[ltp] Input devices seem to freeze or hang periodically
Carl Klitscher
linux-thinkpad@www.bm-soft.com
Wed, 7 Nov 2001 13:23:32 +1300
Hi Eric,
This was one of many things I ended up doing and as a result I never really
got round to testing insert/eject. My setup is 600x with SuSE 7.1 at the
2.4.0 kernel. The machine was basically unusable for a very long time while
I got everything sorted out. This involved (as you have done) recompiling
with and without APM enabled amongst other things. I now have APM enabled
but my battery is knackered so I am usually plugged in to AC and don't
suspend/resume/insert/eject in the normal course of events and so can't
tell if that helps or hinders.
Other things were to fix the CPU clock speed at max (498.somethingMhz) and
to disable the PCI bus power saving 'features'. The theory behind this one
was that some bits of linux had timing dependancies and didn't like having
the clock speeds changed on them. This had almost no effect on stability
but I left them set that way anyway because I run exclusively on AC anyway
(see battery capability as above).
The next thing was to bypass some of the neomagic driver acceleration
features under X 4.0/KDE2.1.1 which was a vast improvement. I went from
3-5minutes Mean Time Between Reboots to almost 2 hours...
The final kicker was patching the kernel with the aforementioned patches. I
had to dig around a bit to get them on as well as they didn't quite match
the kernel I am on. It would pay to check your sources to see if they are
in fact there. I am using the pcmcia-cs package at 3.1.28 level rather than
the kernel based support.
Not very scientific I'm afraid but most of the info did come from this list
so you should be able to find the relevant bits in the archives.
Carl
Carl Klitscher
Systems Specialist
IBM New Zealand Ltd
Phone (64-4) 576-5892. Mobile 021 460-594
Linux - Open. For Business.
Eric
Blouin/Raleigh/IBM@IBM To: linux-thinkpad@www.bm-soft.com
US cc:
Sent by: Subject: Re: [ltp] Input devices seem to freeze or hang periodically
owner-linux-thinkpad@b
m-soft.com
11/07/01 11:10 AM
Please respond to
linux-thinkpad
Carl,
I think you might be on to something. I believe your patches were already
applied to the kernel source and modules I was running.
Did your keyboard/mouse spring back to life when the card was physically
ejected? Did your system continue to degrade with time? What
troubleshooting techniques did you use to track down the mem leak before
(maybe there's a mailing list thread somewhere?)
Thanks,
Eric Blouin
----- The Linux ThinkPad mailing list -----
The linux-thinkpad mailing list home page is at:
http://www.bm-soft.com/~bm/tp_mailing.html