mirrors/pipewire
1
0
Fork 0
mirror of https://gitlab.freedesktop.org/pipewire/pipewire.git synced 2026-07-05 00:06:16 -04:00
Code Issues Projects Releases Packages Wiki Activity Actions

spa: alsa: acp-device: stop 100% CPU spin when a poll fd hangs up

Browse source 
handle_acp_poll() copies the poll revents into ACP and clears them, but
never actually looks at them. When one of the polled fds reports
POLLERR/POLLHUP - which happens when the ALSA card goes away, e.g. the
audio function of a USB dock being unplugged - the source is never
removed, and because that condition is level-triggered the loop just
keeps redispatching the handler.

The result is wireplumber pegged at 100% CPU, spinning on the card's
control fd:

    read(24, ..., 72) = -1 ENODEV (No such device)

with fd 24 pointing at a now-deleted /dev/snd/controlC<n>. It stays like
that until the card comes back or the service is restarted.

Accumulate the returned mask and, if anything reports SPA_IO_ERR or
SPA_IO_HUP, drop the poll sources so we stop hammering a dead descriptor.
The udev monitor still drives the real device removal and setup_sources()
re-arms the poll if the card reappears - the same teardown alsa-pcm.c
already does for its own control sources.

Signed-off-by: Mike Lothian <mike@fireburn.co.uk>
This commit is contained in:
Mike Lothian 2026-07-04 23:58:05 +01:00
parent 85df114eb4
commit b904c3ee15
1 changed files with 18 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

19
spa/plugins/alsa/alsa-acp-device.c
View file

@ -85,17 +85,34 @@ struct impl {
};
static int emit_info(struct impl *this, bool full);
static void remove_sources(struct impl *this);
static void handle_acp_poll(struct spa_source *source)
{
struct impl *this = source->data;
uint32_t rmask = 0;
int i;
for (i = 0; i < this->n_pfds; i++)
for (i = 0; i < this->n_pfds; i++) {
this->pfds[i].revents = this->sources[i].rmask;
rmask |= this->sources[i].rmask;
}
acp_card_handle_events(this->card);
for (i = 0; i < this->n_pfds; i++)
this->sources[i].rmask = 0;
/* A POLLERR/POLLHUP on a poll fd (e.g. the ALSA control device was
* unplugged) is level-triggered and never clears, so the loop would
* redispatch this handler on every iteration and spin at 100% CPU.
* Stop polling the dead descriptors; the udev monitor handles the
* actual device removal, and setup_sources() re-arms on the next
* profile change if the card comes back. */
if (rmask & (SPA_IO_ERR | SPA_IO_HUP)) {
spa_log_warn(this->log, "%p: poll fd error/hangup (card removed?), "
"removing poll sources", this);
remove_sources(this);
return;
}
emit_info(this, false);
}