[pulseaudio-discuss] [RFCv0 20/21] bluetooth: Notify the main thread of a stream fd HUP

[Tanu Kaskinen]

On Sun, 2014-06-01 at 14:09 +0300, Tanu Kaskinen wrote:
> On Tue, 2014-02-04 at 19:04 -0300, jprvita at gmail.com wrote:
> > From: João Paulo Rechi Vita <jprvita at openbossa.org>
> > 
> > ---
> >  src/modules/bluetooth/module-bluez5-device.c | 5 +++++
> >  1 file changed, 5 insertions(+)
> > 
> > diff --git a/src/modules/bluetooth/module-bluez5-device.c b/src/modules/bluetooth/module-bluez5-device.c
> > index f1991a0..e48eaa9 100644
> > --- a/src/modules/bluetooth/module-bluez5-device.c
> > +++ b/src/modules/bluetooth/module-bluez5-device.c
> > @@ -74,6 +74,7 @@ static const char* const valid_modargs[] = {
> >  
> >  enum {
> >      BLUETOOTH_MESSAGE_IO_THREAD_FAILED,
> > +    BLUETOOTH_MESSAGE_STREAM_FD_HUP,
> >      BLUETOOTH_MESSAGE_MAX
> >  };
> >  
> > @@ -1427,6 +1428,8 @@ io_fail:
> >          pending_read_bytes = 0;
> >          writable = false;
> >  
> > +        pa_asyncmsgq_post(pa_thread_mq_get()->outq, PA_MSGOBJECT(u->msg), BLUETOOTH_MESSAGE_STREAM_FD_HUP, NULL, 0, NULL, NULL);
> > +
> >          teardown_stream(u);
> >      }
> >  
> > @@ -1998,6 +2001,8 @@ static int device_process_msg(pa_msgobject *obj, int code, void *data, int64_t o
> >              pa_log_debug("Switching the profile to off due to IO thread failure.");
> >              pa_assert_se(pa_card_set_profile(m->card, pa_hashmap_get(m->card->profiles, "off"), false) >= 0);
> >              break;
> > +        case BLUETOOTH_MESSAGE_STREAM_FD_HUP:
> > +            break;
> >      }
> >  
> >      return 0;
> 
> The fail/io_fail distinction as implemented makes the code hard to
> understand. I'll try to write a patch that makes the code clearer, I'd
> like to rebase this patch then on top of my patch.

I changed my mind a bit: instead of rebasing your patch on top of my
patch, I'll apply your patch and write my patch on top of your patch.

-- 
Tanu