From: Digestifier To: Linux-Development@senator-bedfellow.mit.edu Reply-To: Linux-Development@senator-bedfellow.mit.edu Date: Thu, 1 Sep 94 11:13:12 EDT Subject: Linux-Development Digest #100 Linux-Development Digest #100, Volume #2 Thu, 1 Sep 94 11:13:12 EDT Contents: NeXT on Linux (Steve K Hunter) (none) (Hitzke) MATROX PCI Graphics board supported ?? (Harald von Fellenberg +41-1-801-3578 DTN 760-3578) PCM 1 not installed - Boot Message (C. Engelmann) compiling the kernel (Jonathan Booth) Re: Unix, Unicode, and internationalization (Yasuo Ohgaki) Re: __NEEDS_SHRLIB_libm_4 (Charles Lopes) Re: Kernel change summary 1.1.45 -> 1.1.46 (Remy CARD) lossage with "tar cz" writing to gzip; easy fix? (Chris Metcalf) Re: Does really Linux uses RAM efficiently? Undelete ability? (Alan Cox) AAAAAH - Where Linux.1.1.49 (Robert Mudge) Re: Load-time configurable kernel? (Vassili Leonov) Re: Load-time configurable kernel? (H.J. Lu) Re: Linux console to SCO comp. prob (Harald T. Alvestrand) Re: LSMI CDrom (Steve Smith, Logica UK) Anyone working on ISDN card drivers ?? Re: IDE write bug (Mark Lord) Kernel change summary 1.1.47 -> 1.1.48 (Russell Nelson) Re: [Q] How to use SCSI Generic Driver ? (Drew Eckhardt) Re: Future of Linux (Matt Warnock) Re: Threads for Linux (Steven Buytaert) ---------------------------------------------------------------------------- From: shunter@cais2.cais.com (Steve K Hunter) Subject: NeXT on Linux Date: 31 Aug 1994 22:54:35 GMT Does anybody know if the portable version of the NeXT environment is being ported to linux (i.e. OpenStep for Linux) ? thanks shunter@cais.com ------------------------------ From: Hitzke Subject: (none) Date: 1 Sep 1994 07:50:27 -0400 Reply-To: hitzke@blitz.chemie.uni-ulm.de subscribe ------------------------------ From: harald@zuunix.zuo.dec.com (Harald von Fellenberg +41-1-801-3578 DTN 760-3578) Subject: MATROX PCI Graphics board supported ?? Date: 1 Sep 1994 11:37:09 GMT Reply-To: harald@zuo.dec.com (Harald von Fellenberg +41-1-801-3578 DTN 760-3578) Is anybody working on the MATROX PCI graphics board support? Or how can it be configured using default parameters? I have not found any indication about the board in XFree86-2.1.1... thanks! Harald von Fellenberg Senior Technology Consultant DEC Digital Equipment Corp. AG CH-8600 Dubendorf, Switzerland hvf@zuo.dec.com "Felix qui potuit rerum cognoscere causas" ------------------------------ From: engel@yacc.central.de (C. Engelmann) Subject: PCM 1 not installed - Boot Message Date: Tue, 30 Aug 1994 18:32:39 GMT Today I installed kernel1.1.48 with PAS16-sound-driver and got the following message with DMESG: "PCM device 1 not installed." The sound card seems to work fine. Thank you, Carsten ------------------------------ From: jmxb@dcs.ed.ac.uk (Jonathan Booth) Subject: compiling the kernel Date: Thu, 1 Sep 1994 10:35:28 GMT I am having trouble compiling kernel versions 1.1.35 and 1.1.13. I have a slackware instalation with gcc version 2.5.8. Problem trying to compile either of the kernels produces the following error message. Before here everything seemed ok, no error messages or warnings, then make -C zBoot make[1]: Entering directory `/usr/src/linux/zBoot' gcc -D__KERNEL__ -I/usr/src/linux/include -O2 -DSTDC_HEADERS xtract.c -o xtract In file included from xtract.c:12: /usr/include/stdlib.h:293: parse error before `*' /usr/include/stdlib.h:296: parse error before `wchar_t' /usr/include/stdlib.h:304: parse error before `*' /usr/include/stdlib.h:306: parse error before `*' make[1]: *** [xtract] Error 1 make[1]: Leaving directory `/usr/src/linux/zBoot' make: *** [zBoot/zSystem] Error 1 Any ideas whats going on? Any ideas how to fix it? Please reply using Email thanks Jonathan Booth jmxb@dcs.ed.ac.uk jmxb@tardis.ed.ac.uk ------------------------------ From: yohgaki@mercury.cair.du.edu (Yasuo Ohgaki) Subject: Re: Unix, Unicode, and internationalization Date: Wed, 31 Aug 1994 16:08:37 GMT Alan Cox (iialan@iifeak.swan.ac.uk) wrote: : In article <1994Aug25.212924.3202@midway.uchicago.edu> goer@midway.uchicago.edu writes: : >Unfortunately, it appears that NT, and to some extent Windows (and : >of course the Mac's Word Script), are way, way ahead of anything we find : >in the Unix world - at least when it comes to internationalization. : Not really. There is a lot less publicity for the fact Unix supports : multiple language catalogs (and Linux does this now). But you'll find plenty : of tools for Unix that are multilingual or foreign language - MULE, kterm etc... : Alan : -- : ..-----------,,----------------------------,,----------------------------,, : // Alan Cox // iialan@www.linux.org.uk // GW4PTS@GB7SWN.#45.GBR.EU // : ``----------'`----------------------------'`----------------------------'' As a Japanese user, I really want Unicode to be a world standard. I really disapponted that Win 4.0 dosen't support Unicode like NT.... If win 4.0 supports unicode, then it will be ..... Anyway, multiple language support is not enough. -- Yasuo Ohgaki e-mail: yohgaki@mercury.cair.du.edu ------------------------------ From: tjarls@petrel.infm.ulst.ac.uk (Charles Lopes) Subject: Re: __NEEDS_SHRLIB_libm_4 Date: 31 Aug 1994 18:39:43 GMT In article , skj@oasis.icl.co.uk (Simon Johnston) writes: |> This must be a FAQ, but why is it, now I have moved to slackware 2.0, I |> start getting the __NEEDS_SHRLIB_libm_4 message when trying to link |> anything ? (also libc_4 etc). |> |> Thanks. |> |> |> MODULE Sig; |> FROM ICL IMPORT StdDisclaimer; |> FROM Interests IMPORT Modula2, Modula3, Linux, OS2; |> |> BEGIN |> (* ------------------------------------------------------------------------. |> |Simon K. Johnston - Development Engineer |ICL Retail Systems | |> |------------------------------------------------------|3/4 Willoughby Road| |> |Unix Mail : S.K.Johnston.bra0801@oasis.icl.co.uk |Bracknell, Berks | |> |Telephone : +44 (0)344 476320 Fax: +44 (0)344 476084|United Kingdom | |> |Internal : 7261 6320 OP Mail: S.K.Johnston@BRA0801|RG12 8TJ | |> `------------------------------------------------------------------------ *) |> END Sig. |> You probably don't have the math library stubs (libm.sa) and linked the static library in instead. Try getting image-4.5.26.tar.gz and installing it. -- ------------------ Charles Lopes | Internet: Charles.Lopes@infm.ulst.ac.uk Computing Officer | Phone: +44 (0504) 265621 Faculty of Informatics | Phone extension: 5315 Magee College, University of Ulster | Office: MB023 ------------------------------ From: card@masi.ibp.fr (Remy CARD) Subject: Re: Kernel change summary 1.1.45 -> 1.1.46 Date: 1 Sep 1994 12:33:14 GMT In article <3434oi$4os@clarknet.clark.net>, Marc Fraioli wrote: ] In article 778359874@cs.cornell.edu, nelson@crynwr.crynwr.com (Russell Nelson) writes: ] > - New file attributes: ] > - Immutable files cannot be modified. Data cannot be written to ] > these files. They cannot be removed, renamed and new links cannot ] > be created. Even root cannot modify the files. He has to remove ] > the immutable attribute first. ] > - Append-only files: can only be written in append-mode when writing. ] > They cannot be removed, renamed and new links cannot be created. ] > Note: files may only be added to an append-only directory. ] ] Very cool. Do these work the same way as in BSD 4.4? Mostly, yes. Actually, they were inspired by 4.4BSD flags. ] How can we access these new features? The next e2fsprogs release (version 0.5b) contains support for these new attributes (in the programs chattr and lsattr). As of Linux 1.1.46, the kernel honours them. ] I seem to remember reading that under BSD the immutable ] files can only be modified in single-user mode. Is it necessary to be in ] single-user to remove the immutable attribute as mentioned above? I think ] this would probably be a good idea. Thoughts? In 4.4BSD, immutable and append-only flags can only be set and cleared in single-user mode. It should be easy to make it work the same way in Linux: add a system call which specifies the run-level and modify init to call it when going into single user mode or multi user mode. The problem is that Linux has many versions of init floatting around and such a solution would require that everybody upgrades his init program. In the current Linux implementation, only root can change the immutable and append-only attribute on a file, in single user mode and in multi user mode. ] --- ] Marc Fraioli | "They couldn't hit an elephant at this dist- " ] mjf@clark.net | - Last words of Union General John Sedgwick, ] | Battle of Spotsylvania Court House, U.S. Civil War Remy ------------------------------ From: metcalf@CATFISH.LCS.MIT.EDU (Chris Metcalf) Subject: lossage with "tar cz" writing to gzip; easy fix? Date: 1 Sep 1994 12:17:17 GMT Does anyone have a patch to fix the problem with "tar cfz" where if you suspend and background the process, it dies with a wrong-size return from write()? I'm surprised to see such behavior in Linux, which is pretty well-standardized most of the time. (Clearly the write to the pipe is returning after an atomic-sized write on an interrupt, but it would be nice to suppress that behavior.) This is with tar 1.11.2 and gzip 1.2.4 (and libc 4.5.26 with Linux 1.1.49, but it's been going on a long time); a typical error message would be "tar: only wrote 2048 of 10240 bytes to foo.tgz". Email me with any responses and I will post a summary. Thanks in advance, -- Chris Metcalf, MIT Laboratory for Computer Science metcalf@cag.lcs.mit.edu // +1 (617) 253-7766 ------------------------------ From: iialan@iifeak.swan.ac.uk (Alan Cox) Subject: Re: Does really Linux uses RAM efficiently? Undelete ability? Date: Thu, 1 Sep 1994 11:44:19 GMT In article quinlan@yggdrasil.com writes: >Linux has memory management which is on par with most commercial >operating systems. I know of none with better management, in fact. >(Then, again I can only quote a few names that have significantly >worse memory management.) The Linux memory manager has two problems, both obscure and both present in every other system I've looked at so far. You can't dynamically allocate definitely DMA'able memory, and you can't allocate memory dynamically with a preference for DMA'able memory (to lower bounce buffer usage). Since you have to have a >16Mb ISA bus machine with a bus mastering card for this to matter its not that important.. just a small irritation in the perfection 8) Alan -- ..-----------,,----------------------------,,----------------------------,, // Alan Cox // iialan@www.linux.org.uk // GW4PTS@GB7SWN.#45.GBR.EU // ``----------'`----------------------------'`----------------------------'' ------------------------------ From: mudge@sunny.dab.ge.com (Robert Mudge) Subject: AAAAAH - Where Linux.1.1.49 Date: 1 Sep 1994 11:56:11 GMT Reply-To: mudge@sunny.dab.ge.com I was referenced to linux 1.1.49, can't find it on sunsite... can anyone tell me where it is? AdTHANKSvance Robert email... , mmcmcor@extro.ucc.su.OZ.AU (Mike McCormack) writes: : |> One of the good features of MessDOS is that you can easily reconfigure : |> your system by changing config.sys and rebooting. Doing this can add new : |> device drivers,etc. Why can't linux be made to be configured like that, : |> instead of going to all the trouble of recompiling the kernel? : |> Sorry if I'm just professing my ignorance :-) : Yes, you are. Take a look at the loadable kernel modules. It's always a pleasure to read something by people that know and are confident... :-) I would though say that modules in Linus have a POTENTIAL for having kernel configuration done easily and consistently at the runtime. But so far... Actually all the Unix setup is a terrible mess. Zillions of files in the etc (souldn't they be in var?), being mostly a shell scripts. If you can come up with some sort of .config.sys (name it .unixr :-) containing MOST of the tunable things in some kind of human readable form like: HOSTNAME=darkstart HOSTS=/etc/hosts LOADMODULE ft /lib/mods/floppy_tape.o etc. etc. which is parsed at startup by some rc.readconf file which does all the checks possibled and enforces a certain standards. And of course major and minor numbers should be determined and bound to entries in /dev directory automatically - /dev should be done in the same way as /proc - i.e. show something from inside the kernel. X configuration should also be done in a simular way - though it's much more closer to that now. Somebody was writing about including turnkey X into the distribution. It's only a metter of starting ConfigXF86 after installing X files - in my experience if you answer that questions more or less proper - the X will somehow run after you say startx. Configuration of Unix is unnecessary pain - most commercial brands are mostly concentrating on aiding that. But what need to be done is some redesign of the very ide how it's done - then it would be a real progress. Vassili. ------------------------------ From: hjl@nynexst.com (H.J. Lu) Subject: Re: Load-time configurable kernel? Date: 31 Aug 1994 16:55:58 GMT In article <34188m$h21@news.ysu.edu>, s0017210@cc.ysu.edu (Steve DuChene) writes: |> H.J. Lu (hjl@nynexst.com) wrote: |> : In article , mmcmcor@extro.ucc.su.OZ.AU (Mike McCormack) writes: |> : |> One of the good features of MessDOS is that you can easily reconfigure |> : |> your system by changing config.sys and rebooting. Doing this can add new |> : |> device drivers,etc. Why can't linux be made to be configured like that, |> : |> instead of going to all the trouble of recompiling the kernel? |> : |> Sorry if I'm just professing my ignorance :-) |> |> : Yes, you are. Take a look at the loadable kernel modules. |> |> Ok but why aren't they used more? I have been wondering about this for some |> time (ever since the SLS-1.05 stuff came out and failed to make a dent in |> the Slackware popularity). I am not knocking Slackware as that is what I have |> installed but why didn't the large number of loadable modules in the SLS |> distribution work out? I think it on the way. Right now all my networking and PCMCIA stuff are loadable modules. Give it a little more time. It will catch on. H.J. |> -- |> | Steven A. DuChene sduchene@cis.ysu.edu or s0017210@cc.ysu.edu |> | Youngstown State University | Computer Science / Math / Mech. Eng. |> |They all laughed at Albert Einstein. They all laughed at Columbus. |> |Unfortunately, they also all laughed at Bozo the Clown. ------------------------------ From: hta@uninett.no (Harald T. Alvestrand) Subject: Re: Linux console to SCO comp. prob Date: 1 Sep 1994 13:01:06 GMT 1) Where can I get the SCOANSI doc? 2) Are the ÆØÅ æøå characters properly supported? -- Harald Tveit Alvestrand Harald.T.Alvestrand@uninett.no G=Harald;I=T;S=Alvestrand;O=uninett;P=uninett;C=no +47 73 59 70 94 My son's name is Torbjørn. The letter between "j" and "r" is o with a slash. ------------------------------ Date: Thu, 1 Sep 1994 14:09:45 EST From: Steve Smith, Logica UK Subject: Re: LSMI CDrom Whoever posted the bit about having the lms0.3c driver will they please PLEASE send me it uuencoded. Thnaks very much. I've been looking for a working drive r for months!! Steve, STSMITH@ESOC.BITNET ------------------------------ Date: Thu, 1 Sep 1994 14:55:35 +0200 From: nhead@esoc.bitnet () Reply-To: nhead@esoc.bitnet Subject: Anyone working on ISDN card drivers ?? If anyone is working on or knows of ISDN drivers for Linux systems please would you let me know ?? I'm considering moving into the digital world and I need some feel that my PC won't have to revert to running the Devil's OS .... Thank, Nigel. PS: Do you have ISDN in the US of A yet ?? ------------------------------ From: mlord@bnr.ca (Mark Lord) Crossposted-To: alt.comp.hardware.homebuilt,comp.periphs Subject: Re: IDE write bug Date: 1 Sep 1994 13:19:54 GMT In article <343uhh$ms2@usenet.rpi.edu> wilsonj@alum01.its.rpi.edu writes: >Does anyone know the specifics of the alleged bug in some IDE drives >where if you get an interrupt while writing data to the sector buffer, I know a fair bit about it. A very tiny percentage of IDE drives, mostly very early ones, have this problem (earlier models of Seagate, Quantum, Conner brands). The problem is not "if you get an interrupt", but rather "if you experience any delay/latency during transfering of a sector of data". To prevent it, disable interrupts for the duration of the sector transfer. The vast majority of IDE drives do *not* have this problem, and that IDE spec (AT Attachment Interface..) says nothing about it or about any such weird timing requirements. -- mlord@bnr.ca Mark Lord BNR Ottawa,Canada 613-763-7482 ------------------------------ From: nelson@crynwr.com (Russell Nelson) Crossposted-To: comp.os.linux.announce Subject: Kernel change summary 1.1.47 -> 1.1.48 Date: 01 Sep 1994 02:43:37 GMT Rename sys_call.s to entry.s and put it in arch/i386. Create the entry.S file from the appropriate architechture entry.S. Add Ethernet drivers for ni5210 and ni6510 to alpha test list. Added /proc/dma to report names of drivers that use DMA resources. Change floppy parameters for 800KB, 1.04MB and 1.12MB formats. Add better reporting for the hard disk driver. Reset drives in case we changed the geometry. Make multiple mode and irq-unmasking have default, compiled-in options. Check for disk change before loading ramdisk. Didn't clear all of the Ethenret device structure's memory. SCSI CD-ROM detection notice was printing a number one too high. open/fcntl - O_SYNC is now implemented on blocks devices and on files located on an ext2 file system Added a revision level in the superblock of the ext2 filesystem (starting at revision zero). Oops, caught a sillyness in memset for constant pattern and count. Not a bug, just a waste of time/space. Removed ncp additions. -- -russ http://www.crynwr.com/crynwr/nelson.html Crynwr Software | Crynwr Software sells packet driver support | ask4 PGP key 11 Grant St. | +1 315 268 1925 (9201 FAX) | What is thee doing about it? Potsdam, NY 13676 | LPF member - ask me about the harm software patents do. ------------------------------ From: drew@frisbee.cs.Colorado.EDU (Drew Eckhardt) Subject: Re: [Q] How to use SCSI Generic Driver ? Date: 31 Aug 1994 17:14:00 GMT In article <1994Aug30.133351.24651@rz.uni-hildesheim.de>, Holger Laebe wrote: >i'd like to write a program which talk to an SCSI >Scanner. the brand of the scanner is not known yet >but it is assumed that all documentation is included. > >so my question : > - how can i send/receive data to a known SCSI Id Currently, there is no mechanism to send data to a given SCSI bus/target/lun (it's mostly an anachronism from having a 16 bit dev_t and 8 bit minor numbers). However, it is possible to send arbitrary non-vendor specific commands to any supported (disk, tape, cd, worm) or unsupported (with a change to force detection of the device in scsi.c) devices. > - is this captured by the Generic SCSI Device Yes. -- "The most foolish mistake we could possibly make would be to allow the subject races to possess arms. History shows that all conquerors who have allowed their subject races to carry arms have prepared their own downfall by doing so." -- Adolf Hitler ------------------------------ From: mwarnock@garlic.com (Matt Warnock) Subject: Re: Future of Linux Date: 31 Aug 1994 21:45:55 -0700 In article <3426mm$o8q@bigblue.oit.unc.edu>, Erik Troan wrote: >We could easily put together either a 16 color 640x480 60 Hz server that >would work on most any VGA card. Why hasn't anyone done this? It's SLOW >and UGLY. Okay, but I have a setup like most new Linux users: 14' monitor, 1 meg card. 640x480 is pleasant but not enough real estate. 800x600 is too small to work at for extended periods. 1024x768 is masochistic. So what do I do? I run charmode, that's what. >Is that really a problem? Yes. If we hand users a working X setup they won't >change it for a long time. Linux/X will then suffer from criticism over >the speed of the windowing system, lack of screen real estate, and horrible >refresh rates. Is it really better to give people a half-assed system >or let them take the time to get it working right? While the second >approach is more difficult and time consuming for the user the result >is far superior. I disagree. Let them get a taste of a working system first. I would rather have an inferior solution NOW than a better solution but only if I can grunt it out. Especially since in Unix/Linux there is already so MUCH to learn. I have ashcanned DOS for Linux, and I am now up to about 50-60% of DOS productivity. that's OK, cause I was always at the limits with DOS. But there is a very real initial frustration with having to roll your own everything from the ground up. I couldn't even write a reasonable letter without learning TeX. Sorry, but folks with less patience wouldn't touch it. -- W. Matthew Warnock, Attorney (mwarnock@garlic.com) Tel:408.778.7273 60 West Main Avenue, Suite 12A, Morgan Hill CA 95037-4553 Fax:408.778.7989 ------------------------------ From: buytaert@imec.be (Steven Buytaert) Subject: Re: Threads for Linux Date: Thu, 1 Sep 1994 07:54:35 GMT Nicolas Droux (droux@info.isbiel.ch) wrote: : I need to be able to create threads, or lightweight processes, under : Linux. Is there a good solution to this problem ? Is this feature : integrated in a recent release of the kernel, is it planned, or is it : possible to do this by means of a 3rd party package ? Use archie to locate 'pthread' (use the substring matching). There is a library, rather extensive, that supports Linux with threads, mutexes and reentrant replacements for several functions. It's only drawback is that it isn't documented (yet). You should get a hold on the posix draft 1003.4a (I think) as documentation. I couldn't find it on the net. Recently, the drafts aren't available anymore via ftp. If someone could point me to this draft, please... Stef -- Steven Buytaert WORK buytaert@imec.be HOME buytaert@innet.be 'Imagination is more important than knowledge.' (A. Einstein) ------------------------------ ** FOR YOUR REFERENCE ** The service address, to which questions about the list itself and requests to be added to or deleted from it should be directed, is: Internet: Linux-Development-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux.development) via: Internet: Linux-Development@NEWS-DIGESTS.MIT.EDU Linux may be obtained via one of these FTP sites: nic.funet.fi pub/OS/Linux tsx-11.mit.edu pub/linux sunsite.unc.edu pub/Linux End of Linux-Development Digest ****************************** .