Subject: Re: Call for testers: ACPI suspend/resume, part 2
To: Jared D. McNeill <jmcneill@invisible.ca>
From: Matthias Drochner <M.Drochner@fz-juelich.de>
List: port-amd64
Date: 12/19/2007 21:43:22
[screen switch]
jmcneill@invisible.ca said:
> It should stay in the kernel; joerg has plans to add a 'vbetool post'
> equivalent to the kernel that will resolve this issue. 

OK, if the "vbetool post" equivalent happens before the switch
it should work. There would still be too many assumtions in the
kernel for my taste, eg that X is not running on the first screen --
I'll post an idea in another mail.

[bge]
I haven't looked at that driver, do you have time?

I've tried some more times, and found that it resumes correctly.
Don't know what happened... unfortunately I also found that
the box resets on resume sometimes (one out of 5 or 10 tries),
so there seem to be some strange interactions.

> Ugh, we really need to look at this USB issue

Yes, this is the only nasty and reproducible problem
for me. Shouldn't be too hard to debug... just had a look
at the suspend/resume code, and my first impression was that
it mixes pci/chip and usb hub/bus power management is a way
which makes it hard to get a clean structure in. Hubs should
be handled seperately, and perhaps there are differences between
bus-powered and self-powered hubs, but could you add some comments
to the host controller pm code which tell what it is supposed to do,
so that a reviwer doesn't have to check against register level
documentation all the time?
(I believe I've got some moderate clue about USB power management,
but need to match it against the code.)

> Can you add an entry to the wiki with your results?

There is already an entry for a Dell Latitude D820. We don't agree
about USB obviously, but without more detailed information (hubs,
highspeed devices) one possibly can't compare. Same for graphics -
one can get the D820 with cheap Intel chipset graphics (as I did),
or Nvidea (or was it AIT, don't remenber).

best regards
Matthias




-------------------------------------------------------------------
-------------------------------------------------------------------
Forschungszentrum Juelich GmbH
52425 Juelich

Sitz der Gesellschaft: Juelich
Eingetragen im Handelsregister des Amtsgerichts Dueren Nr. HR B 3498
Vorsitzende des Aufsichtsrats: MinDirig'in Baerbel Brumme-Bothe
Geschaeftsfuehrung: Prof. Dr. Achim Bachem (Vorsitzender),
Dr. Ulrich Krafft (stellv. Vorsitzender), Dr. Sebastian M. Schmidt
-------------------------------------------------------------------
-------------------------------------------------------------------