headers
Yoshihiro Masuda | 2 Jun 2006 01:56
Picon

Re: ibm4xx pci

I think so to.

Since some powerpc device such as 8555 does not have any HOST-PCI
bridge on pci bus, cpu can be seen directory on PCI bus 0 dev 0
function 0. For such device, pci bus should atttach to cpubus(ocb?)
directory.

--
masuda

From: Izumi Tsutsui <tsutsui <at> ceres.dti.ne.jp>
Subject: Re: ibm4xx pci (Re: IBM405GP/GPr OPB bus_space endian (powerpc/ibm4xx/dev/opb.c))
Date: Wed, 10 May 2006 22:25:51 +0900

> In article <4461C1F1.5080300 <at> netbsd.org>
> shige <at> NetBSD.org wrote:
> 
> > Why is pchb below on plb directly?
> > (cf. On i386 port, mainbus -> pci -> pchb.)
> 
> Maybe difference between physical and logical structures, as Allen said.
> 
> From software view, the PCI bridge can be seen via PCI configuration
> space. It's a logical "PCI-HOST bridge" PCI device.
> On real hardware, the PCI bridge is located on CPU local bus.
> While it's called "pchb" on evbppc, it isn't the same device
> with the former PCI-HOST bridge device seen via PCI, but a bus
> controller attached to the local bus.
> 
> We shouldn't call the latter one "pchb" to avoid confusion?
(Continue reading)

Permalink | Reply | 0 comments |
headers
Sanjay Lal | 2 Jun 2006 04:40

NetBSD 3.0 ported to IBM 970/G5 system simulator

I am happy to announce that NetBSD 3.0 has been ported to the IBM 970/ 
G5 (Mambo) system simulator (http://www.alphaworks.ibm.com/tech/ 
systemsim970).  This is a 32 bit port that uses the 970's 64 bit  
bridge mode.  The kernel comes up using the memory disk into single  
user mode. Current device support includes the cpu, interrupt  
controller (openpic) and the console device (com).  I have also added  
preliminary support for the IBM north bridge.

The port is also looking very promising for the Powermac G5, as it  
uses the same chip and north bridge. I have gotten my G5 machine to  
come up and try to spawn init.  I'll have more on that on going  
effort shortly.

Regards and thanks
Sanjay Lal
Kyma Systems
http://www.kymasys.com

The dmesg is enclosed:

Mambo console initialized
[ Kernel symbol table missing! ]
ofmaplen: 0 bytes
openfirmware_entry: 0xf0000000
About to call pmap_bootstrap(), startK: 0x00100000, endK: 0x004da000
********no memory?*********
Allocating 0x80000 bytes for pmap_pvo_table
uvm_page_physload: 0x00083000 -> 0x00100000
uvm_page_physload: 0x004da000 -> 0x00800000
uvm_page_physload: 0x01000000 -> 0x1fffd000
(Continue reading)

Permalink | Reply | 0 comments |
headers
matthew green | 2 Jun 2006 07:38
Picon
Favicon

re: NetBSD 3.0 ported to IBM 970/G5 system simulator


   I am happy to announce that NetBSD 3.0 has been ported to the IBM 970/ 
   G5 (Mambo) system simulator (http://www.alphaworks.ibm.com/tech/ 

cool!
Permalink | Reply | 0 comments |
headers
Hammond, Robin-David%KB3IEN | 2 Jun 2006 10:37

re: NetBSD 3.0 ported to IBM 970/G5 system simulator


I am looking for documentation for creating a unionfs comprised of a read 
only file system with a disposable ram disk (writable) so I can make /var/ 
and such goodies log/conf-writeable while protecting the system's 
integrity.

Can anyone point me at the relevant documentaion?

I have made this tar.bz2 longer (-6), because I lacked the cycles to make it shorter (-9).

Robin-David Hammond	KB3IEN
 	www.aresnyc.org.
Permalink | Reply | 0 comments |
headers
Jachym Holecek | 13 Jun 2006 18:45
Favicon

cntlzw() -> cpu.h

Hello,

I'd like to merge mutiple definitions of cntlzw() to <powerpc/cpu.h>,
OK to commit?

	-- Jachym

Index: macppc/macppc/extintr.c
===================================================================
RCS file: /cvsroot/src/sys/arch/macppc/macppc/extintr.c,v
retrieving revision 1.55
diff -u -r1.55 extintr.c
--- macppc/macppc/extintr.c	24 Dec 2005 22:45:35 -0000	1.55
+++ macppc/macppc/extintr.c	13 Jun 2006 16:35:36 -0000
 <at>  <at>  -87,6 +87,7  <at>  <at> 
 #include <uvm/uvm_extern.h>

 #include <machine/autoconf.h>
+#include <machine/cpu.h>
 #include <machine/intr.h>
 #include <machine/psl.h>
 #include <machine/pio.h>
 <at>  <at>  -103,7 +104,6  <at>  <at> 
 void intr_calculatemasks __P((void));
 int fakeintr __P((void *));

-static inline int cntlzw __P((int));
 static inline uint32_t gc_read_irq __P((void));
 static inline int mapirq __P((int));
 static void gc_enable_irq __P((int));
(Continue reading)

Permalink | Reply | 2 comments |
headers
Allen Briggs | 13 Jun 2006 19:05
Picon

Re: cntlzw() -> cpu.h

On Tue, Jun 13, 2006 at 06:45:50PM +0200, Jachym Holecek wrote:
> I'd like to merge mutiple definitions of cntlzw() to <powerpc/cpu.h>,
> OK to commit?

Looks good to me.

-allen

--

-- 
                  Use NetBSD!  http://www.NetBSD.org/
Permalink | Reply | 1 comment |
headers
Simon Burge | 14 Jun 2006 03:49
Picon

Re: cntlzw() -> cpu.h

Allen Briggs wrote:

> On Tue, Jun 13, 2006 at 06:45:50PM +0200, Jachym Holecek wrote:
> > I'd like to merge mutiple definitions of cntlzw() to <powerpc/cpu.h>,
> > OK to commit?
> 
> Looks good to me.

Same here!

Cheers,
Simon.
Permalink | Reply | 0 comments |
headers
nightbird | 14 Jun 2006 13:56

IBM CBE simulator


IBM provides a simulator for the Cell Broadband Engine that runs on 
Linux (packages provided for Fedora Core). Is it possible to use 
the simulator for porting NetBSD to the Cell processor? Of course, 
you don't have the performance of a real Cell processor, but if you 
don't have access to a Cell-based machine like me, this would be a 
solution, wouldn't it? Has anybody made experience with the 
simulator?

Concerned about your privacy? Instantly send FREE secure email, no account required
http://www.hushmail.com/send?l=480

Get the best prices on SSL certificates from Hushmail
https://www.hushssl.com?l=485
Permalink | Reply | 1 comment |
headers
Peter Seebach | 14 Jun 2006 15:19

Re: IBM CBE simulator

In message <20060614115646.DAE02DA820 <at> mailserver8.hushmail.com>, nightbird <at> mac.
hush.com writes:
>IBM provides a simulator for the Cell Broadband Engine that runs on 
>Linux (packages provided for Fedora Core). Is it possible to use 
>the simulator for porting NetBSD to the Cell processor? Of course, 
>you don't have the performance of a real Cell processor, but if you 
>don't have access to a Cell-based machine like me, this would be a 
>solution, wouldn't it? Has anybody made experience with the 
>simulator?

I have a little experience with the sim, and I will be getting more as soon
as a version compatible with the 2.6.16 Linux kernels is available to me.

I would assume you could use it for porting, given that it was used for
porting Linux, and indeed, you can download the whole patch set for making
Linux run on the Cell.

-s
Permalink | Reply | 0 comments |
headers
Tim Rightnour | 19 Jun 2006 21:04
Gravatar

gcc4 switchover

So I've done some testing with gcc4 on prep, and it seems to work quite well. 
I had heard about some initial problems with gcc4 on macppc, but I understand
those have been fixed.

Since there is discussion of switching the ports that work to gcc4 for the
upcoming 4.0 release, I'd like to bring up the idea that we switch powerpc as
well.

So:

Has anyone else tested thier powerpc port with gcc4?

Has it failed in some way?

What do the other powerpc portmasters think about switching?

---
Tim Rightnour <root <at> garbled.net>
NetBSD: Free multi-architecture OS http://www.netbsd.org/
Genecys: Open Source 3D MMORPG: http://www.genecys.org/
Permalink | Reply | 1 comment |

Gmane