On 5 Jun, this message from Stephen van Egmond echoed through cyberspace: > Michel Lanners ([EMAIL PROTECTED]) wrote: >> 166 BAD interrupts? that's a lot... > > Yeah, I don't know what's up with that. Machine's running fine, > though.
Any funny hardware? >> > they don't >> > do anything quite so mundane as use, you know, interrupts for the >> > keyboard. >> >> Oh, they do, just not in the way that a freaky i386 developer expects. >> When you only have 15 IRQs, reserving one for only the keyboard may not >> be the most efficient use of system resources... > > I knew they were doing something smarter with their proto-USB protocol. > Funny, I didn't know Apple hardware had a 15-IRQ limit. I suppose there > had to be *some* limit. No, PowerPC doesn't have a 15-IRQ limit; PCs have. First-gen PowerMacs have 32 IRQs, later models 64 or more... not all of which are used, of course. IRQ useage is closely related to your hardware: if you have a dedicated keyboard controller, than it might make sense to dedicate an IRQ for that. Apple just chose to do it differently... and often (but not always) better than in PCs. Then again, the cruft in PCs is the old legacy nobody has gotten rid of.... >> > I wonder if anyone knew of a better way to watch for system activity in >> > any meaningful way? >> >> Maybe try and watch /dev/input/* ? That's where keyboard and mouse >> events get delivered. Then again, it all depends on your definition of a >> 'meaningful way'... > > Meaningful = won't result in the machine shutting down unexpectedly, > i.e. while the keyboard or mouse is being used. > > Unfortunately, sleepd doesn't do anything remotely like watching > /dev/input . Maybe you need to get your hands dirty... Cheers Michel ------------------------------------------------------------------------- Michel Lanners | " Read Philosophy. Study Art. 23, Rue Paul Henkes | Ask Questions. Make Mistakes. L-1710 Luxembourg | email [EMAIL PROTECTED] | http://www.cpu.lu/~mlan | Learn Always. " -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
