Received: from SOUTH-STATION-ANNEX.MIT.EDU by po7.MIT.EDU (5.61/4.7) id AA01965; Sat, 9 Dec 95 18:21:01 EST Received: from SENATOR-BEDFELLOW.MIT.EDU by MIT.EDU with SMTP id AA16009; Sat, 9 Dec 95 18:21:04 EST Received: (from root@localhost) by senator-bedfellow.MIT.EDU (8.6.12/2.3JIK) id SAA28486 for Linux-Development-System-Dist@senator-bedfellow.mit.edu; Sat, 9 Dec 1995 18:16:51 -0500 Message-Id: <199512092316.SAA28486@senator-bedfellow.MIT.EDU> From: Digestifier To: Linux-Development-System@senator-bedfellow.MIT.EDU Reply-To: Linux-Development-System@senator-bedfellow.MIT.EDU Date: Sat, 9 Dec 95 18:16:48 EST Subject: Linux-Development-System Digest #100 Linux-Development-System Digest #100, Volume #2 Sat, 9 Dec 95 18:16:48 EST Contents: Re: Motif development libraries for linux, correction :) (Adam B. Meltzer) Re: Help: select() system call unexpected behavior (Matthias Urlichs) Re: Floating point processor - help (Peter Dalgaard BSA) Re: Converting dynamic to static ? (Joel M. Hoffman) Development tools for linux (Rainer Hochgeladen) Re: Library version number bugs (Wolfram Gloger) Re: malloc and security (Roy Lakin) Re: Limiting diskcache/buffers? (Mikael Abrahamsson) Possible bug in .3.43 and .3.45 (Vincent Cojot) Re: Building Linux from scratch....?? (Dimitris Evmorfopoulos) Re: System does not reboot (Dimitris Evmorfopoulos) ldconfig, getcwd() hosed with ELF -- help! (Thomas Boutell) Re: HELP! Need restricted shell for Linux!!!!!!!! (William Evans) ---------------------------------------------------------------------------- From: rampage@ccnet.com (Adam B. Meltzer) Crossposted-To: comp.os.linux.development.apps,comp.os.linux.x Subject: Re: Motif development libraries for linux, correction :) Date: 9 Dec 1995 09:42:15 -0800 On the dark day of 5 Dec 1995 15:42:57 -0500, Charles Meystrik screamed... > I submitted a request for Motif 2.1 about a week ago and was corrected, > thankfully. I am looking for the 2.0 libraries for development work on > my masters thesis, Motif 2.1 has not been released by OSF yet. If anyone > knows of a free library kit that would not take a great deal of porting > from Motif 2.0 that would be appreciated also. Else, if you know where > I can get the Motif development libraries for linux please let me know. > Reply by e-mail is probably best to keep the clutter out of the news > group. As far as I know of you can only get Motif in the store ;). If you want to do Motif developing, you might want to get a copy of LessTif. As far as I know, however I don't know of anyone who has been able to get any Motif apps to run after compiling them with LessTiff (just dummy libs, mostly). -- ___ ___ ____ ___ ___ _____ ___ ~\~ ~/~ '~|~~`'~|~`'~|~` '~|~~~` /~~~` +------------------- --- -- - . \ / ___ | | | | | | rampage@dopeman.com X ~~~ |== | | =|===+ \===\ +------------------- --- -- - . / \ | | | | | | "It is judgement that _/_ _\_ ._|_. ._|_.._|___|._|___| `___/ : defeats us" ~~~ ~~~ ~~~ ~~~ ~~~~~~ ~~~~~~ ~~~ . -- Apocalypse Now isthemostkradfuqink00lshowontvtodayrocks! . ------------------------------ From: smurf@work.smurf.noris.de (Matthias Urlichs) Crossposted-To: comp.os.linux.development.apps Subject: Re: Help: select() system call unexpected behavior Date: 9 Dec 1995 20:17:03 +0100 In comp.os.linux.development.system, article <4a59dc$4g5@news.rain.org>, Fabrice Franceschi writes: > > I'm facing a problem when using select(). > Basically, when I do a select() after a sendto(), the select > does not have the expected behavior. Instead of waiting for the > end of the timeout, it exits right away claiming there is something > to read, which is wrong since nothing is sent to this processus. > It's not "there is something to read", it's "a read on the fd will not block". In other words, you sent to a nonexisting socket and got an error message back, which the kernel is dutifully passing along to you. The other explanation is that select behaves as documented (unless these other Unixes, where the unmolested timeout has been mentioned as a bug-to- -be-fixed for years and years now) and changes the timeout value. -- Matthias Urlichs \ XLink-POP N|rnberg | EMail: urlichs@smurf.noris.de Schleiermacherstra_e 12 \ Unix+Linux+Mac | Phone: ...please use email. 90491 N|rnberg (Germany) \ Consulting+Networking+Programming+etc'ing 42 PGP: 1B 89 E2 1C 43 EA 80 44 15 D2 29 CF C6 C7 E0 DE Click here. ------------------------------ From: pd@kubism.ku.dk (Peter Dalgaard BSA) Subject: Re: Floating point processor - help Date: 08 Dec 1995 18:05:35 +0100 In article <4a622a$dgv@africa.cis.co.za> jvisser@cis.co.za (jvisser) writes: I am porting an app from HP-UX to Linux(1.2.8) and I am having some problems with floating point On the HP I can 1)ignore under/over-flows 2)decide which SIGFPE's I want to trap etc. I have a set of clone routines of the Sun/Solaris sigfpe code. It has helped at least two persons previously. I can send you a copy if you want. I have been reluctant to put them up for public access, because they rely on the way current kernels handle FPU exceptions (by stashing the raised exceptions in another FPU register), which I tend to think is a bit of a kludge and a candidate for change in future releases (using sigcontexts e.g.). Any ideas how to handle this, anyone. -- O_ ---- Peter Dalgaard c/ /' --- Dept. of Biostatistics ( ) \( ) -- University of Copenhagen ~~~~~~~~~~ - (pd@kubism.ku.dk) ------------------------------ From: joel@wam.umd.edu (Joel M. Hoffman) Subject: Re: Converting dynamic to static ? Date: Fri, 8 Dec 1995 20:39:38 GMT In article pkc@scs.carleton.ca (Peter Choynowski) writes: >In order to save some time I like to find out if there is a way to convert >a dynamicly linked binary e.g. /sbin/init into a static version ? Well, you can certainly compile your own version, and statically link it. If you're really tight on space, this might be a good idea anyway, as you can tell the compiler to optimize for size, not speed. -Joel (joel@exc.com) -- ============================================================================= |_|~~ Germany, Europe. 1940's ``A DISTINGUISHED speaker said that we must put __|~| 16 Million DEAD. our children first. We in Bosnia wonder if we have any children left. Seventeen thousand cnc Bosnia, Europe. 1990's children have been killed in Bosnia in the cnc HOW MANY MORE? last three years. Those living, some with gray hair and eyes and hearts of old men, are ``May the world you hardly children any more.'' live in be the world - H. Silajdzic of your dreams.'' - Debbie Friedman Prime Minister of Bosnia ============================================================================= TELL CLINTON TO HELP STOP THE GENOCIDE: PRESIDENT@WHITEHOUSE.GOV ------------------------------ From: hochgeladen@rhnulm.opennet.de (Rainer Hochgeladen) Subject: Development tools for linux Date: 7 Dec 1995 15:28:22 GMT Hi, are there tools for generateing a function calling tree trough different source files at a time? Thanx for help! ------------------------------ From: Gloger@lrz.uni-muenchen.de (Wolfram Gloger) Subject: Re: Library version number bugs Date: 9 Dec 1995 13:30:30 GMT Reply-To: Gloger@lrz.uni-muenchen.de pfnguyen@Viet.Viet.Com (Perry F Nguyen) writes: >In article <4a9b5e$lo0@sparcserver.lrz-muenchen.de>, >Wolfram Gloger wrote: >>Now this complaint directly contradicts your point 1. above. With >>ncurses, the author decided to seperate release version number from >>API version number, and to avoid confusion, he started with API >>version 2.0 (since several 1.x shared libs were in use before he >>supported shared libs officially). So where's the problem ? >ncurses should have had its major version number updated to 2. >ie. ncurses 2.0 And this is exactly what has happened. The major number is now 2. Only the release version has remained 1.9.x, for whatever reason. Wolfram. ------------------------------ From: roy@dsbc.icl.co.uk (Roy Lakin) Subject: Re: malloc and security Date: Tue, 5 Dec 1995 08:59:27 GMT In article <498egj$4vf@hermes.synopsys.com> jbuck@synopsys.com (Joe Buck) writes: > >albert@krakatoa.ccs.neu.edu (Albert Cahalan) writes: >>>Hmmm, I wonder how many bugs are hidden by this feature. I think we could >>>find many bugs if the kernel filled the page with 0xde instead. > >im14u2c@cegt201.bradley.edu (Joe Zbiciak) writes: >>Wait, Linux is a 32-bit OS. Why not 0xdeadbeef instead? > >Use the source, Luke. If you guys think this is a great idea, it's >not that difficult to patch the kernel to do it (and share the patch >so others can have it). > >I think the pages should be filled with random Zippy the Pinhead quotes, >so that if anyone prints uninitialized strings it would be interesting. >:-) > >(Or perhaps Klingon poetry, now that Klingon is an officially supported >language). > Vogon poetry. After some of that, the perpretrator won't want to do it again :-) roy ------------------------------ From: swmike@uplift.sparta.lu.se (Mikael Abrahamsson) Crossposted-To: comp.os.linux.misc Subject: Re: Limiting diskcache/buffers? Date: 9 Dec 1995 18:46:18 GMT In article <4acc1d$kh6@viet.viet.com>, Perry F Nguyen wrote: >In article , >>Well... Yes, that is probably a good idea, but I still dont want it to >>have more than 5 megs of disk buffers. In my opinion that is never >>needed, and there is no reason for it to grow to more than that. Is there >>any chance of this being an option in a future kernel? > >Disk buffers are dynamically allocated as your run the system, it can >be almost all of your ram, to 0 of your ram, depending on when your >system needs it. So, just let your kernel handle the buffer size, it >knows what the best size to use is. No, it does not, when it swaps out my netscape and thinks it wants half of my 48 megs of ram as diskcache. I see absolutely no reason for having more than 5-10 megs of ram as diskbuffer, and 30megs is definately in excess!!!! -- ===== Mikael Abrahamsson email: swmike@df.lth.se Student vid Institutionen f|r Informatik vid Lunds Universitet, Sweden. ------------------------------ From: coyote@step.polymtl.ca (Vincent Cojot) Subject: Possible bug in .3.43 and .3.45 Date: 9 Dec 1995 18:25:05 GMT Hi, this a possible bug report even though I don't consider myself competent enough to say it's a bug. Follows is my sig because this could get lengthish. If someone could tell me what I am doing wrong. To those writing the kernel and its drivers: Keep up the good work! :) ,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-, Vincent S. Cojot, Computer Engineering. Ecole Polytechnique de Montreal, Comite Micro-Informatique. coyote@step.polymtl.ca, coyote@info.polymtl.ca Description of Problem: ************************ Upon bootup, with both 1.3.43 and 1.3.45 compiled with all my standard options (that I have been using since 1.1.95) here is what happens: "Loading Linux...." comes up, the screen then jumps a tiny bit to the left and back, the "d" charcater with the cursor underneath appears at the top left of the screen and it just hangs there. I know of at least one other person who has this exact problem. Complementary information: *************************** Running 1.3.32 at the moment, a.out system, slackware 2.3, gcc 2.6.3, never has such a serious kernel problem with all the 1.3.x kernel I have tried. Note: I have used _exactly_ the same options in both 1.2.13, 1.3.32 and 1.3.43 and only the first two work and without a glitch. Here is my .config file: # # Automatically generated make config: don't edit # # # General setup # # CONFIG_MATH_EMULATION is not set CONFIG_NET=y # CONFIG_MAX_16M is not set CONFIG_PCI=y CONFIG_PCI_OPTIMIZE=y CONFIG_SYSVIPC=y CONFIG_BINFMT_ELF=y # CONFIG_KERNEL_ELF is not set CONFIG_M486=y # # Loadable module support # # CONFIG_MODVERSIONS is not set # # block devices # CONFIG_BLK_DEV_FD=y CONFIG_ST506=y # # Please see drivers/block/README.ide for help/info on IDE drives # # CONFIG_BLK_DEV_HD is not set CONFIG_BLK_DEV_IDE=y CONFIG_BLK_DEV_IDECD=y # CONFIG_BLK_DEV_TRITON is not set # CONFIG_BLK_DEV_XD is not set # # Networking options # # CONFIG_FIREWALL is not set CONFIG_INET=y CONFIG_IP_FORWARD=y # CONFIG_IP_MULTICAST is not set # CONFIG_IP_ACCT is not set # CONFIG_NET_IPIP is not set # # (it is safe to leave these untouched) # # CONFIG_INET_PCTCP is not set # CONFIG_INET_RARP is not set # CONFIG_NO_PATH_MTU_DISCOVERY is not set # CONFIG_TCP_NAGLE_OFF is not set CONFIG_IP_NOSR=y CONFIG_SKB_LARGE=y # CONFIG_IPX is not set # CONFIG_ATALK is not set # CONFIG_AX25 is not set # CONFIG_NETLINK is not set # # SCSI support # CONFIG_SCSI=y # # SCSI support type (disk, tape, CDrom) # CONFIG_BLK_DEV_SD=y CONFIG_CHR_DEV_ST=y CONFIG_BLK_DEV_SR=y CONFIG_CHR_DEV_SG=y # # Some SCSI devices (e.g. CD jukebox) support multiple LUNs # # CONFIG_SCSI_MULTI_LUN is not set # CONFIG_SCSI_CONSTANTS is not set # # SCSI low-level drivers # # CONFIG_SCSI_AHA152X is not set # CONFIG_SCSI_AHA1542 is not set # CONFIG_SCSI_AHA1740 is not set CONFIG_SCSI_AIC7XXX=y CONFIG_SCSI_BUSLOGIC=y # CONFIG_SCSI_EATA_DMA is not set # CONFIG_SCSI_EATA_PIO is not set # CONFIG_SCSI_U14_34F is not set # CONFIG_SCSI_FUTURE_DOMAIN is not set # CONFIG_SCSI_GENERIC_NCR5380 is not set # CONFIG_SCSI_NCR53C7xx is not set # CONFIG_SCSI_IN2000 is not set # CONFIG_SCSI_PAS16 is not set # CONFIG_SCSI_QLOGIC is not set # CONFIG_SCSI_SEAGATE is not set # CONFIG_SCSI_T128 is not set # CONFIG_SCSI_ULTRASTOR is not set # CONFIG_SCSI_7000FASST is not set # CONFIG_SCSI_EATA is not set # CONFIG_SCSI_NCR53C406A is not set # # Network device support # CONFIG_NETDEVICES=y CONFIG_DUMMY=y CONFIG_SLIP=y CONFIG_SLIP_COMPRESSED=y CONFIG_PPP=y # # CCP compressors for PPP are only built as modules. # # CONFIG_SCC is not set CONFIG_PLIP=y # CONFIG_EQUALIZER is not set # CONFIG_NET_ALPHA is not set # CONFIG_NET_VENDOR_SMC is not set # CONFIG_LANCE is not set # CONFIG_NET_VENDOR_3COM is not set # CONFIG_NET_ISA is not set # CONFIG_NET_EISA is not set # CONFIG_NET_POCKET is not set # CONFIG_TR is not set # CONFIG_ARCNET is not set # # CD-ROM drivers (not for SCSI or IDE/ATAPI drives) # # CONFIG_CD_NO_IDESCSI is not set # # Filesystems # CONFIG_MINIX_FS=y # CONFIG_EXT_FS is not set CONFIG_EXT2_FS=y # CONFIG_XIA_FS is not set CONFIG_MSDOS_FS=y CONFIG_UMSDOS_FS=y CONFIG_PROC_FS=y CONFIG_NFS_FS=y # CONFIG_ROOT_NFS is not set CONFIG_ISO9660_FS=y CONFIG_HPFS_FS=y # CONFIG_SYSV_FS is not set CONFIG_SMB_FS=y # # character devices # # CONFIG_CYCLADES is not set # CONFIG_STALDRV is not set CONFIG_PRINTER=y # CONFIG_BUSMOUSE is not set CONFIG_PSMOUSE=y # CONFIG_82C710_MOUSE is not set # CONFIG_MS_BUSMOUSE is not set # CONFIG_ATIXL_BUSMOUSE is not set # CONFIG_QIC02_TAPE is not set # # Sound # CONFIG_SOUND=y # # Kernel hacking # # CONFIG_PROFILE is not set ------------------------------ From: devmorfo@patmos.interport.com (Dimitris Evmorfopoulos) Subject: Re: Building Linux from scratch....?? Date: 8 Dec 1995 14:35:55 GMT Rob Janssen (rob@pe1chl.ampr.org) wrote: : In <4a1ecc$pou@news.xmission.com> proteus@xmission.com (The Proteus) writes: : > Hello all... I was wondering just how rigorous building a Linux : >distribution from scratch would be... what steps to be taken, and if it : >is at all feasible. TIA! : It depends on which distribution you use. Some distributions come with : a shell script or makefile that you just have to run to re-compile and : install everything from the sources. : Of course, all the sources are included on the CD. : Normally this will take about a day of compilation time on a high-end : system... : Rob It seems to me that you missed the point here... He does not want to re-build an existing distribution, he wants to build one from scratch.... Having done that my self 4 - 1/2 years ago, I can tell you that it is painfull. You are better off using a very basic distribution first, and then start building on top of it. Of course you would need a fast internet access so that you can download all these xx MB's of source code that you would have to compile... I have to admit though, that it is fun trying to solve all these little problems you encounter every now and then.... ;) -- ______ _______ | __ \ | _____| devmorfo@mail.hol.gr Dimitris Evmorfopoulos | | \ | | |___ +(30) 1 639-1553 | | | | | ___| | |___| | | |_____ |_______| * |_______| * USER ERROR. Replace user and strike any key. ------------------------------ From: devmorfo@patmos.interport.com (Dimitris Evmorfopoulos) Crossposted-To: comp.os.linux.misc,comp.os.linux.hardware Subject: Re: System does not reboot Date: 8 Dec 1995 14:42:47 GMT Richard S Thomas (rich@) wrote: : In article , : Shawn Ruttledge wrote: : >smurray@earthlight.co.nz wrote: : > : >: But why? I have my linux machine at (M$ banned zone) home and it hangs on : >: reboot (IBM blue lighting 2/66) where as the linux box at work (Intel) is : >: fine. Is this just because of the CPU? : > : >My AMD DX4-100 Award bios system does this too, whereas my old 386-33 : >AMI bios system reboots fine. I would like to fix the problem. : I hate to do this but I have to say "me too". With me it's a blue lightning : dx2-66 and I cant remember the bios but I think it may be award (Has some : daft windows-look interface) : If anyone knows a fix for this, I'd be glad to know. : Rich I used to have the same problem on one of my boards, an ASUS SP3G with an AMD DX4-100. It turned out that I had set the processor jumpers for an Enhanced SL type of processor although the AMD that I have is not.. Changing this to its correct setting solved the problem... All three of you might have a similar problem... Check out the MB's manual and CPU's type and abilities... :) -- ______ _______ | __ \ | _____| devmorfo@mail.hol.gr Dimitris Evmorfopoulos | | \ | | |___ +(30) 1 639-1553 | | | | | ___| | |___| | | |_____ |_______| * |_______| * USER ERROR. Replace user and strike any key. ------------------------------ From: boutell@Linux.boutell.com (Thomas Boutell) Subject: ldconfig, getcwd() hosed with ELF -- help! Date: 8 Dec 1995 21:42:50 GMT Hello, I have two problems with my ELF system which are so bizarre that I suspect only the kernel development/system development crowd will have any idea what is going on. I have looked extensively for any reference to similar problems in the FAQs and HOWTOs, and on the linux irc channels. Please give this message consideration if you know of these mysteries. PROBLEM ONE: I have a functioning ELF system with libc 5.2.16. All my ELF binaries work happily, with one exception: The getcwd() call is broken on my machine for ELF binaries. Strange but true. getcwd() always fails (returns null and sets errno to EFAULT) *unless* the current working directory is /. Permissions seem irrelevant, and even as root I see the problem. I don't expect that all libc 5.2.16 users have this problem, because I have deduced it is the cause of my problems with Java, a set of problems no one else is experiencing (insane things like requests for files in irrational directories -- just what ignoring the result of getcwd() would be very likely to do). Other programs, like ghostview, politely inform me that the current working directory cannot be obtained and exit nicely. Can anyone suggest a course of action? I need libc 5.2.16 in order to run the Linux Java DK, which is essential for me right now. A test program which demonstrates the bug on my system (likely not on yours?) is appended at the end of this message. PROBLEM TWO: No version of ldconfig after 1.7.9 will actually do its job and rebuild ld.so.cache on my system. Versions 1.7.11 and 1.7.12 refuse to build a new ld.so.cache file if it is removed, for instance. Versions 1.7.9 and earlier happily build an ld.so.cache file. TEST PROGRAM FOR PROBLEM ONE (GETCWD()): Short test program follows: #include #include extern int errno; int main(void) { char s[512]; if (!getcwd(s, sizeof(s))) { fprintf(stderr, "Can't get current working directory: %d\n", errno); exit(1); } printf("%s\n", s); } ------------------------------ From: evans@choctaw (William Evans) Subject: Re: HELP! Need restricted shell for Linux!!!!!!!! Date: 08 Dec 1995 11:32:18 GMT In article <4a3van$7hp@renux.freenix.fr> rene@renux.freenix.fr (Rene COUGNENC) writes: Ce brave Anthony J. Stuckey ecrit: > BASH should have a restricted mode, I don't think that pdksh does. Bash has a restricted mode, it is a config-time option : From config.h : /* Define RESTRICTED_SHELL if you want the generated shell to have the ability to be a restricted one. The shell thus generated can become restricted by being run with the name "rbash", or by setting the -r flag. */ /* #define RESTRICTED_SHELL */ I've been looking for an adequate restricted shell, as well, and I've realized that the type of restriction I want is quite hard to come by. Yes, I've read the documentation about rbash. I've even compiled it and tried it, but despite the inability to cd to any directory, I can still less/edit any file that I would normally have read access to (like /etc/passwd, ...). This is exactly what I want to avoid -- being able to have standard rights. I found a shell that really intrigued me, called `osh'. It's designed by Mike Neuman (mcn@EnGarde.com) among a strong security push. It can be found at ftp.c3.lanl.gov:/pub/mcn/osh.tar.gz, along with an accompanying paper, ./osh.sansIII.ps or something like that. I got it to compile (with a few minor changes), but the restriction is not how it is detailed within the paper. Quick summary: nothing can be done unless it is _explicitly_ allowed in a world-unreadable file. This includes which executables to run and exactly what files/directories the user can see/write/etc. I've sent mail to him in an attempt to pursue a linux port, but I have yet to hear back from him (I'll try again today). If anybody else has played with this and actually gotten it to work as detailed, I would greatly appreciate some advice in compiling. Bill -- __ ____________ / /\ /_________ /\ / / \ William Evans \ ______/ / / / / /\ \ evans@cs.tulane.edu \ \ \ / / / / / /\ \ \ Computer Engineering \ \ \/ / / / /_/__\ \ \ Tulane University \ \ \/ / /________\ \ \ New Orleans, Louisiana \ \ / \___________\/ http://www.cs.tulane.edu/~evans \_\/ ------------------------------ ** 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 ******************************