From: Digestifier To: Linux-Development@senator-bedfellow.mit.edu Reply-To: Linux-Development@senator-bedfellow.mit.edu Date: Sat, 15 Oct 94 03:13:10 EDT Subject: Linux-Development Digest #310 Linux-Development Digest #310, Volume #2 Sat, 15 Oct 94 03:13:10 EDT Contents: Re: 3Com 509 Driver Problems - Any fixes - Help (Chris Stoffel) insmod drv_hello.o gives "___moddi3 undefined" (Erik E. Rantapaa) Re: Extreme delays telnetting into linux box (Mark Lord) NFS mounts from linux ? (David Martin) Re: Linux NOT logging people out on hangup (Piet W. Plomp) Re: Just a quick question... (Jeff Kesselman) Re: A badly missed feature in gcc (Bruce Thompson) Re: 1.1.52 bug? (Bruce Thompson) Re: patch in the kernel (Mark Lord) Re: ext2fs vs. Berkeley FFS (Erik Corry) Re: Telnet & ftp freeze! - AND UNFREEZE KLUDGE (Steve Davies) Motif+Linux... broken! (Douglas Frank) ext2, UFS, Linux, and Mach (travis jensen) Cmix,RT -> Linux,DOS (David J Topper) Re: Shared Libs: working toward a permanent solution? (Brandon S. Allbery) Re: We a FAQ: Linux vs. *BSD!!! (Andreas Helke) Re: [fdformat] kernel 1.1.52 (and .53) patch comment (Brent R. Earl) ---------------------------------------------------------------------------- From: cstoffel@memhpc13.sc.ti.com (Chris Stoffel) Crossposted-To: comp.os.linux.help Subject: Re: 3Com 509 Driver Problems - Any fixes - Help Date: 14 Oct 1994 14:30:33 GMT I just read in comp.os.linux.announce that the latest kernel patch has some 3c509 fixes. Maybe it's time for an overhaul. Chris Stoffel e-mail: cstoffel@lobby.ti.com Brian Kramer (bjkramer@pluto.njcc.com) wrote: : I get the following error which pretty much disables my system. Is there : a fix? Or can someone recommend a ethernet card that works flawlessly : with linux? : Sep 27 20:11:56 pluto kernel: eth0: Missed interrupt, status then 2011 now 2000 Tx 00 Rx 8000. : Sep 27 21:56:01 pluto kernel: eth0: Transmitter access conflict. : Sep 27 22:07:24 pluto kernel: eth0: transmit timed out, tx_status 00 status 2000. : Sep 27 22:07:25 pluto kernel: eth0: transmit timed out, tx_status 00 status 2000. : Sep 27 22:33:54 pluto kernel: eth0: Missed interrupt, status then 2011 now 2000 Tx 00 Rx 8000. : Sep 28 01:10:52 pluto kernel: eth0: Missed interrupt, status then 2011 now 2000 Tx 00 Rx 8000. : Sep 28 12:32:12 pluto kernel: eth0: Missed interrupt, status then 2011 now 2000 Tx 00 Rx 8000. : Sep 28 15:39:43 pluto kernel: eth0: Missed interrupt, status then 2011 now 2000 Tx 00 Rx 8000. : -- : Brian Kramer - Owner/Systems Administrator - bjkramer@pluto.njcc.com : New Jersey Computer Connection - Public Access Unix Site - pluto.njcc.com : Voice: 609-896-2799 - Fax: 609-896-2994 - Dialups: 609-896-3191 : Dialup or Telnet to pluto.njcc.com and log in as guest for more information. ------------------------------ From: rantapaa@s6.math.umn.edu (Erik E. Rantapaa) Subject: insmod drv_hello.o gives "___moddi3 undefined" Date: Thu, 13 Oct 1994 01:19:36 GMT I am trying to use the module utilities with the following set up: kernel version: 1.1.51 cc version: 2.5.8 modules.tar.gz: dated June 27, 1994 When attempting to load the sample module dvr_hello.o, however, I get the following error: ___moddi3 undefined Is there something special I need to do when compiling module object files? Here is a list of the symbols in drv_hello.o: zorro /nfs/zorro/x4/build/modules-new: nm drv_hello.o 00000000 t ___gnu_compiled_c U ___moddi3 00000220 T _cleanup_module 000000e0 t _hw_close 00000294 d _hw_fops 00000060 t _hw_lseek 000000d0 t _hw_open 00000000 t _hw_read 00000287 d _hw_reply 00000150 T _init_module 00000280 d _kernel_version U _mod_use_count_ U _printk U _register_chrdev U _unregister_chrdev 00000000 t gcc2_compiled. ------------------------------ From: mlord@bnr.ca (Mark Lord) Crossposted-To: comp.os.linux.admin Subject: Re: Extreme delays telnetting into linux box Date: 15 Oct 1994 01:09:36 GMT In article bart@dunedin.es.co.nz writes: ... <>problems may be related. Running 1.1.49 on a Slackware setup. < I had a run-in with the comp.os.linux.development people over this > several months ago. > > The general consensus is: > Linux strives for POSIX compliance. > POSIX states that the kernel sends HUP signals only to session leaders. > POSIX states that session leaders should relay SIGHUP to children. > Many session leaders (shells, usually) don't follow the POSIX spec. > Lots of processes wind up chewing up CPU time. > > So lots of multi-user admins wind up working around this POSIX "feature" > by writing daemons to kill the processes that Linux doesn't. > > If it's any consolation, I've seen a large-scale commercial provider > running AIX grind to a halt because the OS is POSIX compliant and the > shells or applications [kermit, ytalk, and lynx were the major problems] > aren't. They've got a daemon to work around POSIX now, too. Ok., I got myself a brandnew tcsh 6.0.5. (from sunsite:system/Shells). The docs that come with it don't mention this POSIX compliance problem. The (beautiful) man page says that TERM signals are caught. Well, they are ignored. Both for login tcsh's as for non-login tcsh's kill -15's don't have any effect. For hangup signals (kill -1) the situation is different: they are ignored for login tcsh's, but effectively kill the tcsh on non-login tcsh's. The only way to kill a login tcsh is through SIGKILL (-9). This is true for both the old tcsh (6.0.4) as the new one. Questions: - should the tcsh be considered process group leader? - should it ignore SIGTERM and SIGHUP signals as described above - is there a way to recompile the tcsh, so that it complies with posix and linux signal handling? Btw, kernel 1.0.9, libs 4.5.26. Thanks, piet -- Piet W. Plomp, ICCE, Groningen University, The Netherlands; piet@icce.rug.nl Visit The LINUX SYMPOSIUM: 8th/9th of december 1994 in Amsterdam, Netherlands. SUBSCRIBE NOW: ftp (anonymous) to ftp.icce.rug.nl:pub/symposium/how.to.register ------------------------------ From: jeffpk@netcom.com (Jeff Kesselman) Subject: Re: Just a quick question... Date: Tue, 11 Oct 1994 04:10:25 GMT In article <375afj$rap@acmez.gatech.edu>, Matthew Richard Scopp wrote: > Hello. I have a quick question. I'll apologize in advance if this >is a question that is constantly being asked, but I have just joined this >group this very minute, and it takes far too long to wade through over 1500 >articles looking for what I need. :) I was just wondering what the status >of Linux development is for the PowerPC chip. Specifically, if I go out and >buy myself a PowerMac, will I be able to run Linux on it any time soon, or >will I grow a big grey beard first? I'm just curious as to what kind of >support for PowerMac Linux there is. I don't think I would be able to run >XFree86 :) so I'd have to look elsewhere. They're supposed to be porting AIX >to PowerPC, but AIX is a joke that's not worth laughing at. PowerPCs seem >like good chips to run Linux on, so I'm hopeful about possible support. Thanks >for your time. My e-mail address is gt4374c@prism.gatech.edu. > The buzz I've heard is that Linux for any Mac is a long way off due to the propiatary nature of info on the Mac's hardware (thats just my impression from group reading. (If I am wrong, please correct me, but don't flame, ok?) I have a stoopid question for YOU. WHY buy Mac hardware if you intend to run Linux??? Pound for pound (and $ for $) your price/performance is going to be MUCH better with an intel-type platform. I can understadn buyign a Mac if you want the Mac environment, but if you're just going to install Linux over it, it hardly seems worth it. Myself, I'm waiting for the IBM PPCs, which shoudl be alot cheaper then Power Macs.... JK ------------------------------ From: bruce@mdavcr.mda.ca (Bruce Thompson) Subject: Re: A badly missed feature in gcc Date: 10 Oct 94 07:20:38 GMT Jeff Kesselman (jeffpk@netcom.com) wrote: : In article <373vcn$1da@nz12.rz.uni-karlsruhe.de>, : And this is the cannonical case of why c++ is NOT a true super-set of : ANSI c. (Thanks, I'm going to save the exampel for the next time I have : THAT argument.) : Question, though, if gcc is fed a c++ file, will it accept that (a .C : extension, or one of the other gcc/g++ blessed c++ extensions)? : Seems to me it should. The answer to the poster woudl then be 'Youa re : trying to write c++ code. Stop telling the compiler its C code, and all : will be fine...' : Jeff Kesselman APPLAUSE! It's about time someone said something like this. Please don't take this the wrong way folks, but if you want to write _C_ code, write _C_ (as defined by the ANSI standard). If, on the other hand you want to write _C++_ code, write C++ code (as defined by the ARM) but _PLEASE_ don't complain that C isn't C++! The // comment syntax is not defined to be part of C, therefore gcc should _never_ accept it as a comment. When invoked as g++ though, it's compiling C++ and therefore // is valid syntax for a comment. The fact that gcc is the compiler for C, C++ and even Objective-C (and now Ada too!) is more convenience than anything else. C and C++ and still _distinct_ languages. They have a common history, and C++ is mostly a super-set of C, but that's where the line is drawn. As far as I know, gcc _will_ compile a file called foo.C or foo.cc or foo.cxx etc. as a C++ file. However, if you are linking C++ code you _must_ invoke g++ to do the linking, otherwise the static constructor and destructor lists won't get setup correctly. Cheers, Bruce. -- Bruce Thompson, B.Sc. | "A great many people think they are Software Engineer | thinking when they are merely MacDonald Dettwiler, | rearranging their prejudices." 13800 Commerce Parkway, | -- William James Richmond, BC | (604) 278-3411 | Usual disclaimers apply NAPRA #473 | ------------------------------ From: bruce@mdavcr.mda.ca (Bruce Thompson) Subject: Re: 1.1.52 bug? Date: 10 Oct 94 07:24:45 GMT Riku Saikkonen (riku.saikkonen@compart.fi) wrote: : Hmm... Linux crashes on me, seemingly randomly. Nothing works : after the crash, not virtual console switching, not Ctrl-Alt-Del. : 'Unable to handle kernel paging request' and the normal register dump : appears in the log, for two or three processes. The EIP is in a totally : different location each time... : So, it seems that it's the 1.1.52 kernel which is having some problems. : I noticed some modification of memory.c in the .52 patch; perhaps that's : the culprit? : [...] I haven't had a chance to investigate further, but the bug seems to be present in 1.1.51 too. I'm compiling 1.1.52 as we speak :-) and I've noticed similar behavior to that mentioned above. I can't pinpoint anything that could be causing the problem. It has happened with 'rm', 'atrun' and once with 'inetd'. If I can nail down any additional information I'll post it. Cheers, Bruce. -- Bruce Thompson, B.Sc. | "A great many people think they are Software Engineer | thinking when they are merely MacDonald Dettwiler, | rearranging their prejudices." 13800 Commerce Parkway, | -- William James Richmond, BC | (604) 278-3411 | Usual disclaimers apply NAPRA #473 | ------------------------------ From: mlord@bnr.ca (Mark Lord) Subject: Re: patch in the kernel Date: 15 Oct 1994 01:11:02 GMT In article ncherry@cbnewsg.cb.att.com writes: : Backgammon.app/Backgammon : Backgammon.app/whiteCircle.tiff : Backgammon.app/blackCircle.tiff : Backgammon.app/local-config What nobody has commented on, is that this example shows things being put together that probably shouldn't be together anyway. The TIFFs are architecture independent, the executable is architecture dependent, and the configuration file is probably machine dependent. Given this they shouldn't even be on the same filesystem, never mind packed in the same file/dir. Having got extremely frustrated about VMS's filesystem, where the RMS Record Manangement System stores various meta-information about a file which almost invariably gets left behind when copying between architectures or is set to the wrong thing, and very difficult to alter, it really seems to me that 'a file is an ordered collection of bytes' avoids a lot of trouble. The exception, of course is security-relevant material, which causes lots of trouble, but must be specially handled by the OS. When ext2fs gets Access Control Lists there will probably be a lot of frustration about tar, cp, gzip etc not doing The Right Thing with them. If we introduce resource forks, the problem is only made worse. -- -- Erik Corry, Skagerrakstr. 2, 79100 Freiburg, Germany, +49 761 406637 erik@kroete2.freinet.de ------------------------------ From: steve@iaccess.za (Steve Davies) Crossposted-To: comp.os.linux.help,comp.os.linux.misc Subject: Re: Telnet & ftp freeze! - AND UNFREEZE KLUDGE Date: 14 Oct 1994 23:52:03 +0200 We experience a problem on out Linux 1.1.19 systems where the inetd goes deaf and won't handle any more incoming connections. If you run inetd in debugging mode then as I recall you see that select() starts to return -1 [Dimly remembered] The fix is to kill inetd and start a new one. The cause? I have found that the problem is caused by people connecting with SLIP and using the *wrong IP address* on their end. In other words they have configured their IP stack with an address different from that in the diphosts file. Perhaps that will help someone, Steve ------------------------------ Crossposted-To: alt.uu.comp.os.linux.questions,comp.windows.x.motif From: frankdn@mv.mv.com (Douglas Frank) Subject: Motif+Linux... broken! Date: Thu, 13 Oct 1994 00:33:24 GMT Has anyone successfully installed "Swim," (Motif 1.2.4) on Linux? I have the Slackware 2.0 distribution of Linux, and by golly she don't work too good. Motif apps run fine, but I can't compile a thing using make or imake. The most common failure is 'can't find ' -- make should be compiling with -I/usr/X386/include, but doesn't. Another problem- when I compile manually, the call to XtSetLanguageProc always generates a runtime error message, "Locale not supported by C library." This is a call to an Xt-, not an Xm- function, which leads me to suspect it's Slackware that's broken, not Swim-Motif. Any advice will be welcome (IF it's free :-) --Doug ------------------------------ From: jensen@peruvian.cs.utah.edu (travis jensen) Subject: ext2, UFS, Linux, and Mach Date: 14 Oct 1994 22:04:57 GMT Reply-To: jensen@peruvian.cs.utah.edu I am considering a project to take the ext2fs and put a layer of vaneer over it so it could be mounted as a UFS file system. This would allow Mach to be able to share Linux partitions (and vice-versa). A couple of questions: Is this even possible? I figure anything is possible, but is this a project of weeks, months, years, or millenia? Where can I find documentation on the ext2fs? I would like something a little better than source. :) Is this being done already? If so, who is doing it? Thanks. Travis -- Travis A. Jensen | USER, n.: jensen@peruvian.cs.utah.edu | "The word computer professionals Linux contractor for hire | use when they mean 'idiot'." "Am I really graduating | --Dave Barry next summer??" | "Claw Your Way to the Top" ------------------------------ From: djt1@konichiwa.cc.columbia.edu (David J Topper) Crossposted-To: comp.os.linux.help,comp.os.linux.misc,comp.music Subject: Cmix,RT -> Linux,DOS Date: 14 Oct 1994 21:16:35 GMT Does anyone know any one or more of the following: - cmix for DOS? - where might I find it, I've only seen for Linux. - Linux / Xfree86 based graphical sound editor to view sound files, etc... - Linux / Xfree86 based sound file converter - .snd files to .au (whatever format Linux uses) files? - RT for DOS/Windows or Linux/Xfree86? Thanks, ------------------------------ From: bsa@kf8nh.wariat.org (Brandon S. Allbery) Subject: Re: Shared Libs: working toward a permanent solution? Date: Fri, 14 Oct 1994 21:51:38 GMT In article <37mepf$mei@senator-bedfellow.mit.edu>, tytso@MIT.EDU says: +--------------- | of accessing or taking the address of a static variable. To take the | address of a static variable, just convert the segment-register-relative | offset to a global, absolute offset by adding the base address of the | library, and use the resulting absolute pointer thereafter. This will | | Unfortuantely, life's not that simple. What if you take the address of | a static variable, and store it into a structure? That structure member | might contain a pointer to a static variable local to the shared | library, or it might contain a pointer to memory in the main program's | data segment. +------------->8 I think you misunderstand his intent: he's talking about using a segment register as if it were a general register containing an address. Basically, using e.g. %gs as one would normally use %bp. Segment overrides would not be used in this scheme; instead, the address would be calculated when needed by adding %gs to the library-base-relative address. That said (and I address this to the group, not to you, Ted; I suspect you know more about this than I do :-) I don't think you can get away with using a segment register as a general "address" register, because in 386 mode a segment register points to a segment descriptor, not to the base address of a segment. Moreover, retrieving and setting the value of the register as if it were a general register is impossible because its value gets validated by the hardware as a segment descriptor pointer. System calls would probably have to be used to manipulate the value, which will probably slow things down just as much as if we did it the normal way. Perhaps more so --- the overhead of a system call is likely to be greater than the overhead of pushing and popping registers because there aren't enough to go around with one allocated as a base pointer. ++Brandon -- Brandon S. Allbery KF8NH [44.70.4.88] bsa@kf8nh.wariat.org Linux development: iBCS2, JNOS, MH ~\U Waiting For Godot^H^H^H^H^HRothenberg ------------------------------ From: andreas@orion.mgen.uni-heidelberg.de (Andreas Helke) Crossposted-To: comp.os.386bsd.development Subject: Re: We a FAQ: Linux vs. *BSD!!! Date: 14 Oct 1994 20:55:18 GMT Ken Hughes (hughes@napa.eng.uop.edu) wrote: : I wonder how the Windows-DOS-OS/2 people react to these questions? Does : anyone ever ask over in those groups? Or is everyone there resigned to the : fact that OS/2 is best or Windows is best or Chicago/Windows 95 will be best : or that they're all SOL? :-) You only have to look at the multiple 100 post flamewars that regulary come up in comp.os.ms-windows.advocacy. The Windows solution is simply to assign a special newsgroup for this discussion. Andreas -- * Andreas Helke, Institut fuer molekulare Genetik, Universitaet Heidelberg ** Im Neuenheimer Feld 230, 69122 Heidelberg, Germany *** orion.mgen.uni-heidelberg.de has a ftp server with Unix tools for DOS **** and with the HFM V. 4.02 DOS file manager and archive program shell ------------------------------ From: earl@tenrec.cig.mot.com (Brent R. Earl) Subject: Re: [fdformat] kernel 1.1.52 (and .53) patch comment Date: 14 Oct 1994 15:51:54 GMT >: I justed installed the kernel patch 1.1.52, and now I have problem using >: fdformat. I abort with an IOCTRL error. >[...] > [oops, yet another floppy bug] > > It can be fixed by applying the following small patch: >--- linux-1.1.52/drivers/block/floppy.c Thu Oct 6 20:56:59 1994 >+++ linux/drivers/block/floppy.c Sat Oct 8 13:50:38 1994 >@@ -1745,6 +1745,7 @@ > raw_cmd.track = format_req.track << floppy->stretch; > buffer_track = -1; > setup_format_params(); >+ clear_bit(current_drive, &changed_floppies); > floppy_start(); > #ifdef DEBUGT > debugt("queue format request"); Applied this patch, and it worked for a bit, until I 'rdev'ed a new boot floppy. This locked the drive for fdformat again, until I did a 'mount /dev/fd0 /dos' and got an a mount error, but the drive was freed up again. The setfdprm line didn't work for me. - Brent - earl@rtsg.mot.com (Brent Earl) Motorola CIG +1-708-632-5830 ** Panic! Could not locate coffee - Operator halted ** ------------------------------ ** 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 ****************************** .