User-Agent: Microsoft-Outlook-Express-Macintosh-Edition/5.0.5 Subject: fpga pdp-10 status report From: "David G. Conroy" Newsgroups: alt.sys.pdp10 Message-ID: Mime-version: 1.0 Content-Type: text/plain; charset="US-ASCII" Content-transfer-encoding: 7bit Lines: 27 NNTP-Posting-Host: 64.171.6.237 X-Complaints-To: abuse@prodigy.net X-Trace: newssvr14.news.prodigy.com 1031292869 ST000 64.171.6.237 (Fri, 06 Sep 2002 02:14:29 EDT) NNTP-Posting-Date: Fri, 06 Sep 2002 02:14:29 EDT Organization: Prodigy Internet http://www.prodigy.com X-UserInfo1: [[PAPDONYJPOBQDYMBCD^VX@WB]^PCPDLXUNNHLHEQR@ETUCCNSKQFCY@TXDX_WHSVB]ZEJLSNY\^J[CUVSA_QLFC^RQHUPH[P[NRWCCMLSNPOD_ESALHUK@TDFUZHBLJ\XGKL^NXA\EVHSP[D_C^B_^JCX^W]CHBAX]POG@SSAZQ\LE[DCNMUPG_VSC@VJM Date: Fri, 06 Sep 2002 06:14:29 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.bme.hu!andromeda.datanet.hu!newsfeed.gamma.ru!Gamma.RU!snoopy.risq.qc.ca!ps01-chi1!rip!news.webusenet.com!prodigy.com!newsmst01.news.prodigy.com!prodigy.com!postmaster.news.prodigy.com!newssvr14.news.prodigy.com.POSTED!3a2e9f9f!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:11829 Tonight I logged into an ITS, running on the micromachine simulator, which believes that consoles look like 16c450 UARTS, and disks have ATA-2 interfaces. Fixing ITS to understand a different kind of terminal and disk was easy. Fixing ITS to understand a model of machine which was like a KS, but not exactly a KS, was fairly easy too (once I clued into the fact that "IFE KA10P" really means "IFN KL10P\KS10P"). Finding the microcode bug which caused translation buffer misses on writes to replay incorrectly sometimes (a joy to find; poke at the offending location with DDT and the crash goes away because DDT took the miss on a read), and finding the microcode typo which caused byte pointers to be read/written controlled by bit 11 of XCTR rather than bit 10, took a lot longer. And there are still problems. LOCK crashed the system with what looked like a stack overflow/underflow trap, which is tomorrow's debugging. But at this point I think I believe the architecture of the machine is sound, and it's time to finish designing and then build the hardware version of things. dgc ###### From: jmfbahciv@aol.com Newsgroups: alt.sys.pdp10 Subject: Re: fpga pdp-10 status report Date: Sat, 07 Sep 02 10:50:17 GMT Organization: UltraNet Communications, Inc. Lines: 42 Message-ID: References: X-Trace: UmFuZG9tSVaAeYnMa9f0d9ZQ65WtY3UCb1AUcXqheU5GL7wZvqV8w8+unGm0AWSz X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 7 Sep 2002 12:02:38 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!feed2.news.rcn.net!feed1.news.rcn.net!rcn!207-172-102-142 Xref: chonsp.franklin.ch alt.sys.pdp10:11848 In article , "David G. Conroy" wrote: >Tonight I logged into an ITS, running on the >micromachine simulator, which believes that consoles look like >16c450 UARTS, and disks have ATA-2 interfaces. > >Fixing ITS to understand a different kind of terminal >and disk was easy. Fixing ITS to understand a model of machine which >was like a KS, but not exactly a KS, was fairly easy too (once >I clued into the fact that "IFE KA10P" really means "IFN KL10P\KS10P"). Heh, heh. Those pesky macroes. > >Finding the microcode bug which caused translation buffer >misses on writes to replay incorrectly sometimes (a joy to find; poke >at the offending location with DDT and the crash goes away because >DDT took the miss on a read), Would SET ADDRESS BREAK have helped here? You might think about writing one. > .. and finding the microcode typo >which caused byte pointers to be read/written controlled by bit 11 of >XCTR rather than bit 10, took a lot longer. I hope you didn't bang your head too much. > >And there are still problems. LOCK crashed the system with >what looked like a stack overflow/underflow trap, which is tomorrow's >debugging. > >But at this point I think I believe the architecture of >the machine is sound, and it's time to finish designing and then >build the hardware version of things. Sounds like you're having fun. /BAH Subtract a hundred and four for e-mail. ###### From: jmfbahciv@aol.com Newsgroups: alt.sys.pdp10 Subject: Re: fpga pdp-10 status report Date: Sun, 08 Sep 02 10:22:36 GMT Organization: UltraNet Communications, Inc. Lines: 58 Message-ID: References: <3D7A97ED.6EEC4CE0@ev1.net> X-Trace: UmFuZG9tSVb1taMTlwHrZGfGt/+Zmj0PIPqRLK61KhrQCVTpfJQPi6Lsf3jnNVQg X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 8 Sep 2002 11:35:04 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!triton.net!smallfeed.triton.net!nntp.kreonet.re.kr!newsfeed.gol.com!feed1.news.rcn.net!rcn!207-172-216-141 Xref: chonsp.franklin.ch alt.sys.pdp10:11846 In article <3D7A97ED.6EEC4CE0@ev1.net>, Charles Richmond wrote: >jmfbahciv@aol.com wrote: >> >> In article , >> "David G. Conroy" wrote: >> >Tonight I logged into an ITS, running on the >> >micromachine simulator, which believes that consoles look like >> >16c450 UARTS, and disks have ATA-2 interfaces. >> > >> >Fixing ITS to understand a different kind of terminal >> >and disk was easy. Fixing ITS to understand a model of machine which >> >was like a KS, but not exactly a KS, was fairly easy too (once >> >I clued into the fact that "IFE KA10P" really means "IFN KL10P\KS10P"). >> >> Heh, heh. Those pesky macroes. >> >BAH, were there any "test suites" that DEC ran on the -10's >to beat the hell out of them??? A group got paid to do this. The only people in that group who actually did any useful work didn't stay long. To compensate for the lack of testing out of this group, we found that the Tuesday builds of the monitor gave the system a pretty good workout (I can't say that this exercised the hardware--it sure exercised the montior software). We always wanted ORNL for our field test site because they had a lot of "users" who did all kinds of different things. I've quoted the users because a lot of them thought they were using the IBM machines. > .. I read in _The Soul of a New >Machine_ about such test suites that were used on the MV-8000 >at Data General. These programs loop back and forth, execute >all kinds of instructions in many, many combinations, and >generally do everything they can to expose a flaw in the >hardware. Something like this for the -10 would be helpful >to all the fpga developers out there building their own -10. >IMHO. That was supposed to be provided with the diagnostics. Unfortunately, the diagnostic mind set was based on a hardware p.o.v. and not a software p.o.v. (this is my hypothesis). The last guy in charge of the dianostic group did try to add as many lapses as he could into the diags. The problem with diags is that they were targeted at _pieces_ of hardware, not the interaction of different kinds of hardware. >Did such programs ever exist at DEC??? > Yup. TOPS-10. /BAH Subtract a hundred and four for e-mail. ###### User-Agent: Microsoft-Outlook-Express-Macintosh-Edition/5.0.5 Subject: Re: fpga pdp-10 status report From: "David G. Conroy" Newsgroups: alt.sys.pdp10 Message-ID: References: Mime-version: 1.0 Content-Type: text/plain; charset="US-ASCII" Content-transfer-encoding: 7bit Lines: 28 NNTP-Posting-Host: 64.171.6.237 X-Complaints-To: abuse@prodigy.net X-Trace: newssvr14.news.prodigy.com 1031415236 ST000 64.171.6.237 (Sat, 07 Sep 2002 12:13:56 EDT) NNTP-Posting-Date: Sat, 07 Sep 2002 12:13:56 EDT Organization: Prodigy Internet http://www.prodigy.com X-UserInfo1: [[PAPDONYJPOBQDYMBCD^VX@WB]^PCPDLXUNNHLHEQR@ETUCCNSKQFCY@TXDX_WHSVB]ZEJLSNY\^J[CUVSA_QLFC^RQHUPH[P[NRWCCMLSNPOD_ESALHUK@TDFUZHBLJ\XGKL^NXA\EVHSP[D_C^B_^JCX^W]CHBAX]POG@SSAZQ\LE[DCNMUPG_VSC@VJM Date: Sat, 07 Sep 2002 16:13:56 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!news.stealth.net!news.stealth.net!prodigy.com!newsmst01.news.prodigy.com!prodigy.com!postmaster.news.prodigy.com!newssvr14.news.prodigy.com.POSTED!3a2e9f9f!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:11839 >> Finding the microcode bug which caused translation buffer >> misses on writes to replay incorrectly sometimes (a joy to find; poke >> at the offending location with DDT and the crash goes away because >> DDT took the miss on a read), > > Would SET ADDRESS BREAK have helped here? You might think about > writing one. I suspect not. The bug didn't cause the program to quietly do the wrong thing; the bug caused the micromachine to get lost in the weeds (it replayed the reference incorrectly after the translation buffer miss), and then stumble back into the fetch flow and begin executing instructions "normally" again. > >> .. and finding the microcode typo >> which caused byte pointers to be read/written controlled by bit 11 of >> XCTR rather than bit 10, took a lot longer. > > I hope you didn't bang your head too much. Only a couple of hours. If the ascii string which was being manipulated in this case had started with an ordinary character and the system had typed "AAAAAA..." I would have found it instantly. But it began with a device-independent cursor control code (a ^P), and the command interpreter died cleanly with an I/O error of the type "illegal cursor control sequence" (^P^P). ###### Message-ID: <3D7A97ED.6EEC4CE0@ev1.net> From: Charles Richmond Reply-To: richmond@ev1.net Organization: Canine Computer Center X-Mailer: Mozilla 4.7C-CCK-MCD {C-UDP; EBM-APPLE} (Macintosh; I; PPC) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: alt.sys.pdp10 Subject: Re: fpga pdp-10 status report References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 31 NNTP-Posting-Host: 12.237.69.162 X-Complaints-To: abuse@attbi.com X-Trace: rwcrnsc51.ops.asp.att.net 1031437407 12.237.69.162 (Sat, 07 Sep 2002 22:23:27 GMT) NNTP-Posting-Date: Sat, 07 Sep 2002 22:23:27 GMT Date: Sat, 07 Sep 2002 22:23:27 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!howland.erols.net!usc.edu!attla2!ip.att.net!attbi_feed3!attbi.com!rwcrnsc51.ops.asp.att.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:11861 jmfbahciv@aol.com wrote: > > In article , > "David G. Conroy" wrote: > >Tonight I logged into an ITS, running on the > >micromachine simulator, which believes that consoles look like > >16c450 UARTS, and disks have ATA-2 interfaces. > > > >Fixing ITS to understand a different kind of terminal > >and disk was easy. Fixing ITS to understand a model of machine which > >was like a KS, but not exactly a KS, was fairly easy too (once > >I clued into the fact that "IFE KA10P" really means "IFN KL10P\KS10P"). > > Heh, heh. Those pesky macroes. > BAH, were there any "test suites" that DEC ran on the -10's to beat the hell out of them??? I read in _The Soul of a New Machine_ about such test suites that were used on the MV-8000 at Data General. These programs loop back and forth, execute all kinds of instructions in many, many combinations, and generally do everything they can to expose a flaw in the hardware. Something like this for the -10 would be helpful to all the fpga developers out there building their own -10. IMHO. Did such programs ever exist at DEC??? -- +-------------------------------------------------------------+ | Charles and Francis Richmond | +-------------------------------------------------------------+ ###### User-Agent: Microsoft-Outlook-Express-Macintosh-Edition/5.0.5 Subject: Re: fpga pdp-10 status report From: "David G. Conroy" Newsgroups: alt.sys.pdp10 Message-ID: References: <3D7A97ED.6EEC4CE0@ev1.net> Mime-version: 1.0 Content-Type: text/plain; charset="US-ASCII" Content-transfer-encoding: 7bit Lines: 47 NNTP-Posting-Host: 66.120.160.139 X-Complaints-To: abuse@prodigy.net X-Trace: newssvr14.news.prodigy.com 1031454925 ST000 66.120.160.139 (Sat, 07 Sep 2002 23:15:25 EDT) NNTP-Posting-Date: Sat, 07 Sep 2002 23:15:25 EDT Organization: Prodigy Internet http://www.prodigy.com X-UserInfo1: SCSYASVDLRUORWDYNCOF_W\@PJ_^PBQLGPQRJRQIMASJETAANVW[AKWZE\]^XQWIGNE_[EBL@^_\^JOCQ^RSNVLGTFTKHTXHHP[NB\_C@\SD@EP_[KCXX__AGDDEKGFNB\ZOKLRNCY_CGG[RHT_UN@C_BSY\G__IJIX_PLSA[CCFAULEY\FL\VLGANTQQ]FN Date: Sun, 08 Sep 2002 03:15:25 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!eusc.inter.net!news.tele.dk!small.news.tele.dk!207.115.63.138!newscon04.news.prodigy.com!newsmst01.news.prodigy.com!prodigy.com!postmaster.news.prodigy.com!newssvr14.news.prodigy.com.POSTED!3a2e9f9f!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:11838 The design had already run the DEC core diagnostics for the instructions it implements (AA-AM, CA-CB) which caught a few errors. Finding the two errors I mentioned would require a fairly exhaustive diagnostic which ran with paging enabled. There are no DEC diagnostics which can run with ITS paging enabled, and the diagnostic I wrote from scratch to take a quick swish around the ITS executive mode instructions was far from exhaustive. > From: Charles Richmond > Organization: Canine Computer Center > Reply-To: richmond@ev1.net > Newsgroups: alt.sys.pdp10 > Date: Sat, 07 Sep 2002 22:23:27 GMT > Subject: Re: fpga pdp-10 status report > > jmfbahciv@aol.com wrote: >> >> In article , >> "David G. Conroy" wrote: >>> Tonight I logged into an ITS, running on the >>> micromachine simulator, which believes that consoles look like >>> 16c450 UARTS, and disks have ATA-2 interfaces. >>> >>> Fixing ITS to understand a different kind of terminal >>> and disk was easy. Fixing ITS to understand a model of machine which >>> was like a KS, but not exactly a KS, was fairly easy too (once >>> I clued into the fact that "IFE KA10P" really means "IFN KL10P\KS10P"). >> >> Heh, heh. Those pesky macroes. >> > BAH, were there any "test suites" that DEC ran on the -10's > to beat the hell out of them??? I read in _The Soul of a New > Machine_ about such test suites that were used on the MV-8000 > at Data General. These programs loop back and forth, execute > all kinds of instructions in many, many combinations, and > generally do everything they can to expose a flaw in the > hardware. Something like this for the -10 would be helpful > to all the fpga developers out there building their own -10. > IMHO. > > Did such programs ever exist at DEC??? > > -- > +-------------------------------------------------------------+ > | Charles and Francis Richmond | > +-------------------------------------------------------------+ ###### From: "glen herrmannsfeldt" Newsgroups: alt.sys.pdp10 References: <3D7A97ED.6EEC4CE0@ev1.net> Subject: Re: fpga pdp-10 status report Lines: 29 X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.00.2919.6700 X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2919.6700 Message-ID: <86ef9.266511$aA.46549@sccrnsc02> NNTP-Posting-Host: 12.228.58.87 X-Complaints-To: abuse@attbi.com X-Trace: sccrnsc02 1031630340 12.228.58.87 (Tue, 10 Sep 2002 03:59:00 GMT) NNTP-Posting-Date: Tue, 10 Sep 2002 03:59:00 GMT Organization: AT&T Broadband Date: Tue, 10 Sep 2002 03:59:00 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!news.stealth.net!news.stealth.net!204.127.161.2.MISMATCH!wn2feed!wn3feed!worldnet.att.net!204.127.198.203!attbi_feed3!attbi.com!sccrnsc02.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:11865 wrote in message news:alfcl8$mct$5@bob.news.rcn.net... > In article <3D7A97ED.6EEC4CE0@ev1.net>, > Charles Richmond wrote: (snip) > > .. I read in _The Soul of a New > >Machine_ about such test suites that were used on the MV-8000 > >at Data General. These programs loop back and forth, execute > >all kinds of instructions in many, many combinations, and > >generally do everything they can to expose a flaw in the > >hardware. Something like this for the -10 would be helpful > >to all the fpga developers out there building their own -10. > >IMHO. (snip) > >Did such programs ever exist at DEC??? > > > Yup. TOPS-10. I have some IBM RAM that with a sticker claiming to be OS/2 compatible. It seems that IBM uses a similar system. -- glen