Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

JavaStation running 1.5_A :-)

2 views
Skip to first unread message

Darren Reed

unread,
Jul 7, 2000, 3:00:00 AM7/7/00
to

THanks for the patch pk, the javastation is now running 1.5_ALPHA like
a champ...

almost.

In my bootparams file, I have:
swap=server:/usr/netbsd/swap
root=server:/usr/netbsd/root
with /usr/netbsd exported via nfs. root is mounted, no problems, but
the kernel doesn't appear to get the swap. Do I need to move the swap
file inside /usr/netbsd/root and just add /swap to fstab ?

Darren

Jason R Thorpe

unread,
Jul 7, 2000, 3:00:00 AM7/7/00
to

swap is always configured via swapctl now... no automatic swap.

--
-- Jason R. Thorpe <tho...@zembu.com>

matthew green

unread,
Jul 7, 2000, 3:00:00 AM7/7/00
to

In my bootparams file, I have:
swap=server:/usr/netbsd/swap
root=server:/usr/netbsd/root
with /usr/netbsd exported via nfs. root is mounted, no problems, but
the kernel doesn't appear to get the swap. Do I need to move the swap
file inside /usr/netbsd/root and just add /swap to fstab ?


that's the easiest and preferred solution.

Pete Zaitcev

unread,
Jul 7, 2000, 3:00:00 AM7/7/00
to
> From: Darren Reed <ava...@coombs.anu.edu.au>
> Date: 07/07/2000 10:51:19

> THanks for the patch pk, the javastation is now running 1.5_ALPHA like
> a champ...

I am very glad that our friendly competition is not sleeping.
Now, what about PCI support with IIep? I have certain hopes for
NetBSD on SunRay because Linux has no swap over NFS to speak of,
but NetBSD does. With 8MB of RAM it may be handy, don't you think?
If you have any questions about IIep, feel free to ask.

Here is the Javastation and relatives platform scorecard:

Platform Linux NetBSD
------------------ ------------- ----------------
(SBus/PCI platforms)
JS-1 (OBP 2.30) Solid Works
JS-1 (OBP 3.x) Solid ? works ?

(PCI mass produced platforms)
CP-1200 (OBP 2.30) Lame (DMA error on hme) -
JS-NC Solid -
Boots from flash, too
JE-1 Solid -
SGSC Worked, bitroted -
SR-1 Not planned in 8MB -

(PCI lame duck and one-off platforms)
JS-E Solid, -
includes IDE disk
Roxy Works, but abandoned -
Fox E.B. did something... -

--Pete

Eduardo Horvath

unread,
Jul 7, 2000, 3:00:00 AM7/7/00
to
On Fri, 7 Jul 2000, Pete Zaitcev wrote:

> I am very glad that our friendly competition is not sleeping.
> Now, what about PCI support with IIep? I have certain hopes for
> NetBSD on SunRay because Linux has no swap over NFS to speak of,
> but NetBSD does. With 8MB of RAM it may be handy, don't you think?
> If you have any questions about IIep, feel free to ask.

What PCI controllers to they use?

Eduardo Horvath

Hagbard Seline

unread,
Jul 7, 2000, 3:00:00 AM7/7/00
to

Fox works fine on NetBSD. Been using it for quite some time now as
router and mp3 player. What was Roxy the next revision Fox?

On Fri, 7 Jul 2000, Pete Zaitcev wrote:

> > From: Darren Reed <ava...@coombs.anu.edu.au>
> > Date: 07/07/2000 10:51:19
>
> > THanks for the patch pk, the javastation is now running 1.5_ALPHA like
> > a champ...
>

> I am very glad that our friendly competition is not sleeping.
> Now, what about PCI support with IIep? I have certain hopes for
> NetBSD on SunRay because Linux has no swap over NFS to speak of,
> but NetBSD does. With 8MB of RAM it may be handy, don't you think?
> If you have any questions about IIep, feel free to ask.
>

Pete Zaitcev

unread,
Jul 7, 2000, 3:00:00 AM7/7/00
to
> Date: Fri, 7 Jul 2000 10:33:58 -0700 (PDT)
> From: Eduardo Horvath <e...@turbolinux.com>
> To: Pete Zaitcev <zai...@metabyte.com>
> cc: port-...@netbsd.org, ava...@coombs.anu.edu.au
> Subject: Re: JavaStation running 1.5_A :-)

> On Fri, 7 Jul 2000, Pete Zaitcev wrote:
>
> > I am very glad that our friendly competition is not sleeping.
> > Now, what about PCI support with IIep? I have certain hopes for
> > NetBSD on SunRay because Linux has no swap over NFS to speak of,
> > but NetBSD does. With 8MB of RAM it may be handy, don't you think?
> > If you have any questions about IIep, feel free to ask.
>

> What PCI controllers to they use?
>
> Eduardo Horvath

IIep features an integrated PCI controller on the chip.
It is very well documented, see this URL:
http://www.sun.com/microelectronics/manuals/microSPARC-IIep/802-7100-01.pdf

--Pete

Paul Kranenburg

unread,
Jul 7, 2000, 3:00:00 AM7/7/00
to
> Now, what about PCI support with IIep?

A PCI engine driver for these machines needs to be written. I don't
think that will be a terribly complicated task.

I also hope to borrow much, if not all, of the Ebus stuff emerging
in the sparc64 port. In fact, a common `/sys/dev/{sun/?}ebus'
should be erected a.s.a.p.

Aside: which of those various models of javastations are using the
Openfirmware PROM? (I assume all of the PCI models do). OFW support
has been in NetBSD/sparc for more than a year already, but there
are a few details to be worked out still (read: I need testers or
a test machine..).

-pk

Eduardo Horvath

unread,
Jul 7, 2000, 3:00:00 AM7/7/00
to
On Fri, 7 Jul 2000, Paul Kranenburg wrote:

> > Now, what about PCI support with IIep?
>
> A PCI engine driver for these machines needs to be written. I don't
> think that will be a terribly complicated task.
>
> I also hope to borrow much, if not all, of the Ebus stuff emerging
> in the sparc64 port. In fact, a common `/sys/dev/{sun/?}ebus'
> should be erected a.s.a.p.

I think you will need 64-bit `paddr_t's or `bus_addr_t's first.

Eduardo Horvath

Paul Kranenburg

unread,
Jul 7, 2000, 3:00:00 AM7/7/00
to
> > I also hope to borrow much, if not all, of the Ebus stuff emerging
> > in the sparc64 port. In fact, a common `/sys/dev/{sun/?}ebus'
> > should be erected a.s.a.p.
>
> I think you will need 64-bit `paddr_t's or `bus_addr_t's first.

Why?

The common bus code can import <machine/bus.h> to get those when
needed. And if 64-bit entities are inherent to the ebus design,
those should be coded as such explicitly. The common sbus interface
and drivers work in the same way.

-pk

Eduardo Horvath

unread,
Jul 7, 2000, 3:00:00 AM7/7/00
to

The common parts of sbus may be coded the same way but the sbus drivers
are not. The sysio driver is quite dependent on `bus_addr_t' being
64-bits wide. The same is true of the ebus driver.

Eduardo Horvath

Darren Reed

unread,
Jul 8, 2000, 3:00:00 AM7/8/00
to
In some mail from Paul Kranenburg, sie said:
>
> > Now, what about PCI support with IIep?
>
> A PCI engine driver for these machines needs to be written. I don't
> think that will be a terribly complicated task.
>
> I also hope to borrow much, if not all, of the Ebus stuff emerging
> in the sparc64 port. In fact, a common `/sys/dev/{sun/?}ebus'
> should be erected a.s.a.p.
>
> Aside: which of those various models of javastations are using the
> Openfirmware PROM? (I assume all of the PCI models do). OFW support
> has been in NetBSD/sparc for more than a year already, but there
> are a few details to be worked out still (read: I need testers or
> a test machine..).

If I had anything more advanced than the JS-1, I'd be happy to do beta
testing with it.

Darren

David Brownlee

unread,
Jul 11, 2000, 3:00:00 AM7/11/00
to
If anyone is running NetBSD on hardware not listed on the
main http://www.netbsd.org/Ports/sparc/ page... please mail in
an update including model number and machine type... :)


David/absolute
-- www.netbsd.org: No hype required --

On Fri, 7 Jul 2000, Hagbard Seline wrote:

>
> Fox works fine on NetBSD. Been using it for quite some time now as
> router and mp3 player. What was Roxy the next revision Fox?
>
> On Fri, 7 Jul 2000, Pete Zaitcev wrote:
>
> > > From: Darren Reed <ava...@coombs.anu.edu.au>
> > > Date: 07/07/2000 10:51:19
> >
> > > THanks for the patch pk, the javastation is now running 1.5_ALPHA like
> > > a champ...
> >

> > I am very glad that our friendly competition is not sleeping.
> > Now, what about PCI support with IIep? I have certain hopes for
> > NetBSD on SunRay because Linux has no swap over NFS to speak of,
> > but NetBSD does. With 8MB of RAM it may be handy, don't you think?
> > If you have any questions about IIep, feel free to ask.
> >

Erik E. Fair

unread,
Jul 17, 2000, 3:00:00 AM7/17/00
to
I have a JS-1 here that I've successfully netbooted NetBSD 1.5_ALPHA
on. However, I have not yet succeeded in making the keyboard work.
What's the magic string for "eeprom input-device=" ?

Erik <fa...@clock.org>

Todd Whitesel

unread,
Jul 17, 2000, 3:00:00 AM7/17/00
to
> I have a JS-1 here that I've successfully netbooted NetBSD 1.5_ALPHA
> on. However, I have not yet succeeded in making the keyboard work.
> What's the magic string for "eeprom input-device=" ?

Are you using the keyboard that came with it?

I believe Sun ships a PS/2 compatible keyboard with the JavaStation.
The connector is very similar (but not identical) to the Type 4 and 5.

Also, what about "dev / ls" and other openfirmware silliness.

Todd Whitesel
toddpw @ best.com

Eduardo Horvath

unread,
Jul 17, 2000, 3:00:00 AM7/17/00
to
On Mon, 17 Jul 2000, Todd Whitesel wrote:

> > I have a JS-1 here that I've successfully netbooted NetBSD 1.5_ALPHA
> > on. However, I have not yet succeeded in making the keyboard work.
> > What's the magic string for "eeprom input-device=" ?
>
> Are you using the keyboard that came with it?
>
> I believe Sun ships a PS/2 compatible keyboard with the JavaStation.
> The connector is very similar (but not identical) to the Type 4 and 5.

If it's a PS/2 keyboard then it's probably attached to the keyboard
controller on the SuperIO chip rather than the 16550s or zs8530, so you
would have to use the PC keyboard controller driver rather than the zs
driver or a hacked up com driver.

> Also, what about "dev / ls" and other openfirmware silliness.

What sillyness? You can have a complete picture of all the H/W on the
machine. How do you do that on a PC?

Eduardo Horvath

Jason R Thorpe

unread,
Jul 17, 2000, 3:00:00 AM7/17/00
to
On Mon, Jul 17, 2000 at 04:40:43AM -0700, Todd Whitesel wrote:

> Are you using the keyboard that came with it?
>
> I believe Sun ships a PS/2 compatible keyboard with the JavaStation.
> The connector is very similar (but not identical) to the Type 4 and 5.
>

> Also, what about "dev / ls" and other openfirmware silliness.

On my JS-1, the PS/2 keyboard is simply not recognized by the PROM. Here's
the story.

The JS-1 comes with output-device=screen and input-device=ttya. Hook
a terminal up to the serial port, send a BREAK, and you'll get "ok".
Type away, it's cool.

Set input-device=keyboard, and the PROM will happily report "no keyboard
present" right after it issues the reset command.

I'm guessing that the PROM is trying to get the keyboard ID from the
keyboard, and since it doesn't report "sun-type-4" or "sun-type-5",
so the PROM ignores it.

It all smells like an attempt to keep people from futzing with the PROM.

Jason R Thorpe

unread,
Jul 17, 2000, 3:00:00 AM7/17/00
to
On Mon, Jul 17, 2000 at 08:54:59AM -0700, Eduardo Horvath wrote:

> If it's a PS/2 keyboard then it's probably attached to the keyboard
> controller on the SuperIO chip rather than the 16550s or zs8530, so you
> would have to use the PC keyboard controller driver rather than the zs
> driver or a hacked up com driver.

No, on a JS-1, the keyboard/mouse is on an 8530, and the serial port
is on a 16550, as far as I can tell.

I don't see a junk i/o chip in the JS-1 at all. Looks more or less like
a hacked SS4.

Todd Whitesel

unread,
Jul 17, 2000, 3:00:00 AM7/17/00
to
> No, on a JS-1, the keyboard/mouse is on an 8530, and the serial port
> is on a 16550, as far as I can tell.

Have you tried a real type 4 or 5 on it yet? I happen to have one or
two extras lying around, BTW.

Does it really not work with the shipped keyboard?? That sounds like
a magic firmware setting which got lost in a PRAM reset. Lame, but
it makes more sense than the included keyboard simply not working...

0 new messages