From: Digestifier To: Linux-Development@senator-bedfellow.mit.edu Reply-To: Linux-Development@senator-bedfellow.mit.edu Date: Fri, 30 Sep 94 08:13:06 EDT Subject: Linux-Development Digest #244 Linux-Development Digest #244, Volume #2 Fri, 30 Sep 94 08:13:06 EDT Contents: Re: people using SCSI-IN2000 driver, please read this (Larry Doolittle) Re: [?] DIP with auto-redial? (Harald Milz) Linux for Powermac? (Olaf Ronneberger) Pioneer SCSI Adapter TMC-850MER (Andrew Anderson) Re: Process checkpointing (Matti Aarnio) Re: Korn Shell '93 Now Available from AT&T (Chris Flatters) Please help me ... I can't get InterViews to work !!!!! (k42bc@cunyvm.cuny.edu) dip[6461]: SET_ENCAP(VJ_COMP=0): I/O error (Ron Arts) Re: Korn Shell '93 Now Available from AT&T (Ahmed Naas) Re: i486 Word length, anyone? (Scott A. Laird) Re: [STATUS] Linus Floppy Driver Development (Shannon Hendrix) Re: xircom pocket ethernet support ??? (Bjorn Ekwall) Process checkpointing (Elan Feingold) device driver for all three operating systems (Farrah Solouki-John) Re: Could TCP/IP be implemented over SCSI? (Ian McCloghrie) Re: SMail security hole? (Brian Kramer) NCR-SCSI,Audio-CDs => kernel-panic (Gunnar Kempke) ---------------------------------------------------------------------------- From: doolitt@recycle.cebaf.gov (Larry Doolittle) Subject: Re: people using SCSI-IN2000 driver, please read this Reply-To: doolittle@cebaf.gov Date: Tue, 27 Sep 1994 00:41:53 GMT Stefan Markgraf (stefan@pippi.tu-bs.de) wrote: : Is anybody able to fix the problem with the in2000 driver? : Sometimes while swapping (to a swap-partition) the driver crashes. : No read/writes are possible and therefore every process begins to hang : when it needs access to the disk. The HD light burns constantly. I have seen this, always with the "new" fast driver on tsx-11. If you want something slow, but (more?) stable, try version 0.3 (released by me :-). Version 0.3 needs some hacking to mate with newer kernels. : Console switching (during the driver crash) is possible as long as : Xfree do not use the disk or when working at a plain text console. : However, each console begins to hang as soon as a process needs disk access : (e.g. excuting a new command) : "ping" from other machines result in " is alive." Yup, Yup, Yup. : My in2000-bios version is 1.03, the hardware-version is 27 (bootup message) : kernel is 1.1.50 with 16 MB, 17 MB swap-space on a swap partition. I use : the newest version of the in2000.c. : (My memory "upgrade" 8->16 MB has shown, that the driver crashing occurs : more seldom --- this is a result of fewer swapping requests, I suppose). : I HAVE READ THAT THERE ARE PEOPLE USING THE IN2000 DRIVER WITHOUT : PROBLEMS. PERHAPS WE CAN FIND OUT THE BIOS- AND HARDWARE-VERSIONS : WHICH WORK AND WHICH DO NOT WORK. PLEASE POST (OR SEND ME MAIL) YOUR : CONFIGURATION (KERNEL, RAM, SWAP-TYPE AND SPACE, HARDDRIVE) AND : IN2000-VERSION HERE. PLEASE WRITE IF CRASHING OCCURS OR NOT. Those who use it without crashing either use the slow driver, or use swapping to a non-IN2000 disk, or swap to a file. The IN2000 slow driver comes with patches to the kernel to force swapping to happen in 512 byte chunks. (I know, it's lame, but it works). The "new" driver allows 4096 byte transfers to happen -- most of the time. Eventually a latent bug shows up -- it could be in the firmware, for all I know -- and the bus locks up. Good luck debugging it! : Perhaps we can fix this nasty bug. : Another question: Perhaps one of the gurus can post us how to set up : the in2000 driver for intelligent debugging. I tried to set "#define DEBUG" in : in2000.c, redirected kernel-errors/debug/crit-messages of the syslogd to : /dev/console and recompiled the kernel. However there are so many : messages, that I cannot bootup again. Since the system is still running, - is still functioning. Try patching some IN2000 status routine printer into that routine. If you need help crashing your SCSI, e-mail me for a copy of "thrash" that exercises arbitrary amounts of swap-space. - Larry Doolitle doolittle@cebaf.gov ------------------------------ From: hm@ix.de (Harald Milz) Subject: Re: [?] DIP with auto-redial? Reply-To: hm@seneca.ix.de Date: Tue, 27 Sep 1994 16:33:01 GMT In comp.os.linux.development, Mike Castle (mcastle@umr.edu) wrote: > In article <365o3q$ohk@quartz.ucs.ualberta.ca>, > John Voth wrote: > >I am in search of a DIP or DIP-alike program that has auto redial > >functions built into it. > >I am in constant competition with others trying to connect to my > >university and the DIP I have now just flops at a busy signal. > I'm not sure if all versions of DIP did this, but I know the most > recent versions set $errlvl to different values, depending on why > it failed. I'm not positive where the values for errlvl are > documented, perhaps in one of the sample scripts. I tuned dip-337 some time ago to return an exitcode from the dip script. Now you can say "exit 5" in the script, and dip returns "5" to the shell. This allows me to auto-redial if a recoverable error (such as BUSY) occurs. I sent the respective patch to Uri Blimenthal but didn't receive an answer. Seems he's not too interested. If anyone wants the patch (actually, five or six lines more), drop me a note. -- They're only trying to make me LOOK paranoid! -- Harald Milz (hm@ix.de) WWW: http://www.ix.de/editors/hm.html iX Multiuser Multitasking Magazine phone +49 (511) 53 52-377 Helstorfer Str. 7, D-30625 Hannover fax +49 (511) 53 52-378 Opinions stated herein are my own, not necessarily my employer's. ------------------------------ From: ronneber@galcit.caltech.edu (Olaf Ronneberger) Subject: Linux for Powermac? Date: 28 Sep 1994 23:33:48 GMT Reply-To: ronneber@galcit.caltech.edu Hi! I read in the PowerPC-FAQ, that Linux is beeing ported to PowerPC. Where can I get any further information? Thanks in advance, Olaf ------------------------------ From: andrew@amelia.db.erau.edu (Andrew Anderson) Subject: Pioneer SCSI Adapter TMC-850MER Date: 29 Sep 1994 17:07:12 GMT Can any developer tell me if the code in ~/linux/drivers/scsi/seagate.c will support this card? The board has a TMC-950 chip on it, but the kernel cannot find it. Does the TMC-950 referenced in the comments in the code refer to a *board* or the *chip*? This model of the board does _not_ have a BIOS chip on it, does the driver need one? If this driver does not work, does anyone have a (pre) alpha driver that I can try? Thanks for any and all responces, Andrew -- |===========================================================================| | Andrew Anderson andrew@db.erau.edu | | Novell Network System Administrator "Making the impossible | | Linux System Administrator possible -- daily!" | | | | I don't speak for ERAU, and God knows I don't want them to speak for me! | |===========================================================================| ------------------------------ From: mea@utu.fi (Matti Aarnio) Subject: Re: Process checkpointing Date: Thu, 29 Sep 1994 07:10:27 GMT feingold@avette.zko.dec.com (Elan Feingold) writes: > >I seem to remember reading a while back about a program that >checkpointed running processes. This strikes me as being a >very interesting start for a larger project. One thing that >I've always disliked is that when I reboot I have to set >everything up again (i.e. boot X, run my programs, position >the windows, etc.) I know that there are ways to save my >desktop, but I'm going one step beyond this. In a perfect >OS/World, one could hit a large Red checkpoint button, and have >the state of all running processes saved onto disk, including >the state of IPC/file desciptors/etc. and make there a way for >this to be restored upon reboot (even up to the pids of the >processes -- why not?!) Such facilities do exist, and SparcBook computers act quite much like you describe, however not all the things are saveable, especially this is true with external communication -- say, remote end wants to timeout in X seconds, and the system is off for X+1 seconds ? Another environment/way to do it is implemented on IBM SP2 supercomputer (multi-cpu processor cluster) on which processes can be migrated in between the individual processors! (Which said processors are independent AIX running systems each.) .... >Any comments? Anyone know of the program that did something like >this already? Remember this is just sort of a wishful post, not >a design document or a volunteer notice :) On BSD systems there is a gcore program, which is used to do several "weird" things, like checkpoint dumping of running programs -- generally there is no way to CONTINUE such program from a dump, just a way to make a core-snapshot.. However preservation of process external status, (like files and TCP sockets, ) is going to be a lot more difficult! I think that when a "shutdown" is hit on the SparcBook, it writes ENTIRE core-state to the disk, and when it comes up again, that state is retrieved from the disk as is. (That includes the contents of the display RAM...) Device drivers must also be up to the task, several (all?) need special resetting on powerup, and running kernel can't be thrown in, unless the devices are on proper state. I don't think a kernel change can be done under such conditions. >Regards to all the Linux people out there... > >Elan >-- >|| Elan Feingold (Cornell '94) || "Two of the most famous products of || /Matti Aarnio ------------------------------ From: cflatter@nrao.edu (Chris Flatters) Subject: Re: Korn Shell '93 Now Available from AT&T Date: 29 Sep 1994 22:06:48 GMT Reply-To: cflatter@nrao.edu Hmm. I seem to have missed something here. Could someone mail me the original article. --- ============================================================================== Chris Flatters cflatter@nrao.edu ============================================================================== Well, our problem stems from the fact that we, basically, allow every planet and moon-base this side of Alpha Centauri to make their own version of Spam. "Mystery Meat" Man... or Astroman? ------------------------------ Subject: Please help me ... I can't get InterViews to work !!!!! From: k42bc@cunyvm.cuny.edu Date: 29 Sep 1994 22:10:38 GMT Reply-To: k42bc@cunyvm.cuny.edu Hello, Each time I tried to run InterViews' programs such as Docs, Logo, or the others, I got an error message saied Floating Point Exception. I can't find a way to solve it and please help me. I do have a math coprocesser. I even complied the kernel with math emulation by it still not work. I Linux kernel is 1.1.18. Thanks in advance. ------------------------------ Crossposted-To: comp.os.linux.admin From: raarts@netland.nl (Ron Arts) Subject: dip[6461]: SET_ENCAP(VJ_COMP=0): I/O error Date: Thu, 29 Sep 1994 06:40:41 GMT HI, Subject line says it all. I get this error in the log *only* for every 5th (concurrent) SLIP session. Note I run the 1.1.51 kernel with 16 SLIP channels enabled. I use dip-3.3.7-lilo-3.2 (called as diplogin from /etc/passwd). Any clues? I can provide more info. Email replies preferred please, there's too much noise in the c.o.l.* groups already. Thanks, Ron NetLand Internet Services - Email, Usenet, gopher, WWW, ftp, telnet etc We also set up Internet-connected BBS'es or a WWW server for your organisation Info: +31-(0)20-6943664, dial-up +31-(0)20-6940350, Email: info@netland.nl Info (dutch language only). ------------------------------ From: ahmed@oea.xs4all.nl (Ahmed Naas) Subject: Re: Korn Shell '93 Now Available from AT&T Date: Thu, 29 Sep 1994 12:17:11 GMT Ralph Sims (ralphs@halcyon.halcyon.com) wrote: : chrisb@wombat.cssc-syd.tansu.com.au (Chris Bitmead) writes: : >In article yorton@crawfish.cig.mot.com (James J. Yorton) writes: : >> "The Labs raised some software packaging and porting issues : >> Prices for the binary version of K-shell '93 are $99 per copy : >> (per cpu) until December 31, 1994. Orders placed after that date : >> will be licensed at $149 per copy. We will also write a site : >> license, for a variety of platforms, without restriction as to : >> the number of users or cpus, for $10,000." : >$149 just for a shell? Forget it! : You run a Linux box with more than one user and want to make it : available to everyone, it's $10K, NOT $149. There's a lesson : to be learned somewhere in there. I think you misunderstood the license terms. It says $99/$149 per cpu (i.e. per machine), not per user. (Hmmm. What about multiprocessor machines, you say?) The $10k (source?) license makes sense for sites with hundred(s) of hetrogeneous workstations. -- The above is a result of random neuron activity in the writer's brain. Ahmed M. Naas ahmed@oea.xs4all.nl ====================================================================== ------------------------------ From: lair@kimbark.uchicago.edu (Scott A. Laird) Subject: Re: i486 Word length, anyone? Reply-To: lair@midway.uchicago.edu Date: Thu, 29 Sep 1994 22:28:55 GMT In article <36fan0$92c@nz12.rz.uni-karlsruhe.de>, Thomas Koenig wrote: >Andrew F. Lee (74673.2153@CompuServe.COM) wrote in comp.os.linux.development, > article <36f8f9$75c$1@mhadg.production.compuserve.com>: > >>A long is 32 bits. The registers of a 486DX (as >>opposed to SX) have a 32 bit capacity > >The registers of a 486SX also have 32 bits, it't the external >bus that's 16 bits. > Argh! This is the third time I've seen this this week. The only real difference between a 486SX and a 486DX is the math coprocessor -- the DX has one and the SX doesn't. The _3_86SX has a 16 bit external bus, but even it still has 32-bit registers. Now, I suppose I should mention that Cyrix's 486SLC chips are designed to fit into a 386SX socket, more or less, so they have a 16 bit external bus, but Intel's 486SX does not. Scott. -- Scott A. Laird | "But this goes to 18,446,744,073,709,551,615" lair@midway.uchicago.edu | - Nigel on his 64-bit computer ------------------------------ From: shendrix@escape.widomaker.com (Shannon Hendrix) Subject: Re: [STATUS] Linus Floppy Driver Development Date: Fri, 30 Sep 1994 03:23:33 GMT matthew@crocker.com (Matthew S. Crocker) writes: >So does Fastback for DOS, the drive light is *not* on all the time >but it can tell when you insert the next volume (I think). There must >be a way somehow... Fastback, Norton, and MS-DOS 6.2 poll the drive to do this. -- csh =========================================================================== shendrix@escape.widomaker.com | Linux... that's it for the moment ===================================+ ------------------------------ From: bj0rn@blox.se (Bjorn Ekwall) Subject: Re: xircom pocket ethernet support ??? Date: 29 Sep 94 16:03:45 GMT root@nimir.demon.co.uk wrote: > Is anyone working on a driver for the Xircom parallel > port ethernet adaptor ???? > > > What is the difference between this and the dlink ???? D-Link encourages development of drivers and willingly shows you how. Xircom don't... Bjorn Ekwall == bj0rn@blox.se > Thanks > > Steve Hunt hunt@nimir.demon.co.uk > hunt@nimir.com ------------------------------ From: feingold@avette.zko.dec.com (Elan Feingold) Subject: Process checkpointing Date: 28 Sep 1994 17:07:37 GMT Reply-To: feingold@avette.zko.dec.com (Elan Feingold) I seem to remember reading a while back about a program that checkpointed running processes. This strikes me as being a very interesting start for a larger project. One thing that I've always disliked is that when I reboot I have to set everything up again (i.e. boot X, run my programs, position the windows, etc.) I know that there are ways to save my desktop, but I'm going one step beyond this. In a perfect OS/World, one could hit a large Red checkpoint button, and have the state of all running processes saved onto disk, including the state of IPC/file desciptors/etc. and make there a way for this to be restored upon reboot (even up to the pids of the processes -- why not?!) I realize that this requires support at multiple levels of the kernel to handle this persistance, but as with many ideas, one could take an incremental approach. In OS/2, for example, when I reboot I get the same applications that I was running, but of course it doesn't do checkpointing of DOS/Windows programs, or even OS/2 programs I believe. But it looks nice. Imagine a future where one recompiles a kernel, reboots into DOS just for the heck of it, and then boots up Linux again, to have X come up, the raytrace that has been going for the last week to continue, the TCP/IP connections to reestablish themselves, and the two dosemu sessions to continue with the EPIC Pinball games they were busy with. :) It seems that different applications try to support this persistance in a user visable way (i.e. an option to continue an interrupted raytrace, or a save desktop option in a window manager), but doing it with kernel support (i.e. SIGCHKPNT) is of course much more powerful. Any comments? Anyone know of the program that did something like this already? Remember this is just sort of a wishful post, not a design document or a volunteer notice :) Regards to all the Linux people out there... Elan -- =============================================================================== || Elan Feingold (Cornell '94) || "Two of the most famous products of || || Software Engineer II || Berkeley are LSD and Unix. I don't || || Digital Equipment Corporation || think that is a coincidence." || || Work: 603.881.1115 || - Anonymous || =============================================================================== ------------------------------ From: farrah@farrah.ebay.sun.com (Farrah Solouki-John) Subject: device driver for all three operating systems Date: 29 Sep 1994 22:42:36 GMT Reply-To: farrah@farrah.ebay.sun.com Anyone know how to write a device driver logic for Adaptec 2940 scsi for Linux operating system? Or, Does anyone know of a PCI-Scsi controler that would be compatible with all three (Dos, Linux and OS2) oeprating systems? I would appreciate any info. Thanks much Farrah. Farrah@farrah ------------------------------ From: ianm@qualcomm.com (Ian McCloghrie) Subject: Re: Could TCP/IP be implemented over SCSI? Date: 28 Sep 1994 10:44:16 -0700 jason@idiom.berkeley.ca.us (Jason Venner) writes: >With differential scsi, your scsi bus can extend quite a distance. >With SCSI 2 fast wide/SCSI3 I think you are limited to 256 id's and 256 feet >for the scsi bus. >So: given a little change in technology, networking via scsi will make a lot >of sense. Ummm... it's certainly be possible. But make sense? Last I checked, fast/wide scsi 2 cards weren't exactly cheap. Granted, it's faster than ethernet, but how many people really, seriously, need more than 10Mbits on their LAN? Especially at home (the main Linux presence). The cabling for it would be a nightmare (25 pairs, not just 2 as in 10baseT, *and* you have to worry about termination). It's a cool idea, and perhaps useful in a Desk-Area-Network concept (put an xterminal on your system via scsi and use that as your main display, etc). But it's not well suited as a general LAN technology. -- Ian McCloghrie work: ianm@qualcomm.com home: ian@egbt.org ____ GCS d-- H s+:+ !g p? au a- w+ v- C++$ UL++++ US++$ P+>++ \bi/ L+++ 3 E+ N++ K--- W--- M-- V-- -po+ Y+ t+ 5+++ jx R G''' \/ tv- b+++ D- B-- e- u* h- f+ r n- y* The above represents my personal opinions and not necessarily those of my employer, Qualcomm Inc. ------------------------------ From: bjkramer@pluto.njcc.com (Brian Kramer) Crossposted-To: comp.os.linux.help Subject: Re: SMail security hole? Date: 29 Sep 1994 19:34:02 -0400 I tried setting my transports up exactly as you have them, and still had the same problem. Lee J. Silverman (lee@netspace.students.brown.edu) wrote: : If your version of smail is writing files as root then you : have smail configured incorrectly, and should fix it IMMEDIATELY. If : user xxx has a .forward file that sends output to a file, then smail : should write to that file as user xxx. In order to make sure that : this is true, make sure your /usr/lib/smail/transports file has these : entry: : # pipe - deliver mail to shell commands : # : # This is used implicitly when smail encounters addresses which begin with : # a vertical bar character, such as "|/usr/lib/news/recnews talk.bizarre". : # The vertical bar is removed from the address before being given to the : # transport. : pipe: driver = pipe, # pipe message to another program : return_path, -local, from, -unix_from_hack; : cmd = "/bin/sh -c $user", # use Bourne shell to execute : parent_env, # environment info from parent addr : user=nobody, : group=mail, : -pipe_as_user, # not user-id associated with address : umask=0022, : log_output, # do not log stdout/stderr : ignore_status, # exit status may be bogus, ignore it : ignore_write_errors, # ignore broken pipes : # file - deliver mail to files : # : # This is used implicitly when smail encounters addresses which begin with : # a slash or squiggle character, such as "/usr/info/list_messages" or : # perhaps "~/Mail/inbox". : file: driver = appendfile, : return_path, local, -from, unix_from_hack; : file = $user, # file is taken from address : append_as_user, # use user-id associated with address : expand_user, # expand ~ and $ within address : suffix = "\n", : mode = 0644, : With the drivers set up this way, mail written to a file is written as : user xxx, but mail sent to a pipe is sent as user nobody. It's a : little bit more secure that way. : -- : Lee Silverman, Brown class of '94, Brown GeoPhysics ScM '95 : Email to: Lee_Silverman@brown.edu : Phish-Net Archivist: phish-archives@phish.net : "Nonsense - you only say it's impossible because nobody's ever done it." -- Brian Kramer - Owner/Systems Administrator - bjkramer@pluto.njcc.com New Jersey Computer Connection - Public Access Unix Site - pluto.njcc.com Voice: 609-896-2799 - Fax: 609-896-2994 - Dialups: 609-896-3191 Dialup or Telnet to pluto.njcc.com and log in as guest for more information. ------------------------------ From: gu@gus.kiel.org (Gunnar Kempke) Subject: NCR-SCSI,Audio-CDs => kernel-panic Date: Thu, 29 Sep 1994 09:36:21 GMT Hello all! My system: PCI/I-P5MP3 ASUS-board (BIOS ver 2.01) with NCR53C810-SCSI (BIOS ver 3.04), SCSI-HD, Toshiba 4101B SCSI-CDROM Kernel 1.1.19 with Drew's NCR-Driver Release 3 My problem: Every time I start playing an Audio-CD (with Workbone, Workman, doesn't matter), my system crashes. Everything else (including CD-Rom operations) works just fine. How come? tia, Gunnar -- =============================================================================== Gunnar Kempke, Kopperpahler Allee 41, 24119 Kronshagen, Germany ------------------------------ ** FOR YOUR REFERENCE ** The service address, to which questions about the list itself and requests to be added to or deleted from it should be directed, is: Internet: Linux-Development-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux.development) via: Internet: Linux-Development@NEWS-DIGESTS.MIT.EDU Linux may be obtained via one of these FTP sites: nic.funet.fi pub/OS/Linux tsx-11.mit.edu pub/linux sunsite.unc.edu pub/Linux End of Linux-Development Digest ****************************** .