Re: Code hangs or crashes on Kernel startup (dbau1100)
- From: Silver <thesilverring@hot>
- Date: Thu, 27 Jul 2006 08:09:43 -0700
Is there any debug output at all that we could see?
What are you building? (a full build or the sample provided by AMD or a minkern)
Is this CE5.0?
As for the SDRAM setup, leaving the settings the same may or may not work as your SDRAM setup will have a different row/column layout. Also, is the eboot on your board aware of the different RAM layout? Your config.bib should also reflect your new RAM layout.
For further debugging fire up your JTAG and step through the code, KernelStart is in here (for 5.0 at least):
WINCE500\PRIVATE\WINCEOS\COREOS\NK\KERNEL\MIPS\STARTUP.S
Geoff
--
jbarker@xxxxxxxxxxxxxxxxxx wrote:
Hello-.
I've not brought up a board under Wince before, so please excuse any
ignorance. We have a new board which is based on the AMD Alchemy
DbAu1100 evaluation board. We use a similar procedure that was done on
the Alchemy board, which is a YAMON monitor programmed into flash,
which then runs EBOOT, which will load the kernel via. Platform
Builder, and theoritically run. We had a couple of issues with
correctly setting the Endian mode on the processor, but now YAMON
boots, we can run EBOOT, it connects to Platform Builder, loads the
code, but after it tries to start the kernel, nothing happens.
Unfortunately we don't have a lot of visibility on our board, but we do
have a logic analyzer tied to a bus address we can write to. Using this
and sprinkling debug code it seems that the reset_db1100.s is called
once when eboot comes up, and then again when the kernel is initialized
(from the HAL startup.s file). It seems to make it all the way through
reset_db1100.s even during the kernel initialization, then it goes into
kernel code, and at that point I am at a loss as to what might be going
on.
Does anyone have any ideas on what we can probe or try next? We tried
putting debug code in the OEMInit function and it never makes it there.
Any and all ideas are appreciated as we are not making any progress on
this at the moment. We've tried to isolate any differences between our
board and the AMD eval board- the SDRAM is a different size (larger on
our board), but if we leave things configured for the AMD board, the
extra RAM will be ignored correct?
- References:
- Code hangs or crashes on Kernel startup (dbau1100)
- From: jbarker
- Code hangs or crashes on Kernel startup (dbau1100)
- Prev by Date: Re: Determining physical memory distribution
- Next by Date: Re: NK.Bin and binary extraction.
- Previous by thread: Code hangs or crashes on Kernel startup (dbau1100)
- Next by thread: NK.Bin and binary extraction.
- Index(es):
Relevant Pages
|
