[TUHS] Re: 4.3 BSD version in the Unix Archive

[TUHS] Re: 4.3 BSD version in the Unix Archive

[Gregg C Levine]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 6282 bytes --]

Hello (again) from Gregg C Levine
The RPM file provided didn't build correctly. Remember I did say that
I use Slackware. I don't have a running instance of Red Hat here.
However, I was able to build a copy of SIMH straight from the current
source code.  One thing does get to me, though. This creation moves a
lot slower on SIMH/VAX, then NetBSD, which worked well enough. I'm
guessing that this is indeed a populated pack you have here, but do
you have any idea as to why it's practically moving slower then a
tortoise? This is running on a Pentium 100.

As to your question, yes they are working on it, moving through the
2.4.2x series of kernels. It isn't pretty, but it is working. If you
want to join the list to offer complements, or comments, or just lurk,
go to their website, and tell the list-manager that I sent you. He'll
give you a good seat.
-------------------
Gregg C Levine hansolofalcon at worldnet.att.net
------------------------------------------------------------
"The Force will be with you...Always." Obi-Wan Kenobi
"Use the Force, Luke."  Obi-Wan Kenobi
(This company dedicates this E-Mail to General Obi-Wan Kenobi )
(This company dedicates this E-Mail to Master Yoda )



> -----Original Message-----
> From: Markus Weber [mailto:jmbw at nather.com]
> Sent: Tuesday, September 30, 2003 3:40 PM
> To: Gregg C Levine
> Subject: RE: [TUHS] Re: 4.3 BSD version in the Unix Archive
> 
> Hi Gregg!
> 
> The src.rpm was built on Redhat 7.3, but you should be able to
rebuild in on
> a more recent version of Redhat. There is, however, no reason not to
do a
> straight recompile of the simh sources. I simply prefer to package
> everything and thought I may save somebody else the bother.
> 
> My site runs Postnuke, which uses cookies internally. Unless you
want to
> create an account and login, I doubt that it matters one way or the
other if
> you reject the cookies or not.
> 
> I would guess that the primary objective of simh/vax is to run
OpenVMS,
> which it does just fine for me. If you wish, I can dig up a URL for
a
> beautiful installation-on-simh write-up. You can really follow the
standard
> instructions to the letter once simh is set up properly.
> 
> It looks like simh can run most, if not all, BSD's. I'm happy with
> Quasijaro - it has a certain sentimental value to observe 4.3BSD in
its
> native habitat. Quasijaro has issues, but they can be worked around.
NetBSD
> 1.5.2 works with a very minor install-time glitch. OpenBSD is the
one I
> still haven't figured out; there's an odd fatal error when mounting
the root
> filesystem. Or rather, there's something very odd about the in-core
> disklabel. Come to think of it, there is a common theme of problems
relating
> to disk labels and first-time access of disks. By and large, it's
not clear
> (to me) in any of these cases if the emulation is broken or simply
exposes
> flaws in the OS.
> 
> So the Linux/Vax project is still alive?
> 
> Cheers
> -Markus
> 
> -----Original Message-----
> From: Gregg C Levine [mailto:hansolofalcon at worldnet.att.net]
> Sent: Tuesday, September 30, 2003 2:02 PM
> To: 'Markus Weber'; tuhs at minnie.tuhs.org
> Subject: RE: [TUHS] Re: 4.3 BSD version in the Unix Archive
> 
> 
> Hello (again) from Gregg C Levine
> Just for the sake of argument, Markus, what was your build
environment
> for your SRC RPM version of SIMH? Personally I use Slackware Linux
> here, and the source code files straight from Bob's site. Also, that
> site kept wanting to set a cookie on my machine here. Is that a
normal
> process?
> 
> Incidentally, the folks building Linux for the VAX, also use
SIMH/VAX
> for testing, and sometimes even they have problems. So the comment
> regarding this product, and the VAX simulation is valid, just needs
to
> be further tested.
> 
> I, myself, have used it, to boot either VAX VMS, (Didn’t workout how
> to install it on blank disk though.). or a relative of what we
> discuss, as well.
> -------------------
> Gregg C Levine hansolofalcon at worldnet.att.net
> ------------------------------------------------------------
> "The Force will be with you...Always." Obi-Wan Kenobi
> "Use the Force, Luke."  Obi-Wan Kenobi
> (This company dedicates this E-Mail to General Obi-Wan Kenobi )
> (This company dedicates this E-Mail to Master Yoda )
> 
> 
> 
> > -----Original Message-----
> > From: tuhs-bounces at minnie.tuhs.org
> [mailto:tuhs-bounces at minnie.tuhs.org] On
> > Behalf Of Markus Weber
> > Sent: Tuesday, September 30, 2003 9:05 AM
> > To: Joseph F. Young; tuhs at minnie.tuhs.org
> > Subject: RE: [TUHS] Re: 4.3 BSD version in the Unix Archive
> >
> > Thanks to your help it works. Please see
> >
> >
>
http://www.itsecuritygeek.com/modules.php?op=modload&name=News&file=ar
> ticl
> > e&
> > sid=22
> >
> > for an annotated installation transcript. You'll find a
> pre-installed disk
> > image in the site's Download section.
> >
> >
> >
> > -----Original Message-----
> > From: Joseph F. Young [mailto:jy99 at swbell.net]
> > Sent: Monday, September 29, 2003 1:50 PM
> > To: tuhs at minnie.tuhs.org
> > Cc: jy99 at swbell.net
> > Subject: [TUHS] Re: 4.3 BSD version in the Unix Archive
> >
> >
> >  From my experience, you need to have a recent release of SIMH in
> order to
> > run 4.3BSD.  A few months ago, I managed to build "working"
> Quasijarus and
> > Reno disk images (Quasijarus appears to work fine for me, but Reno
> is as
> > buggy as I remember it being on real hardware).  I had to use
Ultrix
> and
> > Netbsd to do the bootstrap/install; I could not get the tape boot
to
> work
> > at all.
> >
> >
> > ---
> > Outgoing mail is certified Virus Free.
> > Checked by AVG anti-virus system (http://www.grisoft.com).
> > Version: 6.0.520 / Virus Database: 318 - Release Date: 9/18/2003
> >
> > _______________________________________________
> > TUHS mailing list
> > TUHS at minnie.tuhs.org
> > http://minnie.tuhs.org/mailman/listinfo/tuhs
> 
> 
> 
> ---
> Incoming mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.520 / Virus Database: 318 - Release Date: 9/18/2003
> 
> ---
> Outgoing mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.520 / Virus Database: 318 - Release Date: 9/18/2003

[TUHS] Re: 4.3 BSD version in the Unix Archive

[Markus Weber]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 2195 bytes --]

I wasn't even aware that Slackware had rpm... If you send me a typescript of
the build attempt, I can have a look. The version of rpm itself would also
help.

If you mean that the pre-installed system is slow, all I can say is that in
my environment all the BSDs perform well enough for light interactive use,
but running a compiler is not fun. OpenBSD/Vax takes about four hours to
recompile the generic kernel on a dual-1Ghz PIII. BTW, does NetBSD
outperform Quasijarus on real hardware? I suggest, however, to take the
performance analyis offlist and post the results at a later date.

-----Original Message-----
From: Gregg C Levine [mailto:hansolofalcon@worldnet.att.net]
Sent: Tuesday, September 30, 2003 8:01 PM
To: 'Markus Weber'
Cc: tuhs at minnie.tuhs.org
Subject: RE: [TUHS] Re: 4.3 BSD version in the Unix Archive


Hello (again) from Gregg C Levine
The RPM file provided didn't build correctly. Remember I did say that
I use Slackware. I don't have a running instance of Red Hat here.
However, I was able to build a copy of SIMH straight from the current
source code.  One thing does get to me, though. This creation moves a
lot slower on SIMH/VAX, then NetBSD, which worked well enough. I'm
guessing that this is indeed a populated pack you have here, but do
you have any idea as to why it's practically moving slower then a
tortoise? This is running on a Pentium 100.

As to your question, yes they are working on it, moving through the
2.4.2x series of kernels. It isn't pretty, but it is working. If you
want to join the list to offer complements, or comments, or just lurk,
go to their website, and tell the list-manager that I sent you. He'll
give you a good seat.
-------------------
Gregg C Levine hansolofalcon at worldnet.att.net
------------------------------------------------------------
"The Force will be with you...Always." Obi-Wan Kenobi
"Use the Force, Luke."  Obi-Wan Kenobi
(This company dedicates this E-Mail to General Obi-Wan Kenobi )
(This company dedicates this E-Mail to Master Yoda )

---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.520 / Virus Database: 318 - Release Date: 9/18/2003

[TUHS] Re: 4.3 BSD version in the Unix Archive

[Robert Brockway]

On Tue, 30 Sep 2003, Markus Weber wrote:

> I wasn't even aware that Slackware had rpm... If you send me a typescript of
> the build attempt, I can have a look. The version of rpm itself would also
> help.

If worse comes to worse the Linux utility alien will convert debs, rpms
and .tgz files into each other.  Thanks to the Linux Standards Base the
resulting file is _usually_ pretty ok for installation (a bit of munging
may be required, add salt to taste).

Rob

-- 
Robert Brockway B.Sc. email: robert at timetraveller.org, zzbrock at uqconnect.net
Linux counter project ID #16440 (http://counter.li.org)
"The earth is but one country and mankind its citizens" -Baha'u'llah

[TUHS] Re: 4.3 BSD version in the Unix Archive

[Jochen Kunz]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1003 bytes --]

On 2003.10.01 04:46 Markus Weber wrote:

> BTW, does NetBSD outperform Quasijarus on real hardware? 
No. NetBSD has ELF and shared libs. The later saves space but involves
some overhead. When it comes to compiling NetBSD looses. NetBSD uses
GCC. GCC is _very_ slow. GCC 3.x is even slower. 4.3BSD-Tahoe /
Quasijarus uses pcc that is _much_ faster in compiling. I did a "make
world" of 4.3BSD-Tahoe on a KA655 based MicroVAX III and it took a day
or so. I would not even try this with NetBSD (as I have much faster NVAX
machines for this).

The thing with NetBSD is that you get a modern *ix like OS complete with
ssh etc. and features like mmap(2) that 4.3BSD-Tahoe doesn't have. If
you are running a VAX emulator on PeeCee hardware you can use the
exelent cross compile features of NetBSD to cross build kernels on your
PeeCee. (Or Alpha, PowerPC, UltraSparc, ...) You don't even need to run
NetBSD on the compile host. 
-- 


tschüß,
       Jochen

Homepage: http://www.unixag-kl.fh-kl.de/~jkunz/

[TUHS] Re: 4.3 BSD version in the Unix Archive

[Michael Sokolov]

Jochen Kunz <jkunz at unixag-kl.fh-kl.de> wrote:

> The thing with NetBSD is that you get a modern *ix like OS complete with
> ssh etc. and features like mmap(2) that 4.3BSD-Tahoe doesn't have.

Just a nitpick, but ssh is available for 4.3BSD-Quasijarus, look for it in:

ifctfvax.Harhan.ORG:/pub/unix/apps/

But yes, NetBSD is a "modern *ix", not UNIX, which is why I can't stand it.  My
love for UNIX as opposed to "modern *ix"'s is what made me create and run
4.3BSD-Quasijarus.  Since I use it as my sole and only OS for my real world
computing needs, not a hobby, I have everything for it that one needs in
everyday life, including ssh, httpd, Russian support, PostScript support, PPP
support, etc.  But it's still pure 4.3BSD as all of the latter are just user
applications.

MS

[TUHS] Re: 4.3 BSD version in the Unix Archive

[Robert Brockway]

On Wed, 1 Oct 2003, Michael Sokolov wrote:

> But yes, NetBSD is a "modern *ix", not UNIX, which is why I can't stand it.  My

The last thing I want to do is start a "flavour war" but I really am
interested in why you don't like "modern *ix" as opposed to say 4.3BSD
(which you mention later).

For reference I'm probably a youngin relative to alot on this list.  My
first unix was SunOS 4.1.3 in 1992 but much of my experience has been as
an admin for Solaris and Linux.  I've had smaller amounts of experience
with Free/NetBSD, SCO Openserver and AIX.

Rob

-- 
Robert Brockway B.Sc. email: robert at timetraveller.org, zzbrock at uqconnect.net
Linux counter project ID #16440 (http://counter.li.org)
"The earth is but one country and mankind its citizens" -Baha'u'llah

[TUHS] Re: 4.3 BSD version in the Unix Archive

[Joseph F. Young]

 From my experience, you need to have a recent release of SIMH in order to 
run 4.3BSD.  A few months ago, I managed to build "working" Quasijarus and 
Reno disk images (Quasijarus appears to work fine for me, but Reno is as 
buggy as I remember it being on real hardware).  I had to use Ultrix and 
Netbsd to do the bootstrap/install; I could not get the tape boot to work 
at all.

Here's my SIMH ini file for Quasijarus:
---------------------------------------------------------------------------------------------------------------
set cpu 32m
set cpu conhalt
set tto 7b
set tti 7b
at rom ka655.bin
at nvr nvr.bin
set rq0 ra82
at rq0 bsd43q_ra0.dsk
set rq1 ra82
at rq1 /dev/null
set rq2 ra82
at rq2 /dev/null
set rq3 ra82
at rq3 /dev/null
set tq0 locked
at tq0 BSD43Q_TAPE.tap
set dz 7b
at -m dz0 4501
at xq eth0
set lpt dis
set ts dis
set rl dis
boot cpu
--------------------------------------------------------------------------------------------------------------
Here's an example boot log for Quasijarus under SIMH:
 >>>boot dua0
(BOOT/R5:0 DUA0



   2..
-DUA0
   1..0..

loading boot

Boot
: /vmunix
327204+103384+130352 start 0x23a8
4.3 BSD Quasijarus UNIX #0: Sat Oct  2 22:15:38 CDT 1999
     msokolov at luthien:/usr/src/sys/GENERIC
real mem  = 33521664
SYSPTSIZE limits number of buffers to 80
avail mem = 31697920
using 80 buffers containing 655360 bytes of memory
MicroVAX 3000, ucode rev 6
tmscp0 at uba0 csr 174500 vec 774, ipl 15
tms0 at tmscp0 slave 0
tms1 at tmscp0 slave 1
uda0 at uba0 csr 172150 vec 770, ipl 15
uda0: version 3 model 3
uda0: DMA burst size set to 4
ra0 at uda0 slave 0: RA82, size = 1216665 sectors
ra1 at uda0 slave 1: no disk label: ra82, size = 1216665 sectors
ra2 at uda0 slave 2: no disk label: ra82, size = 1216665 sectors
ra3 at uda0 slave 3: no disk label: ra82, size = 1216665 sectors
dz0 at uba0 csr 160100 vec 300, ipl 15
dz1 at uba0 csr 160110 vec 310, ipl 15
dz2 at uba0 csr 160120 vec 320, ipl 15
dz3 at uba0 csr 160130 vec 330, ipl 15
qe0 at uba0 csr 174440 vec 764, ipl 15
qe0: delqa, hardware address 08:00:2b:aa:bb:cc
Changing root device to ra0a
WARNING: clock gained 91 days -- CHECK AND RESET THE DATE!
Automatic reboot in progress...
Mon Sep 29 08:38:03 CDT 2003
/dev/ra0a: 669 files, 5242 used, 25429 free (37 frags, 3174 blocks, 0.1% 
fragmentation)
/dev/rra0g: 12267 files, 71408 used, 376714 free (1634 frags, 46885 blocks, 
0.4% fragmentation)
Mon Sep 29 08:43:51 CDT 2003
checking quotas: done.
starting system logger
preserving editor files
clearing /tmp

standard daemons: update cron accounting.
starting network daemons: routed named inetd printer.
Sep 29 08:44:28 simh named[66]: /etc/named.boot: No such file or directory
starting local daemons: sendmail.
Mon Sep 29 08:44:52 CDT 2003


4.3 BSD UNIX (simh.localdomain) (console)

login:
-----------------------------------------------------------------------------------------------------------
Cheers,
----
Joseph Young
jy99 at swbell.net

[TUHS] Re: 4.3 BSD version in the Unix Archive

[Markus Weber]

Thanks to your help it works. Please see

http://www.itsecuritygeek.com/modules.php?op=modload&name=News&file=article&
sid=22

for an annotated installation transcript. You'll find a pre-installed disk
image in the site's Download section.



-----Original Message-----
From: Joseph F. Young [mailto:jy99@swbell.net]
Sent: Monday, September 29, 2003 1:50 PM
To: tuhs at minnie.tuhs.org
Cc: jy99 at swbell.net
Subject: [TUHS] Re: 4.3 BSD version in the Unix Archive


 From my experience, you need to have a recent release of SIMH in order to
run 4.3BSD.  A few months ago, I managed to build "working" Quasijarus and
Reno disk images (Quasijarus appears to work fine for me, but Reno is as
buggy as I remember it being on real hardware).  I had to use Ultrix and
Netbsd to do the bootstrap/install; I could not get the tape boot to work
at all.


---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.520 / Virus Database: 318 - Release Date: 9/18/2003

[TUHS] Re: 4.3 BSD version in the Unix Archive

[Gregg C Levine]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 2571 bytes --]

Hello (again) from Gregg C Levine
Just for the sake of argument, Markus, what was your build environment
for your SRC RPM version of SIMH? Personally I use Slackware Linux
here, and the source code files straight from Bob's site. Also, that
site kept wanting to set a cookie on my machine here. Is that a normal
process? 

Incidentally, the folks building Linux for the VAX, also use SIMH/VAX
for testing, and sometimes even they have problems. So the comment
regarding this product, and the VAX simulation is valid, just needs to
be further tested.

I, myself, have used it, to boot either VAX VMS, (Didn’t workout how
to install it on blank disk though.). or a relative of what we
discuss, as well.  
-------------------
Gregg C Levine hansolofalcon at worldnet.att.net
------------------------------------------------------------
"The Force will be with you...Always." Obi-Wan Kenobi
"Use the Force, Luke."  Obi-Wan Kenobi
(This company dedicates this E-Mail to General Obi-Wan Kenobi )
(This company dedicates this E-Mail to Master Yoda )



> -----Original Message-----
> From: tuhs-bounces at minnie.tuhs.org
[mailto:tuhs-bounces at minnie.tuhs.org] On
> Behalf Of Markus Weber
> Sent: Tuesday, September 30, 2003 9:05 AM
> To: Joseph F. Young; tuhs at minnie.tuhs.org
> Subject: RE: [TUHS] Re: 4.3 BSD version in the Unix Archive
> 
> Thanks to your help it works. Please see
> 
>
http://www.itsecuritygeek.com/modules.php?op=modload&name=News&file=ar
ticl
> e&
> sid=22
> 
> for an annotated installation transcript. You'll find a
pre-installed disk
> image in the site's Download section.
> 
> 
> 
> -----Original Message-----
> From: Joseph F. Young [mailto:jy99 at swbell.net]
> Sent: Monday, September 29, 2003 1:50 PM
> To: tuhs at minnie.tuhs.org
> Cc: jy99 at swbell.net
> Subject: [TUHS] Re: 4.3 BSD version in the Unix Archive
> 
> 
>  From my experience, you need to have a recent release of SIMH in
order to
> run 4.3BSD.  A few months ago, I managed to build "working"
Quasijarus and
> Reno disk images (Quasijarus appears to work fine for me, but Reno
is as
> buggy as I remember it being on real hardware).  I had to use Ultrix
and
> Netbsd to do the bootstrap/install; I could not get the tape boot to
work
> at all.
> 
> 
> ---
> Outgoing mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.520 / Virus Database: 318 - Release Date: 9/18/2003
> 
> _______________________________________________
> TUHS mailing list
> TUHS at minnie.tuhs.org
> http://minnie.tuhs.org/mailman/listinfo/tuhs

[TUHS] Re: 4.3 BSD version in the Unix Archive

[Michael Sokolov]

Markus Weber <jmbw at nather.com> wrote:

> Stand/copy from tms(0,1) to ra(0,1) fails with a 'disk unlabeled' diagnostic
> on a cold disk image.

It should print the unlabeled diagnostic, but then proceed rather than fail.  I
just double-checked the standalone driver source, and it does provide a default
label for RA82, as long as the MSCP controller actually returns RA82 as the MSCP
disk ID.  Does it give you an error message about disk type (some hex humber)
not supported?  If so, SIMH's MSCP emulation must be lacking in quality.

> If only ra0 is
> configured for simh, it takes the kernel a long time to decide that ra1,
> ra2, and ra3 are offline. Having said that, I do not know how long this
> takes on actual hardware.

On real HW it takes no noticeable time.

> the kernel (and
> not simh) subsequently segfaults (trap type 8, code = c0000200, pc =
> 8003fe16).

Well, obviously this doesn't happen on real HW, since on real HW it works (I'm
typing this message on a MicroVAX 3+ running 4.3BSD-Quasijarus).

But I will grant the possibility that the kernel is not w/o fault either in that
perhaps it's going south (dereferencing a garbage pointer and crashing) when the
MSCP controller (in this case SIMH's poor emulation) is doing something it
doesn't expect.  If this is so, it should be fixed, since even w/o emulators
(which I refuse to support on principle) real HW can be broken and return
garbage on reg reads, and the kernel must handle it gracefully.  I'll look into
it.

MS

[TUHS] Re: 4.3 BSD version in the Unix Archive

[Markus Weber]

I found out a bit more.

For any non-OpenVMS OS I tried on simh/vax (read: different flavors of BSD),
disk initialization is a common problem area. NetBSD glitches on mounting
the installation CD/disk, Quasijarus has a hard time deciding which disks
are online and which are not, and finally OpenBSD installs without problem,
but fails when mounting the root filesystem when it boots off ra0a. I can
however boot off the install-time kernel on the floppy disk and access the
partitions on ra0 just fine.

It became apparent that OpenBSD's in-core disk label defaulted to data
matching the c partition where the a partition should have been. After more
digging, it seemed that ra_putonline was called, but rx_putonline must have
failed somehow. To add insult to injury, after adding a few printf
statements to rx_putonline the OpenBSD sort-of-GENERIC kernel now boots.

Here's the relevant block of code:

rx_putonline(rx)
        struct rx_softc *rx;
{
        struct  mscp *mp;
        struct  mscp_softc *mi = (struct mscp_softc *)rx->ra_dev.dv_parent;
        volatile int i;

printf("entered rx_putonline\n");
        rx->ra_state = DK_CLOSED;
        mp = mscp_getcp(mi, MSCP_WAIT);
        mp->mscp_opcode = M_OP_ONLINE;
        mp->mscp_unit = rx->ra_hwunit;
        mp->mscp_cmdref = 1;
        *mp->mscp_addr |= MSCP_OWN | MSCP_INT;

        /* Poll away */
        i = bus_space_read_2(mi->mi_iot, mi->mi_iph, 0);
        if (tsleep(&rx->ra_dev.dv_unit, PRIBIO, "rxonline", 100*100))
                rx->ra_state = DK_CLOSED;
printf("rx->ra_state = %d\n",rx->ra_state);
        if (rx->ra_state == DK_CLOSED)
                return MSCP_FAILED;
printf("rx_putonline returns DONE\n");
        return MSCP_DONE;
}

My guess is that the very first printf upsets the emulation timing just
enough to make things work. I'd have to read more code and RQDX3
documentation to take this further, but maybe this is enough to go on to
make it obvious to somebody else why Quasijarus on simh exhibits these
awfully long disk probe timeouts. I hope that once this problem has been
narrowed down a bit more, a fix to simh/vax will make all BSDs happy.

-----Original Message-----
From: Michael Sokolov [mailto:msokolov@ivan.Harhan.ORG]
Sent: Friday, September 26, 2003 3:48 PM
To: jmbw at nather.com; tuhs at tuhs.org
Subject: RE: [TUHS] Re: 4.3 BSD version in the Unix Archive


[...] If so, SIMH's MSCP emulation must be lacking in quality.

[...]But I will grant the possibility that the kernel is not w/o fault
either in that
perhaps it's going south (dereferencing a garbage pointer and crashing) when
the
MSCP controller (in this case SIMH's poor emulation) is doing something it
doesn't expect.
---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.520 / Virus Database: 318 - Release Date: 9/18/2003

[TUHS] Re: 4.3 BSD version in the Unix Archive

[Kenneth Stailey]

--- Markus Weber <jmbw at nather.com> wrote:
> I found out a bit more.
> 
> For any non-OpenVMS OS I tried on simh/vax (read: different flavors of BSD),
> disk initialization is a common problem area. NetBSD glitches on mounting
> the installation CD/disk, Quasijarus has a hard time deciding which disks
> are online and which are not, and finally OpenBSD installs without problem,
> but fails when mounting the root filesystem when it boots off ra0a. I can
> however boot off the install-time kernel on the floppy disk and access the
> partitions on ra0 just fine.
> 
> It became apparent that OpenBSD's in-core disk label defaulted to data
> matching the c partition where the a partition should have been. After more
> digging, it seemed that ra_putonline was called, but rx_putonline must have
> failed somehow. To add insult to injury, after adding a few printf
> statements to rx_putonline the OpenBSD sort-of-GENERIC kernel now boots.
> 
> Here's the relevant block of code:
> 
> rx_putonline(rx)
>         struct rx_softc *rx;
> {
>         struct  mscp *mp;
>         struct  mscp_softc *mi = (struct mscp_softc *)rx->ra_dev.dv_parent;
>         volatile int i;
> 
> printf("entered rx_putonline\n");
>         rx->ra_state = DK_CLOSED;
>         mp = mscp_getcp(mi, MSCP_WAIT);
>         mp->mscp_opcode = M_OP_ONLINE;
>         mp->mscp_unit = rx->ra_hwunit;
>         mp->mscp_cmdref = 1;
>         *mp->mscp_addr |= MSCP_OWN | MSCP_INT;
> 
>         /* Poll away */
>         i = bus_space_read_2(mi->mi_iot, mi->mi_iph, 0);
>         if (tsleep(&rx->ra_dev.dv_unit, PRIBIO, "rxonline", 100*100))
>                 rx->ra_state = DK_CLOSED;
> printf("rx->ra_state = %d\n",rx->ra_state);
>         if (rx->ra_state == DK_CLOSED)
>                 return MSCP_FAILED;
> printf("rx_putonline returns DONE\n");
>         return MSCP_DONE;
> }
> 
> My guess is that the very first printf upsets the emulation timing just
> enough to make things work. I'd have to read more code and RQDX3
> documentation to take this further, but maybe this is enough to go on to
> make it obvious to somebody else why Quasijarus on simh exhibits these
> awfully long disk probe timeouts. I hope that once this problem has been
> narrowed down a bit more, a fix to simh/vax will make all BSDs happy.

My first guess would be that there should be more "volatile" keywords used.  I
think the printf()s cause the compiler to not discard some varibles.  Does it
work without the printf()s if you don't use "-O" (or "-O2", etc) optimization? 
I think gcc does not elminitate varibles ever when the code is not optimized.

> -----Original Message-----
> From: Michael Sokolov [mailto:msokolov at ivan.Harhan.ORG]
> Sent: Friday, September 26, 2003 3:48 PM
> To: jmbw at nather.com; tuhs at tuhs.org
> Subject: RE: [TUHS] Re: 4.3 BSD version in the Unix Archive
> 
> 
> [...] If so, SIMH's MSCP emulation must be lacking in quality.
> 
> [...]But I will grant the possibility that the kernel is not w/o fault
> either in that
> perhaps it's going south (dereferencing a garbage pointer and crashing) when
> the
> MSCP controller (in this case SIMH's poor emulation) is doing something it
> doesn't expect.
> ---
> Outgoing mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.520 / Virus Database: 318 - Release Date: 9/18/2003
> 
> _______________________________________________
> TUHS mailing list
> TUHS at minnie.tuhs.org
> http://minnie.tuhs.org/mailman/listinfo/tuhs


__________________________________
Do you Yahoo!?
The New Yahoo! Shopping - with improved product search
http://shopping.yahoo.com

[TUHS] Re: 4.3 BSD version in the Unix Archive

[Markus Weber]

>From: Kenneth Stailey [mailto:kstailey at yahoo.com]

>My first guess would be that there should be more "volatile" keywords used.
I
>think the printf()s cause the compiler to not discard some varibles.  Does
it
>work without the printf()s if you don't use "-O" (or "-O2", etc)
optimization?
>I think gcc does not elminitate varibles ever when the code is not
optimized.

Doesn't seem to make a difference, but I'll double-check.

I could be wrong, but I believe the printf will return as soon as the
physical output of the message has started - if so, it can very conceivably
affect the "real-time" behaviour of the code interfacing with the RQDX3.
Just a guess.
---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.520 / Virus Database: 318 - Release Date: 9/18/2003

[TUHS] Re: 4.3 BSD version in the Unix Archive

[Michael Sokolov]

Markus Weber <jmbw at nather.com> wrote:

> Clearly, stand boots. I'm hopeful that 4.3-Quasijarus0a will indeed run on
> simh.

Just be warned that I have heard from others that the kernel won't boot on SIMH
due to sucky emulation.

> For some follow-up questions... Standalone format supports hp and up disks,
> but simh only emulates RL and RQ controllers - which means that the Ultrix
> utilities to label a cold disk must be ported or recreated to the simh host
> platform. Of course, it may well be that applying a hex editor to the disk
> image will suffice.

Arghh, why don't people just f-ing get it!  Standalone format is what is known
as low-level format in the PeeCee world (magnetic level stuff, certainly not
needed on an emulator), and it has absolutely nothing to do with disk labeling.
4.3BSD-Quasijarus0[a] does not have standalone disklabel, but RA82 is known to
the compiled-in tables and thus will work unlabeled.  Also Ultrix' disk labels
are not useful for 4.3BSD-Tahoe/Quasijarus, since the labels are stored in a
different place in a different format.  (Berkeley and DEC developed pack labels
independently.  BSD stores them in the boot block, Ultrix stores them in the
root fs superblock, which has the disadvantage of requiring you to create some
dummy root fs before you can write a label, even if you want a root fs of
different size.)

> Am I correct that Quasijarus supports RA82 disks (using the ra type) and the
> TK50 tape (as a tms)?

Yes.

MS

[TUHS] Re: 4.3 BSD version in the Unix Archive

[Markus Weber]

Stand/copy from tms(0,1) to ra(0,1) fails with a 'disk unlabeled' diagnostic
on a cold disk image. It turns out that NetBSD's disklabel(8) will create a
template disk image file that makes stand/copy happy, although I haven't
verified that the label is all that it should be.

The boot of the miniroot's GENERIC kernel does indeed fail. If only ra0 is
configured for simh, it takes the kernel a long time to decide that ra1,
ra2, and ra3 are offline. Having said that, I do not know how long this
takes on actual hardware. This part can be speeded up by supplying empty
image files for the remaing three disks, but in either case the kernel (and
not simh) subsequently segfaults (trap type 8, code = c0000200, pc =
8003fe16).

I haven't yet investigated what this diagnostic signifies. There are two
likely scenarios, the kernel either crashes late in the ra driver
initialization or while probing for the next device in line. At this point
I'm not prepared to argue whether this is primarily an emulation error or a
kernel bug.

Owing to the lack of Vax hardware or at least a compatible environment, you
have me at a disadvantage. Which function does pc = 8003fe16 correspond to?

Markus

-----Original Message-----
From: Michael Sokolov [mailto:msokolov@ivan.Harhan.ORG]
Sent: Friday, September 26, 2003 1:36 PM
To: jmbw at nather.com; tuhs at tuhs.org
Subject: RE: [TUHS] Re: 4.3 BSD version in the Unix Archive


Markus Weber <jmbw at nather.com> wrote:

> Clearly, stand boots. I'm hopeful that 4.3-Quasijarus0a will indeed run on
> simh.

Just be warned that I have heard from others that the kernel won't boot on
SIMH
due to sucky emulation.

> For some follow-up questions... Standalone format supports hp and up
disks,
> but simh only emulates RL and RQ controllers - which means that the Ultrix
> utilities to label a cold disk must be ported or recreated to the simh
host
> platform. Of course, it may well be that applying a hex editor to the disk
> image will suffice.

Arghh, why don't people just f-ing get it!  Standalone format is what is
known
as low-level format in the PeeCee world (magnetic level stuff, certainly not
needed on an emulator), and it has absolutely nothing to do with disk
labeling.
4.3BSD-Quasijarus0[a] does not have standalone disklabel, but RA82 is known
to
the compiled-in tables and thus will work unlabeled.  Also Ultrix' disk
labels
are not useful for 4.3BSD-Tahoe/Quasijarus, since the labels are stored in a
different place in a different format.  (Berkeley and DEC developed pack
labels
independently.  BSD stores them in the boot block, Ultrix stores them in the
root fs superblock, which has the disadvantage of requiring you to create
some
dummy root fs before you can write a label, even if you want a root fs of
different size.)

> Am I correct that Quasijarus supports RA82 disks (using the ra type) and
the
> TK50 tape (as a tms)?

Yes.

MS


---
Incoming mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.520 / Virus Database: 318 - Release Date: 9/18/2003

---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.520 / Virus Database: 318 - Release Date: 9/18/2003

[TUHS] Re: 4.3 BSD version in the Unix Archive

[Michael Sokolov]

Warren Toomey <wkt at tuhs.org> wrote:

> The 4.3BSD-Quasijarus dists were compressed with a compression format
> that's not compatible with either gzip nor old compress(1). Michael
> Sokolov should be able to send in some notes on the tools required.

Yes, for political reasons I needed to make the 4.3BSD-Quasijarus compressor a
version of compress, not gzip (can't have any GNU), but I wanted to have the
higher compression ratio of deflation, so I created a new version of compress
that supports deflation in addition to the original LZW algorith.  You can find
the new compress in components/compress.tar either on my FTP site or in Warren's
archive in the 4BSD area.

> > Is there a set of 4.3BSD-Tahoe Vax distribution files that's complete?
>
> Not that I know of.

"4.3BSD-Tahoe Vax distribution" is an oxymoron.  Berkeley released the Tahoe
tape with Tahoe binaries, no VAX binaries.  I was the one who compiled the Tahoe
source for the VAX, and the result was 4.3BSD-Quasijarus0.

OTOH, you may be referring to the fact that the Tahoe distribution in the
archive is broken.  Yes, it is.  Unfortunately there was an unrecoverable tape
read error.

> > Finally, do you know of any 4.3BSD version that will install and run on
> > simh? Quasijarus and Tahoe of the Unix Archive are broken for me and Reno
> > doesn't boot stand. Admittedly, I haven't tried the vanilla version yet.
>
> Not that I know of. I'll cc this to the TUHS list and see if any other
> people know the answer.

This has come up time and again.  SIMH's emulation of VAX is too poor.  VAX is
not an easy architecture to emulate.

MS

[TUHS] Re: 4.3 BSD version in the Unix Archive

[Markus Weber]

Thanks for the help.

So far I managed to uncompress the files with the custom version of compress
on NetBSD/simh. I then created a .tap file according to the reference in
FORMAT.

I'm using the following config file for simh 3.0.1:

load -r ka655.bin
set rq0 ra82
at rq0 ra82.dsk
set tq tk50
at tq0 quasijarus0a-dist.tap
boot cpu

Here's a transcript of the very first attempt of booting off the tape:

$ ./vax 43cfg

VAX simulator V3.0-1


KA655-B V5.3, VMB 2.7
Performing normal system tests.
40..39..38..37..36..35..34..33..32..31..30..29..28..27..26..25..
24..23..22..21..20..19..18..17..16..15..14..13..12..11..10..09..
08..07..06..05..04..03..
Tests completed.
>>>boot mua0:
(BOOT/R5:0 MUA0



  2..
-MUA0
  1..0..

=

Clearly, stand boots. I'm hopeful that 4.3-Quasijarus0a will indeed run on
simh.

For some follow-up questions... Standalone format supports hp and up disks,
but simh only emulates RL and RQ controllers - which means that the Ultrix
utilities to label a cold disk must be ported or recreated to the simh host
platform. Of course, it may well be that applying a hex editor to the disk
image will suffice.

Am I correct that Quasijarus supports RA82 disks (using the ra type) and the
TK50 tape (as a tms)?

Markus

-----Original Message-----
From: Michael Sokolov [mailto:msokolov@ivan.Harhan.ORG]
Sent: Friday, September 26, 2003 6:47 AM
To: jmbw at nather.com; tuhs at tuhs.org
Subject: Re: [TUHS] Re: 4.3 BSD version in the Unix Archive


Warren Toomey <wkt at tuhs.org> wrote:

> The 4.3BSD-Quasijarus dists were compressed with a compression format
> that's not compatible with either gzip nor old compress(1). Michael
> Sokolov should be able to send in some notes on the tools required.

Yes, for political reasons I needed to make the 4.3BSD-Quasijarus compressor
a
version of compress, not gzip (can't have any GNU), but I wanted to have the
higher compression ratio of deflation, so I created a new version of
compress
that supports deflation in addition to the original LZW algorith.  You can
find
the new compress in components/compress.tar either on my FTP site or in
Warren's
archive in the 4BSD area.

> > Is there a set of 4.3BSD-Tahoe Vax distribution files that's complete?
>
> Not that I know of.

"4.3BSD-Tahoe Vax distribution" is an oxymoron.  Berkeley released the Tahoe
tape with Tahoe binaries, no VAX binaries.  I was the one who compiled the
Tahoe
source for the VAX, and the result was 4.3BSD-Quasijarus0.

OTOH, you may be referring to the fact that the Tahoe distribution in the
archive is broken.  Yes, it is.  Unfortunately there was an unrecoverable
tape
read error.

> > Finally, do you know of any 4.3BSD version that will install and run on
> > simh? Quasijarus and Tahoe of the Unix Archive are broken for me and
Reno
> > doesn't boot stand. Admittedly, I haven't tried the vanilla version yet.
>
> Not that I know of. I'll cc this to the TUHS list and see if any other
> people know the answer.

This has come up time and again.  SIMH's emulation of VAX is too poor.  VAX
is
not an easy architecture to emulate.

MS


---
Incoming mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.520 / Virus Database: 318 - Release Date: 9/18/2003

---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.520 / Virus Database: 318 - Release Date: 9/18/2003

[TUHS] ctcompare 1.2 is up

[Warren Toomey]

Hi all,
	It's probably bad form to release a new version of a program
24 hours after the last version, but version 1.2 of my C comparison
tool is now available at http://minnie.tuhs.org/Programs/Ctcompare.

The speed has been vastly improved by using the Rabin-Karp string
comparison algorithm, and thanks go to Roger Willcocks who pointed
me at the algorithm. There are no other bug fixes, and I shouldn't
bring out a new version for quite a while now.

So if you have access to any restricted C source trees, maybe you can
send me the ctf file of the tree so I can do some comparisons!

Cheers,
	Warren

[TUHS] Re: 4.3 BSD version in the Unix Archive

[Warren Toomey]

On Wed, Sep 24, 2003 at 11:12:54AM -0500, Markus Weber wrote:
> Can you decompress the files in either of the two 4.3BSD-Quasijarus dists
> (in /4BSD/Distributions)? If so, how? The oldest versions that I have date
> back to a CD somebody cut me in mid-2000, and I can't decompress these files
> no matter what.

The 4.3BSD-Quasijarus dists were compressed with a compression format
that's not compatible with either gzip nor old compress(1). Michael
Sokolov should be able to send in some notes on the tools required.
 
> Is there a set of 4.3BSD-Tahoe Vax distribution files that's complete?

Not that I know of.
 
> Finally, do you know of any 4.3BSD version that will install and run on
> simh? Quasijarus and Tahoe of the Unix Archive are broken for me and Reno
> doesn't boot stand. Admittedly, I haven't tried the vanilla version yet.

Not that I know of. I'll cc this to the TUHS list and see if any other
people know the answer.

> If I manage to get a version of 4.3 working on simh, I'll offer a turnkey
> system for download on my site.

Yes, I'd appreciate that :-)

Thanks,
	Warren