CVS Commit: vlc

[Samuel Hocevar]

On Fri, Feb 22, 2002, Sigmund Augdal wrote:

> >   * ./plugins/lirc/lirc.c: non-blocking patch from Sigmund Augdal
> >     <sigmunau at stud.ntnu.no>; hope this time my enhancements don't break
> >     everything :-)
> 
> Isn't the need for this patch a sign of bad architecture?

   Of a bad architecture in lirc, yes. A blocking function to which you
cannot specify a timeout is bad, period :-)

> Without this p_intf->pf_manage() (who's content I don't know) was
> executed once for each button pressed and that worked fine except
> for the stop button which needed it to be run twice with som delay
> between. Now it's runed some 20(?) timed a second.

   p_intf->pf_manage() manages a few vlc internal structures, such as
jumping to the next playlist item when the current one is finished. It
only needs to be called a few times a second. This way to do things
spares a thread, but this may change in the future.

   Also, 20 times a second may sound a bit high, but it does not eat
that much CPU and allows for better interface response times.

-- 
Sam.

-- 
This is the vlc-devel mailing-list, see http://www.videolan.org/vlc/
To unsubscribe, please read http://www.videolan.org/lists.html
If you are in trouble, please contact <postmaster at videolan.org>