Received: from PACIFIC-CARRIER-ANNEX.MIT.EDU by po7.MIT.EDU (5.61/4.7) id AA26715; Wed, 20 Dec 95 01:16:47 EST Received: from SENATOR-BEDFELLOW.MIT.EDU by MIT.EDU with SMTP id AA21109; Wed, 20 Dec 95 01:16:46 EST Received: (from root@localhost) by senator-bedfellow.MIT.EDU (8.6.12/2.3JIK) id BAA10926 for Linux-Development-System-Dist@senator-bedfellow.mit.edu; Wed, 20 Dec 1995 01:13:55 -0500 Message-Id: <199512200613.BAA10926@senator-bedfellow.MIT.EDU> From: Digestifier To: Linux-Development-System@senator-bedfellow.MIT.EDU Reply-To: Linux-Development-System@senator-bedfellow.MIT.EDU Date: Wed, 20 Dec 95 01:13:50 EST Subject: Linux-Development-System Digest #137 Linux-Development-System Digest #137, Volume #2 Wed, 20 Dec 95 01:13:50 EST Contents: Re: mount iso9660 -norock: no UPPER CASE? (Andries Brouwer) Where can I find the newest sysvinit? (S. Lee) Why won't SIGSEGV leave a core? (Jay N. Armstrong) Nakamichi CD changer support Re: Linux Security - C2 and beyond (Christer Weinigel) Re: convert c++ to c (Joe Buck) FreeBSD Impression (Rui-Tao Dong ~{6-HpLN~}) Re: Software Engineers Programmer Analysts - sysw94.fly [1/1] (Goetz Kohlberg) Re: mounting /tmp (Juergen Exner) Re: Will the old extfs be removed from the kernel? (Scot E. Wilcoxon) Re: Are libc functions MT-safe? (Hannes Reinecke) Re: Linux has poor memory management? (Robert Brockway) Re: Will Linux work with the P6 ??? (Vaughan Pratt) All proc's run NICE'd (Robert Larsson) Re: ftape and new kenels (1.3.4*s) (C Shamis) [patch] Make 3.74 under 1.3.4x (c580721) Re: Software Engineers Programmer Analysts - sysw94.fly [1/1] (pbe@pobox.com) ---------------------------------------------------------------------------- From: aeb@cwi.nl (Andries Brouwer) Subject: Re: mount iso9660 -norock: no UPPER CASE? Date: Sun, 17 Dec 1995 00:40:53 GMT ludwig@Informatik.TU-Muenchen.DE (Thomas Ludwig) writes: : I recently got some conference proceedings on CD which can be read via : e.g. netscape. Unfortunately all references in the html-files use : upper case file names. When mounting the CD all names are listed in : lower case. Under Solaris this is also the case BUT all files can also : be accessed by their file name given with upper case letters. This : does not work with Linux. The -norock option of mount doesn't change : anything. I suppose you are not happy with mount /dev/cdrom /cd -t iso9660 -o ro,norock,map=off either? [norock will tell isofs not to use Rockridge extensions; map=off tells it to leave the filename alone. This is good: you will have the upper case that you want, but bad: you also have the trailing ";1" that you probably don't want] If you are sufficiently interested, mail. I can send you a tiny kernel patch that will allow you to say mount /dev/cdrom /cd -t iso9660 -o ro,norock,map=strip to remove the ;1 but leave upper case (or mount /dev/cdrom /cd -t iso9660 -o ro,norock,map=lcase to convert to lower case but leave the ;1). ------------------------------ From: sl14@crux1.cit.cornell.edu (S. Lee) Subject: Where can I find the newest sysvinit? Date: 19 Dec 1995 22:07:22 GMT I know this was discussed a while ago, but I lost the address... Can somebody send me the ftp site for the newest sysvinit? I'm running 2.57b and sometimes it still doesn't run the script to unmount the disk when rebooting... sl14@cornell.edu Witty .sig under construction. ------------------------------ From: armstrng@bnr.ca (Jay N. Armstrong) Subject: Why won't SIGSEGV leave a core? Date: 19 Dec 1995 18:36:26 GMT I had a program stuck in a loop, and wanted a core, so I sent a SIGSEGV to it, and it terminated, printed Segmentation Fault, but didn't say "core dumped", and as far as I can tell, no core was left anywhere. There was no problem with the permissions on the directory or anything, I just think it didn't know to core. Is it configurable in the kernel which signals drop a core as a default action? I would think if anything cored, an 11 would. Thanks, Jay Armstrong ------------------------------ From: uri@vnet.ibm.com () Subject: Nakamichi CD changer support Date: 19 Dec 1995 22:04:50 GMT -- Does anyone know of any driver support for the Makamichi MBR-7 7 CDROM changer? These are SCSI-2 2X CDROM drives, The one I have got is sold for less then $130. The changer appears as 7 CD drives with the DOS driver but can be used only as a single drive with the existing SCSI CDROM support with Linux 1.3.42 . Please let me know if there is any support for a CDROM changer. Seems like creating a custom driver may be worth the effort. Thanks Uri ==================================================== | Uri Chamish [ \ o | | IBM ISSC Boca Raton [__\__/] | | uri@bocaraton.ibm.comT T , Erik Corry wrote: >Markus Kuhn (mskuhn@unrza3.dialin.rrze.uni-erlangen.de) wrote: >: But this is no argument against the introduction of more access >: control levels between the all-powerful user and the normal user. >: Linux and classical Unix has NOTHING between these two levels, and >: this causes a LOT of serious security problems. If there is one tiny >: thing which a normal process can not do, then you have to use >: dangerous things like root uid s-bits in order to allow an >: implementation. > >I wonder how much this actually buys you. I heard that in VMS >the various privilege levels were effectively reduced to 2, like >in Unix, because it was fairly trivial to hack your way from >'slightly privileged' to 'allpowerful'. Actually, these are the five levels: Group: Potential to interfere with members of the same group Devour: Potential to consume noncritical systemwide resources System: Potential to interfere with normal system operation Objects: Potential to compromise the security of protected objects (files, devices, logical name tables, global sections, and so on) All: Potential to control the system Anyways, the biggest advantage (IMHO) of the VMS security system is that it's easier to guard against mistakes. For example, the BYPASS privilege bypasses all permission checks, and lets a process do just about anything at all without any safeguards; the SYSPRV privilege doesn't bypass security, but allows the process to access resources through the System protection field. It's still possible to get around the permission checks, because SYSPRV also grants the right to modify permissions of any resource. For example, a hypothetical Mailer-daemon for VMS could run with the SYSPRV privilege, and would thus be able to access the following file belonging to the user RANDOM_J in the group LUSER. There are four protection fields under VMS; System, for processes which have the SYSPRV privilege, Owner, Group, and World are analogous to Unix permissions. MAIL.DAT;1 Owner: [LUSER,RANDOM_J] File protection: System:RW, Owner:RW, Group:, World: Since a Mailer-daemon would have little use for modifying the permissions of files, it would be much harder to trick it into overwriting critical system files than if it's running with the BYPASS privilege. Of course, no sane person would implement a mailer for VMS this way, since there are better ways of doing this, but hey, "bara ett exempel Helge" (if you don't understand Swedish, just ignore this :-) It probably would be able to do something like this for Linux. For example, processes which have to access privileged sockets might be assigned a `NETPRIV' privilege which could be revoked after opening the socket. I have faint memory of someone posting a description of a patch about a year ago, where he had patched an old kernel to do permission checks by a bunch of capabilities instead of root-only. Unfortunately, I can't find any traces of that post... hold that thought... 2 minutes later... Praise glimpse and the harvest project! :-) -- From: mtpa@mtpa.nerc-bas.ac.uk (Matthew Paley) Subject: Re: A Vision for Linux 1.4 -- POSIX.4 Compatibility Message-ID: <3lrpj4$lgt@kwuz.nerc-keyworth.ac.uk> -- It might be possible to find his post via DejaNews, but if that doesn't work, I'd be happy to mail it to anyone who's interested. Cheers, Christer >-- >We believe that God is on our side -- Netscape chief executive James Barksdale >-- >Erik Corry ehcorry@inet.uni-c.dk -- Christer Weinigel, wingel@ctrl-c.liu.se ------------------------------ From: jbuck@synopsys.com (Joe Buck) Subject: Re: convert c++ to c Date: 19 Dec 1995 03:28:04 GMT lee-d@cs.buffalo.edu (Chain-Wu Lee) writes: >is there a way to convert c++ to c program under linux? I heard that cfront >does it but I cannot find for linux. Thanks for your help There is a commercial port of cfront to Linux. It costs $$$$. You should know that cfront supports a several-years-old dialect of the C++ language and is further behind the draft standard than either g++ or other commercial compilers. -- -- Joe Buck (not speaking for Synopsys, Inc) ------------------------------ From: rdong@printrak.com (Rui-Tao Dong ~{6-HpLN~}) Crossposted-To: comp.unix.bsd.freebsd.misc Subject: FreeBSD Impression Date: 19 Dec 95 01:10:41 GMT I have used my Cipher ST-350 (QIC-02) tape drive for backup, restore as well as installation for almost two years now. However, QIC-02 support has been elapsed in the more recent Linux kernels (such as 1.2.13). I have started to develop problems with the drive (CRC error, drive not ready, etc) lately. Apparently, quite a few people in the Linux camp are having the same problem. I installed a minimal configuration of FreeBSD 2.1 and was hoping to use it as a reliable way of backup and restore for Linux. I found the FreeBSD QIC-02 driver to be more robust than Linux's. When tpqic02 sees an error, it simply quits. I can't even get the files after a file mark. While under FreeBSD, it will skip to the next file. And accessing files after a file mark is no problem. I have salvaged two of my tapes so far. The FreeBSD install procedure is quite impressive. Dynamic driver configuration, installation through an ftp server etc, all in a single floppy. I wish I can do that with Linux. However, I've also encountered some minor problems. I chose to install from an FTP server via PPP. The procedure hangs several times saying something like "Resolving hostname...". I assume a more accurate message should be "FTP server busy"? I have 8MB of memory, but the installer insisted on a swap partition, which I accidently selected the Linux root partition (remember my tape drive didn't work? :-). After got everything sorted out, I learned that Linux ext2 file system support is only in FreeBSD-current. Which means I have to go through a common denominator (which happens to be the DOS file system) between Linux and FreeBSD for now. The biggest problem I have with FreeBSD is the lack of DOCUMENTATION. I bet not many UNIX users know what a MBR is. Comparing with Linux, LILO comes with extension documentation. Explains what is a partition, how does a PC boot, how to use LILO for "foreign" OS etc. I suppose most of the FreeBSD as well as Linux users are UNIX users. They don't know much about PC hardwares. A little "hand holding" as in Linux howtos will be extremely useful. Regards, Rui-Tao Dong ~{6-HpLN~} 2760 Kelvin Ave., #3211 rdong@printrak.com Irvine, CA 92714 (714)752-0855(H) ------------------------------ From: g.kohlberg@samson.mbis.de (Goetz Kohlberg) Crossposted-To: comp.databases.sybase,comp.groupware.lotus-notes.misc,comp.os.ms-windows.programmer.misc,comp.unix.admin,comp.ai.philosophy,comp.benchmarks,comp.cad.autocad,comp.databases.paradox,comp.databases.xbase.fox,comp.dcom.isdn,comp.dcom.sys.cisco,comp.fonts,comp.graphics.algorithms,comp.graphics.animation,comp.infosystems.www.misc,comp.lang.ada,comp.lang.asm.x86,comp.lang.clipper,comp.lang.pascal.delphi.components,comp.lang.pascal.delphi.databases,comp.lang.tcl,comp.multimedia,comp.object,comp.org.cpsr.talk,comp.org.eff.talk,comp.os.linux.x,comp.os.ms-windows.app.word-proc,comp.os.ms-windows.misc,comp.os.os2.bugs,comp.os.os2.games,comp.os.os2.programmer.misc,comp.os.vms,comp.periphs.printers,comp.security.unix,comp.soft-sys.powerbuilder,comp.sys.acorn.apps,comp.sys.amiga.applications,comp.sys.amiga.games,comp.sys.amiga.marketplace Subject: Re: Software Engineers Programmer Analysts - sysw94.fly [1/1] Date: Tue, 19 Dec 1995 17:13:27 GMT Reply-To: Goetz Kohlberg gademsky@localnet.com (Career Opportunity) wrote: > SOFTWARE/DATA PROCESSING/COMPUTER > PROFESSIONALS > $25-85,000/YEAR how nice? Can I have the 85000/Year ? Goetz ;-) ------------------------------ From: jue@cs.tu-berlin.de (Juergen Exner) Crossposted-To: comp.os.linux.misc,comp.os.linux.setup Subject: Re: mounting /tmp Date: 19 Dec 1995 18:44:51 GMT In article <4b6obk$k99@solaris.cc.vt.edu> ise118@vtaix.cc.vt.edu (Ramakrishna Rao V. Kalapatapu) writes: I wonder why it is not possible to mount /tmp in linux. There seems to be no such thing as tempfs in linux. I use 1.2.0 and 1.2.1 versions of the kernel. Any ideas? Although I didn't use it myself I can't see any reason why it should NOT be possible to mount another partition as /tmp. I'm pretty sure, that you made some mistake. What do you mean with "tempfs"? jue -- -- Juergen Exner: jue@cs.tu-berlin.de FAX: 030/314-73623 snail-mail: TU Berlin; Sekr. FR 5-13; Franklinstr. 28-29; 10587 Berlin =-=-=-=-= "I'm sorry Dave, but I can't do that" HAL 2000 =-=-=-=-=-= ------------------------------ From: sewilco@fieldday.mn.org (Scot E. Wilcoxon) Subject: Re: Will the old extfs be removed from the kernel? Date: 11 Dec 1995 16:49:30 -0600 The configuration script should at least ask about the new ext fs first, then ask about the old ext fs, and use the term "old" or "original". I had to interrupt my first configuration due to this confusion. -- Scot E. Wilcoxon sewilco@fieldday.mn.org Laws are society's common sense, written down for the stupid. The stupid refuse to read. Their lawyers read to them. ------------------------------ From: hare@mathi.uni-heidelberg.de (Hannes Reinecke) Subject: Re: Are libc functions MT-safe? Date: 19 Dec 1995 13:55:36 GMT Reply-To: hare@mathi.uni-heidelberg.de =>"Paul" == Paul G Fox writes: Paul> Peter Hofmann Paul> (peter@I_should_put_my_domain_in_etc_NNTP_INEWS_DOMAIN) wrote: >> as of libc version 5.2.16 threads are supported in the linux libc. >> You have to compile the shared pthreads library yourself because >> it is not included in the binary distribution of >> libc-5.2.16. Please read the file release.libc-5.2.16 for details. >> Paul> i'll bet you don't really mean "threads are supported in the Paul> linux libc", do you? is all of libc (malloc, printf, ctime Paul> etc) thread-safe? adding thread creation calls (ala pthreads) Paul> does not make the rest safe. Of course not. The libc-maintainers are not that stupid. Have a look at the sources and you'll know. Regards Hannes ======= Hannes Reinecke | | XVII.: WHAT ? | PGP fingerprint available | T.Pratchett: Small Gods see 'finger' for details | ------------------------------ From: robert@slip.cc.uq.oz.au (Robert Brockway) Subject: Re: Linux has poor memory management? Date: 18 Dec 1995 09:41:17 GMT Reply-To: ec531667@student.uq.edu.au : Sheduler: Choose a process in core to run. : Swapper: Steal page not recently/often used. How would u suggest changing one or both to fix this problem? Do FreeBSD/NetBSD/OpenBSD have the same problem? If not, how do they do it? These three BSD systems are free, and we can thus look to see how they did it (and write our own code of course). -Robert --Robert Brockway, email: ec531667@student.uq.edu.au WWW: http://student.uq.edu.au/~ec531667 "Do not sell IP routers to hostile governments." Harley Hahn, 1993. ------------------------------ From: pratt@leland.Stanford.EDU (Vaughan Pratt) Subject: Re: Will Linux work with the P6 ??? Date: 15 Dec 1995 21:29:38 -0800 In article <30CC6565.751834DE@mmm.com>, John D. Sundberg wrote: > >echo 99^9999 | bc -l is 6 times faster on the 686 133 than my p5 60 Hard to believe. Factoring out the clock ratio of 133/60 still leaves 2.7 to account for. I know the P6 architecture improves on the P5, but by 2.7? What are the actual times you're getting? Vaughan Pratt -- Vaughan Pratt FTP: boole.stanford.edu:/pub/ABSTRACTS http://boole.stanford.edu/boole.html Si non e bene trovato, e vero. (If not well told, at least true) ------------------------------ From: robban@one.se (Robert Larsson) Subject: All proc's run NICE'd Date: Mon, 18 Dec 1995 23:14:56 GMT Now hear this, Something strange happened when I upgraded from 1.3.32 to 1.3.45 -- all my processes (yes, init as well) run with a nice value of 15, when I look at them with 'ps lax' or 'top'. The same thing happened in patch levels 46, 47 and 48. Everything else looks nornal. When I boot with my old 1.3.32 all processes have nice 0 again. The kernel is compiled with gcc 2.7.7 as ELF, but I had the same behaviour when compiled as a.out with 2.6.3, so this shouldn't be the reason. Am I using old versions of 'ps' and 'top' or do I have a hardware- related problem? The BogoMips value is normal (approx 33) when I boot the system, a 66 MHz 486. I can't tell if the system really runs "nice" but I don't think so. Strange, isn't it? Merry Christmas, all Linuxers! ------------------------------ From: cshamis@erinet.com (C Shamis) Subject: Re: ftape and new kenels (1.3.4*s) Date: 19 Dec 1995 18:14:41 GMT Reply-To: cshamis@erinet.com Peter Onion (ponion@srd.bt.co.uk) wrote: : Hi, I've just joined the "I must have the latest kernel" race again, after a : quiet few months of trouble free 1.2.13 computing :-) : : What / where is the latest ftape code? Is it easy to get going, will it work : with 1.3.4X or should I try and use it with 1.2.13? I've had luck with kernel 1.3.45 and zftape1.02, from sunsite. ~sunsite/kernel/tape/zftape1.02.tgz I thinks. I remember it was a good idea to set the "set module information" in the kernel compile to "ON" this time around. And although the newest kernel I'm aware of is .48, I couldnt get zftape to work under .46, so I'm hesitant to try, just yet... I'm just going to coast a little while... I'm tired. -- C. Shamis / This space intentionally CShamis@erinet.com / left blank. ------------------------------ From: root@mizzou1.missouri.edu (c580721) Subject: [patch] Make 3.74 under 1.3.4x Date: 19 Dec 1995 17:33:57 GMT The following Patch fixes Make 3.74 for use with the newest 1.3.x kernels. Some info for the kernel gurus out there: Make broke due to a small change in dirent.h, i.e. the D_NAMLEN macro points to a nonexistent structure member dirent.d_namlen. Since I am not that familiar with the kernel yet, I decided it was easier, and safer, to change Make. There is a structure member dirent.d_reclen : is this the same as d_namlen? If it is then the D_NAMLEN macro should be changed to work with dirent.d_reclen. To fix Make 3.74: apply the following patch to the unaltered make 3.74 sources. Reboot under a kernel that your existing make works under, but leave the include files from 1.3.4x as your standard includes. configure. make. make install. Reboot under kernel 1.3.4x remake make just to be sure that your new make now works. The patched make should work under any kernel. Cheers, Have a Merry Christmas and a Happy New Year, Cayle Graumann c580721@mizzou1.missouri.edu =============Cut Here=============================== --- dir.c.orig Mon Nov 7 18:14:50 1994 +++ dir.c Tue Dec 19 11:37:57 1995 @@ -294,8 +294,13 @@ if (!REAL_DIR_ENTRY (d)) continue; - +#ifndef _LINUX_DIRENT_H +/* Non Linux Dirent structure contains d_namlen */ len = D_NAMLEN (d); +#else +/* Linux 1.3.x does not, so substitute following */ + len = strlen(d->d_name); +#endif while (d->d_name[len - 1] == '\0') --len; ------------------------------ From: pbe@pobox.com Crossposted-To: comp.databases.sybase,comp.groupware.lotus-notes.misc,comp.os.ms-windows.programmer.misc,comp.unix.admin,comp.ai.philosophy,comp.benchmarks,comp.cad.autocad,comp.databases.paradox,comp.databases.xbase.fox,comp.dcom.isdn,comp.dcom.sys.cisco,comp.fonts,comp.graphics.algorithms,comp.graphics.animation,comp.infosystems.www.misc,comp.lang.ada,comp.lang.asm.x86,comp.lang.clipper,comp.lang.pascal.delphi.components,comp.lang.pascal.delphi.databases,comp.lang.tcl,comp.multimedia,comp.object,comp.org.cpsr.talk,comp.org.eff.talk,comp.os.linux.x,comp.os.ms-windows.app.word-proc,comp.os.ms-windows.misc,comp.os.os2.bugs,comp.os.os2.games,comp.os.os2.programmer.misc,comp.os.vms,comp.periphs.printers,comp.security.unix,comp.soft-sys.powerbuilder,comp.sys.acorn.apps,comp.sys.amiga.applications,comp.sys.amiga.games,comp.sys.amiga.marketplace Subject: Re: Software Engineers Programmer Analysts - sysw94.fly [1/1] Date: 20 Dec 1995 03:31:10 GMT >to find a position for me. Come to think of it, the salary's too low >anyway. What are you recruiting for, Nebraska? $85,000 is too low? What do you do? SAP? _ This isn't the moon? This is PA? What are all those holes? ======================================================== |Chad Z. Hower - PBE@pobox.com - http://pobox.com/~pbe | |Independent Consultant as Phoenix Business Enterprises| | & | |Senior Analyst Programmer - Seltmann, Cobb & Bryant | ======================================================== >>SQUID - The ultimate database reader, and NO limits. #$737961 **Special Compile: 1.033B (Beta) ------------------------------ ** FOR YOUR REFERENCE ** The service address, to which questions about the list itself and requests to be added to or deleted from it should be directed, is: Internet: Linux-Development-System-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux.development.system) via: Internet: Linux-Development-System@NEWS-DIGESTS.MIT.EDU Linux may be obtained via one of these FTP sites: nic.funet.fi pub/OS/Linux tsx-11.mit.edu pub/linux sunsite.unc.edu pub/Linux End of Linux-Development-System Digest ******************************