Subject: Re: hangs in 1.4Q: more disksort bugs?
To: Jason Thorpe <email@example.com>
From: Jonathan Stone <jonathan@DSG.Stanford.EDU>
Date: 02/07/2000 10:28:01
In message <200002060043.QAA00137@lestat.nas.nasa.gov>Jason Thorpe writes
>On Sat, 05 Feb 2000 13:15:30 -0800
Just following up to Jason again:
> Jonathan Stone <jonathan@DSG.Stanford.EDU> wrote:
> > I just tried a kernel built from today's sup, to see if the
> > power-management in if_tlp_cardbus.c made my NetGear card work after
> > suspends. (thanks yet _again_, Jason!!) But after resuming, I
> > experienced symptoms just like the disksort bug from a week or two
> > back. Processes start wedging up, eventually I cant do anything, and
> > I have to reboot to get the machine back.
> > Is this known? fixed already? Anyone else seen the same thing?
>Please just update to as current as you can get... you can even sort
>of hot-swap your CardBus tlp card.
With Sunday's sup, if the tlp card isn't in at boot time, I didn't get
a peep. I'm trying to rebuild Monday's sup, but (as per Subject:) the
Sunday-source kernel I'm running on my laptop keeps wedging when I do
a make -j. It doesn't seem to be related to suspends after all, but it
does look like a bug with similar symptoms to the previous disksort()
OTOH, if I reboot withtout the tlp card, things seem fine.
Kernel thread bug, maybe?
>I'm still planning on doing the power hooks, but I have a LOT of
>other things people are demanding from me right now.
>For now, do what I do... use an /etc/apm/suspend script to bring
>the interface down when you suspend.
Doesn't seem to have any effect on my NetGear cardbus tlp card. This
newer laptop doesn't have BIOS-menu entries for controlling power; the
BIOs says it cant do that when ACPI is active. I guess it's awlays
doing ACPI. Is that likely to be relevant?