>Going to some non-native implementation would probably be doable but it
>would also make the thing unneccesarily big and slow.
I found C versions of ash(l|r)di3 and moved all the files to libgcc
directory, is this better?
> > In real hardware, the zs nodes are named zs@0,0 and zs@0,100000. Maybe
>the
> > address has some effect to naming?
>
>yes, the parent node needs a decode-unit and encode-unit method. See
>drivers/pci.fs
I added dummy methods, but there is no change. I'm afraid my Forth skills
are so basic that while I can guess what the drivers/pci.fs and util/pci.fs
are doing, I can't make a Sbus version. Could you help me?
_________________________________________________________________
Express yourself instantly with MSN Messenger! Download today it's FREE!
http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/
>But I did not check in the libgcc files. Instead I added inclusion of
>gcc --print-libgcc in the make process. Can you try whether this works
>for you?
Maybe when not cross-compiling, but that's way too slow for me. I have built
four different versions of cross-gcc's and while they all can compile, none
has a working libgcc. That's why I had to remove the linking to libgcc and
add the files. I think not using system libgcc is cleaner way, the same
approach as taken in Linux kernel. Another solution would be getting rid of
long long arithmetic.
Here's my current version against SVN revision 3. I added most of the
devices to the tree.
There are two zs devices, but the other one can't be reached:
2 > cd /obio ok
2 > ls
-2bc228 interrupt@0
-2bc180 counter@0
-2bc0dc eeprom@0
-2bc024 zs@0
-2bbed4 zs@0
ok
2 > cd zs ok
2 > .properties
name "zs"
device_type "serial"
reg -- c : 0 0 0 0 0 0 0 0 0 0 0 8
slave 1
model "mk48t08"
intr -- 8 : 0 0 0 2c 0 0 0 0
keyboard 1
mouse 1
ok
In real hardware, the zs nodes are named zs@0,0 and zs@0,100000. Maybe the
address has some effect to naming?
_________________________________________________________________
Don't just search. Find. Check out the new MSN Search!
http://search.msn.click-url.com/go/onm00200636ave/direct/01/
Paul,
I dont have any test docs or drafts for fibre-channel, iSCSI
(and ethernet as well) for OpenBIOS at this point.
Currently I develope FCode for Sun's and IBM's OpenFirmware.
We hired a contractor to port existing FCode to the Apple server.
But it did not get much recognition.
Anyway, it would be awesome if QLogic at first released a
none supported version of FCode for the OpenBIOS/PC platform.
Then when it gets enough recognition from the OpenBIOS community,
I will further endorse QLogic to release supported versions of
fibre-channel, iSCSI and ethernet FCode for OpenBIOS.
Then it would be even more awesome if companies like Dell/HP and
may be even Sun and IBM realized the value of OpenBIOS for their
x86 machines. To make this happen, the interest should start somewhere.
I am going to attempt to make QLogic the pioneer in trying OpenBIOS.
_____________
Asif Haswarey
C1069W
QLogic Corporation
26650 Aliso Viejo Parkway
Aliso Viejo, CA 92656
Phone: 949.389.7607
Fax: 949.389.6308
-----Original Message-----
From: openbios-bounces(a)openbios.org
[mailto:openbios-bounces@openbios.org] On Behalf Of Paul Nasrat
Sent: Wednesday, April 26, 2006 1:02 PM
To: The OpenBIOS Mailinglist
Subject: Re: [OpenBIOS] Sparc32 progress
On Wed, 2006-04-26 at 12:13 -0700, Asif Haswarey wrote:
> At some point I plan to start testing with OpenBIOS and endorse the
> support of OpenBIOS in our fibre-channel and iSCSI products.
Out of curiosity do you have drafts of how you see fibre-channel/iSCSI
working with OpenBIOS/1275?
Paul
--
OpenBIOS http://openbios.org/
Mailinglist: http://www.openbios.org/mailman/listinfo
Free your System - May the Forth be with you
Hi Segher,
Just wanted to mention that it is always safe to
stick to the OpenFirmware spec as far as tokens
are concerned.
"int-property" might break the current Sun's tokenizer,
unless "int-property" exists in the tokenizer dictionary
or it is explicitly defined earlier in the FCode.
The key to increase the credibility and interest of
OpenBIOS among the masses is to _not break the OF spec_.
(By-the-way, I am not criticizing anyone.)
At some point I plan to start testing with OpenBIOS and
endorse the support of OpenBIOS in our fibre-channel and iSCSI
products.
_____________
Asif Haswarey
C1069W
QLogic Corporation
26650 Aliso Viejo Parkway
Aliso Viejo, CA 92656
Phone: 949.389.7607
Fax: 949.389.6308
-----Original Message-----
From: openbios-bounces(a)openbios.org
[mailto:openbios-bounces@openbios.org] On Behalf Of Segher Boessenkool
Sent: Wednesday, April 26, 2006 11:50 AM
To: The OpenBIOS Mailinglist
Cc: stepan(a)coresystems.de
Subject: Re: [OpenBIOS] Sparc32 progress
> d# 256 encode-int " mmu-nctx" int-property
Use either
d# 256 " mmu-nctx" int-property
or
d# 256 encode-int " mmu-nctx" property
(The former is shorter, the latter is standard OF, can be tokenized).
Segher
--
OpenBIOS http://openbios.org/
Mailinglist: http://www.openbios.org/mailman/listinfo
Free your System - May the Forth be with you
* Blue Swirl <blueswir1(a)hotmail.com> [060425 21:38]:
> 2 > cd / ok
> 2 > ls
> -2c45e0 aliases
> -2c453c openprom
> -2c42e8 options
> -2c4270 chosen
> -2c418c builtin
> -2bf2f8 packages
> -2bdc38 memory
> -2bdb94 STP1012PGA
> -2bda40 iommu
> ok
> 2 > cd iommu ok
> 2 > ls
> -2bd9c8 sbus
> ok
Very cool! Did you hardcode the amount of memory or do you probe it from
the emulator?
> Next step would be adding SBus device tree and SCSI devices.
Yes. Most likely other parts of the device tree are missing as well.
> I have to confess that I don't know much Forth except the RPN notation and
> some Sun OBP commands. Can I print the device attributes somehow, like OBP
> .attributes command?
there's
.properties to show device properties and
ls to show sub device nodes and
words to show methods defined for the current node.
> I can't cd to the STP1012PGA node unless the device is renamed to something
> shorter. Is this a limitation somewhere?
Maybe an unintended one ;) But not a natural limitation. How do you
create the node?
Stefan
--
coresystems GmbH · Brahmsstr. 16 · D-79104 Freiburg i. Br.
Tel.: +49 761 7668825 · Fax: +49 761 7664613
Email: info(a)coresystems.de · http://www.coresystems.de/
Hi,
since nobody except me really dared checking anything into the
repository since I switched to GNU arch/tla I decided to move on
to something more practical: Subversion
It's easier to use, the tools are less buggy and have a less braindead
user interface / command line than GNU arch.
Go ahead and do a clean checkout with:
svn co svn://openbios.org/openbios/openbios-devel
You can download the latest snapshot at:
http://openbios.org/viewcvs/openbios-devel.tar.gz?root=OpenBIOS&view=tar
Watch the tree in your web browser here:
http://openbios.org/viewcvs/openbios-devel/?root=OpenBIOS
The documentation on the website will be updated asap.
Stefan
--
coresystems GmbH · Brahmsstr. 16 · D-79104 Freiburg i. Br.
Tel.: +49 761 7668825 · Fax: +49 761 7664613
Email: info(a)coresystems.de · http://www.coresystems.de/
Hi,
I made a first attempt for Sparc32 support, patch attached. Currently we get
this far:
boot arg = 0x0
0000000000000000-0000000000000000
RAM 0 MB
Checksum invalid (d93e0f8c)!
forth started.
initializing memory...done
panic: no dictionary entry point.
Unhandled Exception 0x00000002
PC = 0x00000008 NPC = 0x0000000c
Stopping execution
Any comments or suggestions?
_________________________________________________________________
Express yourself instantly with MSN Messenger! Download today it's FREE!
http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/
Hi,
I've been working on Sparc32/64 support for Qemu. Currently Qemu for Sparc32
uses Proll as a system boot rom, but it's not a very good starting point for
Sparc64. Basically, Linux expects a bit too much from the OpenFirmware
interface, like Forth support etc.
I guess OpenBIOS would be a better choice. How complete do you think
OpenBIOS is, can it boot Linux for example on some architecture? Is there
any support for 64-bit architectures?
_________________________________________________________________
Express yourself instantly with MSN Messenger! Download today it's FREE!
http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/
Hi. Recently cleared the BIOS of my Abit IS7-V2 motherboard. And the result
was my system wont display anything on the screen. I can hear the cpu
running that is the POST has loaded but not the BIOS. Is that possible that
the BIOS has completely lost and is there a way to restore it...
THANKS
--
View this message in context: http://www.nabble.com/Restoring-BIOS-t1415766.html#a3815516
Sent from the OpenBIOS forum at Nabble.com.