Re: Thread initialization and condition

[Christopher Lang <christopher lang acurana de>]

I was just briefly looking over this email thread: couldn't a glib::dispatcher 
be helpful in this case?


http://www.gtkmm.org/docs/glibmm-2.4/examples/thread/

http://www.gtkmm.org/docs/glibmm-2.4/docs/reference/html/classGlib_1_1Dispatcher.html

Chris

http://www.acurana.de/


Am Freitag 04 Januar 2008 15:45:22 schrieb Joaquim Duran:
> Hello,
>
> Andreas told in his previous email the he was already using a queue. The
> code to synchronize the thread should be similar to (pseucode):
>
> Main thread:
>
> Push a new file to queue (assuming STL like queue):
> mutex.lock()
> queue.push(fine_name);
> condition.signal();
> mutex.unlock();
>
>
> Stop a Job:
> mutex.lock();
> stop= true;
> condition.signal();
> mutex.unlock();
>
> FTP Thread:
>
> Get a file fro queue :
>
> string fname;
>
> mutex.lock();
> while ((!stop) && (queue.empty())
> {
> 	condition.wait();
> }
> if (stop)
> {
> 	mutex.unlock();
> 	exit_thread();
> }
>
> fname = queue.top();
> mutex.unlock();
> return fname;
>
> En/na Naveen Verma ha escrit:
> > Hi,
> >
> > I also tried the similar and very simple program on Ubuntu, and find
> > similar behavior. It seems that thread creation takes longer time,  and
> > we call the other member function i.e. fetch() before that. I did
> > nothing but just printing the debug logs and log in fetch() print before
> > that the loop().
> >
> > And the same behavior with pthread also. May be thread creation and
> > execution time varies between different OS and CPU?
> >
> > May be we can have a queue between threads, one thread put the requests
> > in queue and other thread reads and process from that?
> >
> >
> > -Br
> > Naveen
> >
> > On Jan 4, 2008 3:06 PM, Joaquim Duran <joaquinduran adtelecom es
> > <mailto:joaquinduran adtelecom es>> wrote:
> >
> >     En/na Andreas Volz ha escrit:
> >     > Hello,
> >     >
> >     > I have a problem while running a thread application. Here is my
> >     > code:
> >     >
> >     > Download::Download ()
> >     > {
> >     >   thread = Glib::Thread::create (sigc::mem_fun (*this,
> >     >            &Download::threadFunc), true);
> >     > }
> >     >
> >     > void Download::fetch ()
> >     > {
> >     >   // queue new files to download...
> >     > }
> >     >
> >     > void Download::loop ()
> >     > {
> >     >   do
> >     >   {
> >     >     // download new files
> >     >   }
> >     >   while (1)
> >     > }
> >     >
> >     > Ok, this was working on one machine. But now while testing it on
> >     > another machine (also Linux) I got other results. This is how I
> >
> >     use it:
> >     >   Download dl;
> >     >   dl.fetch (m_url, m_file);
> >     >
> >     > If I do debugging or print some debug messages I see that on one
> >     > machine the thread loop() is entered before the fetch() function. I
> >     > would expect that. But on the other machine the fetch() function is
> >     > called before the thread loop().
> >     >
> >     > Couldn't I trust gtkmm that the thread loop is constructed and
> >     > entered after returning the Glib::Thread::create() call?
> >     >
> >     > I tried to use some conditions to wait until the thread has
> >
> >     entered. It
> >
> >     > worked, but only on one machine. The other one blocked forever. If
> >
> >     this
> >
> >     > is a defined behaviour that I need to think about a more complex
> >     > condition mechanism at initialization time.
> >     >
> >     > Gentoo:
> >     > - gtkmm-2.10.10
> >     > - thread loop() is called *before* fetch()
> >     >
> >     > Ubuntu:
> >     > - libgtkmm-2.4-1c2a 1:2.12.0-0ubuntu1
> >     > - thread loop() is called *after* fetch()
> >     >
> >     > Any ideas about that problem?
> >     >
> >     > regards
> >     > Andreas
> >     > _______________________________________________
> >     > gtkmm-list mailing list
> >     > gtkmm-list gnome org <mailto:gtkmm-list gnome org>
> >     > http://mail.gnome.org/mailman/listinfo/gtkmm-list
> >
> >     This looks like that there is no synchronization between main thread
> > and the FTP thread. The FTP thread should be waiting till main thread has
> > posted a new file to download and then start download.
> >
> >     GLibmm provides mutex and mutex-conditions to implement a
> >     producer-consumer relation between threads.
> >
> >     Joaquim Duran
> >     Thanks and Best Regards,
> >     _______________________________________________
> >     gtkmm-list mailing list
> >     gtkmm-list gnome org <mailto:gtkmm-list gnome org>
> >     http://mail.gnome.org/mailman/listinfo/gtkmm-list
> >     <http://mail.gnome.org/mailman/listinfo/gtkmm-list>
>
> _______________________________________________
> gtkmm-list mailing list
> gtkmm-list gnome org
> http://mail.gnome.org/mailman/listinfo/gtkmm-list