[PATCH 0/3] WOL: Add Wake On LAN support
Holger Macht
hmacht at suse.de
Thu Oct 11 02:31:16 PDT 2007
On Thu 30. Aug - 11:49:24, Holger Macht wrote:
> On Tue 28. Aug - 21:47:12, Bastien Nocera wrote:
> > On Tue, 2007-08-28 at 14:12 -0400, David Zeuthen wrote:
> > > On Tue, 2007-08-14 at 16:54 +0200, Holger Macht wrote:
> > > > Adding the GetSupported(void) method to the coldplug detection of HAL
> > > > might make sense, though. I already have code, partly copied from ethtool,
> > > > to detect this from inside C code. It just makes it more complicated IMO
> > > > even though I'm an advocator of binary versus script code. However, I'll
> > > > check this again.
> > >
> > > Actually I think it's nicer to just have a GetSupported() method instead
> > > of a property. At least badly written drivers won't crap out until it's
> > > called which is preferable to having the system lock up at boot time.
> > > Which is what happened back when hal was trying to detect whether a
> > > cable was plugged in.
> >
> > Do we actually want to care about drivers like that? They'll get fixed,
> > and people will downgrade their HAL in the meanwhile. Given that the
> > property never changes, it doesn't make much sense to check with
> > GetSupported() all the time. HAL checks once, all the apps know it can
> > handle it...
>
> Actually after having done the work already, I also think a property is
> more clean ;-) I also don't think the code would be too risky to break
> drivers. Drivers which don't have ethtool support are unlikely to break,
> drivers which do should be fixed in any case to make use of its
> functionality.
Ok, so where to go from here? David, shall I decide for myself what to do?
;-) Going the secure or the clean/right way? Property or method? ;-)
Regards,
Holger
More information about the hal
mailing list