[ECOS] rattler board questions
Richard.von.Lehe@gd-ais.com
Richard.von.Lehe@gd-ais.com
Mon May 17 22:18:00 GMT 2004
It turns out that there is a configuration word and it sets the IMM base
to be FF000000 (factory setting). I still don't know why my section of
code below doesn't work. Still open to suggestions...
Richard.von.Lehe@gd-ais.com wrote:
> Hello,
>
> Basic questions I am unable to find answers to:
>
> 1) How does the IMM base register get set? I don't see it set
> anywhere in rattler.S, although writes to internal registers are being
> made there. There is obviously an assumption that the internal memory
> map is at 0xFF000000, but somewhere this has to be set! This leads me
> to my second question.
>
> 2) Is there a hardware configuration word in EEPROM that sets the IMM
> base register? If so, should I assume it is 0xFF000000?
> 3) If the above is true, why can't I set internal memory-mapped
> registers OR4/BR4 and OR5/BR5?
>
> Here's some code that I've been trying to get to work. Any ideas on
> why this would segfault on the stw line? I have inserted this
> snippet at the end of hal_hardware_init in rattler.S. Basically the
> original body of hal_hardware_init is passed over in my application
> startup because I am defining CYG_HAL_STARTUP_RAM. My code gets
> executed right before the routine returns (blr).
>
> addis r30,0,0xFF01 # IMMR base addr = 0xFF000000+10000. We add
> # 0x10000 because using relative addressing
> # in load and store instructions only allow a
> # offset from the base of +/-32767.
>
> // CS4 - 1MB external <--- CHANGING TO 32K!!
> lwi r3,0xFFFF8000
> stw r3,CYGARC_REG_IMM_OR4(r30)
> lwi r3,0x80001801
> stw r3,CYGARC_REG_IMM_BR4(r30) // CS5 - NVRAM on
> exp board
> // lwi r3,0xFFFF9030
> lwi r3,0xFFFF8000
> stw r3,CYGARC_REG_IMM_OR5(r30)
> lwi r3,0x800F0801 stw r3,CYGARC_REG_IMM_BR5(r30)
>
>
> Any insight? Gary Thomas?
>
> Thanks,
> Rich
>
--
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss
More information about the Ecos-discuss
mailing list