Kevin O'Connor wrote:
On Wed, Mar 05, 2008 at 01:38:08AM +0000, oliford wrote:
Do you have any suggestions for working out what's going on? If not, should
I write an e-mail to the kernel maintainers?
When that happened to me, I just traced through it. Not much fun.
I've got as far as working out that it's crashing in the scheduler
(__wake_up_common) or the spinlock debugging if it's switched on. I have
a feeling this is indicative of the memory being randomly overwritten,
as it seems like variables are randomly changing as it goes down the
call chain ending in __wake_up_common. I suspected this was due to
random hardware being left on so have been investigating HaRET's
hardware shutdown code. Sure enough, the PXA and PXA27x code in HaRET
isn't at all right for the PXA310 so I have begun writing the PXA310
machine handler for HaRET, for the moment just changing the PXA27x one.
This mostly involves changes of register positions, especially the clock
enables, DMAs and IRQ stuff. If I ever get this all working I'll package
it up as proper new machine entry and look into submitting it for HaRET.
Meanwhile, however, none of the changes seems to have stopped the
kernel oops. I have one last thing to try tonight, which is a
modification to the xscale cache clearing in HaRET, to include the L2
cache that is new to the PXA310 (xscale v3). I have also written to the
arm kernel mailing list with the problem, in case they have any advice.
Further to the not shutting down properly thoughts, I was looking at the
resume-into-boot feature of HaRET, in the hope that letting windows do
the hardware shutdown would be more successful. Unfortunately I can't
work out how to find the required RESUMEADDR. Is there some method for
doing this?
I notice that the resume into boot handler checks for a specific
instruction at the beginning of winCE's resume handler. Is this always
what it is, even for the newest versions of windows mobile? If so, I may
just try searching the entire memory for it and attempting the resume
tracing stuff a each place, see where it gets to. My ipaq is running Win
Mobile 6 (CE v5.2.1711).
Yeah, the hh.org kernel has fallen behind. A couple of months back I
attempted to forward port the changes. I put the repo at:
http://git.linuxtogo.org
However, even that is a bit behind now.
Ok, I'll stick to the kernel.org 2.6.25-rc4 for now then.
Thanks again,
Oliver
_______________________________________________
Haret mailing list
[email protected]
https://handhelds.org/mailman/listinfo/haret
