}
rtp.c
+55 -53
View File
@@ -177,7 +177,7 @@ void *rtp_audio_receiver(void *arg) {
debug(3, "Audio receiver -- Server RTP thread interrupted. terminating.");
close(conn->audio_socket);
*/
debug(1, "Audio receiver thread \"normal\" exit -- this can't happen. Hah!");
pthread_cleanup_pop(0); // don't execute anything here.
debug(2, "Audio receiver thread exit.");
@@ -211,50 +211,50 @@ void *rtp_control_receiver(void *arg) {
(drand48() > config.diagnostic_drop_packet_fraction)) {
ssize_t plen = nread;
if (packet[1] == 0xd4) { // sync data
/*
// the following stanza is for debugging only -- normally commented out.
{
char obf[4096];
char *obfp = obf;
int obfc;
for (obfc = 0; obfc < plen; obfc++) {
snprintf(obfp, 3, "%02X", packet[obfc]);
obfp += 2;
};
*obfp = 0;
// get raw timestamp information
// I think that a good way to understand these timestamps is that
// (1) the rtlt below is the timestamp of the frame that should be playing at the
// client-time specified in the packet if there was no delay
// and (2) that the rt below is the timestamp of the frame that should be playing
// at the client-time specified in the packet on this device taking account of
// the delay
// Thus, (3) the latency can be calculated by subtracting the second from the
// first.
// There must be more to it -- there something missing.
// In addition, it seems that if the value of the short represented by the second
// pair of bytes in the packe is 7
// then an extra time lag is expected to be added, presumably by
// the AirPort Express.
// Best guess is that this delay is 11,025 frames.
// uint32_t rtlt = nctohl(&packet[4]); // raw timestamp less latency
// uint32_t rt = nctohl(&packet[16]); // raw timestamp
// uint32_t fl = nctohs(&packet[2]); //
// debug(1,"Sync Packet of %d bytes received: \"%s\", flags: %d, timestamps %u and
%u,
giving a latency of %d frames.",plen,obf,fl,rt,rtlt,rt-rtlt);
// debug(1,"Monotonic timestamps are: %" PRId64 " and %" PRId64 "
respectively.",monotonic_timestamp(rt, conn),monotonic_timestamp(rtlt, conn));
}
*/
if (packet[1] == 0xd4) { // sync data
/*
// the following stanza is for debugging only -- normally commented out.
{
char obf[4096];
char *obfp = obf;
int obfc;
for (obfc = 0; obfc < plen; obfc++) {
snprintf(obfp, 3, "%02X", packet[obfc]);
obfp += 2;
};
*obfp = 0;
// get raw timestamp information
// I think that a good way to understand these timestamps is that
// (1) the rtlt below is the timestamp of the frame that should be playing at the
// client-time specified in the packet if there was no delay
// and (2) that the rt below is the timestamp of the frame that should be playing
// at the client-time specified in the packet on this device taking account of
// the delay
// Thus, (3) the latency can be calculated by subtracting the second from the
// first.
// There must be more to it -- there something missing.
// In addition, it seems that if the value of the short represented by the second
// pair of bytes in the packe is 7
// then an extra time lag is expected to be added, presumably by
// the AirPort Express.
// Best guess is that this delay is 11,025 frames.
// uint32_t rtlt = nctohl(&packet[4]); // raw timestamp less latency
// uint32_t rt = nctohl(&packet[16]); // raw timestamp
// uint32_t fl = nctohs(&packet[2]); //
// debug(1,"Sync Packet of %d bytes received: \"%s\", flags: %d, timestamps %u and
%u,
giving a latency of %d frames.",plen,obf,fl,rt,rtlt,rt-rtlt);
// debug(1,"Monotonic timestamps are: %" PRId64 " and %" PRId64 "
respectively.",monotonic_timestamp(rt, conn),monotonic_timestamp(rtlt, conn));
}
*/
if (conn->local_to_remote_time_difference) { // need a time packet to be interchanged
// first...
@@ -455,7 +455,7 @@ void *rtp_timing_sender(void *arg) {
request_number++;
if (request_number <= 4)
usleep(500000); //these are thread cancellation points
usleep(500000); // these are thread cancellation points
else
usleep(3000000);
}
@@ -518,9 +518,9 @@ void *rtp_timing_receiver(void *arg) {
return_time = arrival_time - conn->departure_time;
uint64_t rtus = (return_time * 1000000) >> 32;
//uint64_t rtus = (return_time * 1000000) >> 32;
if (rtus < 300000) {
if (((return_time * 1000000) >> 32) < 300000) {
// debug(2,"Synchronisation ping return time is %f milliseconds.",(rtus*1.0)/1000);
@@ -683,7 +683,7 @@ void *rtp_timing_receiver(void *arg) {
//(return_time*1000000)>>32);
} else {
debug(2, "Time ping turnaround time: %lld us -- it looks like a timing ping was lost.",
rtus);
(return_time * 1000000) >> 32);
}
} else {
debug(1, "Timing port -- Unknown RTP packet of type 0x%02X length %d.", packet[1], nread);
@@ -947,8 +947,8 @@ void rtp_request_resend(seq_t first, uint32_t count, rtsp_conn_info *conn) {
(struct sockaddr *)&conn->rtp_client_control_socket, msgsize) == -1) {
char em[1024];
strerror_r(errno, em, sizeof(em));
debug(1,
"Error %d using send-to to an audio socket: \"%s\". Backing off for 1/16th of a second.",
debug(1, "Error %d using send-to to an audio socket: \"%s\". Backing off for 1/16th of a "
"second.",
errno, em);
conn->rtp_time_of_last_resend_request_error_fp = time_of_sending_fp;
} else {
@@ -956,8 +956,10 @@ void rtp_request_resend(seq_t first, uint32_t count, rtsp_conn_info *conn) {
}
} else {
debug(3, "Dropping resend request packet to simulate a bad network. Backing off for 1/16th of a "
"second.");
debug(
3,
"Dropping resend request packet to simulate a bad network. Backing off for 1/16th of a "
"second.");
conn->rtp_time_of_last_resend_request_error_fp = time_of_sending_fp;
}
} else {