public inbox for ecos-discuss@sourceware.org
 help / color / mirror / Atom feed
* [ECOS] xscale ATU direct addressing on secondary PCI bus
@ 2001-05-11 17:54 McAtee, Kip
  2001-05-11 18:36 ` Mark Salter
  0 siblings, 1 reply; 2+ messages in thread
From: McAtee, Kip @ 2001-05-11 17:54 UTC (permalink / raw)
  To: 'ecos-discuss@sourceware.cygnus.com'

Greetings-

I'm trying to configure the secondary PCI bus for
ATU outbound direct addressing on the IQ80310.

The redboot setup for the ATUCR is:
*ATU_reg = 0x00000006;

If I try asserting 'Direct addressing enable' and
'Secondary Direct addressing enable':
*ATU_reg = 0x00000186;
the machine hangs immediately.

I suspect that I have to set something else up.
I think the MMU macros establish a linear mapping
over the ATU regions:

	// nothing interesting here (Address Translation)
	.rept	0xA00 - 0x8
	FL_SECTION_ENTRY __base,0,3,0,0,0,0
	.set	__base,__base+1
	.endr

Can anyone suggest something else to check,
or refer me to somebody that knows about this?

Thanks-
Kip

^ permalink raw reply	[flat|nested] 2+ messages in thread

* Re: [ECOS] xscale ATU direct addressing on secondary PCI bus
  2001-05-11 17:54 [ECOS] xscale ATU direct addressing on secondary PCI bus McAtee, Kip
@ 2001-05-11 18:36 ` Mark Salter
  0 siblings, 0 replies; 2+ messages in thread
From: Mark Salter @ 2001-05-11 18:36 UTC (permalink / raw)
  To: Kip.McAtee; +Cc: ecos-discuss

>>>>> McAtee, Kip writes:

> Greetings-

Hi Kip.

> I'm trying to configure the secondary PCI bus for
> ATU outbound direct addressing on the IQ80310.

> The redboot setup for the ATUCR is:
> *ATU_reg = 0x00000006;

> If I try asserting 'Direct addressing enable' and
> 'Secondary Direct addressing enable':
> *ATU_reg = 0x00000186;
> the machine hangs immediately.

> I suspect that I have to set something else up.

I think that direct ATU addressing may be impossible on the IQ80310.
I have never tried it, though. The doc says that the direct addressing 
window goes from 0x00002000 - 0x7FFFFFFF. The problem is that this
area is also used for flash. Flash starts at 0, has a "hole" for the
80312 i/o chip register from 0x1000 - 0x1fff, then the rest of flash
is from 0x2000 to 0x1ffffff. That conflicts with the direct addressing
window, so it seems that the direct window must remain disabled. 
This is just my impression from reading the docs and dooing the
RedBoot port. Intel developer support would be the place to get the
definitive answer.

--Mark

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2001-05-11 18:36 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2001-05-11 17:54 [ECOS] xscale ATU direct addressing on secondary PCI bus McAtee, Kip
2001-05-11 18:36 ` Mark Salter

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).