[ltp] Re: Suspend on T23: a small step forward

[Martin Steigerwald]

Am Montag, 24. Januar 2005 17:27 schrieb Michael Perry:

> Here are a few interesting differences between Martin and my setup.
> BTW, Martin and I have been working in private email on this as well
> for awhile.  I went through a few changes here to get the T23 working.
> The ideas around unloading usb and alsa I have tried before as well as
> trying to stop pcmcia services and restart them.  I got stuck for
> awhile on the whole clock drift question and that's when I decided to
> get the latest acpi diffs patch and apply it.  I don't use software
> suspend 2 here at all so that is a bit different.  I also don't use an
> initrd but I don't know how that would make a difference...  I guess
> in the truly random nature of things it could though :).

Hello Michael,

well I tried without initrd and the kernel didn't come back after a 
suspend. Do you use hotplug, Michael? That might be another difference. 
Well I thought hotplug would be kind of a standard.

> I think overall the T23's Martin and I have are very close in software
> and bios and kernel yet mine will do suspends and alsa starts back up
> after a suspend event.  After applying the newest acpi patches I don't
> have clock drift any longer.  PCMCIA cards just work now and they did

I think about building exact the same kernel as you do and try with that 
one. Before I added configuration options for stuff I need (ISDN) or 
stuff I want (cpu frequency scaling). It would be enough for testing the 
suspend stuff and then I can add my additions one after another. Once 
thing I have to add straight away is XFS, since my root partition is 
using it. But ideally this wouldn't have an impact on the power 
management stuff.

Ah, so the CONFIG_PM_TIMER is not needed at all probably.

> not with a 2.6.9 kernel during a suspend.  I have never had problems
> with USB; in fact, I have a USB 2.0 controller on a pcmcia card I use
> with an ipod and it comes back fine.

Hmmm, I also have a USB 2.0 controller on a PCMCIA card and this one seems 
to come back fine. But the USB 1.1 controller on the motherboard where I 
connected my Logitech mouse doesn't seem to work after resume.

> I'd like to get this down to a set of conditions where we could help
> others with this laptop and perhaps even provide some corrections to
> linux on laptops or whatever for a T23 laptop configuration using
> ACPI.

I too. I had the idea of installing a Wiki on my webpage to work on these 
issues together, but maybe a better place would be having something like 
this on www.linux-thinkpad.org. Anyway I can make a Wiki on my homepage 
quite quickly if creating one on www.linux-thinkpad.org is not possible 
in short time. 

Regards,
-- 
Martin 'Helios' Steigerwald - http://www.Lichtvoll.de