ceos/espeakup
1
0
Fork 0
mirror of https://github.com/linux-speakup/espeakup.git synced 2026-08-09 09:19:09 +02:00
Code Issues Projects Releases Packages Wiki Activity

softsynth: bound the wait for espeak to acknowledge a stop request

Browse source 
request_espeak_stop() waited forever for the espeak thread to
acknowledge the stop.  If espeak-ng is wedged inside the audio output
(e.g. an ALSA device blocked or stuck returning EBUSY, as reported in
issue #62), the acknowledgement never comes, and the softsynth thread
stops draining /dev/softsynth forever.  Speakup's kernel buffer then
fills up and console output stalls, which matches the "blocks dmesg
output after a couple of pages" observation in issue #45.  The process
goes silent and only SIGKILL gets rid of it.

Wait at most 10 seconds for the acknowledgement (a normal cancellation
takes milliseconds; the timeout can only trigger when espeak is truly
stuck).  On timeout, exit with a clear message so that the init system
respawns espeakup in a clean state: our systemd unit already has
Restart=always.  A one-second restart beats an unkillable silent
daemon, and was explicitly requested by the reporter of issue #62.

Helps: https://github.com/linux-speakup/espeakup/issues/45
Helps: https://github.com/linux-speakup/espeakup/issues/62

Co-Authored-By: Claude <noreply@anthropic.com>
This commit is contained in:
Alexander Epaneshnikov 2026-06-11 11:06:35 +03:00 • committed by Samuel Thibault
commit e92edf96dc
1 changed files with 25 additions and 2 deletions
Download patch file Download diff file Expand all files Collapse all files

27
src/softsynth.c
View file

@ -24,6 +24,7 @@
#include <stdlib.h>
#include <string.h>
#include <sys/select.h>
#include <time.h>
#include <unistd.h>
#include "espeakup.h"
@ -223,15 +224,37 @@ static void process_buffer_acsint(struct synth_t *s, char *buf, ssize_t length)
}
}
/* How long to wait for the espeak thread to acknowledge a stop request
* before concluding that espeak is wedged beyond in-process recovery. */
static const int stopAckTimeout = 10;
static void request_espeak_stop(void)
{
struct timespec timeout;
int err = 0;
pthread_mutex_lock(&queue_guard);
stop_requested = 1;
pthread_cond_signal(&runner_awake); // Wake runner, if necessary.
pthread_cond_signal(&wake_stop); // Wake runner, if necessary.
while (should_run && stop_requested)
clock_gettime(CLOCK_REALTIME, &timeout);
timeout.tv_sec += stopAckTimeout;
while (should_run && stop_requested && err != ETIMEDOUT)
// wait for acknowledgement.
pthread_cond_wait(&stop_acknowledged, &queue_guard);
err = pthread_cond_timedwait(&stop_acknowledged, &queue_guard,
&timeout);
if (should_run && stop_requested) {
/* The espeak thread is stuck in a call into espeak, most likely
* on a wedged audio device. There is no way to recover from
* within the process: exit so that the init system can respawn
* us in a clean state, rather than staying silent, ignoring
* SIGTERM, and stalling the whole console by not draining
* /dev/softsynth anymore. Use _exit because exit could hang in
* library destructors while the audio device is wedged. */
fprintf(stderr, "espeakup: espeak did not acknowledge a stop "
"request within %d seconds, aborting\n", stopAckTimeout);
_exit(3);
}
pthread_mutex_unlock(&queue_guard);
}