Am Montag 04 September 2006 15:36 schrieb Sjoerd Simons: > On Tue, Aug 22, 2006 at 12:54:54PM -0700, Steve Langasek wrote: > > It's clear that hald has the means to work whether acpid is started in > > advance or not, but that acpid has no such mechanism to cope with hald > > blocking its access to /proc/acpi/event. Nor should it, AFAICT. > > > > If we want reliable cooperation between these two packages, the only way > > I can see to do it is for hal to depend on acpid and not try to access > > /proc/acpi/event directly. There's no other way to reliably ensure that > > acpid always has a chance to connect first without hald opportunistically > > blocking the file. > > I don't want hal depending on acpid. I will patch the hal acpi addon to not > directly use the proc interface iff /usr/sbin/acpid exists. > > The only open problem will then be when acpid is installed by the user > while hal is already running. In that case acpid should probably restart > hald. > > Does that sound okay ?
No problem. Btw.: you can get the acpi events by reading from the acpid socket. The events are mirrored there. Cheers, Cajus -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
