From: Keith Moore (moore_at_CS.UTK.EDU)
Date: Mon Oct 03 1994 - 15:48:40 EDT
> > I'm not sure the manual implies that complete system
> > lockup is normal behavior, but I have to admit that this issue
> > isn't directly addressed. It just seems strange to me that IBM
> > would consider a system hang as "normal behavior". In fact, behavior
> > which requires a reboot after attempting to resume or "un-hibernate"
> > seems to defeat the purpose of the suspend/hibernate function
> > altogether.
>
> The problem is that the system locks up in the device driver.
>
> > I don't know about IBM, but I most definitely do NOT
> > consider this normal behavior...........
>From what I understand of PCMCIA, this is not something that can be
addressed in the BIOS, unless (that is) you want the BIOS to leave
the PCMCIA cards powered-up while in suspend mode.
It is up to the PCMCIA device driver to realize that a machine has awakened
>from suspend mode and re-initialize the device. The BIOS doesn't know
anything about the PCMCIA card and so lacks the necessary information to do
the re-initialization.
For my own system, I hacked an existing NetBSD Ethernet driver to work with
IBM's PCMCIA ethernet card. It has no problems recovering from suspend mode.
By a happy accident, it turned out that the driver already had code to
re-initialize the ethernet chip when certain errors occured. Such an error
was produced immediately upon trying to send a packet after re-awakening.
Keith
This archive was generated by hypermail 2.1.3 : Thu Jan 23 2003 - 09:52:26 EST