[Simh] Problem booting RSX11M on pdp11 (halt at loc 0)

Fri Mar 30 07:34:54 EDT 2018

Hi Mike

the point on another Linux is valid -- as I suspect VBox itself is not to blame since I had similar results with VMware Workstation (although I have not attempted to find a way to access "real" store on there I think it would work if I could do so).

Based on other responses and comments from the mailing list that others have run other SimH simulators under virtualisation (but maybe more commercial ones) with no issues I suspect it may be something on my hardware setup on the HP laptop (relating to virtualisation) or the Win 10 itself.

I am not yet good with SimH debugging (it took me a while to get to provide the help it did and I am still not convinced I did it correctly) so any suggestions welcome on that front.  I did notice what seems to be a "bug" in that output is CR (015) terminated which I feel is odd - certainly messes up my looking back at my Putty history.

Phil

----- Original Message -----
From: Mike Stramba <mikestramba at gmail.com>
Reply-To: <mikestramba at gmail.com>
To: Phil Fisher <phil.fisher at peejayeff.co.uk>
Cc: Bob Supnik (simh) <simh at trailing-edge.com>
Sent: 29/03/2018 23:40:40
Subject: Re: [Simh] Problem booting RSX11M on pdp11 (halt at loc 0)
________________________________________________________________________________

Hi Phil,

Maybe try another Linux "flavor" ?

And /or check on the VirtualBox forums.

pdp11/rsx11 works for me on Ubuntu 14 (native Ubuntu).

Hmm, I do have VirtualBox installed on that Ubuntu box ..... Ubuntu
running on Vbox on Ubuntu, running rsx11 ? ;)

Mike

On 3/29/18, Phil Fisher <phil.fisher at peejayeff.co.uk> wrote:
> Dear SimH people
>
> I think I have found a pointer to the issue. The issue apparently being that
> my VirtualBox simulator under which I run Centos 6 Linux seems to have some
> challenge providing disk information to the SimH simulator running on Centos
> 6.
>
> I think I have demonstrated this by first enabling some debugging on the
> compiled pdp11 simulator (DEBUG=1) and enabling DEBUG=TRACE;OPS;DATA on the
> RL disk device.
>
> When I ran the simulator and merely enabled 11/70, 1MB RAM and a RL0 disk
> attached, I found the following results (the last few lines only shown):
>
> PDP-11 simulator V4.0-0 Current        git commit id: 9111a1a6
> ======================================================================
> Disabling XQ
> CPU     11/70, FPP, RH70, autoconfiguration enabled, idle enabled
>         1024KB
> ...
> sim> attach RL0 ./RSX-11M_V3.2_RSX11MBL26_3.2.rl01
> sim> BOOT RL0
> DBG(8281)> RL RWR: >>RL wr: RLDA 000013
> DBG(8281)> RL RWR: >>RL0 write: RLBA=013
> DBG(8283)> RL RWR: >>RL wr: RLCS 000200 new 000004
> DBG(8283)> RL RWR: >>RL GSTA: rlds=000035 drv=0
> ...
> DBG(8320)> RL RRD: >>RL0 read: RLCS=014
> DBG(8322)> RL RRD: >>RL rd: RLCS 000014
> DBG(8322)> RL RRD: >>RL0 read: RLCS=014
> DBG(8324)> RL RRD: >>RL rd: RLCS 000014
> DBG(8324)> RL RRD: >>RL0 read: RLCS=014
> DBG(8325)> RL OPS: >>RL svc: func=RD drv=0 rlda=000000
> DBG(8325)> RL OPS: >>RL svc: cyl 0, sect 0, wc 512, maxwc 5120
> DBG(8325)> RL OPS: RL0 sim_disk_rdsect lbn: 00000000 len: 00000400
> DBG(8325)> RL OPS: 0000 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> ................
> DBG(8325)> RL OPS: 0010 thru 03FF same as above
> DBG(8326)> RL RRD: >>RL rd: RLCS 000215
> DBG(8326)> RL RRD: >>RL0 read: RLCS=0215
> DBG(8328)> RL RRD: >>RL rd: RLCS 000215
> DBG(8328)> RL RRD: >>RL0 read: RLCS=0215
> DBG(8328)> RL RWR: >>RL wr: RLCS 000215 new 000000
> DBG(8328)> RL RWR: >>RL0 write: RLCS=00
>
> HALT instruction, PC: 000002 (HALT)
> sim>
>
> at which point clearly the BOOT RL0 command will fail (and HALT).
> So the buffer returned from the disk is all zeroes and is loaded into memory
> which results as above.
>
> I had a think and considered that perhaps the I/O in the simulator(s) was
> getting confused somehow.  So I attached the SAME DISK IMAGE but fetched it
> from the local PC store by making use of the Shared Folder feature of
> VirtualBox.  This led the disk image to be located on the NTFS Win 10
> filestore.
>
> And lo, behold!  It boots and I get the RSX11M prompt.
>
> sim> SET RL  ENABLE
> sim> SET RL0 ENABLE, RL01, WRITEENABLED
> sim> set RL DEBUG=OPS;TRACE;DATA
> sim> ATTACH RL0 /mnt/phil/phil_dir/Simulators/rsx3_2/rsx11mbl26_3_2.rl01
> RL0: '/mnt/phil/phil_dir/Simulators/rsx3_2/rsx11mbl26_3_2.rl01' Contains an
> ODS1 File system
> RL0: Volume Name:       RSXM26 Format: DECFILE11A   Sectors In Volume: 10240
> sim> boot RL0
>
>
>   RSX-11M V3.2 BL26   28K
>>RED DL:=SY:
>>RED DL:=LB:
>>MOU DL:RSXM26
>>@DL:[1,2]STARTUP
>>* PLEASE ENTER TIME AND DATE (HR:MN DD-MMM-YY) [S]:
>
> Notice that SimH detects it as a native File11 file system whereas before
> the format was apparently SimH based on simulator output.
>
> Now, given the above, has anyone any suggestions on how I look further on
> this on the grounds that I would like the setup to be as normal as possible
> and using "native" files in the VBox/Centos6 system?
>
> Phil
> --
> Phil J Fisher           phil.fisher at peejayeff.co.uk
> _____________________________________________________
> Email and any attachments sent are scanned by McAfee
> Anti-Virus but integrity cannot be guaranteed.
> _____________________________________________________
>
> _______________________________________________
> Simh mailing list
> Simh at trailing-edge.com
> http://mailman.trailing-edge.com/mailman/listinfo/simh