FYI: The latest github code has Bob’s changes which add support to disable BEVENT on the 11/03 and 11/23.
From: Simh [mailto:[email protected]] On Behalf Of Mattis Lind Sent: Wednesday, December 30, 2015 11:50 AM To: SIMH <[email protected]> Subject: Re: [Simh] XXDP v2 2015-12-30 18:57 GMT+01:00 Bob Supnik <[email protected]<mailto:[email protected]>>: It's the MMU. After adding the BEVENT disable logic, I noticed that the MMU option cannot be disabled in any model, although it was in fact optional in the KDF11A module (but not the 11/23 system). So I hacked the configuration table to make the MMU optional and tried an 11/23 with no MMU. It hangs just like the 11/03, 04, 05, and 20. Great that we have a solution for BEVENT! According to the manual XXDP shall resort to the small monitor if the MMU is non present or faulty. (page 2-2 indicate so as far as I read it) I tested the images at http://www.ibiblio.org/pub/academic/computer-science/history/pdp-11/xxdp/ I can boot the two XXDP RL02 images attached to rl0 on 11/04, 11/05 and 11/20. PDP-11 simulator V4.0-0 Beta git commit id: ea4d9a16 sim> set 11/04 Disabling XQ sim> attach rl0 ./xxdp22.rl02.img sim> b rl0 CHMDLD0 XXDP+ DL MONITOR BOOTED VIA UNIT 0 28K UNIBUS SYSTEM ENTER DATE (DD-MMM-YY): Simulation stopped, PC: 150574 (BPL 150652) sim> attach rl0 ./xxdp25.rl02.img sim> b rl0 CPU NOT SUPPORTED BY XXDP-XM BOOTING UP XXDP-SM SMALL MONITOR XXDP-SM SMALL MONITOR - XXDP V2.4 REVISION: D0 BOOTED FROM DL0 28KW OF MEMORY UNIBUS SYSTEM RESTART ADDRESS: 152010 TYPE "H" FOR HELP . Simulation stopped, PC: 151006 (TST R0) sim> set 11/05 sim> b rl0 CPU NOT SUPPORTED BY XXDP-XM BOOTING UP XXDP-SM SMALL MONITOR XXDP-SM SMALL MONITOR - XXDP V2.4 REVISION: D0 BOOTED FROM DL0 28KW OF MEMORY UNIBUS SYSTEM RESTART ADDRESS: 152010 TYPE "H" FOR HELP . Simulation stopped, PC: 151002 (JSR PC,150664) sim> set 11/20 sim> b rl0 CPU NOT SUPPORTED BY XXDP-XM BOOTING UP XXDP-SM SMALL MONITOR XXDP-SM SMALL MONITOR - XXDP V2.4 REVISION: D0 BOOTED FROM DL0 28KW OF MEMORY UNIBUS SYSTEM RESTART ADDRESS: 152010 TYPE "H" FOR HELP . Simulation stopped, PC: 150674 (BPL 150750) So it cannot be the MMU (not alone at least). It has to be a combination with device driver. Why it fails is unknown. Mattis' real 11/04 gets further than the simulator, although it ultimately fails. I am not sure if this is a co-incidence but the real 11/03 behaves very similar to the emulated 11/04 in that sense that the address it stops at when halted are very similar. The emulated 11/04 seems to be in some kind of loop testing for console input. I did some further testing creating DU-device images (RD51) for XXDP+, XXDP V2.5 (2.4?) from the RL02-images mentioned above. They both boot on a 11/04 (and all the others) in SimH. https://dl.dropboxusercontent.com/u/96935524/xxdp-plus-du.dsk.gz https://dl.dropboxusercontent.com/u/96935524/xxdp-v25-du.dsk.gz Then I used the V2.2 XXDP image I posted earlier and made a small RD51 image out of it. https://dl.dropboxusercontent.com/u/96935524/xxdp-v22-du.dsk.gz In SimH it boots on 11/73 but not the 11/04. I have not been able to test real systems yet. I don't think this is such a big issue since SimH can in fact run both XXDP+ and XXDP V2.5 (V2.6) and the real machine also had problems with XXDP V2.2. It can also be bugs in XXDP V2.2 that behaves slightly different? /Mattis /Bob On 12/30/2015 12:00 PM, [email protected]<mailto:[email protected]> wrote: Message: 1 Date: Wed, 30 Dec 2015 11:28:35 -0500 From: Bob Supnik <[email protected]<mailto:[email protected]>> To: "[email protected]<mailto:[email protected]>" <[email protected]<mailto:[email protected]>> Subject: Re: [Simh] XXDP v2 Message-ID: <[email protected]<mailto:[email protected]>> Content-Type: text/plain; charset=utf-8; format=flowed I did a quick hack and removed the clock from the device list, effectively disabling it. With that done, the 11/23 runs XXDP V2 correctly, with 64KB of memory. I'll add BEVENT DISABLE to formalize this. So the failing systems are the 03, 04, 05, and 20. Possible common factors: - no memory management - no EIS I tried an 11/03 with EIS and no clock, and it fails, so EIS does not appear to be the problem. I'm now looking for XXDP monitor sources. The failure happens before anything is printed, but a listing would still be of great help. /Bob ------------------------------ Subject: Digest Footer _______________________________________________ Simh mailing list [email protected]<mailto:[email protected]> http://mailman.trailing-edge.com/mailman/listinfo/simh ------------------------------ End of Simh Digest, Vol 143, Issue 27 ************************************* _______________________________________________ Simh mailing list [email protected]<mailto:[email protected]> http://mailman.trailing-edge.com/mailman/listinfo/simh
_______________________________________________ Simh mailing list [email protected] http://mailman.trailing-edge.com/mailman/listinfo/simh
