Message-ID: <[EMAIL PROTECTED]>

Finally i got around to trying out your suggestion.

I enabled the chipselect by setting the V-bit to 1. Yup - silly me...
I removed the reference to m5272sim.h

No good. Still no worky.

Then I got to think about it - something still made the CPU decide it was 
handed something it wasn't allowed to do.

Looking over the individual bits in CSMR2, it dawned on me: Bits 5 to 1 are 
mask bits, restricting access to resources governed by the chipselect. Cleared 
bits cause access, set bits cause - Illegal instruction?
Since this is a driver module, running in Supervisor mode and i'm trying to 
access data, zeroing the SD-bit *only* seemed like a good idea. I also zeroed 
the AM-bit, so that the beforementioned bits become dont-cares during DMA - if 
i should ever need that :)
I also disabled the Byte Enable Mode bit in CSCR2 to disable /BS activity on 
reads. Seemed like a good idea...

Et voila :)

I'm not entirely clear on which of these modifications did the trick, but it 
works!

Actually i've reduced access granting specs - previously also User Data Acces 
was allowed. AM-bit? Well, i don't use DMA, so it shouldn't matter. THE BEM bit 
governs /BS activity. This I read as /BS[3-0], which i don't use.

Hmmm...

Oh well...

Enough feedback for now. Thanks for your help :)

//Martin Filitenborg

--------------------------------------
Did you enable the chipselect?

>    CSMR2 = 0x00000074;    // BAM=0, AM=1, SD=0, UD=0,
should be CSMR2 = 0x00000075;

and don't use m5272sim.h which is not for 5282


_______________________________________________
uClinux-dev mailing list
[email protected]
http://mailman.uclinux.org/mailman/listinfo/uclinux-dev
This message was resent by [email protected]
To unsubscribe see:
http://mailman.uclinux.org/mailman/options/uclinux-dev

Reply via email to