[PATCH] fix for acpi-addon if use/compile with acpid

[Richard Hughes]

On Wed, 2005-10-19 at 11:57 +0200, Danny Kukawka wrote:
> On Wednesday 19 October 2005 11:17, Richard Hughes wrote:
> > On Wed, 2005-10-19 at 00:15 +0200, Danny Kukawka wrote:
> > > Hi,
> > >
> > > this path fix a problem if you compiled the acpi addon (hald-addon-acpi)
> > > for/with acpid support. This patch should prevent that the addon exit, if
> > > the acpid is not running on startup of HAL. Without the patch you need to
> > > restart hald if the acpid is started later.
> >
> > I thought Ryan fixed this with the "reconnecting" stuff... I've cc'd him
> > as i don't think he's on hal-devel list.
> 
> He fixed the reconnect if you was already connected. I saw the patch, but not 
> the difference to my posted patch for the first connect to acpid. So now this 
> patch to fix also the first connect.

Sure, okay.

> > > With this patch I also changed the position of the defines for acpid and
> > > kernel proc event source to remove unneeded function call. Also added a
> > > debug message, if the addon lost the connection to the acpi proc
> > > interface.
> >
> > I get this sometimes, and then subsequently loose all acpid events (as
> > I'm not running acpid). Is it worth trying to reconnect to the proc
> > event source or is this a kernel bug somewhere?
> 
> Not sure, but if you lost the connection to proc this is maybe a bug in the 
> code of the addon or the kernel. IMO you never should lost the connection to 
> proc.

That's what I thought too..

>  If you take a look at my posted patch [1] you can the patch fixed also 
> this problem. I can take a look on the current source and implement this also 
> for proc.

Yes please. It's really irritating, and seemingly random.

Cheers,

Richard.