This is an old thread and I do apologise for the
delay in the reply - I have no excuse.
On Fri 3, Nov 2000 Derek Glidden wrote:
> Ah, a little too small for a kernel image then. :)
Exactly. Actually early on in the piece Jamey and I spent some time
trying to boot the kernel from RAM but ran into a problem.
We assumed some asynchronous event was happening that caused the
kernel boot sequence to fail - that is execution didn't continue
past a known instruction.
> I presume that any work towards being able to boot into Linux from WinCE
> would require either finding a different section of memory to load the
Pretty close to what I was thinking. Rather than hope that I can find some
memory
the idea is to setup virtual memory correctly so that new code can be loaded
and
executed from it. From looking at the cyace code (LinuxCE folks) or pbsboot
code (PocketBSD folks) they do this sort of thing.
> kernel or by loading a 56K bootstrap program that would then load and
> run the kernel?
BTW the Welly Project is underway which will allow you to 'boot from RAM'.
> didn't think about CVS until later. I looked through the code a little
> bit and see what it's doing, although I don't entirely understand it.
At first it looks wierd - it just copies some code into a free bit of
memory and then jumps to it. The code must be compiled for the particular
address it is being executed from.
>> If you would like to particpate in this project that would be great!
> I would love to, but I would have no idea what to do. I've got
> absolutely zero experience with ARM programming, and very little
I'll have to figure out exactly what direction I'm going to take and
I'll leave it up to you to decide wether you wish to help out :)
> experience at all with any kernel-level stuff... But if you need a beta
> tester, I'd love to!
Great! It is very likely I'll take you up on that.
Regards,
Dirk
Received on Wed Nov 8 17:21:29 2000
This archive was generated by hypermail 2.1.8 : Tue May 04 2004 - 09:43:45 EDT