APMD-List:
Archives

  From: Chris Worley <cworley@liberate.com>
  To  : <apmd-list@worldvisions.ca>
  Date: Thu, 20 Jul 2000 07:57:53 -0600

Re: Using power switch to go into standby mode

David Brownell wrote:
> > This brings about other problems: The pegasus USB nic driver goes nuts
> > (recursive panics),
> 
> Unplugging the adapter should work around that, but I'd also
> be sure to run at least 2.4.0-test4 (which is where I think
> a batch of pegasus updates landed).  Or at least its USB code.

Off topic, but let me rant...

That driver is buggier than Windoze on a bad day.  I hope future revisions fix it, but
I've seen the same problems through many revisions already.  I hope USB and Linux
networking aren't mutually exclusive... but they sure seem like a bad fit.

I'm currently using 2.4.0-test4.

Half the time, at bootup, the driver comes up, the link light turns on, but no activity. 
That's one of the reasons I want to put this machine into a standby mode rather than
shutdown: once the pegasus driver starts working, I don't want to muck with it, although,
it often quits communicating under load (and won't restart so I have to reboot anyway). 
Note that I have 2 I-Openers and 2 Lynksys USB100TX adapters and they behave the same...
so I don't think it's buggy hardware.

If I try to "ifconfig eth0 down", it doesn't fix any of the problems or stop any of the
panics.

If I try to unload the driver, it goes into it's recursive panic.  

If I try unplugging the adapter, the driver causes a recursive panic.  

I have noticed that the resume after suspend recursive panic seems to be caused by the USB
driver trying to instantiate a second instance of the adapter.

Sorry for off topic.

> 
> >    and a "resume" message is sent from the bios about
> > five seconds after the suspend is sent.
> 
> That's odd ... it should be exactly four seconds!

I'm missing something.  Why should a resume be sent?  I interpret the "soft-off by PWRBTN:
[delay 4 sec | Instant-Off]" bios setting as either delay 4 seconds before sending "power
down", or do it immediately.  I'm grateful that "delay four seconds" sends a "suspend"
signal... I can rationalize that behavior, but why should it wait four seconds and send a
"resume"?  That makes no sense at all.

Chris