Fix native PCM startup stall at the queue gate
The PCM queue cap was applied before the first output sample. For any input longer than about two seconds the gate could trigger while samples_out was still zero, emit() had nothing to return, and process_pcm returned without marking filter progress. The decoder thread then stopped and audio never became ready; short clips and E-AC-3 JOC were unaffected. Only apply the cap after the first output sample, and mark progress when the gate cannot emit. This keeps the EOF flush bounded without blocking startup.
This commit is contained in:
1 parent
e5954169b2
commit
587ff24e86
1 file changed
+3
-1
@@ -567,9 +567,11 @@ static void process_pcm(struct mp_filter *f)
|
|||||||
* kernel keeps a rendered-output queue but its input queue is unbounded;
|
* kernel keeps a rendered-output queue but its input queue is unbounded;
|
||||||
* without this gate a whole track can pile up and flush() at EOF renders
|
* without this gate a whole track can pile up and flush() at EOF renders
|
||||||
* all of it at once. */
|
* all of it at once. */
|
||||||
if (!p->eof && p->pcm_input_samples > p->samples_out + PCM_MAX_QUEUE_SAMPLES) {
|
if (p->samples_out > 0 && !p->eof &&
|
||||||
|
p->pcm_input_samples > p->samples_out + PCM_MAX_QUEUE_SAMPLES) {
|
||||||
if (emit(f))
|
if (emit(f))
|
||||||
return;
|
return;
|
||||||
|
mp_filter_internal_mark_progress(f);
|
||||||
return;
|
return;
|
||||||
}
|
}
|
||||||
|
|
||||||
|
|||||||
Reference in new issue
Block a user