Received: from SOUTH-STATION-ANNEX.MIT.EDU by po7.MIT.EDU (5.61/4.7) id AA04415; Fri, 15 Dec 95 18:20:36 EST Received: from SENATOR-BEDFELLOW.MIT.EDU by MIT.EDU with SMTP id AA16396; Fri, 15 Dec 95 18:20:28 EST Received: (from root@localhost) by senator-bedfellow.MIT.EDU (8.6.12/2.3JIK) id SAA27314 for Linux-Development-System-Dist@senator-bedfellow.mit.edu; Fri, 15 Dec 1995 18:13:52 -0500 Message-Id: <199512152313.SAA27314@senator-bedfellow.MIT.EDU> From: Digestifier To: Linux-Development-System@senator-bedfellow.MIT.EDU Reply-To: Linux-Development-System@senator-bedfellow.MIT.EDU Date: Fri, 15 Dec 95 18:13:48 EST Subject: Linux-Development-System Digest #121 Linux-Development-System Digest #121, Volume #2 Fri, 15 Dec 95 18:13:48 EST Contents: Re: Linux on BeBox? (muzo) Sockets library? Help, need to create /proc entry (Garth Nielsen) Re: Are libc functions MT-safe? (William S. Gribble) Re: Linux Security - C2 and beyond (Harald Milz) Re: AMD DX4-120 (Palvolgyi Tibor) Re: Compiling in Linux (Harald Milz) Re: Will Linux work with the P6 ??? (Yan-Song Chen) Re: Converting dynamic to static ? (Rogier Wolff) Can't insmod zftape under 1.3.45 (David Adams) Transputer Linux? (Alan Brown) Re: Kernel resets IDE HD -- Normal? (Albert Cahalan) Re: Kernel protection against device drivers! (Mats Andtbacka) Re: Transputer Linux? (Alberto Vignani) Re: Linux has poor memory management? (Albert Cahalan) Re: 'make' fails with 1.3.47? (Eric Dumazet) ---------------------------------------------------------------------------- From: muzok@msn.com (muzo) Subject: Re: Linux on BeBox? Date: Wed, 13 Dec 1995 05:51:24 GMT Reply-To: muzok@msn.com im14u2c@cegt201.bradley.edu (Joe Zbiciak) wrote: > >Has there been any talk of Linux on the BeBox? > > >From the description I saw of the architecture, it sounds like the perfect >Linux device... multiple PPC chips, open architecture, etc. Frankly I'd think Linux on CHRP would be a much better use of the resources (they are limited you know!). Last friday I went to a demo of Be and they are already talking about porting their own OS to CHRP. I'd think the Be hardware will have a very short life time. muzo standard disclaimer ------------------------------ From: bkoen@bill.anet-stl.com () Subject: Sockets library? Date: 14 Dec 1995 17:53:54 GMT ------------------------------ From: garth@say.acnatsci.org (Garth Nielsen) Subject: Help, need to create /proc entry Date: 12 Dec 1995 01:30:48 GMT I am patching the kernel to add a /proc entry showing throughput on the serial ports. For example cat /proc/serial would yield: ttyS0 23452 234523 ttyS1 2341 241 Could someone please outline for me the process of adding a file to the root of the /proc system? I have been looking over the source files but I have not been able to put it altogether. What files need patching? array.c? root.c? fs_proc.h? TIA Garth Nielsen http://www.acnatsci.org/~gnielsen ------------------------------ From: grib@cs.utexas.edu (William S. Gribble) Subject: Re: Are libc functions MT-safe? Date: 13 Dec 1995 23:30:34 -0600 miguel@athena.nuclecu.unam.mx (Miguel de Icaza) writes: > > 2. Are library functions like malloc MT-safe? Or do I need to use > > a semaphore to ensure that threads do not mess up the heap? > > Libc 5.x has now some support (do not know how much) for threads. The libc support for threads isn't quite ready for public consumption, I believe (but I could be wrong); in any event, the pthreads package (http://www.mit.edu:8001/people/proven/pthreads.html) provides a thread-safe stdio and other basic routines. I use pthreads extensively with new and delete and STL and networking and SYS V IPC, and it's just great. My understanding is that clone() in *not* intended as a user interface to threads, just as a thread-implementor's interface. So you can forget about it unless you want to reimplement the POSIX or another thread API. Good luck, Bill Gribble ------------------------------ From: hm@seneca.han.de (Harald Milz) Subject: Re: Linux Security - C2 and beyond Date: Wed, 13 Dec 1995 07:58:41 GMT Heater WONG (heaterw@hkstar.com) wrote: > > I think this is a great ideal, build up Linux to C2 security can be an important > mile stone in Linux world (I think so.) As long as there's source code available for Linux and its accompanying programs such as GCC (and we do love to play with GPL'd software, don't we?), I don't see any way to get a C2 certification for Linux. -- Harald Milz (Harald.Milz@Linux.org) The biggest difference between time and space is that you can't reuse time. -- Merrick Furst ------------------------------ From: amigo@e006.esh.uni-linz.ac.at (Palvolgyi Tibor) Subject: Re: AMD DX4-120 Date: 14 Dec 1995 03:26:27 GMT root (root) wrote: : I have update to an AMD DX4-120..noticed one problem..System won't boot a : floppy kernel unless I turn off the internal cache (bios function)...I don't : really need the function unless I have to reinstall Linux ..I had to repair an : error and needed to boot from a floppy to get into the system when I discovered : it.. : May just be a quirk of my motherboard/bios or could be the DX4/120.. : Anyone have similiar experience? : The kernel starts to uncompress and then gives a crc error and halts.. : Works fine with lilo and loadlin.. : Fred HI! It is a motherboard problem... It was the same error on an EXPERT Vesa Local Bus board, until we jumpered it to an other CPU-type (I think P24-D, but i am not sure ... :-(( ). The AMD DX4-120 is called V8B (B because it has Write Back first level cache), and it is not the same as a "normal" AMD DX4 ... V8T (T=Through, but look at http://www.amd.com). Or buy an other board... For example ASUS SP3 fully supports the AMD DX4 V8B plus CPU, even the CYRIX DX5 ... amigo ------------------------------ From: hm@seneca.han.de (Harald Milz) Subject: Re: Compiling in Linux Date: Thu, 14 Dec 1995 20:40:47 GMT JC (johnc@sound.net) wrote: > What does this mean? Please help if you can because I have some Unix > development that I need to get done. Thanks. You have to install the kernel sources as well. Otherwise you don't have /usr/include/linux/*. -- Harald Milz (Harald.Milz@Linux.org) "I only touch base with reality on an as-needed basis!" -- Royal Floyd Mengot (Klaus) ------------------------------ From: phys6k@menudo.uh.edu (Yan-Song Chen) Subject: Re: Will Linux work with the P6 ??? Date: 15 Dec 1995 09:57:40 GMT Harald Milz (hm@seneca.han.de) wrote: : Eric Kahler (ekahler@mars.superlink.net) wrote: : > but I was curious about wether or not Linux will run : > on a P6 processor. : I had a short hands-on a device Escom sells over here in Germany, and I can : say Linux boots happily with a P6. The SSBA benchmark suite, however, : stopped at one point we couldn't track down. It appears that unter Linux a : P6 is about 1.5 times as fast as a P5 at the same clock speed - no big step : for mankind. Buy P133's and you're fine for the next year or so. I bet the P6 will run better, at least twice as fast as P5 at the same clock speed, if linux will be tuned very well for P6. Maybe we should wait until the gcc supports P6 optimization, at that time the P6 will become very cheap, I think. : -- : Harald Milz (Harald.Milz@Linux.org) : ... A solemn, unsmiling, sanctimonious old iceberg who looked like he : was waiting for a vacancy in the Trinity. : -- Mark Twain -- yansong chen -- YChen@uh.edu ------------------------------ From: wolff@socrates.et.tudelft.nl (Rogier Wolff) Crossposted-To: comp.os.linux.networking Subject: Re: Converting dynamic to static ? Date: 15 Dec 1995 16:57:25 GMT Reply-To: r.e.wolff@et.tudelft.nl Peter Choynowski (pkc@scs.carleton.ca) wrote: : I am looking into setting up a floppy based router, and would like to : use Linux and gated. The idea is to have just a minimum system with : 2 network cards and a floppy. A ramdisk would be used, so the floppy : is only needed for booting - no mechanical parts to fail :) After : getting some idea of the space needed I have reached a conclusion that : I will need a static version of gated and few other programs for all of : this to fit on a single floppy. : 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 ? I have a ready-made package that does this, except for that I don't have gated on the floppy. How large is the executable (with sharedlibs)? I have a (light) shared lib on the floppy....... While on the subject, does gated speak OSPF? We just tested the performance of these beasts. We can get around 600k/sec through them, while the same hardware only allows 300k/sec when using PCBRIDGE. The hardware by the way is a 386/DX40 with 4Mb RAM and a single floppy drive. Roger. -- *** War doesn't determine who's right ****** War determines who's left. *** ** EMail: R.E.Wolff@et.tudelft.nl * Tel +31-15-2783643 or +31-15-2137459 ** *** my own homepage *** ------------------------------ From: vera2@netcom.com (David Adams) Subject: Can't insmod zftape under 1.3.45 Date: Thu, 14 Dec 1995 14:44:58 GMT Help requested. I just upgraded from version 1.2.13 kernel to 1.3.45 (I wanted more than 4 ide drives). Everything seems to be working fine except I can't load my newly aquired zftape. I made the zftape with just the default settings after I recompiled and installed the new kernel (several times) but I get the following when I try to load it..... ====== $insmod -x zftape.o bad_user_access_length: wrong version Failed to load module! The symbols from kernel 1.3.45 don't match 1.3.45 Here is my /usr/src/linux/.config # # Automatically generated make config: don't edit # # # General setup # CONFIG_MATH_EMULATION=y CONFIG_NET=y # CONFIG_MAX_16M is not set # CONFIG_PCI is not set CONFIG_SYSVIPC=y CONFIG_BINFMT_ELF=y # CONFIG_KERNEL_ELF is not set # CONFIG_M486 is not set # # Loadable module support # CONFIG_MODVERSIONS=y # # 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_XD is not set # # Networking options # # CONFIG_FIREWALL is not set CONFIG_INET=y # CONFIG_IP_FORWARD is not set # CONFIG_IP_MULTICAST is not set # CONFIG_IP_ACCT 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 is not set # CONFIG_IPX is not set # CONFIG_ATALK is not set # CONFIG_AX25 is not set # CONFIG_NETLINK is not set # # SCSI support # # CONFIG_SCSI 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 is not set # CONFIG_EQUALIZER is not set CONFIG_NET_ALPHA=y CONFIG_NET_VENDOR_SMC=y CONFIG_WD80x3=y CONFIG_ULTRA=y # 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 is not set CONFIG_PROC_FS=y CONFIG_NFS_FS=y # CONFIG_ROOT_NFS is not set CONFIG_ISO9660_FS=y # CONFIG_HPFS_FS is not set CONFIG_SYSV_FS=y # CONFIG_SMB_FS is not set # # character devices # # CONFIG_CYCLADES is not set # CONFIG_STALDRV is not set CONFIG_PRINTER=y # CONFIG_BUSMOUSE is not set # CONFIG_PSMOUSE is not set # CONFIG_MS_BUSMOUSE is not set # CONFIG_ATIXL_BUSMOUSE is not set CONFIG_QIC02_TAPE=y CONFIG_QIC02_DYNCONF=y # # >>> Setting runtime QIC-02 configuration is done with qic02conf # # # >>> Which is available from ftp://ftp.funet.fi/pub/OS/Linux/BETA/QIC-02/ # # # Sound # CONFIG_SOUND=y # # Kernel hacking # # CONFIG_PROFILE is not set ===== -- ============================================================================= Official Spokesman for Netcom, IBM, the CIA, the US Govt, the Klingon Empire, the .......Hey! Get Away!....I don't wanna go back!!!...Arrrgh David Adams -- vera2@netcom.com ------------------------------ From: alan@papaioea.manawatu.planet.org.nz (Alan Brown) Subject: Transputer Linux? Date: 15 Dec 1995 01:44:40 +1300 I noticed some Transputer files in the kernel area on Funet. How far has Linux progressed for this platform, if it's being ported to it? -- "But you can't say they have arrived to classify your nervous lies What's right, what's not, what may not be. Don't try your tricky tests on me For I won't live in fear my friend - that every day could be my end" ------------------------------ From: albert@krakatoa.ccs.neu.edu (Albert Cahalan) Subject: Re: Kernel resets IDE HD -- Normal? Date: 15 Dec 1995 21:26:08 GMT >>>>> "M" == Michael Faurot writes: M> Albert Cahalan (albert@krakatoa.ccs.neu.edu) wrote: : : This may surprise M> you, but I think you may have a _serious_ hardware : problem. Replace the M> motherboard, controller, and IDE drive all at : the same time and get rid M> of the old ones. M> I think that approach is a bit extreme at the moment, considering the M> machine does run, and can run for several days at a time without any major M> barfing even though I do get errors occasionally like this: M> Dec 14 00:25:15 phzzzt kernel: hdb: DRDY error: status=0x59 { DriveReady M> SeekComplete DataRequest Error } Dec 14 00:25:16 phzzzt kernel: hdb: DRDY M> error: error=0x10 { SectorIdNotFound }, LBAsect=183595, sector=16442 Dec 14 M> 00:25:16 phzzzt kernel: ide0: do_ide_reset: timed-out, status=0xd0 Dec 14 M> 00:25:16 phzzzt kernel: hdb: DRDY error: status=0xd0 { Busy } Dec 14 M> 00:25:16 phzzzt kernel: ide0: do_ide_reset: success Dec 14 00:25:16 phzzzt M> kernel: hdb: enabled 8-sector multiple mode Dec 14 00:25:16 phzzzt kernel: M> hda: enabled 8-sector multiple mode Dec 14 00:28:22 phzzzt kernel: hdb: M> DRDY error: status=0x59 { DriveReady SeekComplete DataRequest Error } Dec M> 14 00:28:22 phzzzt kernel: hdb: DRDY error: error=0x10 { SectorIdNotFound M> }, LBAsect=120608, sector=248500 M> : What I think happened: an accident with a screwdriver that destroyed : an M> IDE board also zapped the motherboard. Everything plugged into : that slot M> was slowly destroyed - maybe I burned out a capacator. M> If something was actually corrupting my drives I'd think I'd have problems M> with my other IDE drive (hda)--but I don't, just on hdb which is where I've M> had my Maxtors. -- A quality drive can take more abuse. An older drive can withstand voltage spikes better, maybe. The fact that one works does not rule out serious mothorboard problems. I fried two drives and several controllers before I got rid of the motherboard. Two small older drives were never damaged. -- Albert Cahalan albert@ccs.neu.edu ------------------------------ From: mandtbac@news.abo.fi (Mats Andtbacka) Subject: Re: Kernel protection against device drivers! Date: 14 Dec 1995 23:54:42 GMT Reply-To: mandtbac@abo.fi Alain Knaff, in <4am1g3$g98@imag.imag.fr>: [...] > I think VMS did this by using three protection rings: one for the >user, one for the kernel, and one for the device driver. The INTEL >processors have 4 protection rings, so this should be possible. Maybe, but could it be kept portable wrt the other architectures? I'm unclear on whether the 680x0 has any equivalent of 80x86 protection rings, for example; please don't ask about the PPC or Sparc... -- " ... got to contaminate to alleviate this loneliness i now know the depths i reach are limitless... " -- nin ------------------------------ From: alberto.vignani@torino.alpcom.it (Alberto Vignani) Subject: Re: Transputer Linux? Date: 15 Dec 1995 17:26:22 -0500 Reply-To: alberto.vignani@torino.alpcom.it (Alberto Vignani) On 14 Dec 1995 14:59:37 GMT prasad@watson.ibm.com wrote: |In article <4ap67o$86v@papaioea.manawatu.gen.nz>, alan@papaioea.manawatu.planet.org.nz (Alan Brown) writes: ||> I noticed some Transputer files in the kernel area on Funet. ||> ||> How far has Linux progressed for this platform, if it's being ported to it? | | |Should be interesting. |Transputers do their own thing with process management.. |Wonder if there have been any other Unix ports to transputers already, |and how they went about it? | Transputers are microcontroller-like and have no memory management, so it wouldn't be an easy port... but I hope that Linux-8086 will be a good starting point, if and when it will be ready, to get Linux running even on this class of machines. Linux on _multiple parallel_ Transputers is currently only a dream. There are(were?) some UNIX-like OSs available on Transputers, the best known are Helios and Parix. There are no free OS for this chip AFAIK; however, there is a port of gcc 2.6.3 available from ftp.botik.yaroslavl.su (mirrored at unix.hensa.ac.uk), which can be compiled and run under Linux _but_ needs the Parix assembler and linker to be useful. (Other GCC ports are simply not working). Under Linux you can currently find only _drivers_ for the Inmos boards B004/B008. The latest official version is 0.8; I currently run a 0.9 preliminary version I grabbed once from... well, don't remember, maybe ftp.swt.ruhr-uni-bochum.de, the "official" site; but development of transputer drivers under Linux has apparently stopped since a long time. If you are interested enough, look for the hardware manual of the new SGS-Thomson ST20450; it is available (the manual, I mean :-) from unix.hensa.ac.uk, in Postscript format. Alberto ------------------------------ From: albert@krakatoa.ccs.neu.edu (Albert Cahalan) Subject: Re: Linux has poor memory management? Date: 15 Dec 1995 22:12:07 GMT >>>>> "J" == John Burton writes: >> It says that RAM is a lot faster than the hard drive. >> If you want better performance, add more RAM. >> No no no! This really is a problem specific to Linux. You >> missed the part about 'positively feeds back'. The complaint >> is not that swap space is slow (of course it is) but that a >> swapped process can never leave swap under the right conditions. >> No joke. >> To demonstrate: write a program that mallocs most of your RAM, >> say 6 MB on an 8MB machine. The program must (randomly?) touch >> all the pages over and over again. Just touch them, to keep >> the program fast. Now run this program in the background. >> Wait a bit, then run a second copy. >> When one process needs a page, the other one gets to run. The >> running process dirties its pages, so they will not get swapped >> (ok, paged) out. The process that needed a page gets one, but >> also loses a page. This causes the needy process to take another >> page fault, letting the pig run again and dirty more pages. >> The kswap patches added to 1.3.xx indirectly helped, but they did not >> solve the real problem. They did reduce the symptoms a bit though. > Okay, sounds like Linux memory management has a problem, but what is > the "real problem", and how do we solve it? You indicate that the > kswap patches helped, but did not solve the "real problem" so it > sounds like you have a clear understanding of what the real problem > is. I don't, but would like to...could you please explain what the > real problem is here ? Understanding the problem is easy enough, but fixing it is another matter! A peek at the SunOS code might help very much. :-) The problem is that the linux kernel does exactly what you might expect: while getting a page for process X, run process Y. When the page process X needs is available, page out some other page that has not been touched in a long time. This is a problem if the running process keeps touching all of its pages. Process X steals a page from itself, and loses big. To fix the problem, one of those recently touched pages in the other process must be sent out to swap. The problem is not easy to fix because swapping out the most recently touched page is usually a Bad Thing, but needed to solve the problem! I think the current situation works well for batch jobs. It can be a disaster _sometimes_ for interactive work while something big is also running. That might mean 6 users with xemacs and 20 other users running vi. I'm not sure exactly what it takes to cause the problem, and the requirements vary by machine. The Linux memory management problem is really one of a bad interaction between the swap code and the scheduler. Sheduler: Choose a process in core to run. Swapper: Steal page not recently/often used. Both make perfect sense. Together with the right conditions, some processes never get to run much and just sit in swap. What makes this a disaster is the 'positively feeds back' part. -- Albert Cahalan albert@ccs.neu.edu ------------------------------ From: Eric Dumazet Subject: Re: 'make' fails with 1.3.47? Date: Fri, 15 Dec 1995 17:30:17 -0500 I suppose you have libc.so.5.2.16 You didn't saw the release notes H.J Lu provided with this version ? It gives the patches needed on gnu make... ------------------------------ ** 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 ******************************