From: Bob Supnik Newsgroups: alt.sys.pdp10,alt.sys.pdp.11,alt.sys.pdp8 Subject: SIMH 2.8-3 Date: Mon, 17 Dec 2001 09:16:56 -0500 Organization: Posted via Supernews, http://www.supernews.com Message-ID: X-Newsreader: Forte Agent 1.8/32.548 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Complaints-To: newsabuse@supernews.com Lines: 17 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newsfeed00.sul.t-online.de!t-online.de!nntp-relay.ihug.net!ihug.co.nz!telocity-west!TELOCITY!sn-xit-03!sn-post-01!supernews.com!corp.supernews.com!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:8231 New version yesterday, minor tweaks. 1. Fixed a couple of bugs in PDP-8 EAE simulation, thanks to online copies of the EAE diagnostics on another site. 2. Fixed a couple of warnings in the Mac compilation of RQDX3. The Big News: Kevin Handy did a terrific piece of detective work debugging why NetBSD crashes on the VAX simulator. His conclusion: the RQDX3 is 'too fast', it doesn't properly simulate the delays that occur in packet processing; as a result, interrupts are occuring before the software is ready. And indeed, a real MSCP controller is run by some kind of microprocessor, nothing happens instantaneously. I'll retool the RQDX3 with a more realistic threading model and see what happens. /Bob ###### From: klh@panix.com (Ken Harrenstien) Newsgroups: alt.sys.pdp10,alt.sys.pdp11,alt.sys.pdp8 Subject: Re: SIMH 2.8-3 Date: 17 Dec 2001 10:03:31 -0500 Organization: PANIX Public Access Internet and UNIX, NYC Lines: 24 Sender: Ken Harrenstien Message-ID: <9vl1g3$i32$1@panix1.panix.com> References: NNTP-Posting-Host: panix1.panix.com X-Trace: news.panix.com 1008601411 10755 166.84.1.1 (17 Dec 2001 15:03:31 GMT) X-Complaints-To: abuse@panix.com NNTP-Posting-Date: 17 Dec 2001 15:03:31 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.tele.dk!small.news.tele.dk!4.1.16.34!cpk-news-hub1.bbnplanet.com!nycmny1-snh1.gtei.net!news.gtei.net!panix!news.panix.com!panix1.panix.com!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:8220 In article , Bob Supnik wrote: >The Big News: Kevin Handy did a terrific piece of detective work >debugging why NetBSD crashes on the VAX simulator. His conclusion: >the RQDX3 is 'too fast', it doesn't properly simulate the delays that >occur in packet processing; as a result, interrupts are occuring >before the software is ready. And indeed, a real MSCP controller is >run by some kind of microprocessor, nothing happens instantaneously. Ho ho, very amusing! In TOPS-10/20 there are several race conditions -- places where the monitor code makes silent assumptions about how long devices take to do certain things, and it breaks if they are too fast (or too slow!). Every time I found one of those while working on my 10 emulator, (and the similarly-microprocessor-controlled NIA20 ethernet interface was a fruitful source of these), for many frustrating hours I would mumble curses about misguided stone age programmers. And what do you know, looks like the newer generation is just as bad! Thanks, I feel better now... --Ken ###### Path: chonsp.franklin.ch!not-for-mail From: Neil Franklin Newsgroups: alt.sys.pdp10,alt.sys.pdp11,alt.sys.pdp8 Subject: Re: SIMH 2.8-3 Date: 17 Dec 2001 22:02:09 +0100 Organization: My own Private Self Lines: 24 Message-ID: <6uk7vl4l2m.fsf@chonsp.franklin.ch> References: <9vl1g3$i32$1@panix1.panix.com> NNTP-Posting-Host: chonsp.franklin.ch X-Trace: chonsp.franklin.ch 1008622930 472 10.0.3.2 (17 Dec 2001 21:02:10 GMT) X-Complaints-To: news@chonsp.franklin.ch NNTP-Posting-Date: 17 Dec 2001 21:02:10 GMT X-Newsreader: Gnus v5.7/Emacs 20.4 Xref: chonsp.franklin.ch alt.sys.pdp10:8232 klh@panix.com (Ken Harrenstien) writes: > In article , > Bob Supnik wrote: > >occur in packet processing; as a result, interrupts are occuring > >before the software is ready. And indeed, a real MSCP controller is > > In TOPS-10/20 there are several race conditions -- places where the > monitor code makes silent assumptions about how long devices take to > do certain things, and it breaks if they are too fast (or too slow!). > Every time I found one of those while working on my 10 emulator, (and > the similarly-microprocessor-controlled NIA20 ethernet interface was a > fruitful source of these), for many frustrating hours I would mumble > curses about misguided stone age programmers. Any info/list on where all the found ones are? For other emulator/clone writers, or for fixing the OSes if someone one day does such an project. -- Neil Franklin, neil@franklin.ch.remove http://neil.franklin.ch/ Hacker, Unix Guru, El Eng HTL/BSc, Sysadmin, Archer, Roleplayer - Intellectual Property is Intellectual Robbery ###### From: klh@panix.com (Ken Harrenstien) Newsgroups: alt.sys.pdp10,alt.sys.pdp11,alt.sys.pdp8 Subject: Re: SIMH 2.8-3 Date: 18 Dec 2001 03:23:45 -0500 Organization: PANIX Public Access Internet and UNIX, NYC Lines: 27 Sender: Ken Harrenstien Message-ID: <9vmueh$loo$1@panix1.panix.com> References: <9vl1g3$i32$1@panix1.panix.com> <6uk7vl4l2m.fsf@chonsp.franklin.ch> NNTP-Posting-Host: panix1.panix.com X-Trace: news.panix.com 1008663826 1033 166.84.1.1 (18 Dec 2001 08:23:46 GMT) X-Complaints-To: abuse@panix.com NNTP-Posting-Date: 18 Dec 2001 08:23:46 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!cpk-news-hub1.bbnplanet.com!nycmny1-snh1.gtei.net!news.gtei.net!panix!news.panix.com!panix1.panix.com!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:8247 In article <6uk7vl4l2m.fsf@chonsp.franklin.ch>, Neil Franklin wrote: >klh@panix.com (Ken Harrenstien) writes: >> In TOPS-10/20 there are several race conditions -- places where the >> monitor code makes silent assumptions about how long devices take to >> do certain things, and it breaks if they are too fast (or too slow!). >> Every time I found one of those while working on my 10 emulator, (and >> the similarly-microprocessor-controlled NIA20 ethernet interface was a >> fruitful source of these), for many frustrating hours I would mumble >> curses about misguided stone age programmers. > >Any info/list on where all the found ones are? For other >emulator/clone writers, or for fixing the OSes if someone one day does >such an project. Urm. The KLH10 source contains hints -- anytime there is a timeout or delay, it is there for a good reason -- but the comments don't always identify the offensive code, largely because at one point I tried to expunge everything that DEC lawyers might construe as NDA issues. If anyone seriously wants to start a new career as a Free-20 monitor programmer, I would have to go back through my offline notes and compile (or contribute to) a master list. By contrast I only found one race in all of ITS, and that was in the bootstrap. --Ken