Re: [ecasound] Ecasound build process stuck after tests

On Fri, Apr 01, 2011 at 12:38:49AM +0300, Kai Vehmanen wrote:
> Hi,
>
> On Wed, 30 Mar 2011, Alessandro Ghedini wrote:
>
> >Turns out that, when running "ecasound --version" it sometimes does not
> >exits properly (it prints on STDOUT and then it stays there).
> [...]
> > 0x00007f8de15f3be5 in pthread_join () from /lib/libpthread.so.0
>
> ok, this is really interesting. Could you send me the config.h file
> from a machine where ecasound hangs here (maybe in a private mail
> not to spam the list)?

Sent.

> Also, does sending a SIGHUP to the stuck ecasound process help?

Nope, it does not show signs of life.

> This is essentially a problem with closing down the signal watchdog
> thread. To properly handle signals (e.g. terminating processing with
> ctrl-c should cleanly close all devices, updated output files
> headers, etc), ecasound creates a single thread that is dedicated to
> handle only signals. The signals are blocked in all other threads of
> ecasound (this solves the ambiguity of which thread gets the
> signals).
>
> As the last step before exiting, ecasound main thread sends a SIGHUP
> to the watchdog, and waits (with pthread_join()) until the watchdog
> terminates and then exits the process.
>
> Hmm, I have one idea what could happen (and a fix), but it'll take a
> few days before I have time to put together a patch for testing.

No hurry.

Cheers

-- 
perl -E'$_=q;$/= @{[@_]};and s;\S+;<inidehG ordnasselA>;eg;say~~reverse'
------------------------------------------------------------------------------
Create and publish websites with WebMatrix
Use the most popular FREE web apps or write code yourself; 
WebMatrix provides all the features you need to develop and 
publish your website. http://p.sf.net/sfu/ms-webmatrix-sf
_______________________________________________
Ecasound-list mailing list
Ecasound-list@email-addr-hidden
https://lists.sourceforge.net/lists/listinfo/ecasound-list