From: barrett@aminet.uucp (Keith Barrett) Subject: HELP! SLS UUCP/NEWS problems Date: 30 Mar 93 05:13:42 GMT Reply-To: barrett%aminet.uucp@nuconvex.com OK, I've switched from TAMU to SLS (1.0, .99pl6-26), and I have some big time problems that I need solutions for. I HOPE someone out there can help me. 1. I am a victim of these random logout/process kills. Is there a solution? 2. Can someone give me straight-forward instructions for upgrading to the latest version? 3. I can't get getty/uucico to work, nore news in general. It appears that uugetty and uucico don't lock each other out properly. It took me quite a lot of playing around eith inittab and modem settings to get it partially working. -clocal causes any process to HANG on the modem port if CD is not enabled for one. This might be fine if getty was the ONLY process, but bi-directional modem usage is hampered. I have heard that there is a new scheme in Linux, where /dev/cul devices are used? Is the the "approved" method? How do I get this update? 4. There is no cron anywhere. How does one set up a crontab for use? I can't schedule uucp and news stuff without a cron, can I? 5. uucico obeys time restrictions, even if -S is used. 6. There seems to be no method of establishing mail forwarding (i.e. mail -F xxx type stuff). How does one do this? 7. When uucico finishes a call on a news feed, all the batch files vanish and I end up with a core file. I assume it was uuxqt that died, but I have no real idea what's going on. 9. How can I change the "sync" interval? 10. LILO no longer builds an interactive loader for selection partitions. How do I do this now? 11. man getty (uugetty?) does not describe the -r switch. I know what it does (in fact, it ws necessary for me to use it), but it's missing from man. 12. Does nntp work? My BIGGEST problems are getting bi-directional uucp/uucico working (and still allowing shell logins on the same port), and getting news functional. I know how to configure the stuff - it's just not working. Thanks! Keith Barrett barrett@aminet.uucp barrett%aminet.uucp@nuconvex.com -- -kgb ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ UUCP: nuconvex.com!aminet!barrett Keep circulating the tapes - MST3k DDN: barrett%aminet.uucp@decwrl.dec.com // My life is my own - the prisoner Alternate: barrett@pamsrc.enet.dec.com \X/ Amiga 3000UX - The Next Generation ================================================================================ ---------------------------------------------------------------------------- From: barrett@aminet.uucp (Keith Barrett) Subject: .99pl7a problems Date: 1 Apr 93 14:12:57 GMT Reply-To: barrett%aminet.uucp@nuconvex.com OK, I was able to upgrade and rebuild by system, and am now running pl7a. I still have some problems/questions I really could use help with. 1. I created the /cua devices (with valid protection and ownerships), but any attempt to use the cua1 device (my serial port) results in a "device is inuse" error. How does one get around this, and how does this "solve" my uucp/uucico bi-directional modem problem. The /dev/ttys1 seems to work as expected. 2. I created a 512k ram disk when I rebuilt and do get a message about it on bootup. How does one mount and use it though? I tried to create a file system on it, but got a superblock error. 3. Any attempt to use the selection program results now in an IOCTL error. This USE to work -- how do I fix it? 4. Is there ANY way to limit X windows to the vc that starts it? It's TERRIBLE that you lose your vc's when X is running. How about limiting XDM to a single vc? 5. I STILL can't build an interactive loader program. The version of LILO I have only builds a loader that boots the first partition in your config file and doesn't know verbs like "prompt" and "message". Franks Xia loader builds, but won't boot my linux partition (hangs). I read somewhere about building a secondary loader in a reserved area. If I need to do this, how do I do it when I already have a system? I was fortunate enough to build my system with a root using e2fs, but I took default answers (except for file system size n blocks obviously). Can I build a secondary loader without harming my existing system? Thanks!!! Keith ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ UUCP: nuconvex.com!aminet!barrett Keep circulating the tapes - MST3k DDN: barrett%aminet.uucp@decwrl.dec.com // My life is my own - the prisoner Alternate: barrett@pamsrc.enet.dec.com \X/ Amiga 3000UX - The Next Generation ================================================================================ ------------------------------ From: pgf@cayman.com (Paul Fox) Subject: Re: .99pl7a problems Date: 7 Apr 93 18:25:28 GMT Keith Barrett (barrett@aminet.uucp) wrote: : 2. I created a 512k ram disk when I rebuilt and do get a message : about it on bootup. How does one mount and use it though? : I tried to create a file system on it, but got a superblock error. The minor device number on /dev/ram as created by the SLS install is incorrect. It should be minor device 1, not 0. -- paul fox, pgf@cayman.com, (617)494-1999 Cayman Systems, Inc., 26 Landsdowne St., Cambridge, MA 02139 ------------------------------ From: barrett@pamsrc.enet.dec.com (Keith Barrett) Subject: Linux compatible Ethernet Cards? Reply-To: barrett@pamsrc.enet.dec.com Date: Wed, 7 Apr 1993 19:17:44 GMT I searched the FAQ for this, but there's nothing answering it directly. Which ethernet cards are proven to work with Linux? I'm in the market for one and want to get the most compatible one possible (my goal is to use it under DOS, NT, and Linux). Please email me with answers -- I'll post a summary if there's interest. Thanks! Keith Barrett barrett@ ------------------------------ From: barrett@pamsrc.enet.dec.com (Keith Barrett) Subject: Linux NET FAQ needed Reply-To: barrett@pamsrc.enet.dec.com Date: Mon, 12 Apr 1993 19:17:35 GMT I keep coming across references to a Linux "NET" FAQ (an independent posting from the Linux-FAQ, Meta-FAQ, and INFO-SHEET), but cannot find a trace of this in comp.os.linux or news.answers. Could someone email a complet copy of this FAQ? Thanks! Keith Barrett barrett@pamsrc.enet.dec.com ------------------------------ From: barrett@pamsrc.enet.dec.com (Keith Barrett) Subject: Help needed with UUCP packet failures Reply-To: barrett@pamsrc.enet.dec.com Date: Tue, 13 Apr 1993 17:09:32 GMT I'm running .99pl6 (SLS 1.0), with an updated getty_ps package and call-out devices for proper bi-directional modem usage. The modem is a 14.4 device, connected at 9600 baud with no compression or error correction. The serial tty is set to 38400 in gettydefs. Everything pretty much runs smoothly, except that during every connection to my news feed for a news download, I get a "header checksum" error and "got packet xxx; expecting packet yyy" errors about 2 minutes into the call (after a few compressed files have been received). This eventually hangs up the phone. I recently switched to this 486/Linux setup from my old SVR4 Bnews setup. I never had a problem using the EXACT same configuration, sites, phone, and modem - so I do not believe it's a real noise/phone/communication error. Anyone have any ideas what could be going wrong, or what I should do to fix this? Anyone else seeing this? Perhaps there's a problem in uucp "getting back to the phone fast enough" when it creates/renames the temp file into real files? Thanks in advance -- Keith barrett@pamsrc.enet.dec.com ------------------------------ From: barrett@pamsrc.enet.dec.com (Keith Barrett) Subject: Help! - GCC and lib upgrade confusion Date: 13 Apr 93 19:38:32 GMT Reply-To: barrett@pamsrc.enet.dec.com I'm experiencing the random logout problem, and heard that I need to upgrade my sharable libraries to eliminate it. I'm currently running .99pl6, GCC version 2.3.3 As I look around, I see Linux GCC releases and updates marked 2.11, 4.2, 4.3, 4.3.2, and 4.3.3 The various readme files are not clear as to which versions are in fact the current releases. I'm also not sure which ones will work with my current system images and kernel version. Could someone give me some details as to what-is-what here, and what the best upgrade path for me to take is? I can no longer just "wipe" out my system and start over (too much has changed), so I have to deal with component upgrades. I want to get this right, because as you know a wrongly-done sharable library upgrade will hose the system. Thanks in advance Keith Barrett barrett@pamsrc.enet.dec.com P.S. Yes; I have (and have read) the FAQs. While they tell you how to upgrade sharable libs (sort of), they don't tell what the various releases of GCC will work with, or which versions I need. ------------------------------ From: barrett@pamsrc.enet.dec.com (Keith Barrett) Subject: Help - Problems upgrading to .99pl7 Reply-To: barrett@pamsrc.enet.dec.com Date: Tue, 13 Apr 1993 19:25:58 GMT I tried to upgrade the kernel from .99pl6 to .99pl7 It pretty much worked, except for the following: 1. I could no longer shutdown the system. shutdown, halt, etc would all simply display the shutdown message, but then just sit there. I can't use this kernel until I solve this -- anyone know what's going on? 2. SELECTION no longer functions, and returns a IOCTL failure. I have heard this is patchable. Could someone send me the details and pointers? Thanks in advance, Keith Barrett barrett@pamsrc.enet.dec.com P.S. GREAT OS!!!!! :-) ------------------------------ From: hlu@eecs.wsu.edu (HJ Lu) Subject: Re: Help! - GCC and lib upgrade confusion Date: 13 Apr 93 21:46:26 GMT In article <1993Apr13.193832.13383@nntpd.lkg.dec.com>, barrett@pamsrc.enet.dec.com (Keith Barrett) writes: |> |> I'm experiencing the random logout problem, and heard that I need to upgrade |> my sharable libraries to eliminate it. |> |> I'm currently running .99pl6, GCC version 2.3.3 |> |> As I look around, I see Linux GCC releases and updates marked 2.11, 4.2, 4.3, |> 4.3.2, and 4.3.3 |> |> The various readme files are not clear as to which versions are in fact the |> current releases. I'm also not sure which ones will work with my current |> system images and kernel version. |> The Linux C library is named as x.y.z where x is the major release number, y is the minor release number and z is the patch level. The shared library is backward compatible. That is the new release of the Linux C library, x1.y1.z1, will be compatible with x.y.z iff x1 == x and ((y1 >= y) or (y1 == y and z1 >= z)). So 4.3.3 is the latest release and compatible with 4.2, 4.3 and 4.3.2, but not 2.11. You should read the release notes for the compatible kernel versions. For 4.3.3, you should use 0.99.5 or above. |> Could someone give me some details as to what-is-what here, and what the best |> upgrade path for me to take is? I can no longer just "wipe" out my system |> and start over (too much has changed), so I have to deal with component |> upgrades. See above. |> |> I want to get this right, because as you know a wrongly-done sharable |> library upgrade will hose the system. |> Always keep a copy of bootable rootdisk handy. |> |> Thanks in advance |> |> |> Keith Barrett |> barrett@pamsrc.enet.dec.com |> |> |> P.S. Yes; I have (and have read) the FAQs. While they tell you how to upgrade |> sharable libs (sort of), they don't tell what the various releases of GCC |> will work with, or which versions I need. H.J. ------------------------------ From: chuckm@canada.hp.com (Chuck Munro) Subject: Re: New feature for the filesystems. What do you think ? Date: Wed, 14 Apr 1993 00:12:36 GMT Ronnie Sahlberg (ronnie@lysator.liu.se) wrote: > Would it not be great to be able to compress/gzip all files > on the filesystem and still have the files behave like they > were uncompressed ??? =================== This is a great idea, and now is a chance for some brave soul to do it right, rather than using one of those strange DOS-based compression schemes. :-) I am aware that some *real* standardization work is being done as we chat (sorry, don't know where). I heard ISO and/or ANSI was involved. A standards body is dealing with the possibility of having intelligent compression take place for magneto-optical read/write disks. The idea is to do as you suggest in your post .... place a marker in the inode to indicate the file is compressed. That way, you can have both types of files concurrently in the same filesystem. An interesting offshoot of this is the use of a compression chip right in the disk controller. The chip is smart enough to recognize inode structures as the data is read from the disk. If it sees the compression marker, it de-compresses the data in the file *in_hardware* and clears the compression marker before the inode data is passed to the CPU. If the same disk cartridge is mounted in a non-smart disk drive, the kernel code for the filesystem gets the inode structure unmodified, and does decompression in software instead. Furthermore, if your filesystem code can't decompress a file written by another, you can simply 'cat' the file through a standard decompressor command to recover the data. Neat idea! The important thing here is that the *file* data is com- pressed, *not* the inode data. This is quite unlike the DOS schemes. Such an ISO/ANSI filesystem is also portable by design, since filesystems created on any computer can be mounted by any other compliant system. If someone is tempted to try this, I'd like to suggest that now would be a good time to investigate any possible emerging standards *before* creating the filesystem code. Since Linux is headed in the POSIX direction, keeping up with the standards would be a good thing. My $0.02 worth, anyway. Chuck. ------------------------------ From: ericy@hal.gnu.ai.mit.edu (Eric Youngdale) Subject: Re: The dangers of playing with shared libraries Date: 13 Apr 1993 21:20:36 -0400 In article <1993Apr13.161616.24954@kf8nh.wariat.org> bsa@kf8nh.wariat.org (Brandon S. Allbery) writes: >(2) A statically-linked shell has been suggested. Of the currently available > ones, I would have to suggest rc just because it's compact... the concept > of a statically linked "bash" makes me think of a "bashdisk" containing > nothing *but* bash because nothing else will fit :-) This is what the bootable rootdisk is for. I have yet to see a good argument for why the bootable rootdisk cannot be used instead of specially staticly linked binaries to fix screwups with the sharable libraries. -Eric ------------------------------ From: fridlund@mermaid.micro.umn.edu (Jim Fridlund) Subject: Re: Crashing machine: >16M mem patch Date: Wed, 14 Apr 1993 01:55:15 GMT In article kutcha@eos.acm.rpi.edu (Phillip Rzewski) writes: > Path: news1.cis.umn.edu!umn.edu!news-feed-1.peachnet.edu!gatech!rpi!batcomputer!eos.acm.rpi.edu!kutcha > From: kutcha@eos.acm.rpi.edu (Phillip Rzewski) > Newsgroups: comp.os.linux > Date: 11 Apr 1993 21:43:11 GMT > Organization: The Voice of Fate > Lines: 46 > Distribution: world > NNTP-Posting-Host: eos.acm.rpi.edu > Summary: Machine freezes or reboots by itself > Keywords: 16, 32, megs, memory, patch > I recently upgraded the memory on my machine from 8 megs in 1 meg > simms to 32 megs in 4 meg simms. Naturally I would like to be able to > use all of those 32 megs from within linux. Of course that meant > enabling the >16M patch option in the kernel configuration. > > Trouble is, when I enable it, I can't use my machine for long without > it breaking. I looked through the FAQ and the only reference I found > to this sort of thing was "you have to patch the kernel to get >16M". > It said nothing like "it will break on " or > anything. I'm not sure if this is rarely tested code, but I may have > found a bug. My setup is a 386/25dx, 32 megs of memory in 4 meg simms > (I think it's called "x9" type memory, not sure), and the "mainboard " > (motherboard?) is a "PD 386", for what that's worth. I looked through > the whole manual and found no reference to whether the bus was EISA or > ISA, but since I bought the machine a couple years ago and didn't know > anything much about buses at the time, they probably gave me an ISA. I, too, just upgraded from 8 megs to 16 megs and have the same exact motherboard as yours. The only thing different is that I have a 33 Mhz Intel chip & 128K cache. I'm also experiencing some problems with my Linux workstation (ver. 0.99pl7). :) If we do have the same motherboard, it's definitely not EISA. > Here's some idea of how I crashed the machine. I made myself two boot > disks, one with the default "only up 16 megs" and the other with the > ">16 megs" patch enabled. I booted with the >16M one and went into X. > I had a kermit open and a couple of xterms running bash, one of which > I had top in. I then ran idraw once to see what top was saying. It > accepted that one. I ran a second idraw and it accepted that one. So > now I had two open at once. I was about to go to make a third when he > machine froze for a couple seconds and then rebooted by itself. I > crashed it similarly by compiling gnutar and opening xv and emacs and > a couple other things (for that one it just froze totally and I had to > reboot myself), but that would be harder to repeat exactly as I did it > the first time. The amount of free memory on top was still above the > 16 meg level when I was crashing the machine. It was usually in the > 17000k range. I'm have a different problem though. I'm using 'mke2fs' version alpha 0.2b which came with SLS 0.99pl6 package. It worked perfectly before the memory upgrade. I dropped back down to 8 megs of memory and reconfigured the kernel for supporting > 16 megs memory. After creating the kernel, I then powered the machine down and upgraded the machine to 16 megs. For a couple of days, my machine was working fine, but now it is slowly eating away my ext2 partition. I keep getting 'inode already cleared...' Lots of the files have permission bits set to '?---------'. /usr/bin/less is one of the dead files. I understand that there is a new ext2fs 0.2c & 0.99pl8? I'm hoping that I have enough binaries to build a new kernel & re-compile ext2fs... I also had problems with 'rm'. It seemed to hang the shell (bash) when I typed 'rm -rf' on a temporary directory. I tried killing the 'rm' process also, but no luck. The only way out is a 'sync;reboot' on a a different virtual console. > To then get some comparison, I booted with the other disk so I was > limited to only the first 16 megs of memory. Right off the bat I > opened 4 idraws with no problems whatsoever. As I am typing this > message I am using the same session I once had the idraws on. It is > quite obvious that the difference between with the patch an without > the patch is: one works, one doesn't (for me). > I guess the questions that need to be asked are: has anyone else had > this happen? has anyone else done these tests and had it work? if so > what is your config? is it simply a case that no one actually has 32 > megs of memory and hence this code never gets tested? if my problem is > serious is there one person in particular i should be telling this to? > > -- > Phillip Andrew Rzewski Internet: kutcha@acm.rpi.edu > "Destruction is not negative, you must destroy to build." > -- Einsturzende Neubauten "Zeichnungen Des Patienten O.T." > I also added a Toshiba 3101(?) SCSI CD-ROM. I took this off a PS/2 system that had some IBM proprietary SCSI interface. I just pulled out that card and now is a standard SCSI CD-ROM. Does anyone have a drive like this? Does it work with the current port of xcdplayer? I just want to play my music on it. Linux does recognize that it is a CD-ROM, however. I've verified that there is no SCSI address conflict because my system worked with the CD-ROM under 8 megs of ram, but 16 megs gives my hard disk some problem. Any help/suggestion(s) greatly appreciated. Linus? BTW, my $0.01 worth: I've shrunk my DOS partition down to 40 megs (for games). Almost gone! Heck, I haven't played games in a while, maybe I just get rid of DOS completely... 8) -- Jim Fridlund U of MN - Computer Science fridlund@mermaid.micro.umn.edu Institute of Technology ------------------------------ From: dd435157@longs.LANCE.ColoState.Edu (Dan Doner) Subject: Re: X window GUI Builder for linux? Date: Wed, 14 Apr 1993 00:49:13 GMT The interface builder I have found that I really like is SUIT: Simple User Interface Toolkit. Its very easy to learn and produces motif-looking widgets. It is available on sunsite.unc.edu in /pub/linux/X11/devel, called suit.tar.Z. The original (non-linuxed) distribution is available at uvacs.cs.virginia.edu. Another thing I like about this package is that there are versions for DOS and Windoze. I've heard about another interface builder that runs in a kind of shell-interpreted environment (tsx or tcx ?) Anyone care to comment about it? Hope this helps, Dan ------------------------------ Crossposted-To: comp.editors From: channui@austin.ibm.com (Christopher Chan-Nui) Subject: Re: Long lines in vi (elvis) Date: Wed, 14 Apr 1993 00:45:24 GMT Reply-To: channui@austin.ibm.com Charles Hannum (mycroft@hal.gnu.ai.mit.edu) wrote: : In article dmw@teal.csn.org (Dave Warner) writes: : > : > (Put another way, the things that you tend to type over and over : > probably work -- things done less frequently, like configuring a : > .exrc file, are quite different.) : The thing which severely annoys me about all vi clones is that `8dd4..' : doesn't do the same thing as in the original. : I have yet to see a clone which does it right. : For the uninitiated, in vi: : 8 - goes into the global argument register : dd - deletes (8) lines : 4 - replaces the global argument register : . - repeats the `dd'; deletes 4 lines because the register changed : . - repeats the `dd', again deleting 4 lines : One clone repeats the whole `8dd', deleting 32 lines for the first `.'; : one ignores the `4'; one gets the `4' but forgets it later; etc. Take a look at VIM. I've got it compiled on an RS/6000 and on my Linux box. Version 1.24 has a couple bugs which you HAVE to fix to make it useful. But I am told that 1.25 has been sent to the comp.source.misc moderator and these bugs are fixed in that version. But it handles '8dd4..' perfectly :) If your anxious to get it to try it out before it appears on csm I can tell you what I changed in order to get the search/replace and ignorecase stuff to work with 1.24. It appeared on alt.sources (articles 7084-7099 at wuarchive.wustl.edu /usenet/alt.sources/articles). I must say that this is the best VI clone that I've ever seen. It has 1 or 2 annoying things (doing a !! just puts you on the : line with x,x! as the command :( But overall it is faithful to the editing features, and improves (optionally) the features that I really wanted improved. But the most important thing is COMMAND LINE RETRIEVAL AND EDITING! :) I use it as the default vi for my RS/6000, Linux and MessyDOS. Check it out. --- Christopher Chan-Nui | If cars developed like computers they would cost channui@austin.ibm.com | $1.15, go Mach 7, and get 14+ billion MPG... and #include | explode once a year killing everyone inside. ------------------------------ ** 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-Activists-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux) via: Internet: Linux-Activists@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 tupac-amaru.informatik.rwth-aachen.de pub/msdos/replace The current version of Linux is 0.99pl6 released on February 23, 1993 End of Linux-Activists Digest ****************************** 1 worth: I've shrunk my DOS partition down to 40 megs (for games). Almost gone! Heck, I haven't played games in a while, maybe I just get rid of DOS completely... 8) -- Jim Fridlund U of MN - Computer Science fridlund@mermaid.micro.umn.edu Institute of Technology ------------------------------ From: dd435157@longs.LANCE.ColoState.Edu (Dan Doner) Subject: Re: X window GUI Builder for linux? Date: Wed, 14 Apr 1993 00:49:13 GMT The interfacedigest830 644 36676 74430 57014 5366137063 5760 From: Digestifier To: Linux-Activists@senator-bedfellow.mit.edu Reply-To: Linux-Activists@senator-bedfellow.mit.edu Date: Wed, 14 Apr 93 03:30:21 EDT Subject: Linux-Activists Digest #830 Linux-Activists Digest #830, Volume #4 Wed, 14 Apr 93 03:30:21 EDT Contents: How to get rid of LILO? (Ruedi Kneubuehler) Re: Linux on the Amiga? (Larry Granroth) Re: X window GUI Builder for linux? (Dave Warner) LILO-0.10 Hangs (Curtis L. Olson) Using the basic menus with Doc (Zane H. Healy) Re: Linux Mascott/Logo (Clayton Haapala) Re: The dangers of playing with shared libraries (Steve VanDevender) Re: need help with Xconfig settings for Trident 8900c & CTX 5468A (Juha Virtanen) SCSI: Driver Needed. (Terry Evans) Installing LILO in the hard way and makehole (Juha Virtanen) xdtm-2.5 Problems (Erik Heinz) Re: Crashing machine: >16M mem patch (Damien Neil) gcc -traditional and memcpy (Tim Pierce) less and regexps (Tim Pierce) Re: New feature for the filesystems. What do you think ? (Frank T Lofaro) Re: Help - Problems upgrading to .99pl7 (Frank T Lofaro) Re: New feature for the filesystems. What do you think ? (Frank T Lofaro) ---------------------------------------------------------------------------- From: baboon@gccs.imp.com (Ruedi Kneubuehler) Subject: How to get rid of LILO? Date: Tue, 13 Apr 1993 22:57:00 +0200 Hi Steve, Tuesday April 13 1993 12:41, Steve Frampton wrote to All: SF> The answer is to use the /mbr flag of the FDISK program. Thanks to all SF> who responded. SF> Another final question -- if I do this, will I lose the data on my hard SF> drive, or is this specific to the master boot record. I'd hate to have SF> to move all 150+ MB of stuff off the drive before doing the fix. No, did it once, to get rid of the bootmanager from Coherent. All my Data where still there. ruedi :- oe. * Origin: International Origin (2:301/580) ------------------------------ Crossposted-To: comp.sys.amiga.unix From: ljg@space.physics.uiowa.edu (Larry Granroth) Subject: Re: Linux on the Amiga? Date: Wed, 14 Apr 1993 03:21:18 GMT Ed Savage (savage@viper.rtp.dg.com) wrote: : Here's the FAQ from the linux 68k mailing list... --Ed . . . : WHAT PLATFORMS WILL IT RUN ON? . . . : I expect that once we make our first public release of the system, : several groups will begin porting our sources to various platforms. : Interest has been shown by Atari programmers, mostly. There are also : a few people hoping to bring Linux up on Sun 3's. I haven't heard : from any Macintosh folks. In all probability only the startup code, ^^^^^^^^^ : interrupt code, and drivers will need re-writing to get the system : running on these systems, or any others with the necessary basic : hardware. . . . GNU software is a major component of Linux. Please consider the FSF's wishes wrt porting any of the GNU software to A.p.p.l.e. platforms. Here are a few excerpts: From /usr/local/lib/emacs/etc/APPLE: > Some of you may be considering using, buying, or recommending Macintoshes; > you might even be writing programs for them or thinking about it. Please > think twice and look for an alternative. Doing those things means more > success for Apple, and this could encourage Apple to persist in its > aggression. It also encourages other companies to try similar > obstructionism. > > [It is because of this boycott that we don't include support for Macintosh > Unix in GNU software.] From /usr/local/lib/emacs/etc/FAQ: > 101: Where can I get Emacs for my Apple computer? > > The FSF is a participant in a boycott of Apple because of Apple's "look > and feel" copyright suits. See the file etc/APPLE for more details. > Because of this boycott, the FSF doesn't include support in GNU software > for Apple computers such as the Macintosh. > > Please don't help people port or develop software for Apple computers. From /usr/local/lib/emacs/etc/MACHINES: > Macintosh > > We are boycotting Apple because of Apple's efforts to take away > our freedom to write compatible imitations of existing software. > If you value your freedom to write such programs, we urge you > not to buy from Apple, not to develop software for Apple, and > certainly not to accept a job with Apple. larry-granroth@uiowa.edu ------------------------------ From: dmw@teal.csn.org (Dave Warner) Subject: Re: X window GUI Builder for linux? Date: Wed, 14 Apr 1993 03:47:43 GMT dd435157@longs.LANCE.ColoState.Edu (Dan Doner) writes: >I've heard about another interface builder that runs in a kind of shell-interpreted >environment (tsx or tcx ?) Anyone care to comment about it? This must be Tk/Tcl developed by John Ousterhout at UC Berkeley. I can't say enough great things about it, so I won't. Lurk around in comp.lang.tcl for a while and see what its all about. BTW, there are, I believe, Linux shared libraries available for Tk/Tcl. Dave -- | Dave Warner | e-mail: dmw@csn.org | | 151 Sky Trail Road | USWest: 443-9589 | | Boulder, CO 80302 | FAX: 443-6791 | ------------------------------ From: olson@unity.cs.umn.edu (Curtis L. Olson) Subject: LILO-0.10 Hangs Date: Wed, 14 Apr 1993 03:46:51 GMT I am running linux 0.99.8 and attempted to installed LILO 0.10 Everything appeared to work flawlessly during the installation -- no error messages or any other signs of trouble. However, when I attempted to boot from my harddrive, lilo got only as far as printing "LI" then appeared to hang. I installed lilo with only the default /Image using QuickInst with all the preset defaults (nothing fancy ...) Note: lilo-0.9 worked fine for me. Any ideas? Since I am already posting, I have another question. I cannot log in as root from my first virtual console. The log in procedure appears to work fine up to the point where the shell is started, then it hangs. All other consoles (2 ... 8) work fine and I can still access them after the first console hangs. Ordinary users can log in fine from the first console, but if they log out, and then attempt to log back in, they have the same problems as root. (Things appear to hang sometime near when the shell is started.) Note, this problem disappears when I boot linux-0.99.8 from a floppy disk. Thanks for any help. Curt. (olson@cs.umn.edu) (curt@sledge.mn.org) ^^^^^^--> (my linux system) -- . --o0o-- Curtis Olson (olson@cs.umn.edu) ------------------------------ From: healyzh@iat.holonet.net (Zane H. Healy) Subject: Using the basic menus with Doc Date: Wed, 14 Apr 1993 01:32:19 GMT I've installed the complete SLS 0.99pl6 release on my machine, and when I try to use 'doc' I only get menus for file management and cut&paste. The app-defaults file is configured with a large number of menu selections, but doc doesn't seem to be reading them. Does anyone know how to get the full menu bar? Zane in DC -- Zane H. Healy | Once I had a little game healyzh@holonet.net | I liked to crawl, back into my brain on Internet | I think you know, the game I mean scnn49a on Prodigy | I mean the game, called 'go insane' (by Jim Morrison) ------------------------------ From: clay@haapi.mn.org (Clayton Haapala) Subject: Re: Linux Mascott/Logo Date: Wed, 14 Apr 1993 03:26:14 GMT >>In article <1qbvosINNplh@uwm.edu> rick@ee.uwm.edu (Rick Miller, Linux Device Registrar) writes: >>> >>>What's distinctively Finnish? >>>I can't think of a thing off the top of my head. >>>What comes to mind (other than "Linux") when I say "Finland"? >>> The sauna, of course. Linux will steam the DOS right out of your computer! Brings to mind an image of Bill Gates being beaten with birch switches in a 100 degree C sauna until he cries uncle. But I digress. But hey! If the inevitable Dr. Dobb's Linux issue shows a dipper of water (code) being thrown on a stove with names of GNU utilities printed on the rocks, etc., I want some credit! And that is pronounced 'Team Bat' to the rest of you. ;-P -- Clay Haapala "Behold the Holy Pogo Stick" clay@haapi.mn.org ------------------------------ From: stevev@miser.uoregon.edu (Steve VanDevender) Subject: Re: The dangers of playing with shared libraries Date: 13 Apr 93 21:19:18 In article dmw@teal.csn.org (Dave Warner) writes: caywood@wyvern.wyvern.com (John Caywood) writes: >Any other opinions on a minimal set of statically-linked utilities? Yes. Build them yourself. Not intended to be a flame but all these sources are freely available, easily compilable, easily statically linkable, easily installable any way you please. Its YOUR system to do with as you see fit -- try it, you'll like it. I'm certainly going to find the GNU file utilities and make statically linked versions of things like cat and ln. My recommendation was that binary distributions of Linux, such as the SLS, ought to have a minimal set of statically linked utilities so that a problem with the shared libraries won't make the system completely unusable or unfixable. It is very frustrating to find that one weak link and then discover that you can't fix it without something like a bootable rootdisk because the tool you needed to fix it broke along with everything else. The one simple way out of my problem immediately became inaccessible to me; it also seemed impossible to do anything like rename or copy one of the necessary files (although the trick with using "while read x; do; echo $x; done < file" in place of cat is a trick I hadn't thought of). Unfortunately the SLS does not come with source for the GNU file utilities so most SLS users without net access will find your solution impractical. This isn't a flame, either. I've used UNIX systems for a long time and am used to a certain amount of arcana and just having to know things in order to keep the system going. My recommendation for a core set of statically linked utilities is intended in the spirit of making Linux easier to deal with for people without the resources or the knowledge that you or I have. Until there's some agreement on what core utilities ought to be statically linked, though, I highly recommend that new Linux users keep copies of the a1/a2 SLS disks around or get HJ Lu's bootable rootdisk, which will allow you to recover from a variety of problems including mangling your shared libraries. -- Steve VanDevender stevev@greylady.uoregon.edu "Bipedalism--an unrecognized disease affecting over 99% of the population. Symptoms include lack of traffic sense, slow rate of travel, and the classic, easily recognized behavior known as walking." ------------------------------ From: jiivee@hut.fi (Juha Virtanen) Subject: Re: need help with Xconfig settings for Trident 8900c & CTX 5468A Date: 14 Apr 93 00:56:00 GMT Reply-To: jiivee@hut.fi In article johnw@unixhub.SLAC.Stanford.EDU (Jonathan S. Wong) writes: > I was wondering if anyone has gotten X to work with Trident 8900c (BIOS v.2.11) > and a Zeos (CTX model 5468A) SVGA monitor. I have only been able to get it > to work with 640x480 mode -- other settings seem to cause my screen to > flicker or just not sync. Can someone give me some possible Xconfig settings > for this combination? (I don't have specs for the monitor-- just bought it > used) Yes. I've succeeded, except that I have a ADI dmc-2304 SVGA monitor which supports only few clocks. I include parts of my Xconfig here (unnecessary lines deleted to save bandwidth). Behavior of these clocks is somehow strange: if I remove the needless clock "31.5" won't X run any more in 800x600 resolution. # The graphics drivers # First the 8-bit colour SVGA driver # vga256 Chipset "tvga8900c" Clocks 25.2 28.3 31.5 35.5 # clocks supported by ADI dmc-2304 #Clocks 25.2 28.3 31.5 35.2 35.5 36.0 # clocks supported by Trident 8900c2.11 #Clocks 25.2 28.3 40.0 44.9 36.0 50.4 57.3 65.0 75.0 Screenno 0 Modes "800x600" "640x480" Videoram 1024 #ViewPort 0 0 Virtual 1152 900 # # To disable SpeedUp, use NoSpeedUp #NoSpeedUp # # Next the 1-bit mono SVGA driver # vga2 Virtual 800 600 ViewPort 0 0 # Modes "640x480" Modes "800x600" "640x480" # And last, the database of video modes # ModeDB # name clock horizontal timing vertical timing flags "640x480" 25 640 664 760 800 480 491 493 525 "800x600" 36 800 824 896 1024 600 601 603 625 -- Juha ------------------------------ From: tevans%sunset.cs.utah.edu@cs.utah.edu (Terry Evans) Subject: SCSI: Driver Needed. Date: 13 Apr 93 22:41:24 MDT I just bought a computer and It came with a Data Technology EISA SCSI Host Adapter. The model is DTC3292. I have tried everyhting that I know of and I can't get Linux to work with it. Is there anybody out there in the Linux world that knows how to use the Controller with Linux??? Any help, comments, suggestions, etc would be greatly appreciated. Should I get another board? Thanks! tevans@cs.utah.edu ------------------------------ From: jiivee@hut.fi (Juha Virtanen) Subject: Installing LILO in the hard way and makehole Date: 14 Apr 93 02:18:11 GMT Reply-To: jiivee@hut.fi >>>> On 13 Apr 93 12:41:21 GMT, frampton@vicuna.ocunix.on.ca (Steve Frampton) said: :> Another final question -- if I do this, will I lose the data on my hard :> drive, or is this specific to the master boot record. I'd hate to have :> to move all 150+ MB of stuff off the drive before doing the fix. No, you shouldn't. An horror story about my LILO installation follows. I have two IDE drives on my system. I just installed LILO for the first time and I screwed up almost everything with boot sectors and MBRs I could. My first drive (entire drive is in one primary dos partition) wasn't recognized, so I had to do FDISK /MBR several times. Then, my dos partition didn't boot and I had to retransfer the system onto it. At last my dos drive were ok again and I read the README once again (I didn't have the paperdoc available yet). I checked everything once again and got LILO to boot up linux. But it refused to install dos at all, so I checked my linux drive with fdisk and it reported that my root partition's physical and logical start points differ. I had to backup my root and mkfs to it and restore the files. Lucky me at this point because I had just day before downloaded HLu's bootable rootdisk (before that I had only SLS-0.98.5 a1 and a2 disks to bootup from floppy!). In the beginning of re-making my root partition (its size is 16MB) I had there about 3800KB free space. After restoring files back to fresh partition I had only about 1300KB free space! So, I remembered a utility called "makehole" and played with it quite a lot and free space increased dramatically, to about 3700KB. Then it was sleeping time. In the morning I woke up and found my system crashed due to problems with root partition's minix-fs. I had to boot from floppy and do fsck onto it a few times (three or four times, so that it didn't complain anymore). Then I checked free space once again and got really surprised. There were -- and still is -- about 6300KB free space! Now I don't know if I should be worried about that since everything works fine so far... -- Juha ------------------------------ From: pih@rz.uni-jena.de (Erik Heinz) Subject: xdtm-2.5 Problems Date: Tue, 13 Apr 1993 21:02:55 GMT Hi, on sunsite.unc.edu I found a binary version of xdtm-2.5 (X Desktop Manager) for Linux. Great program! Unfortunately, it doesn't run properly on my Linux box. At least the following problems I've noticed: - After calling the file information dialog I get an error message on the xterm that launched xdtm: "xdtm: Internal Error, permissions dialog not free'd" Sometimes xdtm crashes after this and dumps core. - Double clicking an executable file doesn't execute it, but calls the file information dialog. - 'set ignore' in .xdtmrc has no effect. - The long listing mode doesn't work at all. The short listing mode works, but needs an additional 'goto' command to show the directory listing. My attempts to compile the xdtm-2.5 sources failed so far. Although there are the binaries on sunsite, I couldn't find any patches for Linux provided. Any help to fix these problems is greatly appreciated. Thank you in advance. Erik Heinz (pih@rz.uni-jena.de) ------------------------------ From: damien@b63519 (Damien Neil) Subject: Re: Crashing machine: >16M mem patch Date: 14 Apr 1993 01:34:17 GMT Reply-To: dpn2@po.cwru.edu Jim Fridlund (fridlund@mermaid.micro.umn.edu) wrote: : In article kutcha@eos.acm.rpi.edu (Phillip Rzewski) writes: : I, too, just upgraded from 8 megs to 16 megs and have the same exact : motherboard as yours. The only thing different is that I have a 33 : Mhz Intel chip & 128K cache. I'm also experiencing some problems with : my Linux workstation (ver. 0.99pl7). :) A thought -- did you just add memory on to your old memory, or did you replace the old SIMMS? Y'see, if you have memory with different speeds in your machine, things might start going wrong. Just a thought. -- Damien Neil dpn2@po.cwru.edu "Until somebody debugs reality, the best Case Western Reserve University I can do is a quick patch here and there." CMPS/EEAP Linux -- the choice of a GNU generation. - Erik Green ------------------------------ From: twpierce@unix.amherst.edu (Tim Pierce) Subject: gcc -traditional and memcpy Date: Wed, 14 Apr 1993 05:32:19 GMT I haven't seen this mentioned in the FAQ. Please excuse me if it's old territory. I'm running 0.99.7a of the kernel and a fairly recent SLS distribution with gcc 2.3.3 and libc-4.3.3. While trying to make rpcgen, snavtched from wuarchive (why? I had this paranoid vision of losing part of the source for mount and needing to rebuild the various modules from mount.x by hand), I noticed that compiling a program referencing memcpy or memcmp with the -traditional switch yields the following error messages: # gcc -O -traditional test.c In file included from test.c:1: /usr/include/string.h:38: warning: conflicting types for built-in function `memcpy' /usr/include/string.h:38: warning: type mismatch with previous external decl :0: warning: previous external decl of `memcpy' /usr/include/string.h:63: warning: conflicting types for built-in function `memcmp' /usr/include/string.h:63: warning: type mismatch with previous external decl :0: warning: previous external decl of `memcmp' Since experimentation on Ultrix and SunOS systems didn't produce the same problem, I poked around a little in the Linux include files and found the inlined functions for memcmp, memcpy, and the various string functions in include/linux/string.h under the Linux source directory. I presume that what's happening is that the GNU include files in /usr/include provide prototypes that clash with the ones defined in the kernel. Granted, the example above demonstrates only warning messages that don't impede the linking of a working executable, but (a) it would be nice to eliminate the warnings if possible, and (b) it's simply never a good idea to let incompatibilities like this linger, for they may come back to haunt you when you can least deal with it. Might it be a good idea to fiddle with the GNU-supplied include files so that they jibe with the kernel prototypes? Or, for that matter, should the kernel prototypes be changed to work with the GNU include files? -- ____ Tim Pierce / ?Usted es la de la tele, eh? !La madre \ / twpierce@unix.amherst.edu / del asesino! !Ay, que graciosa! \/ (BITnet: TWPIERCE@AMHERST) / -- Pedro Almodovar ------------------------------ From: twpierce@unix.amherst.edu (Tim Pierce) Subject: less and regexps Date: Wed, 14 Apr 1993 05:35:54 GMT Is it just me, or does less (version 1.77, SLS-supplied) not accept regular expressions in searches under Linux? It certainly doesn't behave this way in general. (Yes, I can rebuild it, once I spend a half-hour or so downloading the source over a 9600-bps link, but it seems peculiar that it wouldn't work in the SLS distribution). -- ____ Tim Pierce / ?Usted es la de la tele, eh? !La madre \ / twpierce@unix.amherst.edu / del asesino! !Ay, que graciosa! \/ (BITnet: TWPIERCE@AMHERST) / -- Pedro Almodovar ------------------------------ From: Frank T Lofaro Subject: Re: New feature for the filesystems. What do you think ? Date: Wed, 14 Apr 1993 01:53:36 -0400 Excerpts from netnews.comp.os.linux: 13-Apr-93 Re: New feature for the fil.. HJ Lu@eecs.wsu.edu (463) >I am wondering how a compressed filesystem handles demand paging and >random update. I had just posted wondering about demand-paging, but came up with an idea. Have it uncompress the file, and put it on some temporarily allocated inodes, and exec/demand page it from there. It would require some changing of filesystem allocation/exec/demand paging routines I gather, but should be feasible. ------------------------------ From: Frank T Lofaro Subject: Re: Help - Problems upgrading to .99pl7 Date: Wed, 14 Apr 1993 01:49:03 -0400 Excerpts from netnews.comp.os.linux: 13-Apr-93 Help - Problems upgrading t.. Keith Barrett@pamsrc.ene (572) > 2. SELECTION no longer functions, and returns a IOCTL failure. I have > heard this is patchable. Could someone send me the details and > pointers? I hope selection becomes part of the kernel before the kernel is called version 1.0 ------------------------------ From: Frank T Lofaro Subject: Re: New feature for the filesystems. What do you think ? Date: Wed, 14 Apr 1993 01:35:57 -0400 It sounds like a useful idea for those pressed on diskspace, and seems worth implementing. Some problems that need to be addressed: 1. Speed of access. 2. Demand-paging. It'll probably be really hard to get this to work with it. 3. Backwards compatibility. Anyone out there know if there are any spare fields or bits (seems like this idea only needs to steal one) in the inode structure of the current filesystems? 4. How to have the kernel actually decompress? Exec an outside program? (Hard to do, much kernel internals hacking, memory foo to deal with) Work with a daemon? (What do you do if the daemon dies, etc? How does the daemon start, and talk to the kernel?) Build gunzip into the kernel? (Might be difficult, but then, the kernel does decompress itself, so maybe we can steal that code.....) This does seem a lot better than trying to compress the whole FS (not quite as efficient for space, but probably a hell of a lot easier to implement). Anyway, good luck with this idea. It has been mentioned before, but this time there is some proposed way of handling some of the implementation details; it seems quite feasible. I don't know enough about filesystems to know how easy/hard this will be tho... ------------------------------ ** 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-Activists-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux) via: Internet: Linux-Activists@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 tupac-amaru.informatik.rwth-aachen.de pub/msdos/replace The current version of Linux is 0.99pl6 released on February 23, 1993 End of Linux-Activists Digest ****************************** !La madre \ / twpierce@unix.amherst.edu / del asesino! !Ay, que graciosa! \/ (BITnet: TWPIERCE@AMHERST) / -- Pedro Almodovar ------------------------------ From: twpierce@unix.amherst.edu (Tim Pierce) Subject: less and regexps Date: Wed, 14 Apr 1993 05:35:54 GMT Is it just me, or does less (version 1.77, SLS-supplied) not accept regular expressions in searches under Linux? It certainly doesn't behave this way in general. (Yes, I can rebuild it, once I spend a half-hour or so downloaddigest831 644 36676 74430 56555 5366137064 5773 From: Digestifier To: Linux-Activists@senator-bedfellow.mit.edu Reply-To: Linux-Activists@senator-bedfellow.mit.edu Date: Wed, 14 Apr 93 05:15:14 EDT Subject: Linux-Activists Digest #831 Linux-Activists Digest #831, Volume #4 Wed, 14 Apr 93 05:15:14 EDT Contents: Another Elvis Bug (Bill Reynolds) Re: New feature for the filesystems. What do you think ? (Daniel Aaron Supernaw-Issen) ADAPTEC1542A does it work ? Re: Why Linux should remain Linux and seagulls should not be used. (Simon Patrick Janes) Re: Kernel crash: Need help fixing my Ext2fs (Frederick W. Umminger) Desperately looking for a Linux copy (Janjerdsak Jirapat) Re: The dangers of playing with shared libraries (David I. Bell) Re: Why Linux should remain Linux: Deja Vu (James R. Wiegand) Where is the "freq" program? (John Perry) Re: tape retention? (Mike Frankowski) Re: SCSI Device ERRORS! (Mike Frankowski) SYSV IPC:function not implemented (Damien Neil) Kernel panic, Help !!! (Matthew Roderick - KID01) SLS install notes (Jon Jagger) Re: Dos 6/LILO/Linux -- anyone look? (Remco Treffkorn) Dumb newbie question! (Alex) Re: The dangers of playing with shared libraries (Steve VanDevender) Crashing machine >16M: Update (Phillip Rzewski) Re: Kernel panic, Help !!! (KID01) TCP/IP Networking problems (Hunyue Yau) ---------------------------------------------------------------------------- From: bill@yossarian.ucsd.edu (Bill Reynolds) Subject: Another Elvis Bug Date: 13 Apr 93 19:45:15 Reply-To: bill@goshawk.lanl.gov Try the following :map ^V ^V^F (map space to ^F to use the space bar for paging) and then 0 (go to beginning of a line). Voila! Instant coredump. -- ______________________________________________________________________________ Bill Reynolds | Hey Rocky! Watch me pull a rabbit out of my hat! bill@goshawk.lanl.gov | -Bullwinkle ------------------------------ From: danielsi@cs.utexas.edu (Daniel Aaron Supernaw-Issen) Subject: Re: New feature for the filesystems. What do you think ? Date: 14 Apr 1993 01:20:46 -0500 about paging and random update: it seems to me that the solution is simple. don't compress the file, compress on the block level. You KNOW that when uncompressing a block how big it will be. You have to read in the whole block anyways the issue of how big it is on disk is orthogonal to everything else. prepend the block with a byte identifying what compression alg (if any was used) and all is taken care of. I'm currently working on implementing exactly this on a lfs. In this case, a number of different compression algorithms are used and the one with the best compression (including no compression) wins. the only real change to the surrounding lfs code was that the head position had to be byte indexed rather than block indexed. no biggie. Daniel Supernaw-Issen danielsi@cs.utexas.edu Linux or Die! ------------------------------ From: p7003ag@sun1.LRZ-Muenchen.DE () Subject: ADAPTEC1542A does it work ? Date: Wed, 14 Apr 1993 06:30:23 GMT Yoho, I tried to boot the a1 disk from the SLS packages on a Compaq Deskpro with an ADAPTEC 1542A and a Conner 200MB Harddisk. The Kernel unpacks, checks for and finds my disk correctly, but then goes into an indefinite loop, trying to reset the controller and failing. Has anybody ever had the same problem and if so, what is the solution ? (Other than: get yourself another controller, coz I can't) Glad if I could find some help :) Alaric ------------------------------ From: spj@ukelele.GCR.COM (Simon Patrick Janes) Subject: Re: Why Linux should remain Linux and seagulls should not be used. Date: Wed, 14 Apr 1993 03:48:46 GMT hpa@merle.acns.nwu.edu (H. Peter Anvin N9ITP) writes: >Besides the obvious that it is a tip of the hat to Linus and that the >OS is already becoming known under that name, consider it to be yet >another UNIX like OS with a recursive acronym (see: GNU, XINU): >inux s ot NI I like this. Its kinda like a stepping stone to GNU's Not Unix. I don't like the seagull idea however. Seagulls are too common I think... I don't know... it just makes me think what America would have been like if the national bird was the turkey like it almost became. I would rather have a platypus than a seagull. >;-) > /hpa >-- >INTERNET: hpa@nwu.edu FINGER: hpa@eecs.nwu.edu >BITNET: HPA@NUACC IBM MAIL: 36073 at IBMX400 >HAM RADIO: N9ITP, SM4TKN NeXTMAIL: hpa@speedy.acns.nwu.edu >Most inappropriatly named startup command, winner: Microsoft Windows -- ========================================================================= Simon "Guru Aleph-Null" Janes |Taco Bell Otaku-zoku... | | do they exist? Huh? Huh? ------------------------------ From: umminger@purina.berkeley.edu (Frederick W. Umminger) Subject: Re: Kernel crash: Need help fixing my Ext2fs Date: 14 Apr 1993 06:23:16 GMT I should mention this also, now that I've noticed it; my ext2fs apparently ran out of space, at least for normal users, just before the crash. Taylor uucp is taking up 17 megs of disk space! How absurd. Has ext2fs been tested extensively for its behaviour when it runs out of space? -Frederick Umminger umminger@math.berkeley.edu ------------------------------ From: g9321160@wampyr.cc.uow.edu.au (Janjerdsak Jirapat) Subject: Desperately looking for a Linux copy Date: 14 Apr 1993 11:03:08 +1000 Lines: 4 I've recently heard about Linux. It would be great if someone can tell me how can I get a copy of this fabulous O/S. I'm living in Sydney Australia. Is it possible to get it through the Internet? I'm desperately looking for it. Thanks for your help! ------------------------------ From: dbell@pdact.pd.necisa.oz.au (David I. Bell) Subject: Re: The dangers of playing with shared libraries Date: Wed, 14 Apr 1993 01:52:29 GMT >(2) A statically-linked shell has been suggested. Of the currently available > ones, I would have to suggest rc just because it's compact... the concept > of a statically linked "bash" makes me think of a "bashdisk" containing > nothing *but* bash because nothing else will fit :-) With respect to statically linked shells and a small set of needed utilities, I am about to announce "sash", which is designed to solve all these problems. It should appear within a couple of days. (I don't have ftp write access, so I have to ask to put it on an ftp site.) If that fails, than I'll post it here as a shar file. -dbell- ------------------------------ From: jweigand@astro.ocis.temple.edu (James R. Wiegand) Subject: Re: Why Linux should remain Linux: Deja Vu Date: 14 Apr 93 04:33:32 GMT In article <1993Apr13.045652.23534@news.acns.nwu.edu> hpa@nwu.edu (H. Peter Anvin) writes: >Besides the obvious that it is a tip of the hat to Linus and that the >OS is already becoming known under that name, consider it to be yet >another UNIX like OS with a recursive acronym (see: GNU, XINU): > >inux s ot NI > > /hpa >-- See, I knew it would catch on! Only took six months. jim 30 degrees ahead of the timewarp ------------------------------ Crossposted-To: comp.windows.x.i386unix From: jperry@oahu.cs.ucla.edu (John Perry) Subject: Where is the "freq" program? Date: Wed, 14 Apr 93 07:20:11 GMT I got my mouse to work with Xfree, thanks to those of you who pointed out that the serial devices are ttyS0 through ttyS3 (as opposed to ttys0 thru ttys3). Now I'm trying to get the right clocks for my video card to allow me to do more than just 640x480. Someone mentioned a program called freq that would determine the dot-clocks for me, but they didn't mention where to find it. I also haven't found it in any of the FAQs (linux, Xfree86, etc.). If you know where I can get this program, please let me know. thanks in advance, John p.s. Thanks a million Linus and all you other developers for providing us with a real OS for our intel machines. ------------------------------ From: mfrankow@piper.hamline.edu (Mike Frankowski) Subject: Re: tape retention? Date: Wed, 14 Apr 1993 02:22:50 GMT L.G. "Ted" Stern (stern@amath.washington.edu) wrote: : In article <1993Apr13.192932.12892@nntpd.lkg.dec.com>, barrett@pamsrc.enet.dec.com (Keith Barrett) writes: : |> : |> Dumb question warning -- : |> : |> How do you do a tape retention under Linux? My drive runs very hot : |> and I need to do this often to make sure all works well. : |> : |> : |> Thanks in advance : |> : |> : |> Keith Barrett type mt -f /dev/ retension replace with the tape device. Should work wonders. Mike ------------------------------ From: mfrankow@piper.hamline.edu (Mike Frankowski) Subject: Re: SCSI Device ERRORS! Date: Wed, 14 Apr 1993 02:28:46 GMT Joseph D. Noe (noe@edmund.cs.andrews.edu) wrote: : Thanks : Joe : noe@andrews.edu woudl lvoe to help if we could know what the problems are/were... Mike ------------------------------ From: damien@b63519 (Damien Neil) Subject: SYSV IPC:function not implemented Date: 14 Apr 1993 03:01:04 GMT Reply-To: dpn2@po.cwru.edu I'm trying to get SYSV IPC working on my 0.99.8 kernel. The diffs installed nicely for the most part, and I added the ones that didn't by hand. The kernel compiled without complaint. However, whenever I run a program that uses IPC, I get a `Function not implemented.' error. I'm at a loss. Does anyone have any suggestions? Many thanks! -- Damien Neil dpn2@po.cwru.edu "Until somebody debugs reality, the best Case Western Reserve University I can do is a quick patch here and there." CMPS/EEAP Linux -- the choice of a GNU generation. - Erik Green ------------------------------ From: mat@eccles.dsbc.icl.co.uk (Matthew Roderick - KID01) Subject: Kernel panic, Help !!! Date: 14 Apr 93 07:33:54 GMT Could anyone tell me what this means (other than, 'theres a problem with memory') ++ Unable to handle kernel paging request at address d445c700 Oops: 0000 EIP: 0008:0000B550 EFLAGS: 00010207 fs: 0017 base: 00000000, limit: 000A0000 pid: 0, process nr: 0 06 00 76 58 3f 00 7f 48 8b ++ I'm trying to install SLS pl6 onto my 486/33 with 4/8meg of ram (tried both) & IDE. I get as far as loading the ram disk from the first disk an it bombed out! Cheers Matthew ------------------------------ From: Jon Jagger Subject: SLS install notes Date: 14 Apr 1993 04:49:11 -0400 Reply-To: J.R.Jagger@sheffield-hallam.ac.uk Hi there, just thought I'd post a few offerings after I found while installing the latest SLS 1.0 1. the c proggy to test if the mouse is working (can't remember where it is) is missing the line #inlude 2. The file xinitrc has a line OPENWINHOME=/u/........ which should read OPENWINHOME=/usr/.......... 3. to get some of the man/cat pages to work in X you need to create two links. geqn -> neqn and gtbl -> tbl. 4. A few C proggys that compiled before, did not this time. The problem was the /usr/include/linux directory was a a true directory with a few files in it and not a link to /usr/src/linux/include/linux (I think that's right) I just moved the files in /usr/include/linux, deleted the dir and created the link, and all compiled fine then. I needed to do this for a kernel recompile as well. Apart from that it was a breeze. One thing though. X now takes a LOT longer to start up, then the previous SLS did. I'm using a Trident 8900C card :( (the scrolling is painful). Thanks again. JJ :: Jon Jagger J.R.Jagger@shu.ac.uk :: Sheffield Hallam University, Pond Street, SHEFFIELD S1 1WB :: Tel 0742 533802/432889 (work/home) Fax 0743 533840 :: Newspaper ad: Men wanted for expanding contracting company! ------------------------------ From: remco@hip-hop.suvl.ca.us (Remco Treffkorn) Subject: Re: Dos 6/LILO/Linux -- anyone look? Date: Wed, 14 Apr 1993 01:21:08 GMT djr48312@uxa.cso.uiuc.edu (Dennis Robinson) writes: : bmyers@shyguy.lonestar.org (bob myers) writes: : : >In about two weeks, I'll be rebuilding my existing system (340 mb + : >676 mb ESDI dedicated to SVR4 unix) completely. I plan to install : >DOS on the smaller drive, while fully using the larger for Linux. : : >Question is -- has there been any exposure of DOS 6 with Linux and : >the latest LILO package? : : >Minor side point -- guess that a DOS 6 compressed file structure would : >not be accessable to Linux. Any thoughts on whether MS would : >discuss the compressed file structure with the outside world? : : : >-- : >----------------------------------------------------------------- : >Bob Myers | You gotta watch out for those : >Shyguy's AT&T SVR4 Unix/Uniboard | shy guys....you never know : >Email:bmyers@shyguy.lonestar.org | what they're thinking..:) : : You must not be a real Unix player if you are supporting : MS'es latest doggy operating system. If we complain about their : BS but still buy their fucking products than who is getting : pimped? If everyone buys into their stuff then in the future thre : will be no choice of an os. MS can then call other os'es "propietery". : Fuck em. Pretty good analysis! That's exactly what is going to happen. But who said we are not supposed to having some fun while we are getting screwed? As hard as you might wish: messdos/windoze and friends are not going to go away. But look at it as 'bootcamp' for real gurus. Some will actually make it... Seems to me that many (most?) of the participants here at c.o.l have some background in doze. The really KNOW what they hate and why. Without microschrott this group would be pretty deserted. Some of us here might actually have liked it better this way, but I really get a lot of kicks out of all the 'newbies' questions. Those people are courageous enuff to tackle an operating system with a terrible reputation. Complex, complicated, huge... They have the guts to ask 'silly' questions, take the risk of getting flamed, are willing to read annoyed 'RTFMs' comments AND ultimately learn alot in the process. You want pioneers? There they are! Many of them gained access to the net for the first time. Sure they do not know the 'protocol'. Teach 'em, don't flame 'em. (or at least do it in a fun way, so that I get something out of it!) Seriously: Linux success is created in c.o.l. by hobbyists that probably bit off more than they can swallow. Help 'em and they will bring in more people and teach their friends before they become noise on the net. Wait till these masses hit the market and force (enable) employers to look into **ix. But do not believe for a second that we can make a dent in the doze market. I better stop it here... Remco ------------------------------ From: toejam@panix.com (Alex) Subject: Dumb newbie question! Date: Wed, 14 Apr 1993 04:56:58 GMT Sorry to bother all of you with this but I couldn't find an answer in the FAQ. I currently have the SLS distribution of linux on my hard drive and am ready to install! Now I know I'm supposed to rawrite A1 and A2 to their respective floppies. What I was wondering is whether the installation can be switched to the B: drive after the initial boot. The reason I'm asking is because I despise 5 1/4 floppies and would like to install the rest of linux from the B: drive (3 1/2). Is this possible? Also I've been told that X windows is a beast if you only have 4 megs of RAM. I'm looking to get 4 more megs but I'm kinda broke right now. How much of a beast is X at 4 megs...:) Thanks! -- L I V E F R O M N E W Y O R K - Alex Diamantis ================================================\ Email: ToEjAM@pAnIX.cOM "I don't know if there are men on the moon, but \________________________ if there are, they must be using Earth as their lunatic asylum." - Shaw ------------------------------ From: stevev@miser.uoregon.edu (Steve VanDevender) Subject: Re: The dangers of playing with shared libraries Date: 14 Apr 93 00:51:53 In article <1qfot4$tin@hal.gnu.ai.mit.edu> ericy@hal.gnu.ai.mit.edu (Eric Youngdale) writes: In article <1993Apr13.161616.24954@kf8nh.wariat.org> bsa@kf8nh.wariat.org (Brandon S. Allbery) writes: >(2) A statically-linked shell has been suggested. Of the >currently available ones, I would have to suggest rc just >because it's compact... the concept of a statically linked >"bash" makes me think of a "bashdisk" containing nothing >*but* bash because nothing else will fit :-) This is what the bootable rootdisk is for. I have yet to see a good argument for why the bootable rootdisk cannot be used instead of specially staticly linked binaries to fix screwups with the sharable libraries. A good argument: If you trash the shared libraries you can't even run sync! If you use a bootable rootdisk you have to shut the system down, but you risk filesystem damage because you couldn't sync first. shutdown or halt won't work with the shared libraries gone either. Ideally, if you haven't panicked the kernel then you really shouldn't have to reboot the machine. Something simple like mashing the shared library links doesn't destroy the system to the point of unusability _except_ that absolutely nothing will run, including the commands you could have used to get things going again without a reboot. If only I had thought of using a while/read/echo loop in one of the shells I still had running to copy libc.so.4.3.3 over the top of libc.so.4 . . . Perhaps I am harping on this point too much, since I have now recovered from my mistake and I managed to sleep off the incredible fit of anger I experienced when I discovered that I had made it. I am simply recommending that the basic set of Linux utilities have a few critical statically linked members (like 'ln'), so that some common mistakes can be recovered from more quickly and conveniently. -- Steve VanDevender stevev@greylady.uoregon.edu "Bipedalism--an unrecognized disease affecting over 99% of the population. Symptoms include lack of traffic sense, slow rate of travel, and the classic, easily recognized behavior known as walking." ------------------------------ From: kutcha@eos.acm.rpi.edu (Phillip Rzewski) Subject: Crashing machine >16M: Update Date: 14 Apr 1993 03:14:47 GMT Hello again everyone. For those that have been following the thread, I am the same person who posted the bit about "Crashing machine with more than 16 megs of memory" patch applied and all. I have gotten several replies and tried many things and thought that those of you following the thread would appreciate an update. Actually, I also think I should add a couple things I realized I left out in my first post. Oh, first the things I left out. I have been testing this under 0.99p8 (though I also had made it crash under p6 and p7). Also, the memory itself is all the same speed (80ns) so it's not possible that it could be a speed problem. (Though I would be skeptical of that as a problem in the first place. Someone explained to me that if you mix speeds it all just runs at the speed of the highest in the mix, which makes sense from an architectural standpoint.) As for the updates. Someone suggested that maybe it was a bad SIMM. I didn't want to admit this as a possibility at first (it was easier to blame Linux :) ) so I cracked open the machine and ran a battery of tests. It all came up negative. In the end I just SWITCHED the 4 SIMMS that have been working as my 16 megs for the past month with the other 4, and have been running fine with them for days now. So, even though it would have been an easier solution to the problem, it looks like my memory is fine. Also someone pointed out to me that machines with a small cache and lots of memory tend to run really slow. This explains a lot to me. I have just a 386/25 with a 64k cache, and I noticed that when I upgraded from 8 megs to 16 megs that it spent less time swapping (of course) but it seemed to be running much slower when doing normal work with the memory. So obviously this is linked to the size of my cache. In other words, even if I COULD get all 32 megs to work, with a 64k cache it's likely that it would be slower than I was at 8 megs with swapping! Nevertheless, these answers don't explain why it broke when I applied the patch and tried to use all the memory. As another helpful soul pointed out, the patch is amazingly simple and is only in place to watch out for things like the cache problem. I guess the patch can't be blamed then. What is left? The motherboard? Any other PD 386 owners try to go to 32 megs? -- Phillip Andrew Rzewski Internet: kutcha@acm.rpi.edu "Destruction is not negative, you must destroy to build." -- Einsturzende Neubauten "Zeichnungen Des Patienten O.T." ------------------------------ From: KID01 Subject: Re: Kernel panic, Help !!! Date: 14 Apr 93 08:06:03 GMT Could anyone tell me what this means (other than, 'theres a problem with memory') ++ Unable to handle kernel paging request at address d445c700 Oops: 0000 EIP: 0008:0000B550 EFLAGS: 00010207 fs: 0017 base: 00000000, limit: 000A0000 pid: 0, process nr: 0 06 00 76 58 3f 00 7f 48 8b task[0] (swapper) killed: unable to recover kernel panic: Trying to free up swapper memory space In swapper task - not syncing ++ I'm trying to install SLS pl6 onto my 486/33 with 4/8meg of ram (tried both) & IDE. I get as far as loading the ram disk from the first disk an it bombed out! other times it just resets or hanges and some times I can get as far as making the file systems !!! Cheers Matthew Oops Missed alittle out on the last posting !! -- Matthew Roderick -*- B2 Developments -*- mat@{dsbc,oasis}.icl.co.uk +44 [0]782 771000 x3270 -*- UNIX Centre, ICL, Stoke-on-Trent, England. Spelling & typing errors have been left for orthentisity ------------------------------ From: huny@cco.caltech.edu (Hunyue Yau) Subject: TCP/IP Networking problems Date: 14 Apr 1993 05:16:11 GMT Does anyone have any clue as to what is wrong: System: 386DX40 UMC Chipset, AMI Bios IDE HD, 4S, 2P Generic WD clone network card. Linux 0.99pl8; upgraded from SLS Linux pl 4 disk A,B,C,X Problem: Telneting to another machine works initially, but after a short while, it just hangs. I can ^[ back to my linux machine and 'c'lose the connection. However, the connection does not close on the remote end. FTP also hangs randomly. Telneting remotely to my linux machine also hangs randomly. I tried using NCSA Telnet in MS-DOS and it works fine, so I am assuming it is not a hardware problem. Any ideas as to what may be wrong or what I need to get? Email if possible. I'll summerize if there's any interest. If any relevant information is missing, please let me know Thanks. Hun huny@cco.caltech.edu ------------------------------ ** 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-Activists-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux) via: Internet: Linux-Activists@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 tupac-amaru.informatik.rwth-aachen.de pub/msdos/replace The current version of Linux is 0.99pl6 released on February 23, 1993 End of Linux-Activists Digest ****************************** king as my 16 megs for the past month with the other 4, and have been running fine with them for days now. So, even though it would have been an eadigest832 644 36676 74430 56127 5366137064 5767 From: Digestifier To: Linux-Activists@senator-bedfellow.mit.edu Reply-To: Linux-Activists@senator-bedfellow.mit.edu Date: Wed, 14 Apr 93 06:15:08 EDT Subject: Linux-Activists Digest #832 Linux-Activists Digest #832, Volume #4 Wed, 14 Apr 93 06:15:08 EDT Contents: Re: Why Linux should remain Linux and seagulls should not be used. (Jan Jitze Krol) Re: Compiling the gopher server for linux (Hendrik G. Seliger) SLS or TAMU - help me decide (Dan Doner) Re: NFS Problem (Christopher Kuperman) Re: Do I need intelligent serial I/O?? (Christopher Kuperman) Re: Desperately looking for a Linux copy (Christopher Kuperman) Help, Future Domain SCSI TMC-850 NOT WORK! (slvch@cc.usu.edu) urt Problem (Holger Burde) LINUX and data acquisition (Dan Doner) *** READ THIS BEFORE POSTING *** (Ian Jackson) Re: Port Fido Mailer to Linux (Louis Lagendijk) menu problem solved (Eckard Brauer) Re: The dangers of playing with shared libraries (Keith Smith) probs with 4.3.3-shlibs and solution (Eckard Brauer) Re: The dangers of playing with shared libraries (Dietmar Braun) Searched: cc1obj (Eckard Brauer) Backup libraries (was Re: The dangers of playing with shared libraries) (KID01) ---------------------------------------------------------------------------- From: meajjk@ercx44.Skferc.Nl (Jan Jitze Krol) Subject: Re: Why Linux should remain Linux and seagulls should not be used. Date: 14 Apr 93 08:16:07 GMT |> I don't like the seagull idea however. Seagulls are too common I think... |> I don't know... it just makes me think what America would have been like Does have to be a bird? I was thinking of mustard seed: it starts small but grows big. But then, mustard seed don't look fancy. my $0.10 (high inflation) :-) Jan Jitze Krol jitze@mc.bio.uva.nl meajjk@skferc.nl ------------------------------ Crossposted-To: comp.infosystems.gopher From: hank@Blimp (Hendrik G. Seliger) Subject: Re: Compiling the gopher server for linux Date: Wed, 14 Apr 1993 08:06:03 GMT Reply-To: hank@Blimp.automat.uni-essen.de kwhelan@neptune.uark.edu (Kenneth Whelan) writes: : : The only reference to linger or SO_LINGER found in the Linux code was found : in the tcp/ip source code. net.c I believe. It was an element of the data : structure called sock->data->linger It was a single integer it was not the : same format as what gopher uses : It is in /usr/include/linux/socket.h:#define SO_LINGER 13. The only changes I had to make to the gopherd were the references to getdomainname, which isn't implemented for linux (yet). Othere than that it runs perfect for several weeks now. (I must admit, though, that I haven't tried out real long files, as I only have brief informations to serve). Hank. -- ====================================================================== Hendrik G. Seliger Universitaet Essen hank@Blimp.automat.uni-essen.de Schuetzenbahn 70 Tel.: +49-201-183-2898 4300 Essen, Germany ====================================================================== Pajari: "Handling interrupts is simple." A. Tanenbaum: "Interrupts are an unpleasant fact of life." ------------------------------ From: dd435157@longs.LANCE.ColoState.Edu (Dan Doner) Subject: SLS or TAMU - help me decide Date: Wed, 14 Apr 1993 08:36:04 GMT the subject says it all - please flood my mailbox with your suggestions, TAMU or SLS? I've used the SLS release for about a year now, and I've grown very tired of fixing things. But, I could just be getting lazy.... Thanks in advance, Dan ------------------------------ From: ent811l@monu6.cc.monash.edu.au (Christopher Kuperman) Subject: Re: NFS Problem Date: Wed, 14 Apr 1993 07:28:46 GMT swen@uni-paderborn.de (Swen Thuemmler) writes: : >>>>> About NFS Problem, Mike McDonald said: : : Mike> I have Linux (.99pl6) up and running on a 486. I finally got the : Mike> networking working. (Thanks Don!) Now my problem, I can nfs : Mike> mount only one of our file servers. The other two always hang : Mike> with NFS server Foo not responding. PC-NFS is able to mount all \\snip snip snip\\ \\ : foobar-a foobar-b : ^ == ^ : : foo can mount from bar and the other way round. : The same problem existed (I believe) with PC-NFS 3.0, but not with : 4.0. : : Greetings, Swen : : : -- : Swen Th|mmler * "Wenn im Namen eines Ideals Menschen Schurkereien : An der kleinen Mauer 3 * zugef|gt werden m|ssen, so ist dieses Ideal : 4933 Blomberg * einen Dreck wert." : * -- A. und B. Strugatzki Hey Look at this guys .sig // tell me something swen.... How do you pipe to more if you use the pipe symbol as a letter of the alphabet.. IE: cat readme.txt | more *shrug*.. wierd.. Ckup. _____________________________________________________________________________ [__From_________________________________][ aka: Christopher Kuperman ] [_______________________________________][ The holistic systems consultant ] [____@@@@@@____________________@@_______][------------------------------------] [_______@@____@@@@@___@@_@@@___@@__@@___][ email: zork@yoyo.cc.monash.edu.au ] [______@@____@@___@@__@@@___@__@@@@_____][------------------------------------] [_____@@_____@@___@@__@@_______@@_@@____][ Giv a man a fish & he'll eat for a ] [____@@@@@@@__@@@@@___@@_______@@__@@___][ day, teach him how to fish & he'll ] [_______________________________________][_eat for ever.______________________] ------------------------------ Crossposted-To: comp.sys.ibm.pc.hardware From: ent811l@monu6.cc.monash.edu.au (Christopher Kuperman) Subject: Re: Do I need intelligent serial I/O?? Date: Wed, 14 Apr 1993 07:53:30 GMT 1st: you wont get 14.4k times 4 through put! I never have and never will the best you can expect from V42/bis is 2 times.. or 28K for text only.. 2nd: I have used the system at 14.4k - and dont forgot a lot of file, transfers are compressed - and u will get about 11-12k using Zmodem and the CPU use on a 386/40MHZ was approx 20-30% and i cant tell ya wat the load averages were.. coz i cant remember. This was using a normal (16450 nonfifo) serial card with an extrenal modem. i dont know about internals.. they are probably better because they are direct hardware and the bytes of data go direct to the modem hardware. *shrug* ciao, Christopher. _____________________________________________________________________________ [__From_________________________________][ aka: Christopher Kuperman ] [_______________________________________][ The holistic systems consultant ] [____@@@@@@____________________@@_______][------------------------------------] [_______@@____@@@@@___@@_@@@___@@__@@___][ email: zork@yoyo.cc.monash.edu.au ] [______@@____@@___@@__@@@___@__@@@@_____][------------------------------------] [_____@@_____@@___@@__@@_______@@_@@____][ Giv a man a fish & he'll eat for a ] [____@@@@@@@__@@@@@___@@_______@@__@@___][ day, teach him how to fish & he'll ] [_______________________________________][_eat for ever.______________________] ------------------------------ From: ent811l@monu6.cc.monash.edu.au (Christopher Kuperman) Subject: Re: Desperately looking for a Linux copy Date: Wed, 14 Apr 1993 07:59:11 GMT g9321160@wampyr.cc.uow.edu.au (Janjerdsak Jirapat) writes: : : Lines: 4 : I've recently heard about Linux. It would be great if someone can : tell me how can I get a copy of this fabulous O/S. I'm living in Sydney : Australia. Is it possible to get it through the Internet? I'm desperately : looking for it. Thanks for your help! I am living in Melbourne Australia (which incidently is better than Sydney) and I can help you... We have a site in Australia which provides access to the latest Linux stuff PLEASE DOWNLOAD THE STUFF AFTER BUSINESS HOURS aUST eAST sTANDARD TIME. from Bond University (thank god for Allan Bond - he's a hero!) kirk.bu.oz.au anonymous ftp.. and its in /pub/OS/Linux (i think) get the SLS release.. its good for beginners (kirk is down at the moment .. but should be up soon) I didnt email ya.. becoz your email address is more cryptic than my password :) ciao, Christopher. _____________________________________________________________________________ [__From_________________________________][ aka: Christopher Kuperman ] [_______________________________________][ The holistic systems consultant ] [____@@@@@@____________________@@_______][------------------------------------] [_______@@____@@@@@___@@_@@@___@@__@@___][ email: zork@yoyo.cc.monash.edu.au ] [______@@____@@___@@__@@@___@__@@@@_____][------------------------------------] [_____@@_____@@___@@__@@_______@@_@@____][ Giv a man a fish & he'll eat for a ] [____@@@@@@@__@@@@@___@@_______@@__@@___][ day, teach him how to fish & he'll ] [_______________________________________][_eat for ever.______________________] ------------------------------ From: slvch@cc.usu.edu Subject: Help, Future Domain SCSI TMC-850 NOT WORK! Date: 13 Apr 93 23:38:02 MDT Hi, there, My friend installed the Linux 0.99 recently. He got the whole SLS package from tsx-11 and installed about 1 weeks ago. But he has the problem on the SCSI cd-rom. When he started the machine, Linux told him "0 host, 0 scsi, 0cd-rom" I guess the problem is the interface. The interface is Future Domain TMC-850 (8 BIT) card. I once had read about to change the seagate.c in line about 548-550 to ...=...|80;. And I rebuild the kernel, but it still not work. Any suggestion is appreciated, any idea? ------------------------------ From: lucy@lehre2.informatik.Uni-Bremen.DE (Holger Burde) Subject: urt Problem Date: Wed, 14 Apr 1993 09:55:24 GMT Hi I tried to compile urt-3.1b (utah raster toolkit) and have problems with the generation of the sub-Makefiles. There is an awk script that controlls the Makefile creation form some Files (i don't even have the man pages for awk so i don't know absolutly nothing about awk ...). Can someone please send me his/her Makefile (config/urt) or hints to solve this problem ?! Thanx in advance h.b. ==================================================================== Minister f"ur Games und Nightlife ==================================================================== ------------------------------ From: dd435157@longs.LANCE.ColoState.Edu (Dan Doner) Subject: LINUX and data acquisition Date: Wed, 14 Apr 1993 08:42:02 GMT I realize that *nix's method of interrupt handling is not good for data acquisition, however, is it possible for digital/analog data acquisition in the 1000 Hz range? Has anyone tried this? Thanks in advance, Dan ps. Its been about a year now, and I still grin like an idiot when I use my Linux box :) ------------------------------ From: ijackson@nyx.cs.du.edu (Ian Jackson) Subject: *** READ THIS BEFORE POSTING *** Date: Wed, 14 Apr 1993 06:23:00 GMT To get people to read your post and answer your question: - Read "Linux Documents Explained for Newbies" - posted weekly (see below). - Please do not post questions answered in the FAQ lists. - Please do not post "Please send me the FAQ". See below. - Please do not post "Split this group now". See below. - Please read the documentation for the program you have a problem with. - Be informative, both in subject line and body. See below. - Linux-specific posts only please - Unix, X, C, etc. groups listed below. For more information read on ... *** Linux Documents Explained for Newbies (like me 8-) *** is is posted weekly by Jay MacDonald. If you can't find it email me or Jay . It will tell you about the Linux FAQ, the INFO-SHEET, the META-FAQ, etc. These are posted regularly to comp.os.linux and c.o.l.announce, and can be found in /pub/usenet/news.answers/linux-faq at pit-manager.mit.edu and on the Linux archive sites. If you can't get them there, try mailing "mailserver@nic.funet.fi", "mail-server@pit-manager.mit.edu" or "ftpmail@decwrl.dec.com". If you can't get them any other way, e-mail me. Splitting the group (due to volume or other reasons): It's been tried, and all the new groups except .announce failed. It can't be tried again within 6 months. See news.announce.newgroups. If you insist on discussing it, do it in news.groups (if c.o.l has too much volume already, why make it worse?). With threaded newsreaders now in widespread use using a good, specific, polite subject line will get you many more readers. Most people will ignore posts with subjects like "Linux", "Help", "SLS", etc. Don't shout to get attention. Put as much information in your post as possible - preferably a brief summary followed by session transcripts, etc. as appropriate. Include the precise text of any error messages printed, etc., and appropriate info about your setup. Linux is a Unix clone running on PCs; X has its own groups. One of the following groups is is quite likely to be more appropriate for your question: Unix: comp.unix.questions, .admin, .programmer, .shell etc. X: comp.windows.x, comp.windows.x.i386unix (for XFree86 etc) GNU: gnu.emacs.help, comp.emacs, gnu.gcc.help, gnu.* generally PCs: comp.sys.ibm.pc.hardware C: comp.lang.c -- Ian Jackson ijackson@nyx.cs.du.edu PGP2 public key available on request Home: 35 Molewood Close, Cambridge, UK, CB4 3SR; +44 223 327029. These opin- Work and urgent email: iwj@cam-orl.co.uk; +44 223 343398 ions are my own. Olivetti Research Ltd, Old Addenbrookes Site, Trumpington St, Cambridge, UK; ------------------------------ From: etmelag@dcrosby27.ericsson.se (Louis Lagendijk) Subject: Re: Port Fido Mailer to Linux Date: Wed, 14 Apr 1993 07:38:19 GMT In article <1qecup$fck@cismsun.univ-lyon1.fr> cas@cismibm.univ-lyon1.fr (Laurent Cas) writes: >Louis Lagendijk (etmelag@dcrosby27.ericsson.se) wrote: > >I thought Binkley was running yet under linux ? didn't it ? >by the way, let us know when you will release your fidonet >package as many of us are very interested in it... > Yes, Binkley is up and running since october last year. What I want to release now is an updated version with a number of bug fixes. The new version will also compile on BSD systems. Ben Stuyts ported Binkley to the Next, I used his port to get something working for Linux. The new version is coordinated between us, so it will compile under (at least) the Next and Linux. >what will you use for nodelist support ? RFmail and binkley are >two mailers, aren't they ? why release them in the same package ? Currently both Rfmail and Binkley use their own nodelists. I will try to update Rfmail to use the Binkley (version 6) nodelist. Rfmail also includes a mailer (very simple) but I never used it. Rfmail is primarily intended as a fido <-> news/mail gateway so it requires news, (s)mail, a mail interface (e.g. Elm) and a newsreader. It does work quite well at the moment, even though a lot of fido people would like to see native fido compatible programs a setup with Rfmail ` works for me. > Cheers, Louis ------------------------------ From: brauer@aix520.informatik.uni-leipzig.de (Eckard Brauer) Subject: menu problem solved Date: 14 Apr 1993 09:20:05 GMT Because I got problems with menu distributed with SLS (especially with the 'd' command for setting the directory for execution), I did recompile the menu system. I solved e few other bugs too, caused by the use of NULL instead of '\0'. I'll upload the modified sources and the binaries to a few servers in the next days, as there are saturn.hrz.tu-chemnitz.de and server2.rz.uni- leipzig.de as menu-newsrc.tar.Z. Eckard Brauer brauer@aix520.informatik.uni-leipzig.de =============================================================================== ------------------------------ From: keith@ksmith.com (Keith Smith) Subject: Re: The dangers of playing with shared libraries Date: Wed, 14 Apr 93 03:20:21 GMT In article caywood@wyvern.wyvern.com (John Caywood) writes: >Steve VanDevender (stevev@miser.uoregon.edu) wrote: > > [ stuff deleted ] > >: in to fix the problem, I would like to _strongly_ suggest that >: 'ln' be statically linked or that a statically linked version of >: 'ln' be made available in the distribution so that it is possible >: to recover from this mistake the obvious way. > >I second the motion. I got bit installing gcc486 instead of the gcc386 >that came with SLS 0.99pl2. 'ln' is not the only candidate for static >linking -- I can think of ls, mv, and (maybe) cat in the same got-to-have-it >category. I can't see it if I can't 'ls'; if I can 'mv' it, I don't need Sure you can "see" it : # echo * Will happily list all the files in your cwd. Bare minimum is your shell, 'cat' and 'mv' IMHO. 'ln' would be nice, but not absolutely neccessary, You can always make a COPY temporarily, and then link back correctly later. -- Keith Smith keith@ksmith.com 5719 Archer Rd. Digital Designs BBS 1-919-423-4216 Hope Mills, NC 28348-2201 Somewhere in the Styx of North Carolina ... ------------------------------ From: brauer@aix520.informatik.uni-leipzig.de (Eckard Brauer) Subject: probs with 4.3.3-shlibs and solution Date: 14 Apr 1993 09:32:34 GMT While Installing the libs distributed with image-4.3.3.tar.z I got the problems that neither 'mount' nor 'df' worked correctly. Searching the reason I found, that /etc/mtab was empty. Therefor I changed in /etc/rc the line: /bin/rdev >/etc/mtab to: echo "`/bin/rdev` " > /etc/mtab or: echo -n "`/bin/rdev` " > /etc/mtab where should be replaced by the type of the current root-fs. Eckard Brauer brauer@aix520.informatik.uni-leipzig.de =============================================================================== ------------------------------ From: dietmar@techfak.uni-bielefeld.de (Dietmar Braun) Subject: Re: The dangers of playing with shared libraries Date: Wed, 14 Apr 1993 09:26:22 GMT In article , stevev@miser.uoregon.edu (Steve VanDevender) writes: |> In article <1qfot4$tin@hal.gnu.ai.mit.edu> |> ericy@hal.gnu.ai.mit.edu (Eric Youngdale) writes: |> |> In article <1993Apr13.161616.24954@kf8nh.wariat.org> |> bsa@kf8nh.wariat.org (Brandon S. Allbery) writes: |> |> >(2) A statically-linked shell has been suggested. Of the |> >currently available ones, I would have to suggest rc just |> >because it's compact... the concept of a statically linked |> >"bash" makes me think of a "bashdisk" containing nothing |> >*but* bash because nothing else will fit :-) |> |> This is what the bootable rootdisk is for. I have yet to see |> a good argument for why the bootable rootdisk cannot be used |> instead of specially staticly linked binaries to fix screwups |> with the sharable libraries. |> |> A good argument: If you trash the shared libraries you can't even |> run sync! If you use a bootable rootdisk you have to shut the |> system down, but you risk filesystem damage because you couldn't |> sync first. shutdown or halt won't work with the shared |> libraries gone either. |> I suggest to have a bootdisk/rootdisk handy and _sync_ as statically linked binary. I think it's better to maintain a good list of utilities / files on a root disk then to discuss some static linking. The reason is that static linking doesn't help anyway if you make typing errors and delete something valuable. Some old Unix hackers might use 'echo *' instead of ls. But I would not get such an idea. This is the reason, because I tell everybody to use a good bootdisk, where you find everything you need. I've made a recovery disk as a rootdisk for my 5.25 '' drive. It contains no kernel and I could put _everything_ on this disk I need to install linux. (tar, gzip, rdev, ln, mkdir, dd, beav ...) And there is enough place left for zcat, less and faq.z :-) |> Ideally, if you haven't panicked the kernel then you really |> shouldn't have to reboot the machine. Something simple like |> mashing the shared library links doesn't destroy the system to |> the point of unusability _except_ that absolutely nothing will |> run, including the commands you could have used to get things |> going again without a reboot. If only I had thought of using a |> while/read/echo loop in one of the shells I still had running to |> copy libc.so.4.3.3 over the top of libc.so.4 . . . |> |> Perhaps I am harping on this point too much, since I have now |> recovered from my mistake and I managed to sleep off the |> incredible fit of anger I experienced when I discovered that I |> had made it. I am simply recommending that the basic set of |> Linux utilities have a few critical statically linked members |> (like 'ln'), so that some common mistakes can be recovered from |> more quickly and conveniently. |> |> -- |> Steve VanDevender stevev@greylady.uoregon.edu |> "Bipedalism--an unrecognized disease affecting over 99% of the population. |> Symptoms include lack of traffic sense, slow rate of travel, and the |> classic, easily recognized behavior known as walking." ===== Dietmar Braun; University of Bielefeld (finger dietmar@techfac.techfak.uni-bielefeld.de) ------------------------------ From: brauer@aix520.informatik.uni-leipzig.de (Eckard Brauer) Subject: Searched: cc1obj Date: 14 Apr 1993 09:41:32 GMT Sometime before anyone announced a compiled cc1obj. I can't remember, where it was, if there's someone who knows, please mail me, if some has the cc1obj, please send me. Thanks! Eckard Brauer brauer@aix520.informatik.uni-leipzig.de =============================================================================== ------------------------------ From: KID01 Subject: Backup libraries (was Re: The dangers of playing with shared libraries) Date: 14 Apr 93 09:40:23 GMT In article stevev@miser.uoregon.edu (Steve VanDevender) writes: >In article <1qfot4$tin@hal.gnu.ai.mit.edu> [..] Couldn't future version of linux binarys be compiled to first search for /lib/libc.so.4 and if that fails try /lib/libc.so.4.3, etc, etc. This shouldn't be a massive problem to setup and only if all of /lib was remove would the user be in trouble ! (in which case the binaries could search in /oops_i'v_lost_my_shared_libs/backup_here/libc.so.4) Cheers Matthew -- Matthew Roderick -*- B2 Developments -*- mat@{dsbc,oasis}.icl.co.uk +44 [0]782 771000 x3270 -*- UNIX Centre, ICL, Stoke-on-Trent, England. Spelling & typing errors have been left for orthentisity ------------------------------ ** 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-Activists-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux) via: Internet: Linux-Activists@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 tupac-amaru.informatik.rwth-aachen.de pub/msdos/replace The current version of Linux is 0.99pl6 released on February 23, 1993 End of Linux-Activists Digest ****************************** ly linked binary. I think it's better to maintain a good list of utilities / files on a root disk then to discuss some static linking. The reason is that static linking doesn't help anyway if you make typing errors and delete something valuable. Some old Unix hackers might use 'echo *' instead of ls. But I would not get such an idea. This is the reason, because I tell everybody to use a good bootdisk, where you find evedigest833 644 36676 74430 74074 5366137064 5771 From: Digestifier To: Linux-Activists@senator-bedfellow.mit.edu Reply-To: Linux-Activists@senator-bedfellow.mit.edu Date: Wed, 14 Apr 93 09:15:28 EDT Subject: Linux-Activists Digest #833 Linux-Activists Digest #833, Volume #4 Wed, 14 Apr 93 09:15:28 EDT Contents: Re: The dangers of playing with shared libraries (A. V. Le Blanc) Mailing list for Linux Newbies (Duane Davis) Disk Quotas (was Re: New feature for the filesystems.) (Nick Hilliard) We Want Your Work (Bowyer Jeff) Re: [Q] Linux/SLS "mail" is too smart for its own good (Andreas Busse) Re: Disk Quotas (was Re: New feature for the filesystems.) (Klaus Steinberger) Re: weird vi/less screen insect (Michael O'Reilly) linux logo or mascott ("pboyle@unb.ca") installing SLS/a2/mount unknown error 26240?? HELP! (Martin Koch) Accessing Kodak Photo-CD (Dirk Job) XDM/rlogin problems (Marcus Daniels) Does _anyone_ use a WD-7000FASST with Linux? (Scott A. McMullan) Re: Objectively: 386bsd vs. Linux vs. commercial products? (Alasdair Allan) Basic info on LINUX. (Phil) {Q} ftp performance/SLS (william E Davidsen) Re: NFAQ: Do ISA cards cripple EISA performance ? (Joerg Schlaeger) [SUMMARY] How to get rid of LILO? (Steve Frampton) Re: Linux Serial Saga Or How I Got Kermit Working (James Purdon) Re: Disk Quotas (was Re: New feature for the filesystems.) (Scott W. Adkins) Windoze? What Windoze? (Nick Vargish) Re: the umask of root in SLS should be 022 ?? (Brandon S. Allbery) Re: New feature for the filesystems. What do you think ? (Brandon S. Allbery) Possible gotcha in gcc (william E Davidsen) ---------------------------------------------------------------------------- From: zlsiial@uts.mcc.ac.uk (A. V. Le Blanc) Subject: Re: The dangers of playing with shared libraries Date: 14 Apr 93 09:56:40 GMT Reply-To: LeBlanc@mcc.ac.uk In article <1993Apr13.025258.16555@Princeton.EDU> qpliu@princeton.edu writes: > >while read x; do echo "$x"; done < file > (doesn't work correctly with bash, but does with pdksh) >-- >qpliu@princeton.edu Standard opinion: Opinions are delta-correlated. I tried this with bash and had no problems. I am using bash-1.12 compiled with libc-4.3.3 and gcc-2.3.3. -- Owen LeBlanc@mcc.ac.uk ------------------------------ From: system@micromed.com (Duane Davis) Subject: Mailing list for Linux Newbies Date: Wed, 14 Apr 93 02:24:39 PDT Due to an overwelmingly positive response I have setup a mailing list for new Linux users. To subscribe to the list mail a message to 'lists@micromed.com' with 'SUBSCRIBE linux' as the first line in the body of the message. You will recieve a reply with instructions on how to use the list. --- system@micromed.com (Duane Davis) voice (408)293-3299 Micro-Medic public access usenet (408)279-5240 711 Lincoln Avenue public access pcboard (408)280-1610 San Jose, CA 95126 FileServer - FTPMail@micromed.com ------------------------------ From: nick@symphony.mp.ucc.ie (Nick Hilliard) Subject: Disk Quotas (was Re: New feature for the filesystems.) Date: Wed, 14 Apr 1993 10:29:08 GMT Hi, With all this talk about new features for the various filesystems, I figure no-one's really mentioned anything about Disk Quotas. Ok - it's fine if you're a just using Linux as a single user machine, but if you have a whole load of users hanging out of it, you really _do_ need some form of mechanism for making sure that people don't hog the machine. Is there a chance of seeing code for them being written in the (near) future? Nick -- | Nick Hilliard | e-mail: nick@symphony.mp.ucc.ie | | Systems Administrator | Internet: nick@curia.ucc.ie | | Mathematical Physics Dept. | Bitnet: armt6003@iruccvax | | UCC, Cork, Ireland. | Disclaimer: Opinions - they're mine, all mine! | ------------------------------ From: jbowyer@cis.vutbr.cs (Bowyer Jeff) Subject: We Want Your Work Date: Wed, 14 Apr 1993 10:44:25 GMT Reply-To: jbowyer@cis.vutbr.cs We want you to announce your work on our mailing list! Do you use a program that has a non-English interface? Have you converted any software to support more than one language for its interface? Will you sponsor a conference that might concern software with a non-English interface? Please tell us! INSOFT-L on LISTSERV@CIS.VUTBR.CS Internationalization of Software Discussion List Internationalization of software relates to two subjects: 1. Software that is written so a user can easily change the language of the interface; 2. Versions of software, such as Czech WordPerfect, whose interface language differs from the original product. Topics discussed on this list will include: -- Techniques for developing new software -- Techniques for converting existing software -- Internationalization tools -- Announcements of internationalized public domain software -- Announcements of foreign-language versions of commercial software -- Calls for papers -- Conference announcements -- References to documentation related to the internationalization of software This list is moderated. To subscribe to this list, send an electronic mail message to LISTSERV@CIS.VUTBR.CS with the body containing the command: SUB INSOFT-L Yourfirstname Yourlastname Owner: Center for Computing and Information Services Technical University of Brno Udolni 19, 602 00 BRNO Czech Republic INSOFT-L-REQUEST@CIS.VUTBR.CS ------------------------------ From: andy@resi.waldorf-gmbh.de (Andreas Busse) Subject: Re: [Q] Linux/SLS "mail" is too smart for its own good Date: 14 Apr 93 10:19:10 GMT In article <1993Mar28.004916.19259@jmd386.lonestar.org>, jdoss@jmd386.lonestar.org (Joe M. Doss) writes: |> In article <1993Mar24.190859.16371@eecs.nwu.edu> hpa@nwu.edu (H. Peter Anvin) writes: |> >I have a question about the "mail" program for Linux, specifically |> >from the SLS release, which I believe is the same as in mailpak. The |> >problem is that it bypasses the mail transport agent (smail) when it |> >gets an address it "recognizes" (local or UUCP-style, I believe). |> >Unfortunately that is a Bad Thing[TM] at least on my system. "elm" |> >behaves properly but some users on this machine are used to the BSD |> >Mail interface and don't want to switch. |> > |> >So: is there either a way to force "mail" to send ALL its mail to the |> >mail transport agent, or is there a different version of mail |> >somewhere that does that? (If there is but is not for Linux I'll try |> >to port it -- that's no problem) I had the same problem. My solution was to install the sendmail+ida package. Now I have simple /bin/mail and /bin/rmail as transport agents, elm as a frontend and sendmail as the "router". There were only few things to change in the sources. I don't remeber exactly, but the most important thing was to change the header format of elm. The "from" line must be the first one and the "subject" line the last. Andy ------------------------------ From: k2@promethe.Physik.Uni-Muenchen.DE (Klaus Steinberger) Subject: Re: Disk Quotas (was Re: New feature for the filesystems.) Date: Wed, 14 Apr 1993 19:24:39 GMT > With all this talk about new features for the various filesystems, I figure > no-one's really mentioned anything about Disk Quotas. I would be interested too in this feature. Is somebody out there working on a BSD or a log file system (like the one implemented at berkeley) ? These filesystems are commonly implemented together with quotas. Sincerely, Klaus Steinberger -- Klaus Steinberger Beschleunigerlabor der TU und LMU Muenchen Phone: (+49 89)3209 4287 Hochschulgelaende FAX: (+49 89)3209 4280 D-8046 Garching, Germany Internet: Klaus.Steinberger@Physik.Uni-Muenchen.DE ------------------------------ From: oreillym@tartarus.uwa.edu.au (Michael O'Reilly) Subject: Re: weird vi/less screen insect Date: 14 Apr 1993 11:06:03 GMT David Lesher (wb8foz@skybridge.SCL.CWRU.Edu) wrote: : Linux softland 0.99.pl4-62 01/23/93 i386, Elvis 1.6, less 177 : [ deleted ]. : In any case, it looks as if I end up back in less, as I should. But I : hit a command, and it ACTS as if I'm still in vi! Parts of the screen : overwrite with "vi"-isms, such as line numbers. ":" drops me down to the : proper place. Then, ^L's seem to toggle through various screens - less, : vi, and maybe the shell itself. Looking at 'ps' from another tty shows : less & vi still running. This is not specific to less/vi. Try useing elm, and starting emacs, and then typeing control-G. emacs doesn't exit by elm wakes up! chaos! As far as I am aware, this is actually a problem with bash. Bash is a little silly about intercepting signals in the background, and so it exits, so less/elm think that vi/emacs has exited, and so it starts reading the keyboard. Not wonderful. As yet, I haven't managed to find a patch for bash that will fix the problem. anyone??? Michael ------------------------------ From: "pboyle@unb.ca" Subject: linux logo or mascott Date: Wed, 14 Apr 1993 12:34:57 GMT My 0.02 cents to this thread. I think a lobster would be a good mascott for linux. I like the alliteration "linux lobster" (in English at least). Just as eating lobster is a special treat for humans, linux is an especially good os to feed your computer! Paul ------------------------------ From: myth@diku.dk (Martin Koch) Subject: installing SLS/a2/mount unknown error 26240?? HELP! Date: Wed, 14 Apr 1993 11:13:54 GMT When I tried to install linux, I did the following: (my system is a 486/33mhz/212 seagate hd/VESA local bus VGA/8MB ram) o downloaded series a,b,t,x o 'rawrote' a1 and a2 o inserted a1 into my 3 1/2 drive, and turned on the computer some text appeared (Loading .....) and a prompt to press or . I pressed o I'm directed to insert the utilities disk (a2, I suppose) into the drive I wish to use o the drive is "2" on the list, drive a: 3 1/2", so I select it o I press return, and get the following error: MINIX_fs magic match failed mount: unknown error 26240 o some text appears, directing me to "more README" but more doesn't seem to exist. I do get a prompt, though, with a working "help" command, and the commands described by help. What do I do from here??? I hope this isn't covered by the FAQ, because I can't find this probelm there ;) Thanks (in advance) for your help. Martin myth@diku.dk ------------------------------ From: job@euklid.informatik.uni-dortmund.de (Dirk Job) Subject: Accessing Kodak Photo-CD Date: 13 Apr 1993 14:37:48 GMT Hi folks, I tried to access a Kodak Photo-CD with my Toshiba XM3401 SCSI CD-ROM. But there's no chance to read the files because Linux can't mount the CD-ROM. The Filesystem is unknown to it. Reading "normal" CD-ROMS doen't make any problems at all. Does anybody know about this problem? Who can help? Dirk *-------------------------------------------------------------------* | Dirk Job | job@ls7.informatik.uni-dortmund.de | | | ___ | | University of Dortmund | //// | | Department of | UNI DO// | | Computer Science VII | ___ //// | | (Computer Graphics) | \*\\/// | | P.O. Box 500 500 | \\\\/ | | D 4600 Dortmund 50 | | | Germany | Phone: +49 231 7556122 / -6131 | *-------------------------------------------------------------------* ------------------------------ From: marcus@ursula.ee.pdx.edu (Marcus Daniels) Subject: XDM/rlogin problems Date: 14 Apr 93 12:08:06 GMT Has any experience rlogin losing its I/O when running xdm? rlogin works fine with X running and on a normal console. With xdm, I get a `closed connection' and return to prompt. marcus ------------------------------ From: mcmullan@sd-sun1.vf.ge.com (Scott A. McMullan) Subject: Does _anyone_ use a WD-7000FASST with Linux? Date: 14 Apr 1993 07:12:13 -0500 A while ago, my brother posted for me about a problem I'm having getting my WD-7000FASST SCSI controller to work under Linux. At this point, the only replies I've seen in e-mail were from other people who have the same sort of problems, asking to get copies of any help I get. I don't know about any of the other people, but I bought this card specifically because it was listed as a compatable controller in the hardware FAQ. I'm very bummed; here I sit with a nice system full of supposedly compatible hardware just BEGGING to have Linux installed, and I can't get past booting A1. Does ANYONE have a WD-700FASST controller functioning PROPERLY under Linux? If so, what is the ROM release on the card? What kind of hard disk are you using? What SCSI ID is assigned to the disk? I had all sorts of trouble with the controller when I had the disk as SCSI ID 0, but when I moved it to ID1 it works fine under DOS; maybe the SCSI ID has something to do with it. I'm dying to bring up Linux, but until one of my SCSI controllers is properly handled by the OS, I'm SOL (My other SCSI is an IN-2000; it's looking like that may be supported soon, which would be GREAT). Please, if you're using a WD-7000FASST card with Linux, I need to hear from YOU. Pet Peeve #2: Linux thinks my digiboard is on IRQ4, when it's configured at IRQ3. Of course since I can't get Linux installed, I can't really address this problem just yet... -- -- Scott A. McMullan GE AOD, Valley Forge PA (215) 354-5160 ( Or is that Martin Marietta? "GE WAS ME!") mcmullan@sd-sun1.vf.ge.com ! I speak for myself ONLY/no one speaks for me! mcmullan@scottshouse.hfvd.pa.us [ Try my BBS at (215)789-5748/1298 AT&T Dataport modems, up to 14.4K/V.42bis ] ------------------------------ Crossposted-To: comp.os.386bsd.questions From: aa1@st-andrews.ac.uk (Alasdair Allan) Subject: Re: Objectively: 386bsd vs. Linux vs. commercial products? Date: Wed, 14 Apr 1993 11:35:28 GMT Brandon Vanevery writes: >I caught the tail end of a flame war earlier, I really hope this doesn't >start another. I'm looking for any opinions - hopefully objective - on >the strengths and weaknesses of different 386 UNIX platforms. Some key >concerns: > > - stability > - efficiency/speed > - TCP/IP, SLIP and/or PPP support > - support for 24-bit true color hardware of any sort, > but preferrably on the cheap. > [rest of article deleted] I'm considering moving to a Unix based home system. My main need in SLIP or PPP support so I'd be interested in hearing whats out there for BSD or Linux in that area (ie which has the better support?) So if Brandon could post a summary of the Email he gets on this I'd appreciate it (I'm sure there are a couple of other people out there that are at least *mildy* interested as well). Alasdair Allan Dept of Physics & Astronomy University of St. Andrews Email: aa1@st-andrews.ac.uk =====BEGIN PGP PUBLIC KEY BLOCK===== Version: 2.2 mQCNAivJ7W8AAAEEANZVrJOaRZi0TZI7oBvOUfSHh+q4LE+FZbTtOtwZ4OSoQFEr PQG/w+GeMj8I2gDSXki9x3S6jrWv7RoxnOA+XfhIPgX5a7QVXFjwxl1q11UjTQHc 5ZgNEkNuQ6AcGuwUlZTZ1xmbjQmZKHvdqLqlk9Lr+H4zFsnqrSRKv6twmINHAAUR tCVBbGFzZGFpciBBbGxhbiA8YWExQHN0LWFuZHJld3MuYWMudWs+ =DfhY =====END PGP PUBLIC KEY BLOCK===== ------------------------------ From: phil@pixie.login.qc.ca (Phil) Subject: Basic info on LINUX. Date: Tue, 13 Apr 93 23:29:37 PDT Can anybody out there answer a few elementary questions on LINUX? I am familiar with the general concept of the system but I have quite a few simple questions. Here goes: - What are the minimun hardware requirements for decent operation? ( I have a 386DX40, 4megs and 120megs ) - Recommended hard disk space? - Can Linux run DOS apps? To what extent? - Is cohabitation of DOS on the same HD simple or a headache? - What level of technical knowledge is required to install it? - Where can I find information text files via ftp? (I'm looking for basic info, not a whole manual) I would greatly appreciate any type of answers, I want to escape the Gates of hell! Please respond via email if possible at phil@pixie.login.qc.ca Philippe. Philippe Bechamp phil@pixie.login.qc.ca ------------------------------ From: davidsen@ariel.crd.GE.COM (william E Davidsen) Subject: {Q} ftp performance/SLS Reply-To: davidsen@crd.ge.com (bill davidsen) Date: Wed, 14 Apr 1993 12:45:30 GMT I have the Apr 5 SLS installed and have finally hacked the tcp stuff a bit and gotten it going on my system. The problem I'm seeing is that: 1) telnet works fine 2) ping works fine 3) NFS works fine 4) ftp works only for small file (<100k) and is slow (<30k cps) Is there some parameter I have to set to get this working, or is this a known limitation of the current tcp, or what? Any clues would be useful, I have more time in this already than I can afford. Note, all my changes were in the area of card detection, in wd.c. I did a bit of work to insure that the probe didn't run off singing "I found it" when it hadn't, and didn't use the transfer address in EEROM. The logic was changed not at all, I just have some stuff in the system which could look like a net card. -- bill davidsen, GE Corp. R&D Center; Box 8; Schenectady NY 12345 ------------------------------ From: jschief@finbol.toppoint.de (Joerg Schlaeger) Subject: Re: NFAQ: Do ISA cards cripple EISA performance ? Date: Tue, 13 Apr 93 17:31:23 GMT jschief@finbol.toppoint.de writes in article <734365786snx@finbol.toppoint.de>: > > alecm@coyote.uk.sun.com writes in article <1phdksINNbcu@uk-news.uk.sun.com>: > > > > I've heard from a friend that if you have an EISA machine, and install > > an ISA card of any sort, the bus performance will drop to that of an > > ordinary ISA bus. > yes, it's true and depends on the chipset and layout of your motherboard. > Some motherboards may be unable to make a differ the timing of the slots. I wanted to say: Some motherboards may be unable to make a different timing for the EISA-Slots with ISA-Cards in. They detect the ISA-CARD and do ISA-Cycles for all or a group of EISA-Slots. (this meens you get the timing like org. 8MHz AT ISA-Bus) > Look in the board manual or the EISA-Setup Software to get more infomation !! Sorry for mistake Joerg ------------------------------ From: frampton@vicuna.ocunix.on.ca (Steve Frampton) Subject: [SUMMARY] How to get rid of LILO? Date: 14 Apr 93 10:38:27 GMT Reply-To: frampton@vicuna.ocunix.on.ca Hello! First of all, I wish to express my gratitude to all who responded to my queries. My mailbox has been literally overflowing with answers. The response was so overwhelming I must apologize for not being able to personally thank each of you via private e-mail. My question was simple: After removing Linux from my hard drive, and formatting my drive, I was quite surprised to see LILO still prompting me to enter the boot partition. I wanted to know how I could get rid of that boot message and use the standard DOS one instead. Using 'FDISK /MBR', the master boot record on the hard drive will be replaced with one specific to DOS. No data is lost whatsoever. Surprisingly, a number of you did not believe 'Enter boot partition (1-4)' was a LILO prompt. I am using TAMU Linux .99p4 and perhaps LILO has been patched with this revision to make it easier on the user, but yes, this is definitely a LILO (or at least a Linux) message. Thanks again! +--------+< Unabashed Shabba Ranks Fan! >+------+--------------------+ | Steve Frampton - frampton@vicuna.ocunix.on.ca | Steve Frampton | | I collect postcards! If you send me one from | 501-A Princess St. | | your area, I'll send one in return. Send to >| Kingston, Ontario | | (Don't forget both email and mailing address) | CANADA K7L 1C3 | +-----------------------------------------------+--------------------+ ------------------------------ From: purdon@cons1.mit.edu (James Purdon) Subject: Re: Linux Serial Saga Or How I Got Kermit Working Date: 14 Apr 1993 13:03:52 GMT In article <1qfuuoINNidp@senator-bedfellow.MIT.EDU> purdon@cons1.mit.edu (James Purdon) writes: >Finally, I start mucking about with "stty -a a "-hupcl" in the report. I decide to do a "stty hupcl >By the way, at least in this version, for me, the "stty hupcl >initially. After I've grabbed it though, a "stty -hupcl >With all this talk about new features for the various filesystems, I figure >no-one's really mentioned anything about Disk Quotas. > >Ok - it's fine if you're a just using Linux as a single user machine, but if >you have a whole load of users hanging out of it, you really _do_ need some >form of mechanism for making sure that people don't hog the machine. Well, I have thought about these disk quotas for sometime now. One of these days, I will actually sit down an implement them. Here is what I think is involved: 1) The filesystems must all be modified to handle quotas... Essentially, at least the "writing of files" mechanism must be modified to check for reached quotas before going through with the right. Also, many popular quota systems also start setting file sizes to 0 when any file is a) opened b) over quota c) used up time period or warnings. So, the "opening of files" mechanism may also need to be modified. 2) There must be a way for quotas to be maintained. I think BSD's approach is to include a quota field in the password structure, or something like that (I am not sure, though). Most systems keep track of a file called "quotas" which would determine the quotas of every user on the system. 3) Some library calls may need to be modified to handle quotas. At the minimum, every library call that has to do with writing of files, must at least check for quota violations. I figure that a new errno would be created, something like ERRQUOTA or something like that. (There is probably a standard for it.) I am sure there are others, but this is good enough for now. I don't think a quota daemon (i.e. quotad? :-) ) would be good enough, since it is very hard to detect when a file is going to exceed a quota and stop that write *before* it happens. 2 concepts of most quota systems: 1) Soft Quota -> If you exceed this quota, then the writes will succeed, but you will be given a time period (i.e. 1 week) or a number of login warnings (i.e. 3 logins) before the quota system will take action. Taking action usually means that any file that is written to, and sometimes, any file that is read, will have its filesize effectively changed to 0. This would continue until the user removes enough files to go under the quota, or the quota system changes enough files to reduce the quota for the user... 2) Hard Quota -> This one is set higher than the soft quota, and is the absolute highest amount of disk space the user can use at any time. If the hard quota is reached, whatever write was going on at the time is halted and the file closed. So, in most cases, the user will be left with a file that is incomplete, but uses up the rest of the hard quota. I don't think quotas will be that hard to implement. I am concerned, however, how to handle any modifications to the libraries, such as libc and any of the gnu libraries. HLU? Can you give us any ideas on this? Is it necessary? Scott -- Scott W. Adkins Internet: sadkins@bigbird.cs.ohiou.edu ~~~~~~~~~~~~~~~ ak323@cleveland.freenet.edu Ohio University of Athens Bitnet: adkins@ouaccvma.bitnet ------------------------------ From: vargish@sura.net (Nick Vargish) Subject: Windoze? What Windoze? Date: 14 Apr 1993 13:02:51 GMT Last night I decided my "workstation" (386dx-33, Linux .99pl7A) was ready for real work... root@imrryr# rm -rf /DOS/C/win I slept like a baby :^). Cheers, all. Nick p.s. As I mentioned above, I'm running .99pl7A, SoundBlaster support* and mount patches in kernel, 8 megs RAM, two 2efs and one swap partition (across two hard drives), X on a Trident 8900C, and okay output from my Panasonic KX-P1123 (PostScript!). No real problems so far... Many, many thanks to everyone who made this "revolution" possible. * Can anyone point me in the direction of something that will play .wav or .voc files? -- | Nick Vargish | | SURAnet Operations | Who needs a fancy .signature when I have all this? | vargish@sura.net | ------------------------------ From: bsa@kf8nh.wariat.org (Brandon S. Allbery) Subject: Re: the umask of root in SLS should be 022 ?? Date: Wed, 14 Apr 1993 12:20:17 GMT In article twpierce@unix.amherst.edu (Tim Pierce) writes: >In article <1993Apr13.213801.25918@kf8nh.wariat.org> bsa@kf8nh.wariat.org (Brandon S. Allbery) writes: >>In article <1993Apr13.141146.10590@ccds3.ntu.edu.tw> dhliu@solar.csie.ntu.edu.tw (Der-Hua Liu) writes: >>>I cannot understand why root's umask (profile) in SLS is set to 066. >> >>umask... I'm *still* finding things that should be accessible by anyone but >>are in fact only readable by root :-( > >I did a little looking around after installing SLS and decided that >there was nothing, short perhaps of /proc (and perhaps not even that) /etc/shadow >find / -type d -print will run through the directory tree for you. ...I've been doing *ix for 11 years, I'm quite aware of these things... but I'm also a bit security-conscious as a result (of course, with *ix in general tracking security holes is a bit like being a one-armed paper hanger :-) ++Brandon -- Brandon S. Allbery bsa@kf8nh.wariat.org It's not too late to turn back from the "Gates" of Hell... Linux: the FREE 32-bit operating system, available NOW. Why waaaaaait for NT? ------------------------------ From: bsa@kf8nh.wariat.org (Brandon S. Allbery) Subject: Re: New feature for the filesystems. What do you think ? Date: Wed, 14 Apr 1993 12:23:56 GMT In article <1993Apr13.200801.24748@serval.net.wsu.edu> hlu@eecs.wsu.edu (HJ Lu) writes: >In article , rajat@watson.ibm.com (Rajat Datta) writes: >|> Why not just build a compressed filesystem? That is, compress all the >|> blocks, metadata and all. With MSDOS 6.0 coming with builtin > >I am wondering how a compressed filesystem handles demand paging and >random update. The "proper" way to do a compressed filesystem is in fact not a compressed *filesystem* at all, it's a compressed *disk*. Which means that individual blocks are compressed regardless of contents. Since *ix does updates on entire blocks in all (current) cases, this imposes only a decompression speed penalty on demand paging. Of course, it *does* make it rather difficult to put an FFS on a compressed drive, since you need to sandwich an uncompress/recompress operation around fragment updates.... ++Brandon -- Brandon S. Allbery bsa@kf8nh.wariat.org It's not too late to turn back from the "Gates" of Hell... Linux: the FREE 32-bit operating system, available NOW. Why waaaaaait for NT? ------------------------------ From: davidsen@ariel.crd.GE.COM (william E Davidsen) Subject: Possible gotcha in gcc Reply-To: davidsen@crd.ge.com (bill davidsen) Date: Wed, 14 Apr 1993 13:01:10 GMT I noted an interesting behavior in gcc last night, it doesn't ignore stuff which is #ifdef'd out! I got a program to port to Linux, and I had some text included which I found valuable. Having learned long ago that putting /*...*/ around arbitrary stuff might be a problem due to */ in the text, I put #if 0 ... #endif around it instead. A gcc generated error messages about the text it should have been ignoring! Obviously the preprocessor has to scan for #commands, but it shouldn't be complaining about other things in the text, like over-long strings, etc. This *may* be an enhancement of the Linux version, since the V.4/386 and Sun versions of gcc don't complain about the problem. Therefore I want to check it carefully before I report it to the developers as a generic bug in gcc. Just a warning to jiggle your memory if you ever have something similar happen. -- bill davidsen, GE Corp. R&D Center; Box 8; Schenectady NY 12345 ------------------------------ ** 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-Activists-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux) via: Internet: Linux-Activists@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 tupac-amaru.informatik.rwth-aachen.de pub/msdos/replace The current version of Linux is 0.99pl6 released on February 23, 1993 End of Linux-Activists Digest ****************************** 6.10590@ccds3.ntu.edu.tw> dhliu@solar.csie.ntu.edu.tw (Der-Hua Liu) writes: >>>I cannot understand why root's umask (profile) in SLS is set to 066. >> >>umask... I'm *still* finding things that should be accessible by anyone but >>are in fact only readable by root :-( > >I did a little looking around after installing SLS and decided that >there was nothing, short perhaps of /proc (and perhaps not even that) /etc/shadow >find / -type d -print willdigest834 644 36676 74430 53546 5366137065 5774 From: Digestifier To: Linux-Activists@senator-bedfellow.mit.edu Reply-To: Linux-Activists@senator-bedfellow.mit.edu Date: Wed, 14 Apr 93 10:45:25 EDT Subject: Linux-Activists Digest #834 Linux-Activists Digest #834, Volume #4 Wed, 14 Apr 93 10:45:25 EDT Contents: Networking between two linux machine without Ethernet Card (jonathan chen) Re: Kernel crash: Need help fixing my Ext2fs (Thomas Malik) Re: The dangers of playing with shared libraries (william E Davidsen) Re: The dangers of playing with shared libraries (william E Davidsen) Source of docs? (william E Davidsen) HELP: Joining SLIP mailing list? (Dennis Robinson) Re: linux logo or mascott (KID01) [Q] Has XS3 merged with Xfree? (Joe Broniszewski) Re: Do I need intelligent serial I/O?? (william E Davidsen) installing SLS/a2/mount: unknown error 26240?? HELP! (Martin Koch) Re: Kernel crash: Need help fixing my Ext2fs (william E Davidsen) Re: Xfree 1.2 leaves screen in graph-mode after exit, runx !work (Wolfgang Jung) Re: [Q] Has XS3 merged with Xfree? (Rob Rainton) mice problems with Xwindows (Shane Kelvin Richards) Re: New feature for the filesystems. What do you think ? (Jeremy Fitzhardinge) IOCTL on Serial Lines (Wolfgang Jung) ---------------------------------------------------------------------------- From: jchen@cs.buffalo.edu (jonathan chen) Subject: Networking between two linux machine without Ethernet Card Date: 14 Apr 93 12:59:51 GMT I now have two machine: one 386-40 120mb HD IDE controller, the other 486-33 210mb HD Future Domain 885 SISC controller. I don't have money for the two ethernet card. But I still want to use the hard disk in the 486. Can this be done? How? Did anyone try something like this? Please help because I am REALLY running out of space in my 386. -Jonathan =================================================================== jchen@cs.buffalo.edu ------------------------------ From: malik@dfki.uni-kl.de (Thomas Malik) Subject: Re: Kernel crash: Need help fixing my Ext2fs Date: 14 Apr 1993 12:27:40 GMT Reply-To: malik@dfki.uni-kl.de In article n9o@agate.berkeley.edu, umminger@purina.berkeley.edu (Frederick W. Umminger) writes: > During "make install" for taylor-uucp-gamma1.04 >I had my first kernel crash. The message was > >Unable to handle kernel paging request at address C0840000 >Oops: 0000 >EIP: 0008 : D0034090D >EFLAGS: 0001010202 >fs: 0017 >base: 00000000, limit: C0000000 >Pid: 3122, process nr:11 ... stuff deleted >reproduce it for debugging purposes after I get my filesystem >fixed. > > Thanks, > Frederick Umminger > umminger@math.berkeley.edu I made the same, miserable experience. I tried to repartition the hd. As i realized later, i made a fault by setting the unit size in fdisk to and created partitions, that didn't start and end on cylinder boundaries. I never recognized this error, because (older) fdisk gave no warnings and i had ext fs running stable for 4 months. When i now changed to ext2 fs, mke2fs didn't correctly allocate my 32 MB root partition. While there had been ~8MB left on that fs (output from df and e2fsck), write accesses had 'no space left on device' as result. Beside that, i did get these 'bad directory entry errors' from time to time When i tried to mount that fs from the rootdisk (from tsx-11:/pub/linux/GCC/rootdisk) i got a similar error as you described. When i ran fdisk from that rootdisk, i got many warnings about physical and logical sectors not beeing the same. So i had to delete all the old partitions and started a new installation. I'm not done now, but i didn't have any problems creating my ext2 filesystems. =============================================================================== | _/_/_/_/_/_/_/_/_/_/_/ _/ _/_/ | _/ _/_/ _/ _/ | Thomas Malik _/ _/ _/ _/ _/ | malik@dfki.uni-kl.de _/ _/ _/ _/ | _/ _/ _/ | Richard Wagner Str. 38 _/ _/ _/ | 6750 Kaiserlautern _/ _/ _/ | _/ _/ _/ _/ _/ | ------------------------------ From: davidsen@ariel.crd.GE.COM (william E Davidsen) Subject: Re: The dangers of playing with shared libraries Reply-To: davidsen@crd.ge.com (bill davidsen) Date: Wed, 14 Apr 1993 13:15:48 GMT In article , stevev@miser.uoregon.edu (Steve VanDevender) writes: | In article dmw@teal.csn.org (Dave Warner) writes: | | caywood@wyvern.wyvern.com (John Caywood) writes: | | >Any other opinions on a minimal set of statically-linked utilities? | | Yes. Build them yourself. Not intended to be a flame | but all these sources are freely available, easily | compilable, easily statically linkable, easily installable | any way you please. | | Its YOUR system to do with as you see fit -- try it, you'll like it. | | I'm certainly going to find the GNU file utilities and make | statically linked versions of things like cat and ln. | | My recommendation was that binary distributions of Linux, such as | the SLS, ought to have a minimal set of statically linked | utilities so that a problem with the shared libraries won't make | the system completely unusable or unfixable. It is very | frustrating to find that one weak link and then discover that you | can't fix it without something like a bootable rootdisk because | the tool you needed to fix it broke along with everything else. Given that there are bootable rootdisks, and the SLS bootable plus a rootdisk, I'm not sure that building a suite of oversize utilities makes any sense when you can run from the floppy and do what you need. Sounds like a non-problem to me, I've occasionally whacked the Linux partition so hard it wouldn't boot, and I've always been able to fix it from the A1/A2 SLS disks. -- bill davidsen, GE Corp. R&D Center; Box 8; Schenectady NY 12345 ------------------------------ From: davidsen@ariel.crd.GE.COM (william E Davidsen) Subject: Re: The dangers of playing with shared libraries Reply-To: davidsen@crd.ge.com (bill davidsen) Date: Wed, 14 Apr 1993 13:19:16 GMT In article <1993Apr14.032021.12691@ksmith.com>, keith@ksmith.com (Keith Smith) writes: | Bare minimum is your shell, 'cat' and 'mv' IMHO. 'ln' would be nice, | but not absolutely neccessary, You can always make a COPY temporarily, | and then link back correctly later. Spoken like someone who has a big disk. In practice there are lots of people running Linux who are too tight on disk to make a copy of something as large as a library, so ln is really needed. -- bill davidsen, GE Corp. R&D Center; Box 8; Schenectady NY 12345 ------------------------------ From: davidsen@ariel.crd.GE.COM (william E Davidsen) Subject: Source of docs? Reply-To: davidsen@crd.ge.com (bill davidsen) Date: Wed, 14 Apr 1993 13:24:43 GMT I have a request for info on making a boot/root disk combo and rather than write some instructions I would like to find a set. I know I had some info on it since I've done it, but I don't remember ever seeing a clear description of the process, I think I grabbed code from the TAMU 0.96c release and reverse engineered it. If someone can point me to the right place I'd appreciate it. If not, I'll write something for the person with the question and then post it for anyone to archive as needed. -- bill davidsen, GE Corp. R&D Center; Box 8; Schenectady NY 12345 ------------------------------ From: djr48312@uxa.cso.uiuc.edu (Dennis Robinson) Subject: HELP: Joining SLIP mailing list? Date: Wed, 14 Apr 1993 13:21:33 GMT I was interested in helping the debugging effort of the ALPHA SLIP package. What is said is that I must join the mailing list until I can even see code. So I would like to join the mailing list. ------------------------------ From: KID01 Subject: Re: linux logo or mascott Date: 14 Apr 93 13:30:57 GMT In article <14APR93.09269155.0062@UNBVM1.CSD.UNB.CA> AA24@UNB.CA (pboyle@unb.ca) writes: >My 0.02 cents to this thread. I think a lobster would be a good >mascott for linux. I like the alliteration "linux lobster" (in >English at least). Just as eating lobster is a special treat for >humans, linux is an especially good os to feed your computer! > And a quote :- Keep your hardware happy, Linux. Matthew -- Matthew Roderick -*- B2 Developments -*- mat@{dsbc,oasis}.icl.co.uk +44 [0]782 771000 x3270 -*- UNIX Centre, ICL, Stoke-on-Trent, England. Spelling & typing errors have been left for orthentisity ------------------------------ From: joe@stat.psu.edu (Joe Broniszewski) Subject: [Q] Has XS3 merged with Xfree? Date: Wed, 14 Apr 1993 13:38:21 GMT I am trying to get X working with an Orchid Faherhiet Local bus card. I have XS3 working, but there are a few bugs with it. One of the README files says that XS3 will be merged with Xfree 1.2, but a friend of mine told me that he has can't get Xfree to work with his Orchid Faherhiet card. Any experience/help? -- Joe Broniszewski PSU Statistics Department Systems Administrator ------------------------------ Crossposted-To: comp.sys.ibm.pc.hardware From: davidsen@ariel.crd.GE.COM (william E Davidsen) Subject: Re: Do I need intelligent serial I/O?? Reply-To: davidsen@crd.ge.com (bill davidsen) Date: Wed, 14 Apr 1993 13:50:37 GMT In article <1993Apr13.203450.7748@dcs.warwick.ac.uk>, robin@dcs.warwick.ac.uk (Rob Rainton) writes: | I'm thinking of getting a powerful PC and running Linux and X windows | on it. I also hope to get a fast (V32bis with V42bis compression) | modem to connect to an internet service in the UK. Sounds good so far. | How much CPU time will be consumed by blasting the claimed max data | rate of 57,600 baud (14,400 x 4 times compression - I know this won't | happen most of the time, but just assume best case) down a standard PC | serial port? I would get a card with a smart UART for starters, the 16550 part will cut the interrupt rate by 5-8x, which should let you keep up with the incoming. As far as burst rate goes, one of the users of my BBS uses a 2400 modem with V.42bis and serial line locked at 9600, and gets 670cps on text transfers (avg for a 50k file). That's 2.8x compression average, so I would believe that the burst would hit the 4x claimed at least once in a while. Transferring compressed files will be essentially no compression, of course. | | Will buying an intellegent serial I/O card with it's own processor | help in this particular application? (I don't want my machine to be | tied up by the serial I/O when ftping large sources) I have my BBS on an old 386DX-16, about as slow as you can go, and with two lines coming in at 19200 I see about 6% CPU use. WARNING: I am not running Linux on that machine, I'm running commercial UNIX with the FAS device driver. However, I would not expect more than that using Linux and a reasonable 486, since the serial driver uses the 16550 if present. Maybe someone can give you an exact figure, but I wouldn't expect any problem with a single line. | | Also, is there any advantage over buying a modem card rather than use | an external modem via a serial port? No, buy the serial port and modem. Each can be used for other things, you can use the modem with your next computer (might not be a PC) or upgrade the modem without changing the box. I hear that ISDN is big in Europe, you might be able to get one of those beautiful ISDN modems for less than the pice of a car someday (64k bidirectional). -- bill davidsen, GE Corp. R&D Center; Box 8; Schenectady NY 12345 ------------------------------ From: myth@diku.dk (Martin Koch) Subject: installing SLS/a2/mount: unknown error 26240?? HELP! Date: 14 Apr 93 13:08:04 GMT Hmm. This didn't appear the first time, so I'll give it another shot.. ====================================================================== When I tried to install linux, I did the following: (my system is a 486/33mhz/212 seagate hd/VESA local bus VGA/8MB ram) o downloaded series a,b,t,x o 'rawrote' a1 and a2 o inserted a1 into my 3 1/2 drive, and turned on the computer some text appeared (Loading .....) and a prompt to press or . I pressed o I'm directed to insert the utilities disk (a2, I suppose) into the drive I wish to use o the drive is "2" on the list, drive a: 3 1/2", so I select it o I press return, and get the following error: MINIX_fs magic match failed mount: unknown error 26240 o some text appears, directing me to "more README" but more doesn't seem to exist. I do get a prompt, though, with a working "help" command, and the commands described by help. What do I do from here??? I hope this isn't covered by the FAQ, because I can't find this probelm there ;) Thanks (in advance) for your help. Martin myth@diku.dk ------------------------------ From: davidsen@ariel.crd.GE.COM (william E Davidsen) Subject: Re: Kernel crash: Need help fixing my Ext2fs Reply-To: davidsen@crd.ge.com (bill davidsen) Date: Wed, 14 Apr 1993 13:55:21 GMT In article <1qgakk$15f@agate.berkeley.edu>, umminger@purina.berkeley.edu (Frederick W. Umminger) writes: | | I should mention this also, now that I've noticed it; my | ext2fs apparently ran out of space, at least for normal users, | just before the crash. Taylor uucp is taking up 17 megs of disk | space! How absurd. Has ext2fs been tested extensively for its | behaviour when it runs out of space? I tested it just a little, when I rebooted from a maint disk the filesystem wouldn't mount because it was mangled (so I couldn't release space), and wouldn't e2fsck because it was full (I got some superblock error message about the filesystem having more data than physical filespace). Nothing I couldn't fix with a low level format, repartition, and install ;-) -- bill davidsen, GE Corp. R&D Center; Box 8; Schenectady NY 12345 ------------------------------ From: wong@cs.tu-berlin.de (Wolfgang Jung) Subject: Re: Xfree 1.2 leaves screen in graph-mode after exit, runx !work Date: 14 Apr 1993 14:04:45 GMT Subject: Re: Xfree 1.2 leaves screen in graph-mode after exit, runx !work Newsgroups: comp.os.linux References: <1993Apr11.054139.1007@amscons.com> Organization: Technical University of Berlin, Germany Bob Amstadt (bob@amscons.com) wrote: : ado@bigcomm.gun.de (Christoph Adomeit) writes: : : the newest version of xfree leaves my et4000 card in graphics mode after : : exit. I also can't change VTY's, because they only contain of garbaged data : : in VGA-Mode. : : The old xfree has had similar problems, but I could solve them by using : : "runx" from the vgalib distribution. This doesn't work anymore too. : : Any ideas ? : Add MGA cards to those affected by this problem. : -- : Bob Amstadt : bob@amscons.com Try fixing it with using fontpack I had Success in that. BTW Who knows where a VGA Card stores its Textmode Fonts (CodePages} It might be that the normal console Switching stuff won't restore codepages or save codepages.. Gruss Wolfgang ------------------------------ From: robin@dcs.warwick.ac.uk (Rob Rainton) Subject: Re: [Q] Has XS3 merged with Xfree? Date: Wed, 14 Apr 1993 14:02:43 GMT In article joe@stat.psu.edu (Joe Broniszewski) writes: >I am trying to get X working with an Orchid Faherhiet Local bus card. I >have XS3 working, but there are a few bugs with it. One of the README >files says that XS3 will be merged with Xfree 1.2, but a friend of mine >told me that he has can't get Xfree to work with his Orchid Faherhiet >card. Any experience/help? On a related note - has anybody done/or is planning to try and get Xfree to support other accelerators? I would like to get a linux machine soon (in a couple of months), but don't want to be limited to having to buy an S3 accelerator when there are other cards out there claiming (don't know if this is true or not) to be faster for similar price. I use SUN systems at the mo, which have a variety of accelerators (/dev/cg[two|three|six|etc...]) I suppose if you wanted to develop Xfree for some other chip set we'd have to agree on a device driver to start with. If, when I get a system and there was still nothing done on a card I bought I'd be willing to have a bash at this, but how much info, help, etc. is there about, and how long do you reckon it would take (this bit mainly aimed at whoever wrote XS3)? Just a couply of thoughts, Rob. ------------------------------ From: u9050728@jasper.cs.uow.edu.au (Shane Kelvin Richards) Subject: mice problems with Xwindows Date: 14 Apr 1993 23:52:32 +1000 A quick questino I am sure has a simple answer. I have a 2 button microsoft mouse. Now Xterm seems to want to use the middle button to paste text. What do I do if I am using a microsoft mouse (so I only have 2 buttons), is there a keyboard equivalent? Or some mouse sequence which does the equivalent of pressing the middle button??? -- +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- Shane Richards u9050728@cs.uow.edu.au +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- ------------------------------ From: jeremy@sw.oz.au (Jeremy Fitzhardinge) Subject: Re: New feature for the filesystems. What do you think ? Date: 14 Apr 93 12:06:35 GMT In twpierce@unix.amherst.edu (Tim Pierce) writes: >I think that hacking bits into the existing filesystem code or >designing entirely new compressed filesystems would be an unwise way >of attacking the problem. I agree. I'm beginning to cringe with horror every time a see an article of the form: ---- From: studly@microsoft.con Subject: NEW FILESYSTEM IDEA! I've just started using Linux, and I think it's way cool. I've been using Stacker for a couple of years now, and I want to write a version for Linux. All you need to do is put gzip into xiafs. PS. Has anyone done a version of SmartDrive for Linux? ---- I suppose it's seen as a good idea because it's standard in DOS 6. However, I can't think of many compelling reasons why such a thing is needed in unix, let alone the unix kernel. Uncompressing files on the fly mucks lots of assumptions in the basic workings of the filesystem and vm system, and prevents things from happening that are a bigger win in the long term. It seems to me that a cron job that finds long-unused files (most of them) and compresses them would be more useful. The tcx (?) system is example of this idea. Frequently used files shouldn't be compressed, its not so hard to decompress files that are only used every now and then. Some kernel support would be nice, but not by putting gunzip into /Image. >Better yet might be to look into layering the entire filesystem >structure, like a little microkernelization of the operating system. >Besides the issue of compressing the data in your filesystem, a >transparent fs interface leads the way to allowing users to mount tar >files as if they were their own filesystems, obviating the need for >detarring an entire archive in order to process individual files >therein, or mounting FTP archives or the like. This is like something I'm currently working on. I've written a filesystem which is essentially a stub which communicates with a user process through a pair of file descriptors attached to pipes or sockets. It allows the user process to either completely synthesize a filesystem at the mountpoint, or generate it from another source. Apart from filesystem prototyping, the kind of applications I can see for it are a compressing filesystem (which I can't see as being very useful) or an encrypting one. Other ideas are a /u filesystem that is just appears with a symlink for each user, pointing to their home dirs (a good way of not needing ~ expansion that works for everything), an ftp filesystem (tho Alex has preempted that) or mounting tar files. The nice thing is that the kernel remains small (userfs makes the kernel less than 10k bigger), but offers a facility which has unlimited flexibility. Yes, this is a project influenced by Plan 9 :-). Currently its incomplete, but it might be ready for alpha testing in a few weeks. J ------------------------------ From: wong@cs.tu-berlin.de (Wolfgang Jung) Subject: IOCTL on Serial Lines Date: 14 Apr 1993 14:13:42 GMT Hi === While I tested my mouse (Playing with thos tesutilities . I tried to turn my mouse into MouseSzstems mode (DTR OFF it seems with my mouse. While doing this I added into the testprogram (derived from X} some code to turbn off the DTR. I obseved that DTR is always restored when a ioctl(fd,TCSETA,...} is called. This behavior of ioctl on serial lines is nowhere dokumented I looked at the manpages from SunOS due to unavailable ones for linux :-( Is this a Feature or is this a BUG. It makes it almos impossible to turn off the lines if it is needed. (I donot now if the other lines are affected the same I have no RS232 Line monitor :-( } Gruss Wolfgang ------------------------------ ** 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-Activists-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux) via: Internet: Linux-Activists@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 tupac-amaru.informatik.rwth-aachen.de pub/msdos/replace The current version of Linux is 0.99pl6 released on February 23, 1993 End of Linux-Activists Digest ****************************** e a see an article of the form: ---- From: studly@microsoft.con Subject: NEW FILESYSTEM IDEA! I've just started using Linux, and I think it's way cool. digest835 644 36676 74430 47051 5366137065 5767 From: Digestifier To: Linux-Activists@senator-bedfellow.mit.edu Reply-To: Linux-Activists@senator-bedfellow.mit.edu Date: Wed, 14 Apr 93 13:30:42 EDT Subject: Linux-Activists Digest #835 Linux-Activists Digest #835, Volume #4 Wed, 14 Apr 93 13:30:42 EDT Contents: SOS:Can't install by a1.test & a1 (Carpe Diem.....) Re: LINUX and data acquisition (william E Davidsen) flex vs. lex (Hendrik G. Seliger) NFS-Blues in 0.99-p8 (Hendrik G. Seliger) Re: Disk Quotas (was Re: New feature for the filesystems.) (william E Davidsen) Re: mice problems with Xwindows (Bob Spratt) Logitech S9 BusMouse malefunction (Tein H. Yuan) Re: How to get rid of LIL (John Will) Re: tape retention? (John Will) Re: [Q] Has XS3 merged with Xfree? (Wolfgang Jung) monitor settings for new Xfree 1.2 (Gregg Jensen) test (rmag@simula.efis.ucr.cr) Has anyone run Minix 1.5 in the DOS emulator? (Scott Taylor) RE: Linux on Amiga? (nmspillers@ualr.edu) Help: Netwok problem. ---------------------------------------------------------------------------- Crossposted-To: tw.comp.unix From: jsyu@Uz.nthu.edu.tw (Carpe Diem.....) Subject: SOS:Can't install by a1.test & a1 Date: Wed, 14 Apr 1993 14:17:30 GMT Dear Netters, Now I come accross another problem while installing the SLS package. Here is my partition table: ========================================================================== Disk /dev/hda: 8heads, 39 sectors, 761 cylinders Units = cylinders of 312 * 512 bytes Device Boot Begin Start End Blocks ID System /dev/hda1 * 1 1 493 76888+ 6 DOS 16-bit >=32MB /dev/hda2 494 494 729 36816 81 Linux/MINUX /dev/hda3 730 730 761 4992 82 Linux swap ========================================================================== Boot by a1, I have done the following commands with no error messages: # mkswap /dev/hda3 4992 # swapon /dev/hda3 # mke2fs /dev/hda2 36816 But now when I started to run "doinstall /dev/hda2" , the following messages appeared when installing the a3 disk: MINIX-fsmagic mathc failed [MS-DOS FS Rel.alpha.8, FAT 16, check=n, conv=b, uid=0, gid=0, umask=000] [me=0x50,cs=2,#f=246,fs=23412,fl=12574,ds=36424,de=32962,data=38484,se=22133, ts=1963212550] Unsupported FS Parameters EXT-fs: magic match failed Insert disk a3 into the floppy drive then hit enter, q to quit Then after I pressed "enter", the same message appeared. I tried 3 times and was quite sure that the installation failed when the below appeared: mv:can not move `/root/fstab.tmp' to `/root/etc/fstab':No such file or directory I also tried another way: to install with the LILO a1.test disk. I did the same procedure as above, and went quite fine before the " doinstall /dev/hda2 " command. This time the following message appeared when installing a2 disk: mount: mount point /root doesn't exist. Error: Can not mount /dev/hda2. Did you use: mkfs /dev/hda2 SIZE where SIZE is the number of blocks shown by fdisk? But I did have done "mke2fs /dev/hda2 36816" already... Can anyone help me with this problem? Thanks very much... Sincerely Jeanshall Yu ( jsyu@uz.nthu.edu.tw or b800456@xlab.nthu.edu.tw ) ------------------------------ From: davidsen@ariel.crd.GE.COM (william E Davidsen) Subject: Re: LINUX and data acquisition Reply-To: davidsen@crd.ge.com (bill davidsen) Date: Wed, 14 Apr 1993 14:22:07 GMT In article , dd435157@longs.LANCE.ColoState.Edu (Dan Doner) writes: | | I realize that *nix's method of interrupt handling is not good for data acquisition, | however, is it possible for digital/analog data acquisition in the 1000 Hz range? Has | anyone tried this? Depends on the interrupt rate. On a 386DX-40 Linux seems to live with 400-600 int/sec without dropping (serial data). But the UART has a FIFO so the required response is something like 6ms (12 bytes at 19200). Linux will run faster, but the int rate generally doesn't get higher, the FIFO gets deeper. If your hardware will fill a buffer or something, I'm sire you can make it work. If you can come in via serial port, that will work. If you need better latency than 2ms you may have to limit what you do on the machine to reach that level. If you write an IEEE-488 driver for Linux I know someone who would like to know about it (maybe 9000 systems worth). -- bill davidsen, GE Corp. R&D Center; Box 8; Schenectady NY 12345 ------------------------------ From: hank@Blimp (Hendrik G. Seliger) Subject: flex vs. lex Date: Wed, 14 Apr 1993 14:19:43 GMT Reply-To: hank@Blimp.automat.uni-essen.de I didn't find any direct info on this question, so I "dare" posting it: quite often when I try to compile software on linux there are files for lex or yacc. Usually I think it's a good thing to have scanners and parsers generated automagically, but I seem never to be able to use files written for lex with flex. (Yes, I DO change the library to -lfl!) There are always quite some undefined symbols. Mostly I get this straightened out by pushing the file through lex on a Sparc2 and the hand-editing it so it goes along with libfl. But that certainly can't be the way to go. Could someone push me back on right path, please! How do I do it the right way?? Thanks for any advice, Hank. -- ====================================================================== Hendrik G. Seliger Universitaet Essen hank@Blimp.automat.uni-essen.de Schuetzenbahn 70 Tel.: +49-201-183-2898 4300 Essen, Germany ====================================================================== Pajari: "Handling interrupts is simple." A. Tanenbaum: "Interrupts are an unpleasant fact of life." ------------------------------ From: hank@Blimp (Hendrik G. Seliger) Subject: NFS-Blues in 0.99-p8 Date: Wed, 14 Apr 1993 14:28:23 GMT Reply-To: hank@Blimp.automat.uni-essen.de There seems to be something wrong in the NFS-code in 0.99-p8. Today I got the new kernel and also the new nfs-server. I tried all combinations, so I know the following problem is caused by the new kernel, NOT the new nfs-server (nfs-server 1.5 with the p7-kernel works). So here is the problem: My /usr/spool/mail is a sym-link via a mount-point to another machine (also linux), where all mail is sent to. This works fine. Just with the new kernel elm can't open /usr/spool/mail/username for reading anymore. The user can look at the file using more, can edit it with emacs, can receive mail, just elm won't work. I tracked it down to probably being a group permission problem in the nfs-code. Users belonging who have an additional *group* membership of root can read their mail, others cannot. For now I rebooted using the old kernel. But I guess one should lighten this up a bit. (Yes, IF I should find time I'll have a look myself, to help all those great people out their dedicating their time to Linux.) Hank. -- ====================================================================== Hendrik G. Seliger Universitaet Essen hank@Blimp.automat.uni-essen.de Schuetzenbahn 70 Tel.: +49-201-183-2898 4300 Essen, Germany ====================================================================== Pajari: "Handling interrupts is simple." A. Tanenbaum: "Interrupts are an unpleasant fact of life." ------------------------------ From: davidsen@ariel.crd.GE.COM (william E Davidsen) Subject: Re: Disk Quotas (was Re: New feature for the filesystems.) Reply-To: davidsen@crd.ge.com (bill davidsen) Date: Wed, 14 Apr 1993 14:27:58 GMT In article , nick@symphony.mp.ucc.ie (Nick Hilliard) writes: | With all this talk about new features for the various filesystems, I figure | no-one's really mentioned anything about Disk Quotas. On a public access UNIX I sure would like that!! | | Ok - it's fine if you're a just using Linux as a single user machine, but if | you have a whole load of users hanging out of it, you really _do_ need some | form of mechanism for making sure that people don't hog the machine. Less hassle than sending the boys in the black limo after them, for sure. Or even cleaning up their files and sending them a nastygram. | | Is there a chance of seeing code for them being written in the (near) | future? A related question, the V.4 (ufs) filesystem doesn't by default reserve a lot of room for root only. The 10% reserve to prevent quadradic behavior (slowdown) and fragmentation is not there. Neither are the slowdown or fragmentation. And you get 32 bit inodes, so you don't run out of inodes all the time. Sure would be nice to have this functionality on Linux. -- bill davidsen, GE Corp. R&D Center; Box 8; Schenectady NY 12345 ------------------------------ From: bobs@access.isc-br.com (Bob Spratt) Subject: Re: mice problems with Xwindows Date: 14 Apr 93 14:51:29 GMT Shane Kelvin Richards (u9050728@jasper.cs.uow.edu.au) wrote: : A quick questino I am sure has a simple answer. I have a 2 button : microsoft mouse. Now Xterm seems to want to use the middle button to paste : text. What do I do if I am using a microsoft mouse (so I only have 2 buttons), : is there a keyboard equivalent? Or some mouse sequence which does the : equivalent of pressing the middle button??? : -- : +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- : Shane Richards : u9050728@cs.uow.edu.au : +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- You should have the emulate 3 button set in your Xconfig and then you should simply be able to hit both buttons at the same time to emulate the middle button. -- Bob Spratt EMAIL: bobs@oakbrook.oak.isc-br.com Olivetti North America 1315 West 22nd Street Suite 410 Oak Brook, Il 60521 (708) 573-5923 ------------------------------ From: tein@hawk.cc.as.edu.tw (Tein H. Yuan) Subject: Logitech S9 BusMouse malefunction Date: Wed, 14 Apr 1993 15:02:34 GMT Hi Folks Currently, I installed Linux 0.99p6 and I couldn't run X because my Logitech S9 BusMouse functions weird. Now, I set it to IRQ 5. No matter how I move the mouse, the cursor just move to the lower left corner and then disappear. It acts as the mouse driver doesn't recognize the screen resolution. I use 1024x768 ET4000 interlaced mode. Any suggestion? Please contact me directly. Thanx in advance. Tein ------------------------------ From: john.will@satalink.com (John Will) Subject: Re: How to get rid of LIL Date: 14 Apr 93 09:10:00 GMT Reply-To: john.will@satalink.com (John Will) GD>First: MAKE A BACKUP OF YOUR BOOT/PARTITION SECTOR!! GD> GD>Second: Backup a DOS systems boot partition. (Same drive technology -- GD>IDE -- size may be different). GD> GD>Third: Restore your DOS partition to your original LILO system. GD> GD>Fourth: DELETE all partitions listed in the new table, and change the GD> settings so fdisk understands your disk geometry. GD> GD>THIS METHOD IS NOT RECOMMENDED FOR BEGINNERS. IF YOU DON'T UNDERSTAND GD>ANYTHING ABOUT ANY OF THE STEPS I'VE MENTIONED, DON'T EVEN TRY IT -- IT GD>COULD EASILY MAKE YOUR DRIVE "UNUSABLE". GOOD GOD!!! How about 'FDISK /MBR'??? Takes about 5 seconds and restores the standard DOS boot sector. And my method is specifically for beginners, though you "experts" might consider it a bit easier than your method. :-) --- ~ KingQWK 1.05 # 97 ~ Never test for an error you don't know how to handle. ------------------------------ From: john.will@satalink.com (John Will) Subject: Re: tape retention? Date: 14 Apr 93 09:10:00 GMT Reply-To: john.will@satalink.com (John Will) MF>type mt -f /dev/ retension MF> MF>replace with the tape device. Should work wonders. Only if you've managed to get the tape patches to install in your new version of the kernel! :-( I wonder if there's a chance that the QIC-02 stuff will ever be part of the basic kernel, so it isn't such a raindance to get tapes working each time a release comes out? --- ~ KingQWK 1.05 # 97 ~ Gun control is being able to hit your target! ------------------------------ From: wong@cs.tu-berlin.de (Wolfgang Jung) Subject: Re: [Q] Has XS3 merged with Xfree? Date: 14 Apr 1993 15:44:37 GMT Rob Rainton (robin@dcs.warwick.ac.uk) wrote: : In article joe@stat.psu.edu (Joe Broniszewski) writes: : >I am trying to get X working with an Orchid Faherhiet Local bus card. I : >have XS3 working, but there are a few bugs with it. One of the README : >files says that XS3 will be merged with Xfree 1.2, but a friend of mine : >told me that he has can't get Xfree to work with his Orchid Faherhiet : >card. Any experience/help? : On a related note - has anybody done/or is planning to try and get : Xfree to support other accelerators? This is on the way. I don't knwo when a public release is available, but there are several People working on supporting ACCelorated cards. Gruss Wolfgang ------------------------------ From: gj9362@beorn.sbc.com (Gregg Jensen) Subject: monitor settings for new Xfree 1.2 Date: 14 Apr 1993 10:39:02 -0500 OK, I took the plunge and upgraded my X to 1.2 (via SLS). I was running with a 832x600 at 36 just fine. I put in the EXACT same information in the Xconfig and when I ctl-alt-+ over to that setting the screen is all out of sync and messy. My setup is as follows: Samsung Multi-sync monitor with a Diamond Speedstar Plus. With the previous version of X this set up worked just fine (no problem with the clock). I am now wondering if I am being bitten by the Diamond bug and need to do something else to make my set up work (like run something called freq???)? I have made the Xconfig for this new version have as much of the same information as the previous version to try to reduce any errors. I can't provide the speeds for monitor and card as they are at home. If it helps I purchased the monitor in '88 and the card in 90(maybe early 91). I have been thru the FAQ's and will be going thru the modegen tonight. But, I played with modegen before, thats how I came up with the 832x600 setting. So, if modegen hasn't changed for this new version, it will probably be a futile effort. Any other help? Gregg Jensen gj9362@smaug.sbc.com -or- gj9362%smaug@texbell.sbc.com ------------------------------ From: rmag@simula.efis.ucr.cr Subject: test Date: 14 Apr 1993 13:04:48 -0400 Reply-To: rmag@simula.efis.ucr.cr This is just a test... please ignore. -- ********************************************************************** J.R. Maga\~na (TeX convention) also: Laboratorio de Simulaciones. Escuela de Fisica, Universidad de C.R. ********************************************************************** ------------------------------ From: n217cg@tamuts.tamu.edu (Scott Taylor) Subject: Has anyone run Minix 1.5 in the DOS emulator? Date: 14 Apr 1993 15:47:20 GMT Due to an end-of-semester project in my operating systems class, I need to run Minix 1.5 on my PC. Rather than having to reboot whenever I need to work on it, I thought it would be really cool to run it in the DOS emulator. The problem is that even the universal boot diskette (which only uses BIOS calls to access the HW) won't boot in dosemu 0.48; I suspect that all the BIOS calls that Minix wants aren't there. Has anyone successfully done this? I don't have a lot of time right now to hack dosemu (whatever Linux hacking time I have I intend to spend on the Ultra- Stor 14F/34F driver), but if anyone has any "fast & easy" tips, I would really appreciate it! Scott n217cg@tamuts.tamu.edu ------------------------------ Subject: RE: Linux on Amiga? From: nmspillers@ualr.edu Date: 14 Apr 93 09:58:09 GMT 'Lo Larry, >GNU software is a major component of Linux. Please consider the FSF's >wishes wrt porting any of the GNU software to A.p.p.l.e. platforms. >Here are a few excerpts: > >From /usr/local/lib/emacs/etc/APPLE: > >> Some of you may be considering using, buying, or recommending Macintoshes; >> you might even be writing programs for them or thinking about it. Please >> think twice and look for an alternative. Doing those things means more >> success for Apple, and this could encourage Apple to persist in its >> aggression. It also encourages other companies to try similar >> obstructionism. >> >> [It is because of this boycott that we don't include support for Macintosh >> Unix in GNU software.] Look, I understand the FSF position regarding Apple, *but* I am still playing around with porting Linux to the Mac platform along with other motivated people (much more motivated than I am, as a matter of fact :-). The point is, there are a lot of folks out there with Macs. They are not encouraging facist computing, nor are they a member of Apple's "Evil Empire" and wander around with a Borg-like mentality--"My user interface will assimilate you, resistance is futile". They merely bought a nifty, well built machine that fills their needs, why punish them? Shoot, most of 'em don't even *know* about the FSF or GNU or any boycotts! The last thing I need to hear is, "Ohhh, you're doing a *bad* thing, you'd better stop it". I don't consider a Mac port wrong in any way shape or form. Besides, if the FSF doesn't like Apple's legal tactics regarding their user interface, what would be better than giving everybody a viable alternative to that Mac OS that Apple is so protective of? Ironically, the biggest stumbling blocks have been the lack of *useable* FSF or GNU tools to get the job started in the correct spirit--free tools used to develop a free OS. Actually, there *is* quite a lot of GNU stuff ported to the Mac, my only beef is that gcc needs the MPW to run under. Irregardless, I think I've solved that problem and I'm charging on. I want to learn, I want to do something worthwhile and give a little back to the computing community that has helped me so much, and I think this is a good way of achieving these aims. . . By the way, this is *not* an announcement of *anything*! I'm just starting out, I don't expect anything anytime soon--Heck I don't know how far I'll get or if I or others will even come close to such a beast in this decade!!! So much code, so little free time :-) nate nmspillers@ualr.edu "Life is a never-ending learning curve. . ." ------------------------------ From: u8013535@cc.nctu.edu.tw () Subject: Help: Netwok problem. Date: Wed, 14 Apr 1993 15:50:55 GMT Hi, I have set up my network and I can telnet, ftp and talk to user on other machine now, but at the same time, I got messages full of screen like this: eth_if : RX transfer address mismatch, 0x5740(should be) vs. 0x5748(actual) My ethernet card is DLink-200+, it is compatible to NE2000, IRQ 5, if anyone has experience about this problem, please response. Thanx for any help....... Lin. ------------------------------ ** 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-Activists-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux) via: Internet: Linux-Activists@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 tupac-amaru.informatik.rwth-aachen.de pub/msdos/replace The current version of Linux is 0.99pl6 released on February 23, 1993 End of Linux-Activists Digest ****************************** but if anyone has any "fast & easy" tips, I would really appreciate it! Scott n217cg@tamuts.tamu.edu ------------------------------ Subject: RE: Linux on Amiga? From: nmspillers@ualr.edu Date: 14 Apr 93 09:58:09 GMT 'Lo Larry, >GNU software is a major component of Linux. Please consider the FSF's >wishes wrt porting any of the GNU software to A.p.p.l.e. platforms. >Here are a few excerpts: > >From /usr/local/lib/emacs/etc/APPLE: > >> Some of you may be consdigest836 644 36676 74430 52026 5366137066 5767 From: Digestifier To: Linux-Activists@senator-bedfellow.mit.edu Reply-To: Linux-Activists@senator-bedfellow.mit.edu Date: Wed, 14 Apr 93 16:30:44 EDT Subject: Linux-Activists Digest #836 Linux-Activists Digest #836, Volume #4 Wed, 14 Apr 93 16:30:44 EDT Contents: WRONG (Re: probs with 4.3.3-shlibs and solution) (HJ Lu) Re: Backup libraries (was Re: The dangers of playing with shared libraries) (HJ Lu) recommendations wanted: tape drive (Marc G Fournier) X386: Mode for NEC 4FG (Juergen Peus) problem install SLS Linux (S. W. Chan) Re: New feature for the filesystems. What do you think ? (Brandon S. Allbery) Error using xgdbbin and linux 0.99 pl4 (Michael P. Jarreau) Re: New feature for the filesystems. What do you think ? (Rajat Datta) Re: Kernel crash: Need help fixing my Ext2fs (Karl Buck) Linux mascot.... :-) (Allan Bailey) Re: A modified LESS to accept arrow keys (Seng-Poh Lee, Speedy) Is bootp daemon available for Linux? (Seng-Poh Lee, Speedy) Re: UMC chipset motherboards? (Carsten Schiers) Re: NFS-Blues in 0.99-p8 (Linus Torvalds) Configuration (was Re: Kernel Buffer Size Configuration) (John Steele) Re: [Q] Has XS3 merged with Xfree? (Jon Tombs) Re: The dangers of playing with shared libraries (Charles Hannum) A modified LESS to accept arrow keys (root-bound) ---------------------------------------------------------------------------- From: hlu@luke.eecs.wsu.edu (HJ Lu) Subject: WRONG (Re: probs with 4.3.3-shlibs and solution) Date: Wed, 14 Apr 93 14:17:51 GMT In article <1qglnjINN8sh@server2.rz.uni-leipzig.de> brauer@aix520.informatik.uni-leipzig.de (Eckard Brauer) writes: >While Installing the libs distributed with image-4.3.3.tar.z I got the problems >that neither 'mount' nor 'df' worked correctly. Searching the reason I found, >that /etc/mtab was empty. Therefor I changed in /etc/rc the line: > /bin/rdev >/etc/mtab >to: > echo "`/bin/rdev` " > /etc/mtab >or: > echo -n "`/bin/rdev` " > /etc/mtab >where should be replaced by the type of the current root-fs. > I have warned you in c.o.l. Please dig out my old article. I may miss something here. 1. Get Rick Sladkey's mount 0.99.6. It is in mount-0.99.6.tar.Z and utile-1.4.src.tar.z from sunsite or tsx-11. 2. Remove any rdev > /dev/mtab from /etc/rc. 3. List / partition in /etc/fstab. 4. Add "mount -av" in /etc/rc or somewhere like that. 5. Always provide a fs name in /etc/fstab. After rebooting, you will be fine. H.J. -- School of EECS Internet: hlu@eecs.wsu.edu Washington State University BITNET: 60935893@WSUVM1.BITNET Pullman, WA 99164 Phone: (509) 334-6315 ------------------------------ From: hlu@luke.eecs.wsu.edu (HJ Lu) Subject: Re: Backup libraries (was Re: The dangers of playing with shared libraries) Date: Wed, 14 Apr 93 14:20:31 GMT In article <2033@dsbc.icl.co.uk> KID01 writes: >In article stevev@miser.uoregon.edu (Steve VanDevender) writes: >>In article <1qfot4$tin@hal.gnu.ai.mit.edu> >[..] > >Couldn't future version of linux binarys be compiled to first search for >/lib/libc.so.4 and if that fails try /lib/libc.so.4.3, etc, etc. This >shouldn't be a massive problem to setup and only if all of /lib was remove >would the user be in trouble ! (in which case the binaries could search in >/oops_i'v_lost_my_shared_libs/backup_here/libc.so.4) > > The current one tries /usr/lib, /lib and "" in that order. But not all binaries are linked with the new scheme. H.J. ------------------------------ From: marcf@nexus.yorku.ca (Marc G Fournier) Subject: recommendations wanted: tape drive Date: Wed, 14 Apr 1993 15:54:01 GMT Hi... I figure that it is about time I get a tape backup unit for my system...special now that it has decided it wants to fall apart :( I've been doing backups of my users stuff onto floppies...but taht is just too unbearably time-consuming, especially when you have to change floppies every couple of minutes. So, what I'm posting for is any suggestions or recommendations of what other ppl are using under Linux (ie brands?), with a SCSI interface. As well, how do you do a backup to a tape unit? I tend to use afio for doing to floppies, is there some different way of doing it to a tape? Any suggestions/recommendations are welcome :) thanks... marc ------------------------------ From: grobi@uni-paderborn.de (Juergen Peus) Subject: X386: Mode for NEC 4FG Date: 14 Apr 1993 16:14:19 GMT Hello Folks !! A friend of mine has newly installed Linux last weekend. The most things worked fine, but we had one major problem: we can't start X correctly with the right mode. I mean the mode you have to specify in the 'Xconfig' file in /usr/lib/X11. OK. He ownes the following hardware: Monitor: NEC Multisync 4FG 15" Graphic adapter: TARGA WIN 32K with Tseng et4000 high color chips machine: 486DX33 with 16MB ram and 256kb cache Has anyone of you the correct modes for this monitor, yes we tried to calculate them but it wouldn't work. If you have any positive answers to my questions, please e-mail to: grobi@uni-paderborn.de because I don't read this special newsgroup very often. Thanks in advance, J"urgen -- J"urgen P'eus (grobi@uni-paderborn.de) ------------------------------ From: chasw@essex.ac.uk (S. W. Chan) Subject: problem install SLS Linux Date: 14 Apr 93 16:15:03 GMT Reply-To: chasw@essex.ac.uk (S. W. Chan) I am new to Linux and am trying to install it to my system and experienced some problems. I have the following hardware spec. 486DX - 33MHz 8M RAM Seagate ST3238A IDE 245M hard drive (solely for Linux) Rodime IDE 101M hard drive (solely for DOS) 1.44M floppy drive SVGA display When I put disk a1 into the floppy and reboot, the system reads it without any problem until it comes to the part where the user has to key in drive that installation is done from. Below is the message that appear on the display: ============= ..... 4) Drive B, 5.25in Enter Drive You Will Be Doing The Installation From (1/2/3/4): 2 MINIX -fs magic match failed mount: unknown error 26240 /dev/ram on / type minix () Further instructions can be found in the file /user/README. Type 'more /user/README' to look at it. # ============= I tried to list the directory but no files or directories are found. I have also look at the latest posting of FAQ that suggested using 'rdev' to correct to problem, but the file can not be found. Can anyone tell me what is wrong and what should I do. Thanks in advance for any hint. ..Kenneth ****************************************************************************** Kenneth Chan S.W. e-mail: chasw@sx.ac.uk Department of ESE, University of Essex, Colchester, Essex, CO4 3SQ, UK ------------------------------ From: bsa@kf8nh.wariat.org (Brandon S. Allbery) Subject: Re: New feature for the filesystems. What do you think ? Date: Wed, 14 Apr 1993 16:19:49 GMT In article danielsi@cs.utexas.edu (Daniel Aaron Supernaw-Issen) writes: >change to the surrounding lfs code was that the head position had to be >byte indexed rather than block indexed. no biggie. You'll see a performance hit if you do it that way (byte indexes)... I would arrange for logical blocks (probably 4K) to be stored in any of several physical block sizes, from 1K to 4K (e.g. uncompressed). (I'd include 512 bytes but this is problematic when the kernel only supports 1K blocks.) ++Brandon -- Brandon S. Allbery bsa@kf8nh.wariat.org It's not too late to turn back from the "Gates" of Hell... Linux: the FREE 32-bit operating system, available NOW. Why waaaaaait for NT? ------------------------------ From: jarreau@vuse.vanderbilt.edu (Michael P. Jarreau) Subject: Error using xgdbbin and linux 0.99 pl4 Date: Wed, 14 Apr 1993 16:21:44 GMT Hi! I am trying to use xgdbbin (0.99 pl4) with no luck. I made the correct compiler switch (-g) and I started the debugger with xgdbbin -exec=program -symbol=program & The debugger started and loaded the code into the top window. However, when I click on run or if I try to type in the lower window I get the following error: [ioctl TIOCSPGRP failed in terminal_inferior: Not a typewriter] Do you recognize this error and can you make a suggestion as to how I might solve it? If this posting should not be in comp.os.linux which conference should it be a part of? Thanks, Michael (jarreau@vuse.vanderbilt.edu) ------------------------------ From: rajat@watson.ibm.com (Rajat Datta) Subject: Re: New feature for the filesystems. What do you think ? Date: Wed, 14 Apr 1993 16:13:49 GMT In article jeremy@sw.oz.au (Jeremy Fitzhardinge) writes: >This is like something I'm currently working on. I've written a >filesystem which is essentially a stub which communicates with a user >process through a pair of file descriptors attached to pipes or sockets. >It allows the user process to either completely synthesize a filesystem >at the mountpoint, or generate it from another source. > > J Sort of like the Plan 9 interface for file (or anything else) servers. It would be neat to have something like the bind() call from Plan 9 that would let you attach a process to a particular file. All filesystem calls to that file would be get directed to that process. I guess streams might be another way to achieve the same thing. At the very least this would allow user level testing of filesystem code, and would allow one to develop a transparent networked filesystem with little effort (in fact, whether the filesystem is accessed from over the network or not becomes totally irrelevant). Extensions from filesystems are also obvious. This would be a way to get access to any device. To take an example from one of the the Plan 9 papers, you could implement a line printer daemon that mimics a filesystem. Creating a file in that directory is the same as sending it to the printer queue. rm removes it from the queue. ls -t prints out the queue in the order files will be printed, etc. Something like this would also let you establish specific access methods for specific files. For example, if you are worried about security and the encrypted passwords in /etc/passwd, you could bind a process before that file. Any program now reading that file will get the passwd records it expects, but with the passwd field stripped out. Only properly authorized programs would be allowed to see the real record. This would allow you to enforce an update policy for a database file, for example, by routing everything through a single server, and not have to rely on advisory locking. -- rajat (rajat@watson.ibm.com) ------------------------------ From: kxb@cis.ksu.edu (Karl Buck) Subject: Re: Kernel crash: Need help fixing my Ext2fs Date: 14 Apr 93 16:37:03 GMT umminger@purina.berkeley.edu (Frederick W. Umminger) writes: > I should mention this also, now that I've noticed it; my >ext2fs apparently ran out of space, at least for normal users, >just before the crash. Taylor uucp is taking up 17 megs of disk >space! How absurd. Has ext2fs been tested extensively for its >behaviour when it runs out of space? Well, it is in ALPHA testing. That means you take your own chances. I just had my msdos and root partitions bleed into each other. I'm not sure if it was the new lilo (0.10) or the ext2fs or a combination of the two. The point is if you use stuff in ALPHA testing, you have to expect the best and the worst at the same time. :) -- Karl Buck kxb@cis.ksu.edu USGMRL 913.776.2745 ------------------------------ From: fab9452@nigel.tamu.edu (Allan Bailey) Subject: Linux mascot.... :-) Date: 14 Apr 93 13:10:40 Some folks like to "pay homage" to mascots of their favorite thing/deity/etc. So, it that case, I'll *definately* vote for the Sauna as mascot idea. I mean, we all just gotta go sit in a sauna and talk about coding on Linux right? :^) -- Allan Bailey, UNIX programmer, CSC | "Freedom is not free." fab9452@tamsun.tamu.edu | or: allan.bailey@tamu.edu 616c6c616e206261696c6579 | ------------------------------ From: splee@pd.org (Seng-Poh Lee, Speedy) Subject: Re: A modified LESS to accept arrow keys Date: 14 Apr 1993 18:08:32 GMT root-bound (aehall@calvin.seattleu.edu) wrote: > I've modified the source of LESS to accept up/down arrows, pgup/pgdn, > and home/end from the PC keyboard. > less already supports programmable keys. Why would you hard code the keys in? Just create a file ~/.lessin such as the one below (for a VT100) and run lesskey on it. ^[[A back-line ^[[B forw-line ^[[C next-file ^[[D prev-file ^[[6~ forw-screen ^[[5~ back-screen ^[[1~ forw-search ^[[4~ set-mark ^[[2~ goto-mark ^[[28~ help ^[[29~ status ^[OR repeat-search ^[OP4 goto-end ^[OP5 goto-line e visual -- Seng-Poh Lee ------------------------------ From: splee@pd.org (Seng-Poh Lee, Speedy) Subject: Is bootp daemon available for Linux? Date: 14 Apr 1993 18:10:16 GMT Subject says it all. The tftp daemon is already available, and bootp should follow, right? -- Seng-Poh Lee ------------------------------ Crossposted-To: comp.sys.ibm.pc.hardware,comp.unix.pc-clone.32-bit From: schiers@tpki.toppoint.de (Carsten Schiers) Subject: Re: UMC chipset motherboards? Date: Wed, 14 Apr 1993 18:14:32 GMT I use an VL UMC board with 486DX2-66. As an article in c't stated, the ETEQ chipset is the only one that allows 10 MB/s busmaster DMA. My UMC works with 6.7 MB/s. It should at least work with 5.7 MB/s as the article says. CU Carsten. -- Carsten Schiers schiers@tpki.toppoint.de Lerchenfeld 9 2000 Hamburg 76 ------------------------------ From: torvalds@klaava.Helsinki.FI (Linus Torvalds) Subject: Re: NFS-Blues in 0.99-p8 Date: 14 Apr 93 18:06:18 GMT In article <1993Apr14.142823.25497@aixrs0.hrz.uni-essen.de> hank@Blimp.automat.uni-essen.de writes: >There seems to be something wrong in the NFS-code in 0.99-p8. Today I >got the new kernel and also the new nfs-server. I tried all >combinations, so I know the following problem is caused by the new >kernel, NOT the new nfs-server (nfs-server 1.5 with the p7-kernel >works). So here is the problem: > >My /usr/spool/mail is a sym-link via a mount-point to another machine >(also linux), where all mail is sent to. This works fine. Just with >the new kernel elm can't open /usr/spool/mail/username for reading >anymore. The user can look at the file using more, can edit it with >emacs, can receive mail, just elm won't work. I think it's the 'access()' bug that was uncovered by the pl8 NFS code. It's not actually a NFS bug, but some silly coding in the sys_access() routine in linux/fs/open.c - the following short and simple patch should fix it. The patch is also part of the ALPHA-diffs I made for the msdos people due to the vm86 signal problem.. ===== snip snip ===== --- l99.8/linux/fs/open.c Thu Mar 11 19:39:28 1993 +++ linux/fs/open.c Fri Apr 9 23:31:03 1993 @@ -167,11 +167,11 @@ return res; i_mode = inode->i_mode; res = i_mode & 0777; - iput(inode); if (current->uid == inode->i_uid) res >>= 6; else if (in_group_p(inode->i_gid)) res >>= 3; + iput(inode); if ((res & mode) == mode) return 0; /* ===== snip snip ===== As you can see, it's just a 'iput()' call that is made too early: it works on the other filesystem types due to the inode caching, but under NFS this breaks.. Linus ------------------------------ From: jsteele@netcom.com (John Steele) Subject: Configuration (was Re: Kernel Buffer Size Configuration) Date: 14 Apr 93 17:59:04 GMT In <1q03e6INNq18@senator-bedfellow.MIT.EDU> Theodore Ts'o writes: > From: bobm@anasazi.com (Bob Maccione) > Date: Thu, 1 Apr 1993 04:54:20 GMT > I looked all around but noticed that Linux doesn't have a /etc/conf > type area that can set how much memory is allocated to the buffers. > I'm running a 486 33 (DLC ) and notice that free tells me it's using > 3000+ in the buffer count and even when I first login I only have about > a meg available. Is there a way to change this in the kernel. ( I'm > on ver 99.6 or so ). >Linux uses dynamic buffer allocation --- so any memory which isn't used >for holding processes is used to hold buffers; as you start needing more >memory for your processes, Linux will automatically give up buffer space >to free memory. And, if there is unused memory, Linux will >automatically use it for buffer space. >This is far surperior to fixed allocation systems for buffer space, >since it means that your memory is much more efficiently used. > - Ted Yes, it most certainly is. Although this does bring up the question of configuration... With a NE2000 card, a multi-port serial, a sound-card, a QIC-36 tape controller and the like, it sure would be nice to have configurable IRQs, port addresses, etc available without multi-kernal remakes. Has anyone else done any thinking about this? Just some ideas, John -- #include /* route all flames to /dev/null */ /**************************************************************************/ /* Systems Analyst | John Steele email: jsteele@netcom.com */ /* Video Business Systems | */ /* 1-800-255-3088 | Most people spell COBOLSUX incorrectly... */ /**************************************************************************/ ------------------------------ From: jon@robots.ox.ac.uk (Jon Tombs) Subject: Re: [Q] Has XS3 merged with Xfree? Date: 14 Apr 93 18:24:45 GMT In article <1993Apr14.140243.16353@dcs.warwick.ac.uk> robin@dcs.warwick.ac.uk (Rob Rainton) writes: >In article joe@stat.psu.edu (Joe Broniszewski) writes: >>I am trying to get X working with an Orchid Faherhiet Local bus card. I >>have XS3 working, but there are a few bugs with it. One of the README >>files says that XS3 will be merged with Xfree 1.2, but a friend of mine >>told me that he has can't get Xfree to work with his Orchid Faherhiet >>card. Any experience/help? As of Xfree86-1.2 I moved from maintaining XS3 to adding support for S3 cards in Xfree86, The S3 support will not make it into the next release of xfree86 (be it either 1.2.1 or 1.3), but will definetly be there in 2.0. Don't ask me when v.2.0 will arrive, but it won't be this or next month. There are plenty of bugs and performance problems to be ironed out yet. >On a related note - has anybody done/or is planning to try and get >Xfree to support other accelerators? Support for mach8, ibm8514, and probably cirrus are in the makings. This can probably be better answered in comp.windows.x.i386unix. Jon. ------------------------------ From: mycroft@hal.gnu.ai.mit.edu (Charles Hannum) Subject: Re: The dangers of playing with shared libraries Date: 14 Apr 1993 14:54:25 -0400 In article davidsen@crd.ge.com (bill davidsen) writes: > > Given that there are bootable rootdisks, and the SLS bootable plus a > rootdisk, I'm not sure that building a suite of oversize utilities > makes any sense when you can run from the floppy and do what you > need. Having to reboot from floppies is bloody annoying. I don't understand why people are so vehemently against have a /sbin. Even *Sun* got that right. -- \ / Charles Hannum, mycroft@ai.mit.edu /\ \ PGP public key available on request. MIME, AMS, NextMail accepted. Scheme White heterosexual atheist male (WHAM) pride! ------------------------------ From: aehall@calvin.seattleu.edu (root-bound) Subject: A modified LESS to accept arrow keys Date: 14 Apr 1993 16:38:58 GMT I've modified the source of LESS to accept up/down arrows, pgup/pgdn, and home/end from the PC keyboard. If anyone's interested, I can upload it (ftp) somewhere. Anthony -- Anthony Hall _ _ Unix System Administrator aehall@seattleu.edu /_/ /_/ Physician Micro Systems, Inc. _ _ 2033 6th Ave Suite 707 /_/ /_/ Seattle, WA 98122 206-441-8490 ------------------------------ ** 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-Activists-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux) via: Internet: Linux-Activists@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 tupac-amaru.informatik.rwth-aachen.de pub/msdos/replace The current version of Linux is 0.99pl6 released on February 23, 1993 End of Linux-Activists Digest ****************************** sound-card, a QIC-36 tape controller and the like, it sure would be nice to have configurable IRQs, port addresses, etc available without multi-kernal remakes. Has anyone else done any thinking about this? Just some ideas, John -- #include /* route all flames to /dev/null */ /**************************************************************************/ /* Systems Analyst | John Steele email: jsteele@netcom.com */ /* Video Business Systedigest837 644 36676 74430 51374 5366137073 5773 From: Digestifier To: Linux-Activists@senator-bedfellow.mit.edu Reply-To: Linux-Activists@senator-bedfellow.mit.edu Date: Wed, 14 Apr 93 18:45:27 EDT Subject: Linux-Activists Digest #837 Linux-Activists Digest #837, Volume #4 Wed, 14 Apr 93 18:45:27 EDT Contents: Re: The dangers of playing with shared libraries (Brandon S. Allbery) Re: The dangers of playing with shared libraries (Brandon S. Allbery) Anyone using Orchid VA & S3? (Gord Vreugdenhil) Re: Linux on Macintrashes? (Was: Re: Linux on (Chris Higgins) Re: flex vs. lex (-* FireHawk *-) keyboard w/o X-Windows (German_latin1) (bro@clio.iwr.uni-heidelberg.de) Re: weird vi/less screen insect (Ben Cox) clone(), threads, and POSIX (Thomas McWilliams) Re: What are the .z files? (Zack Evans) Re: Mailing list for Newbies? (Jim Graham) Linux on Macintrashes? (Was: Re: Linux on Amiga?) (Steve Davis) Re: PS/2 style mouse not supported?? (E.A.Neonakis) Re: What are the .z files? (Zack Evans) ---------------------------------------------------------------------------- From: bsa@kf8nh.wariat.org (Brandon S. Allbery) Subject: Re: The dangers of playing with shared libraries Date: Wed, 14 Apr 1993 16:25:46 GMT In article <1qfot4$tin@hal.gnu.ai.mit.edu> ericy@hal.gnu.ai.mit.edu (Eric Youngdale) writes: > This is what the bootable rootdisk is for. I have yet to see a good >argument for why the bootable rootdisk cannot be used instead of specially >staticly linked binaries to fix screwups with the sharable libraries. Nobody said it *couldn't* be used... but a strategic statically-linked binary could be used to fix things more quickly and without a reboot. If you want to repair things *without* blowing off that program that's been churning in the background for an hour, you want tools that are installed on the running system. And while pdksh is certainly smaller than bash, rc is smaller yet... and the bootable rootdisk is already squeezed for space. In any case, I didn't say -- or intend --- that all those suggestions should actually be followed; I was just commenting on the individual suggestions themselves. Personally, I have the bootable rootdisk because I don't really *have* any big background churners on my system (yet :-) ++Brandon -- Brandon S. Allbery bsa@kf8nh.wariat.org It's not too late to turn back from the "Gates" of Hell... Linux: the FREE 32-bit operating system, available NOW. Why waaaaaait for NT? ------------------------------ From: bsa@kf8nh.wariat.org (Brandon S. Allbery) Subject: Re: The dangers of playing with shared libraries Date: Wed, 14 Apr 1993 16:27:59 GMT In article <1993Apr14.032021.12691@ksmith.com> keith@ksmith.com (Keith Smith) writes: >Bare minimum is your shell, 'cat' and 'mv' IMHO. 'ln' would be nice, Assuming an 8-bit-clean shell (so you can handle binary files), cat is while read line; do echo line; done < infile > outfile (Well, assuming your shell can handle LARGE lines... again, significant only if you are going to copy binaries with it.) ++Brandon -- Brandon S. Allbery bsa@kf8nh.wariat.org It's not too late to turn back from the "Gates" of Hell... Linux: the FREE 32-bit operating system, available NOW. Why waaaaaait for NT? ------------------------------ From: gvreugde@plg.uwaterloo.ca (Gord Vreugdenhil) Subject: Anyone using Orchid VA & S3? Date: Wed, 14 Apr 1993 18:15:32 GMT I am upgrading my system at the end of the month. The video card seems to be the hardest decision.... I suspect that the Orchid VA/VLB (Vesa) card will be my choice. Is anyone using this card with the S3 server? If so, let me know. I would prefer to know that it works before buying the card... :-) Alternatively, a local bus card which does 8514/A would be nice, though I would prefer not to spend the money for an ATI Ultra. Gord. ------------------------------ From: chris@Odyssey.ucc.ie (Chris Higgins) Subject: Re: Linux on Macintrashes? (Was: Re: Linux on Reply-To: chris@Odyssey.ucc.ie Date: Wed, 14 Apr 1993 18:21:39 GMT In article 1qhhd7INN7m0@matt.ksu.ksu.edu, strat@matt.ksu.ksu.edu (Steve Davis) writes: >nmspillers@ualr.edu writes: > >[excerpt from a GNU vs. Apple thread deleted] > >>The point is, there are a lot of folks out there with Macs. They are not >>encouraging facist computing, nor are they a member of Apple's "Evil Empire" >>and wander around with a Borg-like mentality--"My user interface will >>assimilate you, resistance is futile". They merely bought a nifty, well >>built machine that fills their needs, why punish them? Shoot, most of 'em >>don't even *know* about the FSF or GNU or any boycotts! > >If they don't know about FSF or GNU it is because they *have* been >'assimilated' by the Macintrash empire. If they were dumb enough to >buy an over-priced-one-mouse-button-tiny-black-and-white-screen- >bloated-awful-looking-os-without-a-cli-and-built-by-fascists computer >in the first place, there's really no reason for them to want or need >UNIX. After all, purchasing a Macintrash is acceptance of its point >and drool interface, AND a token of acceptance for Apple's business >practices-- both of which go against the free software/UNIX tradition. The thing is, that if they have a MAC and they don't know about the FSF/GNU then they won't want a CLI, they won't want Un*x. So I agree that porting Linux to the MAC would be in violation of the FSF recomendation/guideline/rule/whatever. BUT, having said that, I do believe that the FREE OS / FREE Software thing from GNU will have to be clarified.... > >>I want to learn, I want to do something worthwhile and give a little back to >>the computing community that has helped me so much, and I think this is a good >>way of achieving these aims. . . > >Not at all. The best way would be to chuck that Mac out the window >and buy a real computer. The MAC *IS* a real computer. ('real' does not imply 'good')... It's the politics behind it that is the problem. > >Stratocaster >-- > Steve Davis (I'm a student, not a spokesperson!) > strat@cis.ksu.edu - Kansas State University - Manhattan KS > >I assume that as a beginner you'll just want to blow people up. -- Red Shirt Why stop when you are no longer a beginner ? --- Chris. + J.C. Higgins, Sys Admin + Auditor, Computer Science Society, U.C.C. + Chris@csvax1.ucc.ie + + Chris@odyssey.ucc.ie + If you love something, set it free. If it doesn't + C.Higgins@iruccvax.ucc.ie + come back to you, hunt it down and KILL it. ------------------------------ From: firehawk@centauri.unm.edu (-* FireHawk *-) Subject: Re: flex vs. lex Date: 14 Apr 1993 18:44:30 GMT In article <1993Apr14.141943.9799@aixrs0.hrz.uni-essen.de> hank@Blimp.automat.uni-essen.de writes: : :I didn't find any direct info on this question, so I "dare" posting :it: quite often when I try to compile software on linux there are :files for lex or yacc. Usually I think it's a good thing to have :scanners and parsers generated automagically, but I seem never to be :able to use files written for lex with flex. (Yes, I DO change the :library to -lfl!) There are always quite some undefined symbols. I encounter problems going from lex -> flex on just about any system that has it. One of the main incompatabilites I have is the yyless() function. If someone has a decent lex package for linux, would they please post where to find it? If not, is there a flex equivelent for yyless()? -Jeff -- INTERNET: firehawk@carina.unm.edu ------------------------------ From: bk1@aixfile1.urz.uni-heidelberg.de (bro@clio.iwr.uni-heidelberg.de) Subject: keyboard w/o X-Windows (German_latin1) Date: Wed, 14 Apr 93 16:06:48 GMT Having recompiled the kernel the German_latin1 keyboard which I personally prefer worked quite well. Even though there are some changes which I made and which might be a suggestion to others. 1. Control-keys: keyboard.c does them by pressing control+ the combination to the corresponding key. This differs from what is common for German keyboards: ctrl-"u (not ctrl-altgr-8) ^[ ctrl-+ (not ctrl-altgr-9) ^] ctrl-# (not ctrl-altgr-"s) ^\ ctrl-2 (not ctrl-altgr-Q) ^@ ctrl-6 (not ctrl-^) ^^ ctrl-_ (that's the same) ^_ Those who do not know the German keyboard can still get the idea. It seems to be necessary to have a control-map for each keyboard-type. 2. altgr-keys: Some characters (like {[]}@~|) can only be accessed with altgr+something (7890q+<). That is ugly for some of them, because it requires holding the right hand in a strange way. Try, if you don't know. The keyboard driver of DOS 3.2 had the additional way to access these characters by ctrl-alt-+something which works much better for most of them. 3. numlock and shift: traditionally one could get the digits with shift, if numlock was off and the arrows with shift if numlock was on. That still makes sense to me. 4. esc-Sequences for arrow-keys and function-keys: There seem to be many esc-sequences for encoding the arrow-keys and things like insert, page-up, page-down, etc. Some look like \e[x (x is a letter), some like \eOx (x letter), some like \e[n~ (n number) and some like \e[nz (n number). I got an idea which sequences have which meaning but I do not know, when the different encoding methods are changed. The \eO-stuff is used in cursor set mode and application mode, the \e[-stuff otherwise. But the switching to these modes seems to depend on the machine... The \eOx and \e[x-stuff is easier to use for programs so I would tend to prefer it as much as possible. There are sequences of this type for page-up, page-down, PF1, PF2, PF3 and PF4. Why not map F1...F4 to the PF1...PF4? Also the well defined numpad number encoding \eOp..\eOy could be used for the ctrl-numpad-key-combinations and ctrl-arrow-key-combinations. This would make it possible (and makes it possible for me) to use these keys heavily for editing with a configurable editor (like emacs is). So I discribed my personal preferences. I would be interested if the suggestions which I made are shared or if there are other suggestions. In that case it might be useful to ask the author of keyboard.c for some enhancements or to supply them. I want to thank the author of keyboard.c for having written a useful and lucid program which could easily be understood and changed. ++bro@clio.iwr.uni-heidelberg.de++ (Karl Brodowsky) ------------------------------ From: thoth@uiuc.edu (Ben Cox) Subject: Re: weird vi/less screen insect Date: Wed, 14 Apr 1993 18:13:13 GMT Reply-To: thoth@uiuc.edu (Ben Cox) oreillym@tartarus.uwa.edu.au (Michael O'Reilly) writes: >As yet, I haven't managed to find a patch for bash that will fix the >problem. anyone??? It's called tcsh... :-) -- Ben Cox thoth@uiuc.edu ------------------------------ From: tmcwill@ukelele.GCR.COM (Thomas McWilliams) Subject: clone(), threads, and POSIX Date: Wed, 14 Apr 1993 14:12:37 GMT Linus writes regarding clone(): > Over to a different subject: > > The diffs also introduce the 'clone()' system call that I have been > thinking about for some time, but this one is only useful for testing > purposes until there is more user-level support for it. It should be > possible to implement asynchronous IO using this using the 'clone()' > system call to create new threads to do the IO etc. If anybody is > interested in creating a threads library for this, feel free.. > > The semantics of 'clone()' are the same as for 'fork()' - the system > call creates a new process that shares the same process space as the old > one. 'clone()' gets one parameter: top of stack for the new process. > Note that upon return from the 'clone()' call, you should be very > careful not to mess around with the same memory in both processes: it > needs some assembly language wrappers etc. If you are serious about > doing a thread library using this, mail me for details. > > Linus > Anyone interested in implementing a threads library might want to have a look at POSIX 1003.4 Draft 9, which include asynchronous i/o and application threads (aread(), aio(), etc.). (no, I don't have it, and no, I'm not working on a threads library) Thomas -- tmcwill@gcr.com ------------------------------ From: zevans@nyx.cs.du.edu (Zack Evans) Subject: Re: What are the .z files? Date: Wed, 14 Apr 93 18:31:06 GMT In article <1993Apr13.142458.15411@waikato.ac.nz> hamish@waikato.ac.nz writes: > >Hey. What format are the .z files on nic.funet.fi, and tsx-11? They aren't >compressed, and I can't find anything that will undo them. Any hints? Yes, they are gzipped...GNU zip - check archie for availability. If Richard Stallman ever made a mistake it was starting to use gzip IMHO, what was wrong with .tar.Z? Zack -- Zack Evans - Rave Child | zevans@nyx.cs.du.edu or | pyc081@uk.ac.lancs.cent1 if it's Watch yer bass bins lads, I'm tellin' yer | vaguely important ------------------------------ From: jim@n5ial.mythical.com (Jim Graham) Subject: Re: Mailing list for Newbies? Date: Tue, 13 Apr 1993 14:06:20 GMT In article system@micromed.com (Duane Davis) writes: >Every 'Newbie' I have talked to has expressed the same feeling as I. >Comp.os.linux is not a very friendly place for the person new to Linux. well, here's one person (who was recently new to Linux...not to UNIX, but new to Linux) who disagrees with you. I've gotten nothing but good help from this group, and all from people who were quite ready to help however they could. this trend has continued even up to this morning---I have more than a few e-mail messages that came in during last night's UUCP poll regarding some stuff I'd asked about recently. >Most of this is due to the volume of messages in the newsgroup. sure, it's a high-volume newsgroup. but that's due to the fact that there is an awful lot of stuff going on here. there's nothing that says you have to read every single article, though. :-) >In an effort to help these people I am willing to host a 'Linux newbie >mailing list' where there aren't any 'what not to post here' or 'read this >before posting' messages to intimidate you. first off, ignore the ``what not to post here'' post that appeared a while back. it appeared to be either a badly-timed April fool's joke or someone on a power trip. it certainly wasn't intended to be taken seriously. as for the ``READ THIS BEFORE POSTING'' post, I've talked (via e-mail, of course) to the guy who posts this, and it is definitely not intended to intimidate anyone (I'd originally made this mistake, too, btw, when it was worded a bit differently). it's intended to give you pointers to where you can find certain info faster and easier than you would if you posted here. if you haven't looked at it recently, take a look---he's revised the wording now to where it doesn't lend itself so easily to this type of mis-interpretation. >Do you think this is a good idea? a bad idea? Let me know. If I get >enough positive responses I will set it up with a weekly auto-poster >telling c.o.l readers how to find it. personally, I think it's a bad idea. c.o.l isn't hostile toward new Linux users....if it was, there wouldn't be any new Linux users. like I said earlier, I've gotten a lot of great help here. also, if all the new folk leave here, they're in a pretty good position to miss out on a lot of stuff they might be very interested in, if they knew about it. I'd originally heard that c.o.l was hostile toward new folk, but this has turned out to be nothing but pure BS...quite the opposite is, in fact, the case. if anything, people just need to pay attention to how they post their question. I see lots of subject lines with things like ``I NEED HELP'' or ``PLEASE READ THIS'' that, of course, nobody ever reads. then, I see people who, instead of calmly describing a problem and asking for help, go straight to being hostile and say this or that sucks because they couldn't install it....this, of course, is going to make a lot of people get hostile with them (they asked for it, they got it). --jim -- #include 73 DE N5IAL (/4) ============================================================================== INTERNET: jim@n5ial.mythical.com | j.graham@ieee.org ICBM: 30.23N 86.32W AMATEUR RADIO: n5ial@w4zbb (Ft. Walton Beach, FL) AMTOR SELCAL: NIAL ============================================================================== E-mail me for information about KAMterm (host mode for Kantronics TNCs). ------------------------------ From: strat@matt.ksu.ksu.edu (Steve Davis) Subject: Linux on Macintrashes? (Was: Re: Linux on Amiga?) Date: 14 Apr 1993 12:24:55 -0500 nmspillers@ualr.edu writes: [excerpt from a GNU vs. Apple thread deleted] >The point is, there are a lot of folks out there with Macs. They are not >encouraging facist computing, nor are they a member of Apple's "Evil Empire" >and wander around with a Borg-like mentality--"My user interface will >assimilate you, resistance is futile". They merely bought a nifty, well >built machine that fills their needs, why punish them? Shoot, most of 'em >don't even *know* about the FSF or GNU or any boycotts! If they don't know about FSF or GNU it is because they *have* been 'assimilated' by the Macintrash empire. If they were dumb enough to buy an over-priced-one-mouse-button-tiny-black-and-white-screen- bloated-awful-looking-os-without-a-cli-and-built-by-fascists computer in the first place, there's really no reason for them to want or need UNIX. After all, purchasing a Macintrash is acceptance of its point and drool interface, AND a token of acceptance for Apple's business practices-- both of which go against the free software/UNIX tradition. >I want to learn, I want to do something worthwhile and give a little back to >the computing community that has helped me so much, and I think this is a good >way of achieving these aims. . . Not at all. The best way would be to chuck that Mac out the window and buy a real computer. Stratocaster -- Steve Davis (I'm a student, not a spokesperson!) strat@cis.ksu.edu - Kansas State University - Manhattan KS I assume that as a beginner you'll just want to blow people up. -- Red Shirt ------------------------------ From: neonakis@alkistis.csi.forth.gr (E.A.Neonakis) Subject: Re: PS/2 style mouse not supported?? Date: 14 Apr 1993 18:09:19 +0300 bkunze@novell.de (Bkunze) writes: >Finally I got myself an AST 486-66DX/2 machine with a PS/2 style mouse. >Linux works just great on this one except for the mouse with X. The mouse >cursor stays at the very bottom of the graphics screen and any mouse >movement is converted to horizontal moves only. I use a REAL Microsoft Mouse. >On serial port, all OK. Any comments? >Thanks, >Bernd I have a similar problem, only the cursor stays on the top of the screen. If somebody has encountered such behaviour, please share any solution found with us. Thanks in advance! E.A.Neonakis ------------------------------ From: zevans@nyx.cs.du.edu (Zack Evans) Subject: Re: What are the .z files? Date: Wed, 14 Apr 93 20:25:08 GMT In article <1993Apr14.183106.19013@mnemosyne.cs.du.edu> zevans@nyx.cs.du.edu (Zack Evans) writes: >Yes, they are gzipped...GNU zip - check archie for availability. If >Richard Stallman ever made a mistake it was starting to use gzip IMHO, >what was wrong with .tar.Z? Well I have _already had three emails about this so I'll reply to my own post to save /usr filling :-> The people who emailed me pointed out that gzip is a darn site more effective at compression than compress is. I can see the importance of this when Gnu rely so much on the net for distribution. What I meant was using a 'proprietory' (although that term doesn't apply to copylefted stuff very well) compression system is dangerously like the sort of thing that Microsoft would do IMHO. At the time of posting I didn't realise that gzip was a better program anyway, so I shall retract my earlier statement. Zack -- Zack Evans - Rave Child | zevans@nyx.cs.du.edu or | pyc081@uk.ac.lancs.cent1 if it's Watch yer bass bins lads, I'm tellin' yer | vaguely important ------------------------------ ** 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-Activists-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux) via: Internet: Linux-Activists@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 tupac-amaru.informatik.rwth-aachen.de pub/msdos/replace The current version of Linux is 0.99pl6 released on February 23, 1993 End of Linux-Activists Digest ****************************** aging facist computing, nor are they a member of Apple's "Evil Empire" >and wander around with a Borg-like mentality--"My user interface will >assimilate you, resistance is futile". They merely bought a nifty, well >built machine that fills their needs, whydigest838 644 36676 74430 51115 5366137073 5765 From: Digestifier To: Linux-Activists@senator-bedfellow.mit.edu Reply-To: Linux-Activists@senator-bedfellow.mit.edu Date: Wed, 14 Apr 93 20:45:19 EDT Subject: Linux-Activists Digest #838 Linux-Activists Digest #838, Volume #4 Wed, 14 Apr 93 20:45:19 EDT Contents: Re: Kernel crash: Need help fixing my Ext2fs (Tomasz Retelewski) Re: Access control lists and Linux (Stephen Tweedie) Re: A few question, mostly about pl8 (Stephen Tweedie) Re: New feature for the filesystems. What do you think ? (John McDermott) Re: Linux Mascott/Logo (Doug Beattie) Re: fsck on mounted file systems? (Stephen Tweedie) Re: g++ problems (Please help) (Stephen Tweedie) GIF Viewer for Linux? (Sean Gum) Is there a .au player for soundcards? (Damien Neil) Re: Compiling XV; general porting problems (Jonathan Magid) [Q] Which SCSI driver in A1 bootdisk? (Terry Evans) Re: Kernel crash: Need help fixing my Ext2fs (John Hood) Re: flex vs. lex (David Fox) Re: efsck doesn't like virtual memory? (Stephen Tweedie) ---------------------------------------------------------------------------- From: tretelew@ic.sunysb.edu (Tomasz Retelewski) Subject: Re: Kernel crash: Need help fixing my Ext2fs Date: 14 Apr 93 18:36:21 GMT In article <1qfgck$n9o@agate.berkeley.edu> umminger@purina.berkeley.edu (Frederick W. Umminger) writes: > During "make install" for taylor-uucp-gamma1.04 >I had my first kernel crash. The message was Wow! You are lucky! I am getting this error constantly! It is nightmare. I can get a stable system at all. (0.99pl6,7,7A,8...all of them!) >Unable to handle kernel paging request at address C0840000 >Oops: 0000 >EIP: 0008 : D0034090D >EFLAGS: 0001010202 >fs: 0017 >base: 00000000, limit: C0000000 >Pid: 3122, process nr:11 >8b449f0489d1d3e00644 Exactly the same problem. I can just press backspace for longer than a few seconds and...its crashed! Maybe it is my Cyrix 387? I have 386/33 with 64 cache, ESDI 314MB, 8MB 70ns memory + 10megs swap. Is it a kernel bug???? Tom. tomi@ams.sunysb.edu tretelew@ic.sunysb.edu ------------------------------ From: sct@dcs.ed.ac.uk (Stephen Tweedie) Subject: Re: Access control lists and Linux Date: 14 Apr 93 22:15:03 GMT In article <1rj7ug$89r@walt.ee.pdx.edu>, gary@acacia (Gary Moyer) writes: > steve@rama.demon.co.uk (Steve Entwistle) writes: > : One possibility would be to develop a generalised security > : package, such as RACF, used on IBM Mainframe systems. This system uses a > : central database in which all the security information for various > : resources is stored, e.g. Files, Users, Terminals etc. > : Rather than changing the filesystem code, you could insert a > : call to the resource checking routine in all the system calls that > : access the resource you want to protect (in this case, all the > : system calls concerned with file accesses). If there is no entry > : in the database for a particular file, I guess you would then just > : use the normal file permission bits. > : > : The advantages of doing it this way is that not only is it > : filesystem independent, but it is also readily extended to > : protecting other resources. Hmm - the biggest problem I can see here is that filesystems can be pretty dynamic things. For instance, how would a centralised security server cope with maintaining consistency across remote (NFS) filesystems, and how would it cope with dynamicly mounted filesystems or filesystems with removeable media? It may be a good idea to have a security service daemon, for example, but this would have to be done in such a way that the security database could be distributed over the various filesystems so that the security information for any filesystem is kept on that filesystem. > Thoughts? > I've been digging into the kernel as of late and noticed that the author of > the ext2 filesystem already has some ACL structure embedded in the inode. Indeed - the ext2fs has inodes reserved to hold an ACL database, and has reserved space within each inode for file and directory indexes into the ACL database. However, there is as yet not even any proposal as to the precise form these will take - they are currently just placeholders, there to ensure that an ACL extension in the future will be compatibile with existing ext2fs filesystems. Cheers, Stephen Tweedie. --- Stephen Tweedie (Internet: ) Department of Computer Science, Edinburgh University, Scotland. ------------------------------ From: sct@dcs.ed.ac.uk (Stephen Tweedie) Subject: Re: A few question, mostly about pl8 Date: 14 Apr 93 22:26:55 GMT > I would appreciate any assistance anyone could offer with the > following problems with Linux (0.99pl8 Initially installed by SLS, then > I updated the kernel): > 3) Now that I have upgraded my kernel to pl8, I assume that there > are (or will be) updates to some of the affected bins. Most of mine are > probably quite old. Any idea what ftp sites I should check and which > commands need to be updated? If your commands still work then there is probably little need to upgrade. Binary compatibility is as much a function of the system shared libraries as it is of the kernel - kernel upgrades go to great lengths not to break existing code, but rather to fix bugs and to provide extra functionality. You can use the "ldd" command (part of the Linux binary utilities in the GCC directory on tsx-11.mit.edu) to find out which libraries an application depends on. The current main library is now libc.so.4.3.3, available in the same GCC directory; by following the instructions in the release note you should be able to upgrade from any libc.so.4.x without changing your binaries. You might however want to replace any programs that depend on libc older than libc.so.4.0, or on X libraries older than libX11.so.3.0. Between them, sunsite.unc.edu and tsx-11.mit.edu hold pretty much all of the significant Linux binaries and source. > 4) Does Linux support memory, disk space, or number of child > process limitations on a user? Is ulimit supported by Linux? How is it > used? Only partially. There is no support for quotas (although there are patches around which add quota support to older kernels). ulimit is partially supported; currently, only the core-size limit is respected, and (for ext2fs filesystems only) the file-size limit. The easiest way to use these is to use the builtin ulimit command of the bash shell. "help ulimit" will help you get started. > 5) Every once in a while I accidentally cause non-printable text > to be sent to my console. Sometimes this results in all future text from > Linux coming out garbled. Setterm doesn't seem to help, nor does logging out > and then back in. It seems that that virtual console is basically worthless > until I reboot. Any fix? Try "echo ^Q^O". You will need to type control-Q twice to get the ^Q character, and use control-Q control-O to get ^O. (This is again assuming you are using bash.) Cheers, Stephen Tweedie. --- Stephen Tweedie (Internet: ) Department of Computer Science, Edinburgh University, Scotland. ------------------------------ From: mcdermot@daedalus.cs.unm.edu (John McDermott) Subject: Re: New feature for the filesystems. What do you think ? Date: Wed, 14 Apr 1993 20:05:33 GMT >Sort of like the Plan 9 interface for file (or anything else) servers. > >It would be neat to have something like the bind() call from Plan 9 >that would let you attach a process to a particular file. All >filesystem calls to that file would be get directed to that process. >-- >rajat (rajat@watson.ibm.com) This seems to be most reasonable. The technique (of bound programs) was used in some old (CDC's NOS)? opsys developed before the mid 70's as we studied the mechanism in school. The concept is pretty clean and lets the user decide whether to incur additional overhead for a given file or set of files. Compressing disk controllers would be the best bet as the main CPU is not used for the work, but compressing blocks in the system could be quite time consuming, expecially for one using, say, X11... The binding mechanism is also quite general and would allow many more features such as passwords for individual files, locking, etc. This could be especially valuable in an environment where filesystems could be created on arbitrary sets of disk blocks (files) because then one could implement the compress-each-block-of-the-fs type of fs also. --john -- John McDermott 505/897-2064 H/W mcdermot@pele.cs.unm.edu (The facts/opinions above are MINE, MINE, MINE) [ UNM only lets me use this account because they are nice. I have no relationship with them whatsoever (besides being friends).] ------------------------------ From: dbb@unislc.slc.unisys.com (Doug Beattie) Subject: Re: Linux Mascott/Logo Date: 14 Apr 93 18:23:15 GMT Reply-To: dbb@unislc.UUCP (Doug Beattie,D1V03,6814) In article <1qd223$a6n@usenet.INS.CWRU.Edu> wb8foz@skybridge.scl.cwru.edu (David Lesher) writes: >Others said: ># | Why not try something distinctively *Finnish* for the Mascott/Logo? ># | I mean, that's where it all *started* right? > >Shucks... >I was just going to suggest the moose from "Monty Python & the >Search for the Holy Grail" but never mind.... > >Wait - what about the Trojan Cow, or maybe the Killer Rabbit? How about the EverReady Battery rabbit? It just keeps on going and going and going ... Just kidding 8-) >Something about the whole deal reminds me of this movie. Maybe >it's all the questions in the Config file...... >-- >A host is a host from coast to coast..wb8foz@skybridge.scl.cwru.edu >& no one will talk to a host that's close............(301) 56-LINUX >Unless the host (that isn't close).........................pob 1433 >is busy, hung or dead....................................20915-1433 -- Doug Beattie - Computer Development Services Inc. (UUCP) sun!unislc!dbb c/o Unisys Corporation dbb@unislc.slc.unisys.com 322 No. 2200 West M.S. D1V03 Subcontractor 88K MP NFS Group - SLC, UT SLC, UT 84116 (801) 594-6814 Disclaimer - I represent myself and no one else ------------------------------ From: sct@dcs.ed.ac.uk (Stephen Tweedie) Subject: Re: fsck on mounted file systems? Date: 14 Apr 93 22:39:20 GMT In article <10974@lhdsy1.lahabra.chevron.com>, hwrvo@kato.lahabra.chevron.com (W.R. Volz) writes: > I noticed that someone said that fsck should be run on > unmounted file systems. Then how does run it on the > filesystem that contains fsck? Is this a studip question > or what? Inquiring minds want to know. As long as there is no other disk activity going on, it is safe to run fsck on a mounted partition just to check the correctness of the partition. However, if you are using the -r or -a options to repair a filesystem, then there are basically two safe ways to go about it: * Boot from a different filesystem (eg. from HJ's excellent bootable floppy rootdisk, from the GCC directory on tsx-11.mit.edu) and use the fsck there to check your primary root partition; or * Using the latest kernel, elect to mount the root filesystem readonly; upon boot, run fsck on the partition, and once that has finished, remount root for writing and mount the other partitions. The kernel code to support this is already present, and a version of mount(8) which supports the necesary remounting of the root filesystem should hopefully be available soon. Cheers, Stephen Tweedie. --- Stephen Tweedie (Internet: ) Department of Computer Science, Edinburgh University, Scotland. ------------------------------ From: sct@dcs.ed.ac.uk (Stephen Tweedie) Subject: Re: g++ problems (Please help) Date: 14 Apr 93 22:46:03 GMT > Hello eveybody .. > I have Linux 0.99.6 on my system for some time now, and I only > encountered one problem with it. The g++ compiler refuses to work > for me. I used SLS v1.0 with the old runtime libraries and the gcc ^^^^^^^^^^^^^^^^^^^^^ > is version 2.3.3. What happens is that although it compiles normaly, > when it comes to linking the files together, it comes up with a set > of errors that say that _cout and ostream are not defined. Try upgrading your shared libraries to libc.so.4.3.2 or later - version 4.3.3 is the most recent. Read the release notes carefully before installing, and you should have few problems. The libraries are available in the /pub/linux/GCC directory on tsx-11.mit.edu. I use g++-2.3.3 with libc.so.4.3.2 pretty extensively, and have had no problems with it at all. > I desperately need some help since I have to write my thesis in c++, > and I would like to be able to do this on my system and not on the > university's systems. I love it. I can swap code between the University's sun-4 system and my Linux box at home, and the only difference I notice is that the Linux box goes faster. :-) Persevere - I can assure you that it can work, and that it's worth it! Cheers, Stephen Tweedie. --- Stephen Tweedie (Internet: ) Department of Computer Science, Edinburgh University, Scotland. ------------------------------ From: sean@whiting.mcs.com (Sean Gum) Subject: GIF Viewer for Linux? Date: Wed, 14 Apr 1993 16:47:13 GMT Anybody have a GIF viewer for Linux? Preferably in binary. (I hate compiling!) If so, please E-Mail me one. (Or lemme know where I can get it. I don't have FTP access, only dial-up.) Thanks! ------------------------------ From: damien@b63519 (Damien Neil) Subject: Is there a .au player for soundcards? Date: 14 Apr 93 19:45:42 GMT Reply-To: dpn2@po.cwru.edu Rick Miller - Linux Device Registrar (rick@ee.uwm.edu) wrote: : It plays Sun-style audio files ("*.au") through the internal speaker. : No sound-card is needed, and it won't use it even if you've got one. This reminds me -- is there an .au player that _will_ use a soundcard? I have splay from the snd-util package, but the sound it produces when playing an .au file makes it seem like I'm getting a poor radio transmission...from Venus. Not the program's fault of course -- it wasn't designed to use the .au format. But it would be nice to have access to all the Sun audio files out there. Thanks! -- Damien Neil dpn2@po.cwru.edu "Until somebody debugs reality, the best Case Western Reserve University I can do is a quick patch here and there." CMPS/EEAP Linux -- the choice of a GNU generation. - Erik Green ------------------------------ From: jem@sunSITE.unc.edu (Jonathan Magid) Subject: Re: Compiling XV; general porting problems Date: Wed, 14 Apr 1993 21:59:30 GMT In article <1993Apr14.200018.21295@aio.jsc.nasa.gov> mancus@sweetpea.jsc.nasa.gov (Keith Mancus) writes: > I'm trying to compile xv 2.21 under Linux 0.99.6. gcc can make the >libjpeg.a just fine, but it runs out of memory and dumps when I attempt >to make libtiff.a. Since I have 12M RAM and a 10M swap partition, this >must either be a gcc bug or bad options. This is a well known bug with gcc, not just the Linux version. Gcc will take outrageous amounts of memory when compiling large static initialized arrays. Fear not! there are people with such ridiculous amounts of memory and swap space; Rob Hooft is one of them. You can find his jump-tabled compilations of lib in /pub/Linux/apps/graphics/libgr-1.0.tar.z on sunsite.unc.edu. > I was able to get tcsh to compile by using -DVMS_POSIX, but xv, >while supplying a variety of Makefiles, does not include one for POSIX >or Linux. I had very little trouble compiling this same program on the >SGI's at work. Could someone point me in the right general direction? There is ported source for xv in /pub/Linux/X11/xapps/graphics/xv221.tar.z also on sunsite. jem. ------------------------------ From: tevans%sunset.cs.utah.edu@cs.utah.edu (Terry Evans) Subject: [Q] Which SCSI driver in A1 bootdisk? Date: 14 Apr 93 19:33:02 GMT When I use the a1 bootdisk to do my install it recognizes my SCSI driver just fine and let's me install the software. The problem is when it is all done installing and writes out the image to disk to use as a bootdisk. When I use that disk now it comes up and then corectly identifies the scsi disk but then soon after it errors with something about superblock reading and then says that it is unable to mount the root partition. I presume that since I can install the packages that there must be some way to configure the kernel to recognize and work with my driver. Any help will be greatly appreciated! Terry Evans tevans@cs.utah.edu ------------------------------ From: jhood@smoke.marlboro.vt.us (John Hood) Subject: Re: Kernel crash: Need help fixing my Ext2fs Date: Wed, 14 Apr 1993 21:05:30 GMT In article <1qgakk$15f@agate.berkeley.edu> umminger@purina.berkeley.edu (Frederick W. Umminger) writes: > > I should mention this also, now that I've noticed it; my >ext2fs apparently ran out of space, at least for normal users, >just before the crash. Taylor uucp is taking up 17 megs of disk >space! How absurd. Has ext2fs been tested extensively for its >behaviour when it runs out of space? Taylor UUCP is taking up 17 megs of disk space (yes, how absurd) because the default compile flags include debugging information, and apparently there is lots of it in Taylor uucp binaries. Strip the binaries and uucico will go from two-odd MB to 200k. --jh -- John Hood Cthulhu-- just imagine it! jhood@smoke.marlboro.vt.us ------------------------------ From: fox@graphics.cs.nyu.edu (David Fox) Subject: Re: flex vs. lex Date: Wed, 14 Apr 1993 23:07:48 GMT In article <1qhm2eINN9nc@lynx.unm.edu> firehawk@centauri.unm.edu (-* FireHawk *-) writes: In article <1993Apr14.141943.9799@aixrs0.hrz.uni-essen.de> hank@Blimp.automat.uni-essen.de writes: : :I didn't find any direct info on this question, so I "dare" posting :it: quite often when I try to compile software on linux there are :files for lex or yacc. Usually I think it's a good thing to have :scanners and parsers generated automagically, but I seem never to be :able to use files written for lex with flex. (Yes, I DO change the :library to -lfl!) There are always quite some undefined symbols. I encounter problems going from lex -> flex on just about any system that has it. One of the main incompatabilites I have is the yyless() function. If someone has a decent lex package for linux, would they please post where to find it? If not, is there a flex equivelent for yyless()? You won't find a more "decent" lex package than flex. yyless is a macro in the flex output. How this leads to a solution of the problem of undefined yyless symbol is left as an exercise to the reader. ------------------------------ From: sct@dcs.ed.ac.uk (Stephen Tweedie) Subject: Re: efsck doesn't like virtual memory? Date: 14 Apr 93 22:58:46 GMT In article <1993Apr12.225831.4540@cheops.qld.tne.oz.au>, pclink@cheops.qld.tne.oz.au (Rick) writes: > ... After upgrading to 99p7, I've found that efsck on the 300M > partition gives hundreds, nay thousands, of errors. Needless to > say, when this first happened I was mildly disconcerted (read: > horrified). Rebooting with 99p4, efsck reported no errors. After > experimenting, it seems that when efsck starts to make significant > use of swap, everything goes to hell. That's pretty serious. I can't think of anything major which has changed between pl4 and pl7 which could cause this, except for the SCSI drivers. I assume your partitions live on SCSI disks? I think that the 0.99pl8 release cures the problems known to exist in the pl7 SCSI code; you might want to see if this helps. It may well do. > Anybody else noticed this? Should I just switch to ext2? Ext2 ought to give much better performance, but the extfs is probably not responsible for your problems. Try out 0.99pl8 first. Cheers, Stephen Tweedie. --- Stephen Tweedie (Internet: ) Department of Computer Science, Edinburgh University, Scotland. ------------------------------ ** 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-Activists-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux) via: Internet: Linux-Activists@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 tupac-amaru.informatik.rwth-aachen.de pub/msdos/replace The current version of Linux is 0.99pl6 released on February 23, 1993 End of Linux-Activists Digest ****************************** ed help fixing my Ext2fs Date: Wed, 14 Apr 1993 21:05:30 GMT In article <1qgakk$15f@agate.berkeley.edu> umminger@purina.berkeley.edu (Frederick W. Umminger) writes: > > I should mention this also, now that I've noticed it; my >ext2fs apparently ran out of space, at least for normal users, >just before the crash. Taylor uucp is taking up 17 megs of disk >space! How absurd. Has ext2fs been tested extensively for its >behaviodigest839 644 36676 74430 55273 5366137074 6000 From: Digestifier To: Linux-Activists@senator-bedfellow.mit.edu Reply-To: Linux-Activists@senator-bedfellow.mit.edu Date: Wed, 14 Apr 93 21:45:38 EDT Subject: Linux-Activists Digest #839 Linux-Activists Digest #839, Volume #4 Wed, 14 Apr 93 21:45:38 EDT Contents: Re: GIF Viewer for Linux? (Leon Dent) Re: Ext-2 FS bug (Stephen Tweedie) Re: X hangs.. (on occasions) (Stephen Tweedie) Re: Help on X-Windows(mostly) (Stephen Tweedie) Re: Linux Mascott/Logo (Kelly Murray) Re: A modified LESS to accept arrow keys (root-bound) Re: Compiling XV; general porting problems (Clint Jeffery) RE: Linux on Macintrashes? (Was: Re: Linux on the Amiga?) (nmspillers@ualr.edu) Re: Disk Quotas (was Re: New feature for the filesystems.) (william E Davidsen) Re: Networking between two linux machine without Ethernet Card (william E Davidsen) Compiling XV; general porting problems (Keith Mancus) Re: flex vs. lex (-* FireHawk *-) Re: weird vi/less screen insect (pete cervasio) Re: A modified LESS to accept arrow keys (pete cervasio) ---------------------------------------------------------------------------- From: lcd@umcc.umcc.umich.edu (Leon Dent) Subject: Re: GIF Viewer for Linux? Date: 14 Apr 1993 18:36:29 -0400 For plain vga (non-X) use dpg-view. You will also need vgalib12.tar.Z For X use xv. An xv binary exists on sunsite.unc.edu, however it needs a rather large set of shared libs to run. Note to sunsite admins: There was also a binary version of xv that did not require these libs. It went by the name xv-jmp.Z (I think). It would be nice if it could be restored. Leon Dent lcd@umcc.umich.edu ------------------------------ From: sct@dcs.ed.ac.uk (Stephen Tweedie) Subject: Re: Ext-2 FS bug Date: 14 Apr 93 23:08:02 GMT keithd@versyss.com (Keith Doyle) writes: > Hi, > I just encountered what seems to be a serious bug in > the ext-2 filesystem. > [trying to perform rename on many ext2fs files... ] > It got about halfway through, a couple of errors > where the filenames came out to be the same (moving > a file to its same name) and then I got a series of > "file not found" errors. > After that, I found the filesystem to be seriously > screwed up. Thanks very much for the bug report. I have twice noticed an unexpected "file not found" error under ext2fs, but this was never associated with any filesystem corruption. Being unable to reproduce this problem, it has been extremely hard to trace. This is the second report which looks as if it might be able to reproduce the bug - if I can duplicate the effect then I'll hopefully be able to get a bug fix soon. If you any problems with the ext2fs which can be associated with any particular behaviour (such as rename here), or which can be readily reproduced, *please* let Remy Card and myself know - we are trying hard to make ext2fs the best filesystem we can, and your feedback is greatly appreciated. Cheers, Stephen Tweedie. --- Stephen Tweedie (Internet: ) Department of Computer Science, Edinburgh University, Scotland. ------------------------------ From: sct@dcs.ed.ac.uk (Stephen Tweedie) Subject: Re: X hangs.. (on occasions) Date: 14 Apr 93 23:10:06 GMT In article <28580@galaxy.ucr.edu>, atristan@galaxy.ucr.edu (Andrew Tristan) writes: > In article <1993Apr7.181918.1@cc.newcastle.edu.au> c9108932@cc.newcastle.edu.au writes: >>My problem is this - X tends to hang on me qith reular monotony. > I have also experienced this problem. Although, I can't even kill the > server with ctrl-alt-bksp. Which kernel are you using? There was a known keyboard problem in some of the 0.99pl7 releases which caused this type of effect; the problem is cured in 0.99pl8. Cheers, Stephen Tweedie. --- Stephen Tweedie (Internet: ) Department of Computer Science, Edinburgh University, Scotland. ------------------------------ From: sct@dcs.ed.ac.uk (Stephen Tweedie) Subject: Re: Help on X-Windows(mostly) Date: 14 Apr 93 23:19:13 GMT In article <28617@galaxy.ucr.edu>, jmward@ucrengr (jonathan ward) writes: > ... I finally got X up and running(it defaults to the open windows > manager) but I have this problem: The mouse, once started is > generally unresponsive. > I only have 4mb of RAM, and I'm swapping to a 12 mb swapfile. Is > the mouse problems due to the lack of ram ... ? It certainly is. 8MB is the general recomended minimum ram for running X11 on Linux, because you are guaranteed to get a lot of swapping with any less than this. This is not anything specific to Linux - the X server is simply _very_ big, so you will have the same effect on any machine running X. If anything, the problem is less severe with Linux because of the quality of its memory management. However, there is really no substitute for more memory here. > Also, I'm using the e2fs for my main partition. What is the proper > method to run e2fsck? You can run e2fsck with no options just to check your filesystem even if it is mounted, as long as there are no applications writing to the filesystem at the time. If you need to do any repairs, the best way is currently to boot of a floppy rootdisk and e2fsck from there. There is currently work in progress to allow checking of the root filesystem during boot, by mounting the root readonly during kernel startup and only allowing root to be writeable after the fsck. In the mean time, don't worry too much about the "filesystem not valid" messages you get from e2fs at startup. Just run e2fsck whenever you have a system crash. Cheers, Stephen Tweedie. --- Stephen Tweedie (Internet: ) Department of Computer Science, Edinburgh University, Scotland. ------------------------------ From: kem@cis.ufl.edu (Kelly Murray) Subject: Re: Linux Mascott/Logo Date: 14 Apr 93 19:56:19 GMT A mouse as a Linux Mascott makes sense as a comment on evolution -- scurrying little college students create something that over takes those big outdated DOSasours. But I really like the seagull. The sense of freedom is compelling. It also fits with a ``Fly with Linux'' promotional line. IMHO, I don't think Finland is critical to Linux. LINUS is critical to Linux, and that is appropriately reflected by the name ``Linux'' (and which I think should not be changed) -Kelly ------------------------------ From: aehall@calvin.seattleu.edu (root-bound) Subject: Re: A modified LESS to accept arrow keys Date: 14 Apr 1993 22:17:02 GMT I tried the lesskey program that someone pointed out to me... Wow, what a concept! :) Here's what I did: I have a file called .lessrc that contains the following: ^[[B forw-line # down arrow ^[[A back-line # up arrow ^[[6~ forw-scroll # page down ^[[5~ back-scroll # page up ^[[1~ goto-line # home ^[[4~ goto-end # end (note: the ^[ has to be an ESC i.e. in vi, type "i" for input mode, press CTRL-v, and then press the Escape key. I'm not sure if putting the '^' character and then the '[' character works.) After you have this file, enter: lesskey .lessrc and the file ~/.less will be created with the key sequences. Now less recognizes your new key sequences. -Anthony -- Anthony Hall _ _ Unix System Administrator aehall@seattleu.edu /_/ /_/ Physician Micro Systems, Inc. _ _ 2033 6th Ave Suite 707 /_/ /_/ Seattle, WA 98122 206-441-8490 ------------------------------ From: cjeffery@optima.cs.arizona.edu (Clint Jeffery) Subject: Re: Compiling XV; general porting problems Date: 14 Apr 93 22:45:27 GMT From article <1993Apr14.200018.21295@aio.jsc.nasa.gov>, by mancus@sweetpea.jsc.nasa.gov (Keith Mancus): > I'm trying to compile xv 2.21 under Linux 0.99.6. gcc can make the > libjpeg.a just fine, but it runs out of memory and dumps when I attempt > to make libtiff.a. Since I have 12M RAM and a 10M swap partition, this > must either be a gcc bug or bad options. When I compiled it under 0.99.7A on a 16MB 486/33, the machine appeared to "hang" while compiling files for libtiff.a. Under X, the mouse no longer moved; outside X, I could still switch consoles, but could not type anything at any of them. This has to be more than just gcc or bad options. Had to hard-reset to reboot. Fortunately, I didn't really need tiff support, so I just commented it out of the makefile and went on my merry way... ;-) -- Clint Jeffery - cjeffery@cs.arizona.edu ------------------------------ From: nmspillers@ualr.edu Subject: RE: Linux on Macintrashes? (Was: Re: Linux on the Amiga?) Date: 14 Apr 93 14:25:37 GMT >[excerpt from a GNU vs. Apple thread deleted] > >>The point is, there are a lot of folks out there with Macs. They are not >>encouraging facist computing, nor are they a member of Apple's "Evil Empire" >>and wander around with a Borg-like mentality--"My user interface will >>assimilate you, resistance is futile". They merely bought a nifty, well >>built machine that fills their needs, why punish them? Shoot, most of 'em >>don't even *know* about the FSF or GNU or any boycotts! > >If they don't know about FSF or GNU it is because they *have* been >'assimilated' by the Macintrash empire. If they were dumb enough to >buy an over-priced-one-mouse-button-tiny-black-and-white-screen- >bloated-awful-looking-os-without-a-cli-and-built-by-fascists computer >in the first place, there's really no reason for them to want or need >UNIX. After all, purchasing a Macintrash is acceptance of its point >and drool interface, AND a token of acceptance for Apple's business >practices-- both of which go against the free software/UNIX tradition. Oh gee, an intelligent response. You really have no clue, do you? If all you want to do is flame, then take it to e-mail like an adult. >>I want to learn, I want to do something worthwhile and give a little back to >>the computing community that has helped me so much, and I think this is a good >>way of achieving these aims. . . > >Not at all. The best way would be to chuck that Mac out the window >and buy a real computer. I work on all types of machines professionally, bub, and I consider every last one of 'em a "real" computer. And yeah, I consider learning about the low level aspects of the Mac to be a Good Thing. You must have been hangin' out in comp.sys.amiga.advocacy. Get back where you belong and take your silly partisan attitudes with you. Such a precocious child. . . Apologies for upping the noise/signal ratio and all followups to *this* thread to e-mail. . . >Stratocaster ^^^^^^^^^^^^^ Sheesh, shoulda known, a c00l handle and *everything*! nate ------------------------------ From: davidsen@ariel.crd.GE.COM (william E Davidsen) Subject: Re: Disk Quotas (was Re: New feature for the filesystems.) Reply-To: davidsen@crd.ge.com (bill davidsen) Date: Wed, 14 Apr 1993 19:16:51 GMT In article <1993Apr14.123238.11120@oucsace.cs.ohiou.edu>, sadkins@bigbird.cs.ohiou.edu (Scott W. Adkins) writes: | 1) The filesystems must all be modified to handle quotas... Essentially, | at least the "writing of files" mechanism must be modified to check | for reached quotas before going through with the right. Also, many | popular quota systems also start setting file sizes to 0 when any | file is a) opened b) over quota c) used up time period or warnings. | So, the "opening of files" mechanism may also need to be modified. based on how Sun does it, and a week course on Sun Advanced SysAdmin... I never saw that, and I just spent a half day playing with quotas in a course. When you hit hard quota you stop writing. After you have been over soft quota for a given time (in sec up to weeks) you can't write until you get below soft quota. I never heard of or saw the o/s zap data for you. | | 2) There must be a way for quotas to be maintained. I think BSD's approach | is to include a quota field in the password structure, or something | like that (I am not sure, though). Most systems keep track of a file | called "quotas" which would determine the quotas of every user on the | system. There is a quota file for each filesystem, and it must be mounted with the quota keyword to enable quotas. The quota is then kept "per user" in terms of soft limit, hard limit, and time limit. | | 3) Some library calls may need to be modified to handle quotas. At the | minimum, every library call that has to do with writing of files, | must at least check for quota violations. I figure that a new errno | would be created, something like ERRQUOTA or something like that. | (There is probably a standard for it.) Yes. I don't have a pure BSD system here, so I'm not sure if any of this is Sun speciffic. | | I am sure there are others, but this is good enough for now. I don't think | a quota daemon (i.e. quotad? :-) ) would be good enough, since it is very | hard to detect when a file is going to exceed a quota and stop that write | *before* it happens. I think quotad handles the timer after you go over soft limit, and the kernel does the actual enforcement. It pretty much must be to work over multiple processes. | | 2 concepts of most quota systems: | 1) Soft Quota -> If you exceed this quota, then the writes will succeed, | but you will be given a time period (i.e. 1 week) or a number of | login warnings (i.e. 3 logins) before the quota system will take | action. Taking action usually means that any file that is written | to, and sometimes, any file that is read, will have its filesize | effectively changed to 0. This would continue until the user | removes enough files to go under the quota, or the quota system | changes enough files to reduce the quota for the user... I've never seen space deleted for you, you just can't write until you get down below soft limit. Of course that may be an option, but I've not found it in the docs. The sysadm is usually the dark force, through scripts. | | 2) Hard Quota -> This one is set higher than the soft quota, and is the | absolute highest amount of disk space the user can use at any time. | If the hard quota is reached, whatever write was going on at the | time is halted and the file closed. So, in most cases, the user | will be left with a file that is incomplete, but uses up the rest | of the hard quota. It will need a bunch of kernel support, since it has to be done on a per file owner basis. If I give you write on a file I own the size goes against my quota, not yours. Non-trivial! -- bill davidsen, GE Corp. R&D Center; Box 8; Schenectady NY 12345 ------------------------------ From: davidsen@ariel.crd.GE.COM (william E Davidsen) Subject: Re: Networking between two linux machine without Ethernet Card Reply-To: davidsen@crd.ge.com (bill davidsen) Date: Wed, 14 Apr 1993 19:23:11 GMT In article , jchen@cs.buffalo.edu (jonathan chen) writes: | I now have two machine: one 386-40 120mb HD IDE controller, | the other 486-33 210mb HD Future Domain 885 SISC controller. | I don't have money for the two ethernet card. But I still | want to use the hard disk in the 486. | | Can this be done? How? | Did anyone try something like this? Please help because I am REALLY | running out of space in my 386. If you don't mind *really* piggy performance, use SL/IP. Tie the systems together with a serial line and a null modem, start SL/IP and cross mount or whatever. If you don't have smart chips in your serial cards you have to keep the speed way down, but it *will* (eventually) work. Performance will be even worse than it sounds. -- bill davidsen, GE Corp. R&D Center; Box 8; Schenectady NY 12345 ------------------------------ From: mancus@sweetpea.jsc.nasa.gov (Keith Mancus) Subject: Compiling XV; general porting problems Date: 14 Apr 93 20:00:18 GMT I'm trying to compile xv 2.21 under Linux 0.99.6. gcc can make the libjpeg.a just fine, but it runs out of memory and dumps when I attempt to make libtiff.a. Since I have 12M RAM and a 10M swap partition, this must either be a gcc bug or bad options. I was able to get tcsh to compile by using -DVMS_POSIX, but xv, while supplying a variety of Makefiles, does not include one for POSIX or Linux. I had very little trouble compiling this same program on the SGI's at work. Could someone point me in the right general direction? I'm also having major problems get sat-track to compile. This time, everything compiles but the link generates an enormous list of "Unrecognized symbol _ referenced in text segment". This code has no system-dependent makefiles, just the basic makefile, and it compiles fine on a Sun Sparc. -- | Keith Mancus | | N5WVR | | "Black powder and alcohol, when your states and cities fall, | | when your back's against the wall...." -Leslie Fish | ------------------------------ From: firehawk@centauri.unm.edu (-* FireHawk *-) Subject: Re: flex vs. lex Date: 14 Apr 1993 23:17:28 GMT In article fox@graphics.cs.nyu.edu (David Fox) writes: : [My original question deleted] : :You won't find a more "decent" lex package than flex. yyless is a :macro in the flex output. How this leads to a solution of the :problem of undefined yyless symbol is left as an exercise to the :reader. Well I've read through the flex documentation and I can't really locate an answer for it. I have seen it in the flex.skel file. The question I have is how to define yyless() to the outside world for programs that need it? Programs that require the yyless() function compile and work fine under lex, but under flex, the linker always give me an "undefined symbol" on yyless(). I do link the files with libfl.a and still have no luck. -Jeff -- INTERNET: firehawk@carina.unm.edu ------------------------------ From: pete@q106fm.uucp (pete cervasio) Subject: Re: weird vi/less screen insect Date: Wed, 14 Apr 93 17:04:57 CDT oreillym@tartarus.uwa.edu.au (Michael O'Reilly) writes: > David Lesher (wb8foz@skybridge.SCL.CWRU.Edu) wrote: > > : In any case, it looks as if I end up back in less, as I should. But I > : hit a command, and it ACTS as if I'm still in vi! Parts of the screen [stuff deleted] > This is not specific to less/vi. Try useing elm, and starting emacs, > and then typeing control-G. emacs doesn't exit by elm wakes up! chaos! Aha! So that's what happened to me this morning. I recently changed my .emacs file so F12 is save-buffers-kill-emacs where it used to do that with F10... duh... F10 is now find-file. Anyway, I forgot that (even though I made this nice reference card for my keyboard) and when I tried to cancel (hit C-g C-x C-c) I got chaos just as you said. I don't remember exactly what I did to get out of them both, but I did manage it. I thought that maybe I had miskeyed a C-z when I was reaching for the C-x, but apparently not. > As far as I am aware, this is actually a problem with bash. Bash is a > little silly about intercepting signals in the background, and so it > exits, so less/elm think that vi/emacs has exited, and so it starts > reading the keyboard. Not wonderful. > > As yet, I haven't managed to find a patch for bash that will fix the > problem. anyone??? Hear, hear! I'd hate to have to write an elm.el just to avoid this in the future... :-) BTW: If anyone wants to know what their various function and cursor keys send (easily), just start emacs and type C-q and then the key... it works great and is less filling. Pete C. +----------------------------------------------------------------------+ | Pete Cervasio | pete%q106fm.uucp@wupost.wustl.edu | | I fish, therefore I am. | pete.cervasio@f1.n2250.z1.fidonet.org | +----------------------------------------------------------------------+ | If you think of C as a preprocessor for your assembler, it makes | | just as little sense as before. (me) | +----------------------------------------------------------------------+ ------------------------------ From: pete@q106fm.uucp (pete cervasio) Subject: Re: A modified LESS to accept arrow keys Date: Wed, 14 Apr 93 18:05:42 CDT [sent in email, and posted here in case anyone else needs the info] aehall@calvin.seattleu.edu (root-bound) writes: > I've modified the source of LESS to accept up/down arrows, pgup/pgdn, > and home/end from the PC keyboard. > > If anyone's interested, I can upload it (ftp) somewhere. Why?? Use the lesskey program and the following input file like this: lesskey < lesskey.input lesskey creates a file called .less in your home directory. Try "man lesskey" for info about it. Maybe even "man less". =====8<=====8<=====lesskey.input=====8<=====8<===== ^[[B forw-line ^[[A back-line ^[[6~ forw-screen ^[[5~ back-screen ^[[1~ goto-line ^[[4~ goto-end ^[[D prev-file ^[[C next-file =====8<=====8<=====lesskey.input=====8<=====8<===== Make sure the ^[ is an esc character. You can create one of these files with emacs by typing control-q before typing the key you want. i.e.: forw-line back-line and so forth and so on. It's very easy doing it this way, and people that call in to your system with different terminal types can create their own .less files that work with their terminals. No need to recompile the darn thing. Besides, if you get a newer version of less, it should see your key mappings automagically... :-) Pete C. +----------------------------------------------------------------------+ | Pete Cervasio | pete%q106fm.uucp@wupost.wustl.edu | | I fish, therefore I am. | pete.cervasio@f1.n2250.z1.fidonet.org | +----------------------------------------------------------------------+ | If you think of C as a preprocessor for your assembler, it makes | | just as little sense as before. (me) | +----------------------------------------------------------------------+ ------------------------------ ** 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-Activists-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux) via: Internet: Linux-Activists@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 tupac-amaru.informatik.rwth-aachen.de pub/msdos/replace The current version of Linux is 0.99pl6 released on February 23, 1993 End of Linux-Activists Digest ****************************** reference card for my keyboard) and when I tried to cancel (hit C-g C-x C-c) I got chaos just as you said. I don't remember exactly what I did to get out of them both, but I did manage it. I thought that maybe I had miskeyed a C-z when I was reaching for the C-x, but apparently not. > As far as I am aware, this is actualdigest840 644 36676 74430 53406 5366137074 5764 From: Digestifier To: Linux-Activists@senator-bedfellow.mit.edu Reply-To: Linux-Activists@senator-bedfellow.mit.edu Date: Wed, 14 Apr 93 22:45:28 EDT Subject: Linux-Activists Digest #840 Linux-Activists Digest #840, Volume #4 Wed, 14 Apr 93 22:45:28 EDT Contents: Re: Ext-2 FS bug (Curtis Yarvin) ext2fs compressed filesystem *IS* *ALREADY* under development. (Stephen Tweedie) Re: Kernel crash: Need help fixing my Ext2fs (Stephen Tweedie) Re: The dangers of playing with shared libraries (Frank T Lofaro) S3 card with X11 (John Bergin) Re: The dangers of playing with shared libraries (Junio Hamano) Re: What are the .z files? (Bill C. Riemers) Re: Disk Quotas (was Re: New feature for the filesystems.) (Bill C. Riemers) Re: Announcing Babaya 1.00b (janeK@world.std.com) Memory problem with Linux? (Chris Newton) New to Linux (Subbarao Chitturi) Re: Does _anyone_ use a WD-7000FASST with Linux? (Alex Freed) Re: [Q] Has XS3 merged with Xfree? (william E Davidsen) Re: fast fsck ... (Remy CARD) missing some mgr .h files (Patrick K. Ferrick) Re: What are the .z files? (Jonathan Magid) Re: Disk Quotas (was Re: New feature for the filesystems.) (Frank T Lofaro) ---------------------------------------------------------------------------- From: curtis@cs.berkeley.edu (Curtis Yarvin) Subject: Re: Ext-2 FS bug Date: 14 Apr 1993 23:42:14 GMT In article <1993Apr12.163803.20898@versyss.com> keithd@versyss.UUCP (Keith Doyle) writes: > > >After that, I found the filesystem to be seriously >screwed up. I ran e2fsck, which reported thousands >of "32995 marked used, not in any file" or similar >errors, indicating there were a bunch of blocks >that have been orphaned. Hah! If this is the same bug Carsten Fischer and I reported a little while ago - the symptoms are the same - this is the _third_ filesystem which has manifested the bug (besides minix fs and extfs). I think it's definitely of extraterrestrial origin :-) c ------------------------------ From: sct@dcs.ed.ac.uk (Stephen Tweedie) Subject: ext2fs compressed filesystem *IS* *ALREADY* under development. Date: 15 Apr 93 00:09:43 GMT Well, there certainly seems to be a lot of interest in this, so I though I'd post about work currently in progress on the ext2fs. In outline: I am working on adding the option to transparently compress and uncompress ext2fs files. Compression will be selectable on a file-by-file basis. This will be an enhancement to the existing ext2fs. The new facility will be immediately useable by anybody with an ext2fs filesystem without requiring any reformatting. In article , fl0p+@andrew.cmu.edu (Frank T Lofaro) writes: > NNTP-Posting-Host: andrew.cmu.edu > It sounds like a useful idea for those pressed on diskspace, and seems > worth implementing. Some problems that need to be addressed: > 1. Speed of access. The filesystem data structures and directory tree will remain uncompressed for speed. Uncompression of data should not be too slow; the code is based on gzip which is already reasonably fast at decompression. Uncompressed data will be buffer-cached on the same terms as raw device blocks, which will significantly improve performance. > 2. Demand-paging. It'll probably be really hard to get this to work > with it. Done. :-) I have a set of kernel patches which implement an extension to the standard kernel inode operations, to increase the flexibility of the demand paging software. This will also incidentally allow demand paging and caching of NFS data. This work is already complete. As an example, it is now possible to demand-page binaries and shared libraries from an msdos floppy using sector (512-byte) alignment, even though such a filesystem cannot support the bmap() call previously necessary for demand paging. With this extension, demand paging just degenerates to randomly accessing compressed data. I am storing a form of indirection data within the file, similar to the inode's own direct/indirect block pointers, which will allow random read access to the compressed file. Incidentally, I don't intend (at least at first) to allow random writes to the compressed file, but concurrent random reads will be fully supported. > 3. Backwards compatibility. Anyone out there know if there are any spare > fields or bits (seems like this idea only needs to steal one) in the > inode structure of the current filesystems? There is in the ext2fs, which is what I am basing my code on. > 4. How to have the kernel actually decompress? Exec an outside program? > Build gunzip into the kernel? Yup - internal kernel compression/decompression. > This does seem a lot better than trying to compress the whole FS (not > quite as efficient for space, but probably a hell of a lot easier to > implement). Indeed. It is also safer - you don't run nearly as much danger of corrupting your entire filesystem if it is only individual files which get compressed - more flexible, and faster (the high-level filesystem data remains uncompressed). The other advantage of per-file compression is that the superuser will be able to directly access the raw compressed form of the data. This will allow quick backup/restore of compressed files, for example, and will also allow suid-root programs to do their own compression on files (to get better compression than the kernel's default faster compression), or to repair or check for damaged compressed files. > Anyway, good luck with this idea. It has been mentioned before, but this > time there is some proposed way of handling some of the implementation > details; it seems quite feasible. I don't know enough about filesystems > to know how easy/hard this will be tho... It is definitely possible. All of the necessary design details have (I hope) been addressed, and I hope to get something working over the Easter holidays. Cheers, Stephen Tweedie. --- Stephen Tweedie (Internet: ) Department of Computer Science, Edinburgh University, Scotland. ------------------------------ From: sct@dcs.ed.ac.uk (Stephen Tweedie) Subject: Re: Kernel crash: Need help fixing my Ext2fs Date: 15 Apr 93 00:17:16 GMT In article <1qgvvsINNrb4@serv-200.dfki.uni-kl.de>, malik@dfki.uni-kl.de (Thomas Malik) writes: > When i now changed to ext2 fs, mke2fs didn't correctly allocate my > 32 MB root partition. While there had been ~8MB left on that fs > (output from df and e2fsck), write accesses had 'no space left on > device' as result. There is a possible cause for this which has tripped up other users before now. The recent ext2fs now respects the FSIZE resource limit, as set by the "ulimit" command in bash. Type "ulimit -a" to see your current limits. If you have a filesize ulimit in place, then you can get "no space left" errors even if there is space left on the device. The next release of ext2fs will report "filesize limit exceeded" in this situation rather than "no space left". > Beside that, i did get these 'bad directory entry errors' from time > to time. This was a known bug in e2fsprogs-0.2c. Upgrade to version 0.2d. Cheers, Stephen Tweedie. --- Stephen Tweedie (Internet: ) Department of Computer Science, Edinburgh University, Scotland. ------------------------------ From: Frank T Lofaro Subject: Re: The dangers of playing with shared libraries Date: Wed, 14 Apr 1993 15:58:08 -0400 cat executable > copy does not make copy executable. You'll need cp (or chmod) to make executable copies. ------------------------------ From: jbergin@lookout.it.uswc.uswest.com (John Bergin) Subject: S3 card with X11 Date: Wed, 14 Apr 1993 21:07:48 GMT I am having trouble installing X11 on my computer. I have a CTX 1760 monitor and a G-HOSTS3 video card. Question: Is the S3 card supported by Linux X11? When I startx, I get an error saying that a configured device was found, but the video mode could not be determined. When I comment out the Chipset variable in Xconfig, no device is found. I have tried all Chipset variables and still have had no luck. I presume the S3 card needs its own Chipset variable. Is this true? Any help would be greatly appreciated. Please E-mail me or post on this forum. ****************************************************** * John Bergin - jbergin@lookout.it.uswc.uswest.com * ****************************************************** -- **************************************** *John F. Bergin - jbergin@uswest.com * **************************************** ------------------------------ From: junio@twinsun.com (Junio Hamano) Subject: Re: The dangers of playing with shared libraries Date: Wed, 14 Apr 1993 15:26:20 GMT In article <1qhml1$1i5k@hal.gnu.ai.mit.edu> mycroft@hal.gnu.ai.mit.edu (Charles Hannum) writes: Having to reboot from floppies is bloody annoying. Using tools that are on a potentially corrupt harddisk to fix the system annoys me more. What I'd do is to wait for 30 seconds, reboot with a floppy, run fsck on the harddisk before doing anything, and then fix whatever problem I have with my harddisk. -- | __-+-__ Jun Hamano _-+ (|o|) +-_ junio@twinsun.com ! . ! ------------------------------ From: bcr@bohr.physics.purdue.edu (Bill C. Riemers) Subject: Re: What are the .z files? Date: 14 Apr 93 20:24:42 GMT In article <1993Apr14.183106.19013@mnemosyne.cs.du.edu> zevans@nyx.cs.du.edu (Zack Evans) writes: >Yes, they are gzipped...GNU zip - check archie for availability. If >Richard Stallman ever made a mistake it was starting to use gzip IMHO, >what was wrong with .tar.Z? Why not just .tar? It has been my experience that gzip is as much of a saving of space over compress as compress is over normal files. So if we can justify .Z files we can use the same justification to use .z files instead. What I find interesting is that gzip and gunzip seems just as fast as compress and uncompress. So if I had my way, I would add in an option to use old style compress with gzip and then do away with compress on any system that still uses it! Bill ------------------------------ From: bcr@bohr.physics.purdue.edu (Bill C. Riemers) Subject: Re: Disk Quotas (was Re: New feature for the filesystems.) Date: 14 Apr 93 20:33:00 GMT In article davidsen@crd.ge.com (bill davidsen) writes: > I've never seen space deleted for you, you just can't write until you >get down below soft limit. Of course that may be an option, but I've not >found it in the docs. The sysadm is usually the dark force, through >scripts. Yes I haven't seen a system yet that properly handles hard quotas. Most systems will still let you write to /tmp so you just direct the normal programs to put thier output to there and then take care of them before logging out. For the few systems that I've seen that aren't fooled by placing the file in /tmp, they are fooled if you do a chown on the file before you exceed your quota. Installing a hard quota is by no means a simple task. For the most part, if you are using a file system that reserves space for root, these shouldn't be neccissary. Bill ------------------------------ From: janeK@world.std.com Subject: Re: Announcing Babaya 1.00b Date: 14 Apr 93 23:47:53 GMT In article <1qg4t0$cf0@fitz.TC.Cornell.EDU> alsaggaf@athena.mit.edu (M. Saggaf) writes: >I've put the first release of Babaya on the major linux ftp sites: >tsx-11, sunsite, and nic.funet.fi. Following the convention of Seyon, >the ``b'' in the version name means BETA. The README file follows. > >Enjoy! > >/Muhammad M. Saggaf > alsaggaf@athena.mit.edu > >---------------------------------------------------------------------- > >Q: What is Babaya? > >A: Babaya is a simple session manager for the X Window System. Babaya > takes care of starting the user's custom applications when logging > in or starting the server, and of ending all applications > gracefully when the user desires to log out or exit the server. > >---------------------------------------------------------------------- > Pretty cool. I have been waiting for somehting like this for some time. I now have Babaya run in the foreground as the last command in my ~/.xinitrc (well, pretty much the only one, all others I have moved to Babaya's startup file), and besides shutting X gracefully, I can now also quit gwm and start olvwm whenevr I feel bored with the former (it used to be that quitting gwm shuts the server since it was run last from ~/.xinitrc). I have some questions, though: 1) Can I make Babaya use a different mouse key, or possibly a double-click? My hand is pretty quick with the right mouse button (blame olvwm for that!) and I don't want to logout if I happen to click on Babaya by accident. 2) Can I start Babaya as an icon? 3) Since it's pretty small (9k, linux shared libraries do wonders), is it a good idea to compile it with the gcc -N swicth? (there was a recent thread about that, but I'm afraid I didn't absorb all of it). Thanks. -jane ------------------------------ From: u0xh@jupiter.sun.csd.unb.ca (Chris Newton) Subject: Memory problem with Linux? Date: Wed, 14 Apr 1993 23:39:55 GMT I think there may be a memory problem with Linux. I have a 386, 33Mhz machine with 8Megs, and a 5 meg swap partition. However, when I run about 4 instances of xv (and load a JPG in each one) with other things going on too (like seyon, and an xterm), the machine locks up. The mouse doesn't move, I can't switch to the other VCs or anything. This seems to be the same problem that people are having when they try to compile xv under gcc. It is using up all available memory, and locking. What does Linux do when it runs out of swap space, or memory in general. Whatever it does, it doesn't seem to be doing it correctly. Also, I had a 234K JPG picture that would lock up the machine EVERY time I tried to view it. I never have seen it. JPGs are compressed very highly, so I imagine this was a very very large picture. With nothing running, when trying to view this, it all locks up. Also I have locked the machine up while in X by trying to scroll very quickly through a man page using xman. This is recreatable, at least for me. I run xman, and then push the left mouse button very quickly on the scroll bar, and it will lock the whole machine. The mouse still moves, but is the shape from the scroll bar. Nothing else works. No VC switching or anything. I just thought I would point these out. Chris ------------------------------ From: schit@ctp.com (Subbarao Chitturi) Subject: New to Linux Date: Mon, 12 Apr 1993 13:25:01 GMT Hello Linux Users: I heard about Linux OS only yesterday through a friend. The questions I have about this OS are: 1. Is Linux a true clone of UNIX? 2. Where from can I get a copy of this OS and how much does it cost? 3. What are the Hardware requirements(Hard Disk size)? 4. Do you get the X-Window system along with the OS? 5. Is there a C++ compiler along with the system? I have a 486 DX from Gateway2000 with the ATI video card and a FAX/Modem. Can Linux be installed on this system? All email replies are appreciated. Thank you in advance. ______________________________________________________ Subbarao Chitturi schit@ctp.com ------------------------------ From: freed@europa.orion.adobe.com (Alex Freed) Subject: Re: Does _anyone_ use a WD-7000FASST with Linux? Date: Thu, 15 Apr 1993 00:38:40 GMT Yes, I use a WD-7000FASST with Linux and it works great The ROM is only used for identifying the card, so the rev shouldn't matter. What does matter, is the address the ROM is on. The standard code starts looking for the ROM signature from D8000 or smth like that. If your ROM lives at say c800, you lose. Also the driver is hard-programmed to use only one of (3?) possible interrupts. So you either configure the card accordingly or make a trivial changes to the device driver code and recompile. If you do, the system will boot MUCH faster not looking for hardware that YOU KNOW is not there. -- _______________________________________________________ | -Alex Freed (The opinions expressed are my own. | | However everyone is entitled to them.) | | freed%adobe.com@uucp-gw-1.pa.dec.com | ------------------------------------------------------- ------------------------------ From: davidsen@ariel.crd.GE.COM (william E Davidsen) Subject: Re: [Q] Has XS3 merged with Xfree? Reply-To: davidsen@crd.ge.com (bill davidsen) Date: Wed, 14 Apr 1993 19:27:52 GMT In article <1qhbh5$erf@news.cs.tu-berlin.de>, wong@cs.tu-berlin.de (Wolfgang Jung) writes: | : On a related note - has anybody done/or is planning to try and get | : Xfree to support other accelerators? | | This is on the way. I don't knwo when a public release is available, | but there are several People working on supporting ACCelorated cards. I sure would like to see support for the Swan S3 card and 17inch monitor. Actually I'd like to see support for more than 8 bits, too. I don't expect to see it anytime soon, though, the folks doing the work have their own wish list. -- bill davidsen, GE Corp. R&D Center; Box 8; Schenectady NY 12345 ------------------------------ From: card@masi.ibp.fr (Remy CARD) Subject: Re: fast fsck ... Date: Wed, 14 Apr 1993 18:51:20 GMT In article <191.2BC72D69@purplet.demon.co.uk> jaggy@purplet.demon.co.uk (Mike Jagdis) writes: >* In message , Stephen Tweedie said: > >ST> > Does Linux fsck support checking for stable/clean filesystems? In > >ST> However, if an e2fs filesystem is clean when it is mounted, then the >ST> clean flag _will_ be set again upon successful unmounting of >ST> the filesystem, > >Does it (or could it) also set the clean flag if the filesystem is remounted >read only? Then shutdown etc. could do killall, sync, remount readonly to >lock things down? No, it can't. The remount call is handled by the VFS layer and the ext 2 fs code does not even know that a fs has been remounted readonly. > > Mike > Remy -- Remy Card card@masi.ibp.fr ------------------------------ From: ferrick@acsu.buffalo.edu (Patrick K. Ferrick) Subject: missing some mgr .h files Date: 14 Apr 93 23:38:42 GMT Has anybody else had problems with the following files not being in the MGR distribution (from sunsite) ? default_font.h icon_server.h They seem to be a) important :-), and b) missing! I ftp'd all the .tar.Z files that were there, but I must have missed something...can anybody shed any light on this? Thanks! pat ------------------------------ Crossposted-To: gnu.misc.discuss From: jem@sunSITE.unc.edu (Jonathan Magid) Subject: Re: What are the .z files? Date: Thu, 15 Apr 1993 00:10:53 GMT In article <9820@dirac.physics.purdue.edu> bcr@bohr.physics.purdue.edu (Bill C. Riemers) writes: > > What I find >interesting is that gzip and gunzip seems just as fast as compress and >uncompress. So if I had my way, I would add in an option to use old style >compress with gzip and then do away with compress on any system that still >uses it! The FSF says that the reason for replacing the de-facto standard (compress) with gzip, was the potential legal problems with using a compression algorithm which two seperate companies own patents on (IBM and Unisys, I believe). Gzip's technical superiority is very nice, but apparently was a secondary consideration. Followups to gnu.misc.discuss where this has been treated, ad-nauseum. jem. ------------------------------ From: Frank T Lofaro Subject: Re: Disk Quotas (was Re: New feature for the filesystems.) Date: Wed, 14 Apr 1993 16:23:32 -0400 Excerpts from netnews.comp.os.linux: 14-Apr-93 Disk Quotas (was Re: New fe.. Nick Hilliard@symphony.m (767) Hi, >With all this talk about new features for the various filesystems, I figure >no-one's really mentioned anything about Disk Quotas. >Ok - it's fine if you're a just using Linux as a single user machine, but if >you have a whole load of users hanging out of it, you really _do_ need >some form of mechanism for making sure that people don't hog the >machine. >Is there a chance of seeing code for them being written in the (near) >future? >Nick I agree whole-heartedly, such a feature is desperately needed if Linux is to be more than a single-user OS (ala DOS). We don't want people to switch to 386BSD, now do we? The thing that bothers me, is that someone did write a quota patch, but it is out of date, and won't fit any newer kernels. (without hacking on it). :( (This is *not* a flame of Linux. It is a very good OS, it is much better than DOS, but if it is missing a feature someone critically needs, they simply can't use it, and that's a shame. I am afraid to do any major work, as others may be, to improve Linux, for fear that one's work won't make it into the kernel, and will thus break in a couple of months and have to be redone, abandoned, or updated (or the person will have to install the patch, then all the official patched till one reaches the current patchlevel, and hope it still works/patches cleanly, etc.) ------------------------------ ** 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-Activists-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux) via: Internet: Linux-Activists@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 tupac-amaru.informatik.rwth-aachen.de pub/msdos/replace The current version of Linux is 0.99pl6 released on February 23, 1993 End of Linux-Activists Digest ****************************** if the filesystem is remounted >read only? Then shutdown etc. could do killall, sync, remount readonly to >lock things down? No, it can't. The remount call is handled by the VFS layer and the ext 2 fs code does not even know that a fs has digest841 644 36676 74430 53415 5366137074 5765 From: Digestifier To: Linux-Activists@senator-bedfellow.mit.edu Reply-To: Linux-Activists@senator-bedfellow.mit.edu Date: Thu, 15 Apr 93 01:00:16 EDT Subject: Linux-Activists Digest #841 Linux-Activists Digest #841, Volume #4 Thu, 15 Apr 93 01:00:16 EDT Contents: Re: Linux on Macintrashes? (Was: Re: Linux on Amiga?) (william E Davidsen) ka9q 14? (Richnut) Re: Mailing list for Newbies? (Daniel Supernaw-Issen) Sound Driver question. (Hur InHo) anyone have kterm binary? (Hur InHo) Re: Trying to kill getty so I can dial out on ttys2; help (John Henders) Mounting an extended dos partition (Dov Grobgeld) Re: What are the .z files? (David Fox) An quick-fix to the static binaries issue (Stimpson Joel Katz) hooray! voodoo spell broken. network works now! (John Paul Morrison) 99p8 breaks xdm (Guy Thomas) About disk quotas... (Garrett D'Amore) Re: Long lines in vi (elvis) (Einar Indridason) Re: SYSV IPC:function not implemented (Krishna Balasubramanian) Re: Disk Quotas (was Re: New feature for the filesystems.) (Steve Fuller) ---------------------------------------------------------------------------- From: davidsen@ariel.crd.GE.COM (william E Davidsen) Subject: Re: Linux on Macintrashes? (Was: Re: Linux on Amiga?) Reply-To: davidsen@crd.ge.com (bill davidsen) Date: Wed, 14 Apr 1993 19:52:31 GMT In article <1qhhd7INN7m0@matt.ksu.ksu.edu>, strat@matt.ksu.ksu.edu (Steve Davis) writes: | nmspillers@ualr.edu writes: | If they don't know about FSF or GNU it is because they *have* been | 'assimilated' by the Macintrash empire. If they were dumb enough to | buy an over-priced-one-mouse-button-tiny-black-and-white-screen- | bloated-awful-looking-os-without-a-cli-and-built-by-fascists computer | in the first place, there's really no reason for them to want or need | UNIX. After all, purchasing a Macintrash is acceptance of its point | and drool interface, AND a token of acceptance for Apple's business | practices-- both of which go against the free software/UNIX tradition. When you move from the .edu to the .com domain you may find that people buy hardware to run their software, and don't care who makes it as long as it solves the problem at hand. I have software which only runs on Mac or DOS, and I need it to make money. Or I need it to do what I wanted to do with a computer in the first place. For most people computers are tools, not political statements. To insist on using UNIX for everything, or on avoiding commercial software is a little like using a sports car to haul manure instead of a truck. You may like the user interface, but the job takes longer, and overall as a solution it stinks. -- bill davidsen, GE Corp. R&D Center; Box 8; Schenectady NY 12345 Last year I worried that Bush would die and let Quayle take over. This year I worry that Hillary will die and let Bill take over. ------------------------------ From: ran59961@uxa.cso.uiuc.edu (Richnut) Subject: ka9q 14? Date: Thu, 15 Apr 1993 00:15:52 GMT i saw it on tsx-11 but when i run it it says "cannot load /lib//libc.so.4 Incom patible version" So which lib do i need, i currently have 4.3.3 =========================================================================== Richard A. Nuttle Internet: Richnut@uiuc.edu "I'll do it tomorrow...." NeXT: ran59961@sumter.cso.uiuc.edu "or maybe the next day.." Bitnet: FREE3277@UIUCVMD =========================================================================== ------------------------------ From: daniel@caspian.austin.ibm.com (Daniel Supernaw-Issen) Subject: Re: Mailing list for Newbies? Date: 14 Apr 93 20:06:44 GMT Reply-To: daniel@caspian.austin.ibm.com (Daniel Supernaw-Issen) I thought that I would second Jim Graham's position that the Linux community has been incredibly helpfull and responsive. I've been using Linux for 11 months now. When I started off I knew nothing of Linux and little of UNIX. I knew NOTHING of system administration. Now, generally, I'm able to get things done, do a bit of programming, and in general have fun with the beastie. I periodically post requests for help (for instance the mfm/ide thread) and have gotton an enormous amount of help from more people than I can give sufficent credit to. In general I have found that if I wait a week between discovering a 'problem' and posting a request for help I usually a) figure out the solution myself b) find that someone else has had the same problem and send him/her a request for the solution c) find that what I thought was the original problem wasn't. If after a week I'm still hosed, I post a request for help including what I think is going on. Without exception, I have gotton useful and prompt replies from more people than I can give sufficient credit to. What I really think 'newbies' need more than a special forum which alot of people will ignore, are large pink letters saying: DON'T PANIC. Be patient. Read the documentation provided. Try to isolate WHERE things are dying, and *ask* for help, don't demand psychic software that does everything. UNIX and subsequently Linux, is big, MSDOS is simple but for that price, you get a glorified boot loader. Daniel Supernaw-Issen email to danielsi@cs.utexas.edu send all flames to me via email. All opinions are mine, Linux or Die! ------------------------------ From: artemis@eve.kaist.ac.kr (Hur InHo) Subject: Sound Driver question. Date: Wed, 14 Apr 93 17:02:43 GMT Is there any utility using sound driver? and where's the sound driver, and utility? ------------------------------ From: artemis@eve.kaist.ac.kr (Hur InHo) Subject: anyone have kterm binary? Date: Wed, 14 Apr 93 18:10:05 GMT I couldn't compile kterm-5.1......... anyone knows where the binary is, please let me know. ------------------------------ From: jhenders@wimsey.bc.ca (John Henders) Subject: Re: Trying to kill getty so I can dial out on ttys2; help Date: 14 Apr 93 19:14:10 GMT In article <1qcu58INN6u3@iskut.ucs.ubc.ca> jmorriso@rflab.ee.ubc.ca (John Paul Morrison) writes: > >I thought the /dev/cua?? devices were supposed to work for dialout. >they dont work properly though. typically a getty is running, say on ttyS0, >and you should be able to start kermit with /dev/cua0, and the getty shouldnt >hog the line if it's just waiting. anyway, maybe that'll get fixed. instead >of editing inittab every time, just kill the getty and start your dialout >in the same line: > kill pid; kermit -l /dev/modem >it works, and it's easier than editing inittab every bloody time. the >kill gets rid of the getty sitting on the port, and kermit gets started very >quickly afterwards; quicker than init can respawn the getty. Another way that works quite well is to only respawn uugetty at init levels 5 and 6, or even just 6. Then if you want to call out, do an init 4. The new duel package is supposed to fix all this though, so hopefully all this wil be history soon. -- John Henders ------------------------------ From: dov@menora.weizmann.ac.il (Dov Grobgeld) Subject: Mounting an extended dos partition Date: 14 Apr 93 13:51:19 GMT My Dos disk is partitioned into two parts, a primary partion (C disk) and an extended partition, with one logical partition (D disk). There is no problem to mount the C disk with the command mount -t msdos /dev/hda1 /usr/dos but the similar command mount -t msdos /dev/hda2 /usr/ddisk or mount -t ext /dev/hda2 /usr/ddisk fails with the error message mount: Unknown error 26240 I guess this is expected as /hda2 is not an 'msdos' partition but an 'extended msdos' partition. Has anybody overcome this problem? Does mount at all support extended ms-dos partitions? Mtools doesn't manage to access this partition either. My /etc/mtools file looks as follows: A /dev/fd0H1440 12 80 2 18 # A: 5 1/4 C /dev/hda1 16 0 0 0 D /dev/hda2 16 0 0 0 But trying to access d: gives the error message: Probable non-MSDOS disk mdir: Cannot initialize 'D:' Any solution to this? -- ___ ___ / o \ o \ Dov Grobgeld ( o o ) o | The Weizmann Institute of Science, Israel \ o /o o / "Where the tree of wisdom carries oranges" | | | | _| |_ _| |_ ------------------------------ From: fox@graphics.cs.nyu.edu (David Fox) Subject: Re: What are the .z files? Date: Thu, 15 Apr 1993 01:59:36 GMT In article <1993Apr14.183106.19013@mnemosyne.cs.du.edu> zevans@nyx.cs.du.edu (Zack Evans) writes: In article <1993Apr13.142458.15411@waikato.ac.nz> hamish@waikato.ac.nz writes: > >Hey. What format are the .z files on nic.funet.fi, and tsx-11? They aren't >compressed, and I can't find anything that will undo them. Any hints? Yes, they are gzipped...GNU zip - check archie for availability. If Richard Stallman ever made a mistake it was starting to use gzip IMHO, what was wrong with .tar.Z? The compress algorithm is covered by a patent, and may be subject to royalty payments. Besides, gzipped files are usually a third smaller than compressed files. ------------------------------ From: stimpson@panix.com (Stimpson Joel Katz) Subject: An quick-fix to the static binaries issue Date: Wed, 14 Apr 1993 21:47:50 GMT I know many of you have discussed concern about messing up installing shared libraries and then having nothing work. My solutios is to make copies of my essential utilities, rm, sh, ln, ls, and so on with the first letter capitalized. Then I edit those files so that instead of looking for "libc..." they look for "Libc.." and I make a copy of the library and symlinks with the first letter capitalized. Now if I mess up my libraries or symlinks I can use the capitalized versions of those utilities to repair the damage. Easy. Works. SJK ------------------------------ From: jmorriso@rflab.ee.ubc.ca (John Paul Morrison) Subject: hooray! voodoo spell broken. network works now! Date: 14 Apr 1993 23:19:52 GMT a few days ago I posted that our networking had mysteriously died. there were no obvious indications why: all I had done previously was edit /etc/inittab and /etc/gettydefs to try and get a modem working. then a reboot or two later, and the ethernet card would NOT transmit. I would get wd8003 transmit timed out. errors in the logs. I had a backup on tape, and I restored all of /etc, /etc/inet and the kernel. the network still wouldnt work. ping would just generate kernel errors (the wd8003 ...error). I thought this was a bit strange. I restored any kernel I had saved. no luck. I built 99.8 from patches. no luck. I built 99.8 fresh from the sources. no luck. the only thing that worked was an old SLS boot disk. Some e-mail correspondence suggested that I use the ethernet drivers from ftp.super.org. Well whattaya know: that old SLS boot disk pops up with a message attributing the drivers to super.org. I cant explain why *several* kernels later, 99.7, 7A, etc. the net still worked, then *without* even touching the kernel, it mysteriously stops working. Anyway, I followed the instructions on ftp.super.org, and I used those ethernet drivers to make a kernel that works with our SMC 16 bit ethernet card. So Im happy it works now. Anyone have theories about why it *stopped* working, when there was no visble change to the files? (this is weird: I restore files from a tape made from a *working* system, and the system still wont work). maybe the soft settings on the ethernet card got twiddled , making it fail with my previously working kernel. -- __________________________________________________________________________ John Paul Morrison | jmorriso@rflab.ee.ubc.ca VE7JPM | .sigfile without a cause ubc-cs!rflab.ee.ubc.ca!jmorriso | ________________________________________|_________________________________ ------------------------------ From: gthomas@fraser.sfu.ca (Guy Thomas) Subject: 99p8 breaks xdm Date: Wed, 14 Apr 1993 22:46:52 GMT Well, here's to life on the bleeding edge. I initialy installed the new kernel, 99p8, which caused xdm to fail. - xdm comes up and presents login prompt - login takes username and passwd - seems to pass control to xinit - then snaps back to the login So, switch back to 99p7A. Lilo 0.10 comes out. Install lilo boots fine. Now X wont even come up! Both for the 99p8 kernel and the 99p7A kenel. - startx runs and sets userresources and systemresources - xinit gets called - xinit dies with .. Fatal server error: server is already active for display 0 What is that now? Thouroughly baffled. Any help appreciated. Thanks Guy R. Thomas gthomas@native-ed.bc.ca ------------------------------ From: garrett@garnet.berkeley.edu (Garrett D'Amore) Subject: About disk quotas... Date: 15 Apr 1993 01:32:04 GMT I have a thought on quotas... my thinking originates from experience as a Novell programmer. I'm no Un*x expert, and I'm new to Linux, but I do write code and have I think enough understanding of how things work to suggest a possible solution... I think a quotad would work... hang-on, don't flame me just yet... :-) 1) Make a filesystem limit using *directories*. If quotas were based on "max K per directory", this could very easily be translated to a user basis, since most users (that you'd want to restrict) don't have write privs. outside of /usr/spool, /tmp, and their home dir. You wouldn't want to limit the spool and tmp dirs much, except maybe alert root when a user's usage gets large without shrinking. Limit the home directory. 2) Home dirs are limited by belonging to root or somesuch, with g+rwx (where g is just root & the user). When the user's disk space gets large enough, the daemon just turns off g+w in the home dir, and a+w in all subdirs (recursively). 3) Quotad should e-mail both the user & root (or other admin) whenever it takes action. Problems to overcome: 1) quotad should know about mail spool dir, and use that when changing permissions on home dir. Users could still cheat, but they would have to use the spool dir. This is usually more inconvienent than tarring and zipping or removing files in the users home dir (to the user). 2) quotad must know how to restore the rights to a users dirs at any given time. In particular, it should restore the "w" rights only if they were there before. This may be overcome in 1 of two ways (others?) i) use some sort of database file to store original permissions. ii) use extra permissions bits or inode bits if 3 bits can be found. (try to avoid using inode bits to maintain fs portability) 3) user-owned sub-dirs... may be problematic. User could reset the permissions if he owns them. Perhaps this can be solved by having quotad automatically change subdir ownership at the same time that it does chmod.... Advantages: 1) filesystem and kernel independent. (yay!) 2) spool files unaffected. (allows printing of big postscript files and creation of temporary files) We use this type of restriction instead of the user quotas in Netware 3.11 because users frequently spool postscript files that are bigger than any quota we would give them, but these spools are very temporary. 3) No problems if user does chown. Disads: 1) can only be done to certain dirs? user doesn't get write access elsewhere, or if he does it is unrestricted... ideas for fixes.. 2) probably a kludge that is not terribly difficult to get around. A kernel based solution would be more secure. Any Linux hackers out there want to comment on this idea? Is it feasible? Is it worth it? Is anyone working on a kernel or fs-based solution? ==================================================================== Garrett D'Amore | garrett@haas.berkeley.edu Software Co-Ordinator | 68 Barrows Hall, UC Berkeley Haas Computing Services | Ph: 510-643-5923 Fax: 642-4769 ==================================================================== ------------------------------ From: einari@rhi.hi.is (Einar Indridason) Crossposted-To: comp.editors Subject: Re: Long lines in vi (elvis) Date: 14 Apr 93 23:54:54 GMT In <1qevjm$k0s@hal.gnu.ai.mit.edu> mycroft@hal.gnu.ai.mit.edu (Charles Hannum) writes: >The thing which severely annoys me about all vi clones is that `8dd4..' >doesn't do the same thing as in the original. >For the uninitiated, in vi: > 8 - goes into the global argument register > dd - deletes (8) lines > 4 - replaces the global argument register > . - repeats the `dd'; deletes 4 lines because the register changed > . - repeats the `dd', again deleting 4 lines >One clone repeats the whole `8dd', deleting 32 lines for the first `.'; >one ignores the `4'; one gets the `4' but forgets it later; etc. What I would like, (and I feel is logical, but that is just my opinion), is that the above command '8dd4..' would be intepreted as: put 8 as a numeric argument in the Global Counting Register dd delete lines... how many? 8, since that is what is the the GCR (:-) put 4 as a numeric argument in the global counting register (GCR for short) . repeate last 'action' (which was dd) with the GCR (implied: put 1 in GCR) <--------------------- Note this line....... . repeat last 'action' (dd) with the GCR which (IMHO) should contain 1, i.e. delete one line. This is what I think is logical, but you are of course free to have your own views about that :-) (Oh, my.... we are getting into an internal vi bashing. Quick, say something nasty about gemacs, and all other editors, to draw attention away from this thing :-) :-) :-) (Ps regarding gemacs: I think the choice of an editor is personal, and I'll respect your selection of XXX as your editor, if you will respect my choice of YYY as my editor. (Insert suitable XXX and YYY.)) (PPs I have no intention to start yaew (yet another editor war) with this note, so please, please don't write back saying that editor XXX is better than YYY or vice verse!) -- einari@rhi.hi.is ------------------------------ From: balasub@skink.cis.ohio-state.edu (Krishna Balasubramanian) Subject: Re: SYSV IPC:function not implemented Date: 14 Apr 1993 21:39:00 -0400 In article <1qfuph$58c@usenet.INS.CWRU.Edu> damien@b63519 (Damien Neil) writes: I'm trying to get SYSV IPC working on my 0.99.8 kernel. The diffs installed nicely for the most part, and I added the ones that didn't by hand. The kernel compiled without complaint. However, whenever I run a program that uses IPC, I get a `Function not implemented.' error. I'm at a loss. Does anyone have any suggestions? Many thanks! Damien Neil dpn2@po.cwru.edu For ipcdelta.tar.z 1) The patch in fsync.cd is needed only with linux-0.99.7. Back it out with patch -R if you applied it. 2) The diffs in ipc.cd fail in kernel/sys.c The patch is trying to delete the stub for sys_ipc() present in this file. So edit kernel/sys.c and remove the lines int sys_ipc (void) { return -ENOSYS; } Mail me if you have problems. krishna ------------------------------ From: s_fuller@iastate.edu (Steve Fuller) Subject: Re: Disk Quotas (was Re: New feature for the filesystems.) Date: Thu, 15 Apr 1993 01:40:52 GMT In <9821@dirac.physics.purdue.edu> bcr@bohr.physics.purdue.edu (Bill C. Riemers) writes: >In article davidsen@crd.ge.com (bill davidsen) writes: >> I've never seen space deleted for you, you just can't write until you >>get down below soft limit. Of course that may be an option, but I've not >>found it in the docs. The sysadm is usually the dark force, through >>scripts. >Yes I haven't seen a system yet that properly handles hard quotas. Most >systems will still let you write to /tmp so you just direct the normal >programs to put thier output to there and then take care of them before >logging out. For the few systems that I've seen that aren't fooled by >placing the file in /tmp, they are fooled if you do a chown on the file It would seem to me, from a new sysadmin standpoint, to have anyone but root able to use chown. At least that's how it's done here on the Ultrix systems at our site. Our system works as follows as far as quotas: 1) If the file is in the middle of a write when you exceed your quota, the file will be written in its entirety and not truncated. Any furthur write requests will be denied If you don't have any disk space when you start writing a file, you will be told you have exceeded the quota. >before you exceed your quota. Installing a hard quota is by no means a >simple task. For the most part, if you are using a file system that reserves >space for root, these shouldn't be neccissary. > Bill -- ==================================================== Steve Fuller = s_fuller@iastate.edu = No witty quote here yet... =================================================== ------------------------------ ** 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-Activists-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux) via: Internet: Linux-Activists@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 tupac-amaru.informatik.rwth-aachen.de pub/msdos/replace The current version of Linux is 0.99pl6 released on February 23, 1993 End of Linux-Activists Digest ****************************** other editors, to draw attention away from this thing :-) :-) :-) (Ps regarding gemacs: I think the choice of an editor is personal, and I'll respect your selection of XXX as your editor, if you will respect my choice of YYY as my editor. digest842 644 36676 74430 51642 5366137074 5766 From: Digestifier To: Linux-Activists@senator-bedfellow.mit.edu Reply-To: Linux-Activists@senator-bedfellow.mit.edu Date: Thu, 15 Apr 93 02:30:15 EDT Subject: Linux-Activists Digest #842 Linux-Activists Digest #842, Volume #4 Thu, 15 Apr 93 02:30:15 EDT Contents: Why can't X-Win find my VTs??!?? (Eirik Ora) Re: weird vi/less screen insect (Kevin Sanders) Re: X hangs.. (on occasions) (Hwan-Rei Lee) [Q] Printing question (Shen-Wei X Li) Re: [Q] Printing question (Shen-Wei X Li) FAQ? Booting MSDOS from partition 1 fails (Dag H. Wanvik) Re: linux logo or mascott (Michael C. Williams) My mouse has gone crazy (Robert A. Hayden) Re: Possible gotcha in gcc (Rick Sladkey) Re: Compiling XV; general porting problems (Bill Broadley) Re: How to get rid of LILO? (Mark Morley) Re: The dangers of playing with shared libraries (James Michael Chacon) NFS troubles (KENNEDY JAMES SCOT) sudo, libl and liby (Steven C. Saliman) Re: Accessing Kodak Photo-CD (Nigel R Gamble) W32/ET4000 and X: pixels too big (Zippy) Big swap partitions [> 16M] (David Baggett) unclutter exhibiting odd behavior (Garner Halloran) Re: ka9q 14? (HJ Lu) ---------------------------------------------------------------------------- From: Eirik.Ora@newcastle.ac.uk (Eirik Ora) Subject: Why can't X-Win find my VTs??!?? Date: 15 Apr 93 00:38:54 GMT I have been struggling with the following problem since I started installing X-Windows on my PC (I still have not seen X-Windows running on it ... :-( ) I have a 486, 8Mb, with a Trident 8900c video-card, and I have made the XConfig-file and I have put the .xinitrc file in my root, but still ... This is exactly what happens when I type startx : /# startx Setting TCP SO_LINGER: Protocol not available no SIOCGIFCONF XFree86 Version 1.2 / X Window System (protocol Version 11, revision 0, vendor release 5000) Configured drivers: VGA2 (monochrome VGA): et4000, et3000, pvga1, gvga, tvga8900c, tvga9000, tvga8900b, generic HGA2 (monochrome HGA): hga6845 Fatal server error: Cannot find a free VT giving up. xinit: Invalid argument (errno 22): unable to connect to X server xinit: No such process (errno 3): Server error. /# _ Can somebody with a bit more insight that me try to point me in the right direction? I have read all the FAQ's, but none of the hints that I found there solved my problem... It also seems like there are some other Trident 8900 - persons who have met the same problems... Eirik =============================================================================== Email:Eirik.Ora@newcastle.ac.uk Post:Eirik Ora, CSSD group, Department of Computing Science, The University, Newcastle upon Tyne, NE2 7RU, UK Telephone : +44 91 222 6000 ext 8008 (lab) / +44 91 261 1539 (home) ------------------------------ From: kevin@TorreyPinesCA.ncr.com (Kevin Sanders) Subject: Re: weird vi/less screen insect Date: Thu, 15 Apr 93 00:58:35 GMT In article <1qfgg6$mjm@usenet.INS.CWRU.Edu> wb8foz@skybridge.scl.cwru.edu (David Lesher) writes: > >Linux softland 0.99.pl4-62 01/23/93 i386, Elvis 1.6, less 177 > >I'm also looking at a real weird one involving elvis, maybe less, and >Linux. I'll be doing something in vi. I typically get to vi via the 'v' >command inside less, this because I look before I leap, and afterwards. >Then I do SOMETHING that crashes me out of ELVIS. What depends. Hitting >the colon, and another command too quickly sometimes causes it. >Sometimes, I hit ^Z (or try to). > >In any case, it looks as if I end up back in less, as I should. But I >hit a command, and it ACTS as if I'm still in vi! Parts of the screen >overwrite with "vi"-isms, such as line numbers. ":" drops me down to the >proper place. Then, ^L's seem to toggle through various screens - less, >vi, and maybe the shell itself. Looking at 'ps' from another tty shows >less & vi still running. > I have seen a very similar problem involving "elm" and "vi". If you are composing mail under elm, it invokes your EDITOR, in my case vi. If you suspend from the editor with ctrl-Z and then resume it with "fg", it appears input is being read simultaneously from vi and elm. Anyone else have any clues? Perhaps some weirdness with process groups, or with interrupted system calls? By the way, either elm or vi by themselves seem to handle ctrl-Z just fine. -- Kevin Sanders, KN6FQ NCR Torrey Pines kevin.sanders@torreypinesca.ncr.com (619) 597-3602 kevin%beacons@cyber.net ------------------------------ From: eea78555@rangers (Hwan-Rei Lee) Subject: Re: X hangs.. (on occasions) Date: Thu, 15 Apr 1993 01:45:35 GMT Do you try: Alt-Crtl-F2(3-8) ? -- Henry Lee, Hwan-Rei, Lee.... -0^0- . | ~ | .W. \|/.................oO | Oo................\|/... ------------------------------ From: sxl5151@cs.rit.edu (Shen-Wei X Li) Subject: [Q] Printing question Date: Thu, 15 Apr 1993 01:37:08 GMT Hi, folks, Why there is nobody talking about the printing problem? Does everyone work fine? I really have trouble to use the whatever 'lpr' or 'ghostscript'. Can somebody help me? Any pointer or comment will be very appreciated. -Bravo. ------------------------------ From: sxl5151@cs.rit.edu (Shen-Wei X Li) Subject: Re: [Q] Printing question Date: Thu, 15 Apr 1993 01:53:10 GMT Here is the meesage when I tried to print out a regular acsii file by lpr: # lpr Xconfig lpr: connect: Invalid argument jobs queued, but cannot start daemon. # The other one is by using GhostScript: GS> (ljet2p) selectdevice GS> (chess.ps) run Both of the above can not print out anything, but if I tried to print out a PostScript file to a 'file', that works: # gs -sDEVICE=ljet2p -sOutputFile=myfile.out chess.ps But if I use the lpr pipe, then it will die.... # gs -sDEVICE=ljet2p -sOutputFile=|lpr chess.ps Please help me, any pointer or comment will be greatly appreciated. Thanks, ------------------------------ From: Dag.H.Wanvik@kvatro.no (Dag H. Wanvik) Subject: FAQ? Booting MSDOS from partition 1 fails Date: 14 Apr 93 21:40:09 GMT I'm sorry if this is a FAQ. I have successfully partitioned my hard disk and installed Linux SLS 0.99p6 /dev/hda: C: 1 active MSDOS 2 Linux ex2fs 3 Linux ex2fs I then reinstalled MSDOS on partition 1, which first made me reformat partition 1 and then reinstalled, successfully, I was told by MSDOS setup :-(. I can mount the MSDOS partition under Linux, but MSDOS fails to boot from the harddisk. The MSDOS filesystem is OK, as I can 'dir' it when booting MSDOS from floppy. Somebody said MSDOS is touchy and needs to be installed *before* any Unix(Linux) on the other primary partitions. Is is the MBR (master boot record) on the disk which is the problem? When I try to boot, the BIOS tells me "missing operating system". Thanks, Dag -- Dag H. Wanvik, Phone: +47 7 520090 KVATRO AS, Pirsenteret, Fax: +47 7 520140 N-7005 Trondheim, NORWAY E-mail: dag.h.wanvik@kvatro.no ------------------------------ From: mcw1@crux2.cit.cornell.edu (Michael C. Williams) Subject: Re: linux logo or mascott Date: 14 Apr 93 21:03:19 GMT >>My 0.02 cents to this thread. I think a lobster would be a good >>mascott for linux. I like the alliteration "linux lobster" (in >>English at least). Just as eating lobster is a special treat for >>humans, linux is an especially good os to feed your computer! >And a quote :- > Keep your hardware happy, Linux. This is going to have to have my vote.. An animal that is both unique and will look good in 256 colors :) -Michael ============================================ mcw1@cornell.edu, mcw1@crux2.cit.cornell.edu Finger for PGP 2.2 public key. Plus ca change, plus c'est les memes choses. ============================================ ------------------------------ Subject: My mouse has gone crazy From: hayden@krypton.mankato.msus.edu (Robert A. Hayden) Date: 14 Apr 93 21:13:47 -0600 I am using .99pl8 and the most recent version of selection. In the last few days I have noticed that my mouse (MS-compat. on COM 1) will suddenly go completely and utterly crazy; not moving where he's supposed to, randomly clicking buttons without my fingers, driving me essentially crazy :) I do not have anything that has conflicting inturrupts (Modem on com2, no printer. SBlaster card on IRQ7). Any thoughts at all what might be causing this? I've made two recent changes to the system. Upgrade to pl8 and change from efs to e2fs. Thanks for your time. -- [> Robert Hayden ____ <] Black Holes result from God [> \ /__ <] dividing the universe by zero. [> hayden@krypton.mankato.msus.edu \/ / <] [> aq650@slc4.INS.CWRU.Edu \/ <] # include std_disclaimer.h ------------------------------ From: jrs@world.std.com (Rick Sladkey) Subject: Re: Possible gotcha in gcc Date: Thu, 15 Apr 1993 02:20:55 GMT >>>>> On Wed, 14 Apr 1993 13:01:10 GMT, >>>>> davidsen@ariel.crd.GE.COM (william E Davidsen) said: bill> I noted an interesting behavior in gcc last night, it doesn't bill> ignore stuff which is #ifdef'd out! I got a program to port to bill> Linux, and I had some text included which I found valuable. bill> Having learned long ago that putting /*...*/ around arbitrary bill> stuff might be a problem due to */ in the text, I put #if 0 ... bill> #endif around it instead. A gcc generated error messages about bill> the text it should have been ignoring! I noticed this too. An excerpt from the GCC manual: +----- | * GNU CC will flag unterminated character constants inside of | preprocessor conditionals that fail. Some programs have English | comments enclosed in conditionals that are guaranteed to fail; if | these comments contain apostrophes, GNU CC will probably report | an error. For example, this code would produce an error: | | #if 0 | You can't expect this to work. | #endif | | The best solution to such a problem is to put the text into an | actual C comment delimited by `/*...*/'. However, `-traditional' | suppresses these error messages. +----- ------------------------------ From: broadley@pitt.edu (Bill Broadley) Subject: Re: Compiling XV; general porting problems Date: 15 Apr 93 01:57:51 GMT > This is a well known bug with gcc, not just the Linux version. Gcc will > take outrageous amounts of memory when compiling large static initialized > arrays. Fear not! there are people with such ridiculous amounts of memory > and swap space; Rob Hooft is one of them. You can find his jump-tabled > compilations of lib in /pub/Linux/apps/graphics/libgr-1.0.tar.z > on sunsite.unc.edu. As far as I know this isn't related to gcc, other vendors compilers do the same. -- Bill 1st> Broadley@neurocog.lrdc.pitt.edu Broadley@schneider3.lrdc.pitt.edu <2nd 3rd> Broadley+@pitt.edu Linux is great. Bike to live, live to bike. PGP-ok ------------------------------ From: morley@suncad.camosun.bc.ca (Mark Morley) Subject: Re: How to get rid of LILO? Date: Wed, 14 Apr 1993 21:39:34 GMT The DOS command FDISK /MBR will write a new master boot record. Works great to get rid of any multi-boot programs I've ever used (LILO, Coherent's, OS/2's, Windows NT's, etc) =================================================================== Mark Morley, UNIX/SUN Manager NET: morley@camosun.bc.ca Camosun College - Interurban Campus TEL: (604) 370-4601 4461 Interurban Road Room 143-Tech FAX: (604) 370-3660 Victoria, B.C. Canada V8X 3X1 ------------------------------ From: probreak@matt.ksu.ksu.edu (James Michael Chacon) Subject: Re: The dangers of playing with shared libraries Date: 14 Apr 1993 20:31:02 -0500 keith@ksmith.com (Keith Smith) writes: >In article caywood@wyvern.wyvern.com (John Caywood) writes: >>Steve VanDevender (stevev@miser.uoregon.edu) wrote: >> >> [ stuff deleted ] >> >>: in to fix the problem, I would like to _strongly_ suggest that >>: 'ln' be statically linked or that a statically linked version of >>: 'ln' be made available in the distribution so that it is possible >>: to recover from this mistake the obvious way. >> >>I second the motion. I got bit installing gcc486 instead of the gcc386 >>that came with SLS 0.99pl2. 'ln' is not the only candidate for static >>linking -- I can think of ls, mv, and (maybe) cat in the same got-to-have-it >>category. I can't see it if I can't 'ls'; if I can 'mv' it, I don't need >Sure you can "see" it : ># echo * >Will happily list all the files in your cwd. >Bare minimum is your shell, 'cat' and 'mv' IMHO. 'ln' would be nice, >but not absolutely neccessary, You can always make a COPY temporarily, >and then link back correctly later. Actually, since the shell can simulate cat with: while read x; do echo "$x"; done < file You can cp with this, so that should be all you need. Just a bash process, and I am set. James ------------------------------ From: kennejs@a.cs.okstate.edu (KENNEDY JAMES SCOT) Subject: NFS troubles Date: Wed, 14 Apr 93 21:41:03 GMT First let me say that I'm sorry about causing unneeded traffic in this newsgroup by asking questions that I should have been able to figure out for myself. Specifically, I'm referring to my last post in which I asked why networking on my system wasn't working anymore after I upgraded to libc 4.3.3. Yesterday I reread the release note for libc 4.3.3 and I discovered that it indeed mentioned that you need to move the networking files to /etc. Somehow or another I missed seeing this when I read the file before. In the future I will endeavor to carefully read all the release notes and FAQs before posting to this group. This isn't the first time that I've had to eat my own words. :-) I realize that this group gets a lot of traffic so I will try to minimize my postings by going through all the relevant docs first and corresponding with people via email. Whenever possible I do try to figure things out first. As a matter of fact, I've helped out a number of people myself by way of email...the reverse is also true. I suggest that other people do the same whenever possible. BTW, I think the creation of c.o.l.a. was a great idea. If we didn't have this newsgroup then it would take a lot more time to wade through c.o.l. to find info on new releases and important notices. My compliments to the people who thought up this group. Now here is a *real* problem I'm having. After upgrading to the new version of mount I've been unable to mount a directory using NFS. I have my Linux system networked to a Sun 386i. I can't mount any of the Sun's directories onto my Linux system. When I was using an old version of mount with .99pl7 (I'm still using that kernel btw) NFS worked just fine. Now it doesn't. I understand that a new version of the NFS server has been released. If I get this then will my problems go away? I suspect that mount may be the culprit but it's possible that libc 4.3.3 is causing this problem. Here's what happens when I try to mount a remote directory: mount hoek:/disk /mnt mount: No such file or directory I telnetted over to the Sun and mounted one of the directories over on the Linux system (mount stimpy:/ /mnt; stimpy is the name of my Linux box). This worked just fine. When I try to mount one of the Sun's volumes onto the Linux system I get the error message above. Also, I was unable to mount / and /usr. I got the same error message as before. I've checked the exports file on the Sun and everything appears to be in order. This problem is truly puzzling. I'd really appreciate it if someone could help me out. Again, let me say that I greatly appreciate all the help that I've gotten so far. A number of diiferent people told me how to fix networking and install the new libraries. ---Scott kennejs@a.cs.okstate.edu ------------------------------ From: saliman@alumni.cs.colorado.edu (Steven C. Saliman) Subject: sudo, libl and liby Date: Thu, 15 Apr 1993 02:37:36 GMT I am trying to compile a sudo program, but I am having trouble. It uses yacc and lex, and therefore links with the liby and libl librearies, which I don't seem to have. Does anyone know where I could fine them? I am using .99p6 with the gcc that came with the SLS distribution (2.3.3 I think) I tried out the su1 program, but I didn't like having to type a password for each command. I also like the way that sudo uses the user's password for verification, instead of different ones for different commands. Thanks in advance for any help. *Steven C. Saliman Saliman@cs.colorado.EDU * *Thought for the day: * * If a train-station is where a train stops, * * then what happens at a workstation? * ------------------------------ From: nigel@world.std.com (Nigel R Gamble) Subject: Re: Accessing Kodak Photo-CD Reply-To: gamble!nigel@uunet.UU.NET Date: Wed, 14 Apr 1993 22:12:25 GMT job@euklid.informatik.uni-dortmund.de (Dirk Job) writes: >I tried to access a Kodak Photo-CD with my Toshiba XM3401 SCSI CD-ROM. I have a hack to the SCSI CD-ROM driver that allows me to mount either standard CD-ROMs or Photo-CDs on a 3401. I'll make a patch and try to send it to you on the weekend. -- Nigel Gamble gamble!nigel@uunet.uu.net Boca Raton, FL, USA. uunet!gamble!nigel ------------------------------ From: seth@gradient.cis.upenn.edu (Zippy) Subject: W32/ET4000 and X: pixels too big Date: 14 Apr 93 22:12:59 GMT Does anyone know why the pixels in X are 5 to 10 times their normal size when I use my W32/ET4000 (1Mb) LB chipset? Is there a solution to this? If you're using this chipset and not having any problems, PLEASE let me know (by email; I'll summarize if I get responses). Thanks, Seth Bromberger Seth Bromberger seth@gradient.cis.upenn.edu M.S.E. '93, University of Pennsylvania seth@eniac.seas.upenn.edu ------------------------------ From: dmb@case.ai.mit.edu (David Baggett) Subject: Big swap partitions [> 16M] Date: 15 Apr 1993 05:38:55 GMT Reply-To: dmb@ai.mit.edu I've checked the FAQ for this, but no luck: I'm running 0.99.8 and I made a 32M swap partition. mkswap tells me that it's allocated that much room, but swapon tells me it's only added 16M, and free corroborates this. Is this a hard limit, or can I tune some kernel param to get the whole 32M? (BTW, I have 8M of physical memory, if that makes a difference.) Dave Baggett __ dmb@ai.mit.edu Natural Language Processing MIT AI Lab ADVENTIONS: We make Kuul text adventures! Ask about Unnkulian 1, 2, 0, 1/2 PO Box 851 Columbia, MD 21044 USA / CIS: 76440,2671 / GEnie: ADVENTIONS ------------------------------ From: kheldar@prism.gatech.EDU (Garner Halloran) Subject: unclutter exhibiting odd behavior Date: 14 Apr 93 22:12:21 GMT I'm trying to use unclutter under X, but after the mouse pointer goes away I can no longer type anything in the window. I got it to work once but have been unable to repeat my success. I suspect its something in my X setup, but I haven't the faintest idea of what it could be. I tried running unclutter from my .xinitrc and just from the command line. Also none of the options seem to have any effect. Any ideas? -- "A common mistake that people make when trying to design something completely foolproof was to underestimate the ingenuity of complete fools." - Douglas Adams, "Mostly Harmless" Garner Halloran - kheldar@pixie.gatech.edu ------------------------------ From: hlu@luke.eecs.wsu.edu (HJ Lu) Subject: Re: ka9q 14? Date: Thu, 15 Apr 93 03:53:11 GMT In article ran59961@uxa.cso.uiuc.edu (Richnut) writes: >i saw it on tsx-11 but when i run it it says "cannot load /lib//libc.so.4 Incom >patible version" > >So which lib do i need, i currently have 4.3.3 > After reading release.libc-4.3.3, do cd /lib ln -sf libc.so.4.3.3 libc.so.4 H.J. ------------------------------ ** 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-Activists-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux) via: Internet: Linux-Activists@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 tupac-amaru.informatik.rwth-aachen.de pub/msdos/replace The current version of Linux is 0.99pl6 released on February 23, 1993 End of Linux-Activists Digest ****************************** ppens at a workstation? * ------------------------------ From: nigel@world.std.digest843 644 36676 74430 73167 5366137075 5776 From: Digestifier To: Linux-Activists@senator-bedfellow.mit.edu Reply-To: Linux-Activists@senator-bedfellow.mit.edu Date: Thu, 15 Apr 93 03:30:18 EDT Subject: Linux-Activists Digest #843 Linux-Activists Digest #843, Volume #4 Thu, 15 Apr 93 03:30:18 EDT Contents: xvt for Linux??? (Gary McTaggart) *** READ THIS BEFORE POSTING *** (Ian Jackson) Re: Networking between two linux machine without Ethernet Card (Harold Brian Raynor) Compaq Deskpro 386/20 & Linux ? Re: Why Linux should remain Linux and seagulls should not be used. (Peter MacDonald) [Q] D-Link card de200 setting needed (Francis Hsieh (LS)) New version of MCC interim Linux available (A. V. Le Blanc) Re: Linux Mascott/Logo (Garrett D'Amore) Re: Compiling XV; general porting problems (Leon Dent) Re: weird vi/less screen insect (Michael O'Reilly) Re: weird vi/less screen insect (Michael O'Reilly) Re: Linux on Macintrashes? (Was: Re: Linux on Amiga?) (Art Taylor) Re: FAQ? Booting MSDOS from partition 1 fails (Frank T Lofaro) Re: Crashing machine: >16M mem patch (Stephen Tweedie) Re: HELP WITH DD (Stephen Tweedie) Re: New feature for the filesystems. What do you think ? (Stephen Tweedie) Re: Compiling XV; general porting problems (Wolfgang STEGER) Re: Searched: cc1obj (HJ Lu) Re: Basic info on LINUX. (Drew Eckhardt) ---------------------------------------------------------------------------- From: gmt@beach.cis.ufl.edu (Gary McTaggart) Subject: xvt for Linux??? Date: 15 Apr 1993 06:03:56 GMT Has anyone ported xvt (a smaller replacement for xterm) to Linux? (Actually, I'd be interested in anything similar to xvt.) Thanks, Gary ------------------------------ From: ijackson@nyx.cs.du.edu (Ian Jackson) Subject: *** READ THIS BEFORE POSTING *** Date: Thu, 15 Apr 1993 06:23:00 GMT To get people to read your post and answer your question: - Read "Linux Documents Explained for Newbies" - posted weekly (see below). - Please do not post questions answered in the FAQ lists. - Please do not post "Please send me the FAQ". See below. - Please do not post "Split this group now". See below. - Please read the documentation for the program you have a problem with. - Be informative, both in subject line and body. See below. - Linux-specific posts only please - Unix, X, C, etc. groups listed below. For more information read on ... *** Linux Documents Explained for Newbies (like me 8-) *** is is posted weekly by Jay MacDonald. If you can't find it email me or Jay . It will tell you about the Linux FAQ, the INFO-SHEET, the META-FAQ, etc. These are posted regularly to comp.os.linux and c.o.l.announce, and can be found in /pub/usenet/news.answers/linux-faq at pit-manager.mit.edu and on the Linux archive sites. If you can't get them there, try mailing "mailserver@nic.funet.fi", "mail-server@pit-manager.mit.edu" or "ftpmail@decwrl.dec.com". If you can't get them any other way, e-mail me. Splitting the group (due to volume or other reasons): It's been tried, and all the new groups except .announce failed. It can't be tried again within 6 months. See news.announce.newgroups. If you insist on discussing it, do it in news.groups (if c.o.l has too much volume already, why make it worse?). With threaded newsreaders now in widespread use using a good, specific, polite subject line will get you many more readers. Most people will ignore posts with subjects like "Linux", "Help", "SLS", etc. Don't shout to get attention. Put as much information in your post as possible - preferably a brief summary followed by session transcripts, etc. as appropriate. Include the precise text of any error messages printed, etc., and appropriate info about your setup. Linux is a Unix clone running on PCs; X has its own groups. One of the following groups is is quite likely to be more appropriate for your question: Unix: comp.unix.questions, .admin, .programmer, .shell etc. X: comp.windows.x, comp.windows.x.i386unix (for XFree86 etc) GNU: gnu.emacs.help, comp.emacs, gnu.gcc.help, gnu.* generally PCs: comp.sys.ibm.pc.hardware C: comp.lang.c -- Ian Jackson ijackson@nyx.cs.du.edu PGP2 public key available on request Home: 35 Molewood Close, Cambridge, UK, CB4 3SR; +44 223 327029. These opin- Work and urgent email: iwj@cam-orl.co.uk; +44 223 343398 ions are my own. Olivetti Research Ltd, Old Addenbrookes Site, Trumpington St, Cambridge, UK; ------------------------------ From: raynor@cs.scarolina.edu (Harold Brian Raynor) Subject: Re: Networking between two linux machine without Ethernet Card Date: 14 Apr 93 21:44:13 GMT Re: hooking up two Linux machines with SLIP I've also hear of someone working on PLIP (parallel Line IP) which out to be much faster I don't know very much about the SLIP protocol, but I have used Lantastic/Z (the serial port/modem version) over a 56k bps cord (from my 486 to my old Compaq Portable with its 8086) and performance was very acceptable, considering this was over a standard serial line. It was slow, but I wasn't booting something huge over the "network" either. I bet if you had this setup and tried to run X from the "server", then you could probably go get lunch in the meantime waiting for it to come up! -- -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= Brian Raynor Graduate Assistant in Computer Science University of South Carolina - Columbia, SC ------------------------------ From: p7003ag@sun1.LRZ-Muenchen.DE () Subject: Compaq Deskpro 386/20 & Linux ? Date: Thu, 15 Apr 1993 06:12:36 GMT Hiyo, I got a prob with my Deskpro 386/20: I can't install Linux. After checking all the hardware I have (2 serial, 1 parallel, AHA1542B, Conner 200MB SCSI) the kernel tries to reset my harddisk controller and fails. This it does in an indefinite loop. Any controller I care to try (AT-Bus, MFM ...) produces the same effect. On a NONAME 486 with AT-Bus HD it works just fine. Has anyone encountered similar problems or is anybody running Linux on a Deskpro 386/20 already ? Please help me. CU Alaric ------------------------------ From: pmacdona@sanjuan (Peter MacDonald) Subject: Re: Why Linux should remain Linux and seagulls should not be used. Date: Thu, 15 Apr 93 05:01:19 GMT In article <93104.131126ACPS7221@RyeVm.Ryerson.Ca> writes: > How about a Tiger? A Tiger is powerful, fast, sleek, efficient, and so is > Linux. > And also, Linus is a roarin' good programmer:) Nice, but maybe too aggressive. Has anyone suggested a Dolphin (dolFINN :-). Aren't they supposed to be one of the most intelligent creatures on the planet. They are fast, and they are found everywhere, just like Linux these days. Peter ------------------------------ From: mjhsieh@Uz.nthu.edu.tw (Francis Hsieh (LS)) Subject: [Q] D-Link card de200 setting needed Date: Thu, 15 Apr 1993 06:23:21 GMT Dear linuxer. I have trouble on setting my network using D-LINK de200 ethernet card can u tell me ? Please email me. Thanx!!! ------------------------------ From: zlsiial@uts.mcc.ac.uk (A. V. Le Blanc) Subject: New version of MCC interim Linux available Date: 15 Apr 93 06:42:17 GMT I sent a message some days ago to comp.os.linux.announce, but it seems to have got lost. I have released the current MCC interim version of Linux (0.99-p8) for anonymous ftp from ftp.mcc.ac.uk [130.88.203.12]. This version is available only with the UK keyboard, but it takes little effort to recompile the kernel. Having learned a few things from SLS, there are only two disk images. Other software is included as packages. The packages are not divided into subdisk directories, but should be copied to DOS or Minix disks. All packages fit on three 3 1/2 or four 5 1/4 inch disks, as explained in the README files. Software on these 5 or six disks include source and binary for the kernel (0.99 patch level 8), gcc (2.3.3 with libs 4.3.3), the usual GNU software (now including flex), kermit, groff, and tcpip software, though the latter is still somewhat buggy. A new package includes all patches which were used to compile the included software. Source for all software is available for anonymous ftp from ftp.mcc.ac.uk. This release comes with many README files, including a walk-through README.install. All binaries were compiled with this release of gcc and with this version of the libraries, except for gcc and the binutils, which are as HJ Lu released them. MCC interim Linux is not intended to be an 'official' release, or to compete with any other release. It is a small, moderately well tested version into which other packages (like X, TeX, and emacs) plug without difficulties. It was developed and first released over a year ago partly in response to the local need to install Linux quickly onto teaching machines for short periods (12 machines in 1 hour, use it heavily for 3 days, then delete it and restore DOS). Before you install this version, please read the README files. They do contain useful information (how to recompile the kernel being the main one). Also read the BUGS+WARNINGS file. The problem with syslogd may cause your virtual terminal 1 to be unavailable for login until after you kill syslogd. Rik Faith has been informed and is investigating it. -- Owen LeBlanc@mcc.ac.uk ------------------------------ From: garrett@garnet.berkeley.edu (Garrett D'Amore) Subject: Re: Linux Mascott/Logo Date: 14 Apr 93 22:45:37 GMT Seagulls are nasty dirty birds -- sailors call them "flying bilge rats", and with good reason. They are probably as dirty as chickens or pigeons. Hmm... I liked the narwhal idea... same with the platypus. Linux is a little bit unconventional (at least at this stage in its existence), and a narwhal or platypus would reflect that. Sea (C) creatures in general are fun only becuase of the C, so my vote goes for the narwhal. (Besides which, narwhals can be fast and sleek, live in the north, and have spurned stories of mythological creaturs (unicorns)). "What? free Un*x with sources, on a PC? Get out of here !" ==================================================================== Garrett D'Amore | garrett@haas.berkeley.edu Software Co-Ordinator | 68 Barrows Hall, UC Berkeley Haas Computing Services | Ph: 510-643-5923 Fax: 642-4769 ==================================================================== ------------------------------ From: lcd@umcc.umcc.umich.edu (Leon Dent) Subject: Re: Compiling XV; general porting problems Date: 14 Apr 1993 18:40:27 -0400 Remember, if you need more virtual memory you can add a swap file (if you have the disk space). I needed this when I compiled sunclock. The procedure may be in the FAQ. Leon Dent lcd@umcc.umich.edu ------------------------------ From: oreillym@tartarus.uwa.edu.au (Michael O'Reilly) Subject: Re: weird vi/less screen insect Date: 15 Apr 1993 06:20:44 GMT Ben Cox (thoth@uiuc.edu) wrote: : oreillym@tartarus.uwa.edu.au (Michael O'Reilly) writes: : >As yet, I haven't managed to find a patch for bash that will fix the : >problem. anyone??? : It's called tcsh... :-) Sorry. I want to use a real shell. Also, I can't exactly link tcsh to /bin/sh and have everything work. : Ben Cox : thoth@uiuc.edu Michael ------------------------------ From: oreillym@tartarus.uwa.edu.au (Michael O'Reilly) Subject: Re: weird vi/less screen insect Date: Thu, 15 Apr 1993 06:19:39 GMT Ben Cox (thoth@uiuc.edu) wrote: : oreillym@tartarus.uwa.edu.au (Michael O'Reilly) writes: : >As yet, I haven't managed to find a patch for bash that will fix the : >problem. anyone??? : It's called tcsh... :-) Sorry. I want to use a real shell. Also, I can't exactly link tcsh to /bin/sh and have everything work. : Ben Cox : thoth@uiuc.edu Michael ------------------------------ From: root@fab4box.wa.com (Art Taylor) Subject: Re: Linux on Macintrashes? (Was: Re: Linux on Amiga?) Date: Thu, 15 Apr 1993 02:32:06 GMT In article <1qhhd7INN7m0@matt.ksu.ksu.edu> strat@matt.ksu.ksu.edu (Steve Davis) writes: >nmspillers@ualr.edu writes: >If they were dumb enough to >buy an over-priced-one-mouse-button-tiny-black-and-white-screen- >bloated-awful-looking-os-without-a-cli-and-built-by-fascists computer >in the first place, there's really no reason for them to want or need >UNIX. Duh. Like it will ever run on the vast quantity of internal-screen(ie., 68000) Macintosh's. And who are you to determine what they want, anyway? If, in fact, the Mac OS is evil, is there no room for 'enlightenment'? So, because you bought a PC with DOS(with it's crappy interface and loser mem limits), you are too much of an idiot to use UNIX? > After all, purchasing a Macintrash is acceptance of its point >and drool interface, AND a token of acceptance for Apple's business >practices-- both of which go against the free software/UNIX tradition. I hate to say this, but this is the attitude many of us have about MicroSloth. You get their damn OS with your PC(like you have a choice), and their business practices have drawn their share of hatred and spite(unearned or no). > >>I want to learn, I want to do something worthwhile and give a little back to >>the computing community that has helped me so much, and I think this is a good >>way of achieving these aims. . . > >Not at all. The best way would be to chuck that Mac out the window >and buy a real computer. Good response. I think you're wanted over in comp.sys.amiga.advocacy. -- "Give a family foodstamps, and they'll survive for a week; teach them how to forge foodstamps, and they're set for life." Art Taylor root@fab4box.wa.com (509)522-1178;(509)522-1086 126 Otis Street reeses@halcyon.com Walla Walla, WA 99362 ------------------------------ From: Frank T Lofaro Subject: Re: FAQ? Booting MSDOS from partition 1 fails Date: Thu, 15 Apr 1993 02:41:11 -0400 Excerpts from netnews.comp.os.linux: 14-Apr-93 FAQ? Booting MSDOS from par.. Dag H. Wanvik@kvatro.no (972) >Somebody said MSDOS is touchy and needs to be installed *before* any >Unix(Linux) on the other primary partitions. >Is is the MBR (master boot record) on the disk which is the problem? >When I try to boot, the BIOS tells me "missing operating system". >Thanks, MS-DOS is *so* brain-damaged in this respect it isn't even funny, it is really perverse and sad. I get the same "missing operating system"* error if I make the DOS paritition in a politically incorrect manner. To have it work, you probably have to make a DOS partiton under DOS (there are undocumented/poorly documented restrictions on DOS partitions, I think in my case (and yours too, since you mention starting at partition 1) it was starting it right at sector (?) 1, instead of a little ways after there like DOS wanted. Maybe it needs space at the beginning before the real partition begins for bott info/code. If they were smart, that would go *inside* the partition... Sigh. Try making DOS paritions under DOS, and Linux partitions under Linux and you should be okay. P.S. DOS really sucks when it comes to trying to create two DOS partitons too! Actually DOS fdisk is a hose in general. (I needed to when copying my DOS files from one part of the HD to another so I could optimize the placement of my Linux partitions. Not a fun experience). * Missing Operating System, how appropriate. DOS isn't an OS ;) ------------------------------ From: sct@dcs.ed.ac.uk (Stephen Tweedie) Subject: Re: Crashing machine: >16M mem patch Date: 14 Apr 93 23:37:11 GMT In article , davidsen@ariel.crd.GE.COM (william E Davidsen) writes: > In article <1qa3dg$3i8@fitz.TC.Cornell.EDU>, kutcha@eos.acm.rpi.edu (Phillip Rzewski) writes: > | > | I recently upgraded the memory on my machine from 8 megs in 1 meg > | simms to 32 megs in 4 meg simms. Naturally I would like to be able > | to use all of those 32 megs from within linux. Of course that > | meant enabling the >16M patch option in the kernel configuration. > Please check what that patch does... I can't remember if it allows > use of memory >16MB for programs, or turns on i/o to memory >16MB. > If the latter is true and you have an ISA bus disk controller which > can only address 16MB, that is likely to be the problem. As far as I am aware, (and no doubt plenty of people will be willing to jump down my throat if I'm wrong :-) the "enable >16MB memory" option of the kernel just allows the kernel to use more than 16MB of memory. It doesn't do anything special about restricting access to memory above 16MB. For hardware on the ISA bus which can only access memory below the 16MB boundary, it is the responsibility of that hardware's device drivers to ensure that this restriction is respected. All of the relevant Linux device drivers do this anyway, so although the user may see slightly less performance doing IO with the larger memory size, any incompatibilities with more memory should be safely hidden from the affected hardware. Using more than 16MB ram should be safe - the kernel configuration script was recently changed so that allowing larger memory is now the default. Cheers, Stephen Tweedie. --- Stephen Tweedie (Internet: ) Department of Computer Science, Edinburgh University, Scotland. ------------------------------ From: sct@dcs.ed.ac.uk (Stephen Tweedie) Subject: Re: HELP WITH DD Date: 14 Apr 93 23:43:21 GMT In article <1993Apr13.140544.2473@ensmp.fr>, belmouh@idefix.NoSubdomain.NoDomain (Rachid BELMOUHOUB ) writes: > I am a newbie to Linux world. And I am in the process of getting the > SLS release on sunsite.unc.edu. > I have red the FAQ. It is indicated thet one can use 'dd' on his > UNIX systeme to transfert the files directly to floppies in place of > rawrite. While reading the doc of 'dd' I found that a lot of options > can be used to do this. My question is what are the options to > create the firts two floppies a1 and a2, for the SLS release? On many Unix systems it is not necessary to use dd at all - you can just copy the data to the disk using "cp". For example, to copy a plain disk image to disk on a sun4, (where /dev/rfd0c is the raw block device associated with the 3.5" drive,) you can use cp /dev/rfd0c or even cat > /dev/rfd0c and if you have a compressed disk image, you can just use zcat > /dev/rfd0c If you have to use 'dd', then it is probably best to select a 9K block size for 3.5" disks: use dd bs=9k if= of=/dev/rfd0c or dd bs=9k < > /dev/rfd0c Try using cp or cat first - it's probably the easiest way. Cheers, Stephen Tweedie. --- Stephen Tweedie (Internet: ) Department of Computer Science, Edinburgh University, Scotland. ------------------------------ From: sct@dcs.ed.ac.uk (Stephen Tweedie) Subject: Re: New feature for the filesystems. What do you think ? Date: 14 Apr 93 23:51:11 GMT In article <1993Apr14.122356.27773@kf8nh.wariat.org>, bsa@kf8nh.wariat.org (Brandon S. Allbery) writes: > The "proper" way to do a compressed filesystem is in fact not a compressed > *filesystem* at all, it's a compressed *disk*. Which means that individual > blocks are compressed regardless of contents. Since *ix does updates on > entire blocks in all (current) cases, this imposes only a decompression speed > penalty on demand paging. Not really. For example, I might want the directory structure to remain uncomressed for speed, and I might want to have a quick compression routine apply to most files for speed but with the option of doing a slower, more thorough compression on some large, seldomly changed files. The biggest problem with a compressed disk is that as the data on the disk changes, its capacity will vary - some files can be compressed more than others. This means that either the filesystem must be aware of the disk compression system, or that the compressor be aware of the filesystem structure. For example, SuperStor and Stacker for DOS do this - they fill up the filesystem allocation table with dummy used blocks to change the apparent size of the filesystem according to the real free space left. Compression at the filesystem level offers far more flexibility, and - perhaps more importantly - it is less likely to corrupt your entire filesystem. Cheers, Stephen Tweedie. --- Stephen Tweedie (Internet: ) Department of Computer Science, Edinburgh University, Scotland. ------------------------------ From: sw@PROBLEM_WITH_INEWS_GATEWAY_FILE (Wolfgang STEGER) Subject: Re: Compiling XV; general porting problems Date: Thu, 15 Apr 1993 06:28:56 GMT Clint Jeffery (cjeffery@optima.cs.arizona.edu) wrote: > From article <1993Apr14.200018.21295@aio.jsc.nasa.gov>, by mancus@sweetpea.jsc.nasa.gov (Keith Mancus): > > I'm trying to compile xv 2.21 under Linux 0.99.6. gcc can make the > > libjpeg.a just fine, but it runs out of memory and dumps when I attempt > > to make libtiff.a. Since I have 12M RAM and a 10M swap partition, this > > must either be a gcc bug or bad options. > When I compiled it under 0.99.7A on a 16MB 486/33, the machine appeared > to "hang" while compiling files for libtiff.a. Under X, the mouse no > longer moved; outside X, I could still switch consoles, but could not > type anything at any of them. This has to be more than just gcc or > bad options. Had to hard-reset to reboot. > Fortunately, I didn't really need tiff support, so I just commented it > out of the makefile and went on my merry way... ;-) > -- > Clint Jeffery - cjeffery@cs.arizona.edu I don't have the source at hand here at work, but it was only one file in the tiff library - I think one of the CCITT support files. I just had to wait for it to compile - took about 15 to 20 min. on a 486 DX2/50 with 8 MB RAM and 16 MB swap, linux 0.99.6, gcc 2.3.3. Don't forget to arrange to build libtiff.a with -traditional option. Otherwise xv compiled cleanly out of the box. -- Wolfgang Steger Inst. f. Geologie / Techn. Univ. Wien sw@diabas.tuwien.ac.at ------------------------------ From: hlu@luke.eecs.wsu.edu (HJ Lu) Subject: Re: Searched: cc1obj Date: Thu, 15 Apr 93 05:17:02 GMT In article <1qiib4INN19g@matt.ksu.ksu.edu> probreak@matt.ksu.ksu.edu (James Michael Chacon) writes: >brauer@aix520.informatik.uni-leipzig.de (Eckard Brauer) writes: > >>Sometime before anyone announced a compiled cc1obj. I can't remember, where >>it was, if there's someone who knows, please mail me, if some has the cc1obj, >>please send me. > >>Thanks! > >I can compile cc1obj, but since there wasn't a real high demand, I didn't >see why the people who wanted couldn't just compile their own. Compiling >gcc will take ~20M or so, but this way you can configure everything the >way you want. > >H.J. might be able to be persuaded to include it with a later release of >gcc, but I really couldn't speak for him on that. > >James Object-C runtime library is changing/improving all the time since 2.3.3. I could include Object-C in the next release. But I have no idea if they work or not since I don't know Object-C. If someone could send me a comprehensive Object-C program, I would be glad to compile it before I release cc1objc. H.J. ------------------------------ From: drew@juliet.cs.colorado.edu (Drew Eckhardt) Subject: Re: Basic info on LINUX. Date: Thu, 15 Apr 1993 06:55:08 GMT In article <3V9y2B1w165w@pixie.login.qc.ca> phil@pixie.login.qc.ca (Phil) writes: > Can anybody out there answer a few elementary questions on LINUX? >I am familiar with the general concept of the system but I have quite a >few simple questions. Here goes: > >- What are the minimun hardware requirements for decent operation? ( I > have a 386DX40, 4megs and 120megs ) CPU : Linux runs fine on any processor that can run it. It's very fast, disk accesses, vi, the VTxxx console will all blaze on a lowly 386sx-16. For CPU intensive tasks, like compiling, I'd want at least a 386-33. Note that memory is more important than raw CPU speed - without enough, you'll be paging to disk that's 1000 times slower than real memory. Memory : 2M is the *bare* minimum on which Linux will run. We've run it on 2M laptops and not had any problems if we didn't compile anything (ie, just a few shells, vi, and terminal emulation). Note that with only 2M, you won't be able to install directly off of the SLS distribution. Instead, you'll need to boot single user first (ie no memory eating init, etc), make swap space, and then do your install. 4M is the *bare* minimum to do any sort of real work. With 4M, the system will be usable if you aren't running X. While gcc 1.37 ran fine in 4M with no paging, GCC has grown considerably, and will page when optomizing in a mere 4M of memory. X will swap some with the monochrome server, but will still be usable (don't run GCC at the same time!), with the larger color server, you won't be happy at all. 8M lets X run happily. You'll still have some paging when running big applications, but it won't be thrashing. With 8M (I've pared down the kernel, and turned off BIOS shadowing to gain a few hundred K) I can keep the load average > 1.0 when running X and doing builds. IMHO, you should invest in at least another 4M of memory ($100 if you buy used SIMMs). More memory will let your cache grow larger, and will mean you have even less paging, both very good for performance. Disk : It depends on how many goodies you want installed, how much swap you want, etc. If you pare it down enough, you can fit Linux on virtually any size partition. Before I got my .12 kernel happy with the bad blocks on my MFM drive, I ran it from my error free 10M partition, with all of /bin, /usr/bin, vi, gcc 1.37, 'C' libraries and other goodies. However, this isn't very realistic if you want fun toys like X, Tex, etc. To handle memory hogging programs like GCC (when compiling large initialized arrays, GCC can eat 16M of virtual memory!) you should have a reasonable amount of virtual memory, say 24M, with a bare minimum of 12M if you're running X, 8M if not. Since Linux doesn't back store, virtual memory is equal to swap space + real memory, so with 8M of memory you'd want to dedicate a single 16M partition to swapping. Linux binaries are *small* because they use shared libraries. I've run Linux + all of X11R5 + GCC + many other GNU goodies on a 40M partition (with separate swap, 5M to spare after kernel sources were installed). My current set up is 16M of swap + ~95M for Linux. I have a fairly complete installation, with X11R5, TeX, two LISP interpreters, and all sorts of goodies like Ghostscript, both byacc and bison, etc. With kernel sources, .o's and other cruft lying around, I have over 30M of free space. So, I'd say a comfortable minimum would be 40M without X and large numbers of goodies, 60M with X and more goodies, plus enough swap space to get you a reasonable amount of virtual memory. A few other comments on disk : Like memory, more is always better. If you get the urge to compile GCC yourself, you'll need over 30M of disk. A make World on the X11 tape can eat over 150M. Swapping should be to a partition, since swapping through the filesystem is much slower. However, Linux supports multiple swap files/devices, and in an emergency you can add a swap file to suplement a minimally size swap partition. >- Recommended hard disk space? See above. >- Can Linux run DOS apps? To what extent? There is a DOS emulator that can handle some real mode BIOS and DOS using programs. Things that get into hardware beyond the video frame buffer won't work, 386 programs are out. >- Is cohabitation of DOS on the same HD simple or a headache? It's extremely simple - you have separate DOS / Linux partitions, and can use a master boot record that prompts you for what to boot when you powerup. From Linux, you can also directly access your DOS files, with all necesarry translations performed on text files (ie, CR/LF combinations become newlines, etc). >- What level of technical knowledge is required to install it? With the stock SLS distribution, if everything works correctly you plug in a disk, insert the second disk, and follow the instructions. You'll need to understand partitioning, and other PC basics. To install separate goodies that have been ported, you'll need to be familiar with Unix. To install unported goodies, you may need an understanding of 'C', and Unix from a programer's level (ie, terminal handling, replacing /dev/kmem accesses in things like xload and GNU make with the appropriate /proc reads, etc). To keep things running smoothly, you should understand Unix as a user, and have some idea about system Administration - ie, where things go, what should be in /etc/rc, etc. >- Where can I find information text files via ftp? (I'm looking for basic > info, not a whole manual) tsx.11.mit.edu:/pub/linux/docs -- Boycott USL/Novell for their absurd anti-BSDI lawsuit. | Drew Eckhardt Condemn Colorado for Amendment Two. | drew@cs.Colorado.EDU Use Linux, the fast, flexible, and free 386 unix | ------------------------------ ** 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-Activists-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux) via: Internet: Linux-Activists@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 tupac-amaru.informatik.rwth-aachen.de pub/msdos/replace The current version of Linux is 0.99pl6 released on February 23, 1993 End of Linux-Activists Digest ****************************** C' libraries and other goodies. However, this isn't very realistic if you want fun toys like X, Tex, etc. To handle memory hogging programs like GCC (when compiling large initialized arrays, GCC can eat 16M of virtual memory!) you should have a reasonable amount of virtual memory, say 24M, with a bare minimum of 12M if you're running X, 8M if not. Since Linux doesn't back store, virtdigest844 644 36676 74430 57614 5366137075 5776 From: Digestifier To: Linux-Activists@senator-bedfellow.mit.edu Reply-To: Linux-Activists@senator-bedfellow.mit.edu Date: Thu, 15 Apr 93 04:45:17 EDT Subject: Linux-Activists Digest #844 Linux-Activists Digest #844, Volume #4 Thu, 15 Apr 93 04:45:17 EDT Contents: need help moving linux to a new hard drive (RAVET, STEVEN) Re: Trying to kill getty so I can dial out on ttys2; help (Kris Gleason) Re: Memory problem with Linux? (Drew Eckhardt) Re: linux logo or mascott (Jason Yanowitz) Is it possible to use OS/2 & Linux w/o Bootmanager?? (DHALIWAL BIKRAM SINGH) e2fs Problems/Queries (Mike Dowling) ** Linux Documents Explained for Newbies ** Weekly Post (J.A. MacDonald) Re: weird vi/less screen insect (Nicholas Vargish) Re: xlock... (Paul Gortmaker) Unix on pc? (Mr. Ed.) crash: 0.99p7A + emacs -> swapper killed! (Michael Will) Automated Linux PowerDown (Jerome Kaidor) ---------------------------------------------------------------------------- From: s0r1282@sigma.tamu.edu (RAVET, STEVEN) Subject: need help moving linux to a new hard drive Date: 15 Apr 1993 02:13 CST OK, heres the deal: I had two IDE drives, a 120MB and a 40MB. I had DOS stuff on the 120, so I put linux on the 40, just to look at it (no flames, please). liked it, so I bought a 240MB drive. I don't want to re-install, so how do I move it over? here are the details: i have temporarily removed the 120MB. I partitioned the 240 into a 100MB DOS partition, 140 MB linux partition. the 240 is HD0, the 40 is HD1 (the 120 was HD0, the 40 has always been HD1). i boot from my linux boot floppy, no problem. i ran fdisk to create the linux partition. its there, no problem. i changed the linux partition to type efs (extended file system). saveed changes, quit fdisk rebooted did 'mkefs /dev/hda2 140224' (140224 is the size as reported by fdisk) did 'mount /dev/hda2 /new' this is where the problem sets in, i get 'unknown error xxxxx' i also tried modifying the fstab file, to no avail. i am new to this, if it is not obvious. what i want, if someone can tell me how to get there, is: to have the 120 back as HD0, and the 240 as HD1, and have the boot floppy boot to partition 2 on the 240. i really don't want to ftp another 80 disks or whatever, so if someone can help me out i would really appreciate it. btw, i am using the SLS release from mid-march. please reply via e-mail, as i cannot keep up with this group. thanks a lot! --steve ------------------------------ From: gleasokr@rintintin.Colorado.EDU (Kris Gleason) Subject: Re: Trying to kill getty so I can dial out on ttys2; help Date: Thu, 15 Apr 1993 06:23:43 GMT jhenders@wimsey.bc.ca (John Henders) writes: >In article <1qdeh7INNfpb@senator-bedfellow.MIT.EDU> tytso@athena.mit.edu writes: >>If the getty hogs the line, then you won't be able to open the >>/dev/cua?? devices. This is NOT a bug with the /dev/cua?? devices, but >>a problem with your getty program. Get the latest, fixed getty, and >>this problem goes away..... getty_ps2.0.6b.tar.z in >>/pub/linux/sources/sbin on tsx-11.mit.edu. > This is where all the trouble is coming from. In actual fact, the >getty_ps package does not work with the 0.99pl6 SLS release. I have gotten >it succesfully running on the serial port, and as soon as it runs it grabs the >tr and rs line and never lets them go. Trying to call out will not work. >Several people have had the same experience, but when they ask for help, all >they get is this same suggestion. It's very frustrating. yes yes yes... the getty_ps2.0.6 package is BROKE! BROKE! BROKE! getty_ps2.0.7 is coming tomorrow (I hope?) I can't exactly guarantee that it will work on everybody's system, but I guess that's how it goes with free software. I can say this... I am very confident in it, have tested it a _lot_, ironed out all of the bugs that I knew of in the last release, added some new stuff, and on and on and on. It is very worth the wait. I sure hope nobody will ever have to do things like kill -1 pid; kermit to dial out ever again. Look for it tomorrow (late at night if you live in the US... that's my way) Kris -- gleasokr@rintintin.colorado.edu HARDWARE (noun): The equipment that makes up a computer system, not to be confused with software ------------------------------ From: drew@juliet.cs.colorado.edu (Drew Eckhardt) Subject: Re: Memory problem with Linux? Date: Thu, 15 Apr 1993 07:02:25 GMT In article <1993Apr14.233955.16941@jupiter.sun.csd.unb.ca> u0xh@jupiter.sun.csd.unb.ca (Chris Newton) writes: > I think there may be a memory problem with Linux. I have a 386, >33Mhz machine with 8Megs, and a 5 meg swap partition. However, when I run >about 4 instances of xv (and load a JPG in each one) with other things >going on too (like seyon, and an xterm), the machine locks up. The mouse >doesn't move, I can't switch to the other VCs or anything. This seems to be >the same problem that people are having when they try to compile xv under >gcc. It is using up all available memory, and locking. What does Linux >do when it runs out of swap space, or memory in general. It becomes extremely, extremely, extremely slow and might get all the way to dying, I've never waited around long enough to find out . Whatever it >does, it doesn't seem to be doing it correctly. Also, I had a 234K JPG >picture that would lock up the machine EVERY time I tried to view it. I >never have seen it. JPGs are compressed very highly, so I imagine this >was a very very large picture. With nothing running, when trying to view >this, it all locks up. Add more swap space - on my machine, I have 24M of virtual memory, 8M of real and 16M of swap. I had no problems compiling xv, and haven't locked the machine since I upgraded from 14M of virtual memory (4M + 10M swap). -- Boycott USL/Novell for their absurd anti-BSDI lawsuit. | Drew Eckhardt Condemn Colorado for Amendment Two. | drew@cs.Colorado.EDU Use Linux, the fast, flexible, and free 386 unix | ------------------------------ From: jmy@neural.hampshire.edu (Jason Yanowitz) Subject: Re: linux logo or mascott Date: 15 Apr 1993 02:26:36 GMT In article <2036@dsbc.icl.co.uk> KID01 writes: >In article <14APR93.09269155.0062@UNBVM1.CSD.UNB.CA> AA24@UNB.CA (pboyle@unb.ca) writes: >>My 0.02 cents to this thread. I think a lobster would be a good >>mascott for linux. I like the alliteration "linux lobster" (in >>English at least). Just as eating lobster is a special treat for >>humans, linux is an especially good os to feed your computer! >> > >And a quote :- > > Keep your hardware happy, Linux. > >Matthew Course some of think lobster is treif (non-kosher) -- or rather, we know it is, and therefor don't eat it. What halacha says about it as a mascot, I don't know. tongue in cheek, jason -- Jason Yanowitz | PGP Key available by | Know how to output a MIDI jmy@neural.hampshire.edu | fingering my account | file in C? Please email me. ------------------------------ From: dhaliwa@ecf.toronto.edu (DHALIWAL BIKRAM SINGH) Subject: Is it possible to use OS/2 & Linux w/o Bootmanager?? Date: Wed, 14 Apr 1993 23:13:50 GMT After a lot of playing around with some hard-drives I have finally settled down on a 120meg IDE, with possibly another 80mb one waiting in the wings. I have OS/2 2.1 beta currently running smoothly on the 120 meg drive with no boot manager. When I get the other drive running I want to install Linux on it, and I want to use Lilo instead of the Boot Manager (with past experience I found that Bootmanager was screwing up somewhere and not allowing me to boot Linux). Can I use Lilo as I would have under DOS? -- .......................... Bikram Dhaliwal PH: 416-351-9660 dhaliwa@ecf.toronto.edu ------------------------------ From: mike@ws.rz.tu-bs.de (Mike Dowling) Subject: e2fs Problems/Queries Reply-To: i1041301@ws.rz.tu-bs.de Date: Wed, 14 Apr 1993 19:42:06 GMT I have just finished porting my entire linux configuration from old MFM hard disks to a new AT bus disk, and am trying ext2 for the first time. Unfortunately, when turning off the kettle, by force of habit, I turned off the power at the mains, and caused my linux to crash. On rebooting, I had some strange effects. Firstly, e2fsck noted that my root partition was not ok. It corrected the problem. Calling e2fsck twice for the same partition, however, resulted it e2fsck again thinking that the fs was not ok, and again it wanted to correct it. I also used e2fsck on the /usr fs, but with the same effect. Silly of me, thought I; I should boot from a diskette and then run e2fsck on an unmounted partition. I booted from disk, and tried e2fsck on root. Again it corrected errors. The second time, though, after the checking, it reported no further errors. I then ran e2fsck on /usr. After a while, I got "out of memory" errors (I have 8 MB), and my system hanged. Pushed the reset button, rebooted from dikette. I altered /etc/fstab and /etc/rc to avoid mounting /usr automatically when booting, and then booted from the hard disk. I ran e2fsck on both partitions. After initial errors on /usr, e2fsck apparently no longer finds errors. In nevertheless never thinks that the partitions are clean, and continues to insist of checking everything in full before coming to the conclusion that things are ok. Question: How do I get e2fsck to immediately recognise that both partitions are clean? Question: Worse, is the fact the e2fsck (0.2d) does not regognise the partions are being clean an indication that permanent damage to the fs has been done? Mike Dowling i1041301@ws.rz.tu-bs.de ------------------------------ From: jay-m@equinox.unr.edu (J.A. MacDonald) Subject: ** Linux Documents Explained for Newbies ** Weekly Post Date: Thu, 15 Apr 1993 07:00:46 GMT This post is sent weekly to the comp.os.linux newsgroup after numerous people suggested I make it a regular thing. Any comments? Contact me... ======= *** Linux Documents Explained for Newbies (like me 8=) *** ======== Have any of you (relatively) new Linux users felt rather intimidated by all the *** READ THE FAQ BEFORE YOU BOTHER US GURUS *** type posts, then gone out and retrieved the numerous FAQ's (after wasting a lot of time filtering through all the docs etc), read them, then still been flamed for not finding something in the FAQ? Well, I kinda felt that way, so I am compiling a very brief (and hopefully understandable) list of what the various docs are that you may need and why. Here it is: ======================================================================= I. Where to find them: - most of the docs are posted to comp.os.linux (c.o.l) every couple of weeks. If you are patient you'll see them. - By ftp try: tsx-11.mit.edu (18.172.1.2) in /pub/linux sunsite.unc.edu (152.2.22.81) in /pub/Linux nic.funet.fi (128.214.6.100) in /pub/OS/Linux - By mail server try: mailserver@nic.funet.fi mail-server@pit-manager.mit.edu ftpmail@decwrl.dec.com - If none of those work then try asking Ian Jackson (who's post *** READ THIS BEFORE POSTING *** should also be read) by email (ijackson@nyx.cs.du.edu) II. What document files are available and should you get: - INFO-SHEET - This file explains what Linux is, its features, hardware supported, some software ported, and (very important) where to get Linux via ftp or snail mail. (in docs/) - META-FAQ - Lots of useful information on where to find more information about Linux (get this one for sure). (in docs/) - faq.p1.mmmyy - The FAQ itself comes in 4 parts. This is the - faq.p2.mmmyy mother of all Linux documentation. These four - faq.p3.mmmyy files are a *must* if you plan to use Linux. - faq.p4.mmmyy (The FAQ is updated regularly, hence the date extension. e.g. mmmyy = feb93) (in docs/FAQ) - faq.toc.mmmyy - Table of Contents for the FAQ. A must-get in order to avoid massive amounts of time waste picking through the four FAQ volumes. (in docs/FAQ) - NET-FAQ - Documentation on networking Linux. This is in addition to Chapter XI of the FAQ. (in docs/) - LILO-QUICKSTART-1.1 - How to install LILO (the disk booting program). (in docs/) - FAQ - GCC FAQ (rather poorly named IMHO). (in packages/GCC) - ext2fs.faq - FAQ regarding the ext2fs utility (i.e. the second version of the extended filesystem program). (in ALPHA/ext2fs) - drivers.doc.z - a gzipped file containing all the information you you should need to know for writing a driver for Linux. Read this before you even think about writing your own driver. (in ALPHA/drv_guide) - bbs.list - List of various BBS and ftp sites where you can find Linux files. iii. Information for SLS!!! (every newbie needs this 8-) - DOWNLOADING - Information on how to get the SLS disks. (in packages/SLS) - NOTICE - Information on warranty, restrictions, etc for SLS. Check this one out. (in packages/SLS) - COPYING - The GNU General Public License document. (in packages/SLS) - RELEASE - Information on what is in the latest release of SLS. (in packages/SLS) - ChangeLog - History of changes to Linux and SLS. (in packages/SLS) - README - Most important document on SLS installation etc. You *will* need this for installing SLS. (in packages/SLS) - SLS.FAQ - The real thing. Two pages of questions you will come come across (complete with answers, too 8-). (in packages/SLS) iv. Comp.os.linux.announce (c.o.l.a) information: - The newsgroup comp.os.linux.announce is a moderated newsgroup where the *latest, greatest* stuff regarding Linux is posted. - C.o.l.a [I love that acronym for this group 8-)] is archived in several places: - sunsite.unc.edu:/pub/Linux/docs/linux-announce.archive - nic.funet.fi: /pub/OS/Linux/doc/news/COLA ===================================================================== From what I've seen so far there is a lot of information contained in these documents. Also examine every README type file you find (e.g README.tapes explained all I needed to know to get my SCSI tape stuff going). A nice trick for finding these files is to get the find-ls file from the archives then grep it for README, FAQ, DOC, etc. i.e. grep README find-ls | more For information on non Linux specific stuff check the corresponding newsgroup for FAQ information (e.g. comp.unix.questions, comp.lang.c, etc). Also, all the Internet FAQs are archived on rtfm.mit.edu. Check it out if you have questions about other subjects. BTW, for those of you with ftp access, here's a neat little trick you may not know about. To view a file on the archive without first transfering it to your machine type: ftp>get filename |more NOTE: NO SPACE BETWEEN THE | AND more!!! This is really handy for looking at README files etc, and general file browsing. HINT: The unix grep command is *REALLY* useful when looking for specific information in the documents! 8-) Cheers! Jay MacDonald <- Linux convert Geosense Consulting Reno, NV P.S. If I've missed any obvious docs or docs that anybody thinks should be included, please let me know. [newbies.autopost - last changed April 8th, '93] -- ============================================================================== jay-m@equinox.unr.edu | The purpose of the bass player is to help his jaym@thoreau@flame.ctc.fmc.com | audience surf the wave of low frequency sound ============================================================================== ------------------------------ From: vargish@bogus.sura.net (Nicholas Vargish) Subject: Re: weird vi/less screen insect Date: 14 Apr 1993 17:23:35 -0400 In article thoth@uiuc.edu (Ben Cox) writes: >oreillym@tartarus.uwa.edu.au (Michael O'Reilly) writes: >>As yet, I haven't managed to find a patch for bash that will fix the >>problem. anyone??? >It's called tcsh... :-) Which of course has its own bugs. The fix for all the above is: zsh. :^) Nick p.s. It's not flame-bait... really! Though if anyone is actually interested, I'll be happy to tell them why _I_ like zsh. -- | Nick Vargish | | SURAnet Operations | Who needs a fancy .signature when I have all this? | vargish@sura.net | ------------------------------ From: rcopg@minyos.xx.rmit.OZ.AU (Paul Gortmaker) Subject: Re: xlock... Date: 15 Apr 1993 01:43:23 GMT u0xh@jupiter.sun.csd.unb.ca (Chris Newton) writes: [stuff deleted.] >for this, or is there one? Cant xlock be modified to not allow jumping to >other VCs? Just intercepting the right key strokes would be all there is to >it (ctrl-alt-Fn), right? Personally, I have no idea... but it sounds that >simple. ...allowing one to go to other vc's *is* the correct behaviour! What happens if xlock won't read your passwd for some reason, and you can't log in to kill it on another vc ???? If you don't like having the vc still logged in then use "startx&logout" or whatever. Then your system will be fully secure. (...except for someone with a bootable rootdisk ;-) Paul. ------------------------------ From: edb866t@mings2.cc.monash.edu.au (Mr. Ed.) Subject: Unix on pc? Date: Thu, 15 Apr 1993 02:43:37 GMT [ Article crossposted from comp.os.msdos.apps ] [ Author was Dave ] [ Posted on 12 Apr 93 14:37:19 GMT ] cakir@unixg.ubc.ca (Anil Yer-Su Cakir) writes: {> Is there a freeware version of unix to enable a dos pc user to convert his co LINUX is a UNIX clone. Its either freeware or very cheap shareware. from what I have read about it, it is pretty good. I am currently looking for availability, where can I get the latest version ? either by FTP or a dial up system. -David =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= China Cat BBS c-cat!david@sed.csc.com (301)604-5976 1200-14,400 8N1 ...uunet!mimsy!anagld!c-cat!david =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= -- If I am a figment of your imagination and you a figment of mine, then what AAAARGHH!!! this is getting me confused!!! ------------------------------ From: michaelw@desaster.hanse.de (Michael Will) Subject: crash: 0.99p7A + emacs -> swapper killed! Date: Wed, 14 Apr 1993 17:35:33 GMT Hello, I was just downloading my news with uucp, replied to a message with elm+emacs, when suddenly my machine got killed in an unusual way. general protection: 6e70 EIP: 0008:00D152 EFLAGS: 00010002 int = 58 9a 37 2a a8 c6 0e 3e 3a c9 cf 40 46 08 b9 39 fs: 0017 base: C000000 limit: 00200000 Pid: 0, process No. 0 cf: 00 00 00 00 00 00 00 00 task[0] (swapper) killed: unable to recover kernel panic: Trying to free up swapper memory space in swapper task - not syncing The machine stopped - I had to reboot cold. I used 0.99p7A, with scsi tape+disk enabled, a root-partition of 120MB efs on dev/hdb1 2 partitions e2fs on hda1 1 partition msdos on hda1 1 swappartition (happens to be 10MB because I was lazy). 1 DOS-Partition (for games only :-) i386-dx-33 8MB RAM - 2x120MB-At-bus-drives, 1-scsi-tape 'Hope that helps anyone... Michael Will -- -- pgp Key fingerprint = 58 9A 37 2A A8 C6 0E 3E 3A C9 CF 40 46 08 B9 39 Happily using Linux 0.99p7A with X11R5, \LaTeX, cnews/nn/uucp and... pgp! >>> Ask for Linux and / or pgp-Information <<< ------------------------------ From: jkaidor@synoptics.com (Jerome Kaidor) Subject: Automated Linux PowerDown Date: 14 Apr 93 21:51:51 GMT Reply-To: jkaidor@synoptics.com A few weeks ago, I cut my home uucp node over to linux. Goodbye, Dos! And good riddance. Now uucico runs off a crontab that periodically checks for outgoing mail, occasionally polls for incoming mail, and in general, chugs away in the background. I even turned the sound off on the modem, so that uucico's entry onto the PSTN is announced by a mere *click*. Just like a real network! Now, for my purposes, linux has only one disadvantage, as compared to dos. I'd like to address it: with DOS, you can generally dump the computer's power any time that you see that dos prompt. With any unix, you have to be more careful... Why is this important? Previously, my computer was on a 24-hour timer, which turned it on at 3:00 in the morning ( or thereabouts ), and turned it off at 5:00 AM ( or so ). The intervening two hours were enough for the machine launch uucico and pick up my mail. Then during the day, I would turn the computer on manually, and do my usual stuff. According to my calculations, leaving the computer on 24 hours/day will eat $10-$20 worth of electricity. I, personally, have no problem with this, but my wife might well disagree. Accordingly, I have thought up the following hardware/software solution: 1) I build a box that allows the computer to turn itself *off*. 2) I let the 24-hour timer turn the computer *on*. 3) A cron script turns the computer off if: a) The computer has been on for a few hours ( or maybe, that the automated nighttime tasks are finished ) b) It's early in the morning ( 4:00 AM or so ) c) There are NO USERS ( nobody logged in ) Of course, a software-only solution is possible, where a crontab just syncs and halts the computer when the conditions are met, but I don't like this because: 1) The computer and timer could get out of sync, and the timer could power-down in the middle of something. 2) The computer would always be turned on much longer than necessary to do its actual work. For the hardware part of this solution, I prefer to keep it simple. I would still use the mechanical 24-hour timer, but set it so that it only supplied power for a little while, say, 15 minutes. It would plug into a box, built by me, that would contain a relay to bypass the timer as soon as power came on. The relay would be turned off by a circuit that looked at an output of the computer. But what output? I was thinking of using one of the DTR lines on the serial ports. Trouble with them, is that you really don't know what will happen to them during bootup, etc. One glitch, and BZZZzzzzz*. My solution to the problem of unknown states and glitches is to place some very specific requirements on the computer's output. Specifically, a simple circuit consisting of a comparator, a couple of capacitors, a resistor or two, and a diode, would make the turn-off circuit only trip when a large group of *pulses* came out of the computer. Now, it's probably not usual to see a couple thousand pulses come out of a DTR output, is it? Anybody have any comment on all this? I'm sorry, it's not totally Linux-specific, but I figured that some other linux'ers out there might be interested, too.. I don't think my situation is all that unusual. - Jerry Kaidor ( tr2!jerry@dragoman.com, jkaidor@synoptics.com ) ------------------------------ ** 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-Activists-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux) via: Internet: Linux-Activists@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 tupac-amaru.informatik.rwth-aachen.de pub/msdos/replace The current version of Linux is 0.99pl6 released on February 23, 1993 End of Linux-Activists Digest ****************************** ynoptics.com A few weeks ago, I cut my home uucp node over to linux. Goodbye, Dos! And good riddance. Now uudigest845 644 36676 74430 55527 5366137075 6000 From: Digestifier To: Linux-Activists@senator-bedfellow.mit.edu Reply-To: Linux-Activists@senator-bedfellow.mit.edu Date: Thu, 15 Apr 93 06:15:28 EDT Subject: Linux-Activists Digest #845 Linux-Activists Digest #845, Volume #4 Thu, 15 Apr 93 06:15:28 EDT Contents: Binaries on line (was Re: A few question, mostly about pl8) (A. V. Le Blanc) kermit weirdness...? (Patrick K. Ferrick) FAQs - which group should we post to ? (Vince Skahan) Re: crash: 0.99p7A + emacs -> swapper killed! (KID01) Re: NFAQ: Do ISA cards cripple EISA performance ? (Cameron L. Spitzer) Re: color xterm (Charles Hedrick) Re: how safe is minix filesystem in power failure? (Charles Hedrick) Re: Possible gotcha in gcc (Michael O'Reilly) Re: Big swap partitions [> 16M] (Wolfgang Jung) Looking for MIT-scheme (janosch-usenet) Re: SCSI: Driver Needed. (Mike Frankowski) bug in .99p8A char device drivers? (was: bad SIMM?) (Eric Jeschke) Re: Searched: cc1obj (James Michael Chacon) efs fun... (David Lesher) Re: The dangers of playing with shared libraries (Brandon S. Allbery) Re: Has anyone run Minix 1.5 in the DOS emulator? (Drew Eckhardt) Re: Why Linux should remain Linux and seagulls should not be used. Re: X window GUI Builder for linux? ---------------------------------------------------------------------------- From: zlsiial@uts.mcc.ac.uk (A. V. Le Blanc) Subject: Binaries on line (was Re: A few question, mostly about pl8) Date: 15 Apr 93 07:48:05 GMT Reply-To: Le Blanc@mcc.ac.uk (A. V. Le Blanc) In article sct@dcs.ed.ac.uk (Stephen Tweedie) writes: >> 3) Now that I have upgraded my kernel to pl8, I assume that there >> are (or will be) updates to some of the affected bins. Most of mine are >> probably quite old. Any idea what ftp sites I should check and which >> commands need to be updated? Actually, the quote was from someone else, but I haven't any idea who. I should like to mention that there is available for anonymous ftp a large collection of Linux binaries (in fact a whole system). This is at ftp.mcc.ac.uk [130.88.203.12] in /pub/linux/binaries. The area is actually NFS mounted by several local Linux installations to save on disk space. If you want a single command, you might want to check there. However, (1) these commands are sometimes compiled with hard-coded pathnames which may not suit your taste; for example, shutdown looks for /bin/umount, not /etc/umount, and (2) this area is constantly being updated, so please don't assume it will remain the same from day to day. -- Owen LeBlanc@mcc.ac.uk ------------------------------ From: ferrick@acsu.buffalo.edu (Patrick K. Ferrick) Subject: kermit weirdness...? Date: 15 Apr 93 02:50:19 GMT Every once in a while, while logged into a remote unix host via kermit my term goes whacko...everything is still there, except most of the characters come out as dots and vert/horiz lines...! The VC in question stays that way until I reboot. None of the others are affected. Has this been discussed to death yet? If so, sorry...I don't have the bandwidth to scan through as many posts as I'd like :-( tnx, pat ------------------------------ From: vince@victrola.sea.wa.us (Vince Skahan) Subject: FAQs - which group should we post to ? Date: Thu, 15 Apr 1993 05:32:55 GMT I maintain a periodic posting of "Linux uucp/news/mail FAQs" that I have a deamon posting twice per month to comp.os.linux... My question is "should this be posted to c.o.l. or to c.o.l.a?" to make it more visible ? It might make it easier for people to find them if we could agree on some kind of unique Subject that might appear to tell people that this is a periodic posting of FAQ-type material. (in my case, it's a linux-specific FAQ that I personally feel should *not* be cross-posted to news.answers). Also, I have some great perl-based code called 'post_faq' that automates this procedure wonderfully...if anybody else is doing so by hand, you need to query archie and grab it :-) -- ---------- Vince Skahan --------- vince@victrola.sea.wa.us ---------- +++ A Waffle Iron - Linux Division +++ ------------------------------ From: KID01 Subject: Re: crash: 0.99p7A + emacs -> swapper killed! Date: 15 Apr 93 07:59:29 GMT In article <1993Apr14.173533.410@desaster.hanse.de> michaelw@desaster.hanse.de (Michael Will) writes: >Hello, > >I was just downloading my news with uucp, replied to a message with elm+emacs, >when suddenly my machine got killed in an unusual way. > >general protection: 6e70 >EIP: 0008:00D152 >EFLAGS: 00010002 int = 58 9a 37 2a a8 c6 0e 3e 3a c9 cf 40 46 08 b9 39 >fs: 0017 >base: C000000 limit: 00200000 >Pid: 0, process No. 0 >cf: 00 00 00 00 00 00 00 00 >task[0] (swapper) killed: unable to recover >kernel panic: Trying to free up swapper memory space >in swapper task - not syncing > I get a similar error when I boot Linux from floppy (SLS). 486/33,8meg,100ide pl6 Cheers Matthew -- Matthew Roderick -*- B2 Developments -*- mat@{dsbc,oasis}.icl.co.uk +44 [0]782 771000 x3270 -*- UNIX Centre, ICL, Stoke-on-Trent, England. Spelling & typing errors have been left for orthentisity ------------------------------ From: cls@truffula.sj.ca.us (Cameron L. Spitzer) Subject: Re: NFAQ: Do ISA cards cripple EISA performance ? Date: Thu, 15 Apr 93 07:15:17 GMT In article <734722283snx@finbol.toppoint.de> jschief@finbol.toppoint.de (Joerg Schlaeger) writes: > > alecm@coyote.uk.sun.com writes in article <1phdksINNbcu@uk-news.uk.sun.com>: > > > > I've heard from a friend that if you have an EISA machine, and install > > an ISA card of any sort, the bus performance will drop to that of an > > ordinary ISA bus. >yes, it's true and depends on the chipset and layout of your motherboard. >Some motherboards may be unable to make a different timing for the >EISA-Slots withISA-Cards in. >They detect the ISA-CARD and do ISA-Cycles for all or a group of EISA-Slots. >(this means you get the timing like org. 8MHz AT ISA-Bus) I've heard the rumor expressed above, but it doesn't make sense to me. An EISA motherboard begins each bus cycle with EISA signalling, hoping it will select an EISA adapter. The selected EISA adapter sees its address, and it asserts the signal EX32* or the signal EX16* to say "Here I am, an EISA adapter! Let's do a fast EISA cycle!" If there's no EX16* or EX32* pulse, the motherboard says "Oh rats, I have to do a slow ISA cycle for this crummy old ISA adapter I appear to be talking to." An AT-style ISA adapter asserts the signal IOCS16*. If there's no pulse on any of those signals, the motherboard says "Wow, it's a dusty old XT-style adapter, I'll have to do a r-e-a-l-l-y s-l-o-w 8-bit cycle!" I've never heard of a motherboard that remembers from one cycle to the next what kind of cycle each slot prefers. The adapters recognize their addresses and say what they are, on each access. If you know of an EISA motherboard that doesn't start each cycle with EISA signals, please mail me. A motherboard that tries to do an ISA cycle to an EISA adapter because there's an ISA or XT adapter in *another* slot is seriously broken. An adapter that asserts IOCS16* when it's not selected is broken also. (To save cost, some EISA adapters perform ISA cycles when their non-critical resources are selected, but that's irrelevant here.) That's why I can't understand this rumor about ISA boards slowing EISA boards down. Cameron Spitzer in San Jose ------------------------------ From: hedrick@geneva.rutgers.edu (Charles Hedrick) Subject: Re: color xterm Date: 15 Apr 93 03:00:07 GMT In article hedrick@geneva.rutgers.edu (Charles Hedrick) writes: >There are two ways to do color in xterm. One defines the standard >ansi sequences for setting colors. The other lets you define colors >to implement normal VT100 attributes (boldface, underline, etc). A Brandon S. Allbery bsa@kf8nh.wariat.org says >...shades of the Wyse 350 *gag*... It depends upon what you want to do. The problem with the ANSI sequences is that you have to hardcode colors in the termcap entry. Different users have different tastes, and some even want to use different colors on different machines. While it's possible for a user to set up a private termcap, I'd rather have a standard termcap, and then have people decide what color they want for standout, bold, etc. by setting up definitions in .Xdefaults. If you're using the ANSI sequences, you'd probably set up standard colors in termcap, and then let people make changes in .Xdefaults so that the ANSI sequence for blue really generates red, etc. I find that counter-intuitive. ------------------------------ From: hedrick@geneva.rutgers.edu (Charles Hedrick) Subject: Re: how safe is minix filesystem in power failure? Date: 15 Apr 93 03:14:12 GMT jim@n5ial.mythical.com (Jim Graham) writes: >like the subject says.... how safe (I realize this is all on relative >terms, here) is the plain minix filesystem (all I've got right now) in >cases where, say, the power fails? If it's a single user machine, and the user is asleep (which is the case you describe), it's almost completely safe (assuming you're not doing things in the background). update will do a sync every 30 sec., so 30 sec. after the last file activity things will be fine. On a multi-user machine, or one that runs batch jobs things get more complex. But on my home machine, I commonly hit reset or power it off without any shutdown or sync. I've never hard any trouble. (I don't recommend that for novices. It works for me because (1) I have a very simple setup. I run no daemons other than update and getty, (2) I don't leave things running in the background.) >even touching the filesystem. this has happened about 3 times now, and no >damage. can I assume from this that the filesystem is considerably more >tolerant to this sort of thing than other UN*Xes? or have I just been >lucky? I think you're seeing the difference between a simple home machine and a multi-user system. The more interesting question is what happens if there's a power failure while you're actively working. (The only relevant work would be something that's causing files to change.) BSD (and presumably other systems) are careful about the order in which they do updates on disk. If they crash, the file system may be inconsistent, but the kinds of inconsistency that are possible are carefully controlled, and fsck can recover from them all (though possibly by removing files that were in the process of being created when the power failed). I've not seen a clear statement about which Linux file systems (if any) do this. ------------------------------ From: oreillym@tartarus.uwa.edu.au (Michael O'Reilly) Subject: Re: Possible gotcha in gcc Date: 15 Apr 1993 03:07:58 GMT william E Davidsen (davidsen@ariel.crd.GE.COM) wrote: : I noted an interesting behavior in gcc last night, it doesn't ignore : stuff which is #ifdef'd out! I got a program to port to Linux, and I had : some text included which I found valuable. Having learned long ago that : putting /*...*/ around arbitrary stuff might be a problem due to */ in : the text, I put #if 0 ... #endif around it instead. A gcc generated : error messages about the text it should have been ignoring! : Obviously the preprocessor has to scan for #commands, but it shouldn't : be complaining about other things in the text, like over-long strings, : etc. This *may* be an enhancement of the Linux version, since the : V.4/386 and Sun versions of gcc don't complain about the problem. Are they the same version of GCC? : Therefore I want to check it carefully before I report it to the : developers as a generic bug in gcc. Read the ANSI standard. The C preprocessor deals with TOKENS!! It is a requirement that your program be entirely composed to legal tokens, if they have been #if 0'ed out. GCC comforms where possible and sensible to the ansi standard. Check the various FAQ's. I think this is mentioned in the comp.lang.c FAQ. : bill davidsen, GE Corp. R&D Center; Box 8; Schenectady NY 12345 Michael ------------------------------ From: wong@cs.tu-berlin.de (Wolfgang Jung) Subject: Re: Big swap partitions [> 16M] Date: 15 Apr 1993 08:38:32 GMT David Baggett (dmb@case.ai.mit.edu) wrote: : I've checked the FAQ for this, but no luck: : I'm running 0.99.8 and I made a 32M swap partition. mkswap tells me : that it's allocated that much room, but swapon tells me it's only added : 16M, and free corroborates this. Is this a hard limit, or can I tune : some kernel param to get the whole 32M? (BTW, I have 8M of physical : memory, if that makes a difference.) This is a FAQ (I asked it once too :-) )but is it already on the FAQ Lists ?! Answer: use several swappartitions or swapfiles with each <= 16M ! Q: How many Swapareas is Linux able to handle ? Gruss Wolfgang ------------------------------ Crossposted-To: comp.lang.scheme,comp.lang.scheme.c From: news@janosch.toppoint.de (janosch-usenet) Subject: Looking for MIT-scheme Date: Thu, 15 Apr 1993 08:40:31 GMT I'm looking for MIT-Scheme running with Linux(!). I would appreciate if s.o. could tell me (ftp-adress is fine) where I could such a Linux-Implementation. Thank you in advance, VK ------------------------------ From: mfrankow@piper.hamline.edu (Mike Frankowski) Subject: Re: SCSI: Driver Needed. Date: Thu, 15 Apr 1993 02:52:07 GMT Terry Evans (tevans%sunset.cs.utah.edu@cs.utah.edu) wrote: : I just bought a computer and It came with a Data Technology : EISA SCSI Host Adapter. The model is DTC3292. I have tried : everyhting that I know of and I can't get Linux to work : with it. Is there anybody out there in the Linux world : that knows how to use the Controller with Linux??? : Any help, comments, suggestions, etc would be greatly : appreciated. Should I get another board? : Thanks! : tevans@cs.utah.edu Put it into 1542 compat mode and it will work just fine. Mike ------------------------------ From: jeschke@cs.indiana.edu (Eric Jeschke) Subject: bug in .99p8A char device drivers? (was: bad SIMM?) Date: 15 Apr 93 04:02:03 GMT I've noticed a couple of posts recently about crashes that look suspiciously like memory errors (one of them was mine :-). Well I did some more sleuthing and determined that it was not my SIMMs or cache card (by swapping, disabling cache, etc). So I turned my attention to Linux. I can now reproduce the error consistently in one of two ways: [1] Log in, type a few characters, use the back arrow key or ^B to move back into the line a bit, hit backspace---BOOM. General protection error, crash. This happens with BOTH bash and tcsh. [2] Log in, type "man mv" (or "cat" a huge file, or any command that spews lots of data to the screen)--BOOM. General protection error, crash. (Note: no error if I redirect output; i.e. "man mv 2> xxx" ) This leads me to believe that there may be a problem with the keyboard handler/ character device drivers-- possibly buffers getting overrun and wiping out other kernel data structures. The crashes are mostly noticable in .99p8A, serial driver 3.94. If I boot from a .99p7 (s.d. 3.93) both [1] and [2] work fine. I am using libc.so.4.3.3, although some of my binaries are of .99p5 vintage, I think. I do have the kernel compiled for memory >16Mb, although I only have 8Mb installed. I will keep on narrowing this down, but I'd appreciate any comments or speculation from experienced Linuxers. My HW configuration (not that it matters much): 486DX, 8Mb, Micronics VL-B motherboard. Eric Jeschke jeschke@moose.cs.indiana.edu ps. one other question, while I'm posting. "ps" (ver .99p6) broke when I upgraded from .99p7 to .99p8. I have a proc entry in my fstab, although I gather that "ps" does not use that, but rather "kmem". I did a "ps -U" after rebooting off the new kernel, but ps doesn't show anything right now. Any hints? Thanks! -- Eric Jeschke | Indiana University jeschke@cs.indiana.edu | Computer Science Department ------------------------------ From: probreak@matt.ksu.ksu.edu (James Michael Chacon) Subject: Re: Searched: cc1obj Date: 14 Apr 1993 21:47:00 -0500 brauer@aix520.informatik.uni-leipzig.de (Eckard Brauer) writes: >Sometime before anyone announced a compiled cc1obj. I can't remember, where >it was, if there's someone who knows, please mail me, if some has the cc1obj, >please send me. >Thanks! I can compile cc1obj, but since there wasn't a real high demand, I didn't see why the people who wanted couldn't just compile their own. Compiling gcc will take ~20M or so, but this way you can configure everything the way you want. H.J. might be able to be persuaded to include it with a later release of gcc, but I really couldn't speak for him on that. James ------------------------------ From: wb8foz@skybridge.SCL.CWRU.Edu (David Lesher) Subject: efs fun... Date: 15 Apr 1993 03:57:37 GMT Reply-To: wb8foz@skybridge.scl.cwru.edu (David Lesher) I've been having fun with filesystems today. ;-} Mount and others complained about my /dev/sda6 partition. That's a 274432-block extended partition, with the efs filesystem on it. (Yes, I *know* I should have efs2, or xiafs, but I'm waiting for the new SLS..) Anyhow when I looked at it with fdisk, it claimed the partition was -257394672983 or thereabouts, blocks long. But since the only file (a BIG tar called SLS.tar) on it was safe on tape, I used fdisk to delete the partition, and rebuilt it. All was fine. Get out the tape, recover the file. Then I played around with iozone for a while. Iozone builds long sequential files & reads them, then tells you how long it took. Not being one for half measures, I let it build a 200 meg test file. Gee, the response sure slows while THAT runs ;_} Later, guess what, more garbage. What a bore. Unmount, efsck said every block was not empty when it was, or the other way around. -a the thing, and found the result was no file. fdisk, delete, new, exit, reboot, mkefs again, remount. Guess what? The tar file was INTACT. Through the partition, & the mkefs. Hmm. q1) Is the efs, or extended partitions on SCSI, noted for this corruption on large files? Yes, I sync before booting, and brush after every meal. q2) How'd that file survive? q3) When will the all-new SLS, with me-proof filesystems, be available in a store nearby ;-? -- A host is a host from coast to coast..wb8foz@skybridge.scl.cwru.edu & no one will talk to a host that's close............(301) 56-LINUX Unless the host (that isn't close).........................pob 1433 is busy, hung or dead....................................20915-1433 ------------------------------ From: bsa@kf8nh.wariat.org (Brandon S. Allbery) Subject: Re: The dangers of playing with shared libraries Date: 14 Apr 93 21:42:48 GMT In article <1993Apr14.162759.28564@kf8nh.wariat.org> bsa@kf8nh.wariat.org (Brandon S. Allbery) writes: >In article <1993Apr14.032021.12691@ksmith.com> keith@ksmith.com (Keith Smith) writes: > while read line; do echo line; done < infile > outfile "$line" I should know better than to rush a post to avoid being late back from lunch. :-( ++Brandon -- Brandon S. Allbery bsa@kf8nh.wariat.org It's not too late to turn back from the "Gates" of Hell... Linux: the FREE 32-bit operating system, available NOW. Why waaaaaait for NT? ------------------------------ From: drew@juliet.cs.colorado.edu (Drew Eckhardt) Subject: Re: Has anyone run Minix 1.5 in the DOS emulator? Date: Thu, 15 Apr 1993 03:56:56 GMT In article <1qhbm8INNh26@tamsun.tamu.edu> n217cg@tamuts.tamu.edu (Scott Taylor) writes: >Due to an end-of-semester project in my operating systems class, I need to run >Minix 1.5 on my PC. Rather than having to reboot whenever I need to work on >it, I thought it would be really cool to run it in the DOS emulator. The >problem is that even the universal boot diskette (which only uses BIOS calls to >access the HW) won't boot in dosemu 0.48; I suspect that all the BIOS calls >that Minix wants aren't there. I also had an end-of-semester project (add symbolic links to Minix) in my operating class, needed to run Minix 1.5 on my PC, and looked at doing the same thing. I looked at the device drivers, and discovered that while there's a disk driver (kernel/bios_wini.c) that uses the BIOS only (ie, for SCSI disks, etc) all of the other drivers (console, serial, etc) all talk directly to the hardware. I figured that virtualizing Minix (either by hacking DOS emu to do the appropriate hardware emulation, writing BIOS using device drivers, or hacking Minix to run as a *normal* Linux process) would add more time to the project than a virtualized Minix would save, and finished the project in eight hours (four hours of writing code, four of debugging, mostly tracking down a int / off_t switch that lint / gcc would have caught) editing the Minix source from Linux under 'X', rebooting to try things out. > >Has anyone successfully done this? I don't have a lot of time right now to >hack dosemu (whatever Linux hacking time I have I intend to spend on the Ultra- >Stor 14F/34F driver), but if anyone has any "fast & easy" tips, I would really >appreciate it! > It's non-trivial. -- Boycott USL/Novell for their absurd anti-BSDI lawsuit. | Drew Eckhardt Condemn Colorado for Amendment Two. | drew@cs.Colorado.EDU Use Linux, the fast, flexible, and free 386 unix | ------------------------------ Date: Wednesday, 14 Apr 1993 13:11:26 EST From: Subject: Re: Why Linux should remain Linux and seagulls should not be used. How about a Tiger? A Tiger is powerful, fast, sleek, efficient, and so is Linux. And also, Linus is a roarin' good programmer:) But seriously, why not go for a a really cool image like a tiger instead of a seagull, which is regarded as a rodent with wings by some. (not me). Seagulls are not that bad an idea though, especially if they are in mid-flight. -mc. ------------------------------ Date: Wednesday, 14 Apr 1993 13:17:17 EST From: Subject: Re: X window GUI Builder for linux? You can also try ibuild that comes with the interviews package, if you don't mind (like) C++. -mc. ------------------------------ ** 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-Activists-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux) via: Internet: Linux-Activists@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 tupac-amaru.informatik.rwth-aachen.de pub/msdos/replace The current version of Linux is 0.99pl6 released on February 23, 1993 End of Linux-Activists Digest ****************************** read line; do echo line; done < infile > outfile "$line" I should know better than to rush a post to avoid being late back from lunch. digest846 644 36676 74430 53761 5366137076 6000 From: Digestifier To: Linux-Activists@senator-bedfellow.mit.edu Reply-To: Linux-Activists@senator-bedfellow.mit.edu Date: Thu, 15 Apr 93 07:45:18 EDT Subject: Linux-Activists Digest #846 Linux-Activists Digest #846, Volume #4 Thu, 15 Apr 93 07:45:18 EDT Contents: Re: weird vi/less screen insect (HJ Lu) Re: The dangers of playing with shared libraries (q.p.liu) Error codes (Ben J Fornshell) Re: New feature for the filesystems. What do you think ? (Bill C. Riemers) Re: Is there a .au player for soundcards? (Andrew J. Cosgriff) Linux tty's (Leo T. Hippelainen) Re: Linux on Macintrashes? (Steve Madsen) Linux swapping (Yap Kid Mun) Re: kermit weirdness...? (root-bound) Compatibility question (Christopher Collins) Re: bug in .99p8A char device drivers? (was: bad SIMM?) (Eric Jeschke) Re: Help on X-Windows(mostly) (Drew Eckhardt) Re: Linux on Amiga? (Richard L. Goerwitz) CD-ROM question. (Tankgirl) HELP!!! with INET Probs (Steve Goyette) Some question? (H. J. Hong) Tutorial for setting up lpr/lpd????? (Steve Fuller) Re: QIC-02 Tape Patches? (Ton van Rosmalen) Re: Trying to kill getty so I can dial out on ttys2; help (John Henders) cp from diskette to HD ? (Joel Ratsaby) Re: Big swap partitions [> 16M] (Edward Seidl) ---------------------------------------------------------------------------- From: hlu@luke.eecs.wsu.edu (HJ Lu) Subject: Re: weird vi/less screen insect Date: 15 Apr 93 07:26:12 GMT In article <1993Apr15.061939.24400@bilby.cs.uwa.edu.au> oreillym@tartarus.uwa.edu.au (Michael O'Reilly) writes: >Ben Cox (thoth@uiuc.edu) wrote: >: oreillym@tartarus.uwa.edu.au (Michael O'Reilly) writes: > >: >As yet, I haven't managed to find a patch for bash that will fix the >: >problem. anyone??? > >: It's called tcsh... :-) > >Sorry. I want to use a real shell. Also, I can't exactly link tcsh to >/bin/sh and have everything work. > Try pdksh. I link it to /bin/sh. I have fixed two bugs in pdksh 4.8 and reported them to the author. But I did get any reply. BTW, kernel 0.99.8 should fix ^Z bug in bash. H.J. ------------------------------ From: qpliu@phoenix.Princeton.EDU (q.p.liu) Subject: Re: The dangers of playing with shared libraries Reply-To: qpliu@princeton.edu Date: Wed, 14 Apr 1993 19:02:47 GMT In article <1993Apr14.095640.28923@nessie.mcc.ac.uk> LeBlanc@mcc.ac.uk writes: >In article <1993Apr13.025258.16555@Princeton.EDU> qpliu@princeton.edu writes: >>while read x; do echo "$x"; done < file >> (doesn't work correctly with bash, but does with pdksh) >I tried this with bash and had no problems. I am using bash-1.12 compiled >with libc-4.3.3 and gcc-2.3.3. I suppose I should have thought to use: (IFS=; while read x; do echo "$x";done sct@dcs.ed.ac.uk (Stephen Tweedie) writes: >In article <1993Apr14.122356.27773@kf8nh.wariat.org>, bsa@kf8nh.wariat.org (Brandon S. Allbery) writes: >> The "proper" way to do a compressed filesystem is in fact not a compressed >> *filesystem* at all, it's a compressed *disk*. Which means that individual >> blocks are compressed regardless of contents. Since *ix does updates on >> entire blocks in all (current) cases, this imposes only a decompression speed >> penalty on demand paging. >Not really. For example, I might want the directory structure to >remain uncomressed for speed, and I might want to have a quick >compression routine apply to most files for speed but with the option >of doing a slower, more thorough compression on some large, seldomly >changed files. Actually I was thinking the best way would not be to have new file system, but simply to use existing tools. One stradegy I was thinking of trying, is having a special bit to mark files as special. Then you could do something like compress a directory, and replace all the files with one special file hard linked to all the names of the files that where there. Then with a command like "ar" you place the compressed libraries in a random access library. Then when "open" sees that your special bit is set, it would follow the instructions in the special file for restoring the uncompressed version before trying to access it. I don't know how efficient such a method would be, but it seems like less work to impliment that a whole new file system. Bill ------------------------------ From: ins407x@aurora.cc.monash.edu.au (Andrew J. Cosgriff) Subject: Re: Is there a .au player for soundcards? Date: Thu, 15 Apr 1993 08:14:29 GMT damien@b63519 (Damien Neil) writes: >This reminds me -- is there an .au player that _will_ use a soundcard? I have >splay from the snd-util package, but the sound it produces when playing an .au >file makes it seem like I'm getting a poor radio transmission...from Venus. Not >the program's fault of course -- it wasn't designed to use the .au format. But >it would be nice to have access to all the Sun audio files out there. Yep, there is... It's called /dev/audio (have you mknod'ed one ?) cat your sun audio files there and they'll work just fine. It'd be trivial to modify the splay code to use /dev/audio if you told it you had a .au file I suppose. (One day when I've got nothing else to do...) /dev/dsp (what splay uses) is fine for .voc and .wav files (you just have to ignore the header at the start of wav files which comes out as a short burst of static). /dev/audio is for Sun audio files. Enjoy ! -- =============================================================================== * Andrew J. Cosgriff ! * "Connect the goddamn dots !" (Ministry) ins407x@aurora.cc.monash.edu.au #include andrew@bing.apana.org.au Bing ! ------------------------------ From: Leo.Hippelainen@ntc.nokia.com (Leo T. Hippelainen) Subject: Linux tty's Date: Thu, 15 Apr 1993 08:02:35 GMT We would like to have 8 or more asyncromous lines in a Linux PC. Is there any supported io-card configuration allowing this many serial lines in Linux Has any one experiences with connecting a terminal server through TCP/IP to linux lumme@kiloapo.tele.nokia.fi ------------------------------ Subject: Re: Linux on Macintrashes? From: sjmadsen@nextsrv.cas.muohio.edu (Steve Madsen) Date: 15 Apr 93 00:05:09 -0500 Chris Higgins (chris@Odyssey.ucc.ie) wrote: > In article 1qhhd7INN7m0@matt.ksu.ksu.edu, strat@matt.ksu.ksu.edu (Steve Davis) writes: [...] Saying that all Macintosh computer owners are part of Apple's "fascist empire" is just as silly as the claim that the Mac isn't a real computer. It works for it's purpose. If you don't like it, don't buy it, but don't go slamming the people that use them for real purposes. The argument that the Mac is a black-and-white piece of junk is an outdated argument. Replies to my mailbox.. this really has little to do with Linux. -- Steve Madsen sjmadsen@nextsrv.cas.muohio.edu Ask me about Linux, the free 386 unix! ------------------------------ From: kmy@conger.mel.dit.CSIRO.AU (Yap Kid Mun) Subject: Linux swapping Date: Thu, 15 Apr 93 04:32:58 GMT I could not find this in the FAQ, so here goes... Is there anyway to make linux share swap with /tmp? (Most Unix allow this and it makes for faster compilation as most of the time /tmp will be in physical memory, provided you have enough of those in the first place :-). Any answers would be much appreciated. Thanks -- / Email: KidMun.Yap@mel.dit.csiro.au CSIRO Division of Information Technology Australia ------------------------------ From: aehall@calvin.seattleu.edu (root-bound) Subject: Re: kermit weirdness...? Date: 15 Apr 1993 04:05:50 GMT In article ferrick@acsu.buffalo.edu (Patrick K. Ferrick) writes: >Every once in a while, while logged into a remote unix host via kermit >my term goes whacko...everything is still there, except most of the >characters come out as dots and vert/horiz lines...! The VC in question >stays that way until I reboot. None of the others are affected. > >Has this been discussed to death yet? If so, sorry...I don't have the >bandwidth to scan through as many posts as I'd like :-( > >tnx, >pat Try escaping back to kermit on your local system and type: !reset This will run the reset script that contains the line "echo ^[c". This will reset your screen and everything should work fine. This also works with the problem I've encountered of everything showing up on the last line of the VC. You can also do, from another VC "reset > /dev/ttyX" to the tty on which you're having problems. Anthony -- Anthony Hall _ _ Unix System Administrator aehall@seattleu.edu /_/ /_/ Physician Micro Systems, Inc. _ _ 2033 6th Ave Suite 707 /_/ /_/ Seattle, WA 98122 206-441-8490 ------------------------------ From: bcollc@ciba-geigy.ch (Christopher Collins) Subject: Compatibility question Date: Thu, 15 Apr 1993 09:30:45 GMT Hi I have some questions over compatibility of some devices, please dont flame me if they have been asked before. I would prefer if any answers were mailed to me at the email address below. Question, will the following hardware work with Linux, and if so what drivers do I need and where do I get them from: 1) Mitsumi CDROM with 16 bit interface (pretty sure its not SCSI) 2) ESCOM Archiv/Maynard compatible tape streamer (I think it connects to the floppy interface. 3) 3Com 3c505 Ethernet card. I do not have any of the above but I need a CDROM + Tape streamer + ethernet card soon. If anybody has experience of these I would like to know. ======= Christopher Collins Ciba-Geigy AG Technical Infrastructure ++41 61 69 79112 Telephone Polymer Informatics ++41 61 69 78624 Faxsimile R-1045.3.37 4002 Basel email: bcollc@ciba-geigy.ch Switzerland ======================================================================== Boomshanka ======================================================================== ------------------------------ From: jeschke@cs.indiana.edu (Eric Jeschke) Subject: Re: bug in .99p8A char device drivers? (was: bad SIMM?) Date: 15 Apr 93 04:34:25 GMT Just a follow-up note; if I compile the kernel with memory support <=16Mb, the system does not crash on the two scenarios described in the header post, although I do get the occasional Result 0x??????? --memory map destroyed. message. I'm upping to 24Mb tomorrow, hope I can figure out what's going on so I can use it! -- Eric Jeschke | Indiana University jeschke@cs.indiana.edu | Computer Science Department ------------------------------ From: drew@juliet.cs.colorado.edu (Drew Eckhardt) Subject: Re: Help on X-Windows(mostly) Date: Thu, 15 Apr 1993 04:31:28 GMT In article <28617@galaxy.ucr.edu> jmward@ucrengr (jonathan ward) writes: >Hi there. I'm relatively new to Linux(awesome OS), and just got the SLS releaseoff of susnsite(0.99pl6). I'm tring to configure X to run on my machine, which is: generic 486DX-50 VLB motherboard, trident tvga8900c video card with 1meg ram, maxtor LXT340A IDE hard drive, a three button serial mouse. I finally got X >up and running(it defaults to the open windows manager) but I have this problem: >The mouse, once started is generally unresponsive. I can move it repeatedly, >and it occasionally "jumps" around the screen. When I depress a button, it >then takes a while to respond. I only have 4mb of RAM, and I'm swapping to a >12 mb swapfile. Is the mouse problems due to the lack of ram, or something I >overlooked? Yes. The Xserver and Xtoys use an incredible amount of memory, and will definately page (going into disk wait) heavily. The best solution would be to add more memory, at least 4M (you can buy used 1M SIMMs for $25 each). You can also reduce the amount of memory used by switching to X386mono, which only needs one bit for every pixel in a window, instead of the 8 needed by X386. X386mono will also be faster since it's not moving as much data - try it. > >Also, I'm using the e2fs for my main partition. What is the proper method to run >e2fsck? > You should automatically fsck file systems on boot up (in case your machine crashes from a power hit and automagically comes back up - if you write to a corrupt file system, you'll just make things worse) from /etc/rc, before you start any daemons that write to the disk, and before you've mounted any filesystems (and root anything but readonly). Generally, it's also a good idea to reboot if there were any problems with the fsck on /. -- Boycott USL/Novell for their absurd anti-BSDI lawsuit. | Drew Eckhardt Condemn Colorado for Amendment Two. | drew@cs.Colorado.EDU Use Linux, the fast, flexible, and free 386 unix | ------------------------------ From: goer@ellis.uchicago.edu (Richard L. Goerwitz) Subject: Re: Linux on Amiga? Reply-To: goer@midway.uchicago.edu Date: Thu, 15 Apr 1993 04:43:20 GMT In article <1993Apr14.095809.1@ualr.edu> nmspillers@ualr.edu writes: > >The point is, there are a lot of folks out there with Macs. They are not >encouraging facist computing, nor are they a member of Apple's "Evil Empire" >and wander around with a Borg-like mentality--"My user interface will >assimilate you, resistance is futile". This isn't a GUI war. This isn't meant to be a grilling, but I wonder if you grasped why the FSF opposes porting anything to Apple's products. Has nothing to do with one GUI conquering any other. -- -Richard L. Goerwitz goer%midway@uchicago.bitnet goer@midway.uchicago.edu rutgers!oddjob!ellis!goer ------------------------------ From: ph91rl@erwin.ph.surrey.ac.uk (Tankgirl) Subject: CD-ROM question. Date: 15 Apr 93 09:34:19 GMT Does linux support mitsumi CD-ROM drives? Will version 1 of linux be bundled on a CD rom? Thanx in advance. Rob. ------------------------------ From: Steve_Goyette@mindlink.bc.ca (Steve Goyette) Subject: HELP!!! with INET Probs Date: Thu, 15 Apr 1993 04:49:55 GMT Hi all, I am attempting to link an NCD-16 XWindows terminal to my 486 Linux box. On the Linux end of it I am running SLS-99 Patch Level 6. I have 8 Megs of memory and am using a Western Digital WD8013 with an IO address of 280, interrupt of 11 and Memory Address of D0000. The problem I am having is that as soon as I try to telnet from the NCD I get the following error (Repeatedly): kmalloc called with impossibly large argument (65587) dev_ring:dropping packet due to lack of memory eventually on the NCD I get connection closed and that's it. The messages may or may not stop. HELP!!! What do I do to stop this???? I am just about to pull out the reaminder of my hair. I have tried just about everything that I can think of. I re-configured everything in Space.c and re-compiled. Did a Make dep/Make clean and Make disk nada, everything gives me the same response. Any help would be GREATLY appreciated!!! -- ============================================================================ Steve Goyette |I knew something he didnt. No Silicon Heaven?? Software Engineer |Where do all the calculators go?? Human heaven Abstract Automation Limited |is something somebody make up to keep you all Steve_Goyette@MindLine.BC.CA|from going nuts!! Kriten - Red Dwarf ==================== ------------------------------ From: hjhong@ev004.ev.nctu.edu.tw (H. J. Hong) Subject: Some question? Date: Thu, 15 Apr 1993 04:54:43 GMT I consider to using linux and have some questions. Does any one help me ? 1. How much disk space is needed for a complete version of linux ? 2. How many MB of RAM is needed to run linux ? 3. Does X11R5 running on linux or any other x-window version can run ? Thanks for any answers!! H.J. 1993,4,15 ------------------------------ From: s_fuller@iastate.edu (Steve Fuller) Subject: Tutorial for setting up lpr/lpd????? Date: Thu, 15 Apr 1993 04:30:08 GMT Does someone out there have a really easy step by step tutorial on what to do to set up the line printer daemon and ALL associated files in order to get lpr to work??? Any and all leads appreciated. I am also looking for a etc/printcap entry for a Xerox 4045 laser printer. I have a Diablo entry for it, but a native 4045 entry would be oh so much nicer. :) Thanks for the help KEEP UP THE WORK LINUS!!!!!!!!!! Steve -- ==================================================== Steve Fuller = s_fuller@iastate.edu = No witty quote here yet... =================================================== ------------------------------ From: ton@blade.stack.urc.tue.nl (Ton van Rosmalen) Subject: Re: QIC-02 Tape Patches? Date: 15 Apr 1993 10:07:53 GMT John Will (john.will@satalink.com) wrote: : Has anyone applied the QIC-02 tape patches to the 99p7A kernel? It's not : clear to me how to get all the patches applied correctly to get tape support : back, I had the tape running fine in 99p6, but now I have three patch files : instead of one, and the first one spits out lots of errors when it runs. : Do I just forge ahead and apply the others, or is there something that : I'm missing here. :-) : --- : ~ KingQWK 1.05 # 97 ~ (A)bort, (R)etry, (F)ail, (S)lam fist on desk ? : The error messages I got when applying the patches all came from the same source. The RCS revision i.d. in about 4 files. So you should check your "*.c#" and "*.h#" files and make sure that the revision date/time is the only problem. If it isn't you'll probably have to patch the remainder yourself. Greetings, Ton. ------------------------------ From: jhenders@wimsey.bc.ca (John Henders) Subject: Re: Trying to kill getty so I can dial out on ttys2; help Date: Thu, 15 Apr 1993 03:17:54 GMT In article <1qdeh7INNfpb@senator-bedfellow.MIT.EDU> tytso@athena.mit.edu writes: >If the getty hogs the line, then you won't be able to open the >/dev/cua?? devices. This is NOT a bug with the /dev/cua?? devices, but >a problem with your getty program. Get the latest, fixed getty, and >this problem goes away..... getty_ps2.0.6b.tar.z in >/pub/linux/sources/sbin on tsx-11.mit.edu. This is where all the trouble is coming from. In actual fact, the getty_ps package does not work with the 0.99pl6 SLS release. I have gotten it succesfully running on the serial port, and as soon as it runs it grabs the tr and rs line and never lets them go. Trying to call out will not work. Several people have had the same experience, but when they ask for help, all they get is this same suggestion. It's very frustrating. A new package called duel is on sunsite and txs11 and is supposed to fix this, but I haven't got it working correctly yet. I'll post when I figure out how it should work. Right now it runs and allows dial outs fine] but doesn;'t answer the modem when called. Has anyone got a patch/fix for the dial out term screwup yet? When I dial out in minterm, if I edit anything in vi, the screen forgets how to scroll upward. Someone told me it was a know linux bug, but their suggested fix of sending a reset from another console didn't work, and you can't suspend miniterm to reset on that vc. -- John Henders ------------------------------ From: jer@pender.ee.upenn.edu (Joel Ratsaby) Subject: cp from diskette to HD ? Date: 15 Apr 93 10:25:23 GMT Reply-To: jer@pender.ee.upenn.edu (Joel Ratsaby) say I download a file, file.Z. to my DOS diskette (using MS windows) which I got from some ftp site. When I am in Linux, how do I copy the file from the diskette (/dev/fd1H1440) to my hard drive ? How do I uncompress it while copying it to the hard-drive ? What are the commads for uncompressin/compressing the different foramts ie .tar .Z .taz thanks -- Joel Ratsaby EE Dept. University of Pennsylvania jer@eniac.seas.upenn.edu jer@pender.ee.upenn.edu --Joel email: jer@ee.upenn.edu, jer@eniac.seas.upenn.edu Phone: (215) 898-1858 ------------------------------ From: seidl@alw.nih.gov (Edward Seidl) Subject: Re: Big swap partitions [> 16M] Date: Thu, 15 Apr 1993 10:26:08 GMT In article <1qj6u8$s6l@news.cs.tu-berlin.de>, wong@cs.tu-berlin.de (Wolfgang Jung) writes: |> |> Q: How many Swapareas is Linux able to handle |> Eight. -- Edward Seidl DCRT/CBEL Bldg 12A, Rm 2033 seidl@alw.nih.gov NIH, Bethesda, MD 20892 It's spelled Luxury-Yacht, but it's pronounced Throat-Warbler-Mangrove. ------------------------------ ** 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-Activists-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux) via: Internet: Linux-Activists@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 tupac-amaru.informatik.rwth-aachen.de pub/msdos/replace The current version of Linux is 0.99pl6 released on February 23, 1993 End of Linux-Activists Digest ****************************** there somethingdigest847 644 36676 74430 47152 5366137076 5776 From: Digestifier To: Linux-Activists@senator-bedfellow.mit.edu Reply-To: Linux-Activists@senator-bedfellow.mit.edu Date: Thu, 15 Apr 93 09:45:39 EDT Subject: Linux-Activists Digest #847 Linux-Activists Digest #847, Volume #4 Thu, 15 Apr 93 09:45:39 EDT Contents: Re: kermit weirdness...? (Bill Reynolds) Re: Help on X-Windows(mostly) (Linus Torvalds) Re: Lilo on the master boot record (Henrik Lund) Need a LINUX logo... !! (Suominen Timo) Re: Compiling XV; general porting problems (Robert Chen) Does anyone ? (Matthew Roderick) Re: PS/2 style mouse not supported?? (Johan Myreen) Re: fsck on mounted file systems? (Nan Zou) Re: Tutorial for setting up lpr/lpd????? (NSF SF Bay research projects) Re: problems compiling programs with math.h (Ken Pizzini) Re: less and regexps (Ken Pizzini) Re: Disk Quotas (was Re: New feature for the filesystems.) (william E Davidsen) Re: NFAQ: Do ISA cards cripple EISA performance ? (Christian Franke) Re: hooray! voodoo spell broken. network works now! (william E Davidsen) How to GCC (Martin Rehwald) Re: Disk Quotas (was Re: New feature for the filesystems.) (william E Davidsen) Re: Big swap partitions [> 16M] (Linus Torvalds) ---------------------------------------------------------------------------- From: bill@yossarian.ucsd.edu (Bill Reynolds) Subject: Re: kermit weirdness...? Date: 14 Apr 93 23:19:26 Reply-To: bill@goshawk.lanl.gov Sometimes under kermit the character set gets screwed up, I've never seen this under linux, but under mskermit, the following fixes it ^\C set term char trans this sets the character set to transparent, you could also use ascii or norwegian or whatever -- ______________________________________________________________________________ Bill Reynolds | Hey Rocky! Watch me pull a rabbit out of my hat! bill@goshawk.lanl.gov | -Bullwinkle ------------------------------ From: torvalds@klaava.Helsinki.FI (Linus Torvalds) Subject: Re: Help on X-Windows(mostly) Date: Thu, 15 Apr 1993 09:06:25 GMT In article <28617@galaxy.ucr.edu> jmward@ucrengr (jonathan ward) writes: >Hi there. I'm relatively new to Linux(awesome OS), and just got the SLS releaseoff of susnsite(0.99pl6). I'm tring to configure X to run on my machine, which is: generic 486DX-50 VLB motherboard, trident tvga8900c video card with 1meg ram, maxtor LXT340A IDE hard drive, a three button serial mouse. I finally got X >up and running(it defaults to the open windows manager) but I have this problem: >The mouse, once started is generally unresponsive. I can move it repeatedly, >and it occasionally "jumps" around the screen. When I depress a button, it >then takes a while to respond. I only have 4mb of RAM, and I'm swapping to a >12 mb swapfile. Is the mouse problems due to the lack of ram, or something I >overlooked? Any suggestion appreciated. If getting more memory is totally out of the question, may I suggest changing the swapfile into a partition. I know that re-fdisking is a pain, but swapping to a partition is *noticeably* faster than swapping to a file, and it can make a sluggish X11 seem almost usable. Also, as has already been recommended, the mono-X11 is much less memory-hungry than the colour version, and would also help the problem. Note that using a swapfile is easy and painless, but it's really meant only for those cases where you don't really need to dip into swap all the time (when you have enough memory for most things, but need the swapfile for those occasional times when you compile big files while running X at the same time etc). If you are serious about swapping, try to do it using a swap partition. Linus ------------------------------ From: lund@diku.dk (Henrik Lund) Subject: Re: Lilo on the master boot record Date: 15 Apr 93 10:10:39 GMT i03a@rs62.zfn.uni-bremen.de (Thomas Meyer) writes: >H REGAN (H.Regan@bradford.ac.uk) wrote: >: Does anybody know away of restoring my master boot record? >To overwrite the Master Boot Record, start MS-DOS' FDISK command with >the option "/MBR". This will write the standard MS MBR. >No guarantees whatsoever! :) >Bye, >Thomas >-- >Thomas Meyer I e-mail: i03a@alf.zfn.uni-bremen.de >Student of Computer Science I IRC: mnemo >University of Bremen, Germany I -o- INCREASE THE PEACE -o- I think LILO makes a copy of the boot record. Try finding that copy on your disk ( has MSDOS5.0 in the beginning, if you used version 5.0). Had to do find it myself because i wrote 512 bytes onto the first bytes my HD, you know the master boot record, partition table etc. (Norton utilities found it, copied it and my computer is alive). I think this proves how stupid Dos is and how careful you have to be when playing with power (LINUX). Henrik Lund lund@diku.dk ------------------------------ From: ts106984@lehtori.cc.tut.fi (Suominen Timo) Subject: Need a LINUX logo... !! Date: 15 Apr 1993 12:51:19 +0300 Hi there... YAP. I think the title says it all. What I need now, is a LOGO for Linux (.GIF, .EPS, .TIF) or any... Send it to the newsgroup or via e-mail to me... Sorry if this is not the right newsgroup :), but I'm quite a green with this kind of Linux things... Thanx in advance! -tipi- -- Timo Suominen ! Voice: +358 31 3162137 ! There is no Tampere University of Technology ! Fax/Voice: +358 31 128435 ! fear without Korkeakoulunkatu 2, P.O. Box 527 ! Fax: +358 31 3162307 ! FEAR... 33101 Tampere - FINLAND ! E-mail: tipi@rulla.me.tut.fi ! ------------------------------ From: rchen@fraser.sfu.ca (Robert Chen) Subject: Re: Compiling XV; general porting problems Date: Thu, 15 Apr 1993 04:03:57 GMT In article <36972@optima.cs.arizona.edu> cjeffery@optima.cs.arizona.edu (Clint Jeffery) writes: >From article <1993Apr14.200018.21295@aio.jsc.nasa.gov>, by mancus@sweetpea.jsc.nasa.gov (Keith Mancus): >> I'm trying to compile xv 2.21 under Linux 0.99.6. gcc can make the >> libjpeg.a just fine, but it runs out of memory and dumps when I attempt >> to make libtiff.a. Since I have 12M RAM and a 10M swap partition, this >> must either be a gcc bug or bad options. > >When I compiled it under 0.99.7A on a 16MB 486/33, the machine appeared >to "hang" while compiling files for libtiff.a. Under X, the mouse no >longer moved; outside X, I could still switch consoles, but could not >type anything at any of them. This has to be more than just gcc or >bad options. Had to hard-reset to reboot. No, Linux just looks like it is hung when it runs out of memory. This has been kicked around many times in this newsgroup - lots of talk about elaborate ways to avoid out of memory hangs but no really good solution, IMHO. Give yourself 32 MB of swap and go to bed. You will have tiff in the morning. - Rob ------------------------------ From: mat@dsbc.icl.co.uk (Matthew Roderick) Subject: Does anyone ? Date: 15 Apr 93 11:17:20 GMT Does anyone have a SiS chipset mother board running linux ? With an AMI BIOS I have 486/33 and would like information on the setup for linux of the CMOS. I think my setup my be incorrect as Linux keeps panicing on boot ! Cheers Matthew ------------------------------ From: jem@snakemail.hut.fi (Johan Myreen) Subject: Re: PS/2 style mouse not supported?? Date: 15 Apr 93 05:51:30 GMT In article neonakis@alkistis.csi.forth.gr (E.A.Neonakis) writes: >>cursor stays at the very bottom of the graphics screen and any mouse >>movement is converted to horizontal moves only. I use a REAL Microsoft Mouse. >>On serial port, all OK. Any comments? >I have a similar problem, only the cursor stays on the top of the screen. Make sure you specify the correct protocol type in Xconfig, it should be 'PS/2' for a PS/2 type mouse. -- Johan Myreen jem@cs.hut.fi ------------------------------ From: nan@matt.ksu.ksu.edu (Nan Zou) Subject: Re: fsck on mounted file systems? Date: 14 Apr 1993 23:33:40 -0500 sct@dcs.ed.ac.uk (Stephen Tweedie) writes: >hwrvo@kato.lahabra.chevron.com (W.R. Volz) writes: >> I noticed that someone said that fsck should be run on >> unmounted file systems. Then how does run it on the >> filesystem that contains fsck? Is this a studip question >> or what? Inquiring minds want to know. >As long as there is no other disk activity going on, it is safe to run >fsck on a mounted partition just to check the correctness of the >partition. However, if you are using the -r or -a options to repair a >filesystem, then there are basically two safe ways to go about it: [...] >* Using the latest kernel, elect to mount the root filesystem > readonly; upon boot, run fsck on the partition, and once that has > finished, remount root for writing and mount the other partitions. If the root filesystem is mounted readonly, how can fsck write back corrections if it finds errors? -- Nan ------------------------------ From: nsf@tahoe.ced.berkeley.edu (NSF SF Bay research projects) Subject: Re: Tutorial for setting up lpr/lpd????? Date: 15 Apr 1993 05:49:30 GMT >Does someone out there have a really easy step by step tutorial >on what to do to set up the line printer daemon and ALL >associated files in order to get lpr to work??? Any and all leads >appreciated. --- snip ----- > >Steve PLEASE, PLEASE, PLEASE, PLEASE, PLEASE.... post something for me too. I've been didling with lpd for weeks (on occasion). I've tried all the permissions, setuid stuff, I've read the readmes, faqs etc, but no go (cannot start daemon). -Howard -- ====================================================================== Howard Foster (nsf@ced.berkeley.edu) Ph.D. candidate, environmental planning, UC Berkeley. Phone: 510-654-6585 Snail: 217 Monte Vista Ave., Oakland CA 94611-9421 ------------------------------ From: ken@halcyon.com (Ken Pizzini) Subject: Re: problems compiling programs with math.h Date: Thu, 15 Apr 1993 10:12:25 GMT In article <1993Apr13.202943.13060@ousrvr.oulu.fi> tax@tolsun.oulu.fi (Samuli Takala) writes: >Every time I try to compile a program which uses functions defined in math.h >it causes an error message like this: >prog.o: Undefined symbol _log10 referenced from text segment Sounds like you forgot to add a "-lm" to the end of your gcc (or ld, if you run that directly) command -line. This tells the linker to make use of the math library, where most of the functions live. >As I got linux from a friend I'm not sure what distribution it is. It came >on 18 3.5 inch disks and was originally compiled by root@tamu. Can someone >identify that as well? Sounds like the TAMU distribution, one of the popular Linux bundlings. --Ken Pizzini ------------------------------ From: ken@halcyon.com (Ken Pizzini) Subject: Re: less and regexps Date: Thu, 15 Apr 1993 11:50:58 GMT In article twpierce@unix.amherst.edu (Tim Pierce) writes: >Is it just me, or does less (version 1.77, SLS-supplied) not accept >regular expressions in searches under Linux? It certainly doesn't >behave this way in general. (Yes, I can rebuild it, once I spend a >half-hour or so downloading the source over a 9600-bps link, but it >seems peculiar that it wouldn't work in the SLS distribution). Apparently whoever built the less that is distributed with SLS decided not to enable regexps. I've upload my less binary to /pub/Linux/Incoming on sunsite.unc.edu; this has regexps enabled. --Ken Pizzini ------------------------------ From: davidsen@ariel.crd.GE.COM (william E Davidsen) Subject: Re: Disk Quotas (was Re: New feature for the filesystems.) Reply-To: davidsen@crd.ge.com (bill davidsen) Date: Thu, 15 Apr 1993 12:27:52 GMT In article , s_fuller@iastate.edu (Steve Fuller) writes: | 1) If the file is in the middle of a write when you exceed your | quota, the file will be written in its entirety and not | truncated. Any furthur write requests will be denied I certainly don't want a quota like that. If the hard limit isn't hard it doesn't protect the system. unless writes STOP at the hard limit the user could run the system out of disk anyway. -- bill davidsen, GE Corp. R&D Center; Box 8; Schenectady NY 12345 ------------------------------ From: franke@andrej.informatik.rwth-aachen.de (Christian Franke) Crossposted-To: comp.sys.ibm.pc.hardware Subject: Re: NFAQ: Do ISA cards cripple EISA performance ? Date: 15 Apr 93 12:41:27 GMT alecm@coyote.uk.sun.com (Alec Muffett - Sun IS - System Administrator) writes: >I've heard from a friend that if you have an EISA machine, and install >an ISA card of any sort, the bus performance will drop to that of an >ordinary ISA bus. I have asked this question some month ago. Here is my posting and all the replies: =========================== []From: franke@andrej.informatik.rwth-aachen.de (Christian Franke) []Subject: Does ISA board slow down EISA bus ? (FAQ?) []Date: 19 Oct 92 10:45:09 GMT [] []Hello, [] []this is probably a FAQ: [] []I have heard from different sources, that an ISA board []in an EISA PC slows down the access to the remaining EISA boards. [] []An article in an old issue of the german 'CT magazine states []that this is true when ISA and EISA board are on the same "address space" []But they did not tell what "address space" means here. [] []Any comments? [] =========================== []From: Sam Shim [] [] It's not true. ISA boards have no effect on the EISA boards. No effect []whatsoever. That's a rumor that always seems to crop up every once in a []while. [] []-Sam- [] =========================== []From: mccall@mksol.dseg.ti.com (fred j mccall 575-3539) [] []From what I understand, this simply doesn't happen on a properly []implemented EISA bus. The bus will run at 8.33 MHz with a transfer []every clock. This seems to imply that even ISA boards will run twice []as fast as they do in an ISA-bus machine (although I'm not sure about []that bit). Anyway, there is no reason for the EISA bus to slow down []for ISA cards. And it doesn't. [] =========================== []From: aron@taos.ced.berkeley.edu (aron bonar) [] []I thought that the only difference was the size of the data path...ie 16 bit for ISA and []32 bit for EISA. The bus clock speed is the same (more or less) for both so if you stick an []ISA card in an EISA slot then the slot only transfers over a 16 bit path and not a 32 bit one. [] []So..I guess you could say that if you put an ISA card in an EISA slot then your system is slower []than it could be. But then again you knew that. [] =========================== I am still not certain about this topic. Regards, Christian Franke Aachen University of Technology Informatik I Ahornstrasse 55 W-5100 Aachen Germany Tel.: +49.241.80-21111 E-Mail: franke@informatik.rwth-aachen.de ------------------------------ From: davidsen@ariel.crd.GE.COM (william E Davidsen) Subject: Re: hooray! voodoo spell broken. network works now! Reply-To: davidsen@crd.ge.com (bill davidsen) Date: Thu, 15 Apr 1993 12:49:12 GMT In article <1qi66oINNh5q@iskut.ucs.ubc.ca>, jmorriso@rflab.ee.ubc.ca (John Paul Morrison) writes: | I cant explain why *several* kernels later, 99.7, 7A, etc. the net | still worked, then *without* even touching the kernel, it mysteriously | stops working. Anyway, I followed the instructions on ftp.super.org, | and I used those ethernet drivers to make a kernel that works with our | SMC 16 bit ethernet card. | | So Im happy it works now. Anyone have theories about why it *stopped* | working, when there was no visble change to the files? (this is weird: | I restore files from a tape made from a *working* system, and the | system still wont work). maybe the soft settings on the ethernet card | got twiddled , making it fail with my previously working kernel. Did you run any DOS programs which use the card? I found that I got mine working reliably when I forced the i/o, int, and shared memory address, and not until. It seemed to use a random location for the transfer buffer unless I forced it. This stuff is in /linux/net/tcp in the GNUmakefile. -- bill davidsen, GE Corp. R&D Center; Box 8; Schenectady NY 12345 ------------------------------ From: zy0004@urix4.uni-muenster.de (Martin Rehwald) Subject: How to GCC Date: 15 Apr 1993 12:20:48 GMT I can't compile C-programs using the math libary properly. Even adding the '-l m' option to the command line does not link the libary. I'm always told, there is a unknown symbol referenced by my text segment (its name is _sin). Watching my libm.a in /usr/lib shows me, there is a sin in there. The strange about this ist, that he doesn't complain abount not having found the libary but just not semms to link it. One solution was to change the specs file in /usr/lib/gcc-lib....... adding a '-l m' to each entry in the default libs line. Well - as i think this is not the very best solution. So - what's wrong about my installation? Thanks for help, MR ------------------------------ From: davidsen@ariel.crd.GE.COM (william E Davidsen) Subject: Re: Disk Quotas (was Re: New feature for the filesystems.) Reply-To: davidsen@crd.ge.com (bill davidsen) Date: Thu, 15 Apr 1993 12:25:11 GMT In article <9821@dirac.physics.purdue.edu>, bcr@bohr.physics.purdue.edu (Bill C. Riemers) writes: | Yes I haven't seen a system yet that properly handles hard quotas. Most | systems will still let you write to /tmp so you just direct the normal | programs to put thier output to there and then take care of them before | logging out. You do realize that quotas are on a "per filesystem" basis? You might have a quota on /tmp, and that would be different from the quota on the home filesystem, etc. That's not remotely a bug, it's the way it was intended to work. And the reason for quotas is to protect the system, not to punish the user from trying to use resources to get something done... I could see having a soft limit on /tmp of about 100b (for shell history), and a hard limit of 50MB, with a time limit of 1hr or so. That lets people get enough disk to do image processing, and still encourages them to remove files as soon as they are done. Of course I usually run a daemon which encourages /tmp cleanliness, too. -- bill davidsen, GE Corp. R&D Center; Box 8; Schenectady NY 12345 ------------------------------ From: torvalds@klaava.Helsinki.FI (Linus Torvalds) Subject: Re: Big swap partitions [> 16M] Date: Thu, 15 Apr 1993 13:01:17 GMT In article <1993Apr15.102608.25039@alw.nih.gov> seidl@alw.nih.gov (Edward Seidl) writes: >In article <1qj6u8$s6l@news.cs.tu-berlin.de>, wong@cs.tu-berlin.de (Wolfgang Jung) writes: >|> >|> Q: How many Swapareas is Linux able to handle >|> > >Eight. In the default configuration, yes. I think the way I encode the swap-page numbers actually restrict the number of swap-areas to 127 or something like that, but you should be able to get more than 8 by just changing a #define. Linus ------------------------------ ** 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-Activists-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux) via: Internet: Linux-Activists@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 tupac-amaru.informatik.rwth-aachen.de pub/msdos/replace The current version of Linux is 0.99pl6 released on February 23, 1993 End of Linux-Activists Digest ****************************** restore files from a tape made from a *working* system, and the | system still wont work). maybe the soft settings on the ethernet card | got twiddled , making it fail with my previously working kernel. Did you run any DOS programs which use the card? I found that I got mine working reliably when I forced the i/o, int, and shared memory address, and not until. It seemed to use a random location for digest848 644 36676 74430 52762 5366137077 6003 From: Digestifier To: Linux-Activists@senator-bedfellow.mit.edu Reply-To: Linux-Activists@senator-bedfellow.mit.edu Date: Thu, 15 Apr 93 10:45:31 EDT Subject: Linux-Activists Digest #848 Linux-Activists Digest #848, Volume #4 Thu, 15 Apr 93 10:45:31 EDT Contents: Re: Automated Linux PowerDown (william E Davidsen) Re: Linux on Macintrashes? (Was: Re: Linux on (Steve Davis) Re: Automated Linux PowerDown (Drew Eckhardt) Re: TROUBLE USING msdosfs (Matthaeus Zepf) xdb (dpjunk@mmm.com) Grrr! This networking is driving me nuts! (Alan Charlton) Re: Disk Quotas (was Re: New feature for the filesystems.) (Marc G Fournier) Re: Disk Quotas (was Re: New feature for the filesystems.) (Liefting W) Re: Linux on Macintrashes? (Was: Re: Linux on Amiga?) (James Mitchell) Re: Linux Logo Postscript (Nick Holloway) ---------------------------------------------------------------------------- From: davidsen@ariel.crd.GE.COM (william E Davidsen) Subject: Re: Automated Linux PowerDown Reply-To: davidsen@crd.ge.com (bill davidsen) Date: Thu, 15 Apr 1993 12:59:47 GMT In article , jkaidor@synoptics.com (Jerome Kaidor) writes: | Previously, my computer was on a 24-hour timer, which turned it on at 3:00 in the morning | ( or thereabouts ), and turned it off at 5:00 AM ( or so ). The intervening two hours were enough | for the machine launch uucico and pick up my mail. Then during the day, I would turn the computer | on manually, and do my usual stuff. | | According to my calculations, leaving the computer on 24 hours/day will eat $10-$20 worth of | electricity. I, personally, have no problem with this, but my wife might well disagree. | | Accordingly, I have thought up the following hardware/software solution: | | 1) I build a box that allows the computer to turn itself *off*. A reed switch which will run off a parallel port, and a surplus relay. | 2) I let the 24-hour timer turn the computer *on*. | 3) A cron script turns the computer off if: | a) The computer has been on for a few hours ( or maybe, that the automated nighttime tasks are finished ) | b) It's early in the morning ( 4:00 AM or so ) | c) There are NO USERS ( nobody logged in ) | Anybody have any comment on all this? I'm sorry, it's not totally Linux-specific, but I figured that some other linux'ers | out there might be interested, too.. I don't think my situation is all that unusual. Power cycling is hard of equipment. I would turn off the monitor if you must, let the computer comtrol the modem power if you must, and leave the computer on. A typical [34]86 actually draws about 150w from the mains. 150w * 24hr * 30 day/mo = 7200, aka 72kw-hr. Sounds like less than a six-pack of beer to me. -- bill davidsen, GE Corp. R&D Center; Box 8; Schenectady NY 12345 ------------------------------ From: strat@enterprise.ksu.ksu.edu (Steve Davis) Subject: Re: Linux on Macintrashes? (Was: Re: Linux on Date: 15 Apr 1993 08:23:55 -0500 chris@Odyssey.ucc.ie (Chris Higgins) writes: >The thing is, that if they have a MAC and they don't know about the >FSF/GNU then they won't want a CLI, they won't want Un*x. So I agree >that porting Linux to the MAC would be in violation of the FSF >recomendation/guideline/rule/whatever. No, that's not the problem. Linux itself isn't a GNU project, nor are many of the contributers members of the FSF. The sticky part comes when you realize that Linux doesn't really *do* much without the help of tools like gcc, and without the help and encouragement of the FSF these tools might not have ever been ported to Linux in the first place. FSF is boycotting apple, so they will do nothing to help the developers put the system together; nor will they support the system if (or when) it ever runs [at least as I understand it]. >BUT, having said that, I do believe that the FREE OS / FREE Software >thing from GNU will have to be clarified.... Why don't you just read the FSF texts on the matter. Read the CopyLeft, the GNU Manefesto, or the many GNU mail lists. >>Not at all. The best way would be to chuck that Mac out the window >>and buy a real computer. >The MAC *IS* a real computer. ('real' does not imply 'good')... It's >the politics behind it that is the problem. I stand corrected on this issue, of *course* Macintrashes are real computers, there are working UNIX ports to them! (:P) The question is: Is it really worth anyone's time to port Linux over to them knowing that 1. The Linux kernel depends on a 386 protected mode architecture. Meaning, of course, that binary compatibility is out of the question, and that "porting software to Linux" would mean having to work on two completely different architectures. (Part of the reason for Linux's success to date is that all 386 and 486 systems at least share the same architecture.) To put it in simpler terms, a Linux port to the Macintrash *wouldn't be Linux*. 2. FSF wants nothing to do with the project. This means that there won't be any support for the compilers, and none for the tools compiled with them. Macintrash Linux users might have to *ghasp* write their own patches to software or deal solely in precompiled packages (something else FSF isn't very fond of). Having said this, we get to the issue of what Macintrash users want or need. Far be it from me to criticize someone for buying a tool that works perfectly for their particular use, but I fail to see reason in trying to make a Macintrash something that it's not. Macintrashes *have* a stable, multitasking operating system and a reasonable selection of software and tools. You don't have to option to get the OS unbundled as you do in MS-DOGland, nor could you really get far without commercial software packages. The few people I know that bought Macintrashes for personal use (most have used Macintrashes at work, and consequently go to the computer store looking for an alternative), the machines were purchaced for homework (word processing) or business (spreadsheets, et al.) and wouldn't know the difference between UNIX and VM/CMS if their lives depended on it. If the machine was bought for business, its because Macintrashes are a plug-and-play system. They're handy for plunking down on someone's desk with minimal configuration and getting the work done. Perhaps I'm missing something important here, but I fail to see where something like Linux would fit into such an environment. If Macintrashes are bought for their stability, compactness, and ease of use ... it seems to me that most Mac users would run in the opposite direction if confronted with a monster like UNIX. I realize there are a handful of Macs that run UNIX, but the market is tiny and is already saturated with commercial offerings. Finally, I see no particular advantages in running Linux on your home Macintrash rather than purchasing a power PC with the expandibility, compatibility, and portability issues already dealt with. Computer Shopper this month is selling complete 386SX/16 systems (the minimum configuration) for under $500 and complete 486DX2/66 systems (top of the line) for under $2500. If you order your machine UPS Red today and copy the SLS distribution off someone elses disks or your Internet account tonight before you go home from work you could have a working, stable, complete, recognised, supported, *FREE* UNIX look-alike workstation by Wednesday (complete with 3 button mouse and *color* display!). >If you love something, set it free. If it doesn't >come back to you, hunt it down and KILL it. Haha! The new FSF motto? :) Stratocaster -- Steve Davis (I'm a student, not a spokesperson!) strat@cis.ksu.edu - Kansas State University - Manhattan KS Anarchy means having to put up with things that really piss you off. ------------------------------ From: drew@kinglear.cs.colorado.edu (Drew Eckhardt) Subject: Re: Automated Linux PowerDown Date: Thu, 15 Apr 1993 13:20:44 GMT In article jkaidor@synoptics.com writes: > > > Now, for my purposes, linux has only one disadvantage, as compared to dos. I'd like to address >it: with DOS, you can generally dump the computer's power any time that you see that dos prompt. >With any unix, you have to be more careful... Why is this important? Because unices use a write back cache to maximize performance - when you have multiple programs accessing multiple files, you want to reorder the writes to minimize head movement, without blocking any of the processes unecessarily. > > Previously, my computer was on a 24-hour timer, which turned it on at 3:00 in the morning >( or thereabouts ), and turned it off at 5:00 AM ( or so ). The intervening two hours were enough >for the machine launch uucico and pick up my mail. Then during the day, I would turn the computer >on manually, and do my usual stuff. > > According to my calculations, leaving the computer on 24 hours/day will eat $10-$20 worth of >electricity. I, personally, have no problem with this, but my wife might well disagree. Every power cycle may put as much as 50 hours worth of wear on the hard disk, and your wife wouldn't like it if it broke and you had to spend hundreds of dollars on a new drive... -- Boycott USL/Novell for their absurd anti-BSDI lawsuit. | Drew Eckhardt Condemn Colorado for Amendment Two. | drew@cs.Colorado.EDU Use Linux, the fast, flexible, and free 386 unix | ------------------------------ From: db9@aixgraf2.urz.uni-heidelberg.de (Matthaeus Zepf) Subject: Re: TROUBLE USING msdosfs Date: Thu, 15 Apr 93 12:41:04 GMT sgccsns@citec.oz.au (Shayne Noel Smith 2245702) writes: : : This is my second post re this problem : : when trying to mount a 40M msdos 5 partition I get the folloowing errors : : [MS-DOS FS Rel. alpha.10,FAT 16,check=n,conv=b,uid=0,gid=0,umask=066] : [me=0xf8,cs=4,#f=2,fs=1,fl=80,ds=161,de=512,data=193,se=0,ts=82110,ls=512] : Unsupported FS parameters : mount: wrong fs type, /dev/hda1 already mounted, /mnt busy, or other error : : : : : The "Unsupported FS parameters" is the prob. I think.. : : can someone give me a pointer where to look.. : : : Thanks : : --Shayne Smith : What partition _exactly_ was it you wanted to mount ? I got a similar problem, but after I looked at the partition table I found I've tried to mount an unused partition entry ... So, have a look at the partition table ( with 'fdisk' and give 'em a 'p' for 'print partition table'). It'll tell you, which partition you should mount. Sorry, try fdisk /dev/xxx where xxx is the name of the HD ( _without_ partition number ). Good luck and greetings from Heidelberg Hannes Reinecke (hare@vogon.mathi.uni-heidelberg.de) No .sig today, my love has gone away ... ------------------------------ From: dpjunk@mmm.com Subject: xdb Date: 15 Apr 1993 10:07:50 -0400 Reply-To: dpjunk@mmm.com I am having a problem getting X started on my pc. The problem I am having is that I see three duplicate images of the screen on my screen. (i.e. Three copies of the window manager, mouse, etc...) I am using a Western Digital Paradise windows acc. card with 1 MB and my monitor is a VGA CTX (640X480). I tried both the color and monochrome versions of X386 and both of them come up with the same results. BTW, the monochrome version was working before I upgraded to SLS .99p6. Any help would be great. Please reply by e-mail. -- Dean P. Junk "An ounce of perception, a pound of obscure" Internet (dpjunk@mmm.com) -- RUSH ------------------------------ From: agc@weasel.demon.co.uk (Alan Charlton) Subject: Grrr! This networking is driving me nuts! Date: Wed, 14 Apr 1993 20:10:02 +0000 As the title suggests, I'm having a few networking problems. Up until recently, I've had no problems at all running the pre-ALPHA SLIP code with my 0.99.4 kernel with libc.so.4.3.2; all the net config files are in the right place, and they all seem to have the right things in them. `Fine!' you might say, `So what's the problem?. Well, I'll tell you. I recently set up another linux box with a similar network configuration (0.99.7A, libc.so.4.3.3, with a different hostname & IP address), and connected the two together using a couple of WD80x3 clones and some thinnet cable. And although both kernels can see both cards, I can't for the life of me make them talk. I get plenty of `network is unreachable' errors, but it seems to me that the code isn't actually talking to the ethernet cards (at least, no lights flash!). And yes, I've checked the network terminations etc. If someone out there has overcome what seems to me to be a very basic and fundamental configuration problem of some description, I'd be extremely grateful to hear from you - preferably by mail, although I may well send an account of my experiences to the maintainer of the net-FAQ if it's something which isn't in there. Thanks very much (in advance), Alan -- [ Alan Charlton - Bristol, England | "March is on its stomach" ] [ email: agc@weasel.demon.co.uk | Napoleon Bonaparte ] ------------------------------ From: marcf@nexus.yorku.ca (Marc G Fournier) Subject: Re: Disk Quotas (was Re: New feature for the filesystems.) Date: Thu, 15 Apr 1993 13:12:33 GMT davidsen@ariel.crd.GE.COM (william E Davidsen) writes: >In article , s_fuller@iastate.edu (Steve Fuller) writes: >| 1) If the file is in the middle of a write when you exceed your >| quota, the file will be written in its entirety and not >| truncated. Any furthur write requests will be denied > I certainly don't want a quota like that. If the hard limit isn't hard >it doesn't protect the system. unless writes STOP at the hard limit the >user could run the system out of disk anyway. but, isn't that what the ulimit is for? to control the size of the file that a user creates? if that were the case, then why not hhave the 'hard limit' be that a user cannot create a file, period, if they have reached it, and have the ulimit being there to keep them from just continually running. Most ulimit's I've seen are defaulted to about 2Meg or so. Understandably, if their hard limit is 1Meg, then creating a 2Meg file starting at their .5Meg level would allow them to pass their hard limit by 1.5Meg, but in that case, why would their ulimit be larger then their hard limit? just a thought... marc ------------------------------ From: wlieftin@cs.vu.nl (Liefting W) Subject: Re: Disk Quotas (was Re: New feature for the filesystems.) Date: Thu, 15 Apr 1993 13:20:33 GMT davidsen@ariel.crd.GE.COM (william E Davidsen) writes: >In article , s_fuller@iastate.edu (Steve Fuller) writes: >| 1) If the file is in the middle of a write when you exceed your >| quota, the file will be written in its entirety and not >| truncated. Any furthur write requests will be denied > I certainly don't want a quota like that. If the hard limit isn't hard >it doesn't protect the system. unless writes STOP at the hard limit the >user could run the system out of disk anyway. >-- >bill davidsen, GE Corp. R&D Center; Box 8; Schenectady NY 12345 I am certainly not a linux or unix wiz, but why not handle a quota exceed just the same way as a 'full device' error? Wouter. ------------------------------ From: drl@eskimo.com (James Mitchell) Subject: Re: Linux on Macintrashes? (Was: Re: Linux on Amiga?) Date: Thu, 15 Apr 1993 01:47:28 GMT <1qhhd7INN7m0@matt.ksu.ksu.edu> strat@matt.ksu.ksu.edu (Steve Davis) writes: >nmspillers@ualr.edu writes: >[excerpt from a GNU vs. Apple thread deleted] >>The point is, there are a lot of folks out there with Macs. They are not >>encouraging facist computing, nor are they a member of Apple's "Evil Empire" >>and wander around with a Borg-like mentality--"My user interface will >>assimilate you, resistance is futile". They merely bought a nifty, well >>built machine that fills their needs, why punish them? Shoot, most of 'em >>don't even *know* about the FSF or GNU or any boycotts! >If they don't know about FSF or GNU it is because they *have* been >'assimilated' by the Macintrash empire. If they were dumb enough to >buy an over-priced-one-mouse-button-tiny-black-and-white-screen- >bloated-awful-looking-os-without-a-cli-and-built-by-fascists computer >in the first place, there's really no reason for them to want or need >UNIX. After all, purchasing a Macintrash is acceptance of its point >and drool interface, AND a token of acceptance for Apple's business >practices-- both of which go against the free software/UNIX tradition. >>I want to learn, I want to do something worthwhile and give a little back to >>the computing community that has helped me so much, and I think this is a good >>way of achieving these aims. . . >Not at all. The best way would be to chuck that Mac out the window >and buy a real computer. >Stratocaster >-- > Steve Davis (I'm a student, not a spokesperson!) > strat@cis.ksu.edu - Kansas State University - Manhattan KS > >I assume that as a beginner you'll just want to blow people up. -- Red Shirt A plea for peace... ENOUGH TRASHING THE MACINTOSH!!! I'm the first to admit that the Macintosh is NOT the perfect computer, but neither is any other computer. Long, long, long ago in my computing infancy I encouraged my parents to buy a macintosh SE. I have never regretted the decision. It is no longer the greatest computer in the world, but MY DAD IS MAKING ADVERTISMENTS ON IT *W*I*T*H*O*U*T* *M*Y* *H*E*L*P!!!!!!!!! This is a man who needed a command sheet to load telix on the much lamented family XT. That alone is enough to make me a believer. I use a pc. I loath windows, particularly after laboring for days to make it run, and failing. However, apart from windows, and bill gates, and most anything to do with microsoft or big blue, being the embodyment of evil, there are no absolute truths, and like in religion, one persons god is anothers devil. You and your ilk are like fundamentalists, imposing your own strict and repressive idealogy on e innocent people who were living quite hapily until you started making a stink. To summarize: stick your elitest ideas in the same bodily orifice where the suits have their heads firmly placed, and send flames to /dev/null. ------------------------------ From: alfie@dcs.warwick.ac.uk (Nick Holloway) Subject: Re: Linux Logo Postscript Date: Thu, 15 Apr 1993 13:56:12 GMT In <1q3fonINNnb5@nestroy.wu-wien.ac.at> herz@d26.wu-wien.ac.at (Thomas Herz) writes: > %!PS-Adobe-2.0 EPSF-1.2 > %%Title: Untitled > %%Creator: PageMaker 4.0 rocky > %%CreationDate: 8-12-1992, 16:3:45 > %%For: > %%BoundingBox: 0 0 612 792 This looks like the original logo posted by James Callinson. I did a smaller hand-crafted version. If this logo is to be used, I would push for using my version (keeping attribution for the original artist -- my part was mostly technical). This is not just so that I can blow my own trumpet (or nose), but for the following reasons: o people have reported problem with using the 'setcolor' operator -- I use 'setrgbcolor'. o the bounding box above is rubbish -- this makes inclusion in documents difficult. o it is too large -- it is almost 53k -- mine is less than 1k (just :-). o my version should be easily updatable to different releases, even by a novice. My own personal feeling is that I quite like this logo, I think it would look snazzy on a CD. However, I feel the final choice must lie with Linus -- it is his OS. Anyway, here is my version, which has been available on tsx-11 for since 0.97. This is marginally updated -- it is for "Linux 1.0" :-) -- Nick Holloway | `O O' | alfie@dcs.warwick.ac.uk, alfie@warwick.UUCP, [aka `Alfie'] | // ^ \\ | ..!uunet!mcsun!uknet!warwick!alfie %!PS-Adobe-3.0 EPSF-3.0 %%Title: Linux Logo %%For: Nick Holloway %%+ based on original by James P. Callison %%BoundingBox: 100 100 316 292 %%DocumentFonts: Times-Roman %%EndComments gsave 100 292 moveto 208 100 lineto 316 292 lineto closepath clip newpath 192 256 div setlinewidth 1 1 255 { dup 192 mul 256 div 100 add 100 exch moveto 216 0 rlineto 1 sub 254 div dup 1 exch sub 1 index 0.998 mul 1 index 0.366 mul add 3 1 roll % R 1 index 0.997 mul 1 index 0.301 mul add 3 1 roll % G 1 index 1.000 mul 1 index 0.787 mul add 3 1 roll % B pop pop setrgbcolor stroke } for grestore gsave 0.879 0.000 0.000 setrgbcolor 100 195 moveto 100 205 lineto 316 205 lineto 316 195 lineto closepath fill grestore /Times-Roman findfont [1 0 0.21 1 0 0] makefont dup 120 scalefont setfont 110 180 moveto (L) show dup 60 scalefont setfont 147 206 moveto (INUX) show 18 scalefont setfont 192 182 moveto (RELEASE 1.0) show showpage %%EOF ------------------------------ ** 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-Activists-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux) via: Internet: Linux-Activists@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 tupac-amaru.informatik.rwth-aachen.de pub/msdos/replace The current version of Linux is 0.99pl6 released on February 23, 1993 End of Linux-Activists Digest ****************************** ADVERTISMENTSdigest849 644 36676 74430 51410 5366137100 5754 From: Digestifier To: Linux-Activists@senator-bedfellow.mit.edu Reply-To: Linux-Activists@senator-bedfellow.mit.edu Date: Thu, 15 Apr 93 13:00:35 EDT Subject: Linux-Activists Digest #849 Linux-Activists Digest #849, Volume #4 Thu, 15 Apr 93 13:00:35 EDT Contents: Re: Basic info on LINUX. (Alan Cox) Re: NFAQ: Do ISA cards cripple EISA performance ? (Alan Cox) Re: Anyone using Orchid VA & S3? (statnfs) Re: Disk Quotas (was Re: New feature for the filesystems.) (Klaus Steinberger) Re: Compaq Deskpro 386/20 & Linux ? (Kim Ieong) I hate Linus Torvalds (Derek Glidden) Linux "mascot" animal suggestions (Jens Goerke) Re: need help moving linux to a new hard drive (Drew Eckhardt) RE: WHY BOTHER? (Ray Messier) Linux (Rick Christian) Seen-by lines (Dave James) Re: [HELP] XWindows 8 MEG RAM req??? (Chris Dreke) A couple of questions: (Derek Glidden) Re: flex vs. lex (Andrew Bray) dosemu cursor positioning bugfix (Matthias Urlichs) Re: CD ROMS on the dosemulator (Matthias Urlichs) Re: FAQ? Booting MSDOS from partition 1 fails (John Granieri) Re: 99p8 breaks xdm -- authentication problems? (Timo Korvola) Re: menu problem solved (Eckard Brauer) ---------------------------------------------------------------------------- From: iiitac@swan.pyr (Alan Cox) Subject: Re: Basic info on LINUX. Date: 15 Apr 93 12:45:34 GMT > 2Mb is the absolute minimum you need BZZT wrong. If you include only the things you really need - ie no scsi, no tcp/ip, and only minix fs you can run some things in 1MB. If you trim the kernel page tables and file buffers down you can combine this with a shared lib init and getty and run many things. I used a 1Mb 386 like this for a while to play games and edit documents. Granted its no good for compiling (hello world took 4 minutes). It's all a matter of what you use it for. The killer with linux is the applications - notably some of the Gnu stuff and X. If someone ported something like bcc386 from minix I'd expect a 1Mb machine wouldn't be too bad for compiling either. Alan ------------------------------ From: iiitac@swan.pyr (Alan Cox) Subject: Re: NFAQ: Do ISA cards cripple EISA performance ? Date: 15 Apr 93 12:51:14 GMT Yes they do is the quick answer. Only it's not the cards that can cause problems, its the bus access times. The problem is that while your EISA cards are working away chatting to the cpu everything is great. An ISA card still behaves like an ISA card so anything doing a lot of ISA bus cycles reduces your PC to ISA speed for that time. The common big culprit is DMA based ethernet cards. If you are receving a lot of broadcast or normal packets you can spend a lot of the time waiting for the ISA bus operations. You never get worse than ISA, but an ISA card in an EISA machine is as much a slowdown and problem as an ISA card in an ISA machine. Unless you have a lot of heavy traffic to ISA cards forget it, even if you do its not that desperate a problem. Alan ------------------------------ From: statnfs@tina.uoregon.edu (statnfs) Subject: Re: Anyone using Orchid VA & S3? Date: Thu, 15 Apr 1993 10:14:06 GMT Do ordinary VESA boards work with Linux? I have a Boca Research SVGAX2 that allows 640x480x16.8mil and 1024x768x256. Can I use any of these modes in Linux? If not, does anyone know what the SVGAX2 is compatible with? I'd hate to have to use mono VGA with such a nice board... If it comes to the crunch, has anyone written a VESA or SuperVGA driver for Linux? I'd like to look at the sources (although I have no technical info for my board)... Rui Carmo ------------------------------ From: k2@bl.physik.tu-muenchen.de (Klaus Steinberger) Subject: Re: Disk Quotas (was Re: New feature for the filesystems.) Date: Thu, 15 Apr 1993 14:06:50 GMT davidsen@ariel.crd.GE.COM (william E Davidsen) writes: >In article , s_fuller@iastate.edu (Steve Fuller) writes: >| 1) If the file is in the middle of a write when you exceed your >| quota, the file will be written in its entirety and not >| truncated. Any furthur write requests will be denied > I certainly don't want a quota like that. If the hard limit isn't hard >it doesn't protect the system. unless writes STOP at the hard limit the >user could run the system out of disk anyway. I agree, BSD quotas are implemented in this way, a hard limit stops any further write. Sincerely, Klaus >-- >bill davidsen, GE Corp. R&D Center; Box 8; Schenectady NY 12345 -- Klaus Steinberger Beschleunigerlabor der TU und LMU Muenchen Phone: (+49 89)3209 4287 Hochschulgelaende, D-8046 Garching, Germany FAX: (+49 89)3209 4280 !!! after 1. July new ZIP Code 85748 Internet: Klaus.Steinberger@Physik.Uni-Muenchen.DE ------------------------------ From: ieong@fraser.sfu.ca (Kim Ieong) Subject: Re: Compaq Deskpro 386/20 & Linux ? Date: Thu, 15 Apr 1993 14:23:35 GMT p7003ag@sun1.LRZ-Muenchen.DE () writes: I have the same problem with my Deskpro 386/20. But I have a MFM disk controller and 132 MB hard drive. I just will not mount the disk. Anyone has any suggustion? Kim Ieong >Hiyo, >I got a prob with my Deskpro 386/20: I can't install Linux. >After checking all the hardware I have (2 serial, 1 parallel, >AHA1542B, Conner 200MB SCSI) the kernel tries to reset my >harddisk controller and fails. This it does in an indefinite loop. >Any controller I care to try (AT-Bus, MFM ...) produces the same >effect. On a NONAME 486 with AT-Bus HD it works just fine. >Has anyone encountered similar problems or is anybody running >Linux on a Deskpro 386/20 already ? >Please help me. >CU >Alaric ------------------------------ From: Derek.Glidden@f42.n377.z1.fidonet.org (Derek Glidden) Subject: I hate Linus Torvalds Date: Sun, 11 Apr 1993 13:24:38 -0800 >> I think that it would be easier for most Linux-heads to relate to "My >> Wife Hates Linus Torvalds." (Perhaps that should be changed to >> "Spouse," but I get the impression that most Linux enthusiasts, like >> most computer geeks, are male.) There are (I'm sure) loads of us out here who'd be better off with "My Girlfriend Hates Linus Torvalds" logos to wear. And perhaps add, in smaller letters underneath, "But my phone company loves him!" :) * Origin: The Slut Club -=- (813) 689-1150 -=- USR/DS 16.8K (1:377/42.0) ------------------------------ From: Jens.Goerke@p42.f310.n242.z2.fidonet.org (Jens Goerke) Subject: Linux "mascot" animal suggestions Date: Wed, 07 Apr 1993 21:52:56 -0800 Hello, *! Since Linux reminds me of the Peanuts character Linus and was written by someone with this name (a big thanks), why not use "Linus with his blanket" or, if you insist on an animal, Woodstock... Have Fun Jens ===================== What do You get when You multiply six by nine? 42, base 13 * Origin: Point 42 (6*9) (2:242/310.42) ------------------------------ From: drew@kinglear.cs.colorado.edu (Drew Eckhardt) Subject: Re: need help moving linux to a new hard drive Date: Thu, 15 Apr 1993 14:21:06 GMT In article <15APR199302132006@sigma.tamu.edu> you write: >did 'mkefs /dev/hda2 140224' (140224 is the size as reported by fdisk) >did 'mount /dev/hda2 /new' Mount with no file system type given will attempt to mount the file system as the default type - older Linux kernels used Minix, newer kernels may use something different. Do mount -t ext /dev/hda2 /new . Also, you *don't* want to be using the ext file system, it's painfully slow because it uses free lists instead of bitmaps, and it has dumb allocation routines. Instead, use the ext2 filesystem (the code is in recent Linux kernels, the programs are available from tsx-11.mit.edu:/pub/linux/alpha/ ext2 or some other intuitive directory). It uses bitmaps to track free blocks, has a clean flag, replicated super blocks for better reliability, saner block allocation routines that will help to prevent fragmentation and that keep inodes close to their files and indirect blocks, etc. Much better. Also, you'll need swap space, since you don't want to swap to a file because of performance reasons, you should make a separate swap partition. > >this is where the problem sets in, i get 'unknown error xxxxx' >i also tried modifying the fstab file, to no avail. you need to change the fstype field (ie, "minix") to "ext" >i am new to this, if it is not obvious. what i want, if someone can tell >me how to get there, is: >to have the 120 back as HD0, and the 240 as HD1, and have the boot floppy >boot to partition 2 on the 240. Mount whatever you're copying to on new, as root do (cd /; tar cf - `echo * | sed s/new//g` ) | (cd /new; tar xfp -) This will install symlinks and device files correctly on the new partition, but hard-linked files will end up with multiple files on the destination. The following shell script should correct the problem with multiple copies of hardlinked files (albeit a bit inefficiently) : #!/bin/sh src=/ dest=/new current_inode=0 find . -xdev -type f -a -links +1 | xargs ls -i | sort -n | ( while read input; do inode=`echo $input | awk '{print $1}'` file=`echo $input | awk '{print $2}' | sed "s,^./,,g" ` if [ $inode -ne $current_inode ]; then current_inode=$inode link_to=$file else rm ${dest}/$file ln ${dest}/$link_to ${dest}/$file fi done ) -- Boycott USL/Novell for their absurd anti-BSDI lawsuit. | Drew Eckhardt Condemn Colorado for Amendment Two. | drew@cs.Colorado.EDU Use Linux, the fast, flexible, and free 386 unix | ------------------------------ From: Ray.Messier@f52.n203.z1.fidonet.org (Ray Messier) Subject: RE: WHY BOTHER? Date: Sat, 10 Apr 1993 11:25:50 -0800 Thanks for the response, my docs do say MSDOS, but I tried msdos like you said and it worked fine! Thanks! * Origin: 24th Street Exchange - A BBS Since 1983 * (916) 451-7179 (1:203/52) ------------------------------ From: Rick.Christian@f220.n129.z1.fidonet.org (Rick Christian) Subject: Linux Date: Fri, 09 Apr 1993 19:02:04 -0800 Hello All! I would like pros & cons of doing the following: Installing Linux in place of a SCO XENIX 386 System V system, which runs on a Tandy 386 computer, and uses Tandy DT-100 terminals(non graphic). Goal would be to: Get a UNIX type system running in place of the SCO system, more std. than SCO Use the existing software, Wordperfect for Unix & some Excaliber (PD software) Add IBM-PC 386SX+ type computer systems with VGA graphics to system Use DOS based programs on terminals & PC's At present the people who use this system have no knowledge on how to use the SCO system, other than the software installed. If this is a good idea or not. Rick * Origin: R.C. Consulting BBS Upper St. Clair, PA (1:129/220) (1:129/220) ------------------------------ From: Dave.James@f209.n209.z1.fidonet.org (Dave James) Subject: Seen-by lines Date: Mon, 12 Apr 1993 07:55:54 -0800 Hello Greg, here ya go good budy... PATH: 129/220 11 3 109/25 396/1 PATH: 242/310 308 302 2405/100 13/13 PATH: 377/42 15 3603/20 396/1 PATH: 350/341 343/94 34 300 138/112 13/13 Dave * Origin: -=['dBest' Western Star]=- HST 16.8 V32b (1:209/209) ------------------------------ From: Chris.Dreke@f460.n3625.z1.fidonet.org (Chris Dreke) Subject: Re: [HELP] XWindows 8 MEG RAM req??? Date: Mon, 12 Apr 1993 15:00:00 -0800 Well, I am able to run X windows on 4 meg + swap space, but it is too slow to do much more than play tetris. If you want to run multiple applications, you will need more memory. * Origin: Vantage Point BBS (1:3625/460.0) ------------------------------ From: Derek.Glidden@f42.n377.z1.fidonet.org (Derek Glidden) Subject: A couple of questions: Date: Tue, 13 Apr 1993 16:35:27 -0800 ...which I know I've seen the answers to, but I can't find them now. 1) When I mount a DOS partition, it's mounted as: -rwx--x--x Which means root can get to it, but nobody else can. I'd like to be able to mount it 755 so other people can read from it, but not write to it. I know this has something to do with permissions on the mount executable and setuid or something like that, but I can't figure it out. I can chmod 755 /dos the mount point, but that only holds for the first directory under it, then everything is non-write-non-read again. 2) I've got Citadel/UX running just fine under Linux (Which, BTW, can be ftp'd from quartz.rutgers.edu, or F'req from 1:377/42 as CIT314.ZIP) which means I now have to get getty running on my serial port with the modem attached. I've got this done ok, but now about every 10 minutes or so I get a message saying "c7 respawning too fast, shutting down for 5 minutes" or somesuch. I know I've seen an answer to this one, too, but don't remember what it was. Responses via E-Mail to dglidden@bix.com please, since I can't keep up with the traffic here. * Origin: The Slut Club -=- (813) 689-1150 -=- USR/DS 16.8K (1:377/42.0) ------------------------------ From: acb@madhouse.demon.co.uk (Andrew Bray) Subject: Re: flex vs. lex Date: Thu, 15 Apr 1993 14:31:21 +0000 TITLE: Re: flex/lex In article fox@graphics.cs.nyu.edu (David Fox) w rites: > > You won't find a more "decent" lex package than flex. yyless is a > macro in the flex output. How this leads to a solution of the > problem of undefined yyless symbol is left as an exercise to the > reader. I have had some serious problems porting some code to flex - mostly due to the code being very poor. The code assumed that lex copied matched strings to a buffer, and did things like edit the buffer in situ, use yyless to insert extra characters into the stream, and all sorts of nasties. For gruesome things like this, I would really appreciate the availability of a more compatible lex package, though where it worked, I would use flex - it's so much faster. Regards, Andrew Bray ------------------------------ From: urlichs@smurf.sub.org (Matthias Urlichs) Subject: dosemu cursor positioning bugfix Date: 15 Apr 1993 16:22:11 +0200 dosemu mixes up X and Y cursor positioning. With this fix, a program which previously printed ugly nonsense now creates perfectly legible output, other than a few escape codes lying around on the screen... --- emu.c.orig Thu Apr 15 16:27:26 1993 +++ emu.c Thu Apr 15 16:25:14 1993 @@ -1199,8 +1199,8 @@ break; xpos[s] = x; ypos[s] = y; - *(unsigned char *)(0x450 + s) = ypos[s]; - *(unsigned char *)(0x451 + s) = xpos[s]; + *(unsigned char *)(0x450 + s) = xpos[s]; + *(unsigned char *)(0x451 + s) = ypos[s]; if (s == screen) poscur(x, y); break; NB: Said program often does things like this: outb [0x3d4] 0xe outb [0x3d5] 0x3 outb [0x3d4] 0xf outb [0x3d5] 0x1f I don't really want to know anything about DOS hardware, so if somebody knows what this is supposed to do... thanks. -- Never lick a gift horse in the mouth. -- Matthias Urlichs -- urlichs@smurf.sub.org -- urlichs@smurf.ira.uka.de /(o\ Humboldtstrasse 7 -- 7500 Karlsruhe 1 -- Germany -- +49-721-9612521 \o)/ ------------------------------ From: urlichs@smurf.sub.org (Matthias Urlichs) Subject: Re: CD ROMS on the dosemulator Date: 15 Apr 1993 16:35:53 +0200 In comp.os.linux, article <9783@dirac.physics.purdue.edu>, bcr@bohr.physics.purdue.edu (Bill C. Riemers) writes: > > I have a whole CD Rom just filled with DOS text applications that > should work under dosemu0.48. The problem is that I can not figure out > how to make dos recognize the CD Rom under the emulator. I tried mounting > it as /usr/dos, but it seems linux.exe can only recognize minix file systems? Ahem... I just tried with a ISO9660 CD-ROM... # mount -t iso9660 /dev/cda /cd # ln -s /cd /dos/cd # dosemu -A >/dev/null > c:linux D > cd D: > dir Volume in drive D has no label Directory of D:\ AUTOEXEC BAT 42 4-14-93 9:12p CONFIG SYS 45 4-15-93 4:30p CD 1-08-93 6:12p 3 File(s) 18998272 bytes free > cd cd :-) > dir Volume in drive D has no label Directory of D:\CD 00_TRANS TBL 668 1-08-93 6:21p ADMIN 1-07-93 9:30a BOOT TXT 2302 1-06-93 3:48p DOCS 1-07-93 9:30a [ etc. There should be a way to turn the Rock Ridge extensions off -- some file names (not those above) look very ugly.] > type boot.txt [ the file in question ] > This is patch 1 of dosemu 0.48 under Linux 0.99.8 with the sigdiff patch applied. Note that you can't (yet?) run .EXE files from under /usr/dos. -- The United States is like the guy at the party who gives cocaine to everybody and still nobody likes him. -- Jim Samuels -- Matthias Urlichs -- urlichs@smurf.sub.org -- urlichs@smurf.ira.uka.de /(o\ Humboldtstrasse 7 -- 7500 Karlsruhe 1 -- Germany -- +49-721-9612521 \o)/ ------------------------------ From: granieri@graphics.cis.upenn.edu (John Granieri) Subject: Re: FAQ? Booting MSDOS from partition 1 fails Date: 15 Apr 93 14:00:29 GMT In article Dag.H.Wanvik@kvatro.no (Dag H. Wanvik) writes: /dev/hda: C: 1 active MSDOS 2 Linux ex2fs 3 Linux ex2fs I then reinstalled MSDOS on partition 1, which first made me I then reinstallereformat partition 1 and then reinstalled, successfully, I was told by MSDOS setup :-(. I can mount the MSDOS partition under Linux, but MSDOS fails to boot from the harddisk. The MSDOS filesystem is OK, as I can 'dir' it when booting MSDOS from floppy. Somebody said MSDOS is touchy and needs to be installed *before* any Unix(Linux) on the other primary partitions. Is is the MBR (master boot record) on the disk which is the problem? When I try to boot, the BIOS tells me "missing operating system". Thanks, Dag -- Dag H. Wanvik, Phone: +47 7 520090 KVATRO AS, Pirsenteret, Fax: +47 7 520140 N-7005 Trondheim, NORWAY E-mail: dag.h.wanvik@kvatro.no ------------------------------ From: Timo.Korvola@hut.fi (Timo Korvola) Subject: Re: 99p8 breaks xdm -- authentication problems? Date: 15 Apr 93 15:46:30 GMT Reply-To: Timo.Korvola@hut.fi In article gthomas@fraser.sfu.ca (Guy Thomas) writes: I initialy installed the new kernel, 99p8, which caused xdm to fail. - xdm comes up and presents login prompt - login takes username and passwd - seems to pass control to xinit - then snaps back to the login I've had the same problem, had to switch back to 0.99.7 to run xdm. It seems that xinit runs .xinitrc, but the user's clients aren't allowed to connect to the server. Putting xhost + in your Xstartup disables the access control, so your clients (and anybody else's for that matter) can connect, but I wouldn't recommend this on a network machine. And a related question: has anybody gotten MIT-MAGIC-COOKIE-1 authentication to work? It doesn't seem to work (with network connections, That is), regardless of the kernel version. Doesn't the server support it? It seems to accept the -auth option, since xdm manages to start the server. If authentication doesn't work, why are local clients allowed to connect to the server when using 0.99.7 or older versions of the kernel? Also, why did the kernel upgrade change things? Finally, I tried to recompile xdm and noticed some header files missing: Where are linux/if.h, linux/if_arp.h and linux/if_route.h, which are supposed to be included by the respective files in /usr/include/net? Many questions, few answers... -- Timo Korvola Timo.Korvola@hut.fi ------------------------------ From: brauer@aix520.informatik.uni-leipzig.de (Eckard Brauer) Subject: Re: menu problem solved Date: 15 Apr 1993 09:47:56 GMT The file is named menu-new.tar.Z and uploaded to tsx-11 and nic.funet.fi today. E. ------------------------------ ** 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-Activists-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux) via: Internet: Linux-Activists@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 tupac-amaru.informatik.rwth-aachen.de pub/msdos/replace The current version of Linux is 0.99pl6 released on February 23, 1993 End of Linux-Activists Digest ****************************** atch applied. Note that you can't (yet?) run .EXE files from under /usr/dos. -- The United States is like the guy at the party who gives cocaine to everybody and still nobody likes him. -- Jim Samuels -- digest850 644 36676 74430 57740 5366137100 5760 From: Digestifier To: Linux-Activists@senator-bedfellow.mit.edu Reply-To: Linux-Activists@senator-bedfellow.mit.edu Date: Thu, 15 Apr 93 14:00:31 EDT Subject: Linux-Activists Digest #850 Linux-Activists Digest #850, Volume #4 Thu, 15 Apr 93 14:00:31 EDT Contents: Re: SOS:Can't install by a1.test & a1 (Carpe Diem.....) Installing Linux beyond 4th LOGICAL(extended) partition? (Edward Dickinson) can 386sx run linux (Onnei Cheng) Autocad (Vince Niel) Where does one find .au files? (lhanson) printing to/from remote machines (Bill E. Eickmeier) Re: Why Linux should remain Linux and seagulls should not be used. (James C. Tsiao) Re: how safe is minix filesystem in power failure? (Jim Graham) Re: how safe is minix filesys (Jim Graham) Re: Help on X-Windows(mostly) (Dietmar Braun) Re: NFS-Blues in 0.99-p8 (Hendrik G. Seliger) Re: Big swap partitions [> 16M] (Edward Seidl) chsh stopped working after kernel & lib upgrade (Sunando Sen) Re: Problems with Sony CDROM CDU-8013. (E. v. Pappenheim) Re: Tutorial for setting up lpr/lpd????? (Jan Willem Hubbers) ---------------------------------------------------------------------------- Crossposted-To: tw.comp.unix From: jsyu@Uz.nthu.edu.tw (Carpe Diem.....) Subject: Re: SOS:Can't install by a1.test & a1 Date: Thu, 15 Apr 1993 15:21:14 GMT Dear Netters Now I find what caused the following error message: > MINIX-fsmagic mathc failed > [MS-DOS FS Rel.alpha.8, FAT 16, check=n, conv=b, uid=0, gid=0, umask=000] > [me=0x50,cs=2,#f=246,fs=23412,fl=12574,ds=36424,de=32962,data=38484,se=22133, > ts=1963212550] > Unsupported FS Parameters > EXT-fs: magic match failed > Insert disk a3 into the floppy drive then hit enter, q to quit If I format the 1.2MB floppies with PCTOOLS R3.4 , the problem is gone. But if I format floppies with the FORMAT utility of MS-DOS 5.0 , the same error message will appear. It seems that Linux 0.99pl6 can't be installed from the floppies with FAT-16 made by DOS-format utility, while it can get alone well with the DOS FAT-12 created by PCTOOLS. But the other problem occured while installing from a1.test remains unsolved. Anyway, thanks to those who offered me opinions to solve my previous question. :-) Sincerely Jeanshall Yu ( jsyu@uz.nthu.edu.tw ) ------------------------------ From: egd@mole.cis.ufl.edu (Edward Dickinson) Subject: Installing Linux beyond 4th LOGICAL(extended) partition? Date: 15 Apr 1993 15:29:54 GMT On my hard disk, the fifth logical drive in the extended partition is available to install Linux. However, this is equivalent to "hda9" (according to FAQ #50), and the default system only has "hda1" to "hda8". This is a large disk (660meg), and it's inconvenient to move everything around. Is there a way to add a device "hda9"? I have diskettes a1 and a2 in Linux format, via rawrite. The others are DOS format awaiting installation. ------------------------------ From: abc@bass.bu.edu (Onnei Cheng) Subject: can 386sx run linux Date: 15 Apr 93 15:08:48 GMT hi, guys My friend has a 386SX, is it good to run Linux thanks in advance Onnei ------------------------------ From: comrade@hal.gnu.ai.mit.edu (Vince Niel) Subject: Autocad Date: 15 Apr 1993 12:01:03 -0400 Before everyone here jumps to conclusions, here me out first. I want to make my whole hard drive linux, and then use the linux dos emulator to run Autocad release 11 in a Window. My goal is to run autocad while someone can login to the Linux system. I don't really care how this is done. I only have 1 line and a 486/33 with 4 megs of ram. Maybe its possible to make a dos partition and you Desqview386 or windows to task. I will try running autocad under the dos emulator but have said it won't work. Please help me out. Thanks for your time. Mike comrade@gnu.ai.mit.edu -- Comrade Winston | and the three men I admire most, the father son Voice/Data: 615/599/7154 | and the holy ghost, they caught the last train comrade@gnu.aim.it.edu | for the cost, the day the music died... ------------------------------ From: lhanson@moose.uvm.edu (lhanson) Subject: Where does one find .au files? Date: Thu, 15 Apr 1993 15:21:13 GMT I picked up the play-au.c code from sunsite and I would like to try it on my PCs speaker. Are the SUN *.au files PD? If so where can I find them? I've checked ARCHIE but can't come up with anyting. Lee ------------------------------ From: bee@watarts.uwaterloo.ca (Bill E. Eickmeier) Subject: printing to/from remote machines Date: Thu, 15 Apr 1993 15:41:52 GMT I'm having a bit of trouble getting my linux box to send print jobs to remote printers. As well, remote machines cannot successfully send print jobs to my linux machine. At one point, I was able to send remote jobs to my linux machine, but for some reason, things stopped working. Local printing does work though - while logged into my linux machine I can print to the local printer on my linux machine without problems. Here's how things look when I try to print from my linux machine to a remote printer: [29]% lpr -Ppslaser test.ps [30]% lpq -Ppslaser linuxbox: Warning: no daemon present Rank Owner Job Files Total Size 1st bee 19 test.ps 16772 bytes connection to otherbox is down [31]% ping otherbox PING otherbox.uwaterloo.ca (129.97.42.118): 56 data bytes 64 bytes from 129.97.42.118: icmp_seq=0 ttl=59 time=0 ms 64 bytes from 129.97.42.118: icmp_seq=1 ttl=60 time=0 ms --- otherbox.uwaterloo.ca ping statistics --- 2 packets transmitted, 2 packets received, 0% packet loss round-trip min/avg/max/max = 0/0/0 ms [32]% ps -aux | grep lpd root 469 0.0 3.1 68 220 ? S 10:45 0:00 lpd root 449 0.0 2.3 60 164 ? S 10:02 0:00 lpd bee 473 0.0 1.9 76 136 p1 S 10:50 0:00 grep lpd ===== [33]% uname -a Linux linuxbox 0.99.pl6-26 03/01/93 i486 ========= [34]% ls -l /etc/lpc /etc/lpd /etc/lpr /usr/bin/lpq /usr/bin/lpr /usr/bin/lprm -rwxr-sr-x 1 daemon daemon 21508 Nov 19 23:32 /etc/lpc -rwxr-sr-x 1 daemon daemon 37892 Nov 19 23:32 /etc/lpd -rwxr-sr-x 1 daemon daemon 17412 Nov 19 23:32 /etc/lpr -rwxrwsr-x 1 daemon daemon 17412 Nov 19 23:32 /usr/bin/lpq -rwxrwsr-x 1 daemon daemon 17412 Nov 19 23:32 /usr/bin/lpr -rwxrwsr-x 1 daemon daemon 17412 Nov 19 23:32 /usr/bin/lprm ===== [35]% ls -alR /usr/spool/otherbox total 5 drwxrwxr-x 2 daemon daemon 1024 Apr 14 16:52 . drwxrwxr-x 13 root root 1024 Apr 15 09:54 .. -rw-rw-r-- 1 daemon daemon 4 Apr 14 16:35 .seq -rw-rw-r-- 1 daemon daemon 4 Apr 15 09:55 lock -rw-rw-r-- 1 daemon daemon 36 Apr 14 16:45 status ===== [36]% ls -l /etc/printcap -rw-r--r-- 1 root root 486 Apr 15 09:54 /etc/printcap ===== [37]% cat /etc/printcap # # dot matrix on linux box lp|lplinuxbox|lp_linuxbox:lp=/dev/lp1:ra=*:sd=/usr/spool/lp1:sh: # # remote laser printer pslaser|PostScript LaserWriter:\ :rp=pslaser:lp=:\ :rm=otherbox:sd=/usr/spool/otherbox:\ :lf=/usr/adm/lpd-errs ===== [38]% ls -l /etc/hosts.lpd -rw-r--r-- 1 root root 116 Apr 12 11:02 /etc/hosts.lpd ===== [39]% cat /etc/hosts.lpd otherbox.uwaterloo.ca ===== [40]% grep 35/tcp /etc/services lp 35/tcp # any private printer server [JBP] printer 35/tcp # any private printer server [JBP] Here's how things look when I try to print from a remote machine to my linux printer (this originally worked fine, but has since stopped): [23]% lpr temp [24]% lpq Spool queue: lp_linuxbox Local (waiting to be sent to remote site): otherbox: waiting for linuxbox to come up since Wed Apr 14 15:18:55 1993 (errno=60 'Connection timed out') 1st bee 10 temp 3030 bytes Remote(linuxbox): ...right here I hit ctrl-c since it seems to hang... Any and all ideas are welcome... Thanks, Bill Bill E. Eickmeier bee@watarts.uwaterloo.ca Department of Psychology, University of Waterloo, Waterloo, Ontario, N2L 3G1 room: PAS 1102 phone: 519-885-1211 x6638 fax: 519-746-8631 ------------------------------ From: jjctc@lhdsy1.lahabra.chevron.com (James C. Tsiao) Subject: Re: Why Linux should remain Linux and seagulls should not be used. Date: 15 Apr 93 15:47:32 GMT In article <1993Apr15.050119.5106@sol.UVic.CA> pmacdona@sanjuan (Peter MacDonald) writes: >In article <93104.131126ACPS7221@RyeVm.Ryerson.Ca> writes: >> How about a Tiger? A Tiger is powerful, fast, sleek, efficient, and so is >> Linux. >> And also, Linus is a roarin' good programmer:) > >Nice, but maybe too aggressive. Has anyone suggested a Dolphin (dolFINN :-). Umm, I hate to see someone, especially some rabid 386bsd fan :-), changing that into DULL-FINN. Perhaps a porpoise... You know, an OS with a porpoise. James. -- jjctc@chevron.com | The shortest distance between jtsiao@netcom.com | two puns is a straight line. ============================================================================= Ask me about Linux, the freely distributable Unix clone. ------------------------------ From: jim@n5ial.mythical.com (Jim Graham) Subject: Re: how safe is minix filesystem in power failure? Date: Wed, 14 Apr 1993 14:52:14 GMT In article <1993Apr13.042529.12851@monu6.cc.monash.edu.au> ent811l@monu6.cc.monash.edu.au (Christopher Kuperman) writes: >but it sounds like you got a real problem with this power thing... uh huh... we're expecting more thunderstorms tonight (90% chance, they say). of course, we all know what that means (Jim shuts his system down). >I personally would do one of two things... > >a> Buy an Uninterruptable Power Supply (UPS) - not very cheap :( way, way out of my budget.... believe me, that's the first thing I thought of (all I'd need would be a very tiny one capable of keeping the computer up for a few seconds). when I get a job again, that's high on the list (along with more hard disk, more memory, etc.). >b> Do what my friend did and get hold of 3 or 4 car batteries and make > one... well first off, use gell-cells instead of car batteries---they're much safer (acid from the car batteries will destroy anything it touches).... I'm not sure what this does to the cost, but there's no way I'd use car batteries for any kind of indoor power supply. this is also why most Amateur Radio ops I've talked to who want to go portable (e.g., hiking trip, etc.) with their rigs carry gell-cells instead of car batteries. :-) > you can do this in two ways.. but both require you to make > some detection circuitry so it detects the power loss or power unstab- > ility. > 1. Buy a kit or a commercial voltage inverter, they are typically > things that DIck Smith or the like sell for people to run their > 240VAC applications from a car battery, and with 3 or 4 batt's > you'll be cruising, but i dont recommend using these inverters for > long lengthas of time coz they make a very crude sine wave and > even sometimes only make a square wave with lots a harmonics. > 2. make a real simple ``regulator'' (12VDC,5VDC etc..) and run > the computer direct from the batteries when the power fails. > Isnt it great the way computers run off 12V and 5V. actually, the way UPSs normally work is more along the same lines as the battery in a telephone central office (switching center). basically, the entire phone system (including your telephone, provided it isn't one of the ones, such as cordless phones, that wants external AC power) runs off of the battery the whole time, whether external power is there or not. the external power, when it exists, is used to keep the battery charged. (in the case of the telco, there are *MASSIVE* generators to take over the charging of the battery when power failure is sustained.) UPS systems for computers (and other devices, too, of course) work the same way. the computer/whatever runs off of the battery, and the AC power keeps the battery charged. this, btw, is why there isn't a hit when the power fails---the computer never sees a change in its power source, because there isn't a change. I've seen demos of some UPSs where you see a scope showing the output voltage level at the time of the power failure, and it's perfectly clean, w/ no fluctuation at all in voltage or phase....rock solid. same for when the power came back---no change. of course, this was also a very high-end UPS system, too...not sure what the cheaper ones would look like in that same demo. anyways, this is more than a bit off the subject of this group, but I'm not sure where to re-direct followups on this one...especially since it'd almost certainly be a group I have no access to. :-( later, --jim -- #include 73 DE N5IAL (/4) ============================================================================== INTERNET: jim@n5ial.mythical.com | j.graham@ieee.org ICBM: 30.23N 86.32W AMATEUR RADIO: n5ial@w4zbb (Ft. Walton Beach, FL) AMTOR SELCAL: NIAL ============================================================================== E-mail me for information about KAMterm (host mode for Kantronics TNCs). ------------------------------ From: jim@n5ial.mythical.com (Jim Graham) Subject: Re: how safe is minix filesys Date: Wed, 14 Apr 1993 14:56:08 GMT In article <14206.1189.uupcb@satalink.com> john.will@satalink.com (John Will) writes: [I wrote]: >> again, if I do expect power failures, I shut down, period. but sometimes, >> the power fails before the storm hits, etc. [....] >> any thoughts on this? >Can you say UPS? see my other reply.... that was the first thing that came to mind. the second thing that came to mind was my budget (and the lack thereof). when you're in what some choose to call a career transition stage (i.e., you don't have a job), money isn't quite as easy to come by. actually, I've yet to see a UPS that I could have afforded even before getting laid off. later, --jim -- #include 73 DE N5IAL (/4) ============================================================================== INTERNET: jim@n5ial.mythical.com | j.graham@ieee.org ICBM: 30.23N 86.32W AMATEUR RADIO: n5ial@w4zbb (Ft. Walton Beach, FL) AMTOR SELCAL: NIAL ============================================================================== E-mail me for information about KAMterm (host mode for Kantronics TNCs). ------------------------------ From: dietmar@techfak.uni-bielefeld.de (Dietmar Braun) Subject: Re: Help on X-Windows(mostly) Date: Thu, 15 Apr 1993 15:19:38 GMT In article <1993Apr15.090625.4637@klaava.Helsinki.FI>, torvalds@klaava.Helsinki.FI (Linus Torvalds) writes: |> In article <28617@galaxy.ucr.edu> jmward@ucrengr (jonathan ward) writes: |> >Hi there. I'm relatively new to Linux(awesome OS), and just got the SLS releaseoff of susnsite(0.99pl6). I'm tring to configure X to run on my machine, which is: generic 486DX-50 VLB motherboard, trident tvga8900c video card with 1meg ram, maxtor LXT34|> 0A IDE hard drive, a three button serial mouse. I finally got X |> >up and running(it defaults to the open windows manager) but I have this problem: |> >The mouse, once started is generally unresponsive. I can move it repeatedly, |> >and it occasionally "jumps" around the screen. When I depress a button, it |> >then takes a while to respond. I only have 4mb of RAM, and I'm swapping to a |> >12 mb swapfile. Is the mouse problems due to the lack of ram, or something I |> >overlooked? Any suggestion appreciated. |> |> If getting more memory is totally out of the question, may I suggest |> changing the swapfile into a partition. I know that re-fdisking is a |> pain, but swapping to a partition is *noticeably* faster than swapping |> to a file, and it can make a sluggish X11 seem almost usable. Also, as |> has already been recommended, the mono-X11 is much less memory-hungry |> than the colour version, and would also help the problem. |> Another _very fast_ solution is to swap to a partition on an other disk. With linux fs on one disk and swap partition on another disk, head movement is much less and X with 4MB is nearly as usable as with 8MB. I've installed linux on my friends pc using on HD for DOS and Linux swap and the other for linux fs, and had really been impressed. |> Note that using a swapfile is easy and painless, but it's really meant |> only for those cases where you don't really need to dip into swap all |> the time (when you have enough memory for most things, but need the |> swapfile for those occasional times when you compile big files while |> running X at the same time etc). If you are serious about swapping, try |> to do it using a swap partition. |> |> Linus Dietmar ===== Dietmar Braun; University of Bielefeld (finger dietmar@techfac.techfak.uni-bielefeld.de) ------------------------------ From: hank@Blimp (Hendrik G. Seliger) Subject: Re: NFS-Blues in 0.99-p8 Date: Thu, 15 Apr 1993 14:37:44 GMT Reply-To: hank@Blimp.automat.uni-essen.de torvalds@klaava.Helsinki.FI (Linus Torvalds) writes: : I think it's the 'access()' bug that was uncovered by the pl8 NFS code. : It's not actually a NFS bug, but some silly coding in the sys_access() : routine in linux/fs/open.c - the following short and simple patch should : fix it. The patch is also part of the ALPHA-diffs I made for the msdos : people due to the vm86 signal problem.. .... : Linus Thanks, Linus, that patch worked. Hank. -- ====================================================================== Hendrik G. Seliger Universitaet Essen hank@Blimp.automat.uni-essen.de Schuetzenbahn 70 Tel.: +49-201-183-2898 4300 Essen, Germany ====================================================================== Pajari: "Handling interrupts is simple." A. Tanenbaum: "Interrupts are an unpleasant fact of life." ------------------------------ From: seidl@alw.nih.gov (Edward Seidl) Subject: Re: Big swap partitions [> 16M] Date: Thu, 15 Apr 1993 15:39:10 GMT In article <1993Apr15.130117.19367@klaava.Helsinki.FI>, torvalds@klaava.Helsinki.FI (Linus Torvalds) writes: |> In article <1993Apr15.102608.25039@alw.nih.gov> seidl@alw.nih.gov (Edward Seidl) writes: |> >In article <1qj6u8$s6l@news.cs.tu-berlin.de>, wong@cs.tu-berlin.de (Wolfgang Jung) writes: |> >|> |> >|> Q: How many Swapareas is Linux able to handle |> >|> |> > |> >Eight. |> |> In the default configuration, yes. I think the way I encode the |> swap-page numbers actually restrict the number of swap-areas to 127 or |> something like that, but you should be able to get more than 8 by just |> changing a #define. |> |> Linus Although if you do this, you'll have to redefine MAX_SWAPFILES in free.c from the ps-0.99.6 source. Would the procps version of free have to change at all? Ed -- Edward Seidl DCRT/CBEL Bldg 12A, Rm 2033 seidl@alw.nih.gov NIH, Bethesda, MD 20892 It's spelled Luxury-Yacht, but it's pronounced Throat-Warbler-Mangrove. ------------------------------ From: sens@FASECON.ECON.NYU.EDU (Sunando Sen) Subject: chsh stopped working after kernel & lib upgrade Date: Thu, 15 Apr 1993 16:19:12 GMT Just a minor problem. The "chsh" program stopped working recently during the upgrade process (I use this program so little that I can't even tell at what version of the kernel and library it stopped working). Currently, my setup is Linux 0.99.8 and libc 4.3.3. The /etc/passwd and the /usr/bin/chsh both belong to root, and the modes are 0755 and 4755, respectively. The point is, chsh doesn't work, even when I am logged in as root. So the permissions cannot be the problem (and they are correct anyway, I think). When I execute chsh, it lists all the shells from the file /etc/shells. But when I actually choose a shell from the list by its number, it just says "/bin/ksh not found" (if that was the shell chosen) and redisplays the list. The only thing I can do is to press Ctrl-C to break out of the loop. This happens with both the chsh that came with hlu's basedisk and Rik Faith's utila package, by the way. Does anybody have an explanation? Thanks in advance. Sunando Sen ------------------------------ From: edi@edefix.han.de (E. v. Pappenheim) Subject: Re: Problems with Sony CDROM CDU-8013. Date: 14 Apr 93 18:41:06 GMT Eric Youngdale (eric@tantalus.nrl.navy.mil) wrote: : In article <1psblm$6ej@news.cerf.net> glasgow@nic.cerf.net (Michael Glasgow) writes: : >I have a Sony CDROM Model CDU-8013, and can not mount a CD. : >This is a SCSI CD Player, and I am using an Adapted 1542 : >SCSI Controller. : > : >Everytime I try to mount a CD, I get the following message: : > : >Panic: Multiple Volume CD Mounted. : The problem is that the Sony comes up with 512 byte sectors, and most : other cdrom drives come up with 2048 byte sectors. I have made patches to the : kernel which should allow you to mount and use your drive, and they appear in : the pre-pl8 kernel release. : -Eric My CDU 8012 works fine when a CD is in the drive during boot (tried with 0.99-pre8 and 0.99.8), in any other case I get the same message as described above. Cheers, Ecki : -- : "When Grigor Samsa woke up one morning from unsettling dreams, he : found himself changed in his bed into a monstrous vermin." : -F. Kafka -- ======================================================================== * Eckebrecht von Pappenheim Email: edi@edefix.han.de * * Eleonorenstr. 17 Phone: +49 511 443755 * * 3000 Hannover 91 * * Germany * ======================================================================== ------------------------------ From: jwhub@sci.kun.nl (Jan Willem Hubbers) Subject: Re: Tutorial for setting up lpr/lpd????? Date: 15 Apr 93 15:49:49 GMT In s_fuller@iastate.edu (Steve Fuller) writes: >Does someone out there have a really easy step by step tutorial >on what to do to set up the line printer daemon and ALL >associated files in order to get lpr to work??? Any and all leads >appreciated. This works for me (SLS 1.0 with 0.99.6 kernel): find lpr, lpd, lpc and lprm (some appear in 2 directories): chown root lpr lprm lpd lpc chmod 4755 lpr lprm lpd lpc chmod 0660 /usr/spool/lp1 chown daemon.daemon /usr/spool/lp1 And make sure you start lpd somewhere (/etc/rc.local is a nice place). >I am also looking for a etc/printcap entry for a Xerox 4045 laser >printer. I have a Diablo entry for it, but a native 4045 entry >would be oh so much nicer. :) >Thanks for the help >KEEP UP THE WORK LINUS!!!!!!!!!! ^ great >Steve >-- >---------------------=------------------------------ >Steve Fuller = >s_fuller@iastate.edu = No witty quote here yet... >---------------------=----------------------------- -- Jan-Willem Hubbers ( For HiFi [no, not out of a computer] ) Internet: jwhub@cs.kun.nl ( For /usr/local/gnu/bin/bash -login ) hubbers@cs.ruu.nl ( For fail-driven loops ) Fidonet : 2:283/218.2 ( For bouncing mail :-( ) ------------------------------ ** 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-Activists-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux) via: Internet: Linux-Activists@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 tupac-amaru.informatik.rwth-aachen.de pub/msdos/replace The current version of Linux is 0.99pl6 released on February 23, 1993 End of Linux-Activists Digest ****************************** bin/ksh not found" (if that was digest851 644 36676 74430 55342 5366137101 5756 From: Digestifier To: Linux-Activists@senator-bedfellow.mit.edu Reply-To: Linux-Activists@senator-bedfellow.mit.edu Date: Thu, 15 Apr 93 18:45:50 EDT Subject: Linux-Activists Digest #851 Linux-Activists Digest #851, Volume #4 Thu, 15 Apr 93 18:45:50 EDT Contents: anyone have a terminal emulator for televideos ????? (Ron McAllister) PORTING A PROGRAM TO OS/2? (Eric J Balog) X and two screens. (S.G. Theakston) Re: The dangers of playing with shared libraries (Bill C. Riemers) using logical drive of MS-DOS extended partition for Linux (Radivoj Suvacarov) Re: NFS problems, Linux as server (Joerg Lenneis) X11 questions (jehresman@eagle.wesleyan.edu) Compressed filesystems (Edwanji Lemorian) Re: Do I need intelligent serial I/O?? (Christopher Kuperman) Re: Long lines in vi (elvis) (Thomas Wieckhorst) Re: how safe is minix filesystem in power failure? (Mark Monnin) df / mount problem ? (jsacha@uoft02.utoledo.edu) Re: Does anyone ? (James C. Tsiao) Problems getting X running. (dpjunk@mmm.com) gnuplot make problem undre SLSp6 (Joe Panico) Re: Memory problem with Linux? (Thomas McWilliams) gcc and/or as (as86) assembler documentation (Richard Francis Jones III) Re: Announcing Babaya 1.00b (M. Saggaf) MAN: compression method (The Big Cheese) Ext-2 FS bug (Enrico Scotoni) ---------------------------------------------------------------------------- From: mcallist@smdis01.sm.aflc.af.mil (Ron McAllister) Subject: anyone have a terminal emulator for televideos ????? Date: 15 Apr 1993 14:22:10 -0400 Reply-To: mcallist@smdis01.sm.aflc.af.mil (Ron McAllister) Hi I am desperately searching for a terminal emulator that will emulate a televideo terminal, at least a 910 or better, a 921. a Kermit .ini file would be GREAT. Thanks much Ronan ------------------------------ From: balog@eniac.seas.upenn.edu (Eric J Balog) Subject: PORTING A PROGRAM TO OS/2? Date: 15 Apr 93 14:53:46 GMT Hi! I have a program called 'diamond.taz'; I ran uncompress and tar. Now I'm left with 'diamond'. How can I port this program to OS/2? Thanks for your help. I'll summarize and post. Eric Balog balog@eniac.seas.upenn.edu ------------------------------ From: sgt10@cl.cam.ac.uk (S.G. Theakston) Crossposted-To: comp.windows.x.i386unix Subject: X and two screens. Date: 15 Apr 93 17:25:49 GMT Since i had an old Hercules screen and card lying around i thought i'd patch the linux kernel to handle two moniters. It worked and was all very nice and clever. One moniter (the VGA one) is Virtual Consoles 1-4 the the Herc is 5-8, great i thought. Snag is, i don't use text mode much i use X. But when you start X all the virtual consoles disappear ( i knew this anyway). This makes my second moniter useless. Is their anyway i can either; run X on both screens, or run X on the VGA one and still use the hercules one for text??? Any ideas anyone? Setup: 486sx, 25Mhz, 8Mb RAM, generic VGA board, Trident8900C card, Hercules moniter, and an 8-bit Hercules card which must be a decade old if its a day. Cheers, Stu. -- _/_/_/ _/_/_/_/_/ _/ _/ Stuart Theakston Uni of Cambridge _/ _/ _/ _/ Internet: SGT10@phx.cam.ac.uk _/_/_/ _/ _/ _/ JANET: SGT10@uk.ac.cam.phx _/ _/ _/ _/ _/_/_/ _/ _/_/ ------------------------------ From: bcr@bohr.physics.purdue.edu (Bill C. Riemers) Subject: Re: The dangers of playing with shared libraries Date: 15 Apr 93 17:02:45 GMT In article <1993Apr14.190247.27211@Princeton.EDU> you write: >In article <1993Apr14.095640.28923@nessie.mcc.ac.uk> LeBlanc@mcc.ac.uk writes: >>In article <1993Apr13.025258.16555@Princeton.EDU> qpliu@princeton.edu writes: >I suppose I should have thought to use: > >(IFS=; while read x; do echo "$x";done >which works correctly. Nice, assuming you use sh or bash. Use tcsh users would have to use: set command = "\ while ( 1 == 1 ) \ set x = "\"\$\<\"" \ echo "\"\$x:q\"" \ end \ " (echo -n "";eval $command:q) < file >> output_file This would work under tcsh, but not csh. (I suspect a csh user would have no recourse!) Anyways, I think the point of these tricks is that they work, but are not obvious exspecially to a person who is panicing because nothing works! Anyways now that sash is available, this is really a mute point. Bill ------------------------------ From: suvac@edmund.cs.andrews.edu (Radivoj Suvacarov) Subject: using logical drive of MS-DOS extended partition for Linux Date: Thu, 15 Apr 1993 17:07:32 GMT I am curious about one thing. Is it possible to use a logical disk of MS-DOS partition for LINUX. I would like to have my MS-DOS logical drive converted to 2efs file system. This way I can save a partition or two for other operating system. Has anyone had any experience with this? Should/Could I do it? I guess if this is possible I could do the same thing with swap too. Regards, Radivoj -- ***************************************************************** * Radivoj Suvacarov * (616) 471-6882 (home) * * 500 Garland C-4 * 471-3251 (office) * * Berrien Springs MI 49103 * suvac@andrews.edu * ------------------------------ From: lenneis@wu-wien.ac.at (Joerg Lenneis) Subject: Re: NFS problems, Linux as server Date: 15 Apr 1993 17:40:57 GMT Bernhard Buettner (buettneb@guug.de) wrote: : Hello everyone! : : I am posting this on behalf of a friend who is using the : configuration described below. The problem ocurs when the Linux : machine is NFS mounted on the R3000. : : Computer #1 : Workstation, MIPS (R3000), Operating System RISCos 4.51 ~ AT&T UNIX V.4 : : Computer #2 : PC 486DX, 33MHz, wd0003, Linux 99pl5 : : NFS, Linux running as server. Server is version 1.0, also tested : version 1.2 with no difference. File system used on Linux machine : is Minix. We have tried this also with pl6 and pl7 but it made no : difference. : : On Client: : : # mount -t nfs /softland:/ /softland : # ll /softland : total 93 : 1 drwxr-xr-x 18 root root 448 Feb 21 01:57 . : 1 drwxrwxrwx 4 root other 512 Feb 21 11:42 .. : 17 -rw------- 1 root root 7720 Feb 22 10:21 .bash_history : ... : etc. : # ll /softland/etc : NFS readdir failed for server softland: RPC: Can't decode result [Rest deleted] You must mount from a Linux NFS server with rsize and wsize equal to 1024. Also, you should definitely use version 1.2 or the newest version 1.5 of the NFS server. -- Joerg Lenneis University of Economics and Business Adminstration Department for Applied Statistics and Data Processing Augasse 2-6, 1090 Vienna, Austria Tel. *43/222/31336 4758 email: lenneis@wu-wien.ac.at ------------------------------ Subject: X11 questions From: jehresman@eagle.wesleyan.edu Date: 15 Apr 93 12:30:25 EDT A few X-windows questions: 1) Where do I find more fonts for X, specifically postscript (adobe) fonts. I downloaded tk/tcl, but almost all of the test apps want adobe helvetica or adobe times. 2) Has anyone successfully built Interviews for X-free 1.2. I checked tsx-11 but the newest Interviews were built in Dec. (I think). Also is there a libX11.so.2.1 compatible with the new SLS? I have limited diskspace and would prefer to rebuild as few things as possible. -John Ehresman (jehresman@eagle.wesleyan.edu) ------------------------------ From: edwanji@matt.ksu.ksu.edu (Edwanji Lemorian) Subject: Compressed filesystems Date: 15 Apr 1993 11:16:43 -0500 This is a thought which just came to me. I don't know whether it's been proposed before. I've noticed the main concern with 'compressed' filesystems is demand-paging. It would seem to me that a good solution to this concern is to compress files in 4k blocks (I think linux pages in 4k blocks. My entire knowledge of linux specifics comes from this group) These compressed blocks could then be fit into a number of smaller-sized (1k or 512 byte) physical blocks. This would be slightly complicated because while physical disk-usage would be kept track of on a 1K/512 level, the files would be kept track of (what English skills) on a 4k level. Each I-node then would give information on a 4k chunk of a file and would possibly show: 1: the physical block to start on 2: the number of contiguous physical blocks this chunk is stored on 3: whether or not this chunk is compressed 4: number of bytes in the un-compressed chunk. This would be both for the last chunk of a file, and possibly for pages of executibles so that padding won't be un-nessisarily compressed and stored. 5: possibly a checksum to ensure data integrity Some concerns about this method: Since pieces must be stored in nunbers of contiguous (is that the right word?) physical blocks, fragmentation may become a problem. I.e. freed sections near the front of a partition will be too small to hold larger compressed chunks so they must be moved to the end. (This was a concern of mine with an idea someone posted earlier about compressing the 'physical' blocks on a disk) But I don't know much in this area, so that may not be a problem. Speed of access. This would only be a problem depending on what is stored under the filesystem. If it is text/source files. Then speed would not be a major concern. Even if it is exicutibles, as long as the time taken to un-compress the chunks won't have a major impact on preformance, it still won't be a big concern. Obviously, you wouldn't want to put a swap file in this kind of filesystem :-) (Unless having a larger swap space is a much larger concern than the large performance loss that would be incured) Any comments? Any information/assumptions I have incorect that would make this infeasible? -- Edwanji Lemorian ------------------------------ Crossposted-To: comp.sys.ibm.pc.hardware From: ent811l@monu6.cc.monash.edu.au (Christopher Kuperman) Subject: Re: Do I need intelligent serial I/O?? Date: Thu, 15 Apr 1993 18:55:06 GMT Bill writes: : : No, buy the serial port and modem. Each can be used for other things, : you can use the modem with your next computer (might not be a PC) or : upgrade the modem without changing the box. I hear that ISDN is big in : Europe, you might be able to get one of those beautiful ISDN modems for : less than the pice of a car someday (64k bidirectional). : : -- : bill davidsen, GE Corp. R&D Center; Box 8; Schenectady NY 12345 : Unfortunately the curent United States standard on ISDN is 54Kbit.. :( but i suppose whats 10Kbit.. C.Kup. _____________________________________________________________________________ [__From_________________________________][ aka: Christopher Kuperman ] [_______________________________________][ The holistic systems consultant ] [____@@@@@@____________________@@_______][------------------------------------] [_______@@____@@@@@___@@_@@@___@@__@@___][ email: zork@yoyo.cc.monash.edu.au ] [______@@____@@___@@__@@@___@__@@@@_____][------------------------------------] [_____@@_____@@___@@__@@_______@@_@@____][ Giv a man a fish & he'll eat for a ] [____@@@@@@@__@@@@@___@@_______@@__@@___][ day, teach him how to fish & he'll ] [_______________________________________][_eat for ever.______________________] ------------------------------ From: thomas@thwieck.han.de (Thomas Wieckhorst) Crossposted-To: comp.editors Subject: Re: Long lines in vi (elvis) Date: 15 Apr 93 10:26:30 GMT Charles Hannum (mycroft@hal.gnu.ai.mit.edu) wrote: : I have yet to see a clone which does it right. : For the uninitiated, in vi: : 8 - goes into the global argument register : dd - deletes (8) lines : 4 - replaces the global argument register : . - repeats the `dd'; deletes 4 lines because the register changed : . - repeats the `dd', again deleting 4 lines : One clone repeats the whole `8dd', deleting 32 lines for the first `.'; : one ignores the `4'; one gets the `4' but forgets it later; etc. Hallo, As I see your posting, I have tested my vi (elvis) on my linux-machine and it does exactly what you want to do. After an "8dd4.." it has exactly deleted 16 lines from my text. thomas posting from yet another galaxie -- Thomas Wieckhorst +49 5192 18848 thomas@thwieck.han.de Kiefernweg 12 GERMANY 3042 Munster / Breloh 1 ------------------------------ From: mgrmem@NeXTwork.Rose-Hulman.Edu (Mark Monnin) Subject: Re: how safe is minix filesystem in power failure? Date: 15 Apr 1993 18:02:08 GMT Reply-To: mgrmem@NeXTwork.Rose-Hulman.Edu (Mark Monnin) In article <1993Apr14.145214.432@n5ial.mythical.com> jim@n5ial.mythical.com (Jim Graham) writes: > actually, the way UPSs normally work is more along the same lines as the > battery in a telephone central office (switching center). basically, the > entire phone system (including your telephone, provided it isn't one of > the ones, such as cordless phones, that wants external AC power) runs off > of the battery the whole time, whether external power is there or not. > the external power, when it exists, is used to keep the battery charged. > (in the case of the telco, there are *MASSIVE* generators to take over > the charging of the battery when power failure is sustained.) Be careful, your talking about on-line UPS. Many low cost UPS's are off-line UPS (standby power supplies). BEST uses a combination and "mixes" the power of both (to prevent a failure in either the inverter or switch over from killing your system). Small UPS (esp. standby variants) are fairly cheap. About US$150, I think. They won't stay up for long, nor will they keep your monitor and printer going as well. But if all you really need is to shutdown the system, 5 minutes is enough time. Locally, we have more problems with "power flickers" than "power failures." Having a baby unit to keep the system running over those times helps a lot. -Mark Monnin Mark.Monnin@Rose-Hulman.Edu ------------------------------ From: jsacha@uoft02.utoledo.edu Subject: df / mount problem ? Date: 15 Apr 93 13:10:14 EST I am using latest SLS package (from tsx-11). If I am type as root 'df' a message appears: df: cannot read table of mounted file systems what is this table ? where it should be located ? Also, when I try 'mount' without options there is no response, it should print mounted file systems. But there is no ather problem with mount I can found, exept worning: EXT2-fs warning: mounting non valid file system when root f.s. is mounted during boot-up. I checked /dev/hda1 (my root f.s) with e2fsk, it seems to be no errors. Any ideas what is going on ? Any clues ? Thanx Jarek. jsacha@uoft02.utoledo.edu ------------------------------ From: jjctc@lhdsy1.lahabra.chevron.com (James C. Tsiao) Subject: Re: Does anyone ? Date: 15 Apr 93 17:41:33 GMT In article <2051@dsbc.icl.co.uk> mat@dsbc.icl.co.uk (Matthew Roderick) writes: > >Does anyone have a SiS chipset mother board running linux ? Well, I had Linux 0.99pl2 running on a 486DX2-66 with SiS chipset for a while. But, when I added my Elite 16T ethernet card, X start locking up randomly when painting menus or windows. The problem was "solved" by relocating my cards within my machine. Then, on another SiS board with the similar hardware and software, the same problem occurred, and swapping cards around didn't solve the problem (at least not with the combinations that I've tried). Finally, I returned both motherboards for ones with UMC chipsets. No problem ever since. Although I can't prove it, it appears that the SiS chipset is marginal at best for Linux. >With an AMI BIOS > >I have 486/33 and would like information on the setup for linux of the CMOS. >I think my setup my be incorrect as Linux keeps panicing on boot ! I've never received panics. If you provide more details (Linux version, attached hardware, etc.), then we can help you better. James. -- jjctc@chevron.com | The shortest distance between jtsiao@netcom.com | two puns is a straight line. ============================================================================= Ask me about Linux, the freely distributable Unix clone. ------------------------------ From: dpjunk@mmm.com Subject: Problems getting X running. Date: 15 Apr 1993 17:24:39 -0400 Reply-To: dpjunk@mmm.com I figured out why I was having a problem getting X to run correctly on my machine. (i.e. Getting multiple images of the same thing) When I re-installed Linux (.99p6), I forgot to set the correct chipset in my Xconfig file. It should of been set to pvga. Sorry for taking up bandwith on a problem caused by my stupidty. BTW, thanks a million to all who have produced this GREAT product. I have been running it for about 4 months now without any real problems. (I wish I could say that about msdog and microhog!) -- Dean P. Junk "An ounce of perception, a pound of obscure" Internet (dpjunk@mmm.com) --RUSH ------------------------------ From: joe@retina.anatomy.upenn.edu (Joe Panico) Subject: gnuplot make problem undre SLSp6 Date: 15 Apr 93 15:46:45 GMT Hi, I just attempted to compile gnuplot 3.2, from tsx-11:/pub/linux/sources/ usr.bin/gnuplot-3.2.tar.Z, on my linux SLS pl6. The make looks for a standard unix library (well, sorta standard), libplot.a, which doesn't exist in SLS pl6. Is this library required for gnuplot? If so, where can I get it. Joe Panico joe@retina.anatomy.upenn.edu ------------------------------ From: tmcwill@ukelele.GCR.COM (Thomas McWilliams) Subject: Re: Memory problem with Linux? Date: Thu, 15 Apr 1993 14:56:17 GMT u0xh@jupiter.sun.csd.unb.ca (Chris Newton) writes: >It is using up all available memory, and locking. What does Linux >do when it runs out of swap space, or memory in general. Make sure that you are using libc.4.3.3. There is a bug in earlier libraries that can cause memory to be eaten under rare circumstances. Upgrade to the new library and see if your problem goes away. Thomas -- tmcwill@gcr.com ------------------------------ From: morpheus@cs.utexas.edu (Richard Francis Jones III) Subject: gcc and/or as (as86) assembler documentation Date: 15 Apr 1993 12:00:13 -0500 I'm trying to port some DOS (TASM) assembly programs to LINUX, but I'm having trouble figuring out the format of gcc,as, and as86. If anyone knows where I can find the documentation, it would be greatly appreciated. Thanks in advance. Richard Jones morpheus@cs.utexas.edu ------------------------------ From: alsaggaf@athena.mit.edu (M. Saggaf) Subject: Re: Announcing Babaya 1.00b Date: 15 Apr 1993 22:38:06 GMT In article janeK@world.std.com writes: > >Pretty cool. I have been waiting for somehting like this for some >time. I now have Babaya run in the foreground as the last command in >my ~/.xinitrc (well, pretty much the only one, all others I have moved >to Babaya's startup file), and besides shutting X gracefully, I can >now also quit gwm and start olvwm whenevr I feel bored with the former >(it used to be that quitting gwm shuts the server since it was run >last from ~/.xinitrc). > Good to hear that! >I have some questions, though: > >1) Can I make Babaya use a different mouse key, or possibly a > double-click? My hand is pretty quick with the right mouse button > (blame olvwm for that!) and I don't want to logout if I happen to > click on Babaya by accident. > You can change it to anything you like. If you want to logout only by double-clicking the middle mouse button, put the following in your ~/.Xresources file: Bayaba*Command.Translations: #replace \n\ : highlight() \n\ : reset() \n\ (2): notify() unset() >2) Can I start Babaya as an icon? > Yes, either invoke it as babaya -iconic or add the resource Babaya.iconic: on >3) Since it's pretty small (9k, linux shared libraries do wonders), is > it a good idea to compile it with the gcc -N swicth? (there was a > recent thread about that, but I'm afraid I didn't absorb all of > it). > I'm not sure. It makes little difference anyways. All the best, /Muhammad M. Saggaf | Bosnia: a nation annihilated while the alsaggaf@athena.mit.edu | world is watching. Help stop the genocide. ------------------------------ From: root@blast.erlm.siemens.de (The Big Cheese) Subject: MAN: compression method Date: Thu, 15 Apr 1993 16:09:12 GMT I have noticed that all the man pages in the cat sections are compressed with 'compress'. Will a future version be using gzip to compress the man pages? or has someone alrteady done this? Iain ------------------------------ From: Enrico.Scotoni@purplet.demon.co.uk (Enrico Scotoni) Subject: Ext-2 FS bug Date: Thu, 15 Apr 1993 07:58:00 +0000 KD>I just encountered what seems to be a serious bug in >the ext-2 filesystem. KD>Over the weekend, I transferred some 200 files from >another computer over a serial link using kermit >into a 70-80 meg ext-2 spare partition on my Linux >386DX-40 system (Ultrastor 14F, Quantum LP240S). After ^^^^^^^^^^^^^ I think there is the Problem. I WAS using the 14F on my system and ext-fs (NOT ext-2-fs). After Linux 0.99pl7 there is a new driver for the 14F which is supposed to be faster.IT IS VERY FAST (in screwing up my filesystem). I could easily repoduce the problem with the following command: cd /usr ; for i in * ; do (tar cf ${i}.tar ${i} ; rm ${i}.tar) & ; done It starts running and after few minutes I get lots of error messages and finally the system crashes. when trying to fix the filesystem i also get so much errors that there is NO CHANCE to fix it. Iam not sure if there really is to blame the disk-driver or the controller. I already twice sent a problem-description to c.o.l.but it either did not get out or nobody took any notice of it. Anyhow, I am using at the moment a friend's AHA1542B and I ordered a new AHA1542C. With the AHA1542B I have no problems at all, hopefully my new AHA1542C will behave the same. Anybody interested on a used (3 months old) UltraStor 14F ???? :-) Enrico Scotoni ------------------------------ ** 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-Activists-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux) via: Internet: Linux-Activists@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 tupac-amaru.informatik.rwth-aachen.de pub/msdos/replace The current version of Linux is 0.99pl6 released on February 23, 1993 End of Linux-Activists Digest ****************************** hting like this for some >time. I now have Babaya run in the foreground as the last command in >my ~/.xinitrc (well, pretty much the only one, all others I have moved >to Babaya's startup file), and besides shutting X gracefully, I can >now also quit gwm and start olvwm whenevr I feel digest852 644 36676 74430 47235 5366137101 5761 From: Digestifier To: Linux-Activists@senator-bedfellow.mit.edu Reply-To: Linux-Activists@senator-bedfellow.mit.edu Date: Thu, 15 Apr 93 21:00:18 EDT Subject: Linux-Activists Digest #852 Linux-Activists Digest #852, Volume #4 Thu, 15 Apr 93 21:00:18 EDT Contents: How do I tell lpr not to print a blank page after each job??? (Joel M. Hoffman) Can linux be used as a network printer server ? (Stephen O. Berger) Re: [HELP] XWindows 8 MEG RAM req??? (Drew Eckhardt) Two questions (Todd Holbrook) Re: Why Linux should remain Linux and seagulls should not be used. (Richard Gooch) Re: 99p8 breaks xdm -- authentication problems? (Guy Thomas) Re: how safe is minix filesys (Alan Cox) Re: df / mount problem ? (HJ Lu) Re: Automated Linux PowerDown (Garrett D'Amore) Video Card for a Mitsubishi Diamond Scan Monitor (Noahal A. Mundt) Test (Ricardo Gil Barrio) [Q]: Driver for the VideoBlaster (Ricardo Gil Barrio) Re: SLIP for linux? (Jim Lynch) Cache controller - which unix fs ? (Heiko Herold 339778/IL) Re: HELP: Joining SLIP mailing list? (Klaus Steinberger) Re: how safe is minix filesystem in power failure? (Steve K) Re: SLIP for linux? (Matthew Rhoten) Re: gnuplot make problem undre SLSp6 (Leon Dent) Re: Why Linux should remain Linux and seagulls should not be used. (Nathan Otto Siemers) ---------------------------------------------------------------------------- From: joel@wam.umd.edu (Joel M. Hoffman) Subject: How do I tell lpr not to print a blank page after each job??? Date: Thu, 15 Apr 1993 20:28:41 GMT Is there a way to ask lpr not to print a blank page after each job? -Joel (joel@wam.umd.edu) ------------------------------ From: sb@next.neuro.emory.edu (Stephen O. Berger) Subject: Can linux be used as a network printer server ? Date: 15 Apr 1993 17:40:42 GMT Reply-To: sb@next.neuro.emory.edu Hi, Is it possible to set up a linux machine on a network as a printer server ? I know that networking support is now in linux, but does lpr understand how to send data to remote machines ? (i.e. can I just put a rm=hostname line in the printcap file like on most unix hosts). Please reply by internet mail if possible. Steve Berger (sb@onripx.neuro.emory.edu) ------------------------------ From: drew@kinglear.cs.colorado.edu (Drew Eckhardt) Subject: Re: [HELP] XWindows 8 MEG RAM req??? Date: Thu, 15 Apr 1993 20:44:05 GMT In article <734876801.F00076@remote.halcyon.com> Chris.Dreke@f460.n3625.z1.fidonet.org (Chris Dreke) writes: >Well, I am able to run X windows on 4 meg + swap space, but it is too >slow to do much more than play tetris. If you want to run multiple >applications, you will need more memory. More memory is definately better. However, if you are stuck with only 4M : - Use the monochrome server instead, mono images require 12.5% of the memory that color images take. - Pare down the kernel to more workable size by compiling out the features you don't use - ie TCP/IP, SCSI, etc. - Use a swap partition instead of a swap file, it's noticeably faster. -- Boycott USL/Novell for their absurd anti-BSDI lawsuit. | Drew Eckhardt Condemn Colorado for Amendment Two. | drew@cs.Colorado.EDU Use Linux, the fast, flexible, and free 386 unix | ------------------------------ From: tholbroo@malibu.sfu.ca (Todd Holbrook) Subject: Two questions Date: Thu, 15 Apr 1993 20:36:34 GMT First question: In /usr/include, some of the header files refer to other header files in a linux directory. These header files, such as time.h, do not exist in the linux directory off of /usr/include, nor are they in /usr/src/linux/include/linux (or whatever that other directory is) Can someone tell me where to get these header files, please? Second question: When trying to connect my XT to my linux box through a null modem, I've managed to get getty to send the issue file and the login prompt, but as soon as I type in a log in name, it hangs. Using ps, I can see that getty has called login, but I don't get the password prompt, and after 5 minutes it drops back to getty and the issue file and login prompt. The (hopefully) relevant line in my inittab is: c5:5:respawn getty 9600 ttyS1 (From memory, but it should be right) If anyone can help me with either of these problems, please email me at: tholbroo@fraser.sfu.ca Thanx, Todd... ------------------------------ From: rgooch@rp.CSIRO.AU (Richard Gooch) Subject: Re: Why Linux should remain Linux and seagulls should not be used. Date: Wed, 14 Apr 1993 15:29:11 GMT In article , spj@ukelele.GCR.COM (Simon Patrick Janes) writes: > I don't like the seagull idea however. Seagulls are too common I think... > I don't know... it just makes me think what America would have been like > if the national bird was the turkey like it almost became. I would rather > have a platypus than a seagull. Being a True Blue Aussie, I'm rather fond of the platypus. However, seeing as the Linux logo should have a Finnish flavour, this seems inappropriate. Of course, an alternative is to have an International flavour logo. How about a set of laurels ("Linux: peace in our time")? Hm. Probably too obtuse. Regards, Richard Gooch.... ------------------------------ From: gthomas@fraser.sfu.ca (Guy Thomas) Subject: Re: 99p8 breaks xdm -- authentication problems? Date: Thu, 15 Apr 1993 17:37:12 GMT Timo.Korvola@hut.fi (Timo Korvola) writes: >In article gthomas@fraser.sfu.ca (Guy Thomas) >writes: > I initialy installed the new kernel, 99p8, > which caused xdm to fail. > - xdm comes up and presents login prompt > - login takes username and passwd > - seems to pass control to xinit > - then snaps back to the login >I've had the same problem, had to switch back to 0.99.7 to run xdm. >It seems that xinit runs .xinitrc, but the user's clients aren't >allowed to connect to the server. Putting xhost + in your Xstartup >disables the access control, so your clients (and anybody else's for that >matter) can connect, but I wouldn't recommend this on a network machine. This behaviour may be related to NFS. I have created anon privleged user with home directory on the local disk and was able to log in. All the other users on the system have directories mounted via NFS. I will apply the patch that came out for the vm86 problem to see if it is the problem here. Watch this space. Guy P.S. clearing /tmp/.X0-lock and /tmp/.X0-unix/X0 cleared up the Server active error. Thanks to; Tor Arntsen, Jason A Kinner, James C. Tsiao, Ron Visser, Bill Woodward and any one else that I have have missed. ------------------------------ From: iiitac@swan.pyr (Alan Cox) Subject: Re: how safe is minix filesys Date: 15 Apr 93 18:39:05 GMT So long as you check the filesystems at boot time you should be fine even if the machine goes off at the wrong moment. Ok you might lose a few blocks that were being written at the time but nothing serious. I'd think a cron job to do a halt at 30 minutes before the timeswitch would be fine. Also remember that even if the uucp overruns under Linux the halt by cron will occur happily mid phone call.(I do hope your modem drops when that happens 8-)) Alan ------------------------------ From: hlu@eecs.wsu.edu (HJ Lu) Subject: Re: df / mount problem ? Date: Thu, 15 Apr 93 20:00:08 GMT In article <1993Apr15.131014.3230@uoft02.utoledo.edu>, jsacha@uoft02.utoledo.edu writes: |> |> I am using latest SLS package (from tsx-11). If I am type as root 'df' |> a message appears: |> |> df: cannot read table of mounted file systems I have warned you in c.o.l. Please dig out my old article. I may miss something here. 1. Get Rick Sladkey's mount 0.99.6. It is in mount-0.99.6.tar.Z and utile-1.4.src.tar.z from sunsite or tsx-11. 2. Remove any rdev > /dev/mtab from /etc/rc. 3. List / partition in /etc/fstab. 4. Add "mount -av" in /etc/rc or somewhere like that. 5. Always provide a fs name in /etc/fstab. After rebooting, you will be fine. |> |> what is this table ? where it should be located ? Also, when I try 'mount' |> without options there is no response, it should print mounted file systems. |> But there is no ather problem with mount I can found, exept worning: |> |> EXT2-fs warning: mounting non valid file system |> |> when root f.s. is mounted during boot-up. I checked /dev/hda1 (my root f.s) |> with e2fsk, it seems to be no errors. |> |> Any ideas what is going on ? Any clues ? |> That is ok since your / is ext2 which cannot be umounted when you stop Linux. |> Thanx |> |> Jarek. |> jsacha@uoft02.utoledo.edu |> |> H.J. ------------------------------ From: garrett@garnet.berkeley.edu (Garrett D'Amore) Subject: Re: Automated Linux PowerDown Date: 15 Apr 1993 18:02:40 GMT A couple of things... 1) Monitors consume the lions share of power on a system by far more than any other peripheral, and *much* more than the CPU. 2) If you figure that much money, you are probably using the power rating of the CPU. It is important to realize that you probably rarely if ever even come close to the power rating of the power supply in your CPU unit. (Power ratings are typically 100-150 watts -- actual consumption is probably less than 20 watts if you turn off the monitor) Net result: ALWAYS turn the MONITOR off! They cost lots to keep turned on. Almost never turn the CPU off. There is additional strain on the system hardware, and on the filesystems when you do that. Incidentally, the EECS department here at UC Berkeley has used that method to DRASTICALLY reduce their power consumption. Turning the monitors off when not in use SAVES BIG BUCKS! ==================================================================== Garrett D'Amore | garrett@haas.berkeley.edu Software Co-Ordinator | 68 Barrows Hall, UC Berkeley Haas Computing Services | Ph: 510-643-5923 Fax: 642-4769 ==================================================================== ------------------------------ From: nam@controls.ccd.harris.com (Noahal A. Mundt) Subject: Video Card for a Mitsubishi Diamond Scan Monitor Date: 6 Apr 93 17:35:54 GMT Fellow Linux Users, I have attempted with no luck to get XFree 1.2 running in color mode using my somewhat dated ATI VGA Wonder Card (circa 1988). However, I can use X386mono without problems on my Mitsubishi Diamond Scan Monitor. Since video cards are much cheaper than monitors and I want color, I would like to know if someone has XFree 1.2 working, in color, on a Mitsubishi Diamond Scan - 13" 800x600 monitor. If so, what video card are you using? My other requirements are I want VGA, 1MB VRAM, 800x600, and 256 colors. Thanks in advance for your replys. P.S. Many thanks to those of you who responded with suggestions in an effort to get my ATI/Mitsubishi combination working. Regards, Noahal Mundt - System Administration - Harris Controls Division ============================================================================= | "A good magician never reveals his secret; the unbelievable trick becomes | | simple and obvious once it is explained. So too with UNIX." | ============================================================================= | INTERNET: nam@ccd.harris.com - Phone: (407) 242-5459 fax (407) 242-4453 | ============================================================================= ------------------------------ From: rgil@pinpin.tid.es (Ricardo Gil Barrio) Subject: Test Reply-To: rgil@pinpin.tid.es Date: Wed, 14 Apr 1993 06:26:14 GMT Please disregard this. --- ================================================== +=++== Ricardo Gil Barrio e-mail: rgil@tid.es | || Telefonica I+D Voice: +34 1 3374716 +-++-+ Madrid, Spain Fax: +34 1 3374222 || | | ================================================== | \+=+ ------------------------------ From: rgil@pinpin.tid.es (Ricardo Gil Barrio) Subject: [Q]: Driver for the VideoBlaster Reply-To: rgil@pinpin.tid.es Date: Wed, 14 Apr 1993 06:35:34 GMT Does anybody know a driver for the VideoBlaster for any (public domain or comercial) Unix ? Thanks --- ================================================== +=++== Ricardo Gil Barrio e-mail: rgil@tid.es | || Telefonica I+D Voice: +34 1 3374716 +-++-+ Madrid, Spain Fax: +34 1 3374222 || | | ================================================== | \+=+ ------------------------------ From: jwl@ferrari.cray.com (Jim Lynch) Subject: Re: SLIP for linux? Date: 15 Apr 93 17:06:23 GMT Is anyone aware of a product for Linux that works similar to SLIP, but doesn't require SLIP? SLIP had some security holes that are incompatible with our corporate police. I am aware of term, but I believe that only works with X and I can`t run X yet. Is there anything else that will let me have multiple connections to a Unix system, that doesn't require that I have su permission on the Unix system? Has anyone heard of a version of UW for Linux? UW is a package that was written for the Mac and ported to MSDOS, then ported to Windows, but I'm not aware of a Linux port. Thanks.. Jim. - ======================================================================== Jim Lynch, Sales Analyst, Cray Research, Inc. / ARS: K4GVO Southeast District, Phone: (404) 631-2254, Email: jwl@sedist.cray.com Suite 270, 200 Westpark Drive, Peachtree City, GA 30269 ------------------------------ Crossposted-To: alt.uu.comp.os.linux.questions From: hman@esse6.dei.unipd.it (Heiko Herold 339778/IL) Subject: Cache controller - which unix fs ? Date: Thu, 15 Apr 1993 12:59:17 GMT I got a cache controller (longshine ? longxxxx ?) for the Ide-Hd. In his setup it is possible to configure an optimization based on the OS, I think in truth it's file-system based. There are several choices: dos windows os2 novell sco xenix sco unix interactive unix at&t unix Anyone knows about which so has the most similar file system to linux? Actually I'm using a little bit of the minix fs and much more the ext(2)-fs, which is not yet stable...so what should I select (and why since the fs will change..) ? Please if anyone knows something about it mail to hman@paola.dei.unipd.it since we've got several trouble with our newsserver... we're still waiting for the news of about 2 weeks ago :-( Thanks -- ---hman@[paola][maya][claudia].dei.unipd.it - Don't let the SUN go down on me {George Micheal} - Hi Mom. I know nobody reads manuals and FaQs, but I knew _you_ would. I worked really hard on this manual, and...Mom...wake up Mom. ------------------------------ From: k2@bl.physik.tu-muenchen.de (Klaus Steinberger) Subject: Re: HELP: Joining SLIP mailing list? Date: Thu, 15 Apr 1993 18:00:13 GMT djr48312@uxa.cso.uiuc.edu (Dennis Robinson) writes: >I was interested in helping the debugging effort of the ALPHA SLIP >package. What is said is that I must join the mailing list until >I can even see code. So I would like to join the mailing list. I would like to join the SLIP mailing list too, could somebody please enlighten me? Sincerely, Klaus -- Klaus Steinberger Beschleunigerlabor der TU und LMU Muenchen Phone: (+49 89)3209 4287 Hochschulgelaende, D-8046 Garching, Germany FAX: (+49 89)3209 4280 !!! after 1. July new ZIP Code 85748 Internet: Klaus.Steinberger@Physik.Uni-Muenchen.DE ------------------------------ From: stevek@banshee.ecn.uoknor.edu (Steve K) Subject: Re: how safe is minix filesystem in power failure? Date: Thu, 15 Apr 1993 21:02:53 GMT jim@n5ial.mythical.com (Jim Graham) writes: >actually, the way UPSs normally work is more along the same lines as the >battery in a telephone central office (switching center). basically, the >entire phone system (including your telephone, provided it isn't one of >the ones, such as cordless phones, that wants external AC power) runs off >of the battery the whole time, whether external power is there or not. >the external power, when it exists, is used to keep the battery charged. >(in the case of the telco, there are *MASSIVE* generators to take over >the charging of the battery when power failure is sustained.) >UPS systems for computers (and other devices, too, of course) work the >same way. the computer/whatever runs off of the battery, and the AC power >keeps the battery charged. WRONG! I just bought an APC 400 UPS for my box. It explicitly states that it only conditions the power except in the event of a power failure. It then *switches* to the batteries and provides a reduced amplitude modified sine wave for the computer power supply. I was under the same impression you stated above until I spoke with a representative from the *manufacturer*. He stated that the life of the batteries used in such a manner is far too short for small scale (i.e. PC) use. For the larger systems you mentioned, I would guess that they have a maintenance cycle in which they disconnect one or more cells from the battery array to replace them. Steve ------------------------------ From: mrhoten@Xenon.Stanford.EDU (Matthew Rhoten) Subject: Re: SLIP for linux? Date: 15 Apr 93 20:44:40 GMT term does not require X. I run term just fine on a 386SX notebook on which I do not run X. -matt -- Matthew Rhoten | mrhoten@cs.stanford.edu | m_rhoten@leland.stanford.edu "When Adam and Eve first saw each other, that's when the blues started." -John Lee Hooker ------------------------------ From: lcd@umcc.umcc.umich.edu (Leon Dent) Subject: Re: gnuplot make problem undre SLSp6 Date: 15 Apr 1993 18:34:21 -0400 If you want, you can skip that version and get the new beta version. Look on tsx in the TeX dir and you will find a gnuplot3.3 beta source which compiled straight away on my SLS system. Leon Dent lcd@umcc.umich.edu ------------------------------ From: ns14@crux3.cit.cornell.edu (Nathan Otto Siemers) Subject: Re: Why Linux should remain Linux and seagulls should not be used. Date: 15 Apr 93 13:59:46 >>>>> On 15 Apr 93 15:47:32 GMT, jjctc@lhdsy1.lahabra.chevron.com (James C. Tsiao) said: | In article <1993Apr15.050119.5106@sol.UVic.CA> pmacdona@sanjuan (Peter MacDonald) writes: |>In article <93104.131126ACPS7221@RyeVm.Ryerson.Ca> writes: |>> How about a Tiger? A Tiger is powerful, fast, sleek, efficient, and so is |>> Linux. |>> And also, Linus is a roarin' good programmer:) |> |>Nice, but maybe too aggressive. Has anyone suggested a Dolphin (dolFINN :-). | Umm, I hate to see someone, especially some rabid 386bsd fan :-), changing | that into DULL-FINN. Perhaps a porpoise... | You know, an OS with a porpoise. This is beginning to sound like the disgusting essay question on the mit admission application asking you what kind of animal you would be and why..... Just make it a cute little rabbit with NASTY POINTY TEETH. nathan -- nathan@chemres.tn.cornell.edu Department of Chemistry Cornell University ------------------------------ ** 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-Activists-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux) via: Internet: Linux-Activists@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 tupac-amaru.informatik.rwth-aachen.de pub/msdos/replace The current version of Linux is 0.99pl6 released on February 23, 1993 End of Linux-Activists Digest ****************************** im@n5ial.mythical.com (Jim Graham) writes: >actually, the way UPSs normally work is more along the same lines as the >battery in a telephone central office (switching center). basically, the >entire phone system (including your telephone, provided it isn't one of >the ones, such as cordless phones, that wants external AC power) runs off >of the batterdigest853 644 36676 74430 50205 5366137101 5751 From: Digestifier To: Linux-Activists@senator-bedfellow.mit.edu Reply-To: Linux-Activists@senator-bedfellow.mit.edu Date: Thu, 15 Apr 93 22:45:20 EDT Subject: Linux-Activists Digest #853 Linux-Activists Digest #853, Volume #4 Thu, 15 Apr 93 22:45:20 EDT Contents: Re: A couple of questions: (Drew Eckhardt) Re: Where does one find .au files? (Jonathan Magid) Re: Linux Logo Postscript (Leon Dent) Re: The dangers of playing with shared libraries (Brandon S. Allbery) Missing Linux Include Files ... (Noahal A. Mundt) Good Swap Partition Size?? (DHALIWAL BIKRAM SINGH) Re: 99p8 breaks xdm -- 99p8A fixes it! (Guy Thomas) Re: Disk Quotas (was Re: New feature for the filesystems.) (Scott W. Adkins) Re: xvt for Linux??? (Robert Nation) AAACK Xtoolplaces?!?! (kevin fisher U) Weird X lock (S3) (Gregory S. Halbrook) Re: A few things to discuss... (Jim Lynch) Solutions to install Logitec S9 Bus Mouse (Tein H. Yuan) Xtoolplaces from new (3L4) xview release (khamer@sumax.seattleu.edu) Re: Disk Quotas (was Re: New feature for the filesystems.) (Steve Sheldon) problem compiling MGR...help! (Patrick K. Ferrick) Re: Automated Linux PowerDown (H. Peter Anvin N9ITP) ---------------------------------------------------------------------------- From: drew@kinglear.cs.colorado.edu (Drew Eckhardt) Subject: Re: A couple of questions: Date: Thu, 15 Apr 1993 20:55:06 GMT In article <734876801.F00077@remote.halcyon.com> Derek.Glidden@f42.n377.z1.fidonet.org (Derek Glidden) writes: >...which I know I've seen the answers to, but I can't find them now. > >1) When I mount a DOS partition, it's mounted as: > >-rwx--x--x > >Which means root can get to it, but nobody else can. I'd like to be >able >to mount it 755 so other people can read from it, but not write to it. >I know this has something to do with permissions on the mount executable > >and setuid or something like that, but I can't figure it out. I can >chmod 755 /dos the mount point, but that only holds for the first >directory under it, then everything is non-write-non-read again. Nope, it's based on the umask at the time of the mount call. Instead of a plan mount, do umask 22 mount /dos or whatever. For this reason, in my /etc/rc file I have umask 22 mount -a umask 0 instead of just a mount -a. -- Boycott USL/Novell for their absurd anti-BSDI lawsuit. | Drew Eckhardt Condemn Colorado for Amendment Two. | drew@cs.Colorado.EDU Use Linux, the fast, flexible, and free 386 unix | ------------------------------ From: jem@sunSITE.unc.edu (Jonathan Magid) Subject: Re: Where does one find .au files? Date: Thu, 15 Apr 1993 21:12:58 GMT In article <1993Apr15.152113.27670@uvm.edu> lhanson@moose.uvm.edu (lhanson) writes: > I picked up the play-au.c code from sunsite and I would >like to try it on my PCs speaker. Are the SUN *.au files PD? If >so where can I find them? I've checked ARCHIE but can't come up >with anyting. sunsite is your one-step archive! Okay enough marketroid BS (tm), but sunsite has an outrageously large collection of sound stuff (around ~300 megs at last count). look in these places: /pub/multimedia/sun-sounds - all the random samples you could possibly want /pub/multimedia/chinese-music - would you believe a musical history of Chinese music? /pub/talk-radio - the Internet Talk Radio show which started this thread jem. ------------------------------ From: lcd@umcc.umcc.umich.edu (Leon Dent) Subject: Re: Linux Logo Postscript Date: 15 Apr 1993 18:25:27 -0400 My own thought was that someone should do a characature(sp?) of Linus. Judging from his picture (available on one of the sites) a drawing might look something like Doonsbury. Leon Dent lcd@ais.org ------------------------------ From: bsa@kf8nh.wariat.org (Brandon S. Allbery) Subject: Re: The dangers of playing with shared libraries Date: Thu, 15 Apr 1993 22:25:03 GMT In article <1qhml1$1i5k@hal.gnu.ai.mit.edu> mycroft@hal.gnu.ai.mit.edu (Charles Hannum) writes: >I don't understand why people are so vehemently against have a /sbin. >Even *Sun* got that right. (1) Have you looked at the size of SunOS recently? (2) On a Sun it's even *more* annoying: you have to boot from CD-ROM. ++Brandon -- Brandon S. Allbery bsa@kf8nh.wariat.org It's not too late to turn back from the "Gates" of Hell... Linux: the FREE 32-bit operating system, available NOW. Why waaaaaait for NT? ------------------------------ From: nam@controls.ccd.harris.com (Noahal A. Mundt) Subject: Missing Linux Include Files ... Date: Thu, 8 Apr 1993 12:38:15 GMT When I try to compile programs using gcc under Linux 0.99 I get error messages indicating ../linux/*.h cannot be found. Looking in /usr/include/linux, I do not find any include files. I loading the "c" series of disks - Did miss something? Thanks in advance for your helpful suggestions and instructions. BTW, I am a fairly new Linux user and just want to thank all of you who have been patient with my questions. Thanks! :-) P.S. Anyone using Linux on a Mitsubishi Diamond Scan monitor? Regards, Noahal Mundt - System Administration - Harris Controls Division ============================================================================= | "A good magician never reveals his secret; the unbelievable trick becomes | | simple and obvious once it is explained. So too with UNIX." | ============================================================================= | INTERNET: nam@ccd.harris.com - Phone: (407) 242-5459 fax (407) 242-4453 | ============================================================================= ------------------------------ From: dhaliwa@ecf.toronto.edu (DHALIWAL BIKRAM SINGH) Subject: Good Swap Partition Size?? Date: Thu, 15 Apr 1993 21:40:25 GMT I have an 8mb machine, running SLS w/X. What is a good swap partition size for such a configuration. (running really no major apps, perhaps using the compiler every now and then). -- .......................... Bikram Dhaliwal PH: 416-351-9660 dhaliwa@ecf.toronto.edu ------------------------------ From: gthomas@fraser.sfu.ca (Guy Thomas) Subject: Re: 99p8 breaks xdm -- 99p8A fixes it! Date: Thu, 15 Apr 1993 19:13:06 GMT gthomas@fraser.sfu.ca (Guy Thomas) writes: >Timo.Korvola@hut.fi (Timo Korvola) writes: >>In article gthomas@fraser.sfu.ca (Guy Thomas) >>writes: >> I initialy installed the new kernel, 99p8, >> which caused xdm to fail. >> - xdm comes up and presents login prompt >> - login takes username and passwd >> - seems to pass control to xinit >> - then snaps back to the login >>I've had the same problem, had to switch back to 0.99.7 to run xdm. >>It seems that xinit runs .xinitrc, but the user's clients aren't >>allowed to connect to the server. Putting xhost + in your Xstartup >>disables the access control, so your clients (and anybody else's for that >>matter) can connect, but I wouldn't recommend this on a network machine. >This behaviour may be related to NFS. I have created anon privleged >user with home directory on the local disk and was able to log in. >All the other users on the system have directories mounted via NFS. >I will apply the patch that came out for the vm86 problem to see >if it is the problem here. >Watch this space. Yup. The susption was correct. It was related to the access() problem with the NFS code in 99p8. Apply the patch that got released for the vm86 problem and all was well again. >Guy >P.S. clearing /tmp/.X0-lock and /tmp/.X0-unix/X0 cleared up >the Server active error. >Thanks to; Tor Arntsen, Jason A Kinner, James C. Tsiao, Ron Visser, >Bill Woodward and any one else that I have have missed. ------------------------------ From: sadkins@bigbird.cs.ohiou.edu (Scott W. Adkins) Subject: Re: Disk Quotas (was Re: New feature for the filesystems.) Date: 16 Apr 93 00:40:35 GMT In article davidsen@crd.ge.com (bill davidsen) writes: >In article , s_fuller@iastate.edu (Steve Fuller) writes: > >| 1) If the file is in the middle of a write when you exceed your >| quota, the file will be written in its entirety and not >| truncated. Any furthur write requests will be denied > > I certainly don't want a quota like that. If the hard limit isn't hard >it doesn't protect the system. unless writes STOP at the hard limit the >user could run the system out of disk anyway. Right... every quota system I have ever seen truncates the file so that it will *not* exceed the hard quota... Scott -- Scott W. Adkins Internet: sadkins@bigbird.cs.ohiou.edu ~~~~~~~~~~~~~~~ ak323@cleveland.freenet.edu Ohio University of Athens Bitnet: adkins@ouaccvma.bitnet ------------------------------ From: nation@snoopy.sanders.lockheed.com (Robert Nation) Subject: Re: xvt for Linux??? Date: 15 Apr 93 14:30:50 I have ported xvt to linux, but I make a bunch of changes along the way, so you may or may not be interested. Here's a summary of the changes: Just to port xvt: Only a few minor mods: 1. DON'T compile using optimization. In the Makefile, remove -O 2 from cflags. 2. there is one instance of the constant FNDELAY which should be changed to O_NDELAY. 3. One of the include file names, I think needs to be changed to . The error will become obvious when you compile. The sources can be found on ftp.uu.net. This leaves the bug that if, in one terminal window, you type xvt, then cntrl-z, then bg, it croaks. The problem is that Linux returns an error if a select() call is interrupted, but sunos doesn't. My modifications: 1. Stripped out debugging sub-routines to make the code smaller. 2. Cleaned up some code to make it a bit more compact, and a little easier to follow. 3. Removed support for some fairly useless vt-100 commands, like the one that fills your screen with capital E's/ 4. Added partial support for the graphics character sets (enough to run all of my applications), which is a fair amount, but not all graphics character commands. 5. Forced the scroll-bar to always show. Also modified the appearance of the scrollbar (I like it better). 6. removed support for getting default info from the .Xresources files, since I don't use these, and find it annoying to have to look in 29 different config files all over the system to find out why I' getting purple text, or something. 7. Added support for more of the function keys. The original xvt didn't respond to any of F1-F20, or most of the home, pg up, etc keys. Anyway, after doing all that, the final executable has a text size of about 41 Kbytes, and data size of 8 kbytes, about 40% smaller than stock xvt. If one xvt is used instead of 1 xterm, it saves 368 Kbytes of memory. I didn't check into saving on subsequent instances. If you want my source code, I will email it to you. ------------------------------ From: kfisher3@mach1.wlu.ca (kevin fisher U) Subject: AAACK Xtoolplaces?!?! Date: Wed, 14 Apr 1993 20:00:59 GMT I don't know if anyone has asked this before, so please be gentle! I can't get on as much to read the news.... So the question: How the hell can you get xtoolplaces to compile? I notice in the new release of Xview xtoolplaces has been compiled....I've even contacted the author and he doesn't know why it won't compile for me. The problem is with the Xfree() function. If I weren't at the school, I'd dump the entire compile error list to the net but... ...anyway, the compiler is complaining about the wrong 'type' argument for Xfree(). Any help would greatly be appreciated (e-mail please!) as this has been bugging me for months! -- kfisher3@mach1.wlu.ca Guru Meditation: ------------------------------ From: gsh@iti.org (Gregory S. Halbrook) Subject: Weird X lock (S3) Date: Thu, 15 Apr 1993 23:15:02 GMT Hello! I am experiencing a strange lockup about once or twice a day while running X-windows (I am running the tvtwm window manager). What happens is that intermittantly my screen will just *poof* and appear that it lost all horizontal hold. It doesn't flicker or anything but rather just freezes. The screen has slanted stripes, which are rather wide, running down it. Sometimes, more often than not, I can still telnet into my system so it's not dead (sometimes it is dead though). I've tried killing the X-server and miscellaneous other stuff, but only a reboot truely fixes the problem. I'm running on a DEC 466D2LP (a 486/66) and have the S3 chip and a local bus. I also have 24 megs ram and a 11meg swap partition on a 245 meg HD. I'm running the XS3-0.4.4 server. I am very impressed with linux and will quite willing live with this minor inconvenience but am hoping some of the gurus out there will be able to help or point me in the right direction. Please assume I'm an idiot ;-) and if the answer is in the FAQ somewhere then I just proved it to you. Any further information needed will be gladly provided. -- Greg. ------------------------------ From: jwl@ferrari.cray.com (Jim Lynch) Subject: Re: A few things to discuss... Date: 15 Apr 93 12:38:21 CDT In article <1993Apr9.200907.173@tbucks.com>, timothy@tbucks.com writes: |> Chris Newton (u0xh@jupiter.sun.csd.unb.ca) wrote: |> : Is it possible to set up Linux as a packet radio site? Just |> : wondering, a friend of mine is interested in becoming an Internet site and |> : is looking into Linux VERY heavily. Any comments on this? |> : What I want to do eventually, is have my linux system set up with |> : SLIP, so that I can hook up to my friends Linux box, him running the |> : packet radio thing. I would like to have all the normal things running, |> : like email, talk, finger, ftp... so on. Is this possible. Both of us |> : have terminals that would be hooked up to the system too. Does this sound |> : like an interesting setup? |> |> : Thanks for any comments/help... |> |> : Chris |> |> This sounds very interesting, but here are two points to consider. |> Packet radio is amateur radio, and there are restrictions on what may |> be legally broadcast. Also, anyone can monitor the packets, so |> passwords are of little use. I'd suggest dial-in for the uucp site, |> and running something like kaq9 for the packet bbs. Follow-ups to |> rec.radio.amateur.packet |> |> -- |> Timothy Buchanan. |> |> ------- Time is our only resource, humor our only defense. -------- One other minor point, in addition to restrictions on the material that may be broadcast over amateur radio, one must also have a valid amateur license. I don`t know what is happening in Canada these days, but the FCC is levying some pretty large fines for relatively minor infractions down here. I don't want to discourage you from using packet radio, just be sure you are legal. Jim. -- ======================================================================== Jim Lynch, Sales Analyst, Cray Research, Inc. / ARS: K4GVO Southeast District, Phone: (404) 631-2254, Email: jwl@sedist.cray.com Suite 270, 200 Westpark Drive, Peachtree City, GA 30269 ------------------------------ From: tein@hawk.cc.as.edu.tw (Tein H. Yuan) Subject: Solutions to install Logitec S9 Bus Mouse Date: Fri, 16 Apr 1993 01:43:59 GMT After I tried all the combination to install Logitech S9 Bus Mouse, it works now. First of all, IRQ must be set to 5. Secondly, in Xconfig, either MouseSystems "/dev/bmouselogitec" or BusMouse "/dev/bmouselogitec" must be set as well. All set! Tein ------------------------------ From: khamer@sumax.seattleu.edu Subject: Xtoolplaces from new (3L4) xview release Date: 15 Apr 1993 14:55:38 -0700 I got the new xview stuff, nice work kenneth BUT xtoolplaces as set up by the install program a) writes a 0k file .openwin-init if I run it using the "owplaces" script provided, which is fine and dandy except I have 5 apps running and it should mention them in the file b) dumps core if I start it as xtoolplaces Any suggestions? Is there source to the linux version I can get? Thanks, -Ken -- .............................................................................. -Kenneth Hamer khamer@sumax.seattleu.edu ------------------------------ From: sheldon@iastate.edu (Steve Sheldon) Subject: Re: Disk Quotas (was Re: New feature for the filesystems.) Date: Thu, 15 Apr 1993 19:41:51 GMT In s_fuller@iastate.edu (Steve Fuller) writes: >It would seem to me, from a new sysadmin standpoint, to have >anyone but root able to use chown. At least that's how it's done >here on the Ultrix systems at our site. Our system works as >follows as far as quotas: I think you mean, only root can use chown. This is common at most sites, I think. >1) If the file is in the middle of a write when you exceed your >quota, the file will be written in its entirety and not >truncated. Any furthur write requests will be denied Steve is correct, except that he forgot to point out that here he is talking about the Andrew File System which ISU uses, a.k.a. AFS. AFS goes way beyond a typical filesystem tho, and isn't something you'd probably want to implement with Linux. Actually though, there is a hard limit in that if the file exceeds your quota amount by less than 1 Meg it will let you write it out. If it exceeds your quota amount by more than 1 Meg it will return an error: >afs: failed to store file (over quota) You also get this message anytime you are over quota and try to write another file. It's not a particularly friendly environment, and is really only used to establish quota's for the undergraduate students. The research staff and faculty are on NFS mounted disks, and they will begin using the standard BSD quota system in the next month or two. -- sheldon@iastate.edu Steve Sheldon copyright 1993 This signature is Shareware ICSS Resource Unit This .sig is Please send $25 to this address ------> 2142 Agronomy Hall Shareware! Please remember, this is Shareware! Iowa State University Shareware! :-) ------------------------------ From: ferrick@acsu.buffalo.edu (Patrick K. Ferrick) Subject: problem compiling MGR...help! Date: 15 Apr 93 21:10:21 GMT Well, MGR _almost_ compiles...but I'm getting the following unintelligible (to me, anyway :-) error message: /usr/lib/libc.sa.(setsid.o): Undefined symbol "___setsid" referenced What is the problem? I'm running SLS 0.99p6, and I seem to have both a /lib/libc.so.4.1 and a /lib/libc.so.4.3, but neither contains the above mentioned string. If somebody could shed some light on this I would appreciate it. Thanks! pat ------------------------------ From: hpa@merle.acns.nwu.edu (H. Peter Anvin N9ITP) Subject: Re: Automated Linux PowerDown Reply-To: hpa@nwu.edu (H. Peter Anvin) Date: Thu, 15 Apr 1993 17:53:43 GMT In article of comp.os.linux, jkaidor@synoptics.com writes: > trip when a large group of *pulses* came out of the computer. Now, it's probably not usual to see > a couple thousand pulses come out of a DTR output, is it? > My guess would be that a better bet is to require is to: a) Require the power to be on well beyond the bootup period (say 10 minutes) before even monitoring the input. b) Require it to keep an asserted level for several seconds. Just my $0.02 worth. -- INTERNET: hpa@nwu.edu FINGER: hpa@eecs.nwu.edu BITNET: HPA@NUACC IBM MAIL: 36073 at IBMX400 HAM RADIO: N9ITP, SM4TKN NeXTMAIL: hpa@speedy.acns.nwu.edu while ( 1 ) ; cp /dev/zero /dev/null & end ------------------------------ ** 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-Activists-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux) via: Internet: Linux-Activists@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 tupac-amaru.informatik.rwth-aachen.de pub/msdos/replace The current version of Linux is 0.99pl6 released on February 23, 1993 End of Linux-Activists Digest ****************************** tleu.edu ------------------------------ From: sheldon@iastate.edu (Steve Sheldon) Subject: Re: Disk Quotas (was Re: New feature for the filesystems.) Date: Thu, 15 Apr 1993 19:41:51 GMT In s_fuller@iastate.edu (Steve Fuller) writes: >It would seem to me, from a new sysadmin standpoint, to have >anyone but root able to use chown. Atdigest854 644 36676 74430 55150 5366137102 5757 From: Digestifier To: Linux-Activists@senator-bedfellow.mit.edu Reply-To: Linux-Activists@senator-bedfellow.mit.edu Date: Fri, 16 Apr 93 01:00:19 EDT Subject: Linux-Activists Digest #854 Linux-Activists Digest #854, Volume #4 Fri, 16 Apr 93 01:00:19 EDT Contents: Gateway Nomad Field Mouse (Paul Fishwick) Re: can 386sx run linux (Derrick C. Cole) Re: fsck on mounted file systems? (Stephen Tweedie) Help with SLS install... (Gent Hito) Re: Compiling XV; general porting problems (Christoph Adomeit) Re: how safe is minix filesystem in power failure? (Christoph Adomeit) Re: PORTING A PROGRAM TO OS/2? (Christoph Adomeit) Re: Unix OS for 286 (Garrett D'Amore) Re: e2fs Problems/Queries (Stephen Tweedie) Re: Linux on Macintrashes? (Was: Re: Linux on Amiga?) (Kevin Sanders) Re: Big swap partitions [> 16M] (H. Peter Anvin N9ITP) Re: weird vi/less screen insect (Brandon S. Allbery) Re: A couple of questions: (David Engel) Better? XIAFS or EXT2FS? (Greg Bothe) Kernel panic (kowal@stsci.edu) libolgx.a (Vulcan) Re: Unix OS for 286 (David W. Summers) Dexxa 3-button mouse w/Linux and Xfree 1.x (matthew dale porter) Summary of Linux vs. 386BSD vs. Commercial Unixes (Brandon Vanevery) liboldX.a, been done? (Matthew Rhoten) xvt... please (Figmiester) ---------------------------------------------------------------------------- From: fishwick@fish.cis.ufl.edu (Paul Fishwick) Subject: Gateway Nomad Field Mouse Date: 15 Apr 1993 21:45:08 GMT I am looking into putting Linux on a Gateway Nomad 486 DX2/50 and before I do, I'd like to know whether anyone else has put Linux on this machine. I checked the compatibility chart but did not see a "Nomad" or "Texas Instruments" machine there (the Gateway machines being repackaged TI machines). Specifically, the Nomad comes with a "field mouse" (whatever that is :). Does the field mouse emulate the Microsoft bus or serial mouse? Also, I'm wondering how much video memory is on the machine. I would think that a very large virtual console would be invaluable on a notebook computer due to the relatively small display. In that event, even though the thing can only do VGA, it would be great to have at least .5meg of video memory. Do notebooks carry this much video memory? -paul ------------------------------ From: cole@concert.net (Derrick C. Cole) Subject: Re: can 386sx run linux Date: 16 Apr 1993 02:20:39 GMT In article <115444@bu.edu> abc@bass.bu.edu (Onnei Cheng) writes: >hi, guys > >My friend has a 386SX, is it good to run Linux > >thanks in advance > >Onnei > 386sx-16 MHz AMI BIOS 2MB RAM 65MB HD (55MB Linux, 10MB dos) 512KB Paradise 1024i video card Microsoft Bus Mouse WD8003E Ethernet card Compiles are a little on the slow side, but the computer's run just about every version of Linux since 0.12, and even handles X11 with surprising results. I'm preparing to buy another 2MB RAM very soon. Call me a masochist, but it works for me ;) Derrick -- The scene : A 'quiet' frequency on a Sunday afternoon during a DX contest. First station : "Is this frequency in use?" Second station : "Yea, I am listening to the peace and quiet" -- from rec.radio.amateur.misc ------------------------------ From: sct@dcs.ed.ac.uk (Stephen Tweedie) Subject: Re: fsck on mounted file systems? Date: 16 Apr 93 00:40:05 GMT > sct@dcs.ed.ac.uk (Stephen Tweedie) writes: > If the root filesystem is mounted readonly, how can fsck write back > corrections if it finds errors? fsck does not access the partition through the filesystem functions - it directly accesses the block device beneath the filesystem. So, in effect, there are two concurrent users of the block device - the filesystem, which is using it readonly, and the fsck program, which has opened the same device for read-write. This is safe, because the filesystem and fsck will both be seeing exactly the same data - they will share the same buffer cache for the device. Cheers, Stephen Tweedie. --- Stephen Tweedie (Internet: ) Department of Computer Science, Edinburgh University, Scotland. ------------------------------ From: ghito@bvcd.csc.ncsu.edu (Gent Hito) Subject: Help with SLS install... Date: Fri, 16 Apr 1993 01:33:24 GMT I just learned about Linux a couple of days ago. I would like to install it on a PC shared by several people, so I want to be really sure that none of their data is lost. Is that possible? Also, is there any other way of installing it rather than having to back up 100 Mb on the harddrive first? Thanks for all (if any) replies. ------------------------------ From: ado@bigcomm.gun.de (Christoph Adomeit) Subject: Re: Compiling XV; general porting problems Date: Fri, 16 Apr 1993 01:16:29 GMT broadley@pitt.edu (Bill Broadley) writes: >> This is a well known bug with gcc, not just the Linux version. Gcc will >> take outrageous amounts of memory when compiling large static initialized >> arrays. Fear not! there are people with such ridiculous amounts of memory >> and swap space; Rob Hooft is one of them. You can find his jump-tabled >> compilations of lib in /pub/Linux/apps/graphics/libgr-1.0.tar.z >> on sunsite.unc.edu. >As far as I know this isn't related to gcc, other vendors compilers do the same. Thats true, i remember this one file.c in libtiff that compiled about 20 Minutes on a 486/50 Eisa with 16 MB Cache Controller, 64 MB Ram and 7 MB Kernel Buffers on an SCO-Opendesktop 2.0 with a Microsoft C Compiler. It's definitely NOT a gcc-bug. Ciao Christoph ------------------------------ From: ado@bigcomm.gun.de (Christoph Adomeit) Subject: Re: how safe is minix filesystem in power failure? Date: Fri, 16 Apr 1993 01:24:05 GMT hedrick@geneva.rutgers.edu (Charles Hedrick) writes: >jim@n5ial.mythical.com (Jim Graham) writes: >>like the subject says.... how safe (I realize this is all on relative >>terms, here) is the plain minix filesystem (all I've got right now) in >>cases where, say, the power fails? I have NO experiences especially with the Minix-fs, but I use Unix,Linux,Sun, whatever for 5 years now. From about 200 Crashes I have seen at customers or wherever, it only happened 2-times that a fs was so destroyed that it was not reparable. 90 percent where absolutely ok after an fsck, in about 10% there were only 1 or 2 files destroyed. I think, the cheapest way to protect from a crash is a frequently backup. CIao Christoph ------------------------------ From: ado@bigcomm.gun.de (Christoph Adomeit) Subject: Re: PORTING A PROGRAM TO OS/2? Date: Fri, 16 Apr 1993 01:40:53 GMT balog@eniac.seas.upenn.edu (Eric J Balog) writes: >Hi! I have a program called 'diamond.taz'; I ran uncompress and tar. Now I'm >left with 'diamond'. How can I port this program to OS/2? OS/2, Windows, Dos, what's that ? Ask this in comp.os.idiotic or somewhere like that:-) Honestly, i don't know diamond, what type of file is it ? A C-Source ? A Shell-Script ? An executable ? If you are not able to find this out, you'd better not think about porting it to OS/2. CIao Christoph ------------------------------ From: garrett@garnet.berkeley.edu (Garrett D'Amore) Crossposted-To: comp.os.coherent,comp.os.minix,comp.os.386bsd.questions,comp.os.misc Subject: Re: Unix OS for 286 Date: 16 Apr 1993 00:40:50 GMT Get yourself a copy of Coherent. It's sold by Mark Williams Company, and comes in two flavors, v4.0 and v3.2. You won't be able to use v3.2 -- becuase it requires a 386, but v4.0 should work. MWC charges $99 for either version, regardless of price. They do include nice documentation, and their o.s. is much more stable (if it doesn't include as many nice features -- like TCP/IP networking). A few notes -- you will be limited to programs that fit in 64K of ram. You won't be able to run X. There are probably other limitations as well. This is because the 286 is an old cpu which doesn't have as much support for protection and large-space addressing as the 386. Coherent 3.2 is the only 286 based version of Un*x I know of worth messing with. On a side note: if you're going to do anything real with Unix, get yourself a 386 or 486 and save yourself some headaches. :-) PS: There is a group devoted to Coherent on the internet... comp.os.coherent, I think. ==================================================================== Garrett D'Amore | garrett@haas.berkeley.edu Software Co-Ordinator | 68 Barrows Hall, UC Berkeley Haas Computing Services | Ph: 510-643-5923 Fax: 642-4769 ==================================================================== ------------------------------ From: sct@dcs.ed.ac.uk (Stephen Tweedie) Subject: Re: e2fs Problems/Queries Date: 16 Apr 93 01:37:27 GMT In article , mike@ws.rz.tu-bs.de (Mike Dowling) writes: > Unfortunately, ... I turned off the power at the mains, and caused > my linux to crash. On rebooting, I had some strange effects. > ... I then ran e2fsck on /usr. After a while, I got "out of memory" > errors (I have 8 MB), and my system hanged. This is slightly peculiar if you are using e2fsprogs-0.2d - the out-of-memory problem should not appear with this. Are you sure you weren't using an older version at the time? > Pushed the reset button, rebooted from dikette. I altered > /etc/fstab and /etc/rc to avoid mounting /usr automatically when > booting, and then booted from the hard disk. I ran e2fsck on both > partitions. After initial errors on /usr, e2fsck apparently no > longer finds errors. In nevertheless never thinks that the > partitions are clean, and continues to insist of checking everything > in full before coming to the conclusion that things are ok. e2fsck-0.2d and earlier had a slight oversight with the valid flag. If they saw a filesystem with the valid bit not set, they would happily go through and check for any errors on the filesystem. Then, if there were no errors, they would say so and exit - forgetting to set the valid flag. Paradoxically, then, the valid flag would only get set if there were errors to be fixed. Remy Card released e2fs-0.2e today, which cures this problem. It is available in /pub/linux/ALPHA/ext2fs on ftp.ibp.fr, and should be available on tsx-11.mit.edu in the same directory soon. Cheers, Stephen Tweedie. --- Stephen Tweedie (Internet: ) Department of Computer Science, Edinburgh University, Scotland. ------------------------------ From: kevin@TorreyPinesCA.ncr.com (Kevin Sanders) Subject: Re: Linux on Macintrashes? (Was: Re: Linux on Amiga?) Date: Thu, 15 Apr 93 19:12:22 GMT In article <1qhhd7INN7m0@matt.ksu.ksu.edu> strat@matt.ksu.ksu.edu (Steve Davis) writes: > >If they don't know about FSF or GNU it is because they *have* been >'assimilated' by the Macintrash empire. If they were dumb enough to >buy an over-priced-one-mouse-button-tiny-black-and-white-screen- >bloated-awful-looking-os-without-a-cli-and-built-by-fascists computer >in the first place, there's really no reason for them to want or need >UNIX. After all, purchasing a Macintrash is acceptance of its point >and drool interface, AND a token of acceptance for Apple's business >practices-- both of which go against the free software/UNIX tradition. > ... >The best way would be to chuck that Mac out the window >and buy a real computer. > >Stratocaster >-- > Steve Davis (I'm a student, not a spokesperson!) > strat@cis.ksu.edu - Kansas State University - Manhattan KS Ahem, just how do you define a "real" computer? I suppose it's one that uses a microprocessor from the One True Micro Manufacturer, Intel? Bullcorn. I don't currently use a Macintosh, but I bought one of the original "Fat Macs" when they came out. I made a decision at that time to support what has always been a superior chip design, the M68K. The Intel's lack of registers and segmented architecture was retro even then; it's quite unfortunate that IBM chose this chip because this set the microprocessor industry back at least 10 years. My point related to this thread is, there are many reasons for choosing a particular machine. How many of us had ever even heard of GNU prior to becoming involved in computing? I don't think GNU was even around when I bought my first Mac (1983 I think). Even if someone knows of GNU, this is only one of the factors entering into a decision of what computer to buy. Personally I would not buy another Macintosh, but this is due to its higher pricing and lack of decent development tools (which is why I got rid of my old one), not some quasi-religious conviction such as you exhibit. -- Kevin Sanders, KN6FQ NCR Torrey Pines kevin.sanders@torreypinesca.ncr.com (619) 597-3602 kevin%beacons@cyber.net ------------------------------ From: hpa@merle.acns.nwu.edu (H. Peter Anvin N9ITP) Subject: Re: Big swap partitions [> 16M] Reply-To: hpa@nwu.edu (H. Peter Anvin) Date: Fri, 16 Apr 1993 02:56:55 GMT In article <1993Apr15.130117.19367@klaava.Helsinki.FI> of comp.os.linux, torvalds@klaava.Helsinki.FI (Linus Torvalds) writes: > > In the default configuration, yes. I think the way I encode the > swap-page numbers actually restrict the number of swap-areas to 127 or > something like that, but you should be able to get more than 8 by just > changing a #define. > > Linus How about getting larger than 16 Mb partitions? Is that also just to change one (or more) #define's? I can't think of anyone who would need more than, say, 16 swap areas, but larger than 16 Mb seems to be a common desire. How about changing 127x16 Mb to, say, 15x128 Mb? /hpa P.S. I know about "using the source", but I am not at my Linux machine right now. I'll look for it when I get home. -- INTERNET: hpa@nwu.edu FINGER: hpa@eecs.nwu.edu BITNET: HPA@NUACC IBM MAIL: 36073 at IBMX400 HAM RADIO: N9ITP, SM4TKN NeXTMAIL: hpa@speedy.acns.nwu.edu This is a test of the emergency USENET system. ^G^G^G^G^G^G^G^G^G^G^G^G ------------------------------ From: bsa@kf8nh.wariat.org (Brandon S. Allbery) Subject: Re: weird vi/less screen insect Date: Thu, 15 Apr 1993 22:18:50 GMT For those nonbelievers out there... we just tripped over the bash signal bug today --- on a Sun IPX/SunOS 4.1.2. It's *not* Linux's fault.... ++Brandon -- Brandon S. Allbery bsa@kf8nh.wariat.org It's not too late to turn back from the "Gates" of Hell... Linux: the FREE 32-bit operating system, available NOW. Why waaaaaait for NT? ------------------------------ From: david@ods.com (David Engel) Subject: Re: A couple of questions: Date: Thu, 15 Apr 1993 22:43:58 GMT Drew Eckhardt (drew@kinglear.cs.colorado.edu) wrote: : Nope, it's based on the umask at the time of the mount call. Instead : of a plan mount, do : umask 22 : mount /dos : : or whatever. For this reason, in my /etc/rc file I have : umask 22 : mount -a : umask 0 : instead of just a mount -a. Or better yet, just use the uid, gid and umask options I added to the msdos fs some time ago. Just put the appropriate options in /etc/fstab and do a regular mount -a. Here is the relevant entry from my fstab: /dev/hda1 /c msdos uid=100,gid=100,umask=022 0 0 David -- David Engel Optical Data Systems, Inc. david@ods.com 1101 E. Arapaho Road (214) 234-6400 Richardson, TX 75081 ------------------------------ From: gbothe@discover.wright.edu (Greg Bothe) Subject: Better? XIAFS or EXT2FS? Date: Fri, 16 Apr 1993 02:34:44 GMT I have been using XIAFS on my system now for about a month and am pretty happy with it; however, I would like to get the most out of my machine. Which is better, xiafs or ext2fs?? Which one tends to be more efficient at keeping down fragmentation? Which one has better error recovery? Which one seems to be the most 'stable'? Xiafs is certainly faster than the old extfs, but I haven't been able to mess around with the ext2fs. BTW, all my partitions (including root) are xiafs and I have had no problems (yet -- I'm knocking on wood :))... Greg Bothe gbothe@discover.wright.edu ------------------------------ From: kowal@stsci.edu Subject: Kernel panic Date: Fri, 16 Apr 1993 04:23:36 GMT This is a newbie question. I was not able to find an answer in the FAQ. I tried to install Linux from the SLS 99.6 files. I got to the part where it asks for the type of drive. At this point I got the message: Unable to handle kernel paging request at address C0000012 Oops: 0000 EIP: 0008:00118468 EFLAGS: 00010206 fs: 0017 base: 00000000 , limit: C0000000 Pid: 3 , process nr: 3 80 78 12 00 75 17 80 78 11 00 After I type in the disk number, I can proceed with fdisk, but at the end I get exactly the same error message, plus a "Segmentation fault". Does anybody know the cause this problem? My setup is: 486-DX2/66 with 8-Meg 212-Meg Western Digital Caviar drive Genoa Local Bus motherboard Genoa 8500VL video card AMI BIOS Also, the instructions in the FAQ say to write-protect all installation disks, but if I do that the installation program complains that the disk is write protected! Many thanks in advance. --Charles Kowal kowal@stsci.edu ------------------------------ From: matt@tahiti.cs90-dev.amex-trs.com (Vulcan) Subject: libolgx.a Date: Thu, 15 Apr 1993 19:47:07 GMT I have noticed an 'olgx-src' file in the linux source tree; is this an implementation of the openlook widget set? Matt Meola matt@amex-trs.com ------------------------------ Crossposted-To: comp.os.coherent,comp.os.minix,comp.os.386bsd.questions,comp.os.misc From: dws@cseg03.engr.uark.edu (David W. Summers) Subject: Re: Unix OS for 286 Date: Fri, 16 Apr 1993 02:22:19 GMT I have Xenix 286 available for $100 (plus the developement system for $100) if anyone is interested. - David Summers (dws@engr.uark.edu) -- "Never under-estimate the bandwidth of a station-wagon David Summers full of tapes, hurtling down the highway." dws@engr.uark.edu - Tanenbaum, "Computer Networks" ------------------------------ From: mporter@cis.ohio-state.edu (matthew dale porter) Crossposted-To: comp.windows.x.i386unix Subject: Dexxa 3-button mouse w/Linux and Xfree 1.x Date: 15 Apr 1993 22:28:40 -0400 I'm upgrading to a 'real' machine to run Xfree on and I'm wondering if anybody has gotten a Dexxa mouse to work under X with all three buttons functional. I noticed that it is listed in the hardware compatibility guide, but it notes that it acts as a Microsoft mouse which indicates to me that possibily the middle button won't work since the Microsoft design is only two buttons. At any rate, does anybody have the middle button working on this el-cheapo mouse so I don't have to junk it? Thanks, -- Matt Porter mporter@cis.ohio-state.edu mporter@magnus.acs.ohio-state.edu ------------------------------ Crossposted-To: comp.os.386bsd.questions From: Brandon.Vanevery@launchpad.unc.edu (Brandon Vanevery) Subject: Summary of Linux vs. 386BSD vs. Commercial Unixes Date: Thu, 15 Apr 1993 22:53:54 GMT Both free unixes are deemed reliable by many sources. 386BSD has better TCP/IP, for now. Linux is far less resource-hungry, due to shared libraries. 386BSD fixes come out slowly, all at once. Linux fixes come out quickly, a little bit at a time. Neither free unix has 24-bit color support. Both will get it at the same time if it comes available, as both use XFree86. Snittily Graphics Consulting Services makes 24-bit X drivers for S3 928 based boards, for several commercial unixes. Commercial unixes are all very expensive. The cheapest is the new Univel UNIXware - $250 for user version, $695 for the developer version. All others are well over $1000. It's really rather silly, I think. SCO Unix has been recommended to me as "the best" of the commercial systems, by a few folks. That's about all the info I have. Look to the FAQs for more precise breakdowns of the commercial stuff. Cheers, Brandon -- The opinions expressed are not necessarily those of the University of North Carolina at Chapel Hill, the Campus Office for Information Technology, or the Experimental Bulletin Board Service. internet: laUNChpad.unc.edu or 152.2.22.80 ------------------------------ From: mrhoten@Xenon.Stanford.EDU (Matthew Rhoten) Subject: liboldX.a, been done? Date: 16 Apr 93 03:37:53 GMT So has anyone hacked up liboldX.a, supporting things like those great Assoc functions? I personally don't write programs for R10, but a few programs (notably gnuchess 3.1) use liboldX for the Assoc functions, and I wouldn't mind having gnuchess (er, rather, xchess) working. By the way, don't buy a Diamond Stealth, it's too much work. -matt -- Matthew Rhoten | mrhoten@cs.stanford.edu | m_rhoten@leland.stanford.edu "When Adam and Eve first saw each other, that's when the blues started." -John Lee Hooker ------------------------------ From: u0xh@jupiter.sun.csd.unb.ca (Figmiester) Subject: xvt... please Date: Fri, 16 Apr 1993 02:33:07 GMT >nation@snoopy.sanders.lockheed.com (Robert Nation) writes: >>I have ported xvt to linux, but I make a bunch of changes along the >>way, so you may or may not be interested. Here's a summary of the >>changes: >[ porting xvt ] >>If you want my source code, I will email it to you. >Robert, I'd appreciate it if you could send it to me. >By the way I tried to mail you but it bounced. Output from nslookup: >% nslookup snoopy.rocket.sanders.com >Server: matt.ksu.edu >Address: 129.130.12.2 >*** matt.ksu.edu can't find snoopy.rocket.sanders.com: Non-existent domain Me too, please send me a copy too... I had the same results as this guy did. Maybe you could post another email address for us interested folks to get a hold of you at instead of filling up the group. Chris ======================================================================= Hellooooo in there ?!? Anybody home ? ......... Huh! No answer!! ------------------------------ ** 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-Activists-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux) via: Internet: Linux-Activists@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 tupac-amaru.informatik.rwth-aachen.de pub/msdos/replace The current version of Linux is 0.99pl6 released on February 23, 1993 End of Linux-Activists Digest ****************************** he middle button won't work since the Microsoft design is only two buttons. At any rate, does anybody have the middle button working on this el-cheapo mouse so I don't have to junk it? Thanks, -- Matt Porter mporter@cis.ohio-state.edu mporter@magnus.acs.ohio-state.edu ------------------------------ Crossposted-To: comp.os.386bsd.questions From: Brandon.Vanevery@launchpad.unc.edu (Brandon Vdigest855 644 36676 74430 51476 5366137102 5767 From: Digestifier To: Linux-Activists@senator-bedfellow.mit.edu Reply-To: Linux-Activists@senator-bedfellow.mit.edu Date: Fri, 16 Apr 93 01:30:15 EDT Subject: Linux-Activists Digest #855 Linux-Activists Digest #855, Volume #4 Fri, 16 Apr 93 01:30:15 EDT Contents: compressed log structured filesystems (Adam J. Richter) Re: Linux on Macintrashes? (STOP ALREADY!) (Roy Wood) Re: Anyone using Orchid VA & S3? (Carl Witty) Re: Is there a .au player for soundcards? (KENNEDY JAMES SCOT) Modem problems (JJ Lay) ---------------------------------------------------------------------------- From: adam@netcom.com (Adam J. Richter) Subject: compressed log structured filesystems Date: Fri, 16 Apr 1993 04:17:02 GMT In article danielsi@cs.utexas.edu (Daniel Aaron Supernaw-Issen) writes: > >about paging and random update: > >it seems to me that the solution is simple. don't compress the file, >compress on the block level. You KNOW that when uncompressing a block how >big it will be. You have to read in the whole block anyways the issue of how >big it is on disk is orthogonal to everything else. prepend the block with >a byte identifying what compression alg (if any was used) and all is taken >care of. I'm currently working on implementing exactly this on a lfs. In >this case, a number of different compression algorithms are used and the one >with the best compression (including no compression) wins. the only real >change to the surrounding lfs code was that the head position had to be >byte indexed rather than block indexed. no biggie. I am also working, on and off, on a compressed log structured filesystem for Linux. I'm busy with a lot of other things (like making those Linux CDROM's that everyone uses), and it's slow going. For example, just figuring out when you're out of disk space is a tough problem on compressed log structured filesystem, and there are some basic kernel modifications that I need to make to, for example, cache file data that doesn't correspond to a particular disk block. However, I thought that I would write this posting in the hopes that even if I don't get this filesystem written, other people doing Linux filesystems might be able to use these ideas. For the benefit of those of you who have not heard of log structured filesystems, I will explain the basic idea. When a traditional filesystem updates a particular block of a file, it moves the disk head and overwrites the disk block corresponding to that particular block of the file (allocating such a block if it doesn't exist). On a log structured filesystem, the writes are recorded in a contiguous log on the disk. So, rather than moving the disk head to the old location of the corresponding disk block, the file block is appended the end of the log. Once that block is written, the block that points to it and the map of free blocks are updated to reflect the fact that the file block has moved. These blocks later written to the logs, followed by the blocks that keep track of them. Eventually this chain of events terminates with a checkpoint block being written to one of a small number of checkpoint regions (for example, there may be two checkpoint regions so that the previous checkpoint is not overwritten). There aren't that many extra blocks that have to be written, because many of the blocks will belong to the same file, so they will share the same indirect blocks and other bookkeeping blocks. Although a log structured filesystem actually writes more data to the disk, it makes up for this, because the disk head doesn't have to do any seeks. By writing in a sequential log, a log structured filesystem can theoretically do sustained random writes at close to the maximum physical bandwidth of the disk. There are a number of ways in which a lot structured filesystem can be implemented. One way is to treat the whole disk as one gigantic circular log. The file system that Daniel is talking about works that way. On such a system, one can guarantee a certain amount of free space for the log by simply rewriting any live data blocks that are within that amount of space from the front of the log. The problem with this approach is that a disk is generally going to be about 75% full (otherwise the user would have bought a smaller disk), so about 3 blocks will be rewritten for every new block that is written. This problem can be reduced substantially if the filesystem simply skips the blocks that are in use, but, it takes just as long for the disk head to skip over a single block as it does to write one. In order to increase bandwidth, it would be necessary to skip whole tracks, and whether even that is enough is debatable. A more popular way to organize a log structured filesystem is to divide it into segments. The log is written into an empty segment. When the system is running low on empty segments, it reads the live data from a few live segments into one of the remaining empty segments ("cleaning"). In this way, seldomly changed data is stuffed into segments that rarely have to be cleaned. Proponents of log structured file systems argue that as RAM size grows, read hit ratios on disk caches will increase toward 100%, while a steady trickle of disk writes will continue to be necessary, since dirty blocks are written to the disk every 30 seconds. According to this reasoning, the relative performance of a filesystem that maximizes write bandwidth should improve in comparison to a traditional filesystem as RAM sizes grow. Log structured filesystems suffer from blocks that need to be read together being dispersed all over the disk. In comparison, the 4.3 BSD filesystem tries to put the blocks of a all of the files in a directory in the same part of the disk. When files that are read together are also written together, then a log structured filesystem has reasonable locality (e.g., a bunch of .o files are written during a make and are then linked into an executable), otherwise a log structured filesystem can do some other things to improve locality. Firstly, a log structured filesystem can write all of the dirty blocks for a file together. A log structured file system may also be able to get similar locality by selecting how it rearranges blocks when it does segment cleaning (or some other type of garbage collection if it isn't a segmented LFS). The time that it takes to run the segment cleaner is the other thing that brings the performance of log structured filesystems down to the level of traditional filesystems. On the other hand, if the workload of a system is primarily reads, it can run the segment cleaner at night. Whether log structured filesystems are a win in practice is a popular subject of debate. Current benchmarks suggest that log structured filesystems perform somewhat better that traditional filesystems in some areas and somewhat worse than others. Since a log structured filesystem needs a segment cleaner, a fairer comparison would probably be between a log structured filesystem and its cleaner versus a traditional filesystem and a defragmenting program. My primary interest in a log structured filesystem is not for improving performance, but, rather, for doing a compressed filesystem. The advantage of using a log structured filesystem for compression is that the data that is written to the log can be written to byte boundaries rather than disk block boundaries. Squeezing out every byte from compression is not the only benefit of this approach. You can also adopt the convention of truncating every block at its last non-zero byte. On a traditional filesystem, you waste about half of a block per file, because the last block will be, on average, only half full. If your average file size is 10kB, then your losing about 5% due to this internal fragmentation. If you're running a news server, in which case, you'll have article sizes of around 2kB, then you're losing 25% to internal fragmentation. Currently, files larger than about 12kB need to use an indirect block which is generally going to have a few non-zero entries followed by mostly zeroes. By using a log structured filesystem you can save space by making more efficient use of compression and by trimming trailing zeros from each block. In addition, compression of metadata such as indirect blocks and directory entries is possible (but not required), random reads and random writes are fast, and data loss due to media failure does not effect the decompression of other blocks of same files that were corrupted by the media failure. Other possible features that can be added to log structured filesystems include: 1. RAID (Rapid Array of Inexpensive Disks). Given n identical disks, break each segments into n-1 parts. Distribute these n-1 parts to the same regions of each of the first n-1 drives. The last drive will hold the XOR of the n-1 parts of the segment, so that the contents of any drive can be reconstructed in the event of a drive failure. More complex arrangements of drives and parity sectors are possible. The problem with this scheme is that, for writes, all of the disk arms need to be used together, so the segments have to be fairly large for the system not to spend most of its time seeking. 2. REPLICATING SEGMENTS. It may be advantageous to replicate frequently read but rarely modified segments to different parts of the disk to reduce seek times. 3. more flexibility for ENCRYPTION. Some encryption algorithms change the size of their data. It's easier to support such algorithms if your block size is variable. Admittedly, this is a pretty contrived advantage. There are also other advantages of using a filesystem that reallocates blocks instead of overwriting them, such as a log structured filesystem: 1. CRASH PROOFING in the sense that the filesystem will be in a consistent state when the system reboots, provided that a crash will leave any block that was being written to disk either completely written, completely unwritten, or in a detectable error state. 2. FILESYSTEM FORKING: the ability to instantly clone the filesystem into basically two parallel universes that use additional disk sectors on a copy-on-write basis only for the things that they change (useful for taking a snapshot to do a disk backup). 3. TIME TRAVEL (basically this involves holding onto a snapshot of the filesystem at every consistency point for the period that you want to be able to visit). 4. easily adapted to WRITE-ONCE MEDIA 5. DIGITAL SIGNATURES. Add a new field in the data structure that points to a disk block: the digital signature of the contents of that block. When you write a new block, you update not only the address of that block, but also the system's signature of that block. Any attempt to write to alter filesystem without the system's digital signature can be detected (perhaps the signature would have to be provided at mount time). The one exception is that an adversary could overwrite the disk with a snapshot of the disk at some previous time. It may also be possible to extend this scheme to have digital signatures for each user, but I haven't worked all of the details out. The problem is that the system has to be able to relocate a user's files without the user's digital signature without giving an adversary the ability to restore another user's file to a previous state. If these problems can be worked out, it would make it possible to require every user to sign his or her setuid files, for example. 6. OBSCURING DELETED DATA. Similar to #5, but instead of signing each block, a random key is generated every time a new block is to be written. The new block is encrypted with that key, and the key is stored with the pointer the block. The pointers to the indirect blocks form a hierarchical encryption scheme, so that as soon as an old checkpoint is overwritten, the key to decrypt any data that is not pointed to by a current checkpoint is lost. It is worth noting that it would not be too difficult to adapt a traditional filesystem like ext2fs to reallocate blocks instead of overwriting them. The only changes in the disk format would be to use data structures that could be updated by block reallocation for things like the inode table and the free block table. The easiest way to make these changes is to treat the inode table and the free block table as regular files. Before anybody rushes off to change all of the Linux filesystems to do reallocate on write, it is worth noting that reallocating blocks is not without cost. Reallocating blocks requires additional disk writes to update the block that pointed to the block that has been updated, to update the list of free blocks, and to do similar bookkeeping for these other blocks which had to be written. The only things in this posting that are actually my ideas are the suggestions for trimming trailing zeros, encryption, replicated segments, digital signatures, and obscuring deleted data. If you want to learn about log structured filesystems, here are some references. Ousterhout, John and Fred Douglis _Beating the I/O Bottleneck: A Case for Log-Structured File Systems_, Report UCB/CSD 88/467, October 1988 Burrows, Michael, Charles Jerian, Buttler Lampson, Timothy Mann _On-line Data Compression in a Log-structured File System_ Report #85, April 15, 1992, Digital Systems Research Center Rosenblum, Mendel, _The Design and Implementation of a Log-structured File System_, UCB/CSD 92/696, June 1992 Seltzer, Margo, Keith Bostic, Marshall Kirk McKusick and Carl Staelin _An Implementation of a Log-Structured File System for UNIX_ January 1993 USENIX conference proceedings, pp. 307ff -- Adam J. Richter Yggdrasil Computing, Incorporated 409 Evelyn Ave., Apt. 312, Albany CA 94706 PO Box 8418, Berkeley CA 94707-8418 (510) 528-3209 (510) 526-7531, fax: (510) 528-8508 adam@netcom.com yggdrasil@netcom.com Another member of the League for Programming Freedom (lpf@uunet.uu.net). ------------------------------ From: rrwood@canrem.com (Roy Wood) Subject: Re: Linux on Macintrashes? (STOP ALREADY!) Date: Thu, 15 Apr 1993 23:01:11 GMT Look, this sort of adolescent bickering about the merits/shortcomings of a particular brand of computer were fun when I was an adolescent, but now they're just irritating. Please take it somewhere other than comp.os.linux. And since I can't stop myself from adding.... Computers are just tools. Just as most people don't use hacksaws to pound nails, so too do certain computers/OS's have sensible applications. I've got access to Macs and PC clones, and whatever does the job best gets used at that particular moment. Of course, if you're determined to be irrationally biased as to the fact that your particular computer is the greatest all-around computing device since Charles Babbage first coined the expression "Difference Engine"-- well, I'm certainly not going to waste any more time talking to you. -Roy ------------------------------ From: cwitty@ai.mit.edu (Carl Witty) Subject: Re: Anyone using Orchid VA & S3? Date: 16 Apr 93 00:45:34 In article gvreugde@plg.uwaterloo.ca (Gord Vreugdenhil) writes: I am upgrading my system at the end of the month. The video card seems to be the hardest decision.... I suspect that the Orchid VA/VLB (Vesa) card will be my choice. Is anyone using this card with the S3 server? If so, let me know. I would prefer to know that it works before buying the card... :-) Alternatively, a local bus card which does 8514/A would be nice, though I would prefer not to spend the money for an ATI Ultra. Yes, I use the VA/VLB with XS3. I had to add: vendor "nolinear" to my Xconfig to get it to work. I didn't do any benchmarks or anything, but it seemed plenty fast enough to me. Carl Witty cwitty@ai.mit.edu ------------------------------ From: kennejs@a.cs.okstate.edu (KENNEDY JAMES SCOT) Subject: Re: Is there a .au player for soundcards? Date: Thu, 15 Apr 93 23:22:30 GMT From article , by ins407x@aurora.cc.monash.edu.au (Andrew J. Cosgriff): > damien@b63519 (Damien Neil) writes: > >>This reminds me -- is there an .au player that _will_ use a soundcard? I have >>splay from the snd-util package, but the sound it produces when playing an .au >>file makes it seem like I'm getting a poor radio transmission...from Venus. Not >>the program's fault of course -- it wasn't designed to use the .au format. But >>it would be nice to have access to all the Sun audio files out there. > > Yep, there is... > > It's called /dev/audio (have you mknod'ed one ?) > > cat your sun audio files there and they'll work just fine. > > It'd be trivial to modify the splay code to use /dev/audio if you told it > you had a .au file I suppose. (One day when I've got nothing else to do...) > > /dev/dsp (what splay uses) is fine for .voc and .wav files (you just have > to ignore the header at the start of wav files which comes out as a short > burst of static). > > /dev/audio is for Sun audio files. > > > Enjoy ! > > > > -- > ------------------------------------------------------------------------------- > * Andrew J. Cosgriff ! * "Connect the goddamn dots !" (Ministry) > ins407x@aurora.cc.monash.edu.au #include > andrew@bing.apana.org.au Bing ! How about just using a utility to convert Sun audio files over to a format that splay can handle? I have a program called SOX that can translate au sounds into voc sounds. SOX is capable of doing other kinds of conversions as well. Of course you'll have to go to the trouble of converting the sounds over first before you can play them, but that shouldn't be a big deal. Unless I'm mistaken, you can get SOX off of nic.funet.fi. It should be under /pub/unix/sound. The name of the file is sox7.linux.tar.Z. ---Scott ------------------------------ From: csjjlay@knuth.mtsu.edu (JJ Lay) Subject: Modem problems Date: 15 Apr 1993 23:11:44 -0500 Greetings all! I have the SLS version of Linux with a 0.99.6 kernel running on a 386DX with 8Mb RAM a mouse on /dev/ttyS0 and a modem on /dev/ttyS1 When I enter kermit I get a I/O error: can't read from line or something similar. The permissions on this line were originally rwx-w--w- but I played with all variations with no positive results. *NO* communications programs will work. This modem is a Hayes 2400 compatible. I did not have this problem under 0.99.4 TAMU. All help is appreciated. ------ JJ LAY ======================================================================== JJ LAY CENTER FOR HISTORIC PRESERVATION COMPUTER SPECIALIST MIDDLE TENNESSEE STATE UNIVERSITY csjjlay@mtsu.edu MTSU BOX 80 (615) 898-2658 Murfreesboro, TN 37132 ======================================================================== "Marriage is a noble institution... If you want to be institutionalized the rest of your life!" -Dr. Paul Hutcheson, MTSU Computer Science Dept ------------------------------ ** 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-Activists-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux) via: Internet: Linux-Activists@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 tupac-amaru.informatik.rwth-aachen.de pub/msdos/replace The current version of Linux is 0.99pl6 released on February 23, 1993 End of Linux-Activists Digest ****************************** nnejs@a.cs.okstate.edu (KENNEDY JAMES SCOT) Subject: Re: Is there a .au player for soundcards? Date: Thu, 15 Apr 93 23:22:30 GMT From article , by insdigest856 644 36676 74430 53161 5366137102 5761 From: Digestifier To: Linux-Activists@senator-bedfellow.mit.edu Reply-To: Linux-Activists@senator-bedfellow.mit.edu Date: Fri, 16 Apr 93 02:30:10 EDT Subject: Linux-Activists Digest #856 Linux-Activists Digest #856, Volume #4 Fri, 16 Apr 93 02:30:10 EDT Contents: X Windows and TVGA8900C (JJ Lay) Re: Why Linux should remain Linux and seagulls should not be used. (Kevin Sanders) Re: Can't get nnmaster to run. (Ralf Reddemann) Re: Automated Linux PowerDown (Grant Taylor) Questions about virtual mem. (Peter A. Schwenk) Install/Configure help needed (Ted) Re: crash: 0.99p7A + emacs -> swapper killed! (Art Taylor) problems with TERM109 (detailed description) (Keith Smith) problems with TERM109 (SUMMARY) (Keith Smith) Re: Ext-2 FS bug (Scott Taylor) Re: xvt for Linux??? (Nan Zou) Allowing others to mount ? (Pramod Charles John) Will Amiga Linux support AGA? (KENNEDY JAMES SCOT) LHA compression (Figmiester) ---------------------------------------------------------------------------- From: csjjlay@knuth.mtsu.edu (JJ Lay) Subject: X Windows and TVGA8900C Date: 15 Apr 1993 23:14:12 -0500 I have the newest SLS release of Linux running on a 386DX with 8Mb of RAM, a TVGA 8900C chipset and Microsoft 2button serial mouse on /dev/ttyS0 When I start X up it gives me a bunch of "wavey" lines. I followed the hints given by someone to use the VESA timings. Any other suggestions? ------ JJ LAY ======================================================================== JJ LAY CENTER FOR HISTORIC PRESERVATION COMPUTER SPECIALIST MIDDLE TENNESSEE STATE UNIVERSITY csjjlay@mtsu.edu MTSU BOX 80 (615) 898-2658 Murfreesboro, TN 37132 ======================================================================== "Marriage is a noble institution... If you want to be institutionalized the rest of your life!" -Dr. Paul Hutcheson, MTSU Computer Science Dept ------------------------------ From: kevin@TorreyPinesCA.ncr.com (Kevin Sanders) Subject: Re: Why Linux should remain Linux and seagulls should not be used. Date: Fri, 16 Apr 93 01:55:18 GMT In article <93104.131126ACPS7221@RyeVm.Ryerson.Ca> writes: > How about a Tiger? A Tiger is powerful, fast, sleek, efficient, and so is > Linux. > And also, Linus is a roarin' good programmer:) > > But seriously, why not go for a a really cool image like a tiger instead of > a seagull, which is regarded as a rodent with wings by some. (not me). > > Seagulls are not that bad an idea though, especially if they are in > mid-flight. > > -mc. Yes! DOS slowing you down? Put a TIGER in your tank! (I think Esso's/Exxon's copyright on this slogan has run out, hasn't it?) or DOS got you down? Stuck in a rut? Give Linux a try And dump that DOS slut! Burma shave -- Kevin Sanders, KN6FQ NCR Torrey Pines kevin.sanders@torreypinesca.ncr.com (619) 597-3602 kevin%beacons@cyber.net ------------------------------ From: bill@akron.toppoint.de (Ralf Reddemann) Subject: Re: Can't get nnmaster to run. Date: Thu, 15 Apr 1993 20:06:47 GMT cassel@tsoft.net (Greg Cassel) writes: : I can't seem to get nnmaster working on my Linux machine. : : If I try to run it, I get the error: : NNMASTER FATAL ERROR : control (0): cannot open 'x' file (mode=82, errno=13) : : : Anyone have any idead? That is a permission problem. Read /usr/doc/nn6.4.18/PROBLEMS ! First be sure that /usr/local/lib/nn looks like this: drwxr-xr-x 3 news news 184 Apr 14 22:25 . -rw-rw-rw- 1 news news 5144 Apr 15 16:34 Log -rwxr-xr-x 1 news news 8142 Dec 7 20:41 aux -rwxr-xr-x 1 news news 926 Dec 7 20:41 back_act -rw-r--r-- 1 news news 1170 Dec 7 20:41 conf drwxr-xr-x 2 news news 440 Apr 14 22:13 help -rwsr-xr-x 1 news news 40316 Jan 6 23:24 nnmaster -rwxr-xr-x 1 news news 884 Dec 7 20:41 nnspew -rwxr-xr-x 1 news news 1096 Dec 7 20:41 upgrade_rc Then you have to correct the permissions of NN's database, simply type: chown -R news /usr/spool/nndb chgrp -R news /usr/spool/nndb chmod -R 755 /usr/spool/nndb Ralf -- Ralf Reddemann Kiel, Germany Voice: +49 431 85442 e-mail: bill@akron.toppoint.de ------------------------------ From: gtaylor@jade.tufts.edu (Grant Taylor) Subject: Re: Automated Linux PowerDown Date: Thu, 15 Apr 1993 23:08:52 GMT some ups's do this sort of thing the other way around - they have the power supply notify the computer a little before the power goes away. while this is obviously the best way for a ups, mightn't it also be easier for this purpose, too? having the computer look for an asserted something in a serial line and then run halt shouldn't be particularly difficult to swing. -grant ------------------------------ From: schwenk@fred.cis.udel.edu (Peter A. Schwenk) Subject: Questions about virtual mem. Date: Fri, 16 Apr 1993 04:32:29 GMT I have a few questions about Linux. I haven't read a FAQ, so bear with me. Does Linux support virtual memory, or once system RAM is used up, crash city? How stable is Linux? Is it only good for playing with? Where is Linux available? Thanks for listening. Peter Schwenk (schwenk@cis.udel.edu) ------------------------------ From: tlilley@jarthur.claremont.edu (Ted) Subject: Install/Configure help needed Date: Fri, 16 Apr 1993 04:30:34 GMT Hi there, I picked up the SLS Linux at sunsite, and after a number of different repartitions and installs :), I have the proper disk setup, however I have not been able to get lilo or tcp/ip to work properly. First, the lilo problem. I run DOS, OS/2, and now Linux, and have the OS/2 boot manager installed. I used OS/2's fdisk to partition the system as detailed in the FAQ. I have attempted to use lilo to make my Linux partition bootable, but it consistently gives me a message very close to "LILO signature not found on boot sector" and won't install. I used the command line given in the FAQ, to wit "/etc/lilo/lilo -c -b /dev /hda6 -v -v /Image", and I have attempted the install on a primary and logical drive (and the extended partition holding the logical drive), all to no avail. The lilo version is 7, I believe. Second, the tcp/ip problem. The problem really isn't tcp/ip, rather it's the IRQ on my card. I have a 3c503 Etherlink II, on IRQ 3, but autodetect attempts to use IRQ 5. So, I figured I needed to recompile the kernel with the changed IRQ. I attempt to follow the instructions in the /usr/src/linux/README file, however, "make dep" reports a number of missing header files (mktime.h, etc.), and plain "make" fails as well. What's up with that? Thanks in advance, Ted ------------------------------ From: root@fab4box.wa.com (Art Taylor) Subject: Re: crash: 0.99p7A + emacs -> swapper killed! Date: Fri, 16 Apr 1993 03:08:35 GMT In article <1993Apr14.173533.410@desaster.hanse.de> michaelw@desaster.hanse.de (Michael Will) writes: >Hello, > >I was just downloading my news with uucp, replied to a message with elm+emacs, >when suddenly my machine got killed in an unusual way. > >general protection: 6e70 >EIP: 0008:00D152 >EFLAGS: 00010002 int = 58 9a 37 2a a8 c6 0e 3e 3a c9 cf 40 46 08 b9 39 >fs: 0017 >base: C000000 limit: 00200000 >Pid: 0, process No. 0 >cf: 00 00 00 00 00 00 00 00 >task[0] (swapper) killed: unable to recover >kernel panic: Trying to free up swapper memory space >in swapper task - not syncing > >The machine stopped - I had to reboot cold. Ok, I got the same response when dl'ing with UUCP(taylor 1.04), and scrolling up through a command buffer(up arrow, once in bash, once in ATP(.QWK packet offline reader). >I used 0.99p7A, with scsi tape+disk enabled, I have 0.997A, no SCSI, TCP/IP, all FS but NFS and ISO, AutoIRQ, PS2Mouse and MSBusMouse. >a root-partition of 120MB efs on dev/hdb1 61545 block(1024) minix-fs(14 char filenames) on hda1 >2 partitions e2fs on hda1 1 15385 efs(old) on hda2 >1 swappartition (happens to be 10MB because I was lazy). 6 Meg swap on hda3 >i386-dx-33 8MB RAM - 2x120MB-At-bus-drives, 1-scsi-tape i386sx/20(32k cache) WD 85 MB IDE drive. It would seem that UUCP is the common link here. I've only experienced this while running Taylor 1.04, and scrolling up. It's happened twice, and it shocked hell out of me both times. Oh well. If this statistic is of any use to anyone, good luck. -- "Give a family foodstamps, and they'll survive for a week; teach them how to forge foodstamps, and they're set for life." Art Taylor root@fab4box.wa.com (509)522-1178;(509)522-1086 126 Otis Street reeses@halcyon.com Walla Walla, WA 99362 ------------------------------ From: ksmith@monu6.cc.monash.edu.au (Keith Smith) Subject: problems with TERM109 (detailed description) Date: Fri, 16 Apr 1993 02:27:27 GMT G'day all, I've been running Linux (I'm pretty sure that's right).99P6 for a few days now, and have been having a ball. The fun stopped, though when trying to use TERM. I'll do my best to give you a run down on the details of what happens when I try to use it, hopefully someone will be able to point me in the right direction. First off, I HAVE read the docs that come with the source code for TERM109, which I obtained from tartarus.uwa.edu.au. They are quite comprehensive and come with a README that gives a step by step description of how to get it up and running. Here's the system I'm using (I don't think it's too relevant, but I should say in case someone spots a prob with it): 486-33 215K cache 8meg ram 80 Meg Western Digital (IDE) running DOS only 250 Meg Maxtor (IDE): hdb1 150Meg DOS hdb2 65 Meg Linux mounted as / hdb3 32 Meg Linux (roughly 32 meg, anyways), not mounted hdb4 8 Meg Linux SWAP I'm running a standard Hayes 2400 baud modem to a DEC station. The modem does not have MNP4 or 5 (error correction or compression). I don't mean to over simplify things, guys, but I don't want to skip anything. I'm booting a Linux image from a 5.25" A drive, using the disk that the SLS install disks make after they put the A series on the hard drive. I am logged on tty1 as ROOT. I use XC -l /dev/modem (/dev/modem is a link to /dev/ttyS0), and have no problems logging in at 2400 baud. OK from here on, REMOTE system refers to the INTERNET machine LOCAL system refers to MY machine at home. The REMOTE system has no TERMDIR set, and therefore refers to ~/.term to find the termrc file. The REMOTE system is running CSH, though I've been told the shell shouldn't make a difference. The LOCAL system has TERMDIR set to /tmp, and therefore refers to ~/tmp/.term for it's termrc file. The LOCAL system is running TCSH. The contents of the TERMRC file on both machines are identical and are as follows: escape 17 escape 19 escape 147 escape 149 I have also tried it with other ESCAPE clauses (including two that escaped 25% of the character set, as suggested in the term/README). I know that I can run linecheck to see what characters need to be escaped, but this TERMRC has worked for people who live close by to me. Btw, line noise is very low on my exchange, ie. have never seen bad characters on the screen (apart from when someone picks up the phone :) ) I run TERM on the REMOTE machine, and return to the XC command prompt. I enter "$ term -d64" which runs another TCSH shell with stdin and stdout set to /dev/modem by default. Alternatively, I can exit XC completely and type "TERM -d64 < /dev/modem > /dev/modem. Both options give the same results. OK, to my knowledge I've done everything right so far. I run another ROOT login on tty2, and type "TRSH" The modem TX and RX lights flash a couple of times, then the machine freezes. I look at tty1 where TERM is running locally, and I see that a few bytes have gone back and forth between the REMOTE and the LOCAL machine, but tty2 is just sitting there with nothing happening. As a test, I decided to see if I could use TUPLOAD, and I can, it works fine. I can login as ROOT on 5 consoles and run FIVE TUPLOAD sessions, swap back to TTY1 to check the DEBUG output from the LOCAL TERM, and see it furiously sending packets back and forth to the 5 local clients. So how come I can use TUPLOAD but not TRSH? Could someone PLEASE let me know what I might be doing wrong? Thanks for your time, hope I provided enough detail to let you know the situation. EMAIL to ksmith@monu6.cc.monash.edu.au, or to this group, I don't care. Keith -- ===================================================================== - ksmith@monu6.cc.monash.edu.au | Keith Smith, Monash Uni, - - ksmith@vx24.cc.monash.edu.au | Melbourne, Australia - ===================================================================== ------------------------------ From: ksmith@monu6.cc.monash.edu.au (Keith Smith) Subject: problems with TERM109 (SUMMARY) Date: Fri, 16 Apr 1993 02:31:27 GMT This is a summary of the article I posted detailing my problems with TRSH. In short, is there anyone who knows why I might be having problems getting TRSH to give me a shell, considering TUPLOAD works just fine? Some settings: LOCAL: termdir = /tmp shell = /bin/tcsh REMOTE: shell = /bin/csh termdir is not set on the REMOTE machine, so it should use ~/.term as the directory to house TERMRC. TERMRC on both machines: escape 17 escape 19 escape 145 escape 147 If you need more details, check out the other post I made, entitled "problems with TERM109 (detailed description)". ANideas welcome!! Cheers, Keith -- ===================================================================== - ksmith@monu6.cc.monash.edu.au | Keith Smith, Monash Uni, - - ksmith@vx24.cc.monash.edu.au | Melbourne, Australia - ===================================================================== ------------------------------ From: n217cg@tamuts.tamu.edu (Scott Taylor) Subject: Re: Ext-2 FS bug Date: 16 Apr 1993 05:24:45 GMT In article <193.2BCDC2DC@purplet.demon.co.uk> Enrico.Scotoni@purplet.demon.co.uk (Enrico Scotoni) writes: >KD>I just encountered what seems to be a serious bug in > >the ext-2 filesystem. > >KD>Over the weekend, I transferred some 200 files from > >another computer over a serial link using kermit > >into a 70-80 meg ext-2 spare partition on my Linux > >386DX-40 system (Ultrastor 14F, Quantum LP240S). After > ^^^^^^^^^^^^^ >I think there is the Problem. I WAS using the 14F on my system and >ext-fs (NOT ext-2-fs). After Linux 0.99pl7 there is a new driver for the >14F which is supposed to be faster.IT IS VERY FAST (in screwing up my >filesystem). I could easily repoduce the problem with the following This is harsh! ;-) >command: > >cd /usr ; for i in * ; do (tar cf ${i}.tar ${i} ; rm ${i}.tar) & ; done > >It starts running and after few minutes I get lots of error messages and >finally the system crashes. when trying to fix the filesystem i also get >so much errors that there is NO CHANCE to fix it. Iam not sure if there >really is to blame the disk-driver or the controller. I already twice >sent a problem-description to c.o.l.but it either did not get out or >nobody took any notice of it. Anyhow, I am using at the moment a It didn't make it here. I read c.o.l. pretty religiously, and I never saw either of your posts. I have been using the current UltraStor driver ever since I started working on it about three months ago. Since then, I have used it (sometimes heavily) with every available filesystem (minix, ext, xia, ext2) with no problems. Until 99.7A, my /tmp and /usr filesystems were ext; now they are ext2; every- thing has run smoothly on my 14F. Also, I have received several reports of others using the driver with no problems, and only one bug report (fixed in 99.7A). >friend's AHA1542B and I ordered a new AHA1542C. With the AHA1542B I have >no problems at all, hopefully my new AHA1542C will behave the same. > >Anybody interested on a used (3 months old) UltraStor 14F ???? :-) If I needed a really good SCSI adapter, I'd buy it in a second! ;-) But I already have one. Seriously, you should try to return it; it may very well be defective. Scott n217cg@tamuts.tamu.edu ------------------------------ From: nan@matt.ksu.ksu.edu (Nan Zou) Subject: Re: xvt for Linux??? Date: 15 Apr 1993 18:23:20 -0500 nation@snoopy.sanders.lockheed.com (Robert Nation) writes: >I have ported xvt to linux, but I make a bunch of changes along the >way, so you may or may not be interested. Here's a summary of the >changes: [ porting xvt ] >If you want my source code, I will email it to you. Robert, I'd appreciate it if you could send it to me. By the way I tried to mail you but it bounced. Output from nslookup: % nslookup snoopy.rocket.sanders.com Server: matt.ksu.edu Address: 129.130.12.2 *** matt.ksu.edu can't find snoopy.rocket.sanders.com: Non-existent domain -- Nan -- "Linux -- the choice of a GNU generation" -- Steve Frampton ------------------------------ From: pcjg7045@dcl-nxt25 (Pramod Charles John) Subject: Allowing others to mount ? Date: Thu, 15 Apr 1993 22:52:22 GMT Hi, How can I allow other users to mount files like the floppy for example. My problem is that I want to allow other users to use their msdos - floppy s in the drive, but only root can mount. Another solution to this problem would be if someone could tell me how I can get mtools to recognize a low density 5 1/4" floppy. For some reason, it seems that fdprms has the requisite header but it seems to be unable to read it. (unless my fdprm is wrong). Any help would be appreciated. How can I give super-user (limited) status to some? Thanks, Pramod John ------------------------------ Crossposted-To: comp.sys.amiga.hardware,comp.sys.amiga.misc From: kennejs@a.cs.okstate.edu (KENNEDY JAMES SCOT) Subject: Will Amiga Linux support AGA? Date: Fri, 16 Apr 93 00:24:41 GMT I understand that there are some people who are currently working on creating a version of Linux for the Amiga. My question is this: will Amiga Linux be able to take advantage of the new AGA chipset that is in the Amiga 1200 and Amiga 4000? It would really be nice if it could. Then you'd be able to run X with 256 colors in resolutions such as 640x480 and 800x600 (more with overscan). Plus, the video would be a helluva lot faster. As a matter of fact, my Amiga 1200 (14.3Mhz 68020) has video that is *much* faster than my 33Mhz 486DX with SVGA card. Unfortunately, my A1200 doesn't have a MMU in it. Is there anyway that I could install one without having to buy an accelerator card? I'm thinking about purchasing a 68030 board sometime in the future anyhow but it would be nice if I could just buy a MMU and pop that in there instead---this would save me a lot of money. Then later on I could buy an accelerator to get better performance. Perhaps Amiga Linux could come with a set of drivers that would allow you to use either the original, enhanced, or AGA chipsets. Just install the appropriate driver then you're in business. This scheme would also have another advantage: whenever Commodore releases a new chipset a driver could be written to take advantage of it. So, Amiga Linux would be able to run on any new models that C= comes out with. Would it be difficult to do this? Remember, if Amiga Linux will only be able to run on the old Amigas then you'd only be able to bring up X with 16 colors at 640x200 or 640x400. Worse yet you'd be limited to either 512K or 1024K of chip RAM, depending upon what kind of Agnus you have under the hood. I realize that work is still underway on Amiga Linux and that X isn't of course. Probably a long ways off I gather; have to come up with a working kernel first! However, what I'm saying is still something that should be taken into consideration, if for no other reason then for the compatibility issues involved here. ---Scott kennejs@a.cs.okstate.edu ------------------------------ From: u0xh@jupiter.sun.csd.unb.ca (Figmiester) Subject: LHA compression Date: Thu, 15 Apr 1993 23:27:35 GMT I looked on sunsite.unc.edu in the directory /pub/Linux/utils/compress, but could not find a compression package to handle LHA type compression. Reason I am looking for one is that I recently compiled the sound blaster support into my kernel, and I want to get my hands on some .mod files. Most of the ones I know about are compressed using LHA compression. The site I know about that has some is nic.funet.fi in directory /pub/amiga/audio/modules. These are all stored as LHA. Does any know where I can get the linux port of LHA? Or, better yet, a site that has .mods using zip compression. Thanks for any help. Chris ==================================================================== Huh? dot sig what? ------------------------------ ** 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-Activists-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux) via: Internet: Linux-Activists@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 tupac-amaru.informatik.rwth-aachen.de pub/msdos/replace The current version of Linux is 0.99pl6 released on February 23, 1993 End of Linux-Activists Digest ****************************** how I can get mtools to recognize a low density 5 1/4" floppy. For some reason, it seems that fdprms has the requisite header but it seems to be unable to read it. (unless my fdprm is wrong). Any help would be appreciated. How can I give super-user (limited) status to some? Thanks, Pramod John ------------------------------ Crossposted-To: comp.sys.amiga.hardware,comp.sys.amiga.misc Frdigest857 644 36676 74430 75325 5366137103 5771 From: Digestifier To: Linux-Activists@senator-bedfellow.mit.edu Reply-To: Linux-Activists@senator-bedfellow.mit.edu Date: Fri, 16 Apr 93 04:00:14 EDT Subject: Linux-Activists Digest #857 Linux-Activists Digest #857, Volume #4 Fri, 16 Apr 93 04:00:14 EDT Contents: working crond package uploaded to sunsite (Grant Taylor) problems using HD_TYPE to avoid using BIOS (Daniel Aaron Supernaw-Issen) Unix OS for 286 (Scott Philip Sable) EISA video cards supported? (Craig Horman) *** READ THIS BEFORE POSTING *** (Ian Jackson) Linux&Compaq Deskpro same old prob : ( Re: can 386sx run linux (Christopher Kuperman) TIMEOUT ON CDROM INSTALL (Cliff V Collins) Problems configuring news, etc (Gerry George) [Q] Linking thru serial lines. (R. Ramesh) Re: crash: 0.99p7A + emacs -> swapper killed! (Linus Torvalds) Serial communications (Re: Two questions) (Juha Ursin) Re: Big swap partitions [> 16M] (Linus Torvalds) Re: Xfree 1.2 leaves screen in graph-mode after exit, runx !work ( Joel M. Baldwin) problems with tar: broken pipe, Problems with lpd (Juerg Thoeny) ---------------------------------------------------------------------------- From: gtaylor@jade.tufts.edu (Grant Taylor) Subject: working crond package uploaded to sunsite Date: Fri, 16 Apr 1993 06:05:31 GMT I found that the crond in the vixie-cron package on sunsite & tsx was sufficiently old that it no longer works, so i extracted working binaries from the SLS and found REAL manpages elsewhere to get it going on my system. Since this took me some time (finding a specific binary in the SLS in not easy) I figured it would be good to share my results: sunsite.unc.edu:/pub/Linux/Incoming/cron.tar.z is essentially a replacement for .../system/Daemons/vixie-cron.tar.Z, but with man pages and in the usual sls tar format; tar xzvvpf it from / and all will be well once /etc/crond is added to /etc/rc. note that the man pages are gzipped for the new perl man. -grant ------------------------------ From: danielsi@cs.utexas.edu (Daniel Aaron Supernaw-Issen) Subject: problems using HD_TYPE to avoid using BIOS Date: 16 Apr 1993 01:14:37 -0500 hey all. This is yet another post requesting help on the same old saga of getting my mfm and ide to coexist. Numerous people sent me email suggesting that I hard code my drives geometry using the HD_TYPE macro defined in config.h and take my ide off my cmos. This seems like a resonable idea. I tried this and found that the image I generated still used the bios so I was hosed. I confirmed that it was using the bios by noticing the following: When I originally booted up with the mfm in and the ide not specified the system got a kernel panic telling me that it couldn't mount root. When I removed the mfm and left the rest alone, I got the same error. When I reset my cmos and put my ide back on it, the kernel came up fine. strange. I could be using the HD_TYPE macro incorrectly. oh, before I forget, yes I did comment out the line #undefing HD_TYPE. My ide disk has 8 heads, 33 sectors and 776 cylinders, the landing zone is 776. My mfm *I THINK* has the following geometry: 4 heads, 17 sectors 615 cylinders, wpcom 128 and a landing zone of 638. <- I got these because the only documentation on my mfm says that on a compaq (not that I have one) it should be listed in the cmos as a type 2. So I called compaq and asked about the geometry of a type 2 disk. I'm pretty hosed. Any help is greatly appreciated. I'm using the 0.99pl8 kernel. Thanks in advance. Daniel Supernaw-Issen danielsi@cs.utexas.edu Linux or Die! ------------------------------ Crossposted-To: comp.os.coherent,comp.os.minix,comp.os.386bsd.questions,comp.os.misc From: sable@halcyon.com (Scott Philip Sable) Subject: Unix OS for 286 Date: Thu, 15 Apr 1993 22:47:45 GMT I am in desperate need of any and all help!!! I am currently trying to find a Unix OS for an Intel 286 machine. I have tried a number of OS's, including Minix (Shows Booting Minix 1.5 prompt, then garbage), Linux (Seems to be CPU Specific. Does nothing), Xenix (Also seems to be CPU Specific. Runs to the Insert Floppy FileSystem prompt, then reboots the system.) I have also tried SCO Unix, which does the same thing as Linux. (Nothing!) Please, help me out. Any responses should be made to scott@vulpus.atl.ga.us. Much thanks in advance. Scott ------------------------------ From: heron@carson.u.washington.edu (Craig Horman) Subject: EISA video cards supported? Date: 16 Apr 1993 06:15:52 GMT I'm preparing to ditch a MIPS-based workstation and replace it with a 486DX50 w/EISA slots. Please forgive my ignorance, but I'm wondering if an EISA variant of any of the cards listed in the Linux X11 supported hardware list (the Orchid ProDesigner IIs, say) will work. I'm interested in driving a 1024x768x256 NI, or better, display. If the answer is affirmative, I'd love to hear people's impressions of the relative speed of ISA, EISA, and LB variants of similar boards. I have heard the arguments touting local bus over EISA, and I mostly agree with them. Due to circumstances, I will most likely end up with an EISA machine anyway. Thank you. ------------------------------ From: ijackson@nyx.cs.du.edu (Ian Jackson) Subject: *** READ THIS BEFORE POSTING *** Date: Fri, 16 Apr 1993 06:23:01 GMT To get people to read your post and answer your question: - Read "Linux Documents Explained for Newbies" - posted weekly (see below). - Please do not post questions answered in the FAQ lists. - Please do not post "Please send me the FAQ". See below. - Please do not post "Split this group now". See below. - Please read the documentation for the program you have a problem with. - Be informative, both in subject line and body. See below. - Linux-specific posts only please - Unix, X, C, etc. groups listed below. For more information read on ... *** Linux Documents Explained for Newbies (like me 8-) *** is is posted weekly by Jay MacDonald. If you can't find it email me or Jay . It will tell you about the Linux FAQ, the INFO-SHEET, the META-FAQ, etc. These are posted regularly to comp.os.linux and c.o.l.announce, and can be found in /pub/usenet/news.answers/linux-faq at pit-manager.mit.edu and on the Linux archive sites. If you can't get them there, try mailing "mailserver@nic.funet.fi", "mail-server@pit-manager.mit.edu" or "ftpmail@decwrl.dec.com". If you can't get them any other way, e-mail me. Splitting the group (due to volume or other reasons): It's been tried, and all the new groups except .announce failed. It can't be tried again within 6 months. See news.announce.newgroups. If you insist on discussing it, do it in news.groups (if c.o.l has too much volume already, why make it worse?). With threaded newsreaders now in widespread use using a good, specific, polite subject line will get you many more readers. Most people will ignore posts with subjects like "Linux", "Help", "SLS", etc. Don't shout to get attention. Put as much information in your post as possible - preferably a brief summary followed by session transcripts, etc. as appropriate. Include the precise text of any error messages printed, etc., and appropriate info about your setup. Linux is a Unix clone running on PCs; X has its own groups. One of the following groups is is quite likely to be more appropriate for your question: Unix: comp.unix.questions, .admin, .programmer, .shell etc. X: comp.windows.x, comp.windows.x.i386unix (for XFree86 etc) GNU: gnu.emacs.help, comp.emacs, gnu.gcc.help, gnu.* generally PCs: comp.sys.ibm.pc.hardware C: comp.lang.c -- Ian Jackson ijackson@nyx.cs.du.edu PGP2 public key available on request Home: 35 Molewood Close, Cambridge, UK, CB4 3SR; +44 223 327029. These opin- Work and urgent email: iwj@cam-orl.co.uk; +44 223 343398 ions are my own. Olivetti Research Ltd, Old Addenbrookes Site, Trumpington St, Cambridge, UK; ------------------------------ From: p7003ag@sun1.LRZ-Muenchen.DE () Subject: Linux&Compaq Deskpro same old prob : ( Date: Fri, 16 Apr 1993 06:19:42 GMT Hiyo, I still got a prob with my Deskpro 386/20: I can't install Linux. After checking all the hardware I have (2 serial, 1 parallel, AHA1542B, Conner 200MB SCSI) the kernel tries to reset my harddisk controller and fails. This it does in an indefinite loop. Any controller I care to try (AT-Bus, MFM ...) produces the same effect. On a NONAME 486 with AT-Bus HD it works just fine. Has anyone encountered similar problems or is anybody running Linux on a Deskpro 386/20 already ? Please help me. CU Alaric ------------------------------ From: ent811l@monu6.cc.monash.edu.au (Christopher Kuperman) Subject: Re: can 386sx run linux Date: Fri, 16 Apr 1993 05:35:26 GMT abc@bass.bu.edu (Onnei Cheng) writes: : hi, guys : : My friend has a 386SX, is it good to run Linux : : thanks in advance : : Onnei : if your friend has 4MB ram the installation willl work, but it is a little complicated... if your friend has an sx with 8MB than it should work just fine (amaybe a little slow) but to answer your querstion: any 386/486 sx or not with at least 4MB ram will work.. Ckup. _____________________________________________________________________________ [__From_________________________________][ aka: Christopher Kuperman ] [_______________________________________][ The holistic systems consultant ] [____@@@@@@____________________@@_______][------------------------------------] [_______@@____@@@@@___@@_@@@___@@__@@___][ email: zork@yoyo.cc.monash.edu.au ] [______@@____@@___@@__@@@___@__@@@@_____][------------------------------------] [_____@@_____@@___@@__@@_______@@_@@____][ Giv a man a fish & he'll eat for a ] [____@@@@@@@__@@@@@___@@_______@@__@@___][ day, teach him how to fish & he'll ] [_______________________________________][_eat for ever.______________________] ------------------------------ From: cvc@world.std.com (Cliff V Collins) Subject: TIMEOUT ON CDROM INSTALL Date: 16 Apr 93 07:16:47 GMT My system is 486DX266 256K ISA 8 megs ram adaptec 1542B maxtor 7213S SCSI1 200 Meg hd texel 2024 SCSI2 300kb/s cdrom when I try to install the binary only dist of yydrasil beta cdrom the process says SCSI Timeout host 1 operation aborted after copying 20-30 of the bin files. The demo option runs fine. I have been running the softlanding SLS 99p6 kernal for 6-7 weeks with no problem (however while I have used the cdrom under it, I have never realy copied alot of files from cdrom to my drive. My partitions are 100megs dos and 100 megs linux efs. This same type of problem happened months ago when I tried to install a 386bsd system from cdrom. My dos system never seems to crap out on multimedia stuff ( mixed up mother goose and such). All my adaptec jumpers are default. My maxtor is id 1 , cdrom is 3. I have slowed the system down (turbo off) shut the caches off but no go. help, anyone? ------------------------------ From: ggeorge@bu.edu (Gerry George) Subject: Problems configuring news, etc Date: 16 Apr 93 07:04:58 GMT News problems: News programs (/usr/local/lib/news...) is setup as suid. Tin is not. When I run tin as a regular user, posting fails due to permissions. (why do all my problems seem related to permissions - ingres, news, mail, etc?) root is able to post (sometimes) but no-one else can. Scratch that now - noboby can post anymore - (3 days later) Why doesn't tin call config (/usr/local/lib/news/config) with effective uid of 'news'? Should tin be su/gid news? After fiddling with config, I made it a+r so the config problems are now gone. After much hacking around, posts are now accepted from ordinary users, but they are dumped. I still get the error: relaynews: set[ug]id failed (invalid argument) then it dumps the article in dead.article. Lots of the other utilities (news related) do not run properly - some problem or the other, which may all be linked, or not. If those are just regular problems that will be cleared up with "proper" setup (and much reading) then don't bother, but if it is unusual stuff, please reply (post if possible?). Thanks. Note: I do not yet have a feed, but I'm trying to get everything working before I activate my link to avoid trashing things for others when I do connect in the next week. I have SLS 99pl6 and the latest of smail/news/elm/tin/nn, etc (April 10/11) Thanks for any assistance. I might also need help in configuring my UUCP files for my feed. Can someone mail me an example - if it is complicated, or if it is a simple solution (i managed to work my way through TCP/IP and most of the news stuff), then just a simple RTFF (FAQ) will suffice. In case any instructions are forthcoming, it would help to explain what the different options do since I wish to understand (and maybe help someone else?). I hope I'm not asking too much. Thanks again. =========================================================================== Gerry George | Anything good in life is either School of Management, Boston Univ. | illegal, immoral or fattening. Internet: ggeorge@acs.bu.edu | Any item not in the above three Compu$erve: 72607.2560@compuserve.com | categories causes cancer in rats! =========================================================================== ------------------------------ From: ramesh@utdallas.edu (R. Ramesh) Subject: [Q] Linking thru serial lines. Date: Fri, 16 Apr 1993 00:11:34 GMT Hi: I am trying to set a lab full of PC's running Linux. I neither have ethernet card nor a tape drive. So have to configure one Linux box at a time. To make things easy, I had the following plan. First I will setup one PC and then setup others thru serial line download of the files. The only glitch is that I don't know how to set up the cabling between the serial lines for a successful download. I tried the following without success. 1) Connect the com2's together of the two PC's in question with a female-female null modem cable. 2) setup both serial ports (/dev/cua2) for 38400baud, 8bits, crtscts, noparity -ixon clocal. 3) from the source machine (the one which has a complete setup) I did a tar cvf /dev/cua2 and at the sink I did a tar xvf /dev/cua2. Well, tar on the source side starts firing but the destination side gets nothing. What am I doing wrong? BTW, here is my configuration (both m/c are same) Gateway 486/33 (Vesa local bus) 16450 based serial ports (total of 2 serial and 1 parallel) Linux 0.99 pl 8 Actix GEPlus (801 with 2M Dram) video card General setup with the latest of the binaries (as far as I know). Any help would be appreciated. Ramesh ------------------------------ From: torvalds@klaava.Helsinki.FI (Linus Torvalds) Subject: Re: crash: 0.99p7A + emacs -> swapper killed! Date: Fri, 16 Apr 1993 07:19:53 GMT In article root@fab4box.wa.com (Art Taylor) writes: >In article <1993Apr14.173533.410@desaster.hanse.de> michaelw@desaster.hanse.de (Michael Will) writes: >>Hello, >> >>I was just downloading my news with uucp, replied to a message with elm+emacs, >>when suddenly my machine got killed in an unusual way. >> >>general protection: 6e70 >>EIP: 0008:00D152 >>EFLAGS: 00010002 int = 58 9a 37 2a a8 c6 0e 3e 3a c9 cf 40 46 08 b9 39 >>fs: 0017 >>base: C000000 limit: 00200000 >>Pid: 0, process No. 0 >>cf: 00 00 00 00 00 00 00 00 >>task[0] (swapper) killed: unable to recover >>kernel panic: Trying to free up swapper memory space >>in swapper task - not syncing >> >>The machine stopped - I had to reboot cold. > >Ok, I got the same response when dl'ing with UUCP(taylor 1.04), and scrolling >up through a command buffer(up arrow, once in bash, once in ATP(.QWK packet >offline reader). It's a known bug (and a very silly one) in the 0.99.7A keyboard driver routines. It happens mostly when the keyboard does auto-repeat on keys that send multi-byte sequences (ie cursor keys and other special keys), but depends a bit on the hardware as well (ie it won't happen on some hardware, while being almost common on other). >It would seem that UUCP is the common link here. I've only experienced this >while running Taylor 1.04, and scrolling up. It's happened twice, and it >shocked hell out of me both times. It's the up-scrolling, and possibly the race-condition is made worse by serial interrupts going on etc, which is why you see it during UUCP. Others see it only under X11 (probably due to similar reasons), others won't see it at all... The fix is to get pl8 (together with the vm86 diff if you use NFS or dosemu), Linus ------------------------------ From: Juha.Ursin@hut.fi (Juha Ursin) Subject: Serial communications (Re: Two questions) Date: 16 Apr 93 08:15:57 GMT In article <1993Apr15.203634.13720@sfu.ca> tholbroo@malibu.sfu.ca (Todd Holbrook) writes: >Second question: When trying to connect my XT to my linux box through >a null modem, I've managed to get getty to send the issue file and the >login prompt, but as soon as I type in a log in name, it hangs. Using >ps, I can see that getty has called login, but I don't get the >password prompt, and after 5 minutes it drops back to getty and the >issue file and login prompt. The (hopefully) relevant line in my >inittab is: >c5:5:respawn getty 9600 ttyS1 =================================================================== From tytso@Athena.MIT.EDU Thu Apr 15 06:23:44 1993 To: jus@vinkku.hut.fi > >This is a list of the serial problems I have had on my linux system. >Each problem is described shortly and after that is explained how I solved >it. > >----- >Problem: >Trying to allow login on my terminal (on ttyS1 with only three wires >(RX, TX, GND) connected). Getty 9600 /dev/ttyS1 starts and I get the >login prompt, but the password prompt newer appears. > >Solved by: >Changed the /etc/gettydefs line >9600# B9600 CS8 # B9600 SANE -ISTRIP #S login: #9600 >to >9600# B9600 CS8 # B9600 SANE -ISTRIP CLOCAL #S login: #9600 > >Also can be solved by: >Fake the serial singnals. Shortcut (pins 4 and 5), (pins 6, 8 and 20) on D25 >connector plugged to the I/O card. > >Question: >What would be the proper way to solve this? Bug in kernel? Setting CLOCAL is the correct way to solve this --- it is not a bug in the kernel. Since you only have three wires connected, it means that the carrier detect line was left floating, and probably floated to ground. Thus, the /dev/ttyS1 device blocked waiting for carrier detect to go high. The reason why the login prompt appeared is that getty opened /dev/ttyS1 with the NODELAY flag; but when /dev/ttyS1 was reopened for /bin/login, it did not have the NODELAY flag, and you ended up blocking. That's what the CLOCAL flag is for --- it indicates that the serial port is hooked up to a local (hence CLOCAL) terminal, and so the carrier detect line should be ignored. In general, whenever the carrier detect line is not supported by a serial device, or if your RS-232 cable does not support the CD wire, the CLOCAL flag should be set. >----- >Problem: >Where are my Call-out-mode -serial ports (cuaX) > >Solved by: > login as root > cd /dev > mknod cua0 c 5 64 > mknod cua1 c 5 65 > mknod cua2 c 5 66 > mknod cua3 c 5 67 > chmod ugo+rw cua? > ln -sf cua3 modem > >Questions: >When will these be created in install procedure? Ask the person who created the install procedure (Peter MacDonald, in the case of SLS). >When should I use these cuaX ports? When you're dialing out (i.e., kermit, cu, seyon, uucp, etc.). >On my configuration ttyS3 is named ttys3 incorrectly - why? Again, see the person who created the install procedure. Recent versions of SLS create the serial devices using the new convention (ttyS3). ttys3 represents an older convention, and some older install procedures may not have switched over. >------ >Problem: >I got the remote logins on terminal working, but when I start X, the terminal >session hangs. My mouse is on ttyS0, terminal on ttyS1 and modem on ttyS3. >TtyS0 uses IRQ 4, ttyS1 and ttyS3 use IRQ 3. The cards are made in Taiwan. > >Solved by: >IRQ's can't be shared by serial ports. Must configure them differently. >Here's a list of IRQ's and their use. > >No. MSDOS name /dev/... Pin >----------------------------------------------- >IRQ 3 Com2 / Com4 ttyS1 / ttyS3 B25 >IRQ 4 Com1 / Com3 ttyS0 / ttyS2 B24 >IRQ 5 LPT2 lp2 B23 >IRQ 6 Floppy drive B22 >IRQ 7 LPT1 lp1 B21 >IRQ 9 VGA? B4 > >If there the system hasn't a secondary parallel port, IRQ 5 is usually free. > >The simplest I/O cards don't let configure themselves too much. Often you >can only select IRQ 4 or 3 to be used. This prevents from getting more than >2 serial ports in use. > >To make a simple multi-I/O card co-existing with another one, you must >change the IRQ wirings on the card. A simple 8-bit I/O card has 31 connecting >pins on the both sides. Pin 1 is near the plate with connectors and pin 31 is >in the other end. Side A is the component side and side B is the other one. > >To make the card use IRQ 5 instead of IRQ 4 you cut the folio connected to >pin B24 and reconnect it to pin B23 (small soldering needed). > >Questions: >What other possibly free IRQ's there are than IRQ 5? IRQ 2 is also often free, although some networking cares use them. Can I use IRQ 7 when not printing? It's generally a bad idea to have two cards plugged in sharing an IRQ address. It happens to work with most serial cards, but it doesn't alwas work. You can use IRQ 7 if you don't have a parallel port plugged into your system. If you do have a parallel port plugged in, you can try using IRQ 7; it might work, it might not. Be sure to also make sure that the printer works after you plug in your serial card configured to use IRQ 7. Another alternative: If you are using the version of the parallel printer driver that doesn't use interrupts, you can go to your parallel port board, and disable the IRQ, thus freeing IRQ 7. Is pin 1 nearest to the plate or is it 31? If this is a sample PC ISA card, then A1 is the pin closest the "back" of the card, and on the top, if the pins are held down and the "back" of the card is to your right: ---------------------------------------------------------------------- | | | | | | | | | | | | +---------------------------------+ +---+ +---+ | | | | +----------+ +---------------+ A A 3 1 1 The "B1..B31" pins are on the reverse side of the card. - Ted -- . . Juha.Ursin@hut.fi p. 7733119 . . ------------------------------ From: torvalds@klaava.Helsinki.FI (Linus Torvalds) Subject: Re: Big swap partitions [> 16M] Date: Fri, 16 Apr 1993 07:26:47 GMT In article <1993Apr16.025655.10225@news.acns.nwu.edu> hpa@nwu.edu (H. Peter Anvin) writes: > >How about getting larger than 16 Mb partitions? Is that also just to >change one (or more) #define's? I can't think of anyone who would >need more than, say, 16 swap areas, but larger than 16 Mb seems to be >a common desire. How about changing 127x16 Mb to, say, 15x128 Mb? The problem isn't the swap-page encoding (the encoding will allow 127x128MB happily), but due to my lazyness. A swap area needs a "swap page counter array", where each page (4kB) uses one byte as a counter that tells how many different processes use the page. And due to the normal linux memory management routines, it's hard to allocate an array bigger than one page - thus you get a maximum swap area of 4096 pages which is 16MB.. The coding necessary to use several pages for swap counters isn't overwhelming: in fact I have already gotten one patch for it (sorry to have forgotten the name of the person who sent it to me), and the only reason I didn't use it was that I'm a bit nervous about the memory management, so I either want to do it myself or be *very* sure there are no problems with the patch.. So the current 16MB limit isn't anything fundamental (unless you call my lazyness that), and one day you'll wake up and notice that linux automatically takes advantage of more than 16MB of swap. The only problem now is "when?", and I can't say. Linus ------------------------------ From: qumqats@citrus.SAC.CA.US ( Joel M. Baldwin ) Subject: Re: Xfree 1.2 leaves screen in graph-mode after exit, runx !work Date: Thu, 15 Apr 1993 06:08:25 GMT In article <1993Apr11.054139.1007@amscons.com> bob@amscons.com (Bob Amstadt) writes: >ado@bigcomm.gun.de (Christoph Adomeit) writes: >: the newest version of xfree leaves my et4000 card in graphics mode after >: exit. I also can't change VTY's, because they only contain of garbaged data ...... >Add MGA cards to those affected by this problem. Here's a patch to fix that. > From qumqats Sat Mar 6 08:37:47 1993 > Return-Path: > Received: by citrus.SAC.CA.US (Smail3.1.28.1 #6) > id m0nUuOB-00090AC; Sat, 6 Mar 93 08:37 GMT > Sender: qumqats ( Joel M. Baldwin ) > To: root > Newsgroups: comp.os.linux,comp.windows.x.i386unix > Path: citrus!csusac!charnel!rat!usc!rpi!batcomputer!munnari.oz.au!bruce.cs.monash.edu.au!monu6!vaxc.cc.monash.edu.au!apm233m > From: apm233m@vaxc.cc.monash.edu.au > Subject: A fix for Linux XFree86 hga problems > Message-ID: <1993Mar3.001340.1@vaxc.cc.monash.edu.au> > Lines: 85 > Sender: news@monu6.cc.monash.edu.au (Usenet system) > Organization: Computer Centre, Monash University, Australia > Date: Tue, 2 Mar 1993 13:13:40 GMT > Status: RO Hi, Here is a small patch to the kernel to fix the problems which Hercules card owners have with XFree86 1.2. With the patch installed, vt switching works as advertised and the console is returned to text mode upon exit from XFree86. To install the patch, put it in a file (eg "patches"), cd to where you have put the Linux source (usually /usr/src) and do "patch < patches". Re-compile the kernel, install the Image, and re-boot. Remember that XFree grabs the first vt which is not in use for itself, therefore to switch back to XFree from another vt you need to switch to that vt, not the one where you started XFree. On many systems this will probably be vt #5, in which case Alt-F5 gets you back to XFree. On my system, I have logins on 6 vts so I have to do Alt-F7. I have mailed this patch to Linus, so if he approves it may appear in the next release of the kernel. Good luck! --Bill apm233m@vaxc.cc.monash.edu.au =================== >8 Patches start here 8< =============================== *** linux/kernel/chr_drv/console.c.old Sun Feb 28 18:13:08 1993 --- linux/kernel/chr_drv/console.c Tue Mar 2 20:39:32 1993 *************** *** 1356,1361 **** --- 1380,1422 ---- static void set_scrmem(int currcons) { + #define HERCULES + #ifdef HERCULES + /* This works with XFree86 1.2. + It is not clear that this is the best place for the code. + It could be extended and made more generally useful if we could + determine the actual video mode. It appears that this should be + possible on a genuine Hercules card, but I haven't been able to + read from any of the required registers on my clone card. + This code will not be of much benefit with existing non-X Hercules + graphics programs, which are probably not as smart as XFree86 1.2 + about vt changes. However, new programs which follow the techniques + of XFree86 would allow the use of vt changes. + */ + /* This code should work with Hercules and MDA cards. */ + if (video_type == VIDEO_TYPE_MDA) + { + if (vcmode == KD_TEXT) + { + /* Ensure that the card is in text mode. */ + int i; + static char herc_txt_tbl[12] = { + 0x61,0x50,0x52,0x0f,0x19,6,0x19,0x19,2,0x0d,0x0b,0x0c }; + outb_p(0, 0x3bf); /* Back to power-on defaults */ + outb_p(0, 0x3b8); /* Blank the screen, select page 0, etc */ + for ( i = 0 ; i < 12 ; i++ ) + { + outb_p(i, 0x3b4); + outb_p(herc_txt_tbl[i], 0x3b5); + } + } + #define BLINKER_ON 0x20 + #define SCREEN_ON 8 + /* Make sure that the hardware is not blanked */ + outb_p(BLINKER_ON | SCREEN_ON, 0x3b8); + } + #endif HERCULES + video_mem_start = video_mem_base; video_mem_end = video_mem_term; origin = video_mem_start; =================== >8 Patches end here 8< =============================== -- =============== Bill Metzenthen Mathematics Department Monash University Australia -- Joel M. Baldwin System Administrator, Citrus Grove Public Access 916-381-5822 N6NPL H ( 916 ) 383-9470 W ( 916 ) 452-1481 uunet!csusac.ecs.csus.edu!citrus!qumqats Sacramento CA. USA. InterNet: qumqats@citrus.SAC.CA.US FidoNet: 1:203/7707.42 /\../\ ------------------------------ From: thoeny@ito.umnw.ethz.ch (Juerg Thoeny) Subject: problems with tar: broken pipe, Problems with lpd Date: Fri, 16 Apr 1993 07:28:19 GMT Hi there, i have problems with tar. Every time I use tar on big files or to create big files I get the message: tar: some error messages broken pipe Any clue? I have also an other problem. The lpd doesn't accept bigger streams than 1MB. Since I'm using a postscript printer this problem, I encounter this problem every day. I use a remote printer, which is spooled by a SUN-Server. Thanks in advance Juerg ------------------------------ ** 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-Activists-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux) via: Internet: Linux-Activists@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 tupac-amaru.informatik.rwth-aachen.de pub/msdos/replace The current version of Linux is 0.99pl6 released on February 23, 1993 End of Linux-Activists Digest ****************************** e you started XFree. On many systems this will probably be vt #5, in which case Alt-F5 gets you back to XFree. On my system, I have logins on 6 vts so I have to do Alt-F7. I have mailed this patch to Linus, so if he approves it may appear in the next release of the kernel. Good luck! --Bill apdigest858 644 36676 74430 52104 5366137104 5761 From: Digestifier To: Linux-Activists@senator-bedfellow.mit.edu Reply-To: Linux-Activists@senator-bedfellow.mit.edu Date: Fri, 16 Apr 93 07:15:07 EDT Subject: Linux-Activists Digest #858 Linux-Activists Digest #858, Volume #4 Fri, 16 Apr 93 07:15:07 EDT Contents: Linux Journal -- magazine (Linux Journal) Problem installing SLS with Imprimis hard drives (Emil Rainero) mke2fs problem: unable to find a block for the inode table (Michael Elkins) Re: Linux on a diskless workstation? (Leigh M Hart) Re: Does anyone ? (KID01) Re: Mounting an extended dos partition (Hal N. Brooks) Re: New feature for the filesystems. What do you think ? (Keith Smith) Need Help: rw_swap_page: bad swap file (keith eric hellman) respawn prob. serial connection Re: Help on X-Windows(mostly) (llew weaver) (X386) light C shared image??? (Cameron L. Spitzer) Re: Why Linux should remain Linux and seagulls should not be used. (Matt Welsh) ---------------------------------------------------------------------------- From: linux@fylz.com (Linux Journal) Subject: Linux Journal -- magazine Date: Fri, 16 Apr 1993 02:07:10 GMT A few months ago a group of people in the U.S. (some in Seattle but stretching all the way to the east coast) started talking about an idea for a magazine. The initial magazine idea is still undergoing some changes but, in the process, the idea for the Linux Journal was born. Why does the Linux community need a magazine? The publisher of a technical magazine I was talking to put it this way: "No successful movement has ever made it without a journal reporting on its progress." Well, Linux Journal wants to be that journal for the Linux movement. There is a questionnaire at the end of this post. We would like to encourage your input. If this magazine is going to happen it needs to be what people want. "The Net" is one audience we need to hear from. The other, harder to find, audience are all the people that don't know about or have access to the net. If you know people who fit in this category, please either give them a copy of the questionnaire to mail or FAX back or send us their address and we will mail them a flyer. Some details of our plans: * We will be a monthly publication focusing on the world of Linux * We will minimize lead time so the information will be current * We will cover both "free" and commercial software * We will report on the Linux project as a whole * We want to help newcomers as well as those with *ix experience * Our first issue will be out by September 1 The answers we get from this questionnaire will help us complete our advertising policy and editorial policy. Thanks for your time -- Phil, Laurie, Mike, ... ========================================================================= Questionnaire (please return) ========================================================================= Help Direct Linux Journal This flier attempts to tell you what we think should be in Linux Journal. But, what matters is what you want to see in Linux Journal. Please answer the questions below and return this form to help make this effort succeed. 1. Rate your interest in the following features: (1=not interested, 2=possible interest, 3=interested, 4=very interested, 5=that's my subject!) ___ a. Club and organization contacts ___ b. Non-commercial classified ads ___ c. Latest Linux release information ___ d. Letters to the editor ___ e. Software wanted column ___ f. Linux Questions and Answers ___ g. New Linux-related products ___ h. New to the club -- a focus on schools, etc that are getting into Linux ___ i. Linux Profile -- a look at a Linux developer or user ___ j. Tutorial on a particular aspect of Linux (or Unix) ___ k. Current trends -- what's happening with Linux (becoming commercial, support by vendors, etc) ___ l. Introduction to Linux -- basic background information on Linux for first-time or prospective users 2. Are there any other topics you would like to see covered? ______________________________________________________________________ ______________________________________________________________________ ______________________________________________________________________ ______________________________________________________________________ 3. We intend to sell commercial advertising but limit it to vendors whose products and services are in tune with the Statement of Purpose of Linux Journal. We expect advertising from hardware manufacturers, publishers of books and other publications and software vendors of both "free" and commercial products. Comments: ____________________________________________________________ ______________________________________________________________________ ______________________________________________________________________ ______________________________________________________________________ 4. Would you be interested in: [ ] subscribing [ ] advertising [ ] writing articles Name __________________________________________________________ Address _______________________________________________________ _______________________________________________________ City ____________________________State ____ ZIP ____________ E-mail: _______________________________________________________ Currently running: [ ] Linux, [ ] MS-DOS, [ ] Unix, [ ] Other ______________ Have the following system(s): _________________________________ _______________________________________________________________ _______________________________________________________________ Please return this form to: Linux Journal P.O. Box 85867 Seattle, WA 98145-1867 Or, e-mail something like it to: linux@fylz.com ------------------------------ From: rainero@sente.wrc.xerox.com (Emil Rainero) Subject: Problem installing SLS with Imprimis hard drives Reply-To: rainero@wrc.xerox.com Date: Tue, 13 Apr 1993 15:45:22 GMT Gateway 2000/486 DX2 66 Imprimis 93601-15 (1040Mb) hard drives When booting SLS off of disk a1, it says detected 0 SCSI Disks. then everything proceeds nicely until a2 is inserted. At that point # fdisk /dev/hda not found Has anyone gotten Imprimis 1040Mb disks working with SLS? Thanks in advance Emil Rainero rainero@wrc.xerox.com ------------------------------ From: michael@jarthur.claremont.edu (Michael Elkins) Subject: mke2fs problem: unable to find a block for the inode table Date: Fri, 16 Apr 1993 00:51:03 GMT What exactly are the requirements for this to happen? I remember when I initially installed linux on my system that I had real problems getting mke2fs to like my partitioning scheme. Now that I've decided to ditch DOS altogether, I was trying to make another partition, but I can't use the partition that I made. Can anyone tell me what the correct "parameters" (so to speak) are for making an ext2 filesystem?? Thanks, me ------------------------------ From: hart@flash.pax.tpa.com.au (Leigh M Hart) Subject: Re: Linux on a diskless workstation? Date: 16 Apr 1993 17:57:20 +0930 edwanji@matt.ksu.ksu.edu (Edwanji Lemorian) writes: >This may be a dumb question: Not as dumb as my answer might be ;) >Is it possible in its current form or with few modifications to run >linux on a diskless workstation? I.e. a remote server feeds the >workstation everyting from the kernel to all mounted filesystems? >Thanks for any info. In a "DOS" based network, diskless workstations boot (either with special software or a firmware chip in the network card) from a server. The server downloads via ethernet a boot image file (previously setup on the server) which the PC then uses as a virtual drive A. The PC then boots from this disk image, which then sets up all the network o/s specific stuff in the config.sys and autoexec.bat that resides in the boot (disk) image. I don't think it would be terribly hard to write a low-level ethernet listening daemon that listens out for the remote-boot requests that the ethernet card sends out - from there it could send a Linux boot disk image to the PC, which would then proceed to boot from that image. The image you could set up on a floppy (and then image to a file) would have the kernel and some basic utils, and an /etc/rc file - in this you would need to NFS mount the rest of the filesystems that the workstation needed. Has anyone written something like this yet? I might be interested enough to get the specs on some of the more popular brands of ethernet cards and try to implement this - if they are standard enough. If any linux guru's would care to give a few pointers in regard to the writing of the actual "daemon" I would appreciate any input. I'm not sure what level the network card's remote-boot request is sent at - I am pretty sure it isn't netbios and I know it isn't tcp/ip :) does this mean it would need to be written in the kernel? or just a hardware specific daemon that waits for a specific request? To quote a silly robot - INPUT! Take care all Leigh -- Leigh M Hart (hart@pax.tpa.com.au) Ph: +61-8-267-5966 (W) ===================== The Fluffy Rabbit rides again ------------------------------ From: KID01 Subject: Re: Does anyone ? Date: 16 Apr 93 08:29:22 GMT In article <11061@lhdsy1.lahabra.chevron.com> jjctc@lhdsy1.lahabra.chevron.com (James C. Tsiao) writes: >In article <2051@dsbc.icl.co.uk> mat@dsbc.icl.co.uk (Matthew Roderick) writes: >> >>Does anyone have a SiS chipset mother board running linux ? >>With an AMI BIOS >> >>I have 486/33 and would like information on the setup for linux of the CMOS. >>I think my setup my be incorrect as Linux keeps panicing on boot ! > >I've never received panics. If you provide more details (Linux version, >attached hardware, etc.), then we can help you better. > I'm (trying) to run SLS pl6 and now have found the problem, the machine doesn't like the ide drive, I tried it with a friends scsi bits (disabling c: in BIOS) and all works fine ! Anyone come access this ?!! (I'll try moving the cards tonight !!) Matthew -- Matthew Roderick -*- B2 Developments -*- mat@{dsbc,oasis}.icl.co.uk +44 [0]782 771000 x3270 -*- UNIX Centre, ICL, Stoke-on-Trent, England. Spelling & typing errors have been left for orthentisity ------------------------------ From: hal@pollux (Hal N. Brooks) Subject: Re: Mounting an extended dos partition Date: Thu, 15 Apr 1993 23:16:07 GMT In article <1993Apr14.135119.4022@wisipc.weizmann.ac.il> dov@menora.weizmann.ac.il (Dov Grobgeld) writes: >My Dos disk is partitioned into two parts, a primary partion (C disk) >and an extended partition, with one logical partition (D disk). >There is no problem to mount the C disk with the command > > mount -t msdos /dev/hda1 /usr/dos > >but the similar command > > mount -t msdos /dev/hda2 /usr/ddisk As I understand it, the device name for the first logical partition would be /dev/hda5, hda1 through hda4 refer to primary partitions. See the Linux Device list. ================================================== Hal N. Brooks Internet: hal@pollux.cs.uga.edu ================================================== ------------------------------ From: keith@ksmith.com (Keith Smith) Subject: Re: New feature for the filesystems. What do you think ? Date: Fri, 16 Apr 93 03:33:56 GMT In article <0fmuPUm00VopMZhFU7@andrew.cmu.edu> Frank T Lofaro writes: >Excerpts from netnews.comp.os.linux: 13-Apr-93 Re: New feature for the >fil.. HJ Lu@eecs.wsu.edu (463) > >>I am wondering how a compressed filesystem handles demand paging and >>random update. > >I had just posted wondering about demand-paging, but came up with an >idea. Have it uncompress the file, and put it on some temporarily >allocated inodes, and exec/demand page it from there. It would require >some changing of filesystem allocation/exec/demand paging routines I >gather, but should be feasible. > Am I nuts here? Seems like you just need a layer of code between the virtual disk your filesystem see's, and the physical media. This code can maintain and map multiple compressed blocks (say 4K) into physical uncompressed block numbers used by the filesystem code. The only trick I see is having a filesystem that has a dynamic size. This should not be insurmountable either, as one could define the size of the virtual disk as arbitrarily 3X, and then use a non-removeable psuedo file entry to eat up any unavailable space, or reduce it's psuedo "size" from 2X the disk as successful compression kicks in. In this way you always see "real" space available, but space in use would be dynamic shifting from the pseudo file to real files and back as compression occurs or compressed space is free'd. For any kind of data integrity compression should be done on a block by block basis (4K at least), but perhaps with n smaller sub-blocks (256bytes?) being sqeezed into each block. That way a trashed block won't kill you for more than 16K of data (@4:1). For example the filesystem wants block #20, hands off to the virtual translator which looks in a table and finds virtual block 20 is offset 512 characters in physical block 3. It fetches it, uncompresses it, and delivers the appropriate chunk of data back. In the reverse, if the virtual block won't squeeze back in, it will have to drift onto a different physical block until it fits. Just like a variable length ISAM file record. Hell design the directory structure and file system as a B-tree variaton. Oooohhm, It's *LATE* yawn, and I'm rambling .... -- Keith Smith keith@ksmith.com 5719 Archer Rd. Digital Designs BBS 1-919-423-4216 Hope Mills, NC 28348-2201 Somewhere in the Styx of North Carolina ... ------------------------------ From: keh5@ellis.uchicago.edu (keith eric hellman) Subject: Need Help: rw_swap_page: bad swap file Reply-To: keh5@midway.uchicago.edu Date: Fri, 16 Apr 1993 05:44:36 GMT I am running Linux 0.99pl6-26; SLS distribution brought down from tsx-11.mit.edu on 4/11/93. The box: 486DX, 4Meg, 120MB SCSI w/ Adaptec 1542b. I have been running Linux since 0.97 with little (if any) serious problems. Now something has popped up that I just can't resolve. It occurs when I run X, but I don't think X is the problem per se: #1 olwm just doesn't behave correctly from the get go. #2 when I exit (harshly with CTRL-ALT-BS) I have a long list of: > olwm: Warning, X Error: BadGC (invalid GC parameters) > Request Major Code: 74 > Request Minor Code: 0 > ResourceID 0x800030. errors. #3 The machine is then in a pseudo-locked up state. I can get to another tty but cannon login. While I was trying to hash this problem out, I accidently switched to tty3... #4 The screen is _full_ of the single error message: > rw_swap_page: bad swap file I have done little to the SLS distribution startup files, only the following two lines: ># swapfile >mkswap /usr/local/paging/swapfile 5120 >swapon /usr/local/paging/swapfile > >/usr/bin/setterm -blank 10 <-- Beginning of SLS stuff. at the beggining of /etc/rc.local. I don't get any errors at startup and free after login reports: total used free shared buffers Mem: 2920 1672 1248 384 780 Swap: 5116 0 5116 I'm not sure where my mistake has been and would appreciate some quick advice if possible. I _really_ don't wanna dig through source code for the answer. Please reply by email; I've cluttered the c.o.l enough with this post I'm sure. keh5@midway.uchicago.edu ------------------------------ Date: Fri, 16 Apr 1993 03:51:43 EDT From: Subject: respawn prob. serial connection I'm trying to hook another machine to my linux box using the serial port I put the proper addition in the inittab file.. c6:6:respawn:/bin/uugetty 9600 cua1 I'be also tried ttyS1 and getty. I keep getting respawn too fast errors any help would be appreciated Mark ------------------------------ From: lmw@moose.uvm.edu (llew weaver) Subject: Re: Help on X-Windows(mostly) Date: Fri, 16 Apr 1993 08:47:56 GMT From article <28617@galaxy.ucr.edu>, by jmward@ucrengr (jonathan ward): > Hi there. I'm relatively new to Linux(awesome OS), and just got the SLS releaseoff of susnsite(0.99pl6). I'm tring to configure X to run on my machine, which is: generic 486DX-50 VLB motherboard, trident tvga8900c video card with 1meg ram, maxtor LXT340A IDE hard drive, a three button serial mouse. I finally got X > up and running(it defaults to the open windows manager) but I have this problem: > The mouse, once started is generally unresponsive. I can move it repeatedly, > and it occasionally "jumps" around the screen. When I depress a button, it > then takes a while to respond. I only have 4mb of RAM, and I'm swapping to a > 12 mb swapfile. Is the mouse problems due to the lack of ram, or something I > overlooked? Any suggestion appreciated. > > Also, I'm using the e2fs for my main partition. What is the proper method to run > e2fsck? > > Thanks, > Jonathan Ward > > Email to: > drdrums@watserv.ucr.edu > drdrums@csld.ucr.edu > jmward@cs.ucr.edu > -- ------------------------------ From: cls@truffula.sj.ca.us (Cameron L. Spitzer) Subject: (X386) light C shared image??? Date: Fri, 16 Apr 93 07:55:45 GMT Once again, I'm trying to install X386. I have HJ's root and base, and 0.99pl8. I obtained and untarred SLS disks x1-x10. I have an Xconfig I believe is right. My (Logitech) mouse passes HJ's mouse test. Startx produces the following error messages: Undefined C library functions: 1. light C shared image (Use the real one instead.) giving up. xinit: No such file or directory (errno 2): unable to connect to X Server xinit: No such process (errno3): Server error. My questions are: a. What's "the real one" and where might I obtain it? Is there a heavy C shared image? Does SLS use a different set of libraries than HJ supplies? b. What document should I have read that I would know what a "light C shared image" is? What newsgroup? c. How can I redirect the output of startx into a file? When I tried (startx 2>&1 ) > startx.errors & from HJ's ksh, it printed nothing and the file was empty. Any help much appreciated! Cameron in San Jose, still stuck with SCO 3.2.0 but not for long :-) cls@truffula.sj.ca.us ------------------------------ From: mdw@theory.TC.Cornell.EDU (Matt Welsh) Subject: Re: Why Linux should remain Linux and seagulls should not be used. Date: 16 Apr 1993 05:18:23 GMT In article , spj@ukelele.GCR.COM (Simon Patrick Janes) writes: > I don't like the seagull idea however. Seagulls are too common I think... > I don't know... it just makes me think what America would have been like > if the national bird was the turkey like it almost became. I would rather > have a platypus than a seagull. Folks, Michael K. Johnson and I have decided to use the Linux seagull logo on the title page of the Linux Documentation Project manuals as well as other works. We're working right now on integrating the logo with the current linuxdoc.sty LaTeX style format used by the LDP. While this by no means makes it "official", it's about as official as you can get in the Linux community. The seagull represents many things about Linux: its size (Linux is a small UNIX), freedom, internationalization (where in the world aren't seagulls found?), flexibility, and, well... flight (from the world of DOS). If any animal or design represents Linux it is the seagull. Although I'm probably not interested in receiving tons of email trying to persuade me out of this decision, if you design a good-looking Linux logo depicting some other animal (or no animal at all?) feel free to send it along. PostScript is fine, but MetaFont source would be better. The seagull has so far been the most appropriate available design, and we really liked it. A platypus? Okay, if the mammal body represents System V and the duckbill represents BSD, then you may have something there... :) Cheers, mdw -- Matt Welsh, mdw@tc.cornell.edu "I met a girl named Sandoz..." ------------------------------ ** 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-Activists-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux) via: Internet: Linux-Activists@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 tupac-amaru.informatik.rwth-aachen.de pub/msdos/replace The current version of Linux is 0.99pl6 released on February 23, 1993 End of Linux-Activists Digest ****************************** a > 12 mb swapfile. Is the mouse problems due to the lack of ram, or something I > overlooked? Any suggestion appreciated. > > Also, I'm using the e2fs for my main partition. What is the proper method to run > e2fsck? > > Thanks, > Jonathan Ward > > Email to: > drdrums@watserv.ucr.edu > drdrums@csld.ucr.edu > jmward@cs.ucr.edu > -- ------------------------------digest859 644 36676 74430 47354 5366137104 5775 From: Digestifier To: Linux-Activists@senator-bedfellow.mit.edu Reply-To: Linux-Activists@senator-bedfellow.mit.edu Date: Fri, 16 Apr 93 11:00:40 EDT Subject: Linux-Activists Digest #859 Linux-Activists Digest #859, Volume #4 Fri, 16 Apr 93 11:00:40 EDT Contents: Re: Kernel panic (Wolfgang Jung) Farenheit Discontinued - BUY NOW!!! (Mark Robinson) Re: Installation: SLS vs SCO (Colin Jenkins) Re: Why Linux should remain Linux and seagulls should not be used. (Mike Dowling) Re: Crashing machine >16M: Update (Hartmut Pohl) Re: What are the .z files? (Niclas B|rlin) Re: SLIP for linux? (Michael K. Johnson) Re: printing to/from remote machines (Gerald Hanusch) Re: Looking for MIT-scheme (Matteo Frigo) Problem with disk #2 (Denis Perraud) Re: Intel, the Pentium and Linux (Roland Turner) How to GCC part 2 (Martin Rehwald) Re: xvt... please (Robert Nation) X authentication problems & missing headers (Timo Korvola) Problems with 'ld' (Stephen Opal) Re: GIF Viewer for Linux? (Erik Troan) Re: Help with SLS install... ("Mario A. Guerra") Request: which shells (Wolfgang Jung) Where is ipc+ ? (Keith B. Kee {AISG}) C= 64 emulator project (Mast RN) ---------------------------------------------------------------------------- From: wong@cs.tu-berlin.de (Wolfgang Jung) Subject: Re: Kernel panic Date: 16 Apr 1993 09:58:31 GMT kowal@stsci.edu wrote: : Also, the instructions in the FAQ say to write-protect all installation disks, : but if I do that the installation program complains that the disk is write : protected! Since the isstallation script just reads from the disks, why not mount them directly read only ??? This shouldn't make any Problems ! Gruss Wolfganfg ------------------------------ From: mark-r@spec0.ee.man.ac.uk (Mark Robinson) Subject: Farenheit Discontinued - BUY NOW!!! Date: 16 Apr 93 10:46:36 GMT Hi When I went to but an Orchid Farenheit 1280 card to replace my Trident, the dealer told me that this card had been discontinued in favour of the VA version. He then went on to say that he still had a few left, but that they are as rare as hens teeth and I should buy one now, as they are faster than the VA. Since fast video is more important than the gimmicky VA, and I wasn't sure if XS3 could take advantage of the VA feature (or even drive the card) so I bought the 1280. OK, the questions are: 1. Was the dealer just trying to shift the last of his 1280s to some mug, or are they genuinely faster (it would seem logical since the 1280 has VRAM but the VA ordinary DRAM). 2. If it is faster, how much. 3. Are there plans to write a device driver for the VA features (it's not too late to change my card). TIA Mark -- ______________________________________________________________________ Multimedia Information Systems Lab | Ain't no LIFO high enough, Dept. Of Electrical Engineering | Ain't no FIFO deep enough, University of Manchester | Ain't no word size wide enough, Dover St. Manchester. M13 9PL | For my CPU. England | (orig) ------------------------------ From: jenkins@DPW.COM (Colin Jenkins) Subject: Re: Installation: SLS vs SCO Date: 3 Apr 93 07:06:21 GMT In article <1pfba3$2pj@nz12.rz.uni-karlsruhe.de> s_titz@ira.uka.de (Olaf Titz) writes: >Excuse me for being pedantic, but at least this is no valid argument. >First, a compiler is no security problem, since a compiler does not >introduce any new features which could be holes, it is just a >convenience tool. I disagree. A compiler represents a huge security hole. System attacks can be mounted at the lowest level using a compiler. If memory serves, the Internet worm of a few years ago propogated itself by recompiling itself and redistributing the source to other systems for recompilation. >Second, a dedicated hacker could use even use emacs to create a binary >executable. (Followup->alt.folklore.computers :-) Agreed, however few people have the time, patience, or knowledge required to generate any program of significant complexity by hand. The compiler would make attacks easier and faster and more available to hackers of lesser talents. Not that I am arguing against making compilers available, or that linux is inferior by virtue of poor security- simply that compilers do represent potential security problems. Colin ------------------------------ From: i1041301@ws.rz.tu-bs.de (Mike Dowling) Subject: Re: Why Linux should remain Linux and seagulls should not be used. Reply-To: i1041301@ws.rz.tu-bs.de Date: Fri, 16 Apr 1993 12:19:51 GMT In article <11059@lhdsy1.lahabra.chevron.com> jjctc@lhdsy1.lahabra.chevron.com (James C. Tsiao) writes: >> How about a Tiger? A Tiger is powerful, fast, sleek, efficient, and so is >> Linux. >> And also, Linus is a roarin' good programmer:) > >Nice, but maybe too aggressive. Has anyone suggested a Dolphin (dolFINN :-). Umm, I hate to see someone, especially some rabid 386bsd fan :-), changing that into DULL-FINN. Perhaps a porpoise... Sorry, what about poor-puss? Mike Dowling ------------------------------ From: pohl@egk.uni-koblenz.de (Hartmut Pohl) Subject: Re: Crashing machine >16M: Update Date: 16 Apr 1993 09:51:18 GMT In article <1qfvj7$bd9@fitz.TC.Cornell.EDU> kutcha@eos.acm.rpi.edu (Phillip Rzewski) writes: > [stuff deleted...] >memory tend to run really slow. This explains a lot to me. I have just a 386/25 >with a 64k cache, [...stuff deleted] Hello, Do you have an idea how much memory your Cache supports? This depends on the size of Tag-RAM, the size of your Cache-line and the size of your Cache. It is possible, that your Cache supports only 16MB of RAM, especially if you have an older board. Recently, I had the same problem, too. Solutions: Upgrade Cache-RAM if possible or buy new board. I don't know if this *is* the problem, but it might be. Hope this helps, Hartmut ------------------------------ From: niclas@cs.umu.se (Niclas B|rlin) Subject: Re: What are the .z files? Date: 16 Apr 93 07:01:45 GMT In article <1993Apr13.142458.15411@waikato.ac.nz> hamish@waikato.ac.nz writes: > >Hey. What format are the .z files on nic.funet.fi, and tsx-11? They aren't >compressed, and I can't find anything that will undo them. Any hints? > >-- >Trailer courtesy | Hamish Marson, Computer Services, University of Waikato > BOFH.. | hamish@waikato.ac.nz. Fax +64 7 8384066 > | Disclaimer: Remember, you heard it here first! > >Jesus was killed by a Moral Majority. Those should be gzipped (GNU zip) files. -- Niclas Borlin email: niclas@cs.umu.se Department of Computing Science phone: (+46) 90-166832 University of Umea fax: (+46) 90-166126 S-901 87 Umea, SWEDEN timezone: MET DST (UTC + 2h) ------------------------------ From: johnsonm@stolaf.edu (Michael K. Johnson) Subject: Re: SLIP for linux? Date: Fri, 16 Apr 1993 04:38:53 GMT In article <1993Apr15.120624.7276@walter.cray.com> jwl@ferrari.cray.com (Jim Lynch) writes: Is anyone aware of a product for Linux that works similar to SLIP, but doesn't require SLIP? SLIP had some security holes that are incompatible with our corporate police. I am aware of term, but I believe that only works with X and I can`t run X yet. Is there anything else that will let me have multiple connections Bzzt... Term doesn't require X, although it does *allow* you to make X connections if you have X. It works much better than UW (I've tried UW, too...) and is small and fast. Try it, you'll like it... If all you want is a one-user connection, it has all the features of slip that you need, generally. michaelkjohnson ------------------------------ From: root@alijku65 (Gerald Hanusch) Subject: Re: printing to/from remote machines Date: Fri, 16 Apr 1993 14:23:25 CDT Bill E. Eickmeier (bee@watarts.uwaterloo.ca) wrote: : [38]% ls -l /etc/hosts.lpd : -rw-r--r-- 1 root root 116 Apr 12 11:02 /etc/hosts.lpd : ----- : [39]% cat /etc/hosts.lpd : otherbox.uwaterloo.ca : ----- : [40]% grep 35/tcp /etc/services : lp 35/tcp # any private printer server [JBP] : printer 35/tcp # any private printer server [JBP] BSD LPR Service uses 515/tcp instead od 35/tcp, which is used by SysV lp Service. : Here's how things look when I try to print from a remote machine to my : linux printer (this originally worked fine, but has since stopped): : Spool queue: lp_linuxbox : Local (waiting to be sent to remote site): : otherbox: waiting for linuxbox to come up since Wed Apr 14 15:18:55 1993 : (errno=60 'Connection timed out') : 1st bee 10 temp 3030 bytes : Remote(linuxbox): : ...right here I hit ctrl-c since it seems to hang... This error seems to be a Kernel-TCP Bug, a guy an myself had exactly the same effects, when we upgraded from 99pl4 to 99pl7 or 99pl8. I have traced the Ethernet traffic with the PD "Gobbler" Packet Analyzer and found, that the Linux box sends a few tcp packets to the remote printer server and the gets a major timeout problem. If the printer server acknowledges to slow to the linux tcp packets, linux tcp seems not to handle the new SYN packets correctly. When I use the old 99pl4, and trace it, the sequence is correct, and if SYN is needed due to slow printer servers, its recovered ok. On 99pl8 TCP the trace looks like this : 5 - 6 packets sent, Linux TCP doesn't wait for the ACK and opens a new source port without closing the old one. Perhaps Linus can help ? : Any and all ideas are welcome... me too , :-), Temp. solution: use 99pl4 kernel if you want LPR package to print on (slow) remote printers. Kind regards, Gerald (Posted from a Linux Box with 99pl4 using TIN) ------------------------------ Crossposted-To: comp.lang.scheme,comp.lang.scheme.c From: athena@artemide.dei.unipd.it (Matteo Frigo) Subject: Re: Looking for MIT-scheme Date: Fri, 16 Apr 1993 08:25:30 GMT In article <1993Apr15.084031.255@janosch.toppoint.de> news@janosch.toppoint.de (janosch-usenet) writes: >I'm looking for MIT-Scheme running with Linux(!). I would appreciate if s.o. could tell me (ftp-adress is fine) >where I could such a Linux-Implementation. >Thank you in advance, VK > I did a port of MIT-Scheme 7.2.0 (alpha) for linux; it seems to run quite well; edwin and native 386 compiler also run. I did not released the port because it's of an alpha version -- I was waiting for an official release before making the package available. Anyway, if you are really interested in the program (compiler and edwin require at least 10Mb of memory, so on my 8Mb machine they swap almost with continuity), I could upload the program somewhere. In the meantime, you can retrieve my linux port of Scheme-to-C by anonymous ftp on artemide.dei.unipd.it, in the directory /pub/linux . Please let me know if anybody else is interested on a release of the port of C-scheme 7.2.0 (alpha) for linux. Matteo Frigo Matteo.Frigo@dei.unipd.it ------------------------------ From: ura934@cismibm.univ-lyon1.fr (Denis Perraud) Subject: Problem with disk #2 Date: 16 Apr 1993 12:50:57 GMT I use Linux 0.99.6 on my 386 DX 40 Mhz with 4 Meg . I've got a 80 M hard disk on /dev/hda. Yesterday, I bought a 245 M hard disk and configured the setup. DOS is OK but when I boot linux, it tells me : HD time out HD controler reset I can use /dev/hdb , copy files on it , etc ... but sometimes,this error happens. Both disks are Western Digital with an IDE controler card. What does this error means ? Kernel problem ? IDE problem ? HD problem ? (I think my setup is OK) Thanks. =========================== URA934 Denis Perraud e-mail : ura934@cismibm.univ-lyon1.fr dperraud@ens.ens-lyon.fr (fan of Bruce Springsteen) =========================== ------------------------------ From: raz@sleeper.apana.org.au (Roland Turner) Subject: Re: Intel, the Pentium and Linux Date: Thu, 15 Apr 1993 12:52:15 GMT pgf@cayman.com (Paul Fox) writes: >Arnd Burghardt (INABU@ibm.rz.tu-clausthal.de) wrote: >: Hi folks, >: > >: compatible to the i486, and I said I don't believe. I showed him a ONE_DISK_ >: Linux-System (Emergency disk, with patched lilo to boot from disk), and said >: him : Convice me, boot this : No guts, no glory ! A he decided no glory. >: He won't let anybody touch his holy cows, and not even boot a suspect OS. >In fairness, if I was a doing a demo of just about anything, I wouldn't >boot a foreign disk -- what if it _did_ boot, and rewrote the partition table? Hmm? I see no problem with it. I run a PC store and, yes, if someone strolled in and said "can I boot this" on one of my demo machines I'd have no problem with it. If it rewrote my partition table, so what - chuck in the setup disk and cable into the file server - burn 10 mins to rebuild the demo... -- Bye for now. - Raz. (Roland Turner) raz@sleeper.apana.org.au VK2ZRT Raz@3:712/413.1 (OH) 61 2 319 5700 ------------------------------ From: zy0004@obelix.uni-muenster.de (Martin Rehwald) Subject: How to GCC part 2 Date: 16 Apr 1993 12:17:34 GMT Hello out there. Me again. Yesterday I asked for help linking the math-libary using GCC. Many told me to use '-lm' instead of '-l m'. I did this before - and did not detect any difference in GCC's behaviour. Using AIX-GCC on rs6000 both works fine. With me none at all. By the way there is another funny observation: adding -l m.a causes the error: libm.a not found. this should be 'libm.a.a' not found, no? If you need further information: I'm using GCC-2.3.3, lib4.3.2, Linux-0.99.7 OK. Waiting for more ideas, MR ------------------------------ From: nation@snoopy.sanders.lockheed.com (Robert Nation) Subject: Re: xvt... please Date: 16 Apr 93 09:06:45 The correct address is nation@rocket.sanders.lockheed.com. If many more people express interest, maybe I'll upload the program to tsx-11, but first I'd like to hear back from some people who try it out, and tell me how it works, etc. ------------------------------ From: Timo.Korvola@hut.fi (Timo Korvola) Subject: X authentication problems & missing headers Reply-To: Timo.Korvola@hut.fi Date: 16 Apr 93 15:41:41 I applied the NFS patch to 0.99.8 and now xdm lets local non-root clients connect again. Thanks to Guy Thomas for pointing that out. (I have a lot of stuff on NFS disks, including home directories). However, magic cookie authentication still refuses to work for remote clients. I transfer the authentication key to a remote machine with xauth and rsh, it gets trasfered correctly (xauth lists the same key on both ends), yet the server won't accept connections from any client on the remote machine. So I have to resort to using xhost, which is unsecure on hosts with many users. I'd appreciate any help. Another problem, which came up when I tried to recompile xdm, is that the version 4.3.3 header files net/if.h, net/if_arp.h, net/if_route.h now only include the respective files from /usr/include/linux. However, no such headers are distributed with the kernel. -- Timo Korvola Timo.Korvola@hut.fi ------------------------------ From: sno@umcc.umcc.umich.edu (Stephen Opal) Subject: Problems with 'ld' Date: 16 Apr 1993 09:43:48 -0400 Stupid question time... I've sucessfully compiled my kernel several times now, but since installing lib4p3p3 (all 4 packages) I get this complaint from 'ld'. >ld: can't load library '/lib//libc.so.4' > Incompatible version. I am mearly seeking enlightenment on correcting this problem. E-mail is welcome. -- Stephen N. Opal sno@umcc.umich.edu ------------------------------ From: ewtroan@ewt.raleigh.ibm.com (Erik Troan) Subject: Re: GIF Viewer for Linux? Date: Fri, 16 Apr 93 13:16:26 GMT Reply-To: ewtroan@ewt.raleigh.ibm.com (Erik Troan) In article <1qi3ld$2je@umcc.umcc.umich.edu>, lcd@umcc.umcc.umich.edu (Leon Dent) writes: |> |> Note to sunsite admins: There was also a binary version of xv that |> did not require these libs. It went by the name xv-jmp.Z (I think). |> It would be nice if it could be restored. |> |> Leon Dent |> lcd@umcc.umich.edu |> Groan... Sorry, my mistake. Someone told be that xv-jmp.Z was obsolete, so I erased it. If someone would upload a version that doesn't require the other shared libs (i.e. libgr) I'd be happy to have it. Sorry for my mistake... Erik + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + When I was in school, I cheated on my metaphysics exam: I looked into the soul of the boy sitting next to me. -- Woody Allen Erik Troan = ewt@sunsite.unc.edu, ewtroan@vnet.ibm.com ------------------------------ From: "Mario A. Guerra" Subject: Re: Help with SLS install... Date: 16 Apr 1993 10:14:36 -0400 Reply-To: mguerra@inforisc.cr Well, YES!!!!. You can have other way of installing Linux without backuping your 100 MB. There is an utility named FIPS which re-partition your disk in a non-destructive way so you can install Linux afterwards. It works!. I helped a friend to install Linux in a 100 MB. DOS partition. Now he has Linux in a 50 MB p. and DOS in the other 50. Get the file NEW from sunsite.unc.edu, dir. /pub/Linux. There you non-destructive way so you can install Linux afterwards. It works!. I helped a friend to install Linux in a 100 MB. DOS partition. Now he has Linux in a 50 MB p. and DOS in the other 50. FIPS is in sunsite.unc.edu, directory /pub/Linux/system/install/fips01alpha.tar.z. Hope this helps. Mario Guerra mguerra@inforisc.cr ------------------------------ From: wong@cs.tu-berlin.de (Wolfgang Jung) Subject: Request: which shells Date: 16 Apr 1993 13:51:36 GMT Hi --- which shells are available for Linux I am particular interested in bourne shell compatibles.. I just found one drawback of bash on linux: The make depend from Smail 3.1.28 was doing nonsense . checking the sh scripts I found the problem to set the CASE_NO_NEWLINES the bash was not able to work with newlines in arguments inside a case switch... ksh blew on some other kind, it just hang at one place or was that awfully slower than bash. Now> What shells are available for linux and which type they are. I know of: bash, ksh, tcsh, zsh. I know bas is some how bourne shell compatible tcsh somehow csh and ksh also sh.. Gruss Wolfgang ------------------------------ From: keith%accura_nc@concert.net (Keith B. Kee {AISG} ) Subject: Where is ipc+ ? Date: 16 Apr 1993 10:38:32 -0400 Reply-To: keith%accura_nc@concert.net (Keith B. Kee {AISG} ) I cna't find ipc+ any more. Where has it gone to? ------------------------------ From: rmast@cs.vu.nl (Mast RN) Subject: C= 64 emulator project Date: 16 Apr 93 14:21:03 GMT Hi! I found on comp.sys.cbm a message of a project concerning making a C version of a Commodore 64 emulator! I ftp'd it from ccosun.caltech.edu /pub/rknop/INCOMING. for working with it you also need a copy of the copyrighted roms. It is too complex to look like an april fools joke. I haven't got it running on linux yet, but it should work with X in color. Robert (rmast@cs.vu.nl) ------------------------------ ** 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-Activists-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux) via: Internet: Linux-Activists@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 tupac-amaru.informatik.rwth-aachen.de pub/msdos/replace The current version of Linux is 0.99pl6 released on February 23, 1993 End of Linux-Activists Digest ****************************** r Linux? Date: Fri, 16 Apr 93 13:16:26 GMT Reply-To: ewtroan@ewt.raleigh.ibm.com (Erik Troan) In article <1qi3ld$2je@umcc.umcc.umich.edu>, lcd@umcc.umcc.umich.edu (Leon Dent) writes: |> |> Note to sunsite admins: There was also a binary version of xv that |> did not requirdigest860 644 36676 74430 53042 5366137104 5754 From: Digestifier To: Linux-Activists@senator-bedfellow.mit.edu Reply-To: Linux-Activists@senator-bedfellow.mit.edu Date: Fri, 16 Apr 93 15:15:29 EDT Subject: Linux-Activists Digest #860 Linux-Activists Digest #860, Volume #4 Fri, 16 Apr 93 15:15:29 EDT Contents: Compileing Kernels 99.6 99.7 99.8 (BARRY TITMARSH) Re: Automated Linux PowerDown (Erik Troan) five nice little questions (not urgent, but would be nice to get some answers) (Safuat Hamdy) Re: (X386) light C shared image??? (HJ Lu) Re: How to GCC part 2 (HJ Lu) Re: Compileing Kernels 99.6 99.7 99.8 (Tero Laakkonen) About Motif !!?? (H. J. Hong) Re: Anyone using Orchid VA & S3? (Joe Carvalho) Re: Is there a .au player for soundcards? (David Sears) libX11.so.2.1 needs. (Hur InHo) Serial under X (Daniel S. Ridge) Telnetd problem in 99.8 (Kari T. Salmela) Re: Summary of Linux vs. 386BSD vs. Commercial Unixes (Marc Wandschneider) Serial problem in 0.99-p8 (Hendrik G. Seliger) ---------------------------------------------------------------------------- Date: Fri, 16 Apr 1993 10:07:12 CET From: BARRY TITMARSH Subject: Compileing Kernels 99.6 99.7 99.8 I have seen lots of postings about problems compileing 99.6 to 99.8 kernels. Just like to say I have had no problems with any of the kernels or any odd configs etc with 0.99.x release. I will point out im still useing GCC 2.2.2d7 and Libc.4.2 etc.. since i have not seen any problems with my config and compileing and i have seen problems with others useing GCC 2.3 / 2.4 and newer libc's Im wondering if there are some problems with the New Compilers.? Ill upgrade when it stops for me..HI HI Barry... Comments ? ------------------------------ From: ewtroan@ewt.raleigh.ibm.com (Erik Troan) Subject: Re: Automated Linux PowerDown Date: Fri, 16 Apr 93 13:46:04 GMT Reply-To: ewtroan@ewt.raleigh.ibm.com (Erik Troan) In article , jkaidor@synoptics.com (Jerome Kaidor) writes: |> |> I was thinking of using one of the DTR lines on the serial ports. Trouble with them, is that you really don't know |> what will happen to them during bootup, etc. One glitch, and BZZZzzzzz*. My solution to the problem of unknown |> states and glitches is to place some very specific requirements on the computer's output. Specifically, a simple circuit |> consisting of a comparator, a couple of capacitors, a resistor or two, and a diode, would make the turn-off circuit only |> trip when a large group of *pulses* came out of the computer. Now, it's probably not usual to see |> a couple thousand pulses come out of a DTR output, is it? |> A simpler way might be to put a slow timer in the box along with a counter. When the meachanical switch flips, it resets the counter in the box you're building. Then the counter start counting (at say, 1 or two Hz). When it gets to some sufficiently large power of two (say 8 or 16), the counter turns itself off and starts paying attention to the line coming in from the linux box. I think you could probably do this with a count, 555 timer, and probably a couple of and gates (to ignore the linux box for a while and to turn off the timer - might need a not as well). Thus, the total circuitry would probably cost about $2 and be much simpler then building a full state sequencer. If you ever get anything working, would you let me know? I'm interested in seeing how you finally do it. Erik + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + When I was in school, I cheated on my metaphysics exam: I looked into the soul of the boy sitting next to me. -- Woody Allen Erik Troan = ewt@sunsite.unc.edu, ewtroan@vnet.ibm.com ------------------------------ From: hamdy@rzdspc22.informatik.uni-hamburg.de (Safuat Hamdy) Subject: five nice little questions (not urgent, but would be nice to get some answers) Date: Fri, 16 Apr 1993 15:42:45 GMT Dear Linux community, wizards and wizards in spe, I have some questions which are neither critical nor urgent, but i would like to know some things. I'm sorry if one of this questions appeared formerly in this group or in the brand newest FAQ since I don't want to force the fellow news-readers to wade through masses of text more than they already do, but I have still some serious jobs to do, other than reading news all day. Please be patient. But now: 1) In "Essential System Administration", AE. Frisch, O'Reilly, the description of SYS V's init and it's runlevels does not match what LINUX' init (from which it is said to be SYS V like) tells me. E.g. under LINUX' init the runlevel for normal usage is 6. But in EssSysAdm you find runlevel 6 described to be for immediate rebooting, and so on. Could someone explain it (and perhaps give a complete explanation of each LINUX' init's runlevels) ? It would be nice if one of the wizards could write a man-page for init and inittab, unless this has already been done of course. 2) Does someone have a report about all available filesystems with some comparing comments about reliability, speed, pros, cons, special faetures etc.? Where? I think this is something to add in the FAQ (unless this has already been done, I haven't the newest, excuse me). 3) Monthes ago, there was a little discussion about memory above the 16MB limit on the ISA-bus architecture. Someone told that it would be better to give the system all memory (32MB on my box, with AHA 1542B) than to create a ram-disk on the upper 16MB or to use them as a swap-device on the upper 16MB. I recompiled the kernel as described (0.99-pl4 --- not the newest but STABLE!) ... it was really nice to see "free"'s output, and everything ran whitout any trouble, as used before except ... hmmmm ... I found everything running 3-5 times slower!!! (<-- this is really true, and it is more than "slightly less" performance!) This is not acceptable, so I returned to the idea of a swap-device or something like this. Question: Either does any solution in this direction exist? (Where? How?) or should I retry the same with a newer kernel release? (Linus or one of the kernel-wizards might know a good answer :-) but be aware: SCSI !) 4) There were many messages about LILO working together with this-and-that-OS but I never read something about LILO working together with an SCSI- adapter. Could someone reliably tell me if it is possible to use the OS/2 boot-manager so that it calls LILO from the menu choice ON AN AHA-1542B ? =============== I mean, I would like to install LILO on my HD, but only in such a way that I need not to fetch all install-disks for all software that was formerly on HD. 5) Where to get Xview with a nice little installation-script, all neccessary include-files & libs, man-pages, and perhaps some utilities? Thanks in advance S. Hamdy hamdy@rzdspc1.informatik.uni-hamburg.de ------------------------------ From: hlu@luke.eecs.wsu.edu (HJ Lu) Subject: Re: (X386) light C shared image??? Date: Fri, 16 Apr 93 14:21:03 GMT In article <1993Apr16.075545.686@truffula.sj.ca.us> cls@truffula.sj.ca.us (Cameron L. Spitzer) writes: >Once again, I'm trying to install X386. I have HJ's root and >base, and 0.99pl8. >I obtained and untarred SLS disks x1-x10. I have an Xconfig >I believe is right. My (Logitech) mouse passes HJ's mouse test. >Startx produces the following error messages: > >Undefined C library functions: > 1. light C shared image (Use the real one instead.) >giving up. >xinit: No such file or directory (errno 2): unable to connect to X Server >xinit: No such process (errno3): Server error. > > >My questions are: >a. What's "the real one" and where might I obtain it? > Is there a heavy C shared image? Get libc 4.3.3 from tsx-11 under pub/linux/GCC. > Does SLS use a different set of libraries than HJ supplies? Yes. SLS may use a old buggy one. Peter will update it. >b. What document should I have read that I would know what a > "light C shared image" is? What newsgroup? cd /lib ls -l You will see libc-lite.so.4.3.3 and libc.so.4 is linked to it. >c. How can I redirect the output of startx into a file? > When I tried > (startx 2>&1 ) > startx.errors & > from HJ's ksh, it printed nothing and the file was empty. > That is because xinit doesn't run with libc-lite.so.4.3.3. >Any help much appreciated! > H.J. ------------------------------ From: hlu@luke.eecs.wsu.edu (HJ Lu) Subject: Re: How to GCC part 2 Date: Fri, 16 Apr 93 14:27:38 GMT In article <1qm84u$jj1@obelix.uni-muenster.de> zy0004@obelix.uni-muenster.de (Martin Rehwald) writes: >Hello out there. >Me again. >Yesterday I asked for help linking the math-libary using GCC. >Many told me to use '-lm' instead of '-l m'. I did this before - >and did not detect any difference in GCC's behaviour. >Using AIX-GCC on rs6000 both works fine. With me none at all. >By the way there is another funny observation: > adding -l m.a causes the error: libm.a not found. > this should be 'libm.a.a' not found, no? >If you need further information: > I'm using GCC-2.3.3, lib4.3.2, Linux-0.99.7 > You should update to libc 4.3.3. >OK. >Waiting for more ideas, I still don't know what you did. Please try gcc -o foo foo.o bar.o .... xxx.o -lm ^^^^^ -lm has to be after those modules which use it. H.J. ------------------------------ From: laakkone@klaava.Helsinki.FI (Tero Laakkonen) Subject: Re: Compileing Kernels 99.6 99.7 99.8 Date: 16 Apr 93 16:27:51 GMT In <93106.100712BTITMARS@ESOC.BITNET> BARRY TITMARSH writes: >I have seen lots of postings about problems compileing 99.6 to 99.8 >kernels. Just like to say I have had no problems with any of the >kernels or any odd configs etc with 0.99.x release. >I will point out im still useing GCC 2.2.2d7 and Libc.4.2 etc.. >since i have not seen any problems with my config and compileing >and i have seen problems with others useing GCC 2.3 / 2.4 and newer libc's >Im wondering if there are some problems with the New Compilers.? >Ill upgrade when it stops for me..HI HI i managed to compile 0.99.8 under my old 0.96c using gcc-2.3.3 *without* jump tables and stuff... it's a good idea to keep old kernels! -tero ------------------------------ From: hjhong@ev004.ev.nctu.edu.tw (H. J. Hong) Subject: About Motif !!?? Date: Fri, 16 Apr 1993 16:00:27 GMT Is there any one useing Motif on linux/Xfree ? Any message is Nice for me, please .... Thanks! H.J. 1993,4,16 ------------------------------ From: carvalho@tribiet.uucp (Joe Carvalho) Subject: Re: Anyone using Orchid VA & S3? Date: 15 Apr 93 23:41:11 GMT gvreugde@plg.uwaterloo.ca (Gord Vreugdenhil) writes: : : I am upgrading my system at the end of the month. The video : card seems to be the hardest decision.... I suspect that : the Orchid VA/VLB (Vesa) card will be my choice. Is anyone : using this card with the S3 server? If so, let me know. I ...Stuff deleted. Hi Gord! I am using the Orchid VA, the ISA version. I like it. Runs quite nicely with Xs3. I've included the Xconfig I use. It needs a little work. It will help in getting you up and running. Also...the board is VERY fast under MSWindows. =================================8<========================================== # $Header: /home/x_cvs/mit/server/ddx/x386/Xconfig,v 1.14 1992/09/12 07:03:09 dawes Exp $ # $XConsortium: Xconfig,v 1.2 91/08/26 14:34:55 gildea Exp $ # # Copyright 1990,91 by Thomas Roell, Dinkelscherben, Germany. # # Permission to use, copy, modify, distribute, and sell this software and its # documentation for any purpose is hereby granted without fee, provided that # the above copyright notice appear in all copies and that both that # copyright notice and this permission notice appear in supporting # documentation, and that the name of Thomas Roell not be used in # advertising or publicity pertaining to distribution of the software without # specific, written prior permission. Thomas Roell makes no representations # about the suitability of this software for any purpose. It is provided # "as is" without express or implied warranty. # # THOMAS ROELL DISCLAIMS ALL WARRANTIES WITH REGARD TO THIS SOFTWARE, # INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS, IN NO # EVENT SHALL THOMAS ROELL BE LIABLE FOR ANY SPECIAL, INDIRECT OR # CONSEQUENTIAL DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE, # DATA OR PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER # TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR # PERFORMANCE OF THIS SOFTWARE. # # Author: Thomas Roell, roell@informatik.tu-muenchen.de # # some nice paths, to avoid conflicts with other X-servers # RGBPath "/usr/X386/lib/X11/rgb" FontPath "/usr/X386/lib/X11/fonts/misc/,/usr/X386/lib/X11/fonts/75dpi/" # Use this if the Type1 font code is in the server #FontPath "/usr/X386/lib/X11/fonts/misc/,/usr/X386/lib/X11/fonts/75dpi/,/usr/X386/lib/X11/fonts/Speedo/" # NoTrapSignals #Xqueue Keyboard AutoRepeat 500 5 # Xleds 1 2 3 ServerNumLock # DontZap Microsoft "/dev/mouse" BaudRate 1200 # SampleRate 150 Emulate3Buttons # # The graphics drivers # First the colour driver # vga256 Virtual 1024 768 Chipset "S3" Videoram 1024 # Clocks 25 28 40 71 50 77 36 45 80 120 130 32 110 65 75 72 Clocks 25 28 32 36 40 44 45 50 64 65 78 56 75 80 85 89 100 32 # To disable SpeedUp, use NoSpeedUp # # NoSpeedUp ViewPort 0 0 Modes "1" "2" "3" "4" "5" "6" "7" "8" "9" # "10" "11" "12" ModeDB # OFFICIAL VESA Monitor timings + IBM Standards - TRY THESE FIRST # Contributor: Thomas Roell [roell@sgcs.com] # Last Edit Date: 3/29/92 # # name clock horizontal timing vertical timing flags "1" 28 728 736 832 896 501 506 508 519 "2" 65 1016 1032 1240 1384 755 762 765 781 #"3" 40 1024 1024 1152 1216 768 768 770 785 #"4" 64 968 984 1192 1336 771 778 781 797 # name clock horizontal timing vertical timing flags "1" 25 640 664 760 800 480 491 493 525 "2" 31 640 664 704 832 480 489 492 520 "3" 36 800 824 896 1024 600 601 603 625 "4" 40 800 840 968 1056 600 601 605 628 "5" 50 800 856 976 1040 600 637 643 666 "6" 45 1024 1040 1216 1264 768 777 785 817 "7" 65 1024 1032 1176 1344 768 771 777 806 "8" 75 1024 1048 1184 1328 768 771 777 806 #"9" 85 1024 1032 1152 1360 768 784 787 823 "9" 85 1024 1128 1248 1456 768 784 787 823 # "10" 80 1280 1296 1512 1568 1024 1025 1037 1165 Interlace # "11" 110 1280 1328 1512 1712 1024 1025 1028 1054 # "12" 135 1280 1312 1456 1712 1024 1027 1030 1064 =================================8<========================================== -- --joe Beacher Tool & Die...A Hughe Jorgen Company. 1106 2nd Street, Suite 316, Encinitas, CA. 92024 ------------------------------ From: dns@essnj3.essnjay.com (David Sears) Subject: Re: Is there a .au player for soundcards? Date: Fri, 16 Apr 1993 11:41:05 GMT In <1qhpl6$7di@usenet.INS.CWRU.Edu> damien@b63519 (Damien Neil) writes: >Rick Miller - Linux Device Registrar (rick@ee.uwm.edu) wrote: >: It plays Sun-style audio files ("*.au") through the internal speaker. >: No sound-card is needed, and it won't use it even if you've got one. >This reminds me -- is there an .au player that _will_ use a soundcard? I have >splay from the snd-util package, but the sound it produces when playing an .au >file makes it seem like I'm getting a poor radio transmission...from Venus. Not [ ... ] >Damien Neil dpn2@po.cwru.edu "Until somebody debugs reality, the best >Case Western Reserve University I can do is a quick patch here and there." I understand that sox (sound effects) is being released soon to comp.unix.sources. It has linux sound card support and will play .au files and many others. -- =========================================================== David Sears | EMAIL: dns@essnj3.essnjay.com Huntington, CT USA-06484 | 76474.3113@compuserve.com ------------------------------ From: artemis@eve.kaist.ac.kr (Hur InHo) Subject: libX11.so.2.1 needs. Date: Fri, 16 Apr 93 08:50:40 GMT I got xtank binary. but when I executed it, it says ' I need libX11.so.2.1 '... anyone who has libX11.so.2.1, mail me please. ------------------------------ From: newt@eng.umd.edu (Daniel S. Ridge) Subject: Serial under X Date: 16 Apr 1993 14:45:36 GMT I'm a linux neophyte, and I need help. This is my problem: Under X(X binaries released w/ SLS 1.0, any window manager) I cannot properly manage serial/modem connections. When I start a program, such as minicom, the app starts, and hangs at the point where the app generally does something with the com port( such as initalize the modem). Ordinarily, I would just switch to a text console, and things work fine, but I have started to use term, and txconn specifically. I cannot tolerate flipping from X to text and back to see a serial program refresh! I need to see these things refresh in (near)realtime, or txconn will not be a useful venture. This is what I am running: SLS 1.0, w/ Xfree 1.2 386DX-40,8m,120HD 2400baud internal modem on ttyS2(com 3) 2400baud internal modem on ttyS3(com 4) term1.07(also term1.05) minicom Please help if you can, and let me press a little further by asking all of you helpful people out in netland to write me at newt@eng.umd.edu if you have the solution, suggestions, or a similar problem. If I find that several people are having this problem, I will post the definitive solution when I find it. Thank you in advance, Dan Ridge (newt@eng.umd.edu, newt@wam.umd.edu) ------------------------------ From: weasel@mecom.oulu.fi (Kari T. Salmela) Subject: Telnetd problem in 99.8 Date: Fri, 16 Apr 1993 19:04:47 GMT I have a follwing problem with telnetd (could be somewhere else too). It goes like this: when I telnet into my linux box from the network, login process does not recognize :s but prints ^M:s instead. From the console telnet works fine to both directions. Can anybody think of any solutions, please? I'm getting desperate.. //Kari Salmela, Computer Services Centre, University of Oulu, Finland ksalmela at ousrvr.oulu.fi && Kari Salmela at SMTPLink ------------------------------ Crossposted-To: comp.os.386bsd.questions From: storm@cs.mcgill.ca (Marc Wandschneider) Subject: Re: Summary of Linux vs. 386BSD vs. Commercial Unixes Date: Fri, 16 Apr 1993 16:27:49 GMT In article <1993Apr15.225354.18654@samba.oit.unc.edu> Brandon.Vanevery@launchpad.unc.edu (Brandon Vanevery) writes: > >SCO Unix has been recommended to me as "the best" of the commercial >systems, by a few folks. I had the pleasure of using this this product last summer, and have slowly come to the conclusion that anybody who finds this a superior product has either never used ANY other UNIX before, is high, or finds AIX a close second. Never before have I used a UNIX so lacking in features, so annoying to use (even for a SVR? system), and so difficult to get the hang of. However, I've been weaned exclusively on SunOS, with HP/UX and AIX as occasional experiences. Toodlepip! Marc 'em. -- storm@cs.mcgill.ca McGill P.O.W Camp "Oh crap---It's not Marc Wandschneider Montreal, CANADA random enough" 386bsd---Free UNIX! ftp agate.berkeley.edu /pub/386BSD/386bsd-0.1 Ever wonder what Wonder Woman wonders about? ------------------------------ From: hank@Blimp (Hendrik G. Seliger) Subject: Serial problem in 0.99-p8 Date: Fri, 16 Apr 1993 14:47:35 GMT Reply-To: hank@Blimp.automat.uni-essen.de Hi there! Since I upgraded to 0.99-p8 (from p7) one of these days, I am experiencing an instability in the serial driver. After using X-windows for a while and usually after some heavy work of the system while I'm playing around with the mouse pointer the mouse starts - behaving as if the ball was muddy (it isn't. Not really) - the the mouse just quits to move. The mouse keys still are o.k., just the thing won't budge a bit. Leaving X and restarting it doesn't help. So far the only thing I found was rebooting, which I really don't like too much. I'd appreciate ideas! Hank. -- ====================================================================== Hendrik G. Seliger Universitaet Essen hank@Blimp.automat.uni-essen.de Schuetzenbahn 70 Tel.: +49-201-183-2898 4300 Essen, Germany ====================================================================== "Handling interrupts is simple." (G. Pajari) "Interrupts are an unpleasant fact of life." (A. Tanenbaum) ------------------------------ ** 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-Activists-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux) via: Internet: Linux-Activists@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 tupac-amaru.informatik.rwth-aachen.de pub/msdos/replace The current version of Linux is 0.99pl6 released on February 23, 1993 End of Linux-Activists Digest ****************************** u if you have the solution, suggestions, or a similar problem. If I find that several people are having this problem, I will post the definitive solution when I find it. Thank you in advance, Dan Ridge (newt@eng.umd.edu, newt@wam.umd.edu) ------------------------------ From: weasel@mecom.oulu.fi (Kari T. Salmela) Subject: Telnetd problem in 99.8 Date: Fri, 16 Apr 1993 19:04:47 GMT I have a follwing problem with telnetd (could be somewhere else too). It goes like thisdigest861 644 36676 74430 50657 5366137105 5767 From: Digestifier To: Linux-Activists@senator-bedfellow.mit.edu Reply-To: Linux-Activists@senator-bedfellow.mit.edu Date: Fri, 16 Apr 93 20:15:17 EDT Subject: Linux-Activists Digest #861 Linux-Activists Digest #861, Volume #4 Fri, 16 Apr 93 20:15:17 EDT Contents: LILO does not ask for alternative boots TERM109=TERM107, sorry (Keith Smith) Advice on Ext File system (Raghunath K Rao) Re: problems with tar: broken pipe, Problems with lpd (Stefano Ianigro) Re: Why Linux should remain Linux and seagulls should not be used. (Leon Dent) Diamond Video Cards (Ross Bagley) Re: Is there a .au player for soundcards? (Jan Brittenson) Re: Is there a .au player for soundcards? (Zack Evans) Where to get 1542C fix-patched bootable image ? (In Sik Rhee) Re: PS/2 style mouse not supported?? (Kee Chua) Linux Installation problems (Jake Vogelaar) Re: Help with rawrite a1 on 5.25 HD Floppy (Van Zandt) Problem with etc/mtab (Chris {Fig} Newton) Re: C= 64 emulator project (Robert Chen) X keysyms: where do they come from? (Jerome Kaidor) Re: Linux & GUS (Hannu Savolainen) NFS-server for Linux??? (Tuukka Uusheimo) Mailing Lists (Ahmon Dancy) ---------------------------------------------------------------------------- Date: Friday, 16 Apr 1993 14:35:02 TUR From: Subject: LILO does not ask for alternative boots Well, I have installed LINUX, successfuly. The problem is with LILO. As far as the lilo documentation says it should be possible to boot from any bootable partition and at boot time this shall be asked to the user, I have never seen it so I assume it is something like which partition to boot from (1,2) ? : where the user enters his choice. I believe I followed the instructions in README and have created a install script. The problem I encounter is that no such question is asked. The systems goes straight to the LINUX boot and does it Actually there is also a bootable DOS partition which was declared by a script line other= which is following the linux boot info (in the script file). The interesting stuff is when I remove the linux boot lines from this script file and leave the one (other=..) for DOS alone, then DOS bootes automatically. What is the point I miss? (is it only possible to boot automatically?). Do I have to do each time an install to get the next boot from the partition I want? I will appreciate greatly any help. Gokturk Ucoluk ------------------------------ From: ksmith@monu6.cc.monash.edu.au (Keith Smith) Subject: TERM109=TERM107, sorry Date: Fri, 16 Apr 1993 14:22:59 GMT regarding my TERM problem, I was using TERM107....that is the latest version. Apologies to those who have searched the world for 109 Cheers, Keith -- ===================================================================== - ksmith@monu6.cc.monash.edu.au | Keith Smith, Monash Uni, - - ksmith@vx24.cc.monash.edu.au | Melbourne, Australia - ===================================================================== ------------------------------ From: thssrkr@iitmax.iit.edu (Raghunath K Rao) Subject: Advice on Ext File system Date: Fri, 16 Apr 93 18:22:11 GMT Hi , I have experimented with the SLS 1.0 release and found it quite satisfactory. I used the MINIX file system but the only problem is its limitations on fs size and filename size. I want to go for the Ext fs, but am diffident due to the alpha release and the bug reports. Can someone out there give me first hand experience of the EFS? Is it stable and can I safely put my 540 MB - 486 research machine (which I need for serious work not goofin around!) on it? Responses will really be appreciated!! Raghunath Rao ------------------------------ From: w_stef@athalia.unibw-hamburg.de (Stefano Ianigro) Subject: Re: problems with tar: broken pipe, Problems with lpd Date: 16 Apr 93 21:05:54 GMT In article <1993Apr16.072819.28267@bernina.ethz.ch> thoeny@ito.umnw.ethz.ch (Juerg Thoeny) writes: [som stuff about tar deleted...] I have also an other problem. The lpd doesn't accept bigger streams than 1MB. Since I'm using a postscript printer this problem, I encounter this problem every day. I use a remote printer, which is spooled by a SUN-Server. Thanks in advance Hey Juerg, try to use the :mx#0: entry in your printcap (No limitations in file size) Hope ths helps. Cheers! Stef -- ======================================================================== | Stefano Ianigro E-Mail: w_stef@unibw-hamburg.de | | Institut fuer Informatik | | Universitaet der Bundeswehr Hamburg "Ich weiss, dass ich nichts | | Holstenhofweg 85 weiss..." (Sokrates) | | D-2000 Hamburg 70, Germany (ab 01.07.93: D-22043 Hamburg) | | Tel.: 040/6541-2656 oder -3345 (Sekr.) | ======================================================================== ------------------------------ From: lcd@umcc.umcc.umich.edu (Leon Dent) Subject: Re: Why Linux should remain Linux and seagulls should not be used. Date: 16 Apr 1993 16:58:39 -0400 I remarked in another article that Linus looked like Micheal Doonsbury. A caricature on that order might make a nice logo. An apropos homage... Leon Dent lcd@ais.org ------------------------------ Subject: Diamond Video Cards From: rabagley@termite.occ.uc.edu (Ross Bagley) Date: Fri, 16 Apr 1993 20:03:39 GMT I just got off of the phone with one of Diamond Computer Systems' Tech Managers (I think). He seemed to think that the thing Diamond actually objected to was any other party using their data (driver information) as a part of any project/product which would be sold. After perusing the GNU License I came to the conclusion that this possibility was not excluded by the license, and I am wondering if there is a license that exists which can legally preclude the possibility of sale... I am considering purchasing a machine with 2 VESA slots and the upgrade to the Diamond Viper card is extremely attractive in this package deal. If I can manage to copyright a driver for Diamond parts which will satisfy Diamond and GNU I think all could win... Please help. Reply any way you feel is appropriate... -- _____________________________________ _______ | | /\______\ | rabagley@ddt.occ.uc.edu | / / ___ /_____ ______ ______ | | / / /_\/ /______\\_____\\_____\ | Life would be so much easier if | / / _ __// ___ // ___// ___/ | we could just see the source code | / / / \ \ / /_\/ //___ //___ / | -Dave Oleson SGI | \/_/ \_\\_____//_____//_____/ |___________________________________| ------------------------------ From: bson@gnu.ai.mit.edu (Jan Brittenson) Subject: Re: Is there a .au player for soundcards? Date: 16 Apr 93 16:03:45 In article <1993Apr15.232230.17181@a.cs.okstate.edu> kennejs@a.cs.okstate.edu (KENNEDY JAMES SCOT) writes: > How about just using a utility to convert Sun audio files over to a > format that splay can handle? I have written such a program for SVR4. It's free software; you're more than encouraged to port it to Linux. Probably a trivial job. It can handle all .au formats except IEEE floating-point and PCM-24. For 8-bit u-law, it defaults to emulating the MC14400 in Sun workstations. If a value is specified for u (micro) or O, then it switches to idealized North American u-law. It would be nice if it could do European u-law, too (hint, hint!). It can play gzip'ed files (option -z) and can either write .snd output (optional) or fork-exec a player (default). If you or anyone else is interested in porting it, send me a note and I'll mail you the code. I haven't made it publicly available yet. -- -- Jan Brittenson bson@gnu.ai.mit.edu ------------------------------ From: zevans@nyx.cs.du.edu (Zack Evans) Subject: Re: Is there a .au player for soundcards? Date: Fri, 16 Apr 93 19:50:44 GMT In article <1993Apr15.232230.17181@a.cs.okstate.edu> kennejs@a.cs.okstate.edu (KENNEDY JAMES SCOT) writes: >From article , by ins407x@aurora.cc.monash.edu.au (Andrew J. Cosgriff): >> damien@b63519 (Damien Neil) writes: >How about just using a utility to convert Sun audio files over to a format >that splay can handle? I have a program called SOX that can translate au >sounds into voc sounds. SOX is capable of doing other kinds of conversions >as well. Of course you'll have to go to the trouble of converting the >sounds over first before you can play them, but that shouldn't be a big deal. > >Unless I'm mistaken, you can get SOX off of nic.funet.fi. It should be under >/pub/unix/sound. The name of the file is sox7.linux.tar.Z. Sox munged .wav files last time I used it even though it's supposed to handle them. I found a great app for Windows (it has some uses :) - look for sndtool26.zip I think. Zack -- Zack Evans pyc081@cent1.lancs.ac.uk or zevans@nyx.cs.du.edu (Internet) pyc081@uk.ac.lancs.cent1 (JANET) Watch yer bass bins lads, I'm tellin' yer... ------------------------------ From: c169-ab@cephus.berkeley.edu (In Sik Rhee) Subject: Where to get 1542C fix-patched bootable image ? Date: Fri, 16 Apr 1993 21:01:27 GMT I have a 1542C and I don't have linux already installed, and I want to use it so I need to download a boot disk with the patch already applied to it. can someone direct to me a site where I can download this? If none exists, could someone be kind enough to make such a bootable and put it up on the mit or sunsite server? I have seen other users post with the same problem and I am sure they too would appreciate your efforts... thank you very much, Insik. ------------------------------ From: kkc@stat.nursing.arizona.edu (Kee Chua) Subject: Re: PS/2 style mouse not supported?? Date: 16 Apr 93 17:45:37 GMT Reply-To: kkc@stat.nursing.arizona.edu Get the "mconv.c" file and you should be ok. The instructions are in that file. Kee Chua ------------------------------ From: jxv3790@hertz.njit.edu (Jake Vogelaar) Subject: Linux Installation problems Date: 16 Apr 93 22:21:47 GMT Thanks to various people who pointed out places to get Linux in the U.S., my friend finally purchased it on CDROM. However, that's not the end of it. He asked me to post the following question/problem: " Have just purchased Yggdrasil Linux (on CD-ROM) and find that I can't install it... The installation program won't recognize the host adaptor on my system. My machine is a Tangent 486/50 with a cached SCSI-2 host adaptor from "Asustek" (?). It normally runs under an ASPI driver. The adaptor can't be too bizarre as, last week, I installed the latest Windows NT beta which identified it as a "BusLogic" unit and had no problem with it. I phoned Yggdrasil, who said that the installation program was only testing for the presence of a port at an appropriate address and couldn't offer any further advice. As far as I know, the SCSI configuration is dead standard and I haven't changed any of the default settings. Would appreciate any advice toward getting Linux installed or toward getting around the installation program. Thank you. " -- //////////////////////////////////-------\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\ Jake Vogelaar: jxv3790@hertz.njit.edu, jake@aspen.njit.edu Computer Vision R&D, Warner-Lambert Manufacturing Development Laboratory New Jersey Institute of Technology ------------------------------ From: jrv@mbunix.mitre.org (Van Zandt) Subject: Re: Help with rawrite a1 on 5.25 HD Floppy Date: Fri, 16 Apr 1993 17:06:43 GMT johnw@unixhub.SLAC.Stanford.EDU (Jonathan S. Wong) writes: >I'm trying to use rawrite to write a1 and a2 to a 5.25 floppy in DR DOS 6. >I keep getting errors. (forget what it said) but it apparently >didn't work all the way because after booting a1 and pressing >or choosing and selecting resolution, my machine will keep rebooting! > >Do I have to use rawrite on a 3.5 inch floppy? > >-Jonathan > rawrite works for any floppy. For a1 and a2 you need a 1.2M or 1.44M. They should be "freshly formatted". - Jim Van Zandt ------------------------------ From: u0xh@jupiter.sun.csd.unb.ca (Chris {Fig} Newton) Subject: Problem with etc/mtab Date: Fri, 16 Apr 1993 22:10:08 GMT Mount doesn't seem to recreate my etc/mtab file correctly on boot up anymore. When I use df, it dies saying it can't read table of mounted file systems. I check /etc/mtab and it has in it: /dev/hda1 / The proc one is ok. Shouldn't this one read : /dev/hda1 / efs defaults When I run mount rm mtab, and run mount -av agian, it recreates it ok, but it is not doing it at boot time properly. My /etc/rc does remove /etc/mtab, and uses mount -av, but it is not doing it correctly obviously. I have checked my etc/fstab file, and it reads: /dev/hda1 / efs defaults /dev/hda2 none swap none /proc proc defaults I belive this to be correct, but something must be wrong somewhere. Anybosy have any suggestions? Chris ============================================================= Huh? dot sig what?!? ------------------------------ From: rchen@fraser.sfu.ca (Robert Chen) Subject: Re: C= 64 emulator project Date: Fri, 16 Apr 1993 23:02:32 GMT In article rmast@cs.vu.nl (Mast RN) writes: > >Hi! I found on comp.sys.cbm a message of a project concerning making >a C version of a Commodore 64 emulator! I ftp'd it from ccosun.caltech.edu >/pub/rknop/INCOMING. > >for working with it you also need a copy of the copyrighted roms. >It is too complex to look like an april fools joke. >I haven't got it running on linux yet, but it should work >with X in color. > >Robert (rmast@cs.vu.nl) It compiles nearly out of the box and works well. I have had one annoying problem. When using the monitor, I have to type ^M^J to get a newline (ie, to enter each command). "Enter" just gives me "^M" written on the screen. If anyone has a fix for this I would be interested in hearing from you. This emulator look like a lot of fun. I still have some old 6510 machine code sitting on some disks in a closet somewhere. I used to love programming my C=64. Too neat. - Rob ------------------------------ From: jkaidor@synoptics.com (Jerome Kaidor) Crossposted-To: comp.windows.x.i386unix Subject: X keysyms: where do they come from? Date: 16 Apr 93 23:08:18 GMT Reply-To: jkaidor@synoptics.com This week, I have been trying make xmodmap define my scroll-lock key to switch to an alternate keyboard. Xmodmap complains that there is no such keysym. This has lead me to a more general question: where do keysyms come from? BTW, the system in question is Xfree86 1.2 running over Linux 99pl6. Apparently, keysyms are two things: 1) a unique ascii string describing the legend printed on a keycap. 2) a four-byte binary "shorthand" for the system to use when it deals with the "keycap" Now, if there was to be a "Scroll_Lock" keysym in my system, where would it have to be defined? In a config file somewhere? Compiled into the X server? Into the X client? And how is the system to know that this keysym is to be produced when I hit that particular physical button on the keyboard? Here's the sequence I fantasize: 1) I hit a key. 2) My OS's keyboard driver grabs the scan codes, does it's own internal state stuff, and outputs an ascii code. 3) The X server grabs the ascii code, looks up it's internal, compile-time-defined, keysym table, and creates a keysym. 4) The X server converts the keysym to a keycode using a table created during xwindow initialization from config files. 5) The X server sends the one-byte keycode to whichever client happens to have keyboard focus. 6) The client uses xmodmap data to change the keycode into keysyms, taking the modify states into account. 7) If the client happens to be xterm, it changes the keysym back into an ascii code, and sends it to whichever client happens to be running inside the xterm. If some X guru out there could correct and complete this scenario, I'd really appreciate it. I'm especially interested in the names and locations of the config files involved. Step three is especially interesting to me. It seems like that's the area I need to poke around in, to get a new key into the system. Why doesn't xmodmap know about Scroll_Lock, anyway?!? Is a server recompile required, or is there a config file I can goose somewhere? I can do without the answers, if somebody can point me to where to find them. And don't tell me the X-window FAQ, I've been through it. Perhaps somewhere in the "mit-4" documentation tree? What about a book at the bookstore? This is important enough so that I'd be willing to spend $30 or so to find out. - Jerry Kaidor ( tr2!jerry@dragoman.com, jkaidor@synoptics.com ) ------------------------------ From: hsavolai@cs.Helsinki.FI (Hannu Savolainen) Subject: Re: Linux & GUS Date: 16 Apr 1993 23:10:01 GMT In <1993Apr16.141311.24002@doug.cae.wisc.edu> calica@cae.wisc.edu (Carlo James Calica) writes: >I know that Linux has drivers for the Sound Blaster. Do drivers exist for >the Gravis ultraSound? I remember a few people wanted to write one a few >months ago. How has that effort progressed. Thanks. I have added some kind of GUS support to the sound driver for Linux. It supports currently just the wave table synthesizer of GUS. It seems to be reliable but there is couple of problems. The first is that the driver under construction and there will be some kind of incompatibility problems between versions (this version is just a pre pre pre alpha version). The another problem is lack of applications which support GUS. There is some sample programs included with the driver (for example a simple .mod -player). The driver can be found at klingon.epas.utoronto.ca (pub/pc/ultrasound/util/unix). Hannu ============================================================================ Hannu Savolainen (hsavolai@cs.helsinki.fi) Kun Savolainen puhuu, niin vastuu siirtyy valittomasti kuulijalle. (or link("/dev/null", "/dev/flames") in plain C) ------------------------------ From: hogan@marie.physik.tu-berlin.de (Tuukka Uusheimo) Subject: NFS-server for Linux??? Date: 16 Apr 1993 18:05:08 GMT Hi! I'm thinking about installing Linux (or 386bsd) back to my PC, as I would like to use it as a X-terminal for my IBM RT. I have also a CD-RO for the PC, (Sony CDU-31A connected to PAS-16) and I know there is support for those in Linux, but! I would also like to use the CD-ROM from my RT via Ethernet, so does Linux (or 386bsd) have support for NFS-server? That is, is this possible, yet? And does anybody have experience on using two displays with X11 on Linux? I heard it's possible with the new XFree86 v:1.2 ! I think thas about all. Thanks in advance! - Tuge -- +----------------------------------------------------------------------------+ I Tuukka Uusheimo, Lappeenranta University of Technology, Suomi - Finland I I Email : hogan@kannel.lut.fi or hogan@marie.physik.tu-berlin.de I +----------------------------------------------------------------------------+ ------------------------------ From: dancy@uclink.berkeley.edu (Ahmon Dancy) Subject: Mailing Lists Date: 16 Apr 1993 23:22:17 GMT Can someone tell me what site serves the Linux mailing lists? I can't seem to find this information anywhere. I am specifically looking for the SLIP mailing list. ------------------------------ ** 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-Activists-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux) via: Internet: Linux-Activists@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 tupac-amaru.informatik.rwth-aachen.de pub/msdos/replace The current version of Linux is 0.99pl6 released on February 23, 1993 End of Linux-Activists Digest ****************************** nvolved. Step three is especially interesting to me. It seems like that's thdigest862 644 36676 74430 51736 5366137105 5767 From: Digestifier To: Linux-Activists@senator-bedfellow.mit.edu Reply-To: Linux-Activists@senator-bedfellow.mit.edu Date: Fri, 16 Apr 93 22:15:22 EDT Subject: Linux-Activists Digest #862 Linux-Activists Digest #862, Volume #4 Fri, 16 Apr 93 22:15:22 EDT Contents: Re: Installation: SLS vs SCO (CALDERWOOD) Re: C= 64 emulator project (not a joke...) (Paul Mora) Re: [Q] What about the new Adaptec 1542C? (Thomas Wieckhorst) Re: Unix OS for 286 (Garrett D'Amore) [Q] Weird Install Problem (x4 disk) (Jialin Zhong) Another compatible ethernet card (Jonathan Naylor) Can't login with 0.99 p8 (Barracuda) Linux for Amiga? (RW Salnick) Re: Diamond Video Cards (Harry Langenbacher) Xconfig for pvga1 & Qume 835 monitor? HPDeskjet 500 drvrs? (John Edward Kellner) Re: Compiling XV; general porting problems (Keith Mancus) Re: Problem with etc/mtab (HJ Lu) Re: flex vs. lex (Jeffrey M. Simon) TCP/IP ... still not working .... (Fritsch_Wolfgang) Re: Linux on Macintrashes? (Was: Re: Linux on Amiga?) (Keith Mancus) ---------------------------------------------------------------------------- From: wcalderw@dante.nmsu.edu (CALDERWOOD) Subject: Re: Installation: SLS vs SCO Date: 16 Apr 1993 23:29:09 GMT Colin Jenkins (jenkins@DPW.COM) wrote: : In article <1pfba3$2pj@nz12.rz.uni-karlsruhe.de> s_titz@ira.uka.de (Olaf Titz) writes: : >Excuse me for being pedantic, but at least this is no valid argument. : >First, a compiler is no security problem, since a compiler does not : >introduce any new features which could be holes, it is just a : >convenience tool. : : I disagree. A compiler represents a huge security hole. System attacks : can be mounted at the lowest level using a compiler. If memory serves, : the Internet worm of a few years ago propogated itself by recompiling : itself and redistributing the source to other systems for recompilation. : : >Second, a dedicated hacker could use even use emacs to create a binary : >executable. (Followup->alt.folklore.computers :-) : : Agreed, however few people have the time, patience, or knowledge required : to generate any program of significant complexity by hand. The compiler : would make attacks easier and faster and more available to hackers of : lesser talents. : : Not that I am arguing against making compilers available, or that linux : is inferior by virtue of poor security- simply that compilers do represent : potential security problems. : : : : Colin You are wrong about the Internet Worm. "There were 3 methods used by Robert Tappan Morris to accomplish unauthorised access to machines and multiply his worm. 1. Try to run a remote shell using rsh. Some machines will permit access using rsh with out any other authentication. If this worked then the shell uploaded the worm program and continued to infect other machines. 2. Used finger. The worm would finger a site with a specially crafted 536-byte string as the parameter. This string overflowed the daemon's buffer and overwrote its stack. There was a bug in the daemon at the time which did not check for overflow. When the daemon returned from the procedure it was in at the time it got the request it returned to the procedure inside the 536-byte string on the stack. This procedure tried to run /bin/sh. Thus allowing access. 3. Used a bug in the mail system, sendmail, which allowed the worm to mail a copy of the bootstrap and get it executed. Once the worm was running on the new machine it would try to crack passwords and gain access to other machines perhaps even cycling back to the same machine at some time thus creating multiple copies of its self." (PG. 184, Modern Operating Systems, Andrew S. Tanenbaum) William A. Calderwood wcalderw@nmsu.edu ------------------------------ From: cs4gp6aw@maccs.mcmaster.ca (Paul Mora) Subject: Re: C= 64 emulator project (not a joke...) Date: Fri, 16 Apr 1993 23:44:18 GMT It's no joke. The program enables you to emulate a C-64 under XWindows. There are some real sick minds out there... Anyways, I have also been tinkering with it to get it to run under linux. I still have an old 64 at my parent's place, and over Easter, I transferred the kernel and basic to disk. A friend and I compiled it on the sun, but we have yet to try it out. I also have a c64 emulator for a PC; it is really cool. It remaps the keyboard and everything (remember, the " is above the 2). Ahhh, the good old days. Paul Mora McMaster U. Hamilton, Ont. email: morap@mcmail.cis.mcmaster.ca ------------------------------ From: thomas@thwieck.han.de (Thomas Wieckhorst) Subject: Re: [Q] What about the new Adaptec 1542C? Date: Thu, 15 Apr 1993 23:17:10 GMT Carlo Dapor (cdapor@iiic.ethz.ch) wrote: : Hello Linuxers! : Can anybody tell me, if the new easy installable scsi controller 1452C is : or will be supported by linux....I have 0.99pl4 installed. Yes it is supported ! Only take the actuell kernel 0.99p8 or even 0.99pre8, because the AHA 1542 C will need some patches when you have an older kernel. bye thomas posting from jet another universe -- Thomas Wieckhorst +49 5192 18848 thomas@thwieck.han.de Kiefernweg 12 GERMANY 3042 Munster / Breloh 1 ------------------------------ From: garrett@garnet.berkeley.edu (Garrett D'Amore) Crossposted-To: comp.os.coherent,comp.os.minix,comp.os.386bsd.questions,comp.os.misc Subject: Re: Unix OS for 286 Date: 16 Apr 1993 18:47:54 GMT I'm normally not the type to respond to my own posts, but noticed the following error: > >Get yourself a copy of Coherent. It's sold by Mark Williams Company, and >comes in two flavors, v4.0 and v3.2. You won't be able to use v3.2 -- ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ >becuase it requires a 386, but v4.0 should work. MWC charges $99 for ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ That should be that you won't be able to run 4.0, but you should be able to run 3.2. I haven't been getting much sleep lately -- CS classes, programming for work, and hacking Linux have all taken their toll. :) BTW, I own both versions and would be willing to sell them, only becuase Linux is so Cool :). The two versions are against one license, so you would be legally bound to run only one or the other version of the OS, at least at the same time. (No copies on different computers.) If you're interested, make an offer. I'd need to verify that this is legal under MWCs license (i.e. transfer of license). ==================================================================== Garrett D'Amore | garrett@haas.berkeley.edu Software Co-Ordinator | 68 Barrows Hall, UC Berkeley Haas Computing Services | Ph: 510-643-5923 Fax: 642-4769 ==================================================================== ------------------------------ From: jlzhong@vision.csl.uiuc.edu (Jialin Zhong) Subject: [Q] Weird Install Problem (x4 disk) Date: 16 Apr 1993 18:48:48 GMT Although I had all kinds of trouble with Maxtor HD, I managed to get Linux (SLS) installed except one diskette x4. The problem is when I do sysinstall -disk x4, it starts to install and then within 10 seconds, it stopped reading from floppy drive. How do I know? I know it because there is no the normal sound generated when reading from the floppy. However, there is no error message whatsoever. It just simply hanged there. I tried download the x4 from ftp site again to another floppy and got the same result. I had no such experience with other diskettes. It will be a shame if I can't get this x4 installed since I have spent so much effort to get this thing going :). I wonder anyone has experienced samilar problem before and know how to fix it. Thanks in advance. --Jialin -- Jialin Zhong * / l l l l * ---O--O--O--O--- 1614 Beckman Institute * / l-l-l -l-l l-l -l- -l- * /--*--*--*--O--O Urbana, IL 61801 * /l l-l-l / l l-l /l\ /l\ * /--/--*--/--*--O jlzhong@uirvld.csl.uiuc.edu */ l l / \l / l X l \* /--/--/--/--/--/ ------------------------------ From: g4klx@g4klx.demon.co.uk (Jonathan Naylor) Subject: Another compatible ethernet card Reply-To: g4klx@g4klx.demon.co.uk Date: Fri, 16 Apr 1993 17:34:22 +0000 Hello fellow Linuxers I am happy to announce that the Commodore ETHER-16 PLUS COMBO (and by implication the ETHER-16 PLUS) work with the standard Linux kernel ethernet driver code. The card supports four operational modes one of which is WD Plus mode (it also does NE1000, NE2000 and its own mode) and it works great. I had tp move three jumpers to get it work, first to change its mode (see above) another to change its IRQ from 3 to 5 and finally the jumper to switch from twisted pair to cheapernet. After that I had fun with the networking support code due to a nice "little" mistake in my SLS release. In the UK this card is sold by NETLAND(UK) Limited and is sold by them as purely the "NETLAND Combo", they don't mention Commodore in the advert in case people were worried about getting an Amiga by mistake :-) I may point out that I don't work for either Commodore or NETLAND, I'm just chuffed to find a suitable card thats cheaper than the SMC Elite 16 in the UK. Jonathan -- +-------------------------------------+---------------------------------------+ | Internet: g4klx@g4klx.demon.co.uk | The three branches of Government: | | Amprnet: g4klx@g4klx.ampr.org | Money, Television and Bullshit. | | BBS: G4KLX @ GB7HMZ.GBR.EU | P.J.O'Rourke | +-------------------------------------+---------------------------------------+ ------------------------------ From: ez033209@bullwinkle.ucdavis.edu (Barracuda) Subject: Can't login with 0.99 p8 Date: Sat, 17 Apr 1993 00:11:17 GMT I've got a bit of a problem here...I just compiled the 0.99 patchlevel 8 kernel and installed it (and re-installed lilo) and when I reboot, lilo loads the kernel fine, and Linux starts up fine and gets to the Login: prompt, but thats where my problems start: no matter what I try to login as, the system waits for a second and redisplays the login prompt again. If the account that I'm trying to use has a password, I do NOT get a password: prompt. This is only a problem under my newly compiled 0.99.8 kernel...when I boot up my old 0.99.3 kernel (from lilo or from floppy) everything works fine. "halt" at the login: prompt DOES work, strangely enough...its about all that does. Suggestions? Thanks! - rob - ------------------------------ From: salnick@dejavu.spk.wa.us (RW Salnick) Subject: Linux for Amiga? Date: 16 Apr 93 12:57:54 GMT I have heard it rumored that there is such a beast... 1. is it true? 2. what kind of resources does it require on the Amiga 3. is it a complete implementation (including X)? 4. from where might I ftp a copy? Please reply via email... bob -- Like Entropy, bugs can only be created, not destroyed. RW Salnick, Spokane,WA | Home: salnick@dejavu.spk.wa.us Amiga 1000, WB 1.3 | Work: salnick@mfgsys.kaiser.spk.wa.us WA9BVE | ------------------------------ From: harry@neuron6.jpl.nasa.gov (Harry Langenbacher) Subject: Re: Diamond Video Cards Date: Sat, 17 Apr 1993 00:46:46 GMT In article rabagley@termite.occ.uc.edu (Ross Bagley) writes: >... >I am considering purchasing a machine with 2 VESA slots and the upgrade to >the Diamond Viper card is extremely attractive in this package deal. If I I ordered a Micron EISA/486dx2-66 with a Diamond Viper card, and a VLB-SCSI-2 controller - it should be a real screamer, and I'd be willing to put in a hundred or so hours in on writing drivers for these beasts if I just knew where to start. The latest computer shopper has at least a dozen clone-makers (zeos , ares, tc, eps, ariel design, comtrade, fastmicro, santron, zenon, maximus, vektron. micron) who will put a weitek P9000 (the heart of the viper) in the system for you. If we can't get help from Diamond, maybe Weitek can shed some light on the problem ??? My system should leave beautiful downtown Nampa Idaho next week, after they get their ECS EISA motherbord shipment.... I'll be back here looking for more direction and motivation when it arrives on my San Gabriel doorstep. -- H. Langenbacher, Jet Propulsion Laboratory 818-354-9513 Concurrent Processing Devices Group, FAX 818-393-4540 harry%neuron6@jpl-mil.jpl.nasa.gov Mail-stop 302-231, 4800 Oak Grove Dr, Pasadena CA 91109 USA ------------------------------ From: jkellner@galaxy.csc.calpoly.edu (John Edward Kellner) Subject: Xconfig for pvga1 & Qume 835 monitor? HPDeskjet 500 drvrs? Date: 17 Apr 93 00:36:54 GMT I've installed the SLS version of Linux on my 386 clone. Everything seems to be running O.K. except I can not get the Xconfig parameters right. (I.e., very out of sync). I would really appreciate it if any one who is running Xfree with a pvga card with 512Kb would send me a copy of their Xconfig. (esp. if you are also using a Qume 835 SVGA monitor) Also, cat file > /dev/lp1 starts my HP Deskjet 500 printing but it will print a line, spew out an empty page, etc. Are there drivers for this printer or any special configurations I need to do? Thanks for any help, jk ------------------------------ From: mancus@sweetpea.jsc.nasa.gov (Keith Mancus) Subject: Re: Compiling XV; general porting problems Date: Sat, 17 Apr 1993 00:38:14 GMT In article <36972@optima.cs.arizona.edu>, cjeffery@optima.cs.arizona.edu (Clint Jeffery) writes: >From article <1993Apr14.200018.21295@aio.jsc.nasa.gov>, by mancus@sweetpea.jsc.nasa.gov (Keith Mancus): >> I'm trying to compile xv 2.21 under Linux 0.99.6. gcc can make the >> libjpeg.a just fine, but it runs out of memory and dumps when I attempt >> to make libtiff.a. Since I have 12M RAM and a 10M swap partition, this >> must either be a gcc bug or bad options. >When I compiled it under 0.99.7A on a 16MB 486/33, the machine appeared >to "hang" while compiling files for libtiff.a. Under X, the mouse no >longer moved; outside X, I could still switch consoles, but could not >type anything at any of them. This has to be more than just gcc or >bad options. Had to hard-reset to reboot. >Fortunately, I didn't really need tiff support, so I just commented it >out of the makefile and went on my merry way... ;-) I did the same thing and had no problems. A number of people wrote to me to state that there is a gcc bug involving initializing static arrays which causes tif_fax3.c to require ~30M of memory (real and virtual). [The amount of memory claimed as required varied from message to message.] -- Keith Mancus | N5WVR | "Black powder and alcohol, when your states and cities fall, | when your back's against the wall...." -Leslie Fish | ------------------------------ From: hlu@eecs.wsu.edu (HJ Lu) Subject: Re: Problem with etc/mtab Date: Fri, 16 Apr 93 23:01:55 GMT In article <1993Apr16.221008.10189@jupiter.sun.csd.unb.ca>, u0xh@jupiter.sun.csd.unb.ca (Chris {Fig} Newton) writes: |> |> Mount doesn't seem to recreate my etc/mtab file correctly on boot |> up anymore. When I use df, it dies saying it can't read table of mounted |> file systems. I check /etc/mtab and it has in it: |> |> /dev/hda1 / |> |> The proc one is ok. Shouldn't this one read : |> |> /dev/hda1 / efs defaults |> |> When I run mount rm mtab, and run mount -av agian, it recreates it |> ok, but it is not doing it at boot time properly. My /etc/rc does remove |> /etc/mtab, and uses mount -av, but it is not doing it correctly obviously. |> I have checked my etc/fstab file, and it reads: |> |> /dev/hda1 / efs defaults |> /dev/hda2 none swap |> none /proc proc defaults |> |> I belive this to be correct, but something must be wrong |> somewhere. Anybosy have any suggestions? |> I have warned you in c.o.l. Please dig out my old article. I may miss something here. 1. Get Rick Sladkey's mount 0.99.6. It is in mount-0.99.6.tar.Z and utile-1.4.src.tar.z from sunsite or tsx-11. 2. Remove any rdev > /dev/mtab from /etc/rc. 3. List / partition in /etc/fstab. 4. Add "mount -av" in /etc/rc or somewhere like that. 5. Always provide a fs name in /etc/fstab. After rebooting, you will be fine. H.J. -- School of EECS Internet: hlu@eecs.wsu.edu Washington State University BITNET: 60935893@WSUVM1.BITNET Pullman, WA 99164 Phone: (509) 334-6315 ------------------------------ From: jmsimon@acsu.buffalo.edu (Jeffrey M. Simon) Subject: Re: flex vs. lex Date: 17 Apr 93 00:57:14 GMT acb@madhouse.demon.co.uk (Andrew Bray) writes: >TITLE: Re: flex/lex > >I have had some serious problems porting some code to flex - mostly due >to the code being very poor. > >The code assumed that lex copied matched strings to a buffer, and did >things like edit the buffer in situ, use yyless to insert extra >characters into the stream, and all sorts of nasties. > >For gruesome things like this, I would really appreciate the >availability of a more compatible lex package, though where it worked, I >would use flex - it's so much faster. > >Regards, > >Andrew Bray flex also has many additional features that lex can't touch, such as the ability to split a grammar into 'sections' using start conditions. You can use the BEGIN(xx) keyword to start working on section xx of the grammar, where only certain rules are matched against. Take a look at the flex man page for more details. J -- =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= Jeffrey M. Simon INTERNET: jmsimon@cs.buffalo.edu 112 Fairgreen BITNET: jmsimon%cs.buffalo.edu@ubvm.bitnet Amherst, NY 14228 UUCP: ...!{rutgers,uunet}!cs.buffalo.edu!jmsimon ------------------------------ From: pfr@dsun2.hmi.de (Fritsch_Wolfgang) Subject: TCP/IP ... still not working .... Date: 16 Apr 1993 13:02:48 +0200 I had stopped looking into Linux, hoping that things would 'become more stable'. Now, after a few months, I downloaded the most recent SLS distribution, but, alas, I still cannot get tcp/ip to work! This is inspite noticable 'improvements', like my Elite16 configuration is being recognized by the system on bootup (hence no need to rebuild the kernel, I suppose). I believe I followed the instructions step by step, i.e. I edited the configuration files. But when I type 'ftp ' I get 'Network unavailable'. I even don't know how to find the problem, or what details to report here to may-be get help. Notes: when running hostcvt-build I get 'could not find file for router (=add )' where the is the one that I had inserted in the configuration file as our default router (on the line that has the entry 'router'). Running hostcvt.build also destroys my entries for nameservers in the file resolv.conf, and puts in my local host's IP number. Strangely, 'shutdown -r now' leads to shutdown but for reboot I have to reset, or powerdown. Ctrl-Alt-Del does not work. Wolfgang Fritsch pfr@psun2.hmi.de ------------------------------ From: mancus@sweetpea.jsc.nasa.gov (Keith Mancus) Subject: Re: Linux on Macintrashes? (Was: Re: Linux on Amiga?) Date: Sat, 17 Apr 1993 01:16:01 GMT In article , drl@eskimo.com (James Mitchell) writes: > Long, long, long ago in my computing infancy I encouraged my > parents to buy a macintosh SE. I have never regretted the decision. > It is no longer the greatest computer in the world, but MY DAD IS MAKING > ADVERTISMENTS ON IT *W*I*T*H*O*U*T* *M*Y* *H*E*L*P!!!!!!!!! This is a > man who needed a command sheet to load telix on the much lamented family > XT. That alone is enough to make me a believer. And what possible use would such a person have for Unix? This is certainly a legitimate point, but it has NOTHING to do with Unix ports, which necessarily require a certain sophistication on the part of the user that is lacking among the vast majority of Mac users. -- Keith Mancus | N5WVR | "Black powder and alcohol, when your states and cities fall, | when your back's against the wall...." -Leslie Fish | ------------------------------ ** 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-Activists-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux) via: Internet: Linux-Activists@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 tupac-amaru.informatik.rwth-aachen.de pub/msdos/replace The current version of Linux is 0.99pl6 released on February 23, 1993 End of Linux-Activists Digest ****************************** t extra >characters into the streadigest863 644 36676 74430 47645 5366137105 5774 From: Digestifier To: Linux-Activists@senator-bedfellow.mit.edu Reply-To: Linux-Activists@senator-bedfellow.mit.edu Date: Sat, 17 Apr 93 01:00:36 EDT Subject: Linux-Activists Digest #863 Linux-Activists Digest #863, Volume #4 Sat, 17 Apr 93 01:00:36 EDT Contents: Re: working crond package uploaded to sunsite (Marc G Fournier) [Q] ASUS Motherboards? (Bob Craycroft x629) [Q] Is MSDos 6.0 compatable with Linux ? (Jon Scheer) Passwd Problems! HELP!!! (Sean Gum) Xwindow & bott (Tsun-Ming Kao) where to get 0.99p8??? (Tein H. Yuan) Lilo and OS/2 without Boot Manager (DHALIWAL BIKRAM SINGH) Re: Summary of Linux vs. 386BSD vs. Commercial Unixes (Brandon S. Allbery) Re: Automated Linux PowerDown (Marc Christensen) EISA: Anybody working on it? (pmkierst@flash.LakeheadU.Ca) Re: Missing Linux Include Files ... (Andreas Zisowsky) Linux & GUS (Carlo James Calica) Re: Access control lists and Linux (H. Peter Anvin N9ITP) Re: NFS-server for Linux??? (Hwang Gwan-Hwan) mke2fs: "out of memory" (Sherman Brown) Re: where to get 0.99p8??? (Chen Wei Chou) finger....help.. (Jongyoon Lee) info request: RAWRITE (dionysis@u.washington.edu) linux shared libraries often incompatible (Aaron Michael Cohen) Re: Automated Linux PowerDown (Keith Barrett) ---------------------------------------------------------------------------- From: marcf@nexus.yorku.ca (Marc G Fournier) Subject: Re: working crond package uploaded to sunsite Date: Sat, 17 Apr 1993 01:22:01 GMT gtaylor@jade.tufts.edu (Grant Taylor) writes: >I found that the crond in the vixie-cron package on sunsite & tsx was >sufficiently old that it no longer works, so i extracted working >binaries from the SLS and found REAL manpages elsewhere to get it >going on my system. Since this took me some time (finding a specific >binary in the SLS in not easy) I figured it would be good to share my >results: What version of cron is the one in SLS based on? I'd be interested in getting the sources, but not the binaries...there is a bug with the one in SLS that locks up 'mail' because it uses a -d qualifier for it. I ended up getting rid of /bin/mail and linking elm to it, but it still exhibits the same symptoms. I ended up recompiling vixie-cron the last time and got rid of the -d, and it works great. Isn't SLS just using vixie? marc ------------------------------ From: craycrof@ruchbah.rtp.semi.harris.com (Bob Craycroft x629) Crossposted-To: comp.sys.ibm.pc.hardware,comp.unix.pc-clone.32bit Subject: [Q] ASUS Motherboards? Date: 16 Apr 93 19:39:14 GMT I'm considering the purchase of a 486DX-33 VLB system to run linux. The system has an ASUS-brand motherboard. Anyone have any comments on ASUS motherboards? Thanks, -- Bob Craycroft | craycrof@rtp.semi.harris.com Systems Analyst | Phone: (919) 549-3629 Harris Semiconductor - RTP, NC USA | ------------------------------ From: scheer@rtsg.mot.com (Jon Scheer) Subject: [Q] Is MSDos 6.0 compatable with Linux ? Date: Fri, 16 Apr 1993 15:14:31 GMT Greetings, I'll start off by saying that I don't think this site (rtsg.mot.com) is getting all of the c.o.l.a and c.o.l messages. Apparently sometime recently someone asked if MSDos 6.0 and Linux were compatable. Yesterday I saw a couple (nasty) responses to the question, but I never saw the original post. I just checked today, and I didn't see any discussion of the question. So, I'll ask the questions this time: - Is MSDos 6.0 fully compatable with Linux? - If I run the new MSDos 6.0 'dblspace' command, will Linux still be able to mount, use and run MSDos programs ? - Has anybody tried this, or am I going to be the guinea pig? :-? :-) I'll keep watching c.o.l, but feel free to email me your response as well. Thanks. ( Please route MSDos flames to /dev/null ) Jon 4/16/93 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Work: scheer@motcid.rtsg.mot.com or motcid!scheer@UUNET.UU.NET Home: wombat@outback.chi.il.us or outback!wombat@UUNET.UU.NET Just because you're paranoid doesn't mean they AREN'T after you. -- ------------------------------ From: sean@whiting.mcs.com (Sean Gum) Subject: Passwd Problems! HELP!!! Date: Fri, 16 Apr 1993 13:00:29 GMT Help! I am now having a problem with the passwd program. Whenever a user, other than root, tries to change their passwd, they get the following: /home/sean/passwd Changing passwd for sean Old Password: Enter the new password (minimum of 5 characters) Please use a combination of upper and lower case letters and numbers. New Password: Re-enter new password: Cannot change ID to root. And the change fails. Could someone explain to me what the problem is? Any help is greatly appreciated! Thanks! Sean@Whiting.Mcs.Com ------------------------------ Crossposted-To: comp.windows.x.i386unix From: tkao@iastate.edu (Tsun-Ming Kao) Subject: Xwindow & bott Date: Sat, 17 Apr 1993 01:29:12 GMT Could anyone help me with the following problem? 1. I have OS/2 2.0, DOS 5.0, and I just install Linux. I followed the steps shown one the FAQ, but I still can't boot Linux from my Hard Drive. I put Dos and Linux at the primary drive,(also OS/2 Boot Manager) and OS/2 at the logical drive. When I tried to boot Linux from OS/2 Boot Manager, the error message show: "Selected drive not formated, please try another one." I don't where I went wrong; does anyone successfully boot Linux under OS/2 Boot Manager? BTW, When I use /etc/lilo/lilo -c -b /dev/hda4 -v -v /image the message shows that I didn't have a valid LILO signature. any clue about this? I tried /etc/lilo/lilo -c -b -i /etc/lilo/boot.b /dev/hda4 -v -v /Image It still doesn't work. 2. I got my X386 window and a working mouse, but there is nothing in the window, so I hit the Ctrl-Alt-BackSpace, and the message shown as follow: cmdtool: command not found /u/openwin/bin/olwm: No such file or directory exec: /u/openwin/bin/olwm: can not execute: no such file or directory waiting for X server to shutdown shelltool: command not found xinit: Interrupted system call (errno 4): client error Any help will be appreciated Daniel tkao@iastate.edu ------------------------------ From: tein@hawk.cc.as.edu.tw (Tein H. Yuan) Subject: where to get 0.99p8??? Date: Sat, 17 Apr 1993 02:30:39 GMT Hi there Currently, I am using 0.99p6. I would like know where to get 0.99p8. p.s. I cannot ftp files from "nic.funet.fi", so ... ------------------------------ From: dhaliwa@ecf.toronto.edu (DHALIWAL BIKRAM SINGH) Subject: Lilo and OS/2 without Boot Manager Date: Sat, 17 Apr 1993 02:36:32 GMT I just want to make sure before I use Lilo on my machine, but can I use it to boot by default into OS/2, asking only to boot to Linux? this time I don't have Boot Manager installed, so I don't see as many problems, but I did like the fact I could boot from the Boot Manager instead. Is there anybody who has done this?? (there must be) -- .......................... Bikram Dhaliwal PH: 416-351-9660 dhaliwa@ecf.toronto.edu ------------------------------ Crossposted-To: comp.os.386bsd.questions From: bsa@kf8nh.wariat.org (Brandon S. Allbery) Subject: Re: Summary of Linux vs. 386BSD vs. Commercial Unixes Date: Sat, 17 Apr 1993 01:44:53 GMT WARNING: Unix religious war! Followups are sent elsewhere! In article <1993Apr16.162749.29973@sifon.cc.mcgill.ca> storm@cs.mcgill.ca (Marc Wandschneider) writes: >In article <1993Apr15.225354.18654@samba.oit.unc.edu> Brandon.Vanevery@launchpad.unc.edu (Brandon Vanevery) writes: >>SCO Unix has been recommended to me as "the best" of the commercial >>systems, by a few folks. > > I had the pleasure of using this this product last summer, and >have slowly come to the conclusion that anybody who finds this a superior >product has either never used ANY other UNIX before, is high, or finds >AIX a close second. Thank you. I'd been wondering what drugs his informant was doing :-) Mind you, my background is almost exclusively System V --- and the SVR3.1 systems I worked with for several years before encountering SCO "Unix" were far, far better than that purported SVR3.2. Not to mention far more compatible with both System V and (to a limited extent: no networking with it) BSD software. (Unfortunately, it predated POSIX, so there was no compatibility there.) Nicely packaged, expensive spam is still spam, folks.... :-) > Never before have I used a UNIX so lacking in features, so annoying >to use (even for a SVR? system), and so difficult to get the hang of. As far as "annoying to use", that's clearly a matter of one's experience. I started with System III and moved on to System V when it came out; SunOS4 has been a real "trip", and I'm looking forward to our installing Solaris 2.1 so the system will behave as God intended :-) :-) :-) [and I say again for the humor-impaired: :-) ] Let's leave the *ix religious wars to the small- brained, shall we? ++Brandon -- Brandon S. Allbery bsa@kf8nh.wariat.org It's not too late to turn back from the "Gates" of Hell... Linux: the FREE 32-bit operating system, available NOW. Why waaaaaait for NT? ------------------------------ From: mgc@unislc.slc.unisys.com (Marc Christensen) Subject: Re: Automated Linux PowerDown Date: Fri, 16 Apr 1993 16:42:24 GMT Speaking of PC power consumption. I wondered about mine so I measured it. 80485-50 with 16Meg ram and 3.5" IDE drive pulls 65W. this is in a 6 year old case amd power supply, the newer supplies may be better or worse. I would leave the monitor off. laser printers tend to pull a huge amount of power. My machine at home would cost me $3.40/month. Marc Christensen ------------------------------ From: pmkierst@flash.LakeheadU.Ca Subject: EISA: Anybody working on it? Date: 16 Apr 1993 19:33:12 GMT I have been considering my OS future, and Linux looks like _the_ OS to go with....been using it for a couple of months, and enjoying it throughly. Buuut...has anybody been working on EISA support? The 16 Mb ISA barrier for bus operations is a real killer, not to mention the throughput. My primary concern is the 16Mb barrier. In short, is anybody working on some support for EISA cards? In particular, I am looking to HD SCSI support for EISA... Paul Kierstead pmkierst@flash.lakeheadu.ca ------------------------------ From: zisi@marie.physik.tu-berlin.de (Andreas Zisowsky) Subject: Re: Missing Linux Include Files ... Date: 16 Apr 1993 19:39:37 GMT In <1993Apr8.123815.28567@ccd.harris.com> nam@controls.ccd.harris.com (Noahal A. Mundt) writes: >When I try to compile programs using gcc under Linux 0.99 I get error >messages indicating ../linux/*.h cannot be found. Looking in >/usr/include/linux, I do not find any include files. I loading the "c" There probably some missing symbolic links. Check out the README file in /usr/src/linux for more information. Ciao. Andreas ------------------------------ From: calica@cae.wisc.edu (Carlo James Calica) Subject: Linux & GUS Date: 16 Apr 93 14:13:10 CDT I know that Linux has drivers for the Sound Blaster. Do drivers exist for the Gravis ultraSound? I remember a few people wanted to write one a few months ago. How has that effort progressed. Thanks. -- /-------------------------------+----------------------------------\ | Carlo J. Calica | Proud user of OS/2 | | calica@cae.wisc.edu | Dittos from the People's | \ University of Wisconsin | Republic of Madison / ------------------------------ From: hpa@merle.acns.nwu.edu (H. Peter Anvin N9ITP) Subject: Re: Access control lists and Linux Reply-To: hpa@nwu.edu (H. Peter Anvin) Date: Sat, 17 Apr 1993 02:37:33 GMT In article <1rj7ug$89r@walt.ee.pdx.edu> of comp.os.linux, gary@acacia (Gary Moyer) writes: > steve@rama.demon.co.uk (Steve Entwistle) writes: > : One possibility would be to develop a generalised security > : package, such as RACF, used on IBM Mainframe systems. This system uses a > : central database in which all the security information for various > : resources is stored, e.g. Files, Users, Terminals etc. > : > : A sample entry for a File resource might be :- > : > : > : Filename : /etc/foobar > : Default Access : NONE > : > : User fred, Access = READ > : Group Wheel, Access = UPDATE > : > : > : Rather than changing the filesystem code, you could insert a > : call to the resource checking routine in all the system calls that access > : the resource you want to protect (in this case, all the system calls concerned > : with file accesses). If there is no entry in the database for a particular > : file, I guess you would then just use the normal file permission bits. > : > : The advantages of doing it this way is that not only is it filesystem > : independent, but it is also readily extended to protecting other resources. > > Thats an interesting idea. I can see 2 major drawbacks: > 1) what protection scheme would be used for this centralized data base? > 2) if it were stored on a secondary storage device: what would guarantee > security to it ? > > Thoughts? > Well, here is a suggestion: Let each filesystem be mounted with or without access daemon support (with for flexible security, without for speed). If the kernel detects a permission mismatch, it will call up (through some standard interface) `accessd' which (of course) runs as root. It informs accessd of uid, gid(s) and filename, and accessd returns the permissions approved to the kernel. Alternatively `accessd' could be a part of the kernel, but it *would* be bloat.... I would suggest the accessd database is simply a (collection of) file(s) to which only root have access (by the traditional permissiona bits). Of course any such proposal must be carefully examined, especially the interface between the access daemon and the kernel (/dev/access 600 root maybe). /hpa -- INTERNET: hpa@nwu.edu FINGER: hpa@eecs.nwu.edu BITNET: HPA@NUACC IBM MAIL: 36073 at IBMX400 HAM RADIO: N9ITP, SM4TKN NeXTMAIL: hpa@speedy.acns.nwu.edu Heja Sverige friskt humör! EG väntar utanför! :-) ------------------------------ From: ghhwang@csie.nctu.edu.tw (Hwang Gwan-Hwan) Subject: Re: NFS-server for Linux??? Date: Sat, 17 Apr 1993 03:06:12 GMT Dear friend, Surely, if you install your PC with Linux, it can provide it to be a NFS client or server. ghhwang from R.O.C ------------------------------ From: faldor@seq.uncwil.edu (Sherman Brown) Subject: mke2fs: "out of memory" Date: Fri, 16 Apr 1993 20:13:58 GMT I'm at my wits end... I've tried all I can think of to get my extended partition made up with a filesystem. I have a 386/25MHz with 4Mb RAM. I have used mke2fs -c to make my 15Meg root partition and mkswap to make my 20Meg swap partition and used swapon to allow use of of the swap space. When I try to use mke2fs to make my 85Meg partition for /usr and it gives me an out of memory error. It doesn't hang thankfully though... Any help would be greatly appreciated. thanx in advance, Faldor .---------.---------------------------------------------------------. | (__) | | | w \@@/ | Daemons do it with pointers. }:-) | | `/v/-e | | | _/ \_ | Life sucks and then you're a vampire. | ._________._________________________________________________________. | University of North Carolina at Wilmington | | faldor@seq.uncwil.edu | |-------------------------------------------------------------------| | Alien Invasion Division : Integration/Substitution Group. (ISG) | .___________________________________________________________________. ------------------------------ From: u7824501@cc2.nctu.edu.tw (Chen Wei Chou) Subject: Re: where to get 0.99p8??? Date: Sat, 17 Apr 1993 03:37:52 GMT Tein H. Yuan (tein@hawk.cc.as.edu.tw) wrote: : Hi there : Currently, I am using 0.99p6. I would like know where to get 0.99p8. : p.s. I cannot ftp files from "nic.funet.fi", so .. try nctuccca.edu.tw. source/system/. ------------------------------ From: mr2@netcom.com (Jongyoon Lee) Subject: finger....help.. Date: Fri, 16 Apr 1993 20:04:55 GMT Hi netters! How can I fix the finger to show last login information? Mine always says never logged in although it's not true. I tried to change permission to /usr/adm/lastlog but didn't help. Thanks in advance! Jongyoon ------------------------------ From: dionysis@u.washington.edu Subject: info request: RAWRITE Date: 16 Apr 1993 20:29:21 GMT Reply-To: dionysis@u.washington.edu I am looking for the RAWRITE program, yes I have looked thru the FAQ, no location listed in there pls mail, as I do not read this group ------------------------------ From: acohen@world.std.com (Aaron Michael Cohen) Subject: linux shared libraries often incompatible Date: Fri, 16 Apr 1993 20:51:50 GMT Linuxers: I have had some problems recently in trying to use binaries that I have downloaded from tsx-11 and sunsite. This is almost always due to incompatible shared library versions. Now, while I think that shared libraries are great for saving space, distributing executables linked with them without source code isn't working for me. It probably isn't working for some others too. My Linux installation is based on an SLS release about 2 months old, and already there are lots of apps that I can't use. So I'd like to propose a solution. When you post an original app or a port to the net either (1) include the sources and a makefile, or (2) distribute a set of unlinked object files and include the link command in the README.linux file. I think that this would work in most cases. And I sure would be able to use more software. Please give me some feedback on this. Otherwise, I think Linux is the only operating system worth having on your PC (and I have used DOS, OS/2, Windows, Coherent, and Desqview/X and Linux is the only one that has lasted on my hard disk more than 3 weeks). I'm getting really snappy performance running a 40MHz AMD 386 with 8 Megs RAM and an 8 Meg swap partition. I've got X11 with the openlook window manager, and I feel like I'm using the SUNs back at work (well, almost :-)). Count me as another "satisfied customer". Aaron Michael Cohen acohen@world.std.com ------------------------------ From: barrett@pamsrc.enet.dec.com (Keith Barrett) Subject: Re: Automated Linux PowerDown Reply-To: barrett@pamsrc.enet.dec.com Date: Fri, 16 Apr 1993 20:00:57 GMT > Now, for my purposes, linux has only one disadvantage, as compared to dos. >I'd like to address >it: with DOS, you can generally dump the computer's power any time that you see >that dos prompt. >With any unix, you have to be more careful... Why is this important? You can still do this ... Why don't you simply create a crontab entry that shuts your system down at (let's say) 5:00, and set your timer to shut the computer off at 5:30? cron should be able to run shutdown fine. Keith Barrett ------------------------------ ** 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-Activists-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux) via: Internet: Linux-Activists@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 tupac-amaru.informatik.rwth-aachen.de pub/msdos/replace The current version of Linux is 0.99pl6 released on February 23, 1993 End of Linux-Activists Digest ****************************** tion/Substitution Group. (ISG) | .________________________________________________________digest864 644 36676 74430 53664 5366137105 5773 From: Digestifier To: Linux-Activists@senator-bedfellow.mit.edu Reply-To: Linux-Activists@senator-bedfellow.mit.edu Date: Sat, 17 Apr 93 04:15:11 EDT Subject: Linux-Activists Digest #864 Linux-Activists Digest #864, Volume #4 Sat, 17 Apr 93 04:15:11 EDT Contents: Re: Gateway Nomad Field Mouse (Iain Lea) Re: PS/2 style mouse not supported?? (J. Scott Farrow) Thanks... (Hunyue Yau) SOFTWARE WARS (Bill Baker) "640x400" mode timings (Jeff Stamerjohn) Excellent, up and running LINUX!! (z_delgadoap@ccsvax.sfasu.edu) Re: [HELP] XWindows 8 MEG RAM req??? (Mark Cosham) Re: TCP/IP ... still not working .... (Kenneth Hamer,,,) Re: Access control lists and Linux (Gary Moyer) [A] difficulties with libm (was Re: How to GCC part 2) (Hal N. Brooks) Re: Questions about virtual mem. (jonathan ward) Re: X Windows and TVGA8900C (jonathan ward) Re: Summary of Linux vs. 386BSD vs. Commercial Unixes (Charles Hannum) ---------------------------------------------------------------------------- From: Iain.Lea@anl433.erlm.siemens.de (Iain Lea) Subject: Re: Gateway Nomad Field Mouse Date: 16 Apr 93 09:12:20 GMT Paul Fishwick (fishwick@fish.cis.ufl.edu) wrote: : : I am looking into putting Linux on a Gateway Nomad 486 DX2/50 and before I : do, I'd like to know whether anyone else has put Linux on this machine. : I checked the compatibility chart but did not see a "Nomad" or : "Texas Instruments" machine there (the Gateway machines being repackaged : TI machines). Specifically, the Nomad comes with a "field mouse" (whatever : that is :). Does the field mouse emulate the Microsoft bus or serial mouse? I have 0.99p6 running on a TI4000 DX2/50. Tried installing X but could not get the mouse to work but then again I only tried for a short while. Iain -- NAMES Iain Lea iain.lea@erlm.siemens.de SNAIL Siemens AG, ANL A433SZ, Gruendlacher Str. 248, 8510 Fuerth, Germany. PHONE +49-911-3089-407 (work) +49-911-331963 (home) +49-911-3089-290 (FAX) ------------------------------ From: farrow@ucsu.Colorado.EDU (J. Scott Farrow) Subject: Re: PS/2 style mouse not supported?? Date: Fri, 16 Apr 1993 20:41:19 GMT kkc@stat.nursing.arizona.edu (Kee Chua) writes: >Get the "mconv.c" file and you should be ok. The instructions are in that file. >Kee Chua Better yet, get the latest version of XFree86 which supports PS/2 mice and don't worry about it all. Scott -- ========================================================================= J. Scott Farrow - Student SysAdmin/Programmer, University of Colorado Computing & Network Services, Boulder, Colorado, USA, phone:(303)492-4428 Internet: farrow@spot.colorado.edu "Linux Users Do It For Free" ------------------------------ From: huny@cco.caltech.edu (Hunyue Yau) Subject: Thanks... Date: 17 Apr 1993 05:47:19 GMT I want thank the various people that jumped up to help. I was going to send private replies, but I lost all of their address when I hung the POP server/client that I was reading the mail on. In case anyone is interested: My telnet in and out's were hanging and I could figure out why. The suggestions/ideas that I got were: 1. Upgrade to the 4.3.3 libs. I tried this and it seems to help a little(less frequent hangs). 2. inetd is hanging, so the only option short of rewriting/fixing it my self is to wait. 3. The kernel cannot handle fragmentation. My options are the same as #2. Those ideas are helpful, but my system still hangs. However, it seems that I can telnet out properly and remain connected only a freshly booted system. Telneting in still hangs, but it takes a lot longer now. FWIW: I notice that there trying to get a listing of the TCP sockets resulted in an error message of: too many sockets for netinfo to handle. This is from the proc-netinfo patch. It would appear that the kernel is overload from the list of 'stale/unfreed' sockets. Any ideas/suggestions? Again, thanks. Hun huny@cco.caltech.edu ------------------------------ From: baker@spk.hp.com (Bill Baker) Subject: SOFTWARE WARS Date: Fri, 16 Apr 1993 21:48:02 GMT I received this through e-mail (original author unknown) and I thought everyone would enjoy the humor...and decry the underlying truth in it... Monday, 10 AM -- Chicago, Illinois -- Start-up software developer Cuisine International announced CUISINENET, the first working program to seamlessly integrate word and food processing. Called a breakthrough for small restaurants and snack bars, Cuisine Chairman Mark Meigs confidently predicted sales of thousands of copies with shipments soon to begin. Monday, 4 PM -- New York -- Cuisine International shares closed sharply higher on announcement of new CUISINENET product. Tuesday, 9 AM -- Redmond, Washington -- Microsoft Chairman William H. Gates, III announced that Microsoft Food for Windows would soon enter beta testing. Gates described the product as the first of a projected family of products to include Food for Windows, designed for small commercial dining establishments; Personal Food for Windows, designed for home kitchens; Portable Food for Windows, designed for lunchboxes; and, of course, at the high end, Food for Windows NC (Nouvelle Cuisine) designed for large institutional dining rooms. Asked by a reporter about CUISINENET, Gates said that he had never heard of the product, but was not surprised by it, because the software business is highly competitive, and Microsoft has to compete on the merits with many strong competitors, as the FTC had recently concluded. Tuesday, 3 PM -- Chicago, Illinois -- An angry Mark Meigs showed reporters a copy of the nondisclosure agreement signed by Bill Gates, under which Cuisine International had informed Microsoft a year earlier about plans for CUISINENET. Meigs said that in hindsight, he should never have signed the agreement, as the only thing he learned from Microsoft was that Gates was considering making changing to Windows. Wednesday, 9 AM -- Redmond, Washington -- Microsoft Chairman Bill Gates announced that Microsoft would soon publish specifications for the Windows Open Kitchen Architecture (WOKA), a series of design specifications to permit manufacturers of toasters, ranges, and other kitchen appliances to integrate their products into the forthcoming Microsoft Food for Windows line. Asked about reports of a nondisclosure agreement with Cuisine International for a similar product, Gates said that the other product was really at most a niche product, and would probably have less functionality than the food-related features that Microsoft would be building into the new Unsaturated FAT File System which would be part of DOS 7.0. Gates said that he doubted there would be much interest in a dead-end solution that would not be able to keep up to date with advances in WOKA. Gates added that over 11,000 manufacturers of kitchen appliances were already having serious discussions with Microsoft about WOKA, and that he expected almost all important eaters of food to standardize on the WOKA environment. Wednesday, 10 AM -- Redmond, Washington -- Microsoft Chairman Bill Gates announced that he would be giving the keynote speech at the American Bakers annual convention on "Nutrition at Your Fingertips." Gates played down speculation that he would use the Bakers convention to introduce Microsoft Food for Windows, saying only that alpha testing was proceeding ahead of schedule, and the product would be shipped when it was ready. Wednesday, 11 AM -- Redmond, Washington -- Microsoft Corporation announced that its Chairman, William H. Gates, III, had made a donation of over $250 of personal funds to the Cordon Bleu to begin an endowment fund for the Bill Gates Professorship of Advanced Cookery. The famous French cooking school confirmed that it had agreed to be a beta site for the much discussed Food for Windows application sweet. Thursday, 9 AM -- New York -- PCWeek Magazine reported in a copyrighted story that it had obtained a copy of correspondence from Microsoft to Cuisine International, demanding that the small developer of kitchen software cease using the Cuisine name, as it infringes on the trademark for Microsoft Food for Windows NC. Microsoft added that Chairman Mark Meigs would also have to change his own name as Mark infringed a copyright on the Windows Edit menu, Meigs infringed the trademark on Meigs Field in Microsoft Flight Simulator, and Chairman infringed the trademark on Bill Gates's title which he had acquired with personal funds from Mao's estate. Also, Microsoft advised that while the company did not actually have to move out of Chicago, use of the name on press releases infringed a trademark on Windows 4.0 Thursday, 4 PM -- New York -- Cuisine International stock closed at 0-bid, 1/16-asked. Friday, 9 AM -- ? -- An anonymous spokesman for an unnamed Midwestern software developer announced the discontinuation of operations. Undescribed legal problems were cited as the reason. Others speculated that a failure to appreciate the competitive nature of the software business may have led to the company's sudden collapse. Monday, 9 AM -- Microsoft Internal Mail From: billg To: mikem Re: Food Program Please see if you can reassign one of the 3,000 engineers from the OS/2 virus development project to do a feasibility study on a food-related program. Not sure what it would do. Low priority. ------------------------------ From: jstamerj@ux4.cso.uiuc.edu (Jeff Stamerjohn) Subject: "640x400" mode timings Date: Sat, 17 Apr 1993 04:00:30 GMT I'm try to get color X working, but my graphics card only has 256k, and I get the not enough memory error for a 640x480 mode. I'm sure my card supports "640x400" mode, but I'm haveing trouble figureing out the timings to put in ModeDB. Does anyone have any idea how to calculate these, or (even better) does anyone know what they are? Thanks alot, -- Jeff Stamerjohn BTW, I have a cirrus graphics card, and the X driver for that card, if that makes any difference... ------------------------------ From: z_delgadoap@ccsvax.sfasu.edu Subject: Excellent, up and running LINUX!! Date: 16 Apr 93 23:02:51 CST Excellent!!! I am a new user to Linux, in fact I just got it up and running today!! I have performed a base system + X11 SLS install from SLS 1.0 available at TSX-11... I have just a couple of questions that I have not found in any of the docs... 1. When I finally got X configured, for some reason it looked in a dir /u/openwin/bin for all of the openwin stuff, my install did not put it uner /u, but under /usr since I gather that /usr is the right place, how can I fix the X* files to look in the right place?? 2. What is currently available from the install to run on X11?????? I am still finding all kinds of new stuff... Thnks to all who have mae this new joy possible!!!! Andy Delgado | Mac Technical Support - LINC Services __|__ Stephen F. Austin State University | Nacogdoches, Texas | z_delgadoap@ccsvax.sfasu.edu | taco@calvin.sfasu.edu ------------------------------ From: cosham@yoyo.cc.monash.edu.au (Mark Cosham) Subject: Re: [HELP] XWindows 8 MEG RAM req??? Date: Sat, 17 Apr 1993 03:58:05 GMT In <734876801.F00076@remote.halcyon.com> Chris.Dreke@f460.n3625.z1.fidonet.org (Chris Dreke) writes: >Well, I am able to run X windows on 4 meg + swap space, but it is too >slow to do much more than play tetris. If you want to run multiple >applications, you will need more memory. > * Origin: Vantage Point BBS (1:3625/460.0) I'm running X windows on 5 meg + swap space (8 meg) and although it is slow, it is still quite usable. I've got a 486DX-33. Mark Cosham -- Practice random kindness and senseless acts of beauty. ===========> <=======> <=======> <=======> <=======> <=======> <=========== Mark Cosham Monash University, Melbourne, Victoria, Australia cosham@yoyo.cc.monash.edu.au ------------------------------ From: khamer@sumax.seattleu.edu (Kenneth Hamer,,,) Subject: Re: TCP/IP ... still not working .... Date: 16 Apr 1993 22:05:45 -0700 From article <1qm3oo$bko@dsun2.hmi.de>, by pfr@dsun2.hmi.de (Fritsch_Wolfgang): > > I had stopped looking into Linux, hoping that things would 'become more > stable'. Now, after a few months, I downloaded the most recent SLS > distribution, but, alas, I still cannot get tcp/ip to work! > I'll second this. I have the most recent (as far as I know) SLS (0.99pl6), and have been tying for a couple of days to get the loopback stuff working so I can learn about the networking stuff. I have read the FAQ, the NET-FAQ, and all other documentation I can find. I have read some stuff a friend has on tcp/ip networking TCP/IP _should_ be in my kernel, as I recompiled it and marked it to be enabled. But I can't get anything to work after following the instructions in the net-faq. if I try to telnet to 127.0.0.1 (which I thought was myself) I get back "tcp/telnet: unknown service" I get similar responses for ftp, rlogin, and rsh. What am I doing wrong? I apologize if I'm just being stupid, but I'm doing the best that I can... -Ken -- .............................................................................. -Kenneth Hamer khamer@sumax.seattleu.edu ------------------------------ From: gary@acacia (Gary Moyer) Subject: Re: Access control lists and Linux Date: 17 Apr 93 04:58:16 GMT hpa@merle.acns.nwu.edu (H. Peter Anvin N9ITP) writes: : > Thats an interesting idea. I can see 2 major drawbacks: : > 1) what protection scheme would be used for this centralized data base? : > 2) if it were stored on a secondary storage device: what would guarantee : > security to it ? : > : > Thoughts? : : Well, here is a suggestion: : : Let each filesystem be mounted with or without access daemon support : (with for flexible security, without for speed). If the kernel : detects a permission mismatch, it will call up (through some standard : interface) `accessd' which (of course) runs as root. It informs : accessd of uid, gid(s) and filename, and accessd returns the : permissions approved to the kernel. Alternatively `accessd' could be : a part of the kernel, but it *would* be bloat.... : : I would suggest the accessd database is simply a (collection of) : file(s) to which only root have access (by the traditional permissiona : bits). We're back at the square where we started from. The main, and most important advantage I can see, to ACL's is that you have an _independent_ access level that cannot be compromised by the lower level access levels (i.e. unix file permissions). -g.m. ------------------------------ From: hal@pollux.cs.uga.edu (Hal N. Brooks) Subject: [A] difficulties with libm (was Re: How to GCC part 2) Reply-To: hal@pollux.cs.uga.edu (Hal N. Brooks) Date: Sat, 17 Apr 1993 06:15:31 GMT In article <1qm84u$jj1@obelix.uni-muenster.de> zy0004@obelix.uni-muenster.de (Martin Rehwald) writes: >Hello out there. >Me again. >Yesterday I asked for help linking the math-libary using GCC. >Many told me to use '-lm' instead of '-l m'. I did this before - >and did not detect any difference in GCC's behaviour. >Using AIX-GCC on rs6000 both works fine. With me none at all. >By the way there is another funny observation: > adding -l m.a causes the error: libm.a not found. > this should be 'libm.a.a' not found, no? >If you need further information: > I'm using GCC-2.3.3, lib4.3.2, Linux-0.99.7 > >OK. >Waiting for more ideas, > MR > I didn't see your first post, so I'm assuming a lot about what your problem is. Certainly, -lm (at the end of the gcc command line) is the thing to do. That should allow successful compilation. If you subsequently get a runtime error message, then you need to check to see where the symlink /lib/libm.so.4 is pointing. If you're using libc4.3.2, then it should be made to point to libm.so.4.3.2. This works for me anyway. Hopefully HJ won't have to correct me. Others have posted questions about libm, and his response is usually you need to get libc4.3.x. But I don't think HJ's instructions say anything about making the libm.so.4 ---> libm.so.4.3.2 connection. He probably expects more from some of us than we're capable of delivering! I'm going out on a limb here, so if I'm wrong HJ, please be gentle with me. ============================================================================= Hal N. Brooks Voice: (706) 546-7792 Internet: hal@pollux.cs.uga.edu ============================================================================= ------------------------------ From: jmward@ucrengr (jonathan ward) Subject: Re: Questions about virtual mem. Date: 17 Apr 93 02:04:52 GMT Peter A. Schwenk (schwenk@fred.cis.udel.edu) wrote: : I have a few questions about Linux. I haven't read a FAQ, so bear with me. : Does Linux support virtual memory, or once system RAM is used up, crash city? Yes, Linux does support virtual memory, in the form of a swap partition. It can be made as large as you want, although once you get around 12 megs, anything more is a waste of disk space: it won't be used. I run my system on four megs with a 12 meg swap partition, and it will run X - very slowly, but I haven't run out of memory yet( X usually likes eight.). : How stable is Linux? Is it only good for playing with? I haven't crashed Linux yet - I'm using the SLS 0.99pl6 release, and it's very stable. With the ever expanding amount of _FREE_ applications that are becoming available, you can do some serious work with it. It supports POSIX standards mostly, which means that many apps written for other compliant platforms often need only a recompile or a minor change to run on Linux. It also has support for Sun's NFS, and with a generous hard drive and memory, could conceivably be used for a file server on a cross-platform network - and for a lot lower hardware cost than a similarly equipped workstation. : Where is Linux available? Try sunsite.unc.edu or tsx-11.mit.edu. A few words of advice: while you're on these site, pick up a FAQ before posting again here - I don't have a problem with answering you, but many people on this group will flame you for posting questions that are answered in the FAQ; this forum is really for posting questions and solutions to problems not listed in the FAQ. If you are unaware of where to get the FAQ, it is posted every so often in the comp.os.linux.announce forum. After all, that's the whole point of the FAQ - to answer starting questions so that bandwidth isn't wasted on needless common questions(this is a _VERY_ busy group. :-) Hope I could help. -Jonathan Ward University of California, Riverside ========== Email to: drdrums@watserv.ucr.edu drdrums@csld.ucr.edu jmward@cs.ucr.edu NOTE: Please use the above listed addresses, NOT the reply header - I've been told that it spits out invalid addresses. "Linux and X - the GNU cure for Windoze NoT" ------------------------------ From: jmward@ucrengr (jonathan ward) Subject: Re: X Windows and TVGA8900C Date: 17 Apr 93 03:01:37 GMT JJ Lay (csjjlay@knuth.mtsu.edu) wrote: : I have the newest SLS release of Linux running : on a 386DX with 8Mb of RAM, a TVGA 8900C chipset : and Microsoft 2button serial mouse on /dev/ttyS0 : When I start X up it gives me a bunch of "wavey" : lines. I followed the hints given by someone to : use the VESA timings. Any other suggestions? I have had similar problems - If you have SLS release 0.99pl6, switch to the /usr/X386/lib/X11 directory. The file Xconfig.sample contains the standard VESA timings - cut and paste the timings tables into your current Xconfig file, replacing the existing one. These timings work with the tvga8900c. Next make sure that you have the chipset line in Xconfig set to "tvga8900c" - it won't work otherwise. Finally make sure that the modes line reads "1024x768" "800x600" "640x480" - or something in that order(it determines the order and resolutions that the card cycles through with the + and - keys). Note that all of these I use on a 1 meg card - if your card has less that 1 meg, it will probably support lower resolutions only, so be careful!! If you have a fixed frequency monitor, be doubly careful - improper timings can burn it out! If you have a multisync, you don't have to worry as much - it's very hard to destroy a multisync monitor with improper timings. One more note: there are two locations in the Xconfig file: 1 for the vga256 driver and one for vga2(mono) - you should match up the specs mentioned above for both. The chipset line need only by stated once. Hope this helps! -Jonathan Ward University of California, Riverside ========== Email to: drdrums@watserv.ucr.edu drdrums@csld.ucr.edu jmward@cs.ucr.edu Reply to above addresses only. "Linux and X - the GNU cure for Windoze NoT" ------------------------------ From: mycroft@hal.gnu.ai.mit.edu (Charles Hannum) Crossposted-To: comp.os.386bsd.questions Subject: Re: Summary of Linux vs. 386BSD vs. Commercial Unixes Date: 17 Apr 1993 00:22:18 -0400 In article <1993Apr15.225354.18654@samba.oit.unc.edu> Brandon.Vanevery@launchpad.unc.edu (Brandon Vanevery) writes: > > 386BSD has better TCP/IP, for now. > > Linux is far less resource-hungry, due to shared libraries. For now. -- \ / Charles Hannum, mycroft@ai.mit.edu /\ \ PGP public key available on request. MIME, AMS, NextMail accepted. Scheme White heterosexual atheist male (WHAM) pride! ------------------------------ ** 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-Activists-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux) via: Internet: Linux-Activists@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 tupac-amaru.informatik.rwth-aachen.de pub/msdos/replace The current version of Linux is 0.99pl6 released on February 23, 1993 End of Linux-Activists Digest ****************************** ill flame you for posting questions that are answered in the FAQ; this forumdigest865 644 36676 74430 51741 5366137106 5767 From: Digestifier To: Linux-Activists@senator-bedfellow.mit.edu Reply-To: Linux-Activists@senator-bedfellow.mit.edu Date: Sat, 17 Apr 93 07:00:08 EDT Subject: Linux-Activists Digest #865 Linux-Activists Digest #865, Volume #4 Sat, 17 Apr 93 07:00:08 EDT Contents: *** READ THIS BEFORE POSTING *** (Ian Jackson) Re: Linux Logo Postscript (Laszlo Herczeg) Re: Problems configuring news, etc (Vince Skahan) Re: Anyone using Orchid VA & S3? (Bill C. Riemers) [SOLUTION] TERM 107 problem (Keith Smith) STB PowerGraph VL 24 (Tom Wheeler) Re: [A] difficulties with libm (was Re: How to GCC part 2) (HJ Lu) Bourne Shell compatible shells (was: Request: which shells) (Arjan de Vet) Installing Linux (Tsung-lung Li) Internal Modems & Linux (John Edmond Auckett) Re: Why Linux should remain Linux and seagulls should not be used. (Michael K. Johnson) Kernel Panick, PCTOOL's COMPRESS and FIPS (Michael Portz) ---------------------------------------------------------------------------- From: ijackson@nyx.cs.du.edu (Ian Jackson) Subject: *** READ THIS BEFORE POSTING *** Date: Sat, 17 Apr 1993 06:23:01 GMT To get people to read your post and answer your question: - Read "Linux Documents Explained for Newbies" - posted weekly (see below). - Please do not post questions answered in the FAQ lists. - Please do not post "Please send me the FAQ". See below. - Please do not post "Split this group now". See below. - Be informative, both in subject line and body. See below. - Please read the documentation for the program you have a problem with. - Linux-specific posts only please - Unix, X, C, etc. groups listed below. For more information read on ... ** Linux Documents Explained for Newbies ** is is posted weekly by Jay MacDonald. If you can't find it email me or Jay . It will tell you about the Linux FAQ, the INFO-SHEET, the META-FAQ, etc. These are posted regularly to comp.os.linux and c.o.l.announce, and can be found in /pub/usenet/news.answers/linux-faq at pit-manager.mit.edu and on the Linux archive sites. If you can't get them there, try mailing "mailserver@nic.funet.fi", "mail-server@pit-manager.mit.edu" or "ftpmail@decwrl.dec.com". If you can't get them any other way, e-mail me. Splitting the group: It's been tried, and all the new groups except .announce failed. It can't be tried again within 6 months. See the FAQ for details, or read news.announce.newgroups. If you insist on discussing it, do so in news.groups (if c.o.l has too much volume already, why make it worse?). With threaded newsreaders now in widespread use using a good, specific, polite subject line will get you many more readers. Most people will ignore posts with subjects like "Linux", "Help", "SLS", etc. Don't shout to get attention. Put as much information in your post as possible - preferably a brief summary followed by session transcripts, etc. as appropriate. Include the precise text of any error messages printed, etc., and appropriate info about your setup. Linux is a Unix clone running on PCs; X has its own groups. One of the following groups is is quite likely to be more appropriate for your question: Unix: comp.unix.questions, .admin, .programmer, .shell etc. X: comp.windows.x, comp.windows.x.i386unix (for XFree86 etc) GNU: gnu.emacs.help, comp.emacs, gnu.gcc.help, gnu.* generally PCs: comp.sys.ibm.pc.hardware C: comp.lang.c -- Ian Jackson ijackson@nyx.cs.du.edu PGP2 public key available on request Home: 35 Molewood Close, Cambridge, UK, CB4 3SR; +44 223 327029. These opin- Work and urgent email: iwj@cam-orl.co.uk; +44 223 343398 ions are my own. Olivetti Research Ltd, Old Addenbrookes Site, Trumpington St, Cambridge, UK; ------------------------------ From: las@r-node.hub.org (Laszlo Herczeg) Subject: Re: Linux Logo Postscript Date: Fri, 16 Apr 1993 23:12:25 GMT In article <1993Apr12.164431.29710@cs.ucf.edu> tuggle@sg4 (Trenton Tuggle) writes: >Hey, > >There was this postscript logo someone posted here and everyone seemed >to have trouble with it, so I popped it up on my silicon graphics and >converted it to a GIF file... I uploaded linux.gif to tsx-11. It's >a wierd resolution, like 311 x 257; Sorry, my Reality Engine screen >is 1600 x 1200 and it's hard to judge image sizes. For the sake of those without ftp, could you please post the gif here in UUencoded form? Not long ago I managed to get the platypus which was suggested for Linux mascot in the same way. The platypus image took up only 13k of bandwidth, so maybe the Logo (B/W?) wouldn't take up too much space either? Many thanks in advance. -me -- ================================================================================ Laszlo Herczeg E-mail: las@r-node.gts.org Toronto, Canada E-mail (altern): las@eastern.com ------------------------------ From: vince@victrola.sea.wa.us (Vince Skahan) Subject: Re: Problems configuring news, etc Date: Sat, 17 Apr 1993 02:03:58 GMT ggeorge@bu.edu (Gerry George) writes: >News problems: > >News programs (/usr/local/lib/news...) is setup as suid. Tin is not. >When I run tin as a regular user, posting fails due to permissions. >(why do all my problems seem related to permissions - ingres, news, >mail, etc?) >root is able to post (sometimes) but no-one else can. >Scratch that now - noboby can post anymore - (3 days later) you want all of /usr/local/lib/news as news.news except for setnewsids which should be setuid root. >Why doesn't tin call config (/usr/local/lib/news/config) with effective >uid of 'news'? Should tin be su/gid news? mine is... >After fiddling with config, I made it a+r so the config problems are now >gone. > >After much hacking around, posts are now accepted from ordinary users, >but they are dumped. I still get the error: > relaynews: set[ug]id failed (invalid argument) >then it dumps the article in dead.article. set setnewsids to setuid root (see above). >Lots of the other utilities (news related) do not run properly - some problem >or the other, which may all be linked, or not. > >If those are just regular problems that will be cleared up with "proper" >setup (and much reading) then don't bother, but if it is unusual stuff, >please reply (post if possible?). Thanks. > yes. you're not set up right, therefore it's broke. watch for the UUCP-NEWS-MAIL-FAQ which should be coming around in a couple days in c.o.l., which discusses most of what you mentioned. >Thanks for any assistance. I might also need help in configuring my UUCP >files for my feed. Can someone mail me an example - if it is complicated, >or if it is a simple solution (i managed to work my way through TCP/IP and >most of the news stuff), then just a simple RTFF (FAQ) will suffice. In case >any instructions are forthcoming, it would help to explain what the different >options do since I wish to understand (and maybe help someone else?). I hope >I'm not asking too much. ok... RTFF :-) no, you're not asking too much, but yes...you should read the FAQ. Also, there are example config files in the /usr/local/lib/{news|elm|smail} and /usr/lib/uucp dirs that should get you close. Also read and follow the README.linux files. What's in the current SLS works fine if you follow the README.linux files and the FAQ I mentioned. The basic problem is that there's no 'news' group or user in the current SLS, which is something I've mentioned to Peter on a couple occasions. Create a news user and group...make the UID/GID whatever's convenient, then fix the owner.group like indicated above and you'll be in good shape. -- ---------- Vince Skahan --------- vince@victrola.sea.wa.us ---------- +++ A Waffle Iron - Linux Division +++ ------------------------------ From: bcr@bohr.physics.purdue.edu (Bill C. Riemers) Subject: Re: Anyone using Orchid VA & S3? Date: 16 Apr 93 18:12:44 GMT In article <1993Apr15.234111.931@tribiet.uucp> carvalho@tribiet.uucp (Joe Carvalho) writes: >Hi Gord! I am using the Orchid VA, the ISA version. I like it. Runs quite >nicely with Xs3. I've included the Xconfig I use. It needs a little work. >It will help in getting you up and running. Also...the board is VERY fast >under MSWindows. : Xconfig file omited from quote Hi there, This is just a note to comment I've added this Xconfig file to my mail database. If you have an Xconfig file that is not listed please mail a copy to me. Obcourse, in the case that someone asks for something it doesn't hurt to post it on the net, but only a small percentage of the people who want to use the same card-monitor combination are going to read your article. That is why I maintain a mail database of Xconfig files. I'll include a listing of files currently available at the end of this post. Bill ========================================================================== This service is intended to help reduce trafic on comp.os.linux and comp.windows.x.i386unix, by providing an effective service to help users install X. If you have a working Xconfig file not listed, or improvements to one that is listed, please e-mail it to me. If your monitor video card does not exactly match what appears below, please check that the video modes meet your monitor's specifications before using the Xconfig file. ========================================================================== 1 xfree1.2 : ATI Graphics Ultra Pro : 15" Crystal Scan 1572 fs 2 xfree1.0 : ATI mach32 adaptor : Hitachi CM1785 3 xfree1.2 : Diamond Speedstar : KFC 14" 4 xfree1.0 : Diamond Speedstar+ v.5.0 : AOC 14" 5 xfree1.? : TSENG ET 4000 : MD-17 mediascan 6A 6 xfree1.2 : Trident 8900c : Everdata (non interlaced 14") 7 xfree1.? : Trident 8900c : ViewSonic 5E monitor 8 xfree1.0Ya : TSENG ET 4000 : EIZO/FLEXSCAN 5500 8 xfree1.2 : " " : " " 9 xfree1.2 : Paradise : ? 10 xfree1.2 : Paradise WD990C30 : PC4XV-B2 (14" multisync) 11 Xs3 : Orchard VA (ISA) : ? ========================================================================== To recieve a particular file, e-mail bcr@physics.purdue.edu a message with the subject Xconfig-file # ------------------------------ From: ksmith@monu6.cc.monash.edu.au (Keith Smith) Subject: [SOLUTION] TERM 107 problem Date: Sat, 17 Apr 1993 05:18:35 GMT I just want to thank all the guys and girls who wrote to me with solutions...it seems that a few ppl have been in the same sinking boat. Anyways, I found that EVERY other term client was working fine, tupload tmon, termtelnet, termftp, even IRC! TRSH was even working when I ran it on the remote end (giving me a shell on the local machine)...so I figured it must be MY trsh binary. I recompiled it and it flew :) Thanks again, ppl. Keith -- ===================================================================== - ksmith@monu6.cc.monash.edu.au | Keith Smith, Monash Uni, - - ksmith@vx24.cc.monash.edu.au | Melbourne, Australia - ===================================================================== ------------------------------ From: Tom.Wheeler@f2113.n124.z1.fidonet.org (Tom Wheeler) Subject: STB PowerGraph VL 24 Date: Wed, 14 Apr 1993 15:10:11 -0800 AA> Has anyone managed to get the STB PowerGraph VL24 working with AA> X windows? I have made some attempts to get the dot-clocks from AA> the vendor but no results yet. AA> Does anyone have the dot-clock frquencies for this cards or managed AA> to get it working somehow? Ali, I don't know who you've been talking to here, but I have the info that you should need and will put it on our tech support BBS. Look for CLOCKIO.ZIP by the time you get this message. STB Tech Support BBS 214-437-9615 v.32bis 24hrs no parity 8 data 1 stop Please let me know if you have any problems. Regards, Tom Wheeler STB Systems * SLMR 2.1a * A gigolo is a fee-male. * Origin: Lunatic Fringe * Richardson,TX * 214-235-5764 * (1:124/2113) ------------------------------ From: hlu@luke.eecs.wsu.edu (HJ Lu) Subject: Re: [A] difficulties with libm (was Re: How to GCC part 2) Date: Sat, 17 Apr 93 07:25:03 GMT In article hal@pollux.cs.uga.edu (Hal N. Brooks) writes: [...] >If you subsequently get a runtime error message, then you need to >check to see where the symlink /lib/libm.so.4 is pointing. If you're >using libc4.3.2, then it should be made to point to libm.so.4.3.2. >This works for me anyway. > >Hopefully HJ won't have to correct me. Others have posted questions >about libm, and his response is usually you need to get libc4.3.x. >But I don't think HJ's instructions say anything about making the >libm.so.4 ---> libm.so.4.3.2 connection. He probably expects more >from some of us than we're capable of delivering! I'm going out on >a limb here, so if I'm wrong HJ, please be gentle with me. > You are right. I forgot to mention that. BTW, upgrade to 4.3.3 please. It fixes a lot of bugs and adds a few IEEE 754 stuff. H.J. ------------------------------ From: devet@adv.win.tue.nl (Arjan de Vet) Crossposted-To: comp.unix.shell Subject: Bourne Shell compatible shells (was: Request: which shells) Date: 17 Apr 1993 02:00:15 +0200 In article <1qmdl8$sgt@news.cs.tu-berlin.de> wong@cs.tu-berlin.de (Wolfgang Jung) writes: >which shells are available for Linux >I am particular interested in bourne shell compatibles.. Me too :-) >I just found one drawback of bash on linux: >The make depend from Smail 3.1.28 was doing nonsense . >checking the sh scripts I found the problem to set the CASE_NO_NEWLINES > >the bash was not able to work with newlines in arguments inside a case >switch... I also had some problems with smail and make depend. Don't remember how I fixed them. I also detected two other problems with bash 1.12: - it wouldn't run the Configure script of trn 3.0 (beta). This will be fixed in bash 1.13, but the Configure script has already changed so that it works with bash 1.12 too. - it does something wrong with passing $* from other shell scripts which are included by the `.' command. I detected this when INN 1.4 rejected all newgroup messages. >ksh blew on some other kind, it just hang at one place or was that >awfully slower than bash. The pdksh I have (4.7) is absolutely unuseable for running Configure scripts. Try the following script with pdksh 4.7 with two or more parameters: #!/bin/ksh echo "$@" Zsh gives also problems with Configure scripts. >What shells are available for linux and which type they are. >I know of: bash, ksh, tcsh, zsh. >I know bas is some how bourne shell compatible >tcsh somehow csh and ksh also sh.. I use tcsh as interactive shell and bash as /bin/sh substitute because bash is the more-or-less Bourne shell compatible shell which runs most of the (Configure) scripts I have. Running Configure scripts (sometimes >100Kb) is a real good test for testing /bin/sh compatibility of shells. Arjan -- Arjan de Vet (home) Eindhoven University of Technology, the Netherlands (work) ------------------------------ From: quantum@stein.u.washington.edu (Tsung-lung Li) Subject: Installing Linux Date: 17 Apr 1993 09:59:18 GMT Hello, I have gone through the FAQ's and all of the recent postings in c.o.l., but I still could not solve my installation problem. The configuration of my system: 486DX2/66Hz, 16Meg RAM, 240 Meg HD. I am trying to install Linux 0.99pl7. The following was what I did to install Linux. 1. Insert the diskette with file a1.test in drive a:, and boot the system. Some messages about the loading process appears. 2. The system invites me to login as "root", and I do. Then, following the message brought out by "install.info", I do the following. 3. Issue "fdisk", and use the its commands to achieve the following list /dev/hda1 41200 blocks DOS 16bit >= 32M /dev/hda2 197574 blocks Linux/MINIX Finally use the "w" command of fdisk. 4. Reboot the system. I get the following message and the system hangs. task[0](Swapper) killed: unable to recover Kernel panic: Trying to free up swapper memory space In swapper task - not syncing 5. Reboot the system, and login in as "root" again. 6. Issue "mke2fs /dev/hda2 197574" 7. Issue "doinstall /dev/hda2" and use the following options Install from floppy Drive B 3 1/2 inch Install everything (90MB) Is /dev/hda2 an Extended FS (2) used instaed of the default minix(n/y):y Then the system gives me the following error message and gets back to the "#" prompt. No installation occures. mount: mount point /root does not exist Error: can not mount /dev/hda2 Did you use: mkfs /dev/hda2 SIZE where SIZE is the number of blocks shown by fdisk? I am desperate to see Linux run on my system. Any help will be highly appreciated. My e-mail adress is "quantum@u.washington.edu". Tsung ------------------------------ From: s851708@minyos.xx.rmit.OZ.AU (John Edmond Auckett) Subject: Internal Modems & Linux Date: 17 Apr 93 08:43:45 GMT I have come up with a problem with Linux detecting the UART on an internal modem. WHen the system boots up the UART on the modem card is not detected, even though the POST reported its existence. What is different about a UART on a modem card? I have seen simliar problems reported here but no straight forward solutions. I am running SLS .99p6. Any help thanks. JA (s851708@minyos.rmit.oz.au) ------------------------------ From: johnsonm@stolaf.edu (Michael K. Johnson) Subject: Re: Why Linux should remain Linux and seagulls should not be used. Date: Fri, 16 Apr 1993 20:04:01 GMT In article <1qlfiv$h28@fitz.TC.Cornell.EDU> mdw@theory.TC.Cornell.EDU (Matt Welsh) writes: Michael K. Johnson and I have decided to use the Linux seagull logo on the title page of the Linux Documentation Project manuals as well as other works. We're working right now on integrating the logo with the current linuxdoc.sty LaTeX style format used by the LDP. While this by no means makes it "official", it's about as official as you can get in the Linux community. The seagull represents many things about [...] I'd like to emphasize that this is not intended to tell other people what they like in a logo. If "you", whoever you are, are putting out a Linux-related product, and you want to use a blade of grass or an airplane or a cloud or whatever as the logo, that's fine with me. *Unless Linus decides he wants one particular logo, let's not try to make any logo official!* However, since a lot of us in the doc project like the storm petrel logo, and for a few other reasons, we have chosen to put it on our work... "You take the high road, and I'll take the low road..." michaelkjohnson ------------------------------ From: michaelp@mjoli.informatik.rwth-aachen.de (Michael Portz) Subject: Kernel Panick, PCTOOL's COMPRESS and FIPS Date: 17 Apr 93 11:52:18 I have 386/40, 128K Cache, 8MB RAM, 128 MByte HD On HD are two partitions: dosfs on /dev/hda1 (20MB) ext2fs on /dev/hda2 (rest) For booting I use LILO. The versions are SLS 0.99.? and LILO 0.7. Because 20MB DOS is more than I need, I wanted to use the just announced FIPS-system to split the DOS-partition into to 10MB partitions, from which I wanted to reuse the second one mainly for /tmp and swappping. In FIPS-README it is suggested to use a common disk-defragmentor (e.g. PCTOOLS COMPRESS) for preparing the DOS-HD for splitting and then apply FIPS. PROBLEM: After having applied COMPRESS to the DOS-partition I was not able to boot LINUX from /dev/hda2 anymore. LILO comes up with the usual stuff but is not able to mount the root-filesystem. The system hangs with a kernel panick (understandable!). What I will do today is to boot with SLS's disks a1 and a2 and then try to analyze the ext2fs on /dev/hda2. Nevertheless: If anyone knows something about the problem, please post or mail me. Michael -- /***Michael Portz***michaelp@terpi.informatik.rwth-aachen.de*********/ /***Lehrstuhl fuer Angewandte Mathematik insbesondere Informatik*****/ /***RWTH Aachen***Ahornstr.55***5100 Aachen***Germany****************/ /***Phone: +49 241 80 21062***Fax: +49 241 80 21079******************/ ------------------------------ ** 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-Activists-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux) via: Internet: Linux-Activists@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 tupac-amaru.informatik.rwth-aachen.de pub/msdos/replace The current version of Linux is 0.99pl6 released on February 23, 1993 End of Linux-Activists Digest ****************************** adress is "quantum@u.washingtondigest866 644 36676 74430 51667 5366137106 5777 From: Digestifier To: Linux-Activists@senator-bedfellow.mit.edu Reply-To: Linux-Activists@senator-bedfellow.mit.edu Date: Sat, 17 Apr 93 12:00:11 EDT Subject: Linux-Activists Digest #866 Linux-Activists Digest #866, Volume #4 Sat, 17 Apr 93 12:00:11 EDT Contents: Re: LHA compression (J.P. Hillenburg) possible bug in mount() (Todd J. Derr) Re: SLS Boot Disk Oddity (Markfried Fellensiek) Re: possible bug in mount() (Linus Torvalds) New Linux box - Advice? (Benjamin Weste Pearre) Re: linux shared libraries often incompatible (Wolfgang Jung) How to increase inodes? (mlh@lakes.trenton.sc.us) Is curses.h complete? (mlh@lakes.trenton.sc.us) Serial direct connect ttys for > 0.99pl4 (Michael Hamilton) Does Compaq Localbus run X ? (R.R. Watts) Re: EISA: Anybody working on it? (John F Carr) looking for dirt (Johan A. Grape) Re: Questions about virtual mem. (Michael Fuhr) login to X as non-root (Bob Palmer) Telnet problems solved (partly) (Kari T. Salmela) Advice on setting up Linux box on LAN (ronan@bermuda.ucd.ie) MIT Scheme 7.2 Alpha (I ported it, but...) (Charles Fee) Re: C= 64 emulator project (Jarkko Sonninen) ---------------------------------------------------------------------------- From: jph@anaconda.UUCP (J.P. Hillenburg) Subject: Re: LHA compression Date: 16 Apr 93 21:19:41 EST Reply-To: moose.cs.indiana.edu!charon!anaconda!jph (Joseph Hillenburg) In article <1993Apr15.232735.21887@jupiter.sun.csd.unb.ca> u0xh@jupiter.sun.csd.unb.ca (Figmiester) writes: Or, better > yet, a site that has .mods using zip compression. Thanks for any help. Look on ftp.luth.se, in /pub/amiga/unix/something. (Don't remember the exact directory...) Unix LhA is there... Almost all MODs on the Internet are compressed using LhA/LZH because they originated on the Amiga. About the only places they'll be in .ZIP format is on PC BBSes. -- Joseph Hillenburg NPS Technologies charon!anaconda!jph@moose.cs.indiana.edu or jph@gnu.ai.mit.edu [Also running BMS] ------------------------------ From: tjd@neurocog.lrdc.pitt.edu (Todd J. Derr) Subject: possible bug in mount() Date: 17 Apr 93 11:17:36 GMT well, it doesn't seem that passing bad values to mount() should cause a GP fault, so... (yes, I know that I went about this wrong, but that's not the point...) the program: #include main() { mount("/dev/hda1","/", MS_MGC_VAL | MS_REMOUNT); } # ./a.out general protection: 0000 EIP: 0008:0001A1D7 EFLAGS: 00010286 fs: 0017 base: 00000000, limit: C0000000 Pid: 494, process nr: 3 64 8a 00 88 44 14 24 84 c0 74 Segmentation fault this is running a modified (but not in the fs part) .99.8 kernel. 01a1d7 is in _sys_mount() in my kernel. todd. ------------------------------ From: ins413j@mdw037.cc.monash.edu.au (Markfried Fellensiek) Subject: Re: SLS Boot Disk Oddity Date: Sat, 17 Apr 1993 11:24:03 GMT : At one point in the boot, I had become used to seeing a series of : dots (.....) being presented. This behavior changed about a month : ago, and I intermittently see what appears to be assembler debug : info on the screen. When it appears, it looks something like: : : ..... : mov ax,cx : (more asm stuff follows, and then the dots resume) : ........................ : : Once the boot is complete, everything seems to work normally. I've : regenerated the boot disk, but the new one does that too, so it's : not a media problem. The floppy drive itself has never given any Maybe some twisted person has already written a Linux virus!! (Hey, just a thought... this might explain why we're having so many problems...) ;) Mark. ------------------------------ From: torvalds@klaava.Helsinki.FI (Linus Torvalds) Subject: Re: possible bug in mount() Date: Sat, 17 Apr 1993 12:28:40 GMT In article <9246@blue.cis.pitt.edu> tjd@neurocog.lrdc.pitt.edu (Todd J. Derr) writes: >well, it doesn't seem that passing bad values to mount() should cause a GP >fault, so... (yes, I know that I went about this wrong, but that's not the >point...) > >the program: >#include >main() >{ mount("/dev/hda1","/", MS_MGC_VAL | MS_REMOUNT); } > ># ./a.out >general protection: 0000 >EIP: 0008:0001A1D7 >EFLAGS: 00010286 >fs: 0017 >base: 00000000, limit: C0000000 >Pid: 494, process nr: 3 >64 8a 00 88 44 14 24 84 c0 74 >Segmentation fault > >this is running a modified (but not in the fs part) .99.8 kernel. 01a1d7 is >in _sys_mount() in my kernel. Oh, well, it's harmless, but assuming you want the correct error return (EFAULT), you might want to try this patch.. No guarantees - I haven't tried the above program, but it should fix it. Your mount() still won't work, naturally, but at least wrong parameters shouldn't cause a GP fault. Linus ===== snip snip ===== --- l99.8/linux/fs/super.c Tue Apr 6 20:57:20 1993 +++ linux/fs/super.c Sat Apr 17 15:20:56 1993 @@ -379,9 +379,12 @@ return do_remount(dir_name,new_flags & ~MS_MGC_MSK & ~MS_REMOUNT); } if (type) { - for (i = 0 ; i < 100 ; i++) + for (i = 0 ; i < 100 ; i++) { + if (TASK_SIZE <= (unsigned long) type) + return -EFAULT; if (!(tmp[i] = get_fs_byte(type++))) break; + } t = tmp; } else t = NULL; ===== snip snip ===== ------------------------------ From: bwpearre@dial.Princeton.EDU (Benjamin Weste Pearre) Subject: New Linux box - Advice? Date: Sat, 17 Apr 1993 06:27:20 GMT I am about to buy a new computer. Decisions, decisions... Well, does anyone have any advice for me? I am thinking along the lines of a dx2-66 or 50, with 16 meg of ram, an Adaptec 1542B, and ... What else indeed? No-one seems to be able to really reccomend any graphics card. Can anyone here clarify the matter? et4000? ATI Ultra Pro? Orchid something? Has anyone gotten X running on a Gateway with the ATI Ultra Pro? That's what they are trying to sell it to me with, and they seem to think that it might not run with anything else. (!) Anyone? The compatibility list mentions the gateway systems, but does not mention the graphics cards, which I find strange... Zeos claims that their on-board SCSI is fully Adaptec-compatible. Is this so? The compatibility list does not mention it. I am skeptical. Anyone? Any other suggestions are welcome. Please, someone who is happy with your system, tell me what you have! Thanks :) Cheers, Ben Pearre bwpearre@phoenix.princeton.edu ------------------------------ From: wong@cs.tu-berlin.de (Wolfgang Jung) Subject: Re: linux shared libraries often incompatible Date: 17 Apr 1993 13:15:28 GMT Aaron Michael Cohen (acohen@world.std.com) wrote: : Linuxers: : I have had some problems recently in trying to use binaries that I : have downloaded from tsx-11 and sunsite. This is almost always due : to incompatible shared library versions. Now, while I think that : shared libraries are great for saving space, distributing executables : linked with them without source code isn't working for me. It probably : isn't working for some others too. My Linux installation is based on : an SLS release about 2 months old, and already there are lots of apps : that I can't use. So I'd like to propose a solution. : When you post an original app or a port to the net either (1) include the : sources and a makefile, or (2) distribute a set of unlinked object files : and include the link command in the README.linux file. Even better is a pointer to the patches from the original sources (stating which Source version is needed.) I never saw a conf/os/linux for the smail sourcetree :-( Also it might be good to have the Programs accept older lib revisions (i.E all compiled for libc.so.4.3.1 should also run with lib libc.so.4.3.3) or even better libc.so.4 build programs should be able to run with libc.so.4.3.3. Also (I might be incorrect here) I think most of the things done with newer libs are added features and less bugs in the lib itself, so no programm should behave bad, if t has to run with a newer lib. It shoudl mention it ios running with a lib newer revison (or even older revision) But it should at least try to run. Gruss Wolfgang ------------------------------ From: mlh@lakes.trenton.sc.us Subject: How to increase inodes? Date: Sat, 17 Apr 1993 08:01:56 GMT SLS .99pl6 (A disks), pl4 (rest of disks). Using the ext2fs. I get about 1000 newsgroups fed to my system, and find I am running out of inodes on /dev/hdb1 (my usr/spool/news drive) long before running out of volume. I have a 205mb drive mounted as /dev/hdb1, and it shows a total of 51584 inodes. If I'm calculating this right, when I ran out of inodes the last time, using values reported by df: 1024 * / = around 2850 bytes per inode. The man page for mkefs says I can specify the bytes per inode with any number above 1024. Can I re-make the drive and specify 1024, and what would be the consequences? I had thought there was a 65536 (or close to that) limit on inodes. ------------------------------ From: mlh@lakes.trenton.sc.us Subject: Is curses.h complete? Date: Sat, 17 Apr 1993 08:09:22 GMT SLS .99pl6 (A disks) pl4 (rest of disks) When trying to compile a "door" type chess game for the bbs, the compile failed when it encountered A_BOLD and some other, what I understand to be, curses functions. The curses man page says these are defined in curses.h, but grepping it and all the .h files turned up no match. Are these commands defined somewhere? ------------------------------ Subject: Serial direct connect ttys for > 0.99pl4 From: hamilton@golem.wcc.govt.nz (Michael Hamilton) Date: 18 Apr 1993 01:56:00 +1200 I've had a few problems with a direct connect vt320 terminal after upgrading from linux 0.99pl4 to pl8. Someone else is bound to hit the same hassles, so... Here are instructions for getting a direct connect terminal, or PC acting as a terminal, to work over a null modem cable plugged into a ttys? device. After linux 0.99pl5 was released the changes relating to ttyS? and cua? devices made this a little more complex than before. First you need the getty from getty_ps. You install as per instructions, and then add/replace the entry to your /etc/inittab for the terminal, here's mine: ttys2:vt100:/etc/getty_ps ttys2 9600 A slight aside: I normally run the poeigl getty (AKA agetty) and because I've just started using getty_ps, I called it's getty getty_ps so that I can continue to run the poeigl getty on the consoles (I'm playing it safe). You need to use getty_ps for the terminal so that you can use a /etc/gettydef file to decscribe the necessary tty settings for a direct connect terminal. Here's the only entry I changed in the default getty_ps /etc/gettydef file on my system: # default entry: # 9600# B9600 CS8 CLOCAL # B9600 SANE -ISTRIP CLOCAL ECHO #login: #9600 The most important of the tty settings is CLOCAL, which tells getty_ps to turn off modem control signals. If you don't specify CLOCAL getty_ps and poeigl login (I haven't tried other logins) will hang waiting for a modem carrier on any kernel on or above 0.99pl5. This is all I needed to do. But it take a little while to work out what to do with the various getty, tty and device options. "The Unix System Administration Handbook", by Nemeth, Snyder and Seebass put all the pieces into place, I just wish I'd consulted it a few hours earlier (a great book). Where I got my sources: sunsite.unc.edu:/pub/linux/system/Serial/getty_ps2.0.6b.tar.z ftp.daimi.aau.dk:/pub/Linux-source/poeigl-1.15a.tar.z P.S. I initially got around the getty/login hang problems by running getty/login on a cua? device. But I use tcsh, and tcsh calls tcgetpgrp() in libc. tcgetpgrp() doesn't like cua devices (libc 4.3.2) although an ioctl() call to do the same thing does work. tcgetpgrp() returns an invalid file number, possibly because it doesn't know that the major device number of 5 (for cua? devices) are tty's. Should some of the above be an FAQ? It's late, I starting to ramble... -- Michael Hamilton, Information Technology, Wellington City Council, P.O. Box 2199, Wellington, New Zealand. Phone: (64) (4)801-3317 FAX: (64) (4)801-3020 Domain: hamilton@golem.wcc.govt.nz PSImail: PSI%0530147000090::HAMILTON ------------------------------ From: rrw1000@cl.cam.ac.uk (R.R. Watts) Subject: Does Compaq Localbus run X ? Date: 17 Apr 93 14:01:34 GMT Hello, I'm thinking of buying a Linux box, and for several reasons, the best machine that seems to be within my price range is a Compaq Prolinea 50/240 with a Compaq Localbus graphics agcelerator (50/240 = 50 MHz 486DX2, 240MB Hard disk). Question : Will Linux run on such a machine ? Has anyone had any bad experiences with such hardware in the past ? Chipset compatibilities I'm not sure about - my local PC dealer has the brains of a stunned haddock and can only give me Compaq's publicity stuff which isn't very useful. I'm buying this machine primarily for use as a UNIX in general and Linux in particular box, and don't much want to wind up running Wordperfect on it ! Any help would be greatly appreciated. I can't see it on the Linux Compatibility list & I can't seem to get at the cwxi FAQ :-(. Please reply by E-mail to rrw1000@phx.cam.ac.uk [NOT rrw1000@cl.cam.ac.uk]. Thanks in advance. rrw. ------------------------------ From: jfc@athena.mit.edu (John F Carr) Subject: Re: EISA: Anybody working on it? Date: 17 Apr 1993 14:44:28 GMT There are two EISA SCSI adapters supported by Linux: UltraStor 24F and Adaptec 1742. 24F support isn't in the core kernel sources or SLS, but is in the Yggdrasil product. Source code is in: tsx-11:/pub/linux/ALPHA/scsi/u24f-driver.tar.z I only have 16 MB in my system so I haven't verified that the 24F driver works with more memory, but I expect it will. -- John Carr (jfc@athena.mit.edu) ------------------------------ From: jgrape@coos.dartmouth.edu (Johan A. Grape) Subject: looking for dirt Date: Sat, 17 Apr 1993 13:46:23 GMT Hi all you linuxers. I am trying to find out if anyone has ported dirt to linux yet, and if they would be so kind as to point me to a binary, or just provide me with diffs. Thanks a bundle. OK - I'll do a service since I am using bandwidth ;-) For those who wonder, dirt is a user interface builder for X. Allthough the author (Richard Hesketh = rlh@ukc.ac.uk) will probably have me fried for this: it's sort of like a ResEdit for X. I haven't tried it out yet, but it looks interesting, and I want to know if someone else has ported it before I charge ahead. To find out more, get contrib/dirt.README from export.lcs.mit.edu Johan no.... no .sig ------------------------------ From: mfuhr@cwis.unomaha.edu (Michael Fuhr) Subject: Re: Questions about virtual mem. Date: 17 Apr 93 14:27:15 GMT jmward@ucrengr (jonathan ward) writes: >Yes, Linux does support virtual memory, in the form of a swap partition. It >can be made as large as you want, although once you get around 12 megs, >anything more is a waste of disk space: it won't be used. I run my system ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ Depends on what you're doing. I have 8MB and two 16MB swap partitions (okay, THAT's overkill) and I've seen 20MB of swap used up when doing some intense image processing while running X. I've seen compiles of files that have a lot of static data (like a large bitmap) use up 10MB of swap. For what I do, I'd recommend _at least_ 12MB of swap. -- Michael Fuhr mfuhr@cwis.unomaha.edu ------------------------------ From: b645zll@utarlg.uta.edu (Bob Palmer) Subject: login to X as non-root Date: Sat, 17 Apr 1993 15:18:00 GMT Is there any info about how to login to X. I can get my system to recognize only root login, but cannot login any other way. If there is a FAQ, I have not found it. Thanks Bob B645zll@utarlg.uta.edu ------------------------------ From: weasel@rieska.oulu.fi (Kari T. Salmela) Subject: Telnet problems solved (partly) Reply-To: weasel@rieska.oulu.fi Date: Sat, 17 Apr 1993 14:22:08 GMT If somebody is interested: I was able to get telnetd etc. working by just installing SLS network packet over my TAMU set (it is on disk b1, if I remember right.). It still seems to hang when telnetting from outside sometimes, dunno why. Greetings, Kari -- -- Kari Salmela I CS Student. Preferences: -- Virkakatu 8 E 18 90570 Oulu FINLAND I C64,UNIX,OS/2,Beer -- Phone#: 949 68 98 62 / +358 81 5546 275 I ------------------------------ From: ronan@bermuda.ucd.ie Subject: Advice on setting up Linux box on LAN Date: Sat, 17 Apr 1993 16:37:00 GMT I hope to set up a PC as a full time Linux machine over the summer. I also hope to set up a (very) small LAN (the Linux machine and a DOS box). However my practical knowledge of LANs is very limited, so I've a few questions: What recommendations would you make on both the spec for the Linux machine and the networking hardware (bear in mind that I don't _need_ mind-blowing performance, just a cheap price-tag) ? What do I need (on both machines) to be able to access the disk in the Linux machine from the DOS box ? What do I need to be able to use a printer connected to the Linux machine from the DOS machine ? Is it possible to directly access a modem on the Linux machine from a terminal program on the DOS machine (ie without going through something like Kermit or pcomm in Linux). Is it possible to have my modem receive incoming fax calls as well as data calls ? Anything else you think I should know about ??? Any suggestions would be much appreciated (via email if possible). Ronan Mullally =========================================================== 320 Sutton Park, Sutton, | Fidonet: 2:263/151 (TOPPSI) Dublin 13, Ireland | Internet: ronan@bermuda.ucd.ie ------------------------------ Date: Sat, 17 Apr 1993 11:16:49 EDT From: Charles Fee Subject: MIT Scheme 7.2 Alpha (I ported it, but...) I was bored last wednesday night. I ftp'd the i386 source of MIT-Scheme 7.2 Alpha and compiled it for Linux. It seems to work ok. I've run some scheme programs inside it, but haven't extensively tested it. It compiled under GCC 2.3.3 with jumplibs 4.3.3 with very few changes. I hacked the makefile here and there, and commented out some SIGBUS stuff, but that was about it. The source said not even to attempt it unless you are a C and unix god, but I am neither. Hell, I've taught myself C by looking at kernel sources. As I noted before, I haven't extensively tested tested the binaries. I'm not a scheme person. I literally compiled it because I had nothing better to do. If anyone else has compiled this and could give me some hints, or if you would like to help test this, please drop me a line. I have no plans to release this until it can be successfully tested. I don't think I have the necessary skills to sufficiently test this on my own. ------------------------------ From: Jarkko.Sonninen@lut.fi (Jarkko Sonninen) Subject: Re: C= 64 emulator project Date: Sat, 17 Apr 1993 18:34:19 GMT In article <1993Apr16.230232.10343@sfu.ca> rchen@fraser.sfu.ca (Robert Chen) writes: > It compiles nearly out of the box and works well. I have had one > annoying problem. When using the monitor, I have to type ^M^J to > get a newline (ie, to enter each command). "Enter" just gives me "^M" > written on the screen. If anyone has a fix for this I would be > interested in hearing from you. Ouch. there are bugs in command line editor. The following patch for ed.c should fix it (allthought I haven't tested it yet on linux) 194a195 > e_bind_key("\n", "end-of-line"); 201,202c202,203 < e_bind_key("\012", "redraw-line"); < e_bind_key("\011", "kill-to-end"); --- > e_bind_key("\014", "redraw-line"); > e_bind_key("\013", "kill-to-end"); > This emulator look like a lot of fun. I still have some old 6510 > machine code sitting on some disks in a closet somewhere. I used to > love programming my C=64. Too neat. My c64:s keyboard got too full of dust to be usable :( that's one reason for existance of x64 ;) - Jarkko -- Jarkko.Sonninen@lut.fi Free .signature virus -killer: find / -name .signature -exec rm -f {} \; ------------------------------ ** 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-Activists-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux) via: Internet: Linux-Activists@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 tupac-amaru.informatik.rwth-aachen.de pub/msdos/replace The current version of Linux is 0.99pl6 released on February 23, 1993 End of Linux-Activists Digest ****************************** spec for the Linux machine and the networking hardware (bear in minddigest867 644 36676 74430 50761 5366137106 5772 From: Digestifier To: Linux-Activists@senator-bedfellow.mit.edu Reply-To: Linux-Activists@senator-bedfellow.mit.edu Date: Sat, 17 Apr 93 14:15:35 EDT Subject: Linux-Activists Digest #867 Linux-Activists Digest #867, Volume #4 Sat, 17 Apr 93 14:15:35 EDT Contents: Re: Security system for Linux (Gilbert Nardo) Which is the best soundcard supported by 386BSD and Linux ? (Pim Zandbergen) Re: PS/2 style mouse not supported?? (Bob Dupre) Re: the umask of root in SLS should be 022 ?? (Johan Myreen) Re: linux shared libraries often incompatible (Arjan de Vet) Re: ext2fs compressed filesystem *IS* *ALREADY* under development. (Robert L. McMillin) ext2fs feature-itis vs. robustness (Robert L. McMillin) Re: Disk Quotas (was Re: New feature for the filesystems.) (Jim Graham) Re: Summary of Linux vs. 386BSD vs. Commercial Unixes (HJ Lu) Re: possible bug in mount() (HJ Lu) Re: Summary of Linux vs. 386BSD vs. Commercial Unixes (Nate Williams) Compiling Getty********* (Greg Corteville) Changing my address (Greg Corteville) Re: QIC-02 Tape Patches? ( Ian Justman) Re: QIC-02 With 0.99.8 ( Ian Justman) ---------------------------------------------------------------------------- From: gil@netcom.com (Gilbert Nardo) Subject: Re: Security system for Linux Date: Sat, 17 Apr 1993 16:08:20 GMT michaelw@desaster.hanse.de (Michael Will) writes: > mat@eccles.dsbc.icl.co.uk (Matthew Roderick - KID01) writes: > > Secure File system (2 inodes per file) > > What about the journalled-filesystem such as aix uses? > You can pull the plug of a rs6000 and it will never loose a byte. > fsck? You do not need it after a cold-reboot. Journalled file systems belong more under the category "system reliability" than it does under "system security" (fuzzylogically speaking). Highly secure systems that meet requirements for Class A1 or B3 levels include features such as a reference monitor, which is implemented within the kernel. Good references on this material can be found in the Proccedings of the National Computer Security Conference. IMHO, full fledged system security must be well designed in view of the whole system. For Linux, IMHO again, this would lead to system bloating much too early in the game (Please don't interpret this as a "stop this talk on security" but more like "ahh, more reference material"). -- Gil Nardo | gil@netcom.com Migrant Computing Services | (415)664-1032 (voice) 1032 Irving Street, #435 |----------------- San Francisco, 94122 | Save the Universe: Stop Entropy Now! ------------------------------ Crossposted-To: comp.os.386bsd.questions,comp.sys.ibm.pc.soundcard From: pim@cti-software.nl (Pim Zandbergen) Subject: Which is the best soundcard supported by 386BSD and Linux ? Date: Sat, 17 Apr 93 14:27:20 GMT As Linux and 386bsd probably do not have equal support, it's best to rephrase: Which is the best soundcard supported by Linux ? Which is the best soundcard supported by 386BSD ? -- domain : Pim Zandbergen snail : Laan Copes van Cattenburch 70, 2585 GD The Hague, The Netherlands phone : +31 70 3542302 fax : +31 70 3512837 ------------------------------ From: dupre@spades.aces.com (Bob Dupre) Subject: Re: PS/2 style mouse not supported?? Date: 17 Apr 93 16:23:00 GMT In article <1993Apr16.174537.17946@organpipe.uug.arizona.edu>, kkc@stat.nursing.arizona.edu writes... > >Get the "mconv.c" file and you should be ok. The instructions are in that file. > >Kee Chua Don't bother with "mconv.c" if your using XFree386 V1.2. Support for PS/2 mice is already there. I had problems getting the PS/2 mouse to work on my Packard Bell Legend 950. I found that if you use the correct device name in the Xconfig file it works great. My entry for the mouse in Xconfig is as follows. PS/2 "/dev/mouse" Be sure you have a symlink between mouse and ps2aux in your /dev dir. Hope this help 8^) Bob ------------------------------ From: jem@snakemail.hut.fi (Johan Myreen) Subject: Re: the umask of root in SLS should be 022 ?? Date: 17 Apr 93 17:22:06 GMT In article twpierce@unix.amherst.edu (Tim Pierce) writes: >I did a little looking around after installing SLS and decided that >there was nothing, short perhaps of /proc (and perhaps not even that) >that I would not want readable by a non-superuser. So I ran "find / >-exec chmod a+r {} \;" on the whole bunch and it started working fine. If you are running multi-user, you might at least want to change /dev/hda*, /dev/hdb* etc. back to 600... -- Johan Myreen jem@cs.hut.fi ------------------------------ From: devet@adv.win.tue.nl (Arjan de Vet) Subject: Re: linux shared libraries often incompatible Date: 17 Apr 1993 18:46:49 +0200 In article <1qovtg$khg@news.cs.tu-berlin.de> wong@cs.tu-berlin.de (Wolfgang Jung) writes: >Even better is a pointer to the patches from the original sources (stating >which Source version is needed.) I never saw a conf/os/linux for the smail >sourcetree :-( I would suggest the following scheme. Say you port a package foo-1.00.tar.z. You then could make a package named foo-1.00-linux.tar.z containing the following: - binaries - a patch file (context or unified) + Linux specific config files - A README file, saying which version of the libraries the binaries are compiled with and giving instructions how to compile the package yourself (and that includes ungzipping/untarring, patching, putting the config files to the correct location, making and installing). BTW: I ftp-ed a conf/os/linux for smail more than half a year ago from one of the ftp sites. >Also it might be good to have the Programs accept older lib revisions (i.E >all compiled for libc.so.4.3.1 should also run with lib libc.so.4.3.3) or even >better libc.so.4 build programs should be able to run with libc.so.4.3.3. This is already the case :-) The problem was that programs compiled with, say, libc.so.4.3.3 won't run with older versions of the libraries. The best thing to do is to upgrade your C library regularly. Most people who make binaries compile the binaries with the latest C library. Arjan -- Arjan de Vet (home) Eindhoven University of Technology, the Netherlands (work) ------------------------------ From: rlm@helen.surfcty.com (Robert L. McMillin) Subject: Re: ext2fs compressed filesystem *IS* *ALREADY* under development. Date: Sat, 17 Apr 1993 14:55:59 GMT In article sct@dcs.ed.ac.uk (Stephen Tweedie) writes: > Well, there certainly seems to be a lot of interest in this, so I > though I'd post about work currently in progress on the ext2fs. > > In outline: > > I am working on adding the option to transparently compress and > uncompress ext2fs files. Compression will be selectable on a > file-by-file basis. > > This will be an enhancement to the existing ext2fs. The new > facility will be immediately useable by anybody with an ext2fs > filesystem without requiring any reformatting. Hmm. Interesting. e2fsck has been known to hang on me when checking moderately corrupted file partitions. My questions: * I have the SLS 1.0 release. Is there a newer e2 suite available? * Has anyone else had similar troubles with e2fsck hanging while checking a suspect partition? * Why does the e2 filesystem not shut down cleanly, even with a sync? It seems like e2 partitions require a umount before shutting down, lest they get an 'invalid filesystem' error when the partition is remounted after reboot. It seems the e2 filesystem, while nice so long as you don't have any problems with it, tends to degrade rather quickly once you do. Once, when I had a problem with an unreliable SCSI tape controller, I had to *repeatedly* rebuild the e2 partitions. e2fsck just wouldn't fix the filesystems. Typically, it just hung. Anyway, my two cents' worth is that it seems that creeping feature-itis has overtaken robustness as a design goal for the e2 file system. -- Robert L. McMillin | Surf City Software | rlm@helen.surfcty.com | Dude! #include ------------------------------ From: rlm@helen.surfcty.com (Robert L. McMillin) Subject: ext2fs feature-itis vs. robustness Date: Sat, 17 Apr 1993 14:59:31 GMT In article sct@dcs.ed.ac.uk (Stephen Tweedie) writes: > Well, there certainly seems to be a lot of interest in this, so I > though I'd post about work currently in progress on the ext2fs. > > In outline: > > I am working on adding the option to transparently compress and > uncompress ext2fs files. Compression will be selectable on a > file-by-file basis. > > This will be an enhancement to the existing ext2fs. The new > facility will be immediately useable by anybody with an ext2fs > filesystem without requiring any reformatting. Hmm. Interesting. e2fsck has been known to hang on me when checking moderately corrupted file partitions. My questions: * I have the SLS 1.0 release. Is there a newer e2 suite available? * Has anyone else had similar troubles with e2fsck hanging while checking a suspect partition? * Why does the e2 filesystem not shut down cleanly, even with a sync? It seems like e2 partitions require a umount before shutting down, lest they get an 'invalid filesystem' error when the partition is remounted after reboot. It seems the e2 filesystem, while nice so long as you don't have any problems with it, degrades rather quickly once you do. Once, when I had a problem with an unreliable SCSI tape drive (it would hang the SCSI bus), I had to *repeatedly* rebuild the e2 partitions. e2fsck just wouldn't fix the filesystems. Typically, it would hang. Anyway, my two cents' worth is that it seems that creeping feature-itis has overtaken robustness as a design goal for the e2 file system. -- Robert L. McMillin | Surf City Software | rlm@helen.surfcty.com | Dude! #include ------------------------------ From: jim@n5ial.mythical.com (Jim Graham) Subject: Re: Disk Quotas (was Re: New feature for the filesystems.) Date: Fri, 16 Apr 1993 13:22:53 GMT due to what appears to be a substantial interest in this, I'm posting here instead of e-mailing (as I had planned when I read this first article). In article nick@symphony.mp.ucc.ie (Nick Hilliard) writes: >With all this talk about new features for the various filesystems, I figure >no-one's really mentioned anything about Disk Quotas. > >Ok - it's fine if you're a just using Linux as a single user machine, but if >you have a whole load of users hanging out of it, you really _do_ need some >form of mechanism for making sure that people don't hog the machine. this doesn't really address your question exactly, but it might be along the lines of what you had in mind....it doesn't impose limits, as such, but it will automagically complain to users who are hogging the disk. one of the best UNIX admin books I've ever seen (and one of the most recommended books out there) is called ``UNIX System Administration Handbook'' (by Evi Nemeth, Garth Snyder, and Scott Seebass). in this book, there is a csh script called spacegripe (Appendix I, pp. 511-513), that will at least complain to a user who is hogging the system. you could easily modify this to send a copy of the gripe to root, and notify the user that root has been informed...and is going to be watching their disk usage. I can also think of ways to modify it even more to get nastier if they log on and don't do anything about it (have to look at when they last logged on and compare it to when the first/most recent gripe message was sent). btw, if you don't have this book, you need to run (not walk) to the nearest decent bookstore and get or order a copy. I paid $34 at Interop (strangely enough, the receipt is one of my bookmarks in the book), but have no idea what it runs for at bookstores. --jim PS: I have no financial interest in this book...I'm just glad I bought it at the first Interop I went to (I got them to let me look at a copy over breakfast, and after looking through it for about 1 minute, I went back to the counter and bought it...as did my co-worker who also took a look at the copy I had). I have, however, met the primary author, Evi Nemeth (at the Spring 92 Interop)---nice person. very modest, too (I gathered that she didn't see what all the fuss was about over the book, but she likes all the e-mail she's gotten). btw, she mentioned something about a 2nd edition being in the works...no idea when it'll be out, but I want it! -- #include 73 DE N5IAL (/4) ============================================================================== INTERNET: jim@n5ial.mythical.com | j.graham@ieee.org ICBM: 30.23N 86.32W AMATEUR RADIO: n5ial@w4zbb (Ft. Walton Beach, FL) AMTOR SELCAL: NIAL ============================================================================== E-mail me for information about KAMterm (host mode for Kantronics TNCs). ------------------------------ Crossposted-To: comp.os.386bsd.questions From: hlu@luke.eecs.wsu.edu (HJ Lu) Subject: Re: Summary of Linux vs. 386BSD vs. Commercial Unixes Date: Sat, 17 Apr 93 16:15:16 GMT In article <1qo0lq$1hm4@hal.gnu.ai.mit.edu> mycroft@hal.gnu.ai.mit.edu (Charles Hannum) writes: > >In article <1993Apr15.225354.18654@samba.oit.unc.edu> >Brandon.Vanevery@launchpad.unc.edu (Brandon Vanevery) writes: >> >> 386BSD has better TCP/IP, for now. >> >> Linux is far less resource-hungry, due to shared libraries. > >For now. > Linux will use whatever better in 386bsd. But 386bsd has to start from scratch, unless you do it yourself. For example, 387 emulation and C library. H.J. ------------------------------ From: hlu@luke.eecs.wsu.edu (HJ Lu) Subject: Re: possible bug in mount() Date: Sat, 17 Apr 93 16:25:47 GMT In article <9246@blue.cis.pitt.edu> tjd@neurocog.lrdc.pitt.edu (Todd J. Derr) writes: >well, it doesn't seem that passing bad values to mount() should cause a GP >fault, so... (yes, I know that I went about this wrong, but that's not the >point...) > >the program: >#include include >main() >{ mount("/dev/hda1","/", MS_MGC_VAL | MS_REMOUNT); } > Please always use prototype. It will save everybody's time. H.J. ------------------------------ Crossposted-To: comp.os.386bsd.questions From: nate@cs.montana.edu (Nate Williams) Subject: Re: Summary of Linux vs. 386BSD vs. Commercial Unixes Date: Sat, 17 Apr 1993 17:54:31 GMT In article <1993Apr17.161516.2794@serval.net.wsu.edu> hlu@luke.eecs.wsu.edu (HJ Lu) writes: >In article <1qo0lq$1hm4@hal.gnu.ai.mit.edu> mycroft@hal.gnu.ai.mit.edu (Charles Hannum) writes: >> >>In article <1993Apr15.225354.18654@samba.oit.unc.edu> >>Brandon.Vanevery@launchpad.unc.edu (Brandon Vanevery) writes: >>> >>> 386BSD has better TCP/IP, for now. >>> >>> Linux is far less resource-hungry, due to shared libraries. >> >>For now. >> > >Linux will use whatever better in 386bsd. But 386bsd has to >start from scratch, unless you do it yourself. For example, >387 emulation and C library. And then Linux and the GNU folks will take what 386BSD has done and restrict it. Now, isn't that nice....... taking my work and restricting it. I always thought that was awful kind of them. (NOT!) Nate Who really likes GNU utilities, but hates the way alot of folks take other peoples programs and code, fix them and then don't make the changes 'completely redistributable; -- osynw@terra.oscs.montana.edu | Still trying to find a good reason for nate@cs.montana.edu | these 'computer' things. Personally, work #: (406) 994-4836 | I don't think they'll catch on - home #: (406) 586-0579 | Don Hammerstrom ------------------------------ From: gcortevi@nyx.cs.du.edu (Greg Corteville) Subject: Compiling Getty********* Date: Sat, 17 Apr 93 17:28:17 GMT I FTP'd getty_ps2.0.6b.tar.z from TSX-11 yesterday. I was told this would solve my serial port logins by replacing my getty. The thing is, it needs to be compiled. Being new to UNIX in general, I have no idea how to compile it and set it to my specifications. I know it has something to do with Makefile, but that is not executable. Even if it was, it just gives me the famous command not found error. I'd really like to get this serial login thing moving. Please let me know if you can help. My next project is ka9q. -- +---------------------------+----------------------------------------------+ | Greg Corteville | "We are the Knights who say NEE!" | | gcortevi@nyx.cs.du.edu | - Monty Python's QFTHG | +---------------------------+----------------------------------------------+ ------------------------------ From: gcortevi@nyx.cs.du.edu (Greg Corteville) Subject: Changing my address Date: Sat, 17 Apr 93 17:32:20 GMT I discovered the TALK command in Linux and decided to talk to myself. (Yes, I was bored. :)) I found out that my machine name was softland.softland.com (or something like that.) So I decided to poke around in the /etc/inet directory. I found out how to change my IP address and machine name. The thing I was curious about is, can I get rid of my domain name before the login prompt? Also, is the issue file in my /etc directory only displayed for local logins? I tried telnetting to myself and found out it said Linux telnet... -- +---------------------------+----------------------------------------------+ | Greg Corteville | "We are the Knights who say NEE!" | | gcortevi@nyx.cs.du.edu | - Monty Python's QFTHG | +---------------------------+----------------------------------------------+ ------------------------------ From: ianj@citrus.SAC.CA.US ( Ian Justman ) Subject: Re: QIC-02 Tape Patches? Date: Sat, 17 Apr 1993 03:04:35 GMT John Will (john.will@satalink.com) wrote: : Has anyone applied the QIC-02 tape patches to the 99p7A kernel? It's not : clear to me how to get all the patches applied correctly to get tape support : back, I had the tape running fine in 99p6, but now I have three patch files : instead of one, and the first one spits out lots of errors when it runs. : Do I just forge ahead and apply the others, or is there something that : I'm missing here. :-) Most of that stuff can be conveniently ignored; it's almost exclusively revision information. I did have to hand-edit a couple of files to get things working on my system. But I hand-edited all the revision information "just to be safe." BTW, I've got it running just fine with my recently-kibitzed-together setup which includes an Archive 5945L-2 and an Everex EV-811. Hats off to Hennus Bergman to a job well done (so far at least... :-) ). PSS, I'm running 0.99pl7a. -- Born to void warranties! ianj@ijpc.UUCP ------------------------------ From: ianj@citrus.SAC.CA.US ( Ian Justman ) Subject: Re: QIC-02 With 0.99.8 Date: Sat, 17 Apr 1993 03:07:07 GMT Jerry Shekhel (jerry@msi.com) wrote: : Hello folks. Has anyone been successful getting the QIC-02 tape driver : working with Linux kernel 0.99.8? I managed to patch in the driver, and : the resulting kernel can reset and rewind tapes just fine, but screws : up trying to read or write. Yes, I made sure to specify the correct : DMA and IRQ (DMA3 and IRQ5 in my case). Please send me email if you've : been successful. I will be doing that myself shortly. BTW, my setup is DMA3, but my card is set up to IRQ2, but you must sense it at IRQ9 since IRQ2 on the mother- board end is the cascade interrupt. Hennus, I'm running it just fine, but the abovementioned trick MUST be kept in mind. Found that out the hard way when I was dicking around with my XT controller. -- Born to void warranties! ianj@ijpc.UUCP ------------------------------ ** 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-Activists-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux) via: Internet: Linux-Activists@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 tupac-amaru.informatik.rwth-aachen.de pub/msdos/replace The current version of Linux is 0.99pl6 released on February 23, 1993 End of Linux-Activists Digest ****************************** next project isdigest868 644 36676 74430 55061 5366137107 5772 From: Digestifier To: Linux-Activists@senator-bedfellow.mit.edu Reply-To: Linux-Activists@senator-bedfellow.mit.edu Date: Sat, 17 Apr 93 17:00:27 EDT Subject: Linux-Activists Digest #868 Linux-Activists Digest #868, Volume #4 Sat, 17 Apr 93 17:00:27 EDT Contents: ANS: ser. hang under X and QUES: login over serial (Daniel S. Ridge) Problems with 1542 Rev C (Gilbert Callaghan) Re: Which is the best soundcard supported by 386BSD and Linux ? (Greg Lee) Re: Summary of Linux vs. 386BSD vs. Commercial Unixes (Charles Hannum) Re: Compileing Kernels 99.6 99.7 99.8 (Juha Laiho) Re: Summary of Linux vs. 386BSD vs. Commercial Unixes (Lars Wirzenius) Re: how safe is minix filesys (Charles Hannum) [FAQ] Serial logins, UUCP (Stephen Sprunk) Re: Summary of Linux vs. 386BSD vs. Commercial Unixes (Marc WANDSCHNEIDER) Delete Key in Bash? Any way with Termcap? (John Henders) Plea to linux developers! (John Henders) Re: Gateway Nomad Field Mouse (David Fox) Network = Notwork with libc4.3.[3,2] (Tom Van Den Eede) Re: Why Linux should remain Linux and seagulls should not be used. (Simon Patrick Janes) Re: Summary of Linux vs. 386BSD vs. Commercial Unixes (HJ Lu) Re: Summary of Linux vs. 386BSD vs. Commercial Unixes (HJ Lu) Re: Summary of Linux vs. 386BSD vs. Commercial Unixes (Nate Williams) ---------------------------------------------------------------------------- From: newt@eng.umd.edu (Daniel S. Ridge) Subject: ANS: ser. hang under X and QUES: login over serial Date: 17 Apr 1993 07:01:12 GMT I posted yesterday asking for help on my linux communications problems, and I had not realized that X didn't support IRQ sharing, and the mouse on com1 conflicted with the 2400 baud modem on com3 I have neither the time nor the patience to hack IRQ sharing, so I moved the mouse and lost half my modem functionality. Ah well, it is well worth it to run xdaliclock remotely(If slowly! Hint: dont ever do this) Thank you to all who helped me on this, especially Gebran Krikor. I have one more question for the gurus, or even enlightened neophytes. I try to spawn a getty to a serial line to support a terminal, and I encounter difficulties. the getty accepts the login properly, and the login process is spawned on the linux box(as revealed by ps), and the password checking is not handled. Either the password echos on the terminal and does not process, or it never prompts for a password in the first place. This is what I have done so far. I set it(BTW *never* *ever* do this) so that root can login from the serial port, and then root or any other id without passwd authentication works properly. It logs in fine if I dont need to use a password. I have redirected the input , output, and error for a shell over the serial line and that works okay, I have tried the same thing with login, and it exhibits the characteristics of the spawned getty detailed above. This leads me to believe that the problem is in login. Sound good? Have a fix? Thanks in advance for all your help. send replied to newt@eng.umd.edu C C and I have tried the same with login. Login does NOT ------------------------------ From: gilbert@inviso.com (Gilbert Callaghan) Subject: Problems with 1542 Rev C Date: Sat, 17 Apr 1993 16:58:41 GMT Has anybody had any problems running linux with the new revision of the 1542 controller - Rev C? The Rev B board works fine, but when I install the Rev C, it doesn't seem to recognize that a board is even installed and I end up with a kernel panic trying to mount root. I called Adaptec. They don't even sell the Rev B anymore. They claim that the Rev C is 100% downward compatible with Rev B as far as BIOS calls and hardware registers are concerned. Does anyone know where the source code to the adaptec driver is available? Thanks. Oh btw to complicate matters, I'm runnning Linux on a Bernoulli 90MB removable cartridge, but that *shouldn't* cause any problems. -- Gilbert Callaghan gilbert@inviso.com ------------------------------ Crossposted-To: comp.os.386bsd.questions,comp.sys.ibm.pc.soundcard From: lee@Hawaii.Edu (Greg Lee) Subject: Re: Which is the best soundcard supported by 386BSD and Linux ? Date: Sat, 17 Apr 1993 18:17:41 GMT Pim Zandbergen (pim@cti-software.nl) wrote: : As Linux and 386bsd probably do not have equal support, : it's best to rephrase: : : Which is the best soundcard supported by Linux ? I'm currently using both a gus and a pas16 with linux and Hannu Savolainen's sound driver. So I can compare them, but I can't give you a straight answer. Right now, my gus sounds better, except for percussion voices, which come out better on the pas16 -- fm seems to do better at clangy and noisy things. Also, I get more polyphonic voices on the gus. However, I believe a future version of the driver will support more polyphany on the pas16 and also 4-op synthesis (the current version only supports 2-op fm synthesis). At that point, the pas16 might sound overall better -- I can't tell. Now, the driver supports input and output from/to an external midi port only for the pas16, of the two cards, but that will probably change, too. : Which is the best soundcard supported by 386BSD ? Couldn't say. If it's the same driver, maybe the same non-answer would be equally appropriate. -- Greg Lee ------------------------------ From: mycroft@hal.gnu.ai.mit.edu (Charles Hannum) Crossposted-To: comp.os.386bsd.questions Subject: Re: Summary of Linux vs. 386BSD vs. Commercial Unixes Date: 17 Apr 1993 15:01:58 -0400 In article <1993Apr17.161516.2794@serval.net.wsu.edu> hlu@luke.eecs.wsu.edu (HJ Lu) writes: > > Linux will use whatever better in 386bsd. But 386bsd has to start > from scratch, unless you do it yourself. For example, 387 emulation > and C library. Why are you trying to confuse the issue? Yes, Linux will (and has) taken things from 386BSD when it's convenient. Likewise, 386BSD will (and has) taken things from Linux. You make it sound like 386BSD *can't* use code from Linux. This is bogus. It's just very carefully chosen which parts are worthwhile, which aren't, and which need to be redone. -- \ / Charles Hannum, mycroft@ai.mit.edu /\ \ PGP public key available on request. MIME, AMS, NextMail accepted. Scheme White heterosexual atheist male (WHAM) pride! ------------------------------ From: jlaiho@ichaos.nullnet.fi (Juha Laiho) Subject: Re: Compileing Kernels 99.6 99.7 99.8 Date: Sat, 17 Apr 1993 09:57:37 GMT In article <93106.100712BTITMARS@ESOC.BITNET> BARRY TITMARSH writes: >I have seen lots of postings about problems compileing 99.6 to 99.8 >kernels. Just like to say I have had no problems with any of the >kernels or any odd configs etc with 0.99.x release. >I will point out im still useing GCC 2.2.2d7 and Libc.4.2 etc.. >since i have not seen any problems with my config and compileing >and i have seen problems with others useing GCC 2.3 / 2.4 and newer libc's >Im wondering if there are some problems with the New Compilers.? I've been able to keep up with the new kernel and library releases most of the time, and have had almost no problems. I missed the actual 4.3 version of the C library, and didn't upgrade when I read about the problems. After I saw the announcement for 4.3.2 libraries, I upgraded quite soon. Again, no problems. ..Wolf ------------------------------ Crossposted-To: comp.os.386bsd.questions From: wirzeniu@klaava.Helsinki.FI (Lars Wirzenius) Subject: Re: Summary of Linux vs. 386BSD vs. Commercial Unixes Date: Sat, 17 Apr 1993 19:30:29 GMT nate@cs.montana.edu (Nate Williams) writes: >And then Linux and the GNU folks will take what 386BSD has done and >restrict it. Now, isn't that nice....... taking my work and restricting >it. I always thought that was awful kind of them. (NOT!) I would be interested if you would be kind enough to elaborate on this. As far as I know, nothing that Linux or GNU (which have little to do with each other, except that Linux uses a lot of GNU programs) has imported has been made available under a different copyright notice than what the copyright holder (typically the author) has agreed to. Anything else would be illegal. One similar example would be when 386bsd imported Linus' math emulator: the copyright was changed to the same style that the rest of the 386bsd kernel uses, when it originally was under the GPL. (The change was done with Linus' complete approval, I haste to add.) (The article I'm replying to might have been a complete flame-bait, articles of that type usually are, but sometimes I can't resist answering such articles. Color me a sucker.) -- Lars.Wirzenius@helsinki.fi (finger wirzeniu@klaava.helsinki.fi) MS-DOS, you can't live with it, you can live without it. ------------------------------ From: mycroft@hal.gnu.ai.mit.edu (Charles Hannum) Subject: Re: how safe is minix filesys Date: 17 Apr 1993 15:20:31 -0400 In article <14206.1189.uupcb@satalink.com> john.will@satalink.com (John Will) writes: > > Can you say UPS? `A hardware solution to a software bug.' Yah, I can say it. It is well known how to make crash-proof *and* fairly efficient file systems. While I won't pick on Linux for this, I will say that I would never consider buying a commercial system with an unreliable file system again. -- \ / Charles Hannum, mycroft@ai.mit.edu /\ \ PGP public key available on request. MIME, AMS, NextMail accepted. Scheme White heterosexual atheist male (WHAM) pride! ------------------------------ From: ssprunk@nyx.cs.du.edu (Stephen Sprunk) Subject: [FAQ] Serial logins, UUCP Date: Sat, 17 Apr 93 19:20:14 GMT I have installed SLS 0.99pl6 on my 386 and have been looking around in the system to see what all it can do. I know you guys probably answer these questions ten million times a day, but: 1) How can I set up Linux to access serial logins? I have a modem on /dev/ttyS2 (/dev/modem). How do I set my machine up to automatically let users login? I have the modem set for S0=1 and it will connect fine with other modems, but it won't give a login prompt. 2) How do I get UUCP set up? I want to be able to receive news (10-15 groups) and mail on my box at home. I also want to be able to distribute news and mail to friends running Linux in my area who don't have connections like I do. I have an account at Denver Univ. I can dial into to get stuff from using kermit, but how do I get stuff using UUCP and (I hope) cron? Stephen -- -- Stephen Sprunk \ "All I want is peace on earth and goodwill ssprunk@nyx.cs.du.edu \ towards men." -- Whistler Voice: (713) 893-1225 \ "We're the government, we don't do that ------------------------------ Crossposted-To: comp.os.386bsd.questions From: storm@cs.mcgill.ca (Marc WANDSCHNEIDER) Subject: Re: Summary of Linux vs. 386BSD vs. Commercial Unixes Date: Sat, 17 Apr 1993 19:42:35 GMT In article <1qpk76$bbl@hal.gnu.ai.mit.edu> mycroft@hal.gnu.ai.mit.edu (Charles Hannum) writes: > >You make it sound like 386BSD *can't* use code from Linux. This is >bogus. It's just very carefully chosen which parts are worthwhile, >which aren't, and which need to be redone. > On this note, I'm curious to hear if there is any plan to make shared libraries standard in 386bsd...? I have seen one patch for this out on the net, but I don't know if many people are using it, and if wfj plans on adding this feature in 0.2 (official)... Anybody a little more clued in on this one than I? toodlepip! Marc 'em. ------------------------------ From: jhenders@wimsey.bc.ca (John Henders) Subject: Delete Key in Bash? Any way with Termcap? Date: Sat, 17 Apr 1993 20:03:36 GMT As the subject says, is there any way to get the delete key to work with bash, as supplied on the .99-6 SLS release, or do I have to get the source and fix it that way? -- John Henders ------------------------------ From: jhenders@wimsey.bc.ca (John Henders) Subject: Plea to linux developers! Date: Sat, 17 Apr 1993 20:07:10 GMT After several frustraiong experiences with linux software, I'd like to propose that all linux authors and porters try to put a version number in their binaries that can be called with -v, -V, -h or -?. It would be to everyone's benifit if there was an easy way to find out what version of a package, instead of sometimes having to go by the file datw or what directory on what archive site it was on. What do people think? -- John Henders ------------------------------ From: fox@graphics.cs.nyu.edu (David Fox) Subject: Re: Gateway Nomad Field Mouse Date: Sat, 17 Apr 1993 20:41:50 GMT In article <1993Apr16.091220.21224@anl433.uucp> Iain.Lea@anl433.erlm.siemens.de (Iain Lea) writes: Paul Fishwick (fishwick@fish.cis.ufl.edu) wrote: : : I am looking into putting Linux on a Gateway Nomad 486 DX2/50 and before I : do, I'd like to know whether anyone else has put Linux on this machine. : I checked the compatibility chart but did not see a "Nomad" or : "Texas Instruments" machine there (the Gateway machines being repackaged : TI machines). Specifically, the Nomad comes with a "field mouse" (whatever : that is :). Does the field mouse emulate the Microsoft bus or serial mouse? I have 0.99p6 running on a TI4000 DX2/50. Tried installing X but could not get the mouse to work but then again I only tried for a short while. I have 0.99p6 running on a TI4000 DX2/50, and I have installed X but I too can't get the PS/2 mouse running (and I've tried for longer than Iain.) The standard solutions for the PS/2 mouse don't work on the TI/Gateway (mconv, kernal patch to send init sequence) So, I have a mouse in the 9 pin serial port except for the times when I kill X so I can run my modem. ------------------------------ From: vdeede@wins.uia.ac.be (Tom Van Den Eede) Subject: Network = Notwork with libc4.3.[3,2] Date: Sat, 17 Apr 1993 19:28:56 GMT Since I'he installed libc4.3.2 (and 4.3.3 afterwards), all my network programs seem to fail. Functions as 'gethostbyname' 'getprotocolbyname' and others, always return NULL-pointers. Are there any patches to solve this problem ?!? TOM vdeede@wins.uia.ac.be ------------------------------ From: spj@ukelele.GCR.COM (Simon Patrick Janes) Subject: Re: Why Linux should remain Linux and seagulls should not be used. Date: Sat, 17 Apr 1993 18:12:45 GMT mdw@theory.TC.Cornell.EDU (Matt Welsh) writes: >In article , spj@ukelele.GCR.COM (Simon Patrick Janes) writes: >> I don't like the seagull idea however. Seagulls are too common I think... >> I don't know... it just makes me think what America would have been like >> if the national bird was the turkey like it almost became. I would rather >> have a platypus than a seagull. >Folks, >Michael K. Johnson and I have decided to use the Linux seagull logo on the >title page of the Linux Documentation Project manuals as well as other >works. We're working right now on integrating the logo with the current >linuxdoc.sty LaTeX style format used by the LDP. Ok, this is OK by me, I just didn't want the "sea-gull" to be *the* Linux Logo, but the sea-gull as the Linux Doc. Proj. logo is very good. (I'm thinking of the BSD Devil-with-sneakers book and comparing it with the sea-gull.) >While this by no means makes it "official", it's about as official as you >can get in the Linux community. The seagull represents many things about >Linux: its size (Linux is a small UNIX), freedom, internationalization (where >in the world aren't seagulls found?), flexibility, and, well... flight (from >the world of DOS). If any animal or design represents Linux it is the seagull. >Although I'm probably not interested in receiving tons of email trying to >persuade me out of this decision, if you design a good-looking Linux logo >depicting some other animal (or no animal at all?) feel free to send it >along. PostScript is fine, but MetaFont source would be better. The seagull >has so far been the most appropriate available design, and we really liked >it. A platypus? Okay, if the mammal body represents System V and the duckbill >represents BSD, then you may have something there... :) I could never design a good logo. If nothing better comes up, the sea-gull will probably end up being it. (Man, but they're so common.. ;) >Cheers, >mdw >-- >Matt Welsh, mdw@tc.cornell.edu >"I met a girl named Sandoz..." -- ========================================================================= Simon "Guru Aleph-Null" Janes |Taco Bell Otaku-zoku... | | do they exist? Huh? Huh? ------------------------------ Crossposted-To: comp.os.386bsd.questions From: hlu@luke.eecs.wsu.edu (HJ Lu) Subject: Re: Summary of Linux vs. 386BSD vs. Commercial Unixes Date: Sat, 17 Apr 93 19:05:17 GMT In article <1993Apr17.175431.25015@coe.montana.edu> nate@cs.montana.edu (Nate Williams) writes: >In article <1993Apr17.161516.2794@serval.net.wsu.edu> hlu@luke.eecs.wsu.edu (HJ Lu) writes: >>In article <1qo0lq$1hm4@hal.gnu.ai.mit.edu> mycroft@hal.gnu.ai.mit.edu (Charles Hannum) writes: >>> >>>In article <1993Apr15.225354.18654@samba.oit.unc.edu> >>>Brandon.Vanevery@launchpad.unc.edu (Brandon Vanevery) writes: >>>> >>>> 386BSD has better TCP/IP, for now. >>>> >>>> Linux is far less resource-hungry, due to shared libraries. >>> >>>For now. >>> >> >>Linux will use whatever better in 386bsd. But 386bsd has to >>start from scratch, unless you do it yourself. For example, >>387 emulation and C library. > >And then Linux and the GNU folks will take what 386BSD has done and >restrict it. Now, isn't that nice....... taking my work and restricting >it. I always thought that was awful kind of them. (NOT!) > > >Nate > >Who really likes GNU utilities, but hates the way alot of folks take >other peoples programs and code, fix them and then don't make the >changes 'completely redistributable; If everybody did that, the GNU copyright would not be a problem. Is your purpose to let someone make "completely non-redistributable" stuff? Am I missing something? You have gave us freedom (Thanks). We have made our decision and you can get our code for free with some reasonable conditions. Why do you complain? That is what you ask for. Please blame it on yourself. If someone did something to your code and didn't tell your about that or wanted to charge you $$$$ or didn't want you to give it away to others, would you complain? H.J. ------------------------------ Crossposted-To: comp.os.386bsd.questions From: hlu@eecs.wsu.edu (HJ Lu) Subject: Re: Summary of Linux vs. 386BSD vs. Commercial Unixes Date: Sat, 17 Apr 93 19:43:41 GMT In article <1qpk76$bbl@hal.gnu.ai.mit.edu>, mycroft@hal.gnu.ai.mit.edu (Charles Hannum) writes: |> |> In article <1993Apr17.161516.2794@serval.net.wsu.edu> |> hlu@luke.eecs.wsu.edu (HJ Lu) writes: |> > |> > Linux will use whatever better in 386bsd. But 386bsd has to start |> > from scratch, unless you do it yourself. For example, 387 emulation |> > and C library. |> |> Why are you trying to confuse the issue? Yes, Linux will (and has) |> taken things from 386BSD when it's convenient. Likewise, 386BSD will |> (and has) taken things from Linux. |> |> You make it sound like 386BSD *can't* use code from Linux. This is No. I am trying to say the "official" 386bsd release is not willing to use GNU copyrighted code. Of courses, you can make/release your own 386bsd :-). I know you have done that for yourself. |> bogus. It's just very carefully chosen which parts are worthwhile, ^^^^^^^^^^^^ That means no GNU copyrighted code in the "official" 386bsd release. BTW, I'd love to see the "official" 386bsd release uses GNU copyrighted code. |> which aren't, and which need to be redone. |> H.J. ------------------------------ Crossposted-To: comp.os.386bsd.questions From: nate@cs.montana.edu (Nate Williams) Subject: Re: Summary of Linux vs. 386BSD vs. Commercial Unixes Date: Sat, 17 Apr 1993 20:57:15 GMT HJ - Linux will use whatever better in 386bsd. But 386bsd has to HJ - start from scratch, unless you do it yourself. For example, HJ - 387 emulation and C library. Nate - And then Linux and the GNU folks will take what 386BSD has done and Nate - restrict it. Now, isn't that nice....... taking my work and restricting Nate - it. I always thought that was awful kind of them. (NOT!) Nate - Nate - Who really likes GNU utilities, but hates the way alot of folks take Nate - other peoples programs and code, fix them and then don't make the Nate - changes 'completely redistributable; HJ - HJ - If everybody did that, the GNU copyright would not be a problem. Is HJ - your purpose to let someone make "completely non-redistributable" HJ - stuff? Am I missing something? No, my purpose is to make 386BSD completely re-distrubutable, with NO strings attached. That means Sun, DEC, HP, Ren and Stimpy, or whoever can take this code and sell a binary copy of it. The GPL does not allow this. HJ - You have gave us freedom (Thanks). We have made our decision and you HJ - can get our code for free with some reasonable conditions. Why do you HJ - complain? That is what you ask for. Please blame it on yourself. If HJ - someone did something to your code and didn't tell your about that HJ - or wanted to charge you $$$$ or didn't want you to give it away to HJ - others, would you complain? No, I would not complain. I believe the work should be in the same vein as it was distributed. COMPLETELY AND UTTERLY free. Not copylefted, which places restrictions on its distribution. The original BSD copyright has been this way, but unfortunately a group of people take the code, fix the code, and then place restrictions on it. Nate -- osynw@terra.oscs.montana.edu | Still trying to find a good reason for nate@cs.montana.edu | these 'computer' things. Personally, work #: (406) 994-4836 | I don't think they'll catch on - home #: (406) 586-0579 | Don Hammerstrom ------------------------------ ** 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-Activists-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux) via: Internet: Linux-Activists@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 tupac-amaru.informatik.rwth-aachen.de pub/msdos/replace The current version of Linux is 0.99pl6 released on February 23, 1993 End of Linux-Activists Digest ****************************** Unixes Date: Sat, 17 Apr 93 19:43:41 GMT In article <1qpk76$bbl@hal.gnu.ai.mit.edu>, mycroft@hal.gnu.ai.mit.edu (Charles Hannum) writes: |> |> In article <1993Apr17.161516.2794@serval.net.wsu.edu> |> hlu@luke.eecs.wsu.edu (HJ Lu) writes: |> > |> > Linux will use whatever better in 386bsd. But 386bsd has to start |> > from scratch, unless you do it yourself. For example, 387 emulation |> > and C library. |> |> Why are you trying to confuse the issue? Yes, digest869 644 36676 74430 56526 5366137107 6002 From: Digestifier To: Linux-Activists@senator-bedfellow.mit.edu Reply-To: Linux-Activists@senator-bedfellow.mit.edu Date: Sat, 17 Apr 93 19:30:13 EDT Subject: Linux-Activists Digest #869 Linux-Activists Digest #869, Volume #4 Sat, 17 Apr 93 19:30:13 EDT Contents: Re: Questions about virtual mem. (Yonik Christopher Seeley) Re: Summary of Linux vs. 386BSD vs. Commercial Unixes (Nate Williams) Re: Gateway Nomad Field Mouse (David Fox) Re: Xtoolplaces from new (3L4) xview release (Thomas Davis) doc dumps core (Ray Mendonsa) Bernoulli Box with Linux? *** (Fractalman) Re: Can't login with 0.99 p8 (Barracuda) linux on PS2 MCA & SCSI disk (Peter C. Tam) Re: Bernoulli Box with Linux? *** (Drew Eckhardt) Re: linux on PS2 MCA & SCSI disk (Drew Eckhardt) Re: Linux on the Amiga? (Martin P. Ibert) Re: Summary of Linux vs. 386BSD vs. Commercial Unixes (Doug Quale) Re: Is curses.h complete? (Zeyd M. Ben-Halim) Color X-term (Steve Fuller) Re: Is curses.h complete? (Brandon S. Allbery) Re: possible bug in mount() (Brandon S. Allbery) Re: Summary of Linux vs. 386BSD vs. Commercial Unixes (Charles Hannum) Re: Summary of Linux vs. 386BSD vs. Commercial Unixes (Charles Hannum) Re: Summary of Linux vs. 386BSD vs. Commercial Unixes (Joe Emenaker) MINIX DEFRAG errors -- trashed all my static libs. (Scott D. Heavner) ---------------------------------------------------------------------------- From: yseeley@leland.Stanford.EDU (Yonik Christopher Seeley) Subject: Re: Questions about virtual mem. Date: Sat, 17 Apr 93 20:48:10 GMT >Yes, Linux does support virtual memory, in the form of a swap partition. Virtual memory is much more than just adding some disk space to physical ram. Virtual memory without any swap space enabled at all would still be very useful. - Yonik Seeley yseeley@cs.stanford.edu ------------------------------ Crossposted-To: comp.os.386bsd.questions From: nate@cs.montana.edu (Nate Williams) Subject: Re: Summary of Linux vs. 386BSD vs. Commercial Unixes Date: Sat, 17 Apr 1993 21:03:03 GMT In article <1993Apr17.193029.5707@klaava.helsinki.fi> wirzeniu@klaava.Helsinki.FI (Lars Wirzenius) writes: >nate@cs.montana.edu (Nate Williams) writes: >>And then Linux and the GNU folks will take what 386BSD has done and >>restrict it. Now, isn't that nice....... taking my work and restricting >>it. I always thought that was awful kind of them. (NOT!) > >I would be interested if you would be kind enough to elaborate on >this. As far as I know, nothing that Linux or GNU (which have little >to do with each other, except that Linux uses a lot of GNU programs) >has imported has been made available under a different copyright >notice than what the copyright holder (typically the author) has >agreed to. Anything else would be illegal. The stdio code that HJ has done lots of work on(and I commend him on it). The original code, before he fixed some of the bugs, was originally distributed by AT&T, but HJ's fixes are now GPL code, so in order to get working stdio code we have to do those fixes all over in order to make anything compiled against those libraries ABSOLUTELY FREELY REDISTRUTABLE. > >One similar example would be when 386bsd imported Linus' math >emulator: the copyright was changed to the same style that the rest of >the 386bsd kernel uses, when it originally was under the GPL. (The >change was done with Linus' complete approval, I haste to add.) And I commed Linus, and his work highly for that. I wish that more of the Linux people were more willing to do that. (And to be honest, I haven't asked very many folks about this). But, some people believe with distributing anything but GPL code, because the 'end all/be all' way to distribute code is to GPL/LGPL it. I disagree, and am considered a 'bad guy' because I distribute my fixes w/out any added restrictions that the original code had. Flames by email, Nate -- osynw@terra.oscs.montana.edu | Still trying to find a good reason for nate@cs.montana.edu | these 'computer' things. Personally, work #: (406) 994-4836 | I don't think they'll catch on - home #: (406) 586-0579 | Don Hammerstrom ------------------------------ From: fox@graphics.cs.nyu.edu (David Fox) Subject: Re: Gateway Nomad Field Mouse Date: Sat, 17 Apr 1993 21:26:07 GMT In article fox@graphics.cs.nyu.edu (David Fox) writes: In article <1993Apr16.091220.21224@anl433.uucp> Iain.Lea@anl433.erlm.siemens.de (Iain Lea) writes: Paul Fishwick (fishwick@fish.cis.ufl.edu) wrote: : : I am looking into putting Linux on a Gateway Nomad 486 DX2/50 : and before I do, I'd like to know whether anyone else has put : Linux on this machine. I checked the compatibility chart but : did not see a "Nomad" or "Texas Instruments" machine there : (the Gateway machines being repackaged TI machines). : Specifically, the Nomad comes with a "field mouse" (whatever : that is :). Does the field mouse emulate the Microsoft bus or : serial mouse? I have 0.99p6 running on a TI4000 DX2/50. Tried installing X but could not get the mouse to work but then again I only tried for a short while. I have 0.99p6 running on a TI4000 DX2/50, and I have installed X but I too can't get the PS/2 mouse running (and I've tried for longer than Iain.) The standard solutions for the PS/2 mouse don't work on the TI/Gateway (mconv, kernal patch to send init sequence) So, I have a mouse in the 9 pin serial port except for the times when I kill X so I can run my modem. One piece of good news: I can set my machine up to run X-1.2 with the PS/2 mouse. The mouse won't work, but I can create two windows (one for my modem) and bind some functions keys to switch between them. Unfortunately when I exit X my machine dies, but otherwise it is an improvement (I don't use the mouse much anyway...) -david ------------------------------ From: tdavis@beeble.omahug.org (Thomas Davis) Subject: Re: Xtoolplaces from new (3L4) xview release Date: Sat, 17 Apr 1993 12:21:54 GMT khamer@sumax.seattleu.edu writes: >a) writes a 0k file .openwin-init if I run it using the "owplaces" > script provided, which is fine and dandy except I have 5 apps running and > it should mention them in the file >b) dumps core if I start it as xtoolplaces I'll bet you have Xsysinfo running. I just checked it here, and that program cause xtoolplaces to core dump. Haven't checked why yet.. -- Thomas Davis, ECNE | Internet: tdavis@beeble.omahug.org Systems Engineer, | UUCP: uunet!ivgate!beeble!tdavis Huntel Communications, Inc. | Snail Mail: 8803 Webster Plaza Omaha, NE | Omaha, NE 68114 ------------------------------ From: ray@rgm2.rgm.com (Ray Mendonsa) Subject: doc dumps core Date: Sat, 17 Apr 1993 20:50:27 GMT Every binary I get for doc dumps core with Segmentation Fault on startup before opening its window. Other interviews applications work ok, including idraw. I've tried the one that came with the current SLS distribution and another I grabbed from sunsite (IVdoc3.1.tar.Z) I'm using v0.99.8-56 kernel on a 386-33 w/387. Any tips/pointers/flames deeply appreciated via email. Thanks. -- /-Ray Mendonsa--ray@rgm.com--Administrator, rgm.com--Sacramento, CA---------> | 916/923-5013 8n1, v.32bis ; login: new ; Public access Usenet * I ain't | Anon UUCP: system rgm; chat ogin: nuucp ; myname bbsuser * a cheesedog | get ~/ls-lR.Z for file list; PGP pubkey avail req or finger rgm@netcom.com ------------------------------ From: txk1109@ultb.isc.rit.edu (Fractalman) Subject: Bernoulli Box with Linux? *** Date: Sat, 17 Apr 1993 21:09:52 GMT Is it possible to use an Iomega Bernoulli Box under Linux? How about Iomega's LaserSafe? (LaserSafe is an optical drive but behaves somewhat like a Bernoulli Box and uses same drivers.) I know it might be possible using an Adaptec or other SCSI card but how about with Iomega's own SCSI card? Anyone tried yet? Thanks in advance for any responses. E-mail or post are fine. -Fractalman txk1109@ultb.isc.rit.edu ------------------------------ From: ez033209@chip.ucdavis.edu (Barracuda) Subject: Re: Can't login with 0.99 p8 Date: Sat, 17 Apr 1993 21:52:00 GMT ez033209@bullwinkle.ucdavis.edu (Barracuda) writes: : I've got a bit of a problem here...I just compiled the 0.99 patchlevel 8 : kernel and installed it (and re-installed lilo) and when I reboot, lilo : loads the kernel fine, and Linux starts up fine and gets to the Login: : prompt, but thats where my problems start: no matter what I try to login : as, the system waits for a second and redisplays the login prompt again. : If the account that I'm trying to use has a password, I do NOT get a : password: prompt. : This is only a problem under my newly compiled 0.99.8 kernel...when I : boot up my old 0.99.3 kernel (from lilo or from floppy) everything works : fine. : "halt" at the login: prompt DOES work, strangely enough...its about all that : does. Well, I hate to followup to my own post, but I took care of the problem. Seems that getty or login were messed up. I recompiled everything in my /etc directory (for good measure :^) with the latest versions and the problem went away. Thanks to all who mailed me suggestions! -Rob ------------------------------ From: br.pct@RLG.Stanford.EDU (Peter C. Tam) Subject: linux on PS2 MCA & SCSI disk Date: Sat, 17 Apr 1993 18:38:07 GMT Hi, Pardon me if it is a FAQ, but: Can linux run off IBM MicroChannel PS2s, with SCSI disks? & How stable & performance is the TCP/IP, NFS, ARPA services like telnet, ftp, ping, lpr, lpd, etc.? Thanks in advance!! ------------------------------ From: drew@ladymacb.cs.colorado.edu (Drew Eckhardt) Subject: Re: Bernoulli Box with Linux? *** Date: Sat, 17 Apr 1993 21:59:57 GMT In article <1993Apr17.210952.10135@ultb.isc.rit.edu> txk1109@ultb.isc.rit.edu (Fractalman) writes: > >Is it possible to use an Iomega Bernoulli Box under Linux? Yes, as long as you're running it off of a supported SCSI board. >How about Iomega's LaserSafe? (LaserSafe is an optical drive but >behaves somewhat like a Bernoulli Box and uses same drivers. People have successfully used other MO drives. > >I know it might be possible using an Adaptec or other SCSI card but how >about with Iomega's own SCSI card? No. -- Boycott USL/Novell for their absurd anti-BSDI lawsuit. | Drew Eckhardt Condemn Colorado for Amendment Two. | drew@cs.Colorado.EDU Use Linux, the fast, flexible, and free 386 unix | ------------------------------ From: drew@ladymacb.cs.colorado.edu (Drew Eckhardt) Subject: Re: linux on PS2 MCA & SCSI disk Date: Sat, 17 Apr 1993 22:30:27 GMT In article br.pct@RLG.Stanford.EDU (Peter C. Tam) writes: >Hi, > Pardon me if it is a FAQ, but: > > Can linux run off IBM MicroChannel PS2s, with SCSI disks? No. > & How stable & performance is the TCP/IP, NFS, ARPA services like > telnet, ftp, ping, lpr, lpd, etc.? Packet fragmenetation isn't supported, but otherwise everything works reasonably well. -- Boycott USL/Novell for their absurd anti-BSDI lawsuit. | Drew Eckhardt Condemn Colorado for Amendment Two. | drew@cs.Colorado.EDU Use Linux, the fast, flexible, and free 386 unix | ------------------------------ Crossposted-To: comp.sys.amiga.unix From: martini@heaven7.in-berlin.de (Martin P. Ibert) Subject: Re: Linux on the Amiga? Date: Sat, 17 Apr 1993 21:26:41 GMT In , Larry Granroth (ljg@space.physics.uiowa.edu) writes: : : a few people hoping to bring Linux up on Sun 3's. I haven't heard : : from any Macintosh folks. In all probability only the startup code, : ^^^^^^^^^ : GNU software is a major component of Linux. Please consider the FSF's : wishes wrt porting any of the GNU software to A.p.p.l.e. platforms. : Here are a few excerpts: The FSF is in a hell of a fix here, since their own rules (GPL) gives anyone the right to port their software to Apple computers, and to support and maintain those ports. It is not prohibited to port Linux, Emacs or what- have-you to Apple boxes. The FSF doesn't like it, but then again, they can't do a thing about it. -- |||| ~~~~ ALWAYS ~~~~ | Martin P. Ibert, Fürstenweg 11 | __ |||| |\/| ~~~~ LOOK ON ~~~ | D-13589 Berlin, old: D-1000B20 | ( )__ |\/| |/\| ~~ THE BRIGHT ~~ | Fon +4930-3753479 Fax -3753699 | ( )_ |/\| |||| ~ SIDE OF LIFE ~ | | (_________) |||| ------------------------------ Crossposted-To: comp.os.386bsd.questions From: quale@spock.cs.wisc.edu (Doug Quale) Subject: Re: Summary of Linux vs. 386BSD vs. Commercial Unixes Date: Sat, 17 Apr 1993 23:46:07 GMT In article <1993Apr17.210303.12001@coe.montana.edu> nate@cs.montana.edu (Nate Williams) writes: The stdio code that HJ has done lots of work on(and I commend him on it). The original code, before he fixed some of the bugs, was originally distributed by AT&T, but HJ's fixes are now GPL code, so in order to get working stdio code we have to do those fixes all over in order to make anything compiled against those libraries ABSOLUTELY FREELY REDISTRUTABLE. Well Nate, the stdio code in Linux is GNU iostream, part of the g++ library, has always been under the GPL, and was *never* distributed by AT&T. Of course your argument is so confused that I don't expect you to pay any attention to accuracy or the facts in your posts, but I thought I should point that out for the benefit of the people who aren't familiar with the Linux libraries. -- Doug Quale quale@spock.cs.wisc.edu ------------------------------ From: zmbenhal@netcom.com (Zeyd M. Ben-Halim) Subject: Re: Is curses.h complete? Date: Sat, 17 Apr 1993 22:29:48 GMT In article <1993Apr17.080922.1028@lakes.trenton.sc.us> mlh@lakes.trenton.sc.us writes: > >SLS .99pl6 (A disks) pl4 (rest of disks) > >When trying to compile a "door" type chess game for the bbs, the compile >failed when it encountered A_BOLD and some other, what I understand to >be, curses functions. The curses man page says these are defined in >curses.h, but grepping it and all the .h files turned up no match. Are >these commands defined somewhere? For some reason the man pag refers to SYSVR3.2 curses whilst the library is in fact BSD. I suggest you get ncurses (which is SYSVR4-compatible) from netcom.com:pub/zmbenhal. Zeyd ------------------------------ From: s_fuller@iastate.edu (Steve Fuller) Subject: Color X-term Date: Sat, 17 Apr 1993 22:57:06 GMT Does anyone have a compiled (binary) color xterm that they would be willing to tar and send to me? I'm having problems locating one for use with Seyon. thanks -- ==================================================== Steve Fuller = s_fuller@iastate.edu = No witty quote here yet... =================================================== ------------------------------ From: bsa@kf8nh.wariat.org (Brandon S. Allbery) Subject: Re: Is curses.h complete? Date: Sat, 17 Apr 1993 21:58:53 GMT In article <1993Apr17.080922.1028@lakes.trenton.sc.us> mlh@lakes.trenton.sc.us writes: >When trying to compile a "door" type chess game for the bbs, the compile >failed when it encountered A_BOLD and some other, what I understand to >be, curses functions. The curses man page says these are defined in >curses.h, but grepping it and all the .h files turned up no match. Are >these commands defined somewhere? There are two variants of curses: the BSD one, which uses termcap and only supports standout(), and the System V one, which uses terminfo and has support for multiple attributes like A_BOLD, etc. The standard Linux curses is the BSD one. There is a port (and general improvement) of Pavel Curtis's "ncurses" available. It's still under development, so there may be occasional problems; for example, I just discovered today that its color support appears to be hardwired for ANSI terminals (no setf, setb entries). "ncurses" is based on the System V curses, so it supports multiple attributes, color, etc. netcom.com:pub/zmbenhal/ncurses0.7.tar.Z ++Brandon -- Brandon S. Allbery bsa@kf8nh.wariat.org It's not too late to turn back from the "Gates" of Hell... Linux: the FREE 32-bit operating system, available NOW. Why waaaaaait for NT? ------------------------------ From: bsa@kf8nh.wariat.org (Brandon S. Allbery) Subject: Re: possible bug in mount() Date: Sat, 17 Apr 1993 22:00:03 GMT In article <1993Apr17.162547.3040@serval.net.wsu.edu> hlu@luke.eecs.wsu.edu (HJ Lu) writes: >In article <9246@blue.cis.pitt.edu> tjd@neurocog.lrdc.pitt.edu (Todd J. Derr) writes: >>well, it doesn't seem that passing bad values to mount() should cause a GP >>fault, so... (yes, I know that I went about this wrong, but that's not the >>point...) >> >>{ mount("/dev/hda1","/", MS_MGC_VAL | MS_REMOUNT); } > >Please always use prototype. It will save everybody's time. Reread the first paragraph, H.J. I don't care *how* screwed up the user-mode code is, the kernel should NEVER panic as a result of it. ++Brandon -- Brandon S. Allbery bsa@kf8nh.wariat.org It's not too late to turn back from the "Gates" of Hell... Linux: the FREE 32-bit operating system, available NOW. Why waaaaaait for NT? ------------------------------ From: mycroft@hal.gnu.ai.mit.edu (Charles Hannum) Crossposted-To: comp.os.386bsd.questions Subject: Re: Summary of Linux vs. 386BSD vs. Commercial Unixes Date: 17 Apr 1993 19:06:32 -0400 You are just muddying the issue by throwing in random flame baits. -- \ / Charles Hannum, mycroft@ai.mit.edu /\ \ PGP public key available on request. MIME, AMS, NextMail accepted. Scheme White heterosexual atheist male (WHAM) pride! ------------------------------ From: mycroft@hal.gnu.ai.mit.edu (Charles Hannum) Crossposted-To: comp.os.386bsd.questions Subject: Re: Summary of Linux vs. 386BSD vs. Commercial Unixes Date: 17 Apr 1993 19:13:00 -0400 In article <1993Apr17.194341.4562@serval.net.wsu.edu> hlu@eecs.wsu.edu (HJ Lu) writes: > > That means no GNU copyrighted code in the "official" 386bsd release. You are full of shit. The `official' release include GCC, GDB, GNU tar, and several other pieces of GNU code. -- \ / Charles Hannum, mycroft@ai.mit.edu /\ \ PGP public key available on request. MIME, AMS, NextMail accepted. Scheme White heterosexual atheist male (WHAM) pride! ------------------------------ From: jemenake@trumpet.calpoly.edu (Joe Emenaker) Crossposted-To: comp.os.386bsd.questions Subject: Re: Summary of Linux vs. 386BSD vs. Commercial Unixes Date: 17 Apr 93 23:10:00 GMT nate@cs.montana.edu (Nate Williams) writes: >HJ - Linux will use whatever better in 386bsd. But 386bsd has to >HJ - start from scratch, unless you do it yourself. For example, >HJ - 387 emulation and C library. > >Nate - And then Linux and the GNU folks will take what 386BSD has done and >Nate - restrict it. Now, isn't that nice....... taking my work and restricting >Nate - it. I always thought that was awful kind of them. (NOT!) >Nate - >Nate - Who really likes GNU utilities, but hates the way alot of folks take >Nate - other peoples programs and code, fix them and then don't make the >Nate - changes 'completely redistributable; > >HJ - >HJ - If everybody did that, the GNU copyright would not be a problem. Is >HJ - your purpose to let someone make "completely non-redistributable" >HJ - stuff? Am I missing something? > >No, my purpose is to make 386BSD completely re-distrubutable, with NO >strings attached. That means Sun, DEC, HP, Ren and Stimpy, or whoever >can take this code and sell a binary copy of it. The GPL does not allow ^^^^ >this. I have a big problem with that and I'll tell you why. This is how the whole GNU project was explained to me by a good friend. Forgive me if it's not exactly right: Apparently the guy who wrote Emacs released it into the public domain. He made it completely redistributable. As the story was told to me, DEC Corporation saw Emacs and liked it and decided that they were going to take it. I don't recall if they just started selling it or if they somehow "assumed" the rights of emacs. This pissed the guy off... who decided he was going to write a NEW or "GNU" version of emacs and was going to reserve the rights just enough so that nobody else could claim the rights to it. Now, it really, Really, REALLY angers me to think of these big corporations taking public-domain and otherwise free software and distributing it as their own and actually getting money for it. How DEVOID of work-ethic does some have to be to pull a stunt like that? And you're saying that you're pleased as punch if DEC can just ftp a copy of 386BSD and start selling it for $500/copy as DEC-BSD/PC or something?!?! That makes me ill. It really does. When I code stuff and release it to the public domain, I stipulate that the stuff is free and is free to modify, but ANYTHING that is derived from my code or that USES the binaries has to be free as well. I think that the GNU agreement is something similar to this. Now, if a company wants to charge for SUPPORT, or if they want to charge a media fee for distribution.... that's okay (SoftLanding does this.... as far as I can tell). Anyway.... that's what I wanted to say. -- Joe Emenaker - Sexual Engineer | Our infernal mailer daemon has been quite jemenake@nike.calpoly.edu | insitent that my signature be limited to just ..or.. @bslab65.calpoly.edu | 4 lines. However, as you can see, I have ..or.. @cash.calpoly.edu | figured out an elegant way to put as many as ------------------------------ From: sdh@po.CWRU.Edu (Scott D. Heavner) Subject: MINIX DEFRAG errors -- trashed all my static libs. Date: 17 Apr 1993 23:20:08 GMT Reply-To: sdh@po.CWRU.Edu (Scott D. Heavner) Sometime within the last few weeks, I tried defrag-0.4 (at least I think it was 0.4) on my minix fs. I did everything from a floppy with the root partition unmounted. Everthing appeared OK, I got no error messages, but then some of my binaries stopped working, I replaced my libc.so.4 and everthing was OK. Then I noticed some X apps weren't working, they all crashed with segmentation faults, so I replaced all the static X libs. Eventually, I found something wrong with all my static libs and had to replace them all. It was just a real pain trying to figure out waht went wrong. Is there a reason behind this? Should one never defragment a static lib, or should there be a better defragmenting system that preserves whatever holes might exist in the libs. Scott sdh@po.cwru.edu -- HELP! I'm being held prisoner in a .sig file factory. ------------------------------ ** 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-Activists-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux) via: Internet: Linux-Activists@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 tupac-amaru.informatik.rwth-aachen.de pub/msdos/replace The current version of Linux is 0.99pl6 released on February 23, 1993 End of Linux-Activists Digest ****************************** ke "completely non-redistributable" >HJ - stuff? Am I missing something? > >No, my purpose is to make 386BSD completely re-distrubutable, with NO >strings attached. Thatdigest870 644 36676 74430 50561 5366137110 5755 From: Digestifier To: Linux-Activists@senator-bedfellow.mit.edu Reply-To: Linux-Activists@senator-bedfellow.mit.edu Date: Sat, 17 Apr 93 22:15:18 EDT Subject: Linux-Activists Digest #870 Linux-Activists Digest #870, Volume #4 Sat, 17 Apr 93 22:15:18 EDT Contents: Re: Summary of Linux vs. 386BSD vs. Commercial Unixes (Jim Winstead Jr.) Re: finger....help.. (Jacob DeGlopper) Upgrading libc.so.4 library causes problems (Joshua Burnett Goldsmith) Re: possible bug in mount() (Linus Torvalds) Re: Summary of Linux vs. 386BSD vs. Commercial Unixes (Adam J. Richter) Re: linux on PS2 MCA & SCSI disk (Steven A Bairstow) Commpiling mpeg_play (Greg Bothe) Re: Why Linux should remain Linux and seagulls should not be used. (pete cervasio) Re: ANS: ser. hang under X and QUES: login over serial (Wolfgang Jung) ApplicationDEC monitor Xconfig (James Croall) Re: Summary of Linux vs. 386BSD vs. Commercial Unixes (The Knight in Silicon Armor.) Re: [Q] Linux/SLS "mail" is too smart for its own good (Klaus Guhr) Re: Is curses.h complete? (Brandon S. Allbery) Re: possible bug in mount() (Brandon S. Allbery) ---------------------------------------------------------------------------- Crossposted-To: comp.os.386bsd.questions From: jwinstea@fenris.claremont.edu (Jim Winstead Jr.) Subject: Re: Summary of Linux vs. 386BSD vs. Commercial Unixes Date: Sun, 18 Apr 1993 00:31:21 GMT In article <1qq2ts$1lo8@hal.gnu.ai.mit.edu> mycroft@hal.gnu.ai.mit.edu (Charles Hannum) writes: In article <1993Apr17.194341.4562@serval.net.wsu.edu> hlu@eecs.wsu.edu (HJ Lu) writes: > > That means no GNU copyrighted code in the "official" 386bsd release. You are full of shit. The `official' release include GCC, GDB, GNU tar, and several other pieces of GNU code. I believe HJ may have spoken a little hastily. I believe, however, that the Jolitz' have said they do not want any copylefted code in the *kernel proper*. Hence, Linus was asked (or volunteered?) to make the math emulation code available under a BSDish copyright, which he did. I remember a flap in the 386BSD community a little while ago about an improved serial driver (?) that the author had copylefted, which angered a good many people, and the author had to loosen the copyright to the BSDish form for it to be accepted. Just trying to do a little interpreting and clarifying, and not leaping into this absolutely stupid and childish series of flame wars. (And that's coming from someone in the .edu domain, too! :) -- loveritablessencentipedependentalism+ Jim Winstead Jr. (CSci '95) andaterrificklengtherealityearguessy| Harvey Mudd College, WIBSTR mpathybridgenerationiceremonymphysic| jwinstea@jarthur.Claremont.EDU alendareadvertisexpresshothoughthend+ or jwinstea@fenris.Claremont.EDU ------------------------------ From: jacob@mayhem.cwru.edu (Jacob DeGlopper) Subject: Re: finger....help.. Date: 17 Apr 93 23:21:16 GMT In mr2@netcom.com (Jongyoon Lee) writes: >How can I fix the finger to show last login information? >Mine always says never logged in although it's not true. >I tried to change permission to /usr/adm/lastlog but didn't help. The finger that comes with SLS (and the finger on sunsite) is broken in this regard. To fix it, take the BSD finger code from sunsite, and the lastlog.h file from the shadow password sources, remove the #ifdef linux from around the code in finger to read the lastlog file, and compile. You'll probably have to remove a reference to ll_host, or some similarly-named field, since the shadow code isn't writing that field to /usr/adm/lastlog. I may look at that at some point in the future. -- Jacob DeGlopper, EMT-A | CWRU Biomedical Engineering jacob@mayhem.cwru.edu | Wheaton (MD) Volunteer Rescue Squad +1 216 754 1638 | Opinions my own... ------------------------------ From: joshin@cats.ucsc.edu (Joshua Burnett Goldsmith) Subject: Upgrading libc.so.4 library causes problems Date: 17 Apr 1993 23:39:54 GMT Since I've upgraded the libc.so.4 library to .3 nothing seems to work, instead i have been getting an "Incompatible version" error. I just installed the new libdisks and nothing I compile works. Any help is appreciated. Josh ------------------------------ From: torvalds@klaava.Helsinki.FI (Linus Torvalds) Subject: Re: possible bug in mount() Date: Sat, 17 Apr 1993 23:35:57 GMT In article <1993Apr17.220003.3133@kf8nh.wariat.org> bsa@kf8nh.wariat.org (Brandon S. Allbery) writes: > >Reread the first paragraph, H.J. I don't care *how* screwed up the user-mode >code is, the kernel should NEVER panic as a result of it. I'd like to point out that what you saw wasn't even close to a panic - it's just an ordinary debugging message that means that the kernel hasn't done enough checking. In some cases, that can be problematic (if this happens in a interrupt handler or if it happens in a place where the kernel has busy resources), but generally a general protection is just the equivalent of a normal SIGSEGV that happens to be in kernel code. That was indeed the result of all this - just a sigsegv (which was quite ok - you get sigsegvs in user mode too when you give wrong parameters to functions) that happened to have some debug information added to it. The debugging messages used to occur with *every* sigsegv in older versions of linux - I think I removed the user level debug info in 0.97 or so, as it made people nervous to see a register dump when a program did something it shouldn't have. So now linux prints out the error messages only when the problem is in kernel mode (and yes, it should be considered a bug when this happens, but it's not necessarily a panic() or even a "real" problem). Note that most of these have been fixed a long time ago, so now only the really bad things result in debugging info - these are usually also things that actually *do* result in a panic sooner or later (usually sooner), so yes, there is some correlation with the debugging info and a panic. It's just not quite 1:1, Linus ------------------------------ Crossposted-To: comp.os.386bsd.questions From: adam@netcom.com (Adam J. Richter) Subject: Re: Summary of Linux vs. 386BSD vs. Commercial Unixes Date: Sun, 18 Apr 1993 00:02:41 GMT In article <1993Apr17.175431.25015@coe.montana.edu> nate@cs.montana.edu (Nate Williams) writes: >And then Linux and the GNU folks will take what 386BSD has done and >restrict it. Now, isn't that nice....... taking my work and restricting >it. I always thought that was awful kind of them. (NOT!) When speaking at SVNet meetings, Bill Jolitz has repeatedly stated that he sees 386BSD as a sort of publicly accessible research vehicle for operating systems. The 386BSD core distribution is not designed for the average end user. Instead, the 386BSD copyright is unrestricted specifically for the purpose of allowing the results of 386BSD to be assimilated into any other operating system project, such as Windows, turnkey BSD distributions or Linux. The public benefit is supposed to come from this assimilation, not unlike the way most unix users now benefit from the Berkeley filesystem and networking work. Assimilating 386BSD software into Linux without Linux having to give up its copyright policy is more than just allowed; it is specifically the point of 386BSD. -- Adam J. Richter Yggdrasil Computing, Incorporated 409 Evelyn Ave., Apt. 312, Albany CA 94706 PO Box 8418, Berkeley CA 94707-8418 (510) 528-3209 (510) 526-7531, fax: (510) 528-8508 adam@netcom.com yggdrasil@netcom.com Another member of the League for Programming Freedom (lpf@uunet.uu.net). ------------------------------ From: bairstow@white.crayola.cs.psu.edu (Steven A Bairstow) Subject: Re: linux on PS2 MCA & SCSI disk Date: Sun, 18 Apr 1993 00:32:01 GMT In article br.pct@RLG.Stanford.EDU (Peter C. Tam) writes: >Hi, > Pardon me if it is a FAQ, but: > > Can linux run off IBM MicroChannel PS2s, with SCSI disks? I don't thinnk you can install any of the distributed packages directly to a PS/2. There are patches around for a 508 controller. (I think that is the number.) You need to have someone compile these into a new boot disk for you. > & How stable & performance is the TCP/IP, NFS, ARPA services like > telnet, ftp, ping, lpr, lpd, etc.? The answer to this seems to be a yes and no. I have no problems with net programs, another guy 2 floors below me with the same card is getting all sorts of problems. I don't understand why though. Steve ------------------------------ From: gbothe@discover.wright.edu (Greg Bothe) Subject: Commpiling mpeg_play Date: Sun, 18 Apr 1993 00:21:46 GMT I was trying to compile mpeg_play, and it compiles with no errors, EXCEPT that I get unknown symbol errors when linking... Has anyone compiled this for Linux and am I forgetting a library? video.o: Undefined symbol _XShmCreateImage referenced from text segment video.o: Undefined symbol _shmget referenced from text segment video.o: Undefined symbol _shmat referenced from text segment video.o: Undefined symbol _XShmAttach referenced from text segment video.o: Undefined symbol _shmctl referenced from text segment video.o: Undefined symbol _XShmDetach referenced from text segment video.o: Undefined symbol _shmdt referenced from text segment gdith.o: Undefined symbol _XShmGetEventBase referenced from text segment gdith.o: Undefined symbol _XShmPutImage referenced from text segment main.o: Undefined symbol _XShmQueryExtension referenced from text segment Greg Bothe gbothe@discover.wright.edu ------------------------------ From: pete@q106fm.uucp (pete cervasio) Subject: Re: Why Linux should remain Linux and seagulls should not be used. Date: Sat, 17 Apr 93 19:12:48 CDT spj@ukelele.GCR.COM (Simon Patrick Janes) writes: > mdw@theory.TC.Cornell.EDU (Matt Welsh) writes: ....... > >in the world aren't seagulls found?), flexibility, and, well... flight (from > >the world of DOS). If any animal or design represents Linux it is the seagul ........... > I could never design a good logo. If nothing better comes up, the sea-gull > will probably end up being it. (Man, but they're so common.. ;) What do you think we want Linux to become... :-) Pete C. +----------------------------------------------------------------------+ | Pete Cervasio | pete%q106fm.uucp@wupost.wustl.edu | | I fish, therefore I am. | pete.cervasio@f1.n2250.z1.fidonet.org | +----------------------------------------------------------------------+ | GOAL: Linux on every desktop in our lifetime. | +----------------------------------------------------------------------+ ------------------------------ From: wong@cs.tu-berlin.de (Wolfgang Jung) Subject: Re: ANS: ser. hang under X and QUES: login over serial Date: 18 Apr 1993 01:01:52 GMT Daniel S. Ridge (newt@eng.umd.edu) wrote: : I posted yesterday asking for help on my linux communications : problems, and I had not realized that X didn't support IRQ sharing, : and the mouse on com1 conflicted with the 2400 baud modem on com3 Since the serial IRQ handling is inside the kernel and is not build into X. X just open the mouse port sets speed and afterwords just reads from the FD(which points to (/dev/mouse)) So i would think the Problem lies in the IRQ Conflict on the hardware where your Mouse Ports are getting Problems with your Serial card. This is due to some vendors are building unsharable IRQ lines, and this I thnik is the Problem here. (Just take a look at the kernel source (kernel/chr_dev/serial.c) there is the serial IRQ handling done. If the Hardware works it is able to work with at least the four normal COM Ports (with COM1 & 3 on IRQ 3 and COM2 & 4 on IRQ 4) Probably your Modems or your Serialcard is not able to handle shared interups. A soltion would be to put the Modems each on some own IRQ line, Just check the hardware if you are with Open Colector or Totem Pole IRQ Outputs .. Gruss Wolfgang ------------------------------ From: jcroall@ukelele.GCR.COM (James Croall) Subject: ApplicationDEC monitor Xconfig Date: Sun, 18 Apr 1993 00:11:32 GMT well, I think the summary sums it up. i guess thats whay sumarries do hmm. -- ÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄ ³ James Croall -/- jcroall@ukelele.GCR.COM -/- 1@1 sense/net -/- 9@4 CRiMEnet ³ ³ CoSysOp: The Hylton BBS [703/670.5223] Armageddon [703/803.6289] ³ ÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄ ------------------------------ From: venger+@pitt.edu (The Knight in Silicon Armor.) Crossposted-To: comp.os.386bsd.questions Subject: Re: Summary of Linux vs. 386BSD vs. Commercial Unixes Date: 18 Apr 93 01:10:38 GMT In article <1993Apr17.231000.103368@zeus.calpoly.edu> jemenake@trumpet.calpoly.edu (Joe Emenaker) writes: [ Story about DEC ripping off EMACS author deleted...] >Now, it really, Really, REALLY angers me to think of these big >corporations taking public-domain and otherwise free software and >distributing it as their own and actually getting money for it. How >DEVOID of work-ethic does some have to be to pull a stunt like that? And >you're saying that you're pleased as punch if DEC can just ftp a copy of >386BSD and start selling it for $500/copy as DEC-BSD/PC or something?!?! >That makes me ill. It really does. When I code stuff and release it to >the public domain, I stipulate that the stuff is free and is free to >modify, but ANYTHING that is derived from my code or that USES the >binaries has to be free as well. I think that the GNU agreement is >something similar to this. Now, if a company wants to charge for >SUPPORT, or if they want to charge a media fee for distribution.... >that's okay (SoftLanding does this.... as far as I can tell). > >Anyway.... that's what I wanted to say. I think what's griping Nate is that if someone contributes code to a LIBRARY under GPL and someone else creates software and uses that library to compile said software of their own creation, they are not permitted under the GPL to sell it. It's kind of like Borland coming along and saying that because you used TurboC to compile software, they are entitled to a share in any profits since you used their libraries. While I agree that it would be totally reprehensable for anyone to sell the libraries, I do not think it is right to tell someone that wrote 10,000 lines of code that may have 1 or 2 calls to libraries that are under the GPL that they are not allowed to try to get recompense for their efforts. IMHO, if I release something into the public domain, my concern as to the use made of it has ended. This is my opinion. You are free to use and distribute it provided you do not charge for the distribution. If you should, however, derive an opinion of your own from mine, you may charge whatever you can get for it. :-) Snuffy -- Just say NO to .signatures!!! venger+@pitt.edu ------------------------------ From: kg@gnusys.GUN.de (Klaus Guhr) Subject: Re: [Q] Linux/SLS "mail" is too smart for its own good Date: Fri, 16 Apr 1993 12:27:29 GMT Andreas Busse (andy@resi.waldorf-gmbh.de) wrote: : In article <1993Mar28.004916.19259@jmd386.lonestar.org>, jdoss@jmd386.lonestar.org (Joe M. Doss) writes: : |> In article <1993Mar24.190859.16371@eecs.nwu.edu> hpa@nwu.edu (H. Peter Anvin) writes: : |> >I have a question about the "mail" program for Linux, specifically : |> >from the SLS release, which I believe is the same as in mailpak. The : |> >problem is that it bypasses the mail transport agent (smail) when it : |> >gets an address it "recognizes" (local or UUCP-style, I believe). : |> >Unfortunately that is a Bad Thing[TM] at least on my system. "elm" : |> >behaves properly but some users on this machine are used to the BSD : |> >Mail interface and don't want to switch. : |> > : |> >So: is there either a way to force "mail" to send ALL its mail to the : |> >mail transport agent, or is there a different version of mail : |> >somewhere that does that? (If there is but is not for Linux I'll try : |> >to port it -- that's no problem) : I had the same problem. My solution was to install the sendmail+ida package. : Now I have simple /bin/mail and /bin/rmail as transport agents, elm as a : frontend and sendmail as the "router". There were only few things to change : in the sources. I don't remeber exactly, but the most important thing was to : change the header format of elm. The "from" line must be the first one and : the "subject" line the last. I've ported bsd-mail (version 5.5 / 6/1/90) to Linux. It uses /usr/lib/sendmail as MTA. The stuff can be received from gnusys (++49 +228 +432433). Login twice as gast. Online: 20.00 - 08.00h. -- o o ---oOO-(_)-OOo-------------------------------------------------------------- | Klaus Guhr, Marktstr.33 | Phone : +49 +228 48 53 60 kg@gnusys.GUN.de | 5300 Bonn 3, Germany | FAX/Data: +49 +228 43 24 33 | | FAX: 08.00 - 20.00 h | Data:? ============================================================================ uid 0 is expired. Shutdown in 1 minute! ------------------------------ From: bsa@kf8nh.wariat.org (Brandon S. Allbery) Subject: Re: Is curses.h complete? Date: Sun, 18 Apr 1993 01:37:30 GMT In article zmbenhal@netcom.com (Zeyd M. Ben-Halim) writes: >For some reason the man pag refers to SYSVR3.2 curses whilst the library is >in fact BSD. I suggest you get ncurses (which is SYSVR4-compatible) from >netcom.com:pub/zmbenhal. ncurses is missing about half of the color routines (no HLS, and it assumes ANSI emulation --- it doesn't accept the "setf" and "setb" terminfo codes) and *all* of the printer entries in terminfo (which, luckily for you, aren't used by curses proper). Not sure if I'm allowed to send you an SVR3.2 manpage, but I have one. (I'll have to wait until they load Solaris on the machine at work before I can get an SVR4 manpage.) ++Brandon -- Brandon S. Allbery bsa@kf8nh.wariat.org It's not too late to turn back from the "Gates" of Hell... Linux: the FREE 32-bit operating system, available NOW. Why waaaaaait for NT? ------------------------------ From: bsa@kf8nh.wariat.org (Brandon S. Allbery) Subject: Re: possible bug in mount() Date: Sun, 18 Apr 1993 01:43:22 GMT In article <1993Apr17.233557.10285@klaava.Helsinki.FI> torvalds@klaava.Helsinki.FI (Linus Torvalds) writes: >In article <1993Apr17.220003.3133@kf8nh.wariat.org> bsa@kf8nh.wariat.org (Brandon S. Allbery) writes: >I'd like to point out that what you saw wasn't even close to a panic - Okay, sorry. My experience has always been that kernel GP faults are immediately followed by panics... even once or twice under Linux, but with older kernels. In any case, I still think it's improper for the kernel to get into trouble over bad user-mode actions (within limits; some things can't be caught, but most of those are reserved for root), but I'm not complaining about the kernel so much as HJ's assumption that the fix is just to use prototypes in user-mode code. I've seen programs which scramble their own memory and then execute it (jump through an NFG function pointer) crash systems in the past, and there is even a program available which deliberately tries to do that to test the resilience of a kernel. (I hear that most commercial *ixes fail the test....) ++Brandon -- Brandon S. Allbery bsa@kf8nh.wariat.org It's not too late to turn back from the "Gates" of Hell... Linux: the FREE 32-bit operating system, available NOW. Why waaaaaait for NT? ------------------------------ ** 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-Activists-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux) via: Internet: Linux-Activists@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 tupac-amaru.informatik.rwth-aachen.de pub/msdos/replace The current version of Linux is 0.99pl6 released on February 23, 1993 End of Linux-Activists Digest ****************************** to change : in the sources. I don't remeber exactly, but the most important thing was to : change the header format of elm. The "from" line mudigest871 644 36676 74430 50244 5366137121 5756 From: Digestifier To: Linux-Activists@senator-bedfellow.mit.edu Reply-To: Linux-Activists@senator-bedfellow.mit.edu Date: Sun, 18 Apr 93 00:30:09 EDT Subject: Linux-Activists Digest #871 Linux-Activists Digest #871, Volume #4 Sun, 18 Apr 93 00:30:09 EDT Contents: **URGENT HELP Wanted 'init: exec failed' ** (Gerry George) Re: New feature for the filesystems. What do you think ? (Laszlo Herczeg) Re: New feature for the filesystems. What do you think ? (Michael O'Reilly) Re: C= 64 emulator project (Eric Blade) X81514scal & Graphics Ultra Pro: vertical stripes ("17-APR-1993 19:26:24.14") TIOCGETP & TIOCSETP ("17-APR-1993 19:20:43.85") Re: Summary of Linux vs. 386BSD vs. Commercial Unixes (Jay A. Carlson) Re: possible bug in mount() (HJ Lu) Re: bugs (Vladimir Vukicevic) Re: Help - how to create /dev/sda9? (Vladimir Vukicevic) Re: Help - how to create /dev/sda9? (Vladimir Vukicevic) Re: LILO does not ask for alternative boots (Something longer than life) Re: Commpiling mpeg_play (Leon Dent) Re: possible bug in mount() (Jonathan Magid) how to grind your system to a halt .. vitrual mem, tcsh (Kris Gleason) Choosing an init (Damien Neil) ---------------------------------------------------------------------------- From: ggeorge@bu.edu (Gerry George) Subject: **URGENT HELP Wanted 'init: exec failed' ** Date: 18 Apr 93 01:32:47 GMT I was trying to "fix" the permissions on my system (SLS 0.99pl6). I changed 'login' and a couple more programs to non-suid. I rebooted my system and it came up as normal - almost. rc ran, called rc.local, which in turn called rc.net. All of the drives are mounted, swap partition is initialized, and the date prompt is echoed on screen. However, just before giving the login prompt, I get the error messages 'init: exec failed' 9 times, then the system reboots. I logged in with a fresh diskette, changed the permissions of login back to suid root, commented out rc.net - same thing. I also commented out rc.local from rc, no luck. Apparently, it is something at the very end of the boot/initialize process which dies. I can't figure out what it is. My rc file (only relevant lines shown): #!/bin/sh PATH=/bin:/usr/bin:/etc /etc/update & /bin/ctrlaltdel soft /bin/rm -f /etc/mtab* /etc/nologin /etc/utmp >/etc/utmp:/bin/chmod 444 /etc/utmp touch /etc/utmp /bin/mount -av /bin/mount /dev/hda3 /usr /bin/mount /dev/hda4 /home /bin/mount /dev/hdb1 /user /bin/mount /dev/hdb2 /pub /bin/swapon /dev/swap rm -f /usr/spool/uucp/LCK* rm -f /tmp/* # /bin/sh /etc/rc.local (commented out for diagnostics) -- =========================================================================== Gerry George | Anything good in life is either School of Management, Boston Univ. | illegal, immoral or fattening. Internet: ggeorge@acs.bu.edu | Any item not in the above three Compu$erve: 72607.2560@compuserve.com | categories causes cancer in rats! =========================================================================== ------------------------------ From: las@r-node.hub.org (Laszlo Herczeg) Subject: Re: New feature for the filesystems. What do you think ? Date: Sat, 17 Apr 1993 23:27:33 GMT I am currently using a "transparent" compression scheme - the tcx program which was posted a few weeks ago in comp.sources.unix. TCX compiled clean for Linux, except for a few #define changes in untcx.c . For the time I have been using it, TCX has behaved fine and it saves about 50 % space on my /usr/bin. I can compress heavily used files such as gcc and p2c (for me:) and it keeps track of the status of the files without a glitch. Based on my testing experience, I would recommend tcx until such time as someone comes up with an OS-based compression scheme. Cheers! Laszlo -- ================================================================================ Laszlo Herczeg E-mail: las@r-node.gts.org Toronto, Canada E-mail (altern): las@eastern.com ------------------------------ From: oreillym@tartarus.uwa.edu.au (Michael O'Reilly) Subject: Re: New feature for the filesystems. What do you think ? Date: 18 Apr 1993 02:23:41 GMT Keith Smith (keith@ksmith.com) wrote: : In article <0fmuPUm00VopMZhFU7@andrew.cmu.edu> Frank T Lofaro writes: : >Excerpts from netnews.comp.os.linux: 13-Apr-93 Re: New feature for the : >fil.. HJ Lu@eecs.wsu.edu (463) : > : >>I am wondering how a compressed filesystem handles demand paging and : >>random update. : > : >I had just posted wondering about demand-paging, but came up with an : >idea. Have it uncompress the file, and put it on some temporarily : >allocated inodes, and exec/demand page it from there. It would require : >some changing of filesystem allocation/exec/demand paging routines I : >gather, but should be feasible. : > : [ stuff about intermediate layer deleted. ] I think someone already suggested this, but it got ignored.. PLEASE look into the structure of the LFS (log file system) written for the Sprite OS. This is a journalling filesystem. This makes it absolutely ideal for makeing a compressed file system. (not to mention that the fact that it is very fast to write to, and nicely crash resistant). The thing that makes LFS so nice is that ALL writes are sequential. inodes, directory, map blocks, data all get written sequentially into the log. Makes compression kinda easy. If you byte-align the blocks, and work with 8K pseudo blocks, then you never need to uncompress more than 8K to get the data you want. There was another benifit but I have forgotten it. Oh well. Some other time. michael. ------------------------------ From: ak853@cleveland.Freenet.Edu (Eric Blade) Subject: Re: C= 64 emulator project Date: 18 Apr 1993 03:00:21 GMT Reply-To: ak853@cleveland.Freenet.Edu (Eric Blade) In a previous article, rmast@cs.vu.nl (Mast RN) says: > >Hi! I found on comp.sys.cbm a message of a project concerning making >a C version of a Commodore 64 emulator! I ftp'd it from ccosun.caltech.edu >/pub/rknop/INCOMING. > >for working with it you also need a copy of the copyrighted roms. >It is too complex to look like an april fools joke. >I haven't got it running on linux yet, but it should work >with X in color. > >Robert (rmast@cs.vu.nl) > Where the heck are you supposed to get a copy of the roms? read 'em from your c-64 in yer closeet? :) ------------------------------ From: "17-APR-1993 19:26:24.14" Subject: X81514scal & Graphics Ultra Pro: vertical stripes Date: 17 Apr 1993 23:28:55 -0400 Reply-To: nmp08@rz.uni-kiel.dbp.de Hi again! I have another problem with X11 running the X8514scale server with a Graphics Ultra Pro (1 MB): sometimes i'll get vertical stripes on my screen. They disappear when screen was redrawn. Is there anybody else there with this problem? Another question: when leaving X11 the text mode isn't restored correctly. Any solutions? Any help would be greatly appreciated. Please reply to nmp08@rz.uni-kiel.dbp.de Klaus Hoppe ------------------------------ From: "17-APR-1993 19:20:43.85" Subject: TIOCGETP & TIOCSETP Date: 17 Apr 1993 23:28:55 -0400 Reply-To: nmp08@rz.uni-kiel.dbp.de Hi Linuxers! I have a little problem with a program i wish to compile: in ioctl(0, TIOCGETP, &tty); and ioctl(0, TIOCSETP, &tty); TIOCGETP and TIOCSETP are undefined. What is the Linux-aequivalent for these expressions to set RAW and ~ECHO mode of a terminal? Any help would be appreciated. Klaus Hoppe Please reply directly to nmp08@rz.uni-kiel.dbp.de as i have no news access. ------------------------------ Crossposted-To: comp.os.386bsd.questions Subject: Re: Summary of Linux vs. 386BSD vs. Commercial Unixes From: nop@theory.Mankato.MSUS.EDU (Jay A. Carlson) Date: 17 Apr 93 22:14:44 Reply-To: nop@theory.cs.mankato.msus.edu In article <1993Apr17.205715.11278@coe.montana.edu> nate@cs.montana.edu (Nate Williams) writes: No, my purpose is to make 386BSD completely re-distrubutable, with NO strings attached. That means Sun, DEC, HP, Ren and Stimpy, or whoever can take this code and sell a binary copy of it. The GPL does not allow this. You and I, and your list of luminaries, can sell binary copies of GPL'd code. What you can't do is sell 'em without source availability. The original BSD copyright has been this way, but unfortunately a group of people take the code, fix the code, and then place restrictions on it. How is this worse than, say, DEC fixing the code and not releasing source at all? -- Jay Carlson nop@theory.cs.mankato.msus.edu Flat text is just *never* what you want. ---stephen p spackman ------------------------------ From: hlu@luke.eecs.wsu.edu (HJ Lu) Subject: Re: possible bug in mount() Date: Sun, 18 Apr 93 02:28:33 GMT In article <1993Apr17.220003.3133@kf8nh.wariat.org> bsa@kf8nh.wariat.org (Brandon S. Allbery) writes: >In article <1993Apr17.162547.3040@serval.net.wsu.edu> hlu@luke.eecs.wsu.edu (HJ Lu) writes: >>In article <9246@blue.cis.pitt.edu> tjd@neurocog.lrdc.pitt.edu (Todd J. Derr) writes: >>>well, it doesn't seem that passing bad values to mount() should cause a GP >>>fault, so... (yes, I know that I went about this wrong, but that's not the >>>point...) >>> >>>{ mount("/dev/hda1","/", MS_MGC_VAL | MS_REMOUNT); } >> >>Please always use prototype. It will save everybody's time. > >Reread the first paragraph, H.J. I don't care *how* screwed up the user-mode >code is, the kernel should NEVER panic as a result of it. > May I say oops? Sorry for that. I am too used to prototype. Linus has posted a patch for kernel. Sorry for that. H.J. ------------------------------ Crossposted-To: answer,spawn,inittab,in,inttab,syntax From: vladimir@intrepid.com (Vladimir Vukicevic) Subject: Re: bugs Date: Sun, 18 Apr 1993 01:41:15 GMT In article ucc02aa@sun3.LRZ-Muenchen.DE (Pilch) writes: >I have almost become a Linux freak by now. Of course it would still be >better, if one could do everything without becoming a freak. In order >to make my PC accessible by modem from outside, I had to maneuver >around several near-collapses of the harddisk and experiment until >I finally found the solution, which is somewhat different from the >hazardous proposals in FAQ. > >The relevant passage of my file /etc/inittab now looks like this: > >c1:123456:respawn:/bin/getty 9600 tty1 >c2:23456:respawn:/bin/getty 9600 tty2 >c3:3456:respawn:/bin/getty 9600 tty3 >c4:456:respawn:/bin/getty 9600 tty4 >c5:6:respawn:/bin/getty 9600 ttyS1 >c6:6:respawn:/bin/getty 38400 ttyS0 > >This, however, only allows access for 9600 baud modems and a 38400 baud >nullmodem. Anyone with a 2400 baud-only modem will be excluded as will >be 14400 baud modem users, whom I could otherwise connect with. This >is because the getty command does not accept any comma-separated lists >of baudrates nor does it recognize the option -m. Any attempt to >change the terminal name c5, c6 to something like vt100 or ttysx >will inevitably lead to a near-catastrophe, which only a freak knows >how to handle. The screen is inremittantly clobbered with error-messages, >as soon as the file is saved or as soon as the init procedure is activated >during booting. > Well, the way you would do this you would lock the modem's interface speed to the highest speed your modem can do; in this case, 38400. Then, you'd use hardware handshaking between the modem & computer, so that the speed between the modem & computer stays the same, but the people who dial up to your modem can connect at any speed. Also, 38400 is not a 'real' modem-to- modem speed; it is actually 14.4Kbps with V.42/V.42bis compression. Also, the c5,c6,etc. are *not* terminal types; they are id strings for the processes. They have nothing to do with terminal types. When a user logs in through getty (getty can dynamically adapt itself to a user's terminal capabilities), he/she can set the term by doing 'setenv TERM whatever'. (I hope this did not sound like a RT*M; if it did - sorry) - Vladimir -- #---- Vladimir Vukicevic #include #--------- vladimir@intrepid.com #-------- ------------------------------ From: vladimir@intrepid.com (Vladimir Vukicevic) Subject: Re: Help - how to create /dev/sda9? Date: Sun, 18 Apr 1993 02:56:45 GMT In article <1qc30fINN1to@tamsun.tamu.edu> tpradeep@cs.tamu.edu (Pradeep K Tapadiya) writes: >Howdy netters, > >I am trying to install SLS on g: and h: ( /dev/sda8 and /dev/sda9) >partition of my SCSI drive. > >/dev/sda8 is to be my main filesystem. >/dev/sda9 is to be my swap filesystem. > >"mke2fs" works fine with /dev/sda8. > >However, "mkswap" doesn't recognize /dev/sda9. I think its because the scsi >device entries in a2 installation disk are limited to /dev/sda8. > >Q1. How can I create a new entry /dev/sda9? > >Q2. Is it necessary to create the entry on the floppy? There doesn't seem >to be any space left on the floppy. > >Thank you for your help. > >Pradeep >tpradeep@cs.tamu.edu Well, for SCSI, the major number is 8 & the minor number is the partition #. I think 'mknod /dev/sda9 8 9' will do what you want. I think you can have up to 15 partitions; minor 16 is partition 1 of sdb. - Vladimir -- #---- Vladimir Vukicevic #include #--------- vladimir@intrepid.com #-------- ------------------------------ From: vladimir@intrepid.com (Vladimir Vukicevic) Subject: Re: Help - how to create /dev/sda9? Date: Sun, 18 Apr 1993 03:09:59 GMT In article <1993Apr18.025645.1941@intrepid.com> vladimir@intrepid.com (Vladimir Vukicevic) writes: >In article <1qc30fINN1to@tamsun.tamu.edu> tpradeep@cs.tamu.edu (Pradeep K Tapadiya) writes: >>Howdy netters, >> >>I am trying to install SLS on g: and h: ( /dev/sda8 and /dev/sda9) >>partition of my SCSI drive. >> >>/dev/sda8 is to be my main filesystem. >>/dev/sda9 is to be my swap filesystem. >> >>"mke2fs" works fine with /dev/sda8. >> >>However, "mkswap" doesn't recognize /dev/sda9. I think its because the scsi >>device entries in a2 installation disk are limited to /dev/sda8. >> >>Q1. How can I create a new entry /dev/sda9? >> >>Q2. Is it necessary to create the entry on the floppy? There doesn't seem >>to be any space left on the floppy. >> >>Thank you for your help. >> >>Pradeep >>tpradeep@cs.tamu.edu > >Well, for SCSI, the major number is 8 & the minor number is the partition #. >I think 'mknod /dev/sda9 8 9' will do what you want. I think you can have >up to 15 partitions; minor 16 is partition 1 of sdb. > Oops. The correct command is 'mknod /dev/sda9 b 8 9'. Sorry. -- #---- Vladimir Vukicevic #include #--------- vladimir@intrepid.com #-------- ------------------------------ From: devmorfo@mtu.edu (Something longer than life) Subject: Re: LILO does not ask for alternative boots Date: Sun, 18 Apr 1993 03:12:46 GMT Read the manual ... ! You need to put a 'delay = tsecs' in your config fileand then while the system is booting you wiil see the word LILO pausing for as many tsecs as you sepcified in your config file. Then if you press any of the SHIFT CTRL ALT CAPS LOCK or SCROLL LOCK you will see a prompt that looks like this 'LILO boot: '. This is where you type your label for the OS that you want LILO to boot. Simple if you read the manuals .... Dimitris ------------------------------ From: lcd@umcc.umcc.umich.edu (Leon Dent) Subject: Re: Commpiling mpeg_play Date: 18 Apr 1993 00:02:15 -0400 I have compiled mpeg_play several times with no problems (hence I can't really help you). Are you using the versions from sunsite.unc.edu in pub/multimedia/utilities/unix ? Leon Dent lcd@umcc.umich.edu ------------------------------ From: jem@sunSITE.unc.edu (Jonathan Magid) Subject: Re: possible bug in mount() Date: Sun, 18 Apr 1993 04:05:33 GMT In article <1993Apr18.014322.3720@kf8nh.wariat.org> bsa@kf8nh.wariat.org (Brandon S. Allbery) writes: > I've seen programs which scramble their own memory and then execute it >(jump through an NFG function pointer) crash systems in the past, and there is >even a program available which deliberately tries to do that to test the >resilience of a kernel. (I hear that most commercial *ixes fail the test....) This reminded me of an article I saved a long time ago... It is included for hysterical interest.... jem. (who is a hopeless packrat) ----8<---From the Archives--------- From: scot@catzen.GUN.de (Scot W. Stevenson) Newsgroups: comp.os.linux Subject: Crashme and Linux Message-ID: Date: 28 Jun 92 20:29:04 GMT Reply-To: scot@catzen.GUN.de Organization: Interactive! Hi, The German computer magazine iX has an article on the crashme program used to test (well, hopefully) the stability of Unix systems. Of the 20 tested systems, only three didn't crash: Esix V4.0.3, Sun SPARC as of OS 4.1.2, and - Linux V0.95 =8). The list of those who bit the cybernetic dust includes names like Coherent, AIX, Ultrix... Nice work, Linus =8)! T, Scot [Source: 'Weichmacher - crashme fuer Unix-Systeme', Thomas Neser, iX Juli (7) 1992, pp. 88] ------------------------------ From: gleasokr@rintintin.Colorado.EDU (Kris Gleason) Subject: how to grind your system to a halt .. vitrual mem, tcsh Date: Sun, 18 Apr 1993 03:21:02 GMT I happened upon an interesting discovery the other day. If you're curious, try this... in tcsh, hold down a key for a while; let the key repeat until the buffer is full (no more keystrokes will be accepted). Then, hit return. You should probably run top in another virtual terminal before you hit return. Then, watch tcsh's memory usage climb and climb and climb, until the system grinds to a halt (swap space fills up, then the remaining space in the buffer cache gets eaten until the system can no longer function). You might also want to set up an infinite loop that calls free, so you can watch your memory usage directly. Careful, 'cause the only way out of this is to cycle the power (if you let things get out of hand enough). I let it run until it took 10 minutes between free's updates... 0 swap, 300 or so cache, 40 free. Yikes! I didn't check into the reason that tcsh eats up so much memory. But, it may be important that I am running with libc4.3.3. If you're curious, bash is a bit more graceful... segmentation fault instead of the grinding. I didn't try any other shells. I did this with 8 M of ram, and 16 M of swap (so don't tell me to increase my swap :) I just found it interesting, and a _huge_ security hole (unless you want other users to halt your system). Just an observation. Kris -- gleasokr@rintintin.colorado.edu HARDWARE (noun): The equipment that makes up a computer system, not to be confused with software ------------------------------ From: damien@b63519.student.cwru.edu (Damien Neil) Subject: Choosing an init Date: 18 Apr 1993 04:28:12 GMT Looking on sunsite, I found a rather large number of inits. Unfortunately, I have little enough experience with UN*X systems administration (read: none) that I find it difficult to figure out which one is best for me. Would it be possible for some kind soul to offer some advice to me and other quasi-newbies on making this decision? I will summarize email responses. Thanks! -- Damien Neil dpn2@po.cwru.edu "Until someone debugs reality, the best Case Western Reserve University I can do is a quick patch here and there." CMPS/EEAP Linux -- the choice of a GNU generation. -Erik Green ------------------------------ ** 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-Activists-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux) via: Internet: Linux-Activists@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 tupac-amaru.informatik.rwth-aachen.de pub/msdos/replace The current version of Linux is 0.99pl6 released on February 23, 1993 End of Linux-Activists Digest ****************************** test the >resilience of a kernel. (I hear that most commercial *ixes fail the test....) This reminded me of an article I saved a long time ago... It is included for hysterical interest.... jem. (who is a hopeless packrat) ----8<---From the Archives--------- From: scot@catzen.GUN.de (Scot W. Stevenson) Newsgroups: comp.os.linux Subject: Crasdigest872 644 36676 74430 50266 5366137122 5764 From: Digestifier To: Linux-Activists@senator-bedfellow.mit.edu Reply-To: Linux-Activists@senator-bedfellow.mit.edu Date: Sun, 18 Apr 93 04:00:19 EDT Subject: Linux-Activists Digest #872 Linux-Activists Digest #872, Volume #4 Sun, 18 Apr 93 04:00:19 EDT Contents: txconn (dpjunk@mmm.com) Thanks to all! (Jonathan Ward) Re: Summary of Linux vs. 386BSD vs. Commercial Unixes (Tim Pierce) X!! and LILO (Tsun-Ming Kao) testing (Manish Gupta) Re: the umask of root in SLS should be 022 ?? (Gilbert Callaghan) *** READ THIS BEFORE POSTING *** (Ian Jackson) Re: Summary of Linux vs. 386BSD vs. Commercial Unixes (Garrett D'Amore) Re: New feature for the filesystems. What do you think ? (Brian Edmonds) Re: Summary of Linux vs. 386BSD vs. Commercial Unixes (James R. Wiegand) Re: Summary of Linux vs. 386BSD vs. Commercial Unixes (Thomas McWilliams) Re: Summary of Linux vs. 386BSD vs. Commercial Unixes (Thomas McWilliams) Re: Telnetd problem in 99.8 (Dan Graham) Re: Linux on Macintrashes? (STOP ALREADY!) (Philip Rhoades) Re: Questions about virtual mem. (James Michael Chacon) ---------------------------------------------------------------------------- From: dpjunk@mmm.com Subject: txconn Date: 18 Apr 1993 00:59:11 -0400 Reply-To: dpjunk@mmm.com I need help getting txconn working. I am connecting to a remote HP-UX computer that I will call node a. I am currently running term with no problems (term 1.05). On node a, I set my display variable by doing export DISPLAY=a:9. I then run txconn. Next I try to start xclock on node a and only get "Error: Can't open Display". What am I doing wrong? I am running olwm on the SLS release .99p6. Please help me. Thanks in advance. -- Dean P. Junk "An ounce of perception, a pound of obscure" Internet (dpjunk@mmm.com) --RUSH ------------------------------ From: drdrums@mozart.ucr.edu (Jonathan Ward) Subject: Thanks to all! Date: 18 Apr 93 04:50:56 GMT Reply-To: drdrums@csld.ucr.edu I posted a question several weeks ago regarding a problem I was having with my mouse in XWindows. First I want to thank all of you who sent me a response - I did finally get the problem fixed. Second I want to thank everyone who posts and maintains this group: You are providing an invaluable service to all in the Linux community. May this group live long, along with its OS. Last, and most certianly not least, I wish to thank all who participated in the development of such a fantastic(and FREE) OS. -Jonathan Ward University of California, Riverside ===================================================================== E-Mail to: | "Pentium - SMP in a chip" drdrums@watserv.ucr.edu | drdrums@csld.ucr.edu | "Linux and X - the GNU cure jmward@cs.ucr.edu | for Windoze NoT" ------------------------------ Crossposted-To: comp.os.386bsd.questions From: twpierce@unix.amherst.edu (Tim Pierce) Subject: Re: Summary of Linux vs. 386BSD vs. Commercial Unixes Date: Sun, 18 Apr 1993 04:30:04 GMT In article <1993Apr17.205715.11278@coe.montana.edu> nate@cs.montana.edu (Nate Williams) writes: >No, my purpose is to make 386BSD completely re-distrubutable, with NO >strings attached. That means Sun, DEC, HP, Ren and Stimpy, or whoever >can take this code and sell a binary copy of it. The GPL does not allow >this. You really think that? Snort. -- ____ Tim Pierce / ?Usted es la de la tele, eh? !La madre \ / twpierce@unix.amherst.edu / del asesino! !Ay, que graciosa! \/ (BITnet: TWPIERCE@AMHERST) / -- Pedro Almodovar ------------------------------ Crossposted-To: comp.windows.x.i386unix From: tkao@iastate.edu (Tsun-Ming Kao) Subject: X!! and LILO Date: Sun, 18 Apr 1993 05:16:01 GMT Could anyone help me with the following problem? 1. I have OS/2 2.0, DOS 5.0, and I just install Linux. After followed the steps shown one the FAQ, but I still can't boot Linux from my Hard Drive. I put Dos and Linux at the primary drive and OS/2 at the logical drive. Here is the command I used to build the Image file: etc/lilo/lilo -b /dev/hda4 -i /etc/lilo/boot.b -v -v -v /Image Everything seems fine so far, but when I try to boot up( after I added it to the Boot Manager) from the hard drive, system hangs during the boot procedure. There are some differences when I load Linux from A dirve: HARD DRIVE: EXT2-fs: magic match false [MS-DOS FS Rel alpha.8 FAT 16 check n,conv=b,uid=0,gid=0,umask=002] [me=0xf8,cs=4,#f=2,fs=1,fl=200,ds=401,d0=512,data=433,se=0,ts=204600] System Hang...... A: DRIVE: EXT2-fs warning: mounting none valid file system [EXT II FS alpha 0.2b 93/03/01,bs=1024,gc=16,bpg=8192,lpg=1992] Boot up continues...... 2. When I exit from X Window, my screen started flickering. I can't even make a warm boot. I tried several speed setting, but still no results. Please help me with my problems. Thanks in advance. Daniel tkao@iastate.edu ------------------------------ From: manish@ms.uky.edu (Manish Gupta) Subject: testing Date: 18 Apr 93 05:16:15 GMT please ignore this -- :;8B--=) ------------------------------ From: gilbert@inviso.com (Gilbert Callaghan) Subject: Re: the umask of root in SLS should be 022 ?? Date: Sun, 18 Apr 1993 05:10:23 GMT Johan Myreen (jem@snakemail.hut.fi) wrote: : In article twpierce@unix.amherst.edu (Tim Pierce) writes: : >I did a little looking around after installing SLS and decided that : >there was nothing, short perhaps of /proc (and perhaps not even that) : >that I would not want readable by a non-superuser. So I ran "find / : >-exec chmod a+r {} \;" on the whole bunch and it started working fine. I did that too. *THEN* I found the script /etc/sysperms which sets up all the correct file permissions. ------------------------------ From: ijackson@nyx.cs.du.edu (Ian Jackson) Subject: *** READ THIS BEFORE POSTING *** Date: Sun, 18 Apr 1993 06:23:01 GMT To get people to read your post and answer your question: - Read "Linux Documents Explained for Newbies" - posted weekly (see below). - Please do not post questions answered in the FAQ lists. - Please do not post "Please send me the FAQ". See below. - Please do not post "Split this group now". See below. - Be informative, both in subject line and body. See below. - Please read the documentation for the program you have a problem with. - Linux-specific posts only please - Unix, X, C, etc. groups listed below. For more information read on ... ** Linux Documents Explained for Newbies ** is is posted weekly by Jay MacDonald. If you can't find it email me or Jay . It will tell you about the Linux FAQ, the INFO-SHEET, the META-FAQ, etc. These are posted regularly to comp.os.linux and c.o.l.announce, and can be found in /pub/usenet/news.answers/linux-faq at pit-manager.mit.edu and on the Linux archive sites. If you can't get them there, try mailing "mailserver@nic.funet.fi", "mail-server@pit-manager.mit.edu" or "ftpmail@decwrl.dec.com". If you can't get them any other way, e-mail me. Splitting the group: It's been tried, and all the new groups except .announce failed. It can't be tried again within 6 months. See the FAQ for details, or read news.announce.newgroups. If you insist on discussing it, do so in news.groups (if c.o.l has too much volume already, why make it worse?). With threaded newsreaders now in widespread use using a good, specific, polite subject line will get you many more readers. Most people will ignore posts with subjects like "Linux", "Help", "SLS", etc. Don't shout to get attention. Put as much information in your post as possible - preferably a brief summary followed by session transcripts, etc. as appropriate. Include the precise text of any error messages printed, etc., and appropriate info about your setup. Linux is a Unix clone running on PCs; X has its own groups. One of the following groups is is quite likely to be more appropriate for your question: Unix: comp.unix.questions, .admin, .programmer, .shell etc. X: comp.windows.x, comp.windows.x.i386unix (for XFree86 etc) GNU: gnu.emacs.help, comp.emacs, gnu.gcc.help, gnu.* generally PCs: comp.sys.ibm.pc.hardware C: comp.lang.c -- Ian Jackson ijackson@nyx.cs.du.edu PGP2 public key available on request Home: 35 Molewood Close, Cambridge, UK, CB4 3SR; +44 223 327029. These opin- Work and urgent email: iwj@cam-orl.co.uk; +44 223 343398 ions are my own. Olivetti Research Ltd, Old Addenbrookes Site, Trumpington St, Cambridge, UK; ------------------------------ From: garrett@garnet.berkeley.edu (Garrett D'Amore) Subject: Re: Summary of Linux vs. 386BSD vs. Commercial Unixes Date: 18 Apr 1993 06:38:49 GMT In article <9292@blue.cis.pitt.edu> venger+@pitt.edu (The Knight in Silicon Armor.) writes: >In article <1993Apr17.231000.103368@zeus.calpoly.edu> jemenake@trumpet.calpoly.edu (Joe Emenaker) writes: > [a whole BUNCH of stuff deleted] > >I think what's griping Nate is that if someone contributes code to >a LIBRARY under GPL and someone else creates software and uses that >library to compile said software of their own creation, they are not permitted >under the GPL to sell it. It's kind of like Borland coming along and saying >that because you used TurboC to compile software, they are entitled to >a share in any profits since you used their libraries. While I agree that >it would be totally reprehensable for anyone to sell the libraries, I do >not think it is right to tell someone that wrote 10,000 lines of code >that may have 1 or 2 calls to libraries that are under the GPL that >they are not allowed to try to get recompense for their efforts. IMHO, if >I release something into the public domain, my concern as to the use >made of it has ended. > I believe that if you write LIBRARY code using the GLPL (or whatever) you impart the ability to developers using your libraries the ability to sell or freely distribute "derived works". This is different from the standard GPL, which is intended to protect applications. This can be considered analagous to Borland allowing to freely restribute code containing calls to the Turbo C libraries, but prohibiting you from distributing code derived from reverse engineering Paradox. The assumption is that Library code *should* be freely redistributable, at least insofar as part of an executable. But, I think it's fair to expect developers of applications to protect their apps. As far as Linux goes, I'm not sure. Some of the stuff is "library" code, but other stuff should be considered "application" code. In any case, the idea is to protect not just the author, but the general public (i.e. users) from getting ripped off by someone who'd steal code, change a line or two, and charge exorbitant rates for it. If say DEC were to "borrow" the Linux or BSD kernel, change a few lines (like maybe the copyright...) and market it, the people who developed it would feel *very* cheated. GPL protects against that sort of thing. #include ==================================================================== Garrett D'Amore | garrett@haas.berkeley.edu Software Co-Ordinator | 68 Barrows Hall, UC Berkeley Haas Computing Services | Ph: 510-643-5923 Fax: 642-4769 ==================================================================== ------------------------------ From: edmonds@cs.ubc.ca (Brian Edmonds) Subject: Re: New feature for the filesystems. What do you think ? Date: 17 Apr 1993 23:27:40 -0700 oreillym@tartarus.uwa.edu.au (Michael O'Reilly) wrote: . PLEASE look into the structure of the LFS (log file system) written . for the Sprite OS. This is a journalling filesystem. This makes it . absolutely ideal for makeing a compressed file system. (not to mention . that the fact that it is very fast to write to, and nicely crash . resistant). Just to make things easy for everyone, here are _the_ two references on LFS. The first paper I would recommend to everyone to read, as it is my favourite paper of all time (from a grad student who doesn't like reading papers :). M. Rosenblum and J.K. Ousterhout, "The Design and Implementation of a Log-Structured File System," ACM Transactions on Computer Systems, February 1992, Volume 10, Number 1, pages 26-52. (also printed earlier in ACM SIGOPS, OSR October 1992) M. Burrows, C. Jerian, B. Lampson and T. Mann, "On-line Data Compression in a Log-structured File System," Proceedings of the Fifth International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS-V), October 1992, pages 2-9. ======================================================================== Brian Edmonds (MSc CompSci) edmonds@cs.ubc.ca ------------------------------ From: jweigand@astro.ocis.temple.edu (James R. Wiegand) Subject: Re: Summary of Linux vs. 386BSD vs. Commercial Unixes Date: 18 Apr 93 04:59:57 GMT This thread has degenerated into a meaningless stream of name-calling. For the life of me I have no idea what this has to do with Linux. As for the topic, there is nothing appropriate to say about it except take it somewhere else. This kind of rhetoric does nothing other than provoke normally reasonable people into emotional responses. May I suggest the creation of a newsgroup to handle this discussion: alt.flamewares.linux.vs.bsd.kill.kill.kill. jim rm -f ~/.sig ------------------------------ Crossposted-To: comp.os.386bsd.questions From: tmcwill@ukelele.GCR.COM (Thomas McWilliams) Subject: Re: Summary of Linux vs. 386BSD vs. Commercial Unixes Date: Sun, 18 Apr 1993 04:44:52 GMT nate@cs.montana.edu (Nate Williams) writes: >Who really likes GNU utilities, but hates the way alot of folks take >other peoples programs and code, fix them and then don't make the >changes 'completely redistributable; I suppose you are referring to BSDI and BSD/386? Or do you mean Sun? Or DEC? FSF/GNU protects your freedom. The FSF fights the software hoarding that you dislike. Thomas -- tmcwill@gcr.com ------------------------------ Crossposted-To: comp.os.386bsd.questions From: tmcwill@ukelele.GCR.COM (Thomas McWilliams) Subject: Re: Summary of Linux vs. 386BSD vs. Commercial Unixes Date: Sun, 18 Apr 1993 05:20:13 GMT nate@cs.montana.edu (Nate Williams) writes: >The stdio code that HJ has done lots of work on(and I commend him on >it). The original code, before he fixed some of the bugs, was originally >distributed by AT&T, but HJ's fixes are now GPL code, so in order to >get working stdio code we have to do those fixes all over in order to Why do you need Linux developers to fix BSD code? It seems that the logical choice would be to take the fixes from BSDI. They are using the same NET/2 sources as you. BSD/386 is available with source code. If the NET/2 sources had been protected by the GPL then you wouldn't have various BSD factions re-inventing the wheel, duplicating efforts, wasting time, stalling progress. The purpose the GPL is to prevent this lunacy. The GPL encourages the sharing of improvements, a good thing in my opinion. Thomas -- tmcwill@gcr.com ------------------------------ From: graham@prospero.uoregon.edu (Dan Graham) Subject: Re: Telnetd problem in 99.8 Date: 18 Apr 93 00:25:25 I have the same thing, but I've been too lazy to track it down. Basically, somewhere the terminal program is failing to translate CR's to LF's. All you have to do to log in is use ^J ( ctrl-J) to terminate the lines instead of CR or ^M. Most programs either put the terminal back into CR to LF translation, so once you are logged in, CR works for almost everything. ------------------------------ From: philipr@extro.ucc.su.OZ.AU (Philip Rhoades) Subject: Re: Linux on Macintrashes? (STOP ALREADY!) Date: Sun, 18 Apr 1993 07:16:20 GMT rrwood@canrem.com (Roy Wood) writes: >Look, this sort of adolescent bickering about the merits/shortcomings of >a particular brand of computer were fun when I was an adolescent, but now >they're just irritating. Please take it somewhere other than comp.os.linux. >And since I can't stop myself from adding.... >Computers are just tools. Just as most people don't use hacksaws to pound >nails, so too do certain computers/OS's have sensible applications. I've >got access to Macs and PC clones, and whatever does the job best gets used >at that particular moment. >Of course, if you're determined to be irrationally biased as to the fact that >your particular computer is the greatest all-around computing device since >Charles Babbage first coined the expression "Difference Engine"-- well, I'm >certainly not going to waste any more time talking to you. >-Roy The "tools" argument is valid but it is not the whole story - it seems to me that there is an important philosophical (dare I say political?) component to this whole discussion - a very rough analogy: A car with a polluting internal combustion engine might (?) get you to work every day faster than the train but is it a good idea to use the car? Is not the general good better served by using the train? I am not going to argue the merits of corporations like Apple, IBM, Microsoft etc but on balance, with all things taken into account, surely open systems are better for the general good? To be specific: In Australia, Macs are 100% imported but some IBM compatibles have up to 50% Australian content in their manufacture - is the marginal improvement in productivity for some computer jobs (by using Macs) worth the substantial negative effect on the country's current account deficit? I know this is getting further away from the main topic for discussion for this newsgroup but I had to put my 2c worth in. Phil. -- Philip Rhoades First Year Biology University of Sydney phil@biox.bio.su.OZ.AU or philipr@extro.ucc.su.OZ.AU ------------------------------ From: probreak@matt.ksu.ksu.edu (James Michael Chacon) Subject: Re: Questions about virtual mem. Date: 18 Apr 1993 02:50:57 -0500 jmward@ucrengr (jonathan ward) writes: >Peter A. Schwenk (schwenk@fred.cis.udel.edu) wrote: >: I have a few questions about Linux. I haven't read a FAQ, so bear with me. >: Does Linux support virtual memory, or once system RAM is used up, crash city? >Yes, Linux does support virtual memory, in the form of a swap partition. It >can be made as large as you want, although once you get around 12 megs, >anything more is a waste of disk space: it won't be used. I run my system >on four megs with a 12 meg swap partition, and it will run X - very slowly, >but I haven't run out of memory yet( X usually likes eight.). Actually, this completely depends on what you plan on doing with linux. I run a 486-33, with 8M, and 16M swap and can run out of memory easily just by working long enough. I have a couple of applications that use 2-3M, and I don't use X. Compile something large on top af that and you can quickly run out. Remember, with a unix style system more is always better when it comes to memory. This is my current average right now, with 6 VC's open, and 1 telnet connection. total used free shared buffers Mem: 7340 5308 2032 572 1440 Swap: 16376 5896 10480 This is during compilation, so thats why so much free shows up, but it would be pretty easy to run up the other 10M as well. James ------------------------------ ** 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-Activists-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux) via: Internet: Linux-Activists@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 tupac-amaru.informatik.rwth-aachen.de pub/msdos/replace The current version of Linux is 0.99pl6 released on February 23, 1993 End of Linux-Activists Digest ****************************** fun when I was an adolescent, but now >they're just irritating. Please take it somewhere other than comp.os.linux. >And since I can't stop myself from adding.... >Computers are just tools. Just as most people don't use hacksaws to pound >nails, so too do certain computers/OS's have sensible applications. I've >got access tdigest873 644 36676 74430 50323 5366137122 5757 From: Digestifier To: Linux-Activists@senator-bedfellow.mit.edu Reply-To: Linux-Activists@senator-bedfellow.mit.edu Date: Sun, 18 Apr 93 10:00:08 EDT Subject: Linux-Activists Digest #873 Linux-Activists Digest #873, Volume #4 Sun, 18 Apr 93 10:00:08 EDT Contents: Re: how to grind your system to a halt .. vitrual mem, tcsh (HJ Lu) Installing DOS 6.0 and Linux (Dwayne Moore u) Re: how to grind your system to a halt .. vitrual mem, tcsh (Drew Eckhardt) Re: Installing DOS 6.0 and Linux (Terry Evans) Re: Version numbers (W. Tait Cyrus) [Q] Xconfig for ViewSonic6 and Diamond SpeedStar (Tung-Mei X Kuo) Installation question: Kernel panic .... (Tsung-lung Li) Waffle & Linux (Philippe Bonal) CD-ROM NeWS (Bill C. Riemers) Xtrek (X Windows permissions) (Albert King) CDROM help NEC SCSI + CDR-80 (BARRY TITMARSH) Unix Workstation Decision (Mike J. Bell) xdtm problem: xdtm: can't load library '/lib//libc.so.4' (Ron Daisy) Re: ANNOUNCEMENT: Macintosh port of LINUX (free UNIX) now available (alane@wozzle.linet.org) Re: how to grind your system to a halt .. vitrual mem, tcsh (Arjan de Vet) Trouble installing SLS .99p8 (Joe Emenaker) ---------------------------------------------------------------------------- From: hlu@luke.eecs.wsu.edu (HJ Lu) Subject: Re: how to grind your system to a halt .. vitrual mem, tcsh Date: Sun, 18 Apr 93 05:58:57 GMT In article gleasokr@rintintin.Colorado.EDU (Kris Gleason) writes: > >I happened upon an interesting discovery the other day. If you're >curious, try this... > >in tcsh, hold down a key for a while; let the key repeat until the buffer >is full (no more keystrokes will be accepted). Then, hit return. You >should probably run top in another virtual terminal before you hit return. >Then, watch tcsh's memory usage climb and climb and climb, until the >system grinds to a halt (swap space fills up, then the remaining space >in the buffer cache gets eaten until the system can no longer function). >You might also want to set up an infinite loop that calls free, so you >can watch your memory usage directly. Careful, 'cause the only way out >of this is to cycle the power (if you let things get out of hand enough). >I let it run until it took 10 minutes between free's updates... 0 swap, >300 or so cache, 40 free. Yikes! > >I didn't check into the reason that tcsh eats up so much memory. But, >it may be important that I am running with libc4.3.3. > >If you're curious, bash is a bit more graceful... segmentation fault >instead of the grinding. I didn't try any other shells. > >I did this with 8 M of ram, and 16 M of swap (so don't tell me to increase >my swap :) I just found it interesting, and a _huge_ security hole >(unless you want other users to halt your system). > >Just an observation. > I believe rlimit is not 100% enforced. Maybe it is the time to do that. You can set hard limits in files in /etc. The actual limits should depend on your tastes. H.J. ------------------------------ From: moor9881@mach1.wlu.ca (Dwayne Moore u) Subject: Installing DOS 6.0 and Linux Date: Fri, 16 Apr 1993 16:38:17 GMT Has anyone tried installing DOS 6.0 and Linux on the same system as of yet? I was thinking it would be great as my Linux partition could be larger and then I could DoubleSpace my DOS 6.0 partition and have the same amount of storage for DOS 6.0 plus have Linux on my system! :) I'd be glad to hear from anyone who has done this! ------------------------------ From: drew@ladymacb.cs.colorado.edu (Drew Eckhardt) Subject: Re: how to grind your system to a halt .. vitrual mem, tcsh Date: Sun, 18 Apr 1993 08:51:34 GMT In article <1993Apr18.055857.8813@serval.net.wsu.edu> hlu@luke.eecs.wsu.edu (HJ Lu) writes: > >I believe rlimit is not 100% enforced. Maybe it is the time >to do that. You can set hard limits in files in /etc. The actual >limits should depend on your tastes. > Linux handles RLIMIT_CORE, RLIMIT_STACK, RLIMIT_DATA, and on ext2 file systems RLIMIT_FSIZE. You're missing RLIMIT_CPU, RLIMIT_RSS, RLIMIT_NPROC, RLIMIT_OFILE, and RLIMIT_MEMLOCK. -- Boycott USL/Novell for their absurd anti-BSDI lawsuit. | Drew Eckhardt Condemn Colorado for Amendment Two. | drew@cs.Colorado.EDU Use Linux, the fast, flexible, and free 386 unix | ------------------------------ From: tevans%sunset.cs.utah.edu@cs.utah.edu (Terry Evans) Subject: Re: Installing DOS 6.0 and Linux Date: 18 Apr 93 03:05:59 MDT Dwayne Moore u (moor9881@mach1.wlu.ca) wrote: : Has anyone tried installing DOS 6.0 and Linux on the same system as of yet? I have and it seems to work just fine. : I was thinking it would be great as my Linux partition could be larger and : then I could DoubleSpace my DOS 6.0 partition and have the same amount of storage for DOS 6.0 plus have Linux on my system! :) Of course I don't think Linux will be able to read the DoubleSpace partitions. Maybe in the future. : I'd be glad to hear from anyone who has done this! Terry ------------------------------ From: cyrus@jemez.eece.unm.edu (W. Tait Cyrus) Subject: Re: Version numbers Date: 18 Apr 1993 09:37:57 GMT In article jhenders@wimsey.bc.ca (John Henders) writes: > After several frustraiong experiences with linux software, I'd >like to propose that all linux authors and porters try to put a version >number in their binaries that can be called with -v, -V, -h or -?. It >would be to everyone's benifit if there was an easy way to find out what >version of a package, instead of sometimes having to go by the file datw >or what directory on what archive site it was on. > What do people think? How about one of the following? 1) install 'what' under Linux (what looks at files searching for SCCS headers printing them out - hopefully they have version information) 2) add an extra field to the a.out to contain a version number. This could go at the end of the a.out structure can might be called "version_id" and be 32 bits in length producing a version number of the form #.#.#.# where each number can be 0-255 (unsigned). Option #1 advantages 1) just port 'what' to Linux 2) ANY ASCII file can have a legel SCCS field meaning that 'what' can be run on ASCII files in addition to binary files Option #1 disadvantages 1) multiple output lines meaning it would be impossible to parse the output of 'what' to obtain a single version number For example a 'what /usr/ucb/what' on a Sun produces: what.c 1.7 88/08/02 SMI any.c 1.5 88/02/08 SMI 2) no easy way to assign a version number to a library Option #2 advantages 1) ALL programs can have a version number WITHOUT having to be modified like a -v option would require 2) Libraries can also have version numbers (done by adding a dummy file called __Version which contains the version number) Option #2 disadvantages 1) General ASCII files can't have a version number assigned to them. 2) If done the same way that Convex does it 'ld' would have to be modified to add a "default" version number equaling the version number of the kernel. 2A) There are other programs which would need to be modified or added such as 'vers' and 'install' (see below). 3) Would require the version number of a program to be of one of the following formats: #.# [major.minor] #.#.# or [major.minor.patch] #.#.#.# [major.minor.patch.rev] where all unspecified fields are assumed to be set to zero. This is also an advantage since it forces a set way to represent version numbers. Personally I like #2, though I am biased since I used to work at Convex and Convex does things like this and IHMO it worked VERY well. Basically under ConvexOS you have commands 'vers' which set/viewed the version number, 'install' which had a '-v' option allowing you to set the version during an install, and a '-v' option for 'ld' allowing you to set the version number during a compile/load. If #2 is something that is implemented I would be willing to write 'vers'. Comments??? -- W. Tait Cyrus e-mail: cyrus@su.com Solutions Unlimited Phone: 719-260-7227 4710 Nightingale Dr. #M202 Colorado Springs, CO 80918 ------------------------------ From: txk1109@cs.rit.edu (Tung-Mei X Kuo) Subject: [Q] Xconfig for ViewSonic6 and Diamond SpeedStar Date: Sun, 18 Apr 1993 07:39:00 GMT Hi, folks, I am new to Linux. And I really had BIG trouble to setup my monitor and vga card. I am using Diamond SpeedStar 24 and ViewSonic 6. Does anybody have the examples for this kind of combination? Please help me. Any pointer and comment will be greatly appreciated. -Joyce ------------------------------ From: quantum@stein.u.washington.edu (Tsung-lung Li) Subject: Installation question: Kernel panic .... Date: 18 Apr 1993 10:37:39 GMT Hello, After my failure to install linux from a1.test, I tried to start from a1. The followings are what I did. 1. Insert a1 in drive a: and boot the system. The following message appears. Linux version 0.99 pl6-26 Loading 655360 bytes to ram disk 2. Insert a2 in drive b: (I assume a2 is what the "utility disk" prompted by the installation program.) Drive b: starts to run forever. And the following message shows up. task[0](Swapper) killed: unable to recover Kernel panic: Trying to free up swapper memory space In swapper task - not syncing My system is a 486DX2/66 with 16 MB RAM and 240 MB HD. It is a VESA local bus machine. Is my hardware suitable for Linux?(I am asking this because I found some of the documents say that Linux does not support VESA, but some otehr documents say that it does. ..... I have been reading all of these docs, they are too many for me to remember from which doc I read what.) I have been very frustrated by this installation. Any help will be very much appreciated. Tsung ------------------------------ From: philb@blips.fr.mugnet.org (Philippe Bonal) Subject: Waffle & Linux Date: Sat, 17 Apr 93 14:32:07 +002 Reply-To: philb@frmug.fr.mugnet.org Does someone re-write Waffle (uucp) for Linux ? same question for Tass. I know they are yet written for Un*x but not exactly for Linux, and...I don't know C language yet to transpose code. thank you for reading me :-) --- Philippe Bonal philb@frmug.fr.mugnet.org ESI supinfo philb@blips.fr.mugnet.org France ------------------------------ From: bcr@bohr.physics.purdue.edu (Bill C. Riemers) Subject: CD-ROM NeWS Date: 18 Apr 93 06:29:51 GMT For those of you holding off ordering from jana@canrem.com, I just got the first issue and it is great! Basically, it looks like the ftp directories from a mirror site! However, since this only contains 150MB or so of linux stuff, so some of the neat stuff is missing. However, that is why there are 5 more issues in a years subscription. For the price it is great, if just 2 issues out of the 6 this year are this good, it is well worth the cost of subscription! In addition to all this good linux stuff (it has already allowed me to save 34MB of hard disk space I was using) there is about a 250 MB section of NeXT stuff. I'll probably be to lazy to take this CD to work to try it on a NeXT machine, but I found a good portion of this stuff is handy to have with Linux. Just to mention a few things I've found under the NeXT section: Lots of Cooking Repies (That saved me the price of a cookbook.) "Green Eggs and Ham", and a couple of other Dr Suez stories Some *.el files for emacs that I didn't have yet. So anyways, if you were waiting for some recommendations before ordering, here is one. This offer seemed so great to me, that I went out and paid $200 for a Mitsumi CD-ROM drive just so I could subscribe. (OK, I was already planning on eventually buying a CD-ROM drive, I just didn't plan to get one quite so soon.) Bill ------------------------------ From: ak@dcs.ed.ac.uk (Albert King) Subject: Xtrek (X Windows permissions) Date: 18 Apr 93 11:15:51 GMT I am trying to get Xtrek to work on my linux system, I know this is not the serious type of problem you normally find here, but I think it might affect other things aswell. The problem is that when I run xtrek it replies `cannot connect to king:0.0'. king is the name of the server, and I have xtrekd running etc, is this to do with xauth or xhost? (my man page for xauth is not here so I can't check there!) ------------------------------ Date: Sun, 18 Apr 1993 13:16:25 CET From: BARRY TITMARSH Crossposted-To: Subject: CDROM help NEC SCSI + CDR-80 Hi i have a NEC CD-ROM CDR-80 with the NEC scsi iface that comes with the Kit. its an NEC XT001 SCSI works fine with DOS but i have no idear what clone or card it is when used with linux... the linux with all scsi compiled in dont see it at all.. Does any one have any help or idears on getting an NEC - SCSI + CD-ROM CDR-80 drive to work under linux.. Mail to btitmars@esoc.bitnet (best) or C.O.L Thanks Barry. ------------------------------ From: mikeb@umbc.edu (Mike J. Bell) Crossposted-To: comp.sys.amiga.hardware,comp.sys.mac.hardware,comp.sys.ibm.pc.hardware,comp.sys.sgi.hardware,comp.sys.sun.hardware,comp.sys.workstations,comp.sys.next.hardware,comp.os.386bsd.questions Subject: Unix Workstation Decision Date: 18 Apr 1993 07:43:27 -0400 First of all: Thank you everyone who replied. As you can well imagine, I got nearly one hundred responses, so I figured I'd thank you all here instead of individually. The tallies: I got one vote for A/UX, which runs on the Macintosh. I got several nays for Commodore's 3000UX system. Most people agreed that SGI's were nice, and very very expensive. Some (devoted) Sun users nearly insisted that I try to find a used Sun, say a 3/60 or 3/80. About a dozen people told me how they love their NeXT...only now NeXT is out of the hardware business (see below). And most of the other 85+ people said: Get a 486 box, and run some version of Unix that's available either in the Public Domain or commercially. ============= And so I weighed my options. What stood out, in my opinion, was that there were several different versions of Unix available on one system: the 486 (PC) architecture. If I bought a Sun, I'd have to use SunOS. If I bought a (used) NeXT, I'd have to use NeXTSTEP (see below). Ad nauseum. But with a PC, I can run Linux, which is a small user-friendly (?) flavor that uses shared libraries to cut down on volume, although I hear its networking abilities aren't up to speed with 386bsd, another PD (==free) port (which takes up a lot more space). I can purchase BSD386 (commercial product) if I choose. There are several other commercial packages available. So, what I've found out (after pricing systems, etc.), is that I can get a really good Unix workstation for as little as ~$2500. Including source code! Or, for ~$3500-4000, I can get an unbelievable EISA bus 486-66DX2 system, fully blown, and run single user BSD386. What I'm really excited about, though, is NEXTSTEP for Intel. NeXT changed into a software company only, and they are releasing their OS for PC's. I've used NeXT's before, and I absolutely loved them. Everyone who wrote me about NeXT (or NEXTSTEP) commented on the developer's environment, and how well designed the whole package is. =========== So, my cho