Re: Re: [9fans] bootalpha and the no valid stack error

Re: Re: [9fans] bootalpha and the no valid stack error

[Brian L.Stuart]

> The thing about DEC's Unix on the Alpha was that it never ran on either
> the MIPS or VAX....  It was essentially OSF/1 with some DEC
> customizations and a custom compiler.

So I just spent a little time digging when I should have been writing.
And it's all your fault :-)  Seriously, you're essentially right.  The history
is more rich than I had realized.

- In the begining (well once Olsen got over his anti-UNIX thing), there
was ULTRIX on the 11, the VAX and MIPS.
- When DEC decided to swich from ULTRIX to OSF/1, it started its
development on MIPS.
- To what extent OSF/1 on MIPS was seen in the wild is not clear.  Claims
range from, it was never officially released to, support for MIPS existed
in the source tree for several versions.
- When Alpha was released, OSF/1 was the UNIX for it.
- When OSF (at least as an OS purveyor) imploded, DEC changed
the name to Digital UNIX.
- When the farm was sold to Compaq, it was again renamed to Tru64.

But I stray OT.  Sorry.

BLS

Re: [9fans] bootalpha and the no valid stack error

[Lyndon Nerenberg]

> - In the begining (well once Olsen got over his anti-UNIX thing),
> there
> was ULTRIX on the 11, the VAX and MIPS.

On the /11?  Ultrix seemed very VM-centric. (I.e. wanting post-11
TLBs and such.)

> - When DEC decided to swich from ULTRIX to OSF/1, it started its
> development on MIPS.
> - To what extent OSF/1 on MIPS was seen in the wild is not clear.
> Claims
> range from, it was never officially released to, support for MIPS
> existed
> in the source tree for several versions.

My hazy recollections of the first MIPS workstations involved Ultrix.
(1.1?)  No OSF until ...

> - When Alpha was released, OSF/1 was the UNIX for it.
> - When OSF (at least as an OS purveyor) imploded, DEC changed
> the name to Digital UNIX.
> - When the farm was sold to Compaq, it was again renamed to Tru64.

I think Tru64 pre-dated Compaq.  Nothing much changed between OSF/DUX/
Tru (from the standpoint of someone who had to keep reasonably large
apps running throughout).

Of all the commercial Unixen of the era, the Ultrix follow-on
variants were the least painful to deal with.  And the DEC MIPS-
derived C compiler kicked ass when it came to spitting out
grandmother-guilt-fed diagnostics :-)  (Was it DEC or MIPS who were
responsible for the Spanish Inquisition error messages?  While noisy,
the verbiage shortened a lot of other conversations :-)

--lyndon

Re: [9fans] bootalpha and the no valid stack error

[Dan Cross]

On Wed, Apr 26, 2006 at 08:29:42PM -0700, Lyndon Nerenberg wrote:
> >- In the begining (well once Olsen got over his anti-UNIX thing),
> >there
> >was ULTRIX on the 11, the VAX and MIPS.
>
> On the /11?  Ultrix seemed very VM-centric. (I.e. wanting post-11
> TLBs and such.)

Yes, there was Ultrix on the 11.  I believe you can download it....
Ultrix was a succession of different systems, starting on the PDP-11
but moving to the VAX and eventually MIPS.

> >- When DEC decided to swich from ULTRIX to OSF/1, it started its
> >development on MIPS.
> >- To what extent OSF/1 on MIPS was seen in the wild is not clear.
> >Claims range from, it was never officially released to, support for MIPS
> >existed in the source tree for several versions.

I know that CMU did a port of OSF/1 to MIPS, but I don't think it was
generally available.  And I certainly believe that a lot of the development
work was hosted on MIPS and possibly even targeted MIPS before the Alpha
was ready, but I don't ever recall it being a commercial offering from
DEC.  Even after the Alpha was released, if you bought a MIPS or VAX-based
machine from DEC, the only Unix offered was Ultrix.

Maybe if you were one of those special customers you could get them to
give you OSF for MIPS, but I never ranked that high.  :-)

> My hazy recollections of the first MIPS workstations involved Ultrix.
> (1.1?)  No OSF until ...

That's right.

> I think Tru64 pre-dated Compaq.  Nothing much changed between OSF/DUX/
> Tru (from the standpoint of someone who had to keep reasonably large
> apps running throughout).

Nothing much changed, but I'm pretty sure the Tru64 name came after the
Compaq buy-out.  The name Digial Unix didn't fly anymore since they didn't
keep the Digital name (though, if I recall, in the original buy-out
agreement they said they would).  For a while, when you bought a Compaq
Alpha, it came with the D|I|G|I|T|A|L logo still on the front-panel; I
guess branding wasn't so important as using back-inventory of parts.

> Of all the commercial Unixen of the era, the Ultrix follow-on
> variants were the least painful to deal with.  And the DEC MIPS-
> derived C compiler kicked ass when it came to spitting out
> grandmother-guilt-fed diagnostics :-)  (Was it DEC or MIPS who were
> responsible for the Spanish Inquisition error messages?  While noisy,
> the verbiage shortened a lot of other conversations :-)

Oh, Ultrix was a pain to administer, especially after DEC dropped support
for it.  I didn't especially like the compiler for it, either, though I
have a vague recollection that it had good diagnostics.  I don't think it
was fully ANSI C89 compliant, though, and other things in the OS left a
lot to be desired.  The monicker Uglix wasn't all together inappropriate.

That said, I still have a MIPS-based Ultrix machine somewhere in storage.
I should dig it out and see if it still boots.... (and if I still have
any data on it!).

	- Dan C.

Re: [9fans] bootalpha and the no valid stack error

[Lyndon Nerenberg]

> Yes, there was Ultrix on the 11.  I believe you can download it....

No thanks.

> Ultrix was a succession of different systems, starting on the PDP-11
> but moving to the VAX and eventually MIPS.

 From 7th Edition, 4.1, or ???  I don't have the poster handy.  What
I recall of Ultrix 1.0 said it was 4.2BSD plus stuff (on the 785 at
least).

> I know that CMU did a port of OSF/1 to MIPS, but I don't think it was
> generally available.  And I certainly believe that a lot of the
> development
> work was hosted on MIPS and possibly even targeted MIPS before the
> Alpha
> was ready, but I don't ever recall it being a commercial offering from
> DEC.  Even after the Alpha was released, if you bought a MIPS or
> VAX-based
> machine from DEC, the only Unix offered was Ultrix.
>
> Maybe if you were one of those special customers you could get them to
> give you OSF for MIPS, but I never ranked that high.  :-)

We were hit with VAX-based workstations, early MIPS-based
workstations, and the new Sun Sparc pizza boxes, all at the same
time.  It was very confusing.  It was also a lot more entertaining
than chasing the generic foo86 hardware dragons that live today.

> Oh, Ultrix was a pain to administer, especially after DEC dropped
> support
> for it.

How so?  Back then it was 4.2BSD + tools-to-be-ignored.  You didn't
actually use the 'admin tools', did you?!?  I fell in love with AIX
when I discovered the switch that made it convert the boot files from
SYSV to /etc/rc.*  As I admin machines with /bin/ed that same crowd
is still trying to hack XML into vi.

--lyndon

Re: [9fans] bootalpha and the no valid stack error

[Moritz Kiese]

On Wed, 26 Apr 2006, Dan Cross wrote:

[snip]

> I know that CMU did a port of OSF/1 to MIPS, but I don't think it was
> generally available.  And I certainly believe that a lot of the development
> work was hosted on MIPS and possibly even targeted MIPS before the Alpha

Sort of. First development kits (ADU) where a rather large Alpha box with
one of the bigger DECStations (5000 ?) as I/O frontend (connected via
Turbochannel IIRC). Lots of the OSF/1 development took place on these
boxen...

++mbk

Re: Re: [9fans] bootalpha and the no valid stack error

[Gabriel Diaz]

[-- Attachment #1: Type: text/plain, Size: 474 bytes --]

Hello

i think it is the last bootalpha available (i pull every day).

i put os_type to UNIX, but still hangs.


thanks you very much,

gabi

On 4/27/06, Benjamin Huntsman <BHuntsman@mail2.cu-portland.edu> wrote:
>
> Are you using the newest bootalphapc?  I was working on it a while back
> and noticed that it didn't get all the PALcode entry points right.  Also,
> you need to use the OSF/1 PALcode, not the VMS one.  You can change it by
> set os_type unix.
>
> -Ben
>
>

[-- Attachment #2: Type: text/html, Size: 819 bytes --]

RE: Re: [9fans] bootalpha and the no valid stack error

[Benjamin Huntsman]

[-- Attachment #1: Type: text/plain, Size: 247 bytes --]

Are you using the newest bootalphapc?  I was working on it a while back and noticed that it didn't get all the PALcode entry points right.  Also, you need to use the OSF/1 PALcode, not the VMS one.  You can change it by set os_type unix.

-Ben

[-- Attachment #2: winmail.dat --]
[-- Type: application/ms-tnef, Size: 2935 bytes --]

Re: Re: [9fans] bootalpha and the no valid stack error

[LiteStar numnums]

[-- Attachment #1: Type: text/plain, Size: 1842 bytes --]

You installed NetBSD from ARC? It really will only install from SRM...
There is an option from one of the menus (haven't used ARC in a while)
that will allow you to choose whether or not to boot SRM or ARC.
I can boot up a PWS600 here and start ARC if you need it...


On 4/27/06, Gabriel Diaz <gabidiaz@gmail.com> wrote:
>
> Hello
>
> I have OpenVMS 7.3 installed, can I suppose i have SRM?
>
> I installed yesterday NetBSD and it complains about a
> correctable proccesor fail. That claim comes from some
> interrupt.c related to the network card. I suppose Andrey
> know about this. Can be that related to the fact that bootalpha
> doesn't load or print any message?
>
>
> I will try some print statements, but have no time to dig too much.
>
> thanks all
>
> Gabi
>
>
> On 4/27/06, geoff@collyer.net <geoff@collyer.net> wrote:
> >
> > You may be right.  I had thought that the console monitor and the PAL
> > code were lashed together.  The whole thing is a mess.  I'd have
> > preferred no PAL code and 001 console monitors.  It's not obvious to
> > me that the whizzo VMS queue instructions are actually worth having.
> >
> >
>


--
Nietzsche's first step is to accept what he knows. Atheism for him goes
without saying and is "contructive and
radical". Nietzsche's supreme vocation, so he says, is to provoke a kind of
crisis and a final decision about the
problem of atheism. The world continues on its course at random and there is
nothing final about it. Thus God
is useless, since He wants nothing in particular. If he wanted something --
and here we recognize the traditional
forumlation of the problem of evil -- He would have to assume responsiblity
for "a sum total of pain and inconsistency
which would debase the entire value of being born."
-- Albert Camus, L'Homme révolté

[-- Attachment #2: Type: text/html, Size: 2775 bytes --]

Re: Re: [9fans] bootalpha and the no valid stack error

[LiteStar numnums]

[-- Attachment #1: Type: text/plain, Size: 1314 bytes --]

Actually, when you boot an alpha from the SRM, it will install & set the PAL
code for the specific OS it is booting (VMS or OSF/1). On really old
machines, such as my Multia and one of the Miatas, the ARC lifts off of SRM
(which is fixed in later releases of the firmware it seems).

On 4/26/06, geoff@collyer.net <geoff@collyer.net> wrote:
>
> You may be right.  I had thought that the console monitor and the PAL
> code were lashed together.  The whole thing is a mess.  I'd have
> preferred no PAL code and 001 console monitors.  It's not obvious to
> me that the whizzo VMS queue instructions are actually worth having.
>
>


--
Nietzsche's first step is to accept what he knows. Atheism for him goes
without saying and is "contructive and
radical". Nietzsche's supreme vocation, so he says, is to provoke a kind of
crisis and a final decision about the
problem of atheism. The world continues on its course at random and there is
nothing final about it. Thus God
is useless, since He wants nothing in particular. If he wanted something --
and here we recognize the traditional
forumlation of the problem of evil -- He would have to assume responsiblity
for "a sum total of pain and inconsistency
which would debase the entire value of being born."
-- Albert Camus, L'Homme révolté

[-- Attachment #2: Type: text/html, Size: 1717 bytes --]

Re: Re: [9fans] bootalpha and the no valid stack error

[Paweł Lasek]

On 4/27/06, Brian L. Stuart <blstuart@bellsouth.net> wrote:

> I've got kind of mixed opinions on the subject of the PALcode.  The
> multiple console monitors do seem a little pointless.  As near as
> I can tell, the main motivation behind multiple PALcodes was to ease
> porting VMS from the VAX and DigitalUNIX (or whatever it was called
> that week) from the MIPS.  One had a lot of built-in assumptions about
> the VAX MMU and the four VAX processor modes and the other assumed
> a lot about MIPS.  That naturally led to the usual CS narcotic.  "We'll
> just abstract it away with an interface layer."  In retrospect, it looks
> to be an unnecessary bit of complexity.  But at least it's vaguely interesting,
> unlike say the 7000 variations of how to specify which block you want
> from an IDE drive.

If I understand well, PAL code is used to change the behaviour of AXP
cpu, much like microcode - Linux PAL code is basically unchanged for
all versions of Alpha, different mostly in interrupt handling and such
things. Basically it can allow you to compile one plan9 kernel and
boot it on every version of Alpha, changing only the bootloader part
:)

Basically, treat it like a microcode - except that its standard AXP
assembly + 5 instructions.

Boot process is if I remember:

1. <board specific init>
2. SRM
3. load OS bootloader
4. bootloader setups PAL and all the low level things and prepares
environment for OS
5. kernel in control...

At least that's how it was supposed to work in MILO (Linux Alpha bootloader)

> BLS

--
Paul Lasek

Re: Re: [9fans] bootalpha and the no valid stack error

[andrey mirtchovski]

netbsd's alpha faq claims it can only be installed on SRM. i'm sorry
but i can't remember anymore how to get to the SRM prompt on boot.
there are various help sites available.

i never did solve my network card problem -- it was initialized at IRQ
0 and driven (however slowly) by the clock interrupt. i managed for a
while and then just gave up. others have ran alpha machines for much
longer, i suppose they can be of better help...

andrey

On 4/27/06, Gabriel Diaz <gabidiaz@gmail.com> wrote:
> Hello
>
> I have OpenVMS 7.3 installed, can I suppose i have SRM?

Re: Re: [9fans] bootalpha and the no valid stack error

[Gabriel Diaz]

[-- Attachment #1: Type: text/plain, Size: 763 bytes --]

Hello

I have OpenVMS 7.3 installed, can I suppose i have SRM?

I installed yesterday NetBSD and it complains about a
correctable proccesor fail. That claim comes from some
interrupt.c related to the network card. I suppose Andrey
know about this. Can be that related to the fact that bootalpha
doesn't load or print any message?


I will try some print statements, but have no time to dig too much.

thanks all

Gabi

On 4/27/06, geoff@collyer.net <geoff@collyer.net> wrote:
>
> You may be right.  I had thought that the console monitor and the PAL
> code were lashed together.  The whole thing is a mess.  I'd have
> preferred no PAL code and 001 console monitors.  It's not obvious to
> me that the whizzo VMS queue instructions are actually worth having.
>
>

[-- Attachment #2: Type: text/html, Size: 1166 bytes --]

Re: Re: [9fans] bootalpha and the no valid stack error

[Dan Cross]

On Wed, Apr 26, 2006 at 10:19:00PM -0400, Brian L.Stuart wrote:
> > You may be right.  I had thought that the console monitor and the PAL
> > code were lashed together.  The whole thing is a mess.  I'd have
> > preferred no PAL code and 001 console monitors.  It's not obvious to
> > me that the whizzo VMS queue instructions are actually worth having.
>
> I've got kind of mixed opinions on the subject of the PALcode.  The
> multiple console monitors do seem a little pointless.  As near as
> I can tell, the main motivation behind multiple PALcodes was to ease
> porting VMS from the VAX and DigitalUNIX (or whatever it was called
> that week) from the MIPS.  One had a lot of built-in assumptions about
> the VAX MMU and the four VAX processor modes and the other assumed
> a lot about MIPS.  That naturally led to the usual CS narcotic.  "We'll
> just abstract it away with an interface layer."  In retrospect, it looks
> to be an unnecessary bit of complexity.  But at least it's vaguely interesting,
> unlike say the 7000 variations of how to specify which block you want
> from an IDE drive.

The thing about DEC's Unix on the Alpha was that it never ran on either
the MIPS or VAX....  It was essentially OSF/1 with some DEC
customizations and a custom compiler.

At the time, I thought it was pretty nice.  It `felt' a lot more like
BSD, which was what I was used to coming from a SunOS 4/straight BSD
world.  It was certainly a better offering than Solaris.  The thing is,
DEC was the only company that bought into the koolaid of a unified Unix
to take on Sun/AT&T; the idea of a standard OS across DEC, HP, and IBM
hardware was nice and could have potentially forced Sun's hand to adopt
it, too, but since DEC was struggling and no one else took it
seriously, it just never happened.  Now we're stuck with Linux.

I suspect that you're right: the PALcode was to help ease porting of
VMS from the VAX to Alpha.  There was no such backwards compatibility
requirement for their Unix, so it's unclear why they had a separate
module for it, other than that perhaps the VMS version was too
complicated and/or made too many VMS-centric assumptions to fit the
Unix model neatly.

	- Dan C.

Re: Re: [9fans] bootalpha and the no valid stack error

[Brian L.Stuart]

> You may be right.  I had thought that the console monitor and the PAL
> code were lashed together.  The whole thing is a mess.  I'd have
> preferred no PAL code and 001 console monitors.  It's not obvious to
> me that the whizzo VMS queue instructions are actually worth having.

I've got kind of mixed opinions on the subject of the PALcode.  The
multiple console monitors do seem a little pointless.  As near as
I can tell, the main motivation behind multiple PALcodes was to ease
porting VMS from the VAX and DigitalUNIX (or whatever it was called
that week) from the MIPS.  One had a lot of built-in assumptions about
the VAX MMU and the four VAX processor modes and the other assumed
a lot about MIPS.  That naturally led to the usual CS narcotic.  "We'll
just abstract it away with an interface layer."  In retrospect, it looks
to be an unnecessary bit of complexity.  But at least it's vaguely interesting,
unlike say the 7000 variations of how to specify which block you want
from an IDE drive.

BLS

Re: Re: [9fans] bootalpha and the no valid stack error

[geoff]

You may be right.  I had thought that the console monitor and the PAL
code were lashed together.  The whole thing is a mess.  I'd have
preferred no PAL code and 001 console monitors.  It's not obvious to
me that the whizzo VMS queue instructions are actually worth having.

Re: Re: [9fans] bootalpha and the no valid stack error

[Brian L.Stuart]

> As I understand it, you need to have the Digital Unix PAL code (SRM)
> installed to run Plan 9.  I don't know if it can have SRM and the VMS
> PAL code (ARC?) installed simultaneously.

I certainly don't claim to be an expert on this, but my understanding
is a little different.  First, as far as I can tell ARC and AlphaBIOS do
the same thing, and I think ARC may be a predecessor or earlier
name for AlphaBIOS.  These and SRM are basically console monitors
much like OpenBoot on Sun (but without Forth).  ARC/AlphaBIOS
is for WindowsNT and SRM is for UNIX and OpenVMS.  The PALcode
is a separate thing loaded by the boot loader usually.  I can't find
a reference to tell me for sure, but I think that OpenVMS and WinNT
use the same PALcode.  At least I wouldn't be surprised since
they both use things like the MMU pretty much the same way.
As to whether all these components can be on a machine
simultaneously, it depends on the actual machine.  As far as I
can tell, it depends on the size of the flash more than anything.
My DS5300 has it all there, even though officially it's only supposed
to run WinNT.  But don't tell my NetBSD installation.  One of these
days, though, I'd love to see if I can get Plan 9 on it.

BLS