Re: [microblaze-uclinux] More details of my problem - uClinux freezestoo

[John Williams <jwilliams@xxxxxxxxxxxxxx>]

Hi Aurash,
Aurelian Lazarut wrote:

> I'm using a version of your demo, on which I've modified the MHS (DCM
> uarts etc) and is working @ 75Mhz with the existing image, without
> modifying any settings for the kernel, so my question is: why is
> working  for me it doesn't for Ivan? or it works in the real hardware if
> is faster than the 66 Mhz setting on the kernel but not the other way
> around?

When the kernel calculates the timer interrupt period, it does so with 
respect to the specified clock frequency.  I guess on your 75 MHz system 
time will be running that little bit faster than the wall clock.

As for why yours works and Ivan's doesn't - I was guessing - it probably 
just means that the clock period is not relevant to his problem!

Ivan's problem is strange, the character corruption in the output 
suggests something like rogue memory writes, which can be from problems 
with link scripts, bootmem reserves at startup, all sorts of things. :-(

It is possible to single step the kernel with mb-gdb, which can be very 
useful in these situations.  I've been meaning to write up for a while, 
maybe this will spur me on to do so.

John

___________________________
microblaze-uclinux mailing list
microblaze-uclinux@itee.uq.edu.au
Project Home Page : http://www.itee.uq.edu.au/~jwilliams/mblaze-uclinux
Mailing List Archive : http://www.itee.uq.edu.au/~listarch/microblaze-uclinux/