Date: Mon, 16 Jan 95 11:15:42 EST From: Dwight McKay (The Moderator) Reply-To: Suns-at-Home@harbor.ecn.purdue.edu Subject: Suns-at-Home Digest V8 #2 To: Suns-at-Home-List Suns-at-Home Digest Mon, 16 Jan 95 Volume 8 : Issue 2 Today's Topics: No Subject Problems building X11R6 on Sun 3/80 using gcc 2.6.1 SLIP for 1+ running 4.1.2; tia Sun-4 SLC Ethernet troubles +--------------------------------------------------------------------+ | Submissions: suns-at-home \ @harbor.ecn.purdue.edu | | Requests: suns-at-home-request > -- or -- | | Archives: suns-at-home-archives / ...rutgers!pur-ee!... | +--------------------------------------------------------------------+ ---------------------------------------------------------------------- Date: Wed, 11 Jan 95 11:58:41 +0100 From: jvl@it.et.tudelft.nl (Jan van Lieshout) Subject: No Subject To: suns-at-home@ecn.purdue.edu Hi, This is probably not a new question --I found it twice in the sah and sunmanager archives-- but I still don't know the answer. I'am trying to use an vpc-2200 versatec/centronics interface in my sun 3/160 (an oldy, yes..) But I don't know how to set the dipswitches on the VME to Multibus convertor board for this. Actually I have have currently two versions of such a conversion board: 1. an older version which (dip switches perpendicular to the VME P1, P2 and P3 connectors) is said to have its VME P2 connector not free from the backplane, and thus --according to the FAQ's-- its slot position should carefully be planned not to interfere with other P2 using boards. 2. A newer one with different dipswitches (parallel to the vme connectors. I would like to use this one but I have no indication of how to set these dipswitches. For the older conversion board I found some dipswitch settings (to have it strapped for address 0x480 interrupt 0x80 if I remember correctly.., anyhow the same as in the kernel config file) and when put in slot 8 (with xy450 in slot 9 and scsi in slot 7 and ALM in slot 12) it was detected at boot time. BUT after "catting" a string to /dev/lp0 this string is printed on the printer connected to the the VPC board, and then there is a kernel PANIC and dumping of pages etc. What do I do wrong? should the VPC board be placed in a higher slot then the XY451 board? Should I leave de BGIN/ BGACK jumpers on the backplane IN for the vpc board? To conclude: can onyone provide me with information on how to set the dipswitches for these newer (:-)) VME to Multibus conversion boards (501-1054) ? (preferably some general information such that I can also use some other multibus boards in the system). There seem to have been a program: vme2mbus around that calculated this but I was not able to locate it anywhere. Thanks, Jan - ------------------------------ Date: Thu, 12 Jan 1995 01:25:28 +1100 (EST) From: craig@orb.apana.org.au (Craig Dewick) Subject: Problems building X11R6 on Sun 3/80 using gcc 2.6.1 To: Suns-at-Home@ecn.purdue.edu (Suns at Home Mailing List) Hello all, I posted the following article to a few newsgroups, however I'll post it here as well since I'm sure there are people who are having, or have had, similar problems to what I'm experiencing with trying to build X11R6 shared libraries.... ------------------------------------------------------------------------------ > From: craig@orb.apana.org.au (Craig Dewick) > Newsgroups: comp.windows.x,comp.sys.sun.apps,alt.sys.sun > Subject: Problems building X11R6 on Sun 3/80 using gcc 2.6.1 > Date: 11 Jan 1995 15:36:34 +1100 I'm having no end of trouble trying to build X11R6 on this machine (a Sun 3/80). I've posted articles about my troubles with this here before and received very little response from readers of these groups (except for one person who's been very supportive, but still I can't get good results). I've fairly well established that the pic generation in gcc is broken, so I'm trying to force the libraries to get built with Sun's default C compiler instead. Only problem is, no matter what I've tried in the config files (sun.cf and site.def) to try and force that to happen, it's not working. Every time the makefiles for the libraries are created at the beginning of the build session the 'CC' and 'PREPROCESSCMD' variables always get set with 'gcc -fpcc-struct-return' as the compiler invocation. I've tried modifiying sun.cf to force it to use 'cc' via the SharedLibraryCcCmd and to add /usr/bin to the search path, but this is always ignored (or overridden - I'm not sure which is actually occuring). I've tried kludging it be explicity #defining 'HasGcc' to YES and 'HasGcc2' to NO, but this doesn't work. I've tried explicity defining 'HasSunC' to YES and the two gcc defines to NO, but even this doesn't work. No matter what I try (short of modifying the base definitions in the template files in xc/config/cf, which I haven't done yet since I shouldn't have to do that) I *cannot* force the shared libraries to build with 'cc'. As far as I've found from reading RELNOTES.TXT there's no mention of how to force the shared libraries to be built with a different compiler to the one used for everything else, so all I can assume is that this part of the build process must be very badly broken. The net effect is that I cannot compile X11R6 for my machine. If gcc's Sun-3 support was OK this wouldn't be a problem since I could just use gcc for everything, but it's Sun-3 support seems very lacking in the area of position-independent-code generation. Sooooo..... can someone give me the rundown on how to get the shared libraries to compile on a Sun-3. I'm not using GNU binutils, just gcc by itself on top of the standard Sun C compiler. I've got the most recent patched version of /bin/ld from Sun's patch 100170-10, and I don't recall seeing any patches for /bin/as, so if I can actually convince the build process to use 'cc' for building the shared libraries I should be on the way to being able to successfully build X11R6 on this machine. This is starting to get very frustrating since I've been trying to get this part of the build to work for at least 3 weeks now without any fruitful leads. I'd appreciate any advice and/or assistance that people can offer. Regards, Craig. ------------------------------------------------------------------------------ I don't know if anyone reading the digest will be able to help out, but I can only try.... Regards, Craig. - ------------------------------ Date: Fri, 13 Jan 95 17:51:55 PST From: starnet!bass!lakin@uucp-gw-2.pa.dec.com (Fred Lakin) Subject: SLIP for 1+ running 4.1.2; tia To: Suns-at-Home@ecn.purdue.edu The last word I saw on this list for SLIP was: Date: Mon, 15 Nov 93 14:15:43 -0500 From: Bob Sutterfield Subject: SLIP on SunSparc2/os 4.1.3 To: Suns-at-Home@ecn.purdue.edu I hear that ftp.ee.lbl.gov:cslip-2.7.tar.Z is considered the best (certainly the most recent) freely-available SLIP for your sort of system. I am about to try to install it, along with tia on the host at the Internet end. Any opinions, options or comments before I start? I am junior grade unix hacker, so hints appreciated. tnx, -f Fred Lakin :: lakin@pgc.com :: (415) 424-1086 - ------------------------------ Date: Tue, 10 Jan 1995 16:31:05 +0100 From: Markus Armbruster Subject: Sun-4 SLC Ethernet troubles To: Suns-at-Home@ecn.purdue.edu My grumpy old SLC doesn't boot any more from Ethernet. Looks like my Ethernet adapter went bananas: ok test-net Internal loopback test -- succeeded External loopback test -- Did not receive expected loopback packet I checked this on different Ethernets, and yes, they are correctly terminated. It boots fine from CD-ROM. Sun charges more for repairs than I paid for the complete machine, so this is not an option for me. My local hardware guru (generic one, not Sun) had a look at the board and concluded that the following chip is probably fried: This one Sun-4 SLC main board | <<--+ +-------------------------|-+ << | | Connector | | +-+ << +-+ v +-+*| << #### *| << | << ******| << | << ***** *****| << ***** *****| << ***** *****| << ***** *****| << | The chip reads: bel 9022 0553-0756-AB HIPOTTED I can't get this chip. I'd really appreciate some help: (1) you can tell me where to get the chip, (2) or where to get a cheap, moderately fried main board, (3) tell me that this chip really can't cause the symptoms, and looking for it is futile. If I can't get any advice, I'll have a SLC for sale: 16MB RAM, keyboard, mouse, external transceiver for thin Ethernet (Cabletron TMS-3), bad Ethernet, otherwise in good shape. Markus - ------------------------------ End of Suns-at-Home Digest ******************************