From: Digestifier To: Linux-Activists@senator-bedfellow.mit.edu Reply-To: Linux-Activists@senator-bedfellow.mit.edu Date: Fri, 20 Aug 93 14:13:16 EDT Subject: Linux-Activists Digest #147 Linux-Activists Digest #147, Volume #6 Fri, 20 Aug 93 14:13:16 EDT Contents: Re: Unix close for 486 - commens requested (Larry D Snyder) sdsafdasdf (Projecto VLSI) MetroLink Motif/Linux::Some Resource files not traced::Help (SURYARAO TIPPARAJU) Re: TI travelmate laptop problems? (Jon Gefaell) Re: NetBSD's ash as /bin/sh substitute on Linux (Chet Ramey) how to set the "dot-clock match mode" ?? (Sun Jyh-bang) Re: OAK/VGA 512K Question (Miguel Alvarez Blanco) Targa Cube VL, any experiences? (Robert_Weissenfels) Networking (Harald Finnaas) FAQ Please ... (H. Mike Rice) Re: SCSI Performance (Piercarlo Grandi) twm window manager question (John Fauerbach) SCSI doesnt find my hd in Linux Re: twm window manager question (Robert Nation) Re: Compiling kernel (Richard Kasperowski) Re: Unix close for 486 - commens requested (Bdale Garbee) Term security flaw (Re: Term limitation...) (Olaf Titz) Re: Modem XTERM emulation? (Olaf Titz) Segmentation Fault (Brian Quandt) SCSI compatibility and local bus (Brian Quandt) ---------------------------------------------------------------------------- Crossposted-To: comp.unix.pc-clone.32bit,comp.unix.bsd,comp.unix.questions,comp.os.mach,comp.unix.solaris From: larry@gator.oau.org (Larry D Snyder) Subject: Re: Unix close for 486 - commens requested Date: Fri, 20 Aug 1993 13:44:28 GMT guyd@austin.ibm.com (Guy Dawson) writes: >I use BSD/386 from BSDi - it's based on the BSD Net-2 release. >It costs $1000 with source ( $595 ish ) without and is proving >to be a reliable system. I'm pretty sure it will run on a I've heard excellent things about BSD/386 from several folks here in the net -- and I'm just about ready to send off for information what type of documentation comes with it? -- Larry Snyder Internet: larry@gator.oau.org Orlando, Florida UUCP: ..!uunet!tarpit!gator!larry ------------------------------ From: vlsi@inesc.inesc.pt (Projecto VLSI) Subject: sdsafdasdf Date: Fri, 20 Aug 1993 09:09:24 GMT ------------------------------ From: tipparaju@lheavx.gsfc.nasa.gov (SURYARAO TIPPARAJU) Subject: MetroLink Motif/Linux::Some Resource files not traced::Help Date: 20 Aug 1993 10:26 EST Hai: I have MetroLink Motif installed. Everything seemed O.K but, I observe the following problems: Some X applications (xterm etc.) find their resource files (in /usr/lib/X11/app-defaults directory) but xman does not see its file. I observed this problem only when I found my own application was not tracing its resources. Is there something that I am missing? XFILESEARCHPATH, XUSERFILESEARCHPATH, XAPPLRESDIR are set as follows /usr/X386/lib/X11/app-defaults/%N $HOME/app-defaults/%N $HOME/app-defaults I appreciate someone giving me some insight and help Thanks Suri Tipparaju rao@heasfs.gsfc.nasa.gov tipparaju@lheavx.gsfc.nasa.gov ------------------------------ From: jeg7e@livia.acs.Virginia.EDU (Jon Gefaell) Subject: Re: TI travelmate laptop problems? Date: Fri, 20 Aug 1993 13:40:46 GMT In article <1993Aug20.075622.22566@rp.csiro.au> mchivers@rp.CSIRO.AU (Michael Chivers (Summer Student)) writes: >A few months back a group of people here tried to get Linux AND X11 running >on a TI Travelmate active TFT colour machine. The main problem then seemed to >be the video driver (they could only get monochrome). > >Aparently there are drivers for the Cirrus chip inside the Travelmate and >I see in linux.announce just now that there is an alpha release of the ballpoint No, there are not drivers for the SCirrus chipset in the TI. There ARE drivers for 54xx Cirrus chips, but the ti has the _6_4xx chips... -- ______ \ \ / Jon Gefaell, Computer Systems Engineer | Amateur Radio - KD4CQY \/\/ Information Technology and Communications | -Will chmod for food- \/ The University of Virginia, Charlottesville | Hacker@Virginia.EDU Any opinions expressed herein are not intended to be construed as those of UVA ------------------------------ From: chet@odin.ins.cwru.edu (Chet Ramey) Crossposted-To: comp.unix.shell Subject: Re: NetBSD's ash as /bin/sh substitute on Linux Date: 20 Aug 1993 14:38:54 GMT In article <109551@hydra.gatech.edu>, Howlin' Bob wrote: >>Minimalism, I suspect. Anything to keep that /bin/sh binary small. > >To harp on "modern," it should be pointed out that modern Unices usually >demand-page their executables. If the test portion of the code were not >used, then it wouldn't be paged in. You're right, but that's not what I'm talking about. As it was explained to me, the BSD folks want the binary itself to be as small as possible, to fit on extremely small root partitions and take up as little space as possible on boot disks. For this they are willing to sacrifice some speed. -- ``Times are hard, and you're afraid to pay the fee. So you find yourself somebody who can do the job for free.'' Chet Ramey, Case Western Reserve University Internet: chet@po.CWRU.Edu ------------------------------ Crossposted-To: nctu.talk,tw.comp.unix From: u7923081@cc.nctu.edu.tw (Sun Jyh-bang) Subject: how to set the "dot-clock match mode" ?? Date: Fri, 20 Aug 1993 14:31:17 GMT hi, recently i was confused when setting up Xconfig for my Linux OpenWin, my screen is non-interlaced with 1024x768 for clock 60Hz and 800x600 for 72Hz, how can i set the dot-clock match mode for "1024x768" ? thanks for your answers in advance. Aries in NCTU, HsinChu, Taiwan ------------------------------ Subject: Re: OAK/VGA 512K Question From: miguel@pinon.ccu.uniovi.es (Miguel Alvarez Blanco) Date: 20 Aug 93 15:27:27 +0100 Craig (Craig.Humphrey@comp.vuw.ac.nz) wrote: : But still only on X386mono, though I've heard that Xfree1.3 : supports more chipsets, including the OAK, but I haven't been able to confirm : this. Nay, Xfree1.3 doesn't support OAK cards, because OAK didn't lean their specs. Last month they sent the specs to the Xfree team, and they're supposedly working on an OAK driver. Nevertheless, I wouldn't expect it to be out until the Xfree2.0 release (currently in early beta). Don't know when would this be, but I'm looking every day at the X group to see if it's out. I've a 1MB OAK card, and use it only in mono ! Miguel Alvarez Blanco "All that is gold does not glitter, miguel@hobbit.quimica.uniovi.es not all those who wander are lost." miguel@pinon.ccu.uniovi.es Bilbo Baggins. ------------------------------ Crossposted-To: comp.os.linux.misc,comp.os.linux.help From: robert@luzi.e-technik.uni-dortmund.de (Robert_Weissenfels) Subject: Targa Cube VL, any experiences? Date: Fri, 20 Aug 1993 14:43:54 GMT The mentioned graphics-card (S3/911 a.f.a.i.k.) is quite affordable at the moment. Does anybody have any experiences with that thing? Thanx Robert P.S. Please answer by mail, I'll post a summary later on. ------------------------------ From: harald-f@fidonet.bbs.no (Harald Finnaas) Subject: Networking Date: Fri, 20 Aug 93 17:40:59 GMT I have a couple of questions: I've got a server running LAN Server 3.0 (OS/2), and I'm wondering if there is any way I can connect my Linux machine to the server? I've got a 3Com 507 card in the Linux machine at the moment btw. I also have a problem with BOOTLIN. I've copied the zImage file onto my HD using MTOOLS, but just after it's has checked my HD's (2xIDE), it hangs with some SCSI errors. And I don't even have a SCSI drive in that machine! Anyone? _/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/ _/ Harald Finnaas (Stinger) | Romeo November ...... +47-4-536698 _/ _/ | N-4026 Stavanger, Norway. _/ _/ harald-f@fidonet.bbs.no | FIDO: 2:211/5@fidonet.org _/ _/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/ ------------------------------ From: dsc9hmr@bumed10 (H. Mike Rice) Subject: FAQ Please ... Date: Fri, 20 Aug 1993 14:09:42 GMT Sorry ... Over Done Question .. Where can I get the FAQ for Linux ? Also ... Where can I get all of Linux ?? Sources and Binaries ... I just got an OOooold Beatup 386 mother board ... gonna attempt Linux ! Thanks, -- rice -- ~~ | H. Mike Rice (@@) __o Whistle'n | | dsc9hmr@bumed10.med.navy.mil < _`\<,_ -n- |_|_____ BUMED: (202) 653-0413 o) (*)/'(*) Wheel'n |____________________________________________________________________ ------------------------------ From: pcg@aber.ac.uk (Piercarlo Grandi) Subject: Re: SCSI Performance Reply-To: pcg@aber.ac.uk (Piercarlo Grandi) Date: Fri, 20 Aug 1993 15:33:17 GMT >>> On Wed, 18 Aug 1993 00:20:05 GMT, danny@dragon.stgt.sub.org (Daniel >>> T. Schwager) said: Daniel> IOZONE writes a 8 Megabyte sequential file consisting of Daniel> 16384 records which are each 512 bytes in length. Daniel> It then reads the file. It prints the bytes-per-second Daniel> rate at which the computer can read and write files. Daniel> Writing the 8 Megabyte file, 'iozone.tmp'...20.670000 seconds Daniel> Reading the file...35.630000 seconds Daniel> IOZONE performance measurements: Daniel> 405834 bytes/second for writing the file Daniel> 235436 bytes/second for reading the file Daniel> --------------------------- Daniel> I don't know the problem, but for a syncron transfer it is a Daniel> little bit too slow ! Probably IOZONE is not the most interesting benchmark here. What's likely happening is that first the IOZONE writes the 8MB file into the buffer cache, at least partially, then 'sync' starts writing it out, and while this is going on IOZONE rereads the file. The *only* accurate way to test effective read and write rates is: 1) mkfs an empty disk partitions 2) mount the filesystem 3) write the file to the filesystem a longer than total memory; fsync 4) umount the filesystem 5) mount it again 6) read the file from the filesystem. 7) ummount the filesystem You should time the time from the start of writing to the completion of fsync in step 3). Also, step 4) is essential. This said, the ext2fs filesystem does not give the optimal bandwidth (even if I think it is better than the BSD UFS, as it is less complicated); it has mainly two defects: the allocation strategy becomes suboptimal with full disks, and, more worryingly, it does not do dynamic (at read/write time) clustering, which is the worst defect. I also think that looking for runs of 8 free blocks is not a particularly effective static (at allocation time) clustering strategy, bu without having a look at the block addresses for a few files it is hard to be sure. If anybody feels like investigating this latter point, it is very easy to instrument fsck so that for each file it reports the runs of absolute block addresses it occupies. This would make it easy to gauge how effective is static clustering. I cannot remember well, but perhaps dynamic clustering would require some changes in the strategy routine of the drivers. If not, adding it is probably a two minute change: just figure out whether accesses to the file are being sequential, and if so preallocate and then read/write ahead/behind N (where N may be fixed, say 8-32, or variable) blocks instead of 1. If this is done supporting large blocks and fragments thereof becomes useless. ------------------------------ From: fauerbac@canton.cs.unca.edu (John Fauerbach) Subject: twm window manager question Date: 20 Aug 1993 15:28:04 GMT How can I get pager to work with twm. I have seen it in openlook and fvwm, but not twm. Thanks, John Fauerbach ------------------------------ Date: Fri, 20 Aug 1993 17:58:19 CDT From: Subject: SCSI doesnt find my hd in Linux hello, i have a problem with the SCSI driver. it doesnt recognize my harddisc. configuration: OS-version: linux-pl12, aug-15 (without the net22 patches) Controller: Seagate ST02: HDs: a) ST-296N jumpered at address 1 b) CDC 94221-5 59780015480 , cly: 1515 head: 5 sect: 47 CDC = control data jumpered at address 0 I have already been able to read/write the disc: a) took the disc and put it in a Sparc. no problem. using the sun-monitor command "probe-scsi" reported all attached SCSI devices; and that's where I got the disc description from, (the string saying CDC 94221-5 ....) b) took the disc, and put it in the PC. low-levelformatted it (with debug;g=c800:5) used dos/fdisk to generate a partition on the CDC used dos/format to generate a MS/Dos filesystem did a copy a:*.* c: , which worked okay. a "dir c:" showed all files on the CDC disc being present. c) when I boot Linux, only the SCSI Adapter is recognized. the SCSI Disk is NOT recognized, it says: scsi: 1 hosts. scsi detect 0 discs total. Even when booting the machine , the ST02 BIOS will report that there IS a disc present. I think, the problem is in the SCSI driver in Linux. On the other hand, when I connect the ST-296N disc, this disc wil be recognized, but still the CDC disc will be missing. Summarize: worked with Sparc, worked with MS/Dos, did not work with Linux. I think the problem could be in the SCSI device driver... Can anyone tell me where to start searching for the bug ? Greetings, Herp ------------------------------ From: nation@snoopy.sanders.lockheed.com (Robert Nation) Subject: Re: twm window manager question Date: 20 Aug 93 12:54:36 twm does not come with a pager. You would need to get tvtwm or ctwm to get a virtual desktop. Rob (nation@rocket.sanders.lockheed.com) ------------------------------ From: richk@icad.com (Richard Kasperowski) Subject: Re: Compiling kernel Date: 20 Aug 1993 13:40:31 -0400 Reply-To: richk@icad.com (Richard Kasperowski) It's true. I went so far as to download the _Kernel_Hacker's_Guide_, hoping to find out what to do after compiling a new kernel, but there's no information there either. So what do you do after compiling a new kernel in order to use it? -- Rich Kasperowski richk@icad.com ------------------------------ From: bdale@col.hp.com (Bdale Garbee) Crossposted-To: comp.unix.pc-clone.32bit,comp.unix.bsd,comp.unix.questions,comp.os.mach,comp.unix.solaris Subject: Re: Unix close for 486 - commens requested Date: 20 Aug 1993 17:45:07 GMT Larry D Snyder (larry@gator.oau.org) wrote: : I've heard excellent things about BSD/386 from several folks here : in the net -- and I'm just about ready to send off for information : what type of documentation comes with it? A printed installation guide that's well balanced between terseness and completeness. The remainder of the documentation is all online, as man pages or other document files... not to mention the sourcecode itself. If you can, go for the CDROM distribution, as you can put the disk online and have all the goodies up for perusal without soaking your hard disk. If you've ever admined a BSD'ish machine before, you'll not have much trouble. If you haven't, then I suggest you troll the bookstores for a book on system administration in a BSD'ish environment. That and the man pages should take care of anyone. I don't have a good book recommendation off the cuff, though. Bdale, another very satisfied BSDI customer. ------------------------------ From: s_titz@ira.uka.de (Olaf Titz) Crossposted-To: comp.os.linux.misc Subject: Term security flaw (Re: Term limitation...) Date: 20 Aug 1993 17:53:23 GMT In article <2521mn$a7o@uniwa.uwa.edu.au> oreillym@tartarus.uwa.edu.au (Michael O'Reilly) writes: > : and I have a 14.4 modem connected to my local school, I want to run > : term as root right before I run xdm on my console at startup so that ^^^^ > : any of my users can termtelent or termftp or termfinger to any > : machine in the universe... well my problem is that whoever runs > If you run term as user 'joe', then anyone other than joe should do: > 'export TERMDIR=~joe' (under bash) or > 'setenv TERMDIR ~joe' (under tcsh) Right, (actually I would recommend to put it in /tmp in this case) but it should be recommended to NOT run term as root. Reason? This would give the root on the school machine root on your home box as well (actually anybody who can access the socket). Or have I missed something important here? Perhaps an option to term to deny trsh connections would be nice (you still can get in via trsh and telnet, but then you have to provide the proper password.) Olaf -- olaf titz o olaf@bigred.ka.sub.org praetorius@irc comp.sc.student _>\ _ s_titz@ira.uka.de LINUX - the choice karlsruhe germany (_)<(_) uknf@dkauni2.bitnet of a GNU generation what good is a photograph of you? everytime i look at it it makes me feel blue ------------------------------ From: s_titz@ira.uka.de (Olaf Titz) Crossposted-To: comp.dcom.modems,comp.os.ms-windows.apps,comp.os.os2.apps,comp.os.linux.misc Subject: Re: Modem XTERM emulation? Date: 20 Aug 1993 17:56:30 GMT In article gmt@beach.cis.ufl.edu (Gary McTaggart) writes: > >> Anyway, my quest is to find a modem terminal emulation that would do xterm > >> when dialing into a unix box. > Since Seyon uses XTerm basically as a vt100 terminal emulator, couldn't > xterm be modified in this case to send vt100 escape codes to move the > cursor and stuff like that? It seems like it would be fairly easy to at Isn't it just simpler to run term/txconn and have a genuine xterm from the other side on your screen? Olaf -- olaf titz o olaf@bigred.ka.sub.org praetorius@irc comp.sc.student _>\ _ s_titz@ira.uka.de LINUX - the choice karlsruhe germany (_)<(_) uknf@dkauni2.bitnet of a GNU generation what good is a photograph of you? everytime i look at it it makes me feel blue ------------------------------ From: quandt@cs.umr.edu (Brian Quandt) Subject: Segmentation Fault Date: Sun, 15 Aug 1993 20:08:20 GMT Okay, this probably is not going to help. But, if anyone can give some advice I'd really appreciate it. I have a piece of software that runs great under SunOS 4.1.2 but every time I try to run (after of course recompiling) on Linux I get a segmentation fault. I have tried to debug this code to no avail. Bascially, it appears to be a moving bug, the code that I get the annoying sig-11 is nothing special, either some assignments (pointer copies), or a function call. Again, I said this thing moves based on how I try to debug, (I have tried gdb, and the old fashion way of printf's). I can of course comment out the call, and the code runs appropriately (minus what the function was supposed to do). Tried to trace the function back to the beginning and of course did not see anything useful. Basically, its beginning to look like a bug in either the compiler or the OS (nahhh, its got to be in my code somewhere, but why would it work without a hitch on a sun). As I said if someone can tell me that there are or are not bugs in the compiler/os... or what to look for, or just give me a word of incouragement... Frustrated Programmer if anyone ------------------------------ From: quandt@cs.umr.edu (Brian Quandt) Subject: SCSI compatibility and local bus Date: Mon, 16 Aug 1993 17:00:55 GMT Looked through the compat.list did not see anything about local bus and 486's. Has anyone tried a local bus and SCSI controller under LINUX? If so which one? (That is I'm looking to purchase a local bus SCSI controller.) Brian Quandt PS (not linux specific) Also looking for recommendations on fast lasrge SCSI drives. I'm looking for sustained through-put, not just fast random access times which are of course aided by on board caches. The key figure I need is at least 2MB/s writes and reads. ( ------------------------------ ** 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 0.99pl9 released on April 23, 1993 End of Linux-Activists Digest ******************************