Received: from SOUTH-STATION-ANNEX.MIT.EDU by po7.MIT.EDU (5.61/4.7) id AA20207; Mon, 18 Dec 95 04:19:43 EST Received: from SENATOR-BEDFELLOW.MIT.EDU by MIT.EDU with SMTP id AA21280; Mon, 18 Dec 95 04:19:40 EST Received: (from root@localhost) by senator-bedfellow.MIT.EDU (8.6.12/2.3JIK) id EAA02378 for Linux-Development-System-Dist@senator-bedfellow.mit.edu; Mon, 18 Dec 1995 04:14:01 -0500 Message-Id: <199512180914.EAA02378@senator-bedfellow.MIT.EDU> From: Digestifier To: Linux-Development-System@senator-bedfellow.MIT.EDU Reply-To: Linux-Development-System@senator-bedfellow.MIT.EDU Date: Mon, 18 Dec 95 04:13:58 EST Subject: Linux-Development-System Digest #131 Linux-Development-System Digest #131, Volume #2 Mon, 18 Dec 95 04:13:58 EST Contents: Re: Linux Security - C2 and beyond (William Burrow) Re: parity error with dos but ok in linux (Bob Lorenzini) Re: parity error with dos but ok in linux (Adam J. Richter) Re: AMD DX4-120 (fbroce) Re: 3Com 3c595 patch (Robert Megee) Re: Linux has poor memory management? (Jonathan George) Re: IDE Seagate 157A - nastyness.. (Mark Lord) Re: ld won't dynamically link libc.so.2.5.16 (Mike Norman) Re: iBCS for 1.3.45 (Daniel Hermans) Re: Unusual disk activity under 1.3.45... (fuZZy) Re: If Linux can mount a file system exceed 300 MBytes? (root) Sound and 1.3.45 don't work for me (John Timmers) kmalloc/mmap bug? (Lok Tin Liu) Re: Linux Security - C2 and beyond (Joe Buck) ---------------------------------------------------------------------------- From: aa126@fan1.csd.unb.ca (William Burrow) Subject: Re: Linux Security - C2 and beyond Date: 16 Dec 1995 21:09:56 GMT Scott Johnson (johnsos@holmes.ece.orst.edu) wrote: : In article , Harald Milz wrote: : >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. : Why? Is one of the requirements of C2 security that source code for the : applications be unavailable? I am pretty sure that is NOT the case. Its been 4 or 5 years since I looked at documentation for the security levels, but I believe that whether source code is available or not is irrelevant for class C security. C2 can't be all that hard to get, its just that the certification probably wouldn't apply to newer kernels/distributions and free Unix being what it is.... -- -- William Burrow -- Fredericton Area Network ------------------------------ Crossposted-To: comp.os.linux.hardware From: Bob Lorenzini Subject: Re: parity error with dos but ok in linux Date: Sun, 17 Dec 1995 19:57:24 GMT There is a boot sector virus that gives false parity errors. Bob hwm@netcom.com _______________________________________________________________________________ The Microsoft Network is prohibited from redistributing this work in any form, in whole or in part, without a license. Such a license is available to Microsoft for $500/msg. Redistribution without permission constitutes an agreement to the above terms. If a violation is noted, notify postmaster@microsoft.com and hwm@netcom.com. (c) 1995 Ryan Tucker. _______________________________________________________________________________ ------------------------------ From: adam@adam.yggdrasil.com (Adam J. Richter) Crossposted-To: comp.os.linux.hardware Subject: Re: parity error with dos but ok in linux Date: 14 Dec 1995 21:10:33 GMT In article <30CF46D0.2C9A@gserver.grads.vt.edu>, Thomas Dunbar wrote: >i just upgraded my son's computer to 16megs so he can work >with Java. > >if he boots in dos and runs windows (or another app that >takes a good bit of memory), the system halts with a parity >error. > >However, i can boot Linux, load X, Emacs, Netscape, gcc till >i'm using all 16megs of ram plus some swap and everything runs >fine. > >how can this be? are the simm chips i got smart enough to >know that fathers shouldnt permit their children to do DOS? >that's what i'd prefer but he does need dos for some things.. While we're on the subject of parity and memory problems, you may be interested in knowing that we've had reports of Triton Pentium motherboards, both Endeavor and ASUS, having bit corruption which goes away when you put in a different set of 70ns SIMM's. Apparently, there is no parity checking in the Triton chipset *and* it also pushes the timing on DRAM's more than in the past. So far, we haven't gotten any reports of this problem when 60ns DRAM's are used in these motherboards. -- Adam J. Richter Yggdrasil Computing, Incorporated (408) 261-6630 "Free Software For The Rest of Us." ------------------------------ From: fbroce Subject: Re: AMD DX4-120 Date: 17 Dec 1995 00:46:20 GMT Not true re other operating systems..It will boot dos ok from floppy..It may be a function of the hard drive..I am not sure. It could be the floppy I once had this problem and discovered I had the floppy drive not plugged in properly..I have a DX4-100 (AMD) w/o the writeback cache. I will try it again sometime when I have the case off. I suspect it is some kind of timing problem. It is funny that Linux will uncompress fine from hd or from dos drive but won't otherwise unless I turn off the internal cache. Since others have noted similiar problems. I am guessing it does have to do with the write back cache. I am pleased with the speed..259.4 on Norton ver 8 59.8 bogomips with Linux..compiles kernels fine and quickly..I also have 16 mb ram. Any other experiences? fb ------------------------------ From: megee@flopn2.dseg.ti.com (Robert Megee) Crossposted-To: comp.os.linux.networking,comp.os.linux.setup,comp.os.linux.misc Subject: Re: 3Com 3c595 patch Date: 15 Dec 1995 03:54:21 GMT In article <30D05160.2781E494@cs.unm.edu>, boyd@cs.unm.edu says... > >Has anyone used the vortex.patch for 1.2.13 for the >3COM 3C595 ethernet card successfully? I'm using I used the patch for a 3c595. It worked for me. Robert Megee megee@flopn2.dseg.ti.com ------------------------------ From: jgeorge@cftnet.com (Jonathan George) Subject: Re: Linux has poor memory management? Date: 18 Dec 1995 06:05:57 GMT Nick Kralevich (nickkral@parker.EECS.Berkeley.EDU) wrote: : This isn't a problem with Linux in specific -- this is a problem : with all computer systems. : In all of operating systems, there is a fundamental tradeoff between : fairness and throughput. The two have to be balenced. [...cut lots of OS 101 stuff...] True. : Compounding this problem is the fact that the operating system doesn't : know how long the programs are going to run. Though things like priorities, threads, process type (event realtime, time sliced, opportunistic) _could_ be used to improve behavior. : Anyway, the point of this example was just to point out somthing: : Fairness and throughput are not compatiable. The fairest operating : systems will usually be the ones that are least efficient. The : most efficient operating systems will be the least fair. It is : almost impossible to get around this problem, short of adding : more CPUs. So you have to make a tradeoff between throughput and : fairness. The tradeoff can be minimized by taking the general preferred case algorithm and using another algorithm to anticipate and level worst cases. -Jonathan- http://www.cftnet.com/~jgeorge ------------------------------ From: mlord@bnr.ca (Mark Lord) Subject: Re: IDE Seagate 157A - nastyness.. Date: 11 Dec 1995 21:23:37 GMT In article <4ahle8$8q@cantaloupe.srv.cs.cmu.edu> jmcc@m5.vi.ri.cmu.edu writes: > I have an old 40Mb ST157A that I wanted to hang off my linux machine >(as a swap device mainly). However, although it's detected by >my kernel 1.3.36 and 1.3.45, I keep getting timeouts on reads or >writes (most often "write - interrupt timed out") only on > *) Swapspace > or > *) Dynamically paged executables Mmm.. I have tested Linux 1.3.xx (for "xx" up to about 25 or so) with an ST157A -- no problems seen. I don't think anything new since then would have broken it, but that doesn't solve *your* problem. Mmmm.. -- mlord@bnr.ca Mark Lord BNR Ottawa,Canada 613-763-7482 For Linux IDE (big/many) help, see: /usr/src/linux/drivers/block/README.ide For latest Linux kernels: ftp.cs.helsinki.fi:/pub/Software/Linux/Kernel/v1.[23] ------------------------------ From: mwnorman@foobar.ocunix.on.ca (Mike Norman) Subject: Re: ld won't dynamically link libc.so.2.5.16 Date: Mon, 11 Dec 1995 11:46:36 GMT In article <4agepa$278@charm.magnus.acs.ohio-state.edu>, Sergei Viznyuk wrote: >While screwing around with compilation stuff and >binutils (2.6.0.2) at some point I found >ld won't link dynamically any more. >I re-installed binutils, and ld.so.1.7.12 >but it didn;t help. > >Could anybody give a hint please? > >Thanks, >Serge I asked the same question in c.o.l.d.apps - it seems that there is a new 'feature' in how the linker regards the *.so symbolic links. In /lib, I had Gidday [~]> cd /lib Gidday [/lib]> ll libc.so.5* lrwxrwxrwx 1 root root 14 Dec 10 14:43 libc.so.5 -> libc.so.5.2.16* -rwxr-xr-x 1 bin bin 562683 May 18 1995 libc.so.5.0.9* -rwxr-xr-x 1 bin bin 549124 Dec 1 21:24 libc.so.5.2.16* I was told to add: Gidday [~]> ln -s libc.so.5 libc.so Now, things linked against the 'C' library are done so dynamically (as one would expect in an ELF-system). I had to do this for everything in /lib, /usr/X11R6/lib, and /usr/lib Ciao, -- Mike Norman - mwnorman@foobar.ocunix.on.ca "...strange women lying in ponds, distributing swords is no basis for a system of government. Supreme executive power derives from a mandate from the masses, not from some farcical aquatic ceremony!" (Monty Python's _Holy Grail_) -- Mike Norman - mwnorman@foobar.ocunix.on.ca "...strange women lying in ponds, distributing swords is no basis for a system of government. Supreme executive power derives from a mandate from the masses, not from some farcical aquatic ceremony!" (Monty Python's _Holy Grail_) ------------------------------ From: DHermans@cocam.com.au (Daniel Hermans) Subject: Re: iBCS for 1.3.45 Date: Wed, 13 Dec 95 00:55:04 GMT In article <4aca4i$mn2@news00.btx.dtag.de>, 0616250291-0001@t-online.de (Juergen Pfeifer) wrote: >Where can I get a working iBCS module for 1.3.45 ? > >Regards >Juergen Pfeifer > > The standard iBCS from tsx-11 works with the 1.3.45 kernels with a few mods, just install the source and : - change all references of current->sigaction to current->sig->action in the source code in the iBCSemul directory ( should be 3 or 4 files with this ) - add '-D__NO_VERSION__' to the Makefile in iBCSemul, CFLAGS section The system will then compile and work normally.... Thanks to jaggy@purplet.demon.co.uk for giving me this info..here is the text of the message that he sent me re: the same problem: > > I have used ibcs2 extensively before but it now > > no longer builds on my 1.3.45 kernel ( which I realise is a bit too new!! ). > You need to replace each occurence of current->sigaction with > current->sig->action plus you need to add -D__NO_VERSION__ to > the CFLAGS in iBCSemul/Makefile and compile your kernel with > the module versions config option on. You may also want to > uncomment the i486-linuxaout in x286emul/Makefile if you want > the Xenix 286 kludge to actually compile. Mike > -- > .----------------------------------------------------------------------. > | Mike Jagdis | Internet: jaggy@purplet.demon.co.uk | > | 280, Silverdale Road, Earley, | Voice: +44 1734 266996 | > | Reading RG6 7NU ENGLAND | Work: +44 1734 890403 | > `----------------------------------------------------------------------' -- Daniel Hermans, Co-Cam Pty. Limited, Melbourne, OZ Ph : 61 3 9200 3735 Fx : 61 3 9200 3852 Em : DHermans@cocam.com.au ------------------------------ From: tjimenez@site.gmu.edu (fuZZy) Subject: Re: Unusual disk activity under 1.3.45... Date: 18 Dec 1995 06:37:22 GMT Sangria (sangria@inlink.com) wrote: : Has anyone notice an unusal disk activity under the 1.3.45 : kernel? : Almost like clock work, I come home from work, sit infront of : another machine, and while working, my Linux machine would : suddenly start with a barrage of disk activity. : This would go on for about 5 minutes and everything would : return to normal once again. : At the time when this activity is going on, there is nothing : running on Linux except a copy of bash, which is sitting : idle... I've been experiencing the exact same thing, for a good long bit of the 1.3.x development cycle (donna remember if it occured under 1.[12].x) scared the bloody hell out of me the first couple of times. -- -fuZZy ------------------------------ From: root Subject: Re: If Linux can mount a file system exceed 300 MBytes? Date: 17 Dec 1995 23:12:12 GMT mcchen@hntp2.hinet.net () wrote: >I have a Linux (Version 1.3.25) machine with a 500 MBytes hard disk. >The partition table as follows. > >Disk /dev/sda: 64 heads, 32 sectors, 514 cylinders >Units = cylinders of 2048 * 512 bytes > > Device Boot Begin Start End Blocks Id System >/dev/sda1 1 1 16 16368 82 Linux swap >/dev/sda2 17 17 317 308224 83 Linux native >/dev/sda3 318 318 514 201728 83 Linux native > >It works fine. > >Now I add another 500 Mbytes hard disk(the same type as previous) >to this machine. > >The first case : > This new disk only have one partiton exceed 300 Mbytes and mount it to >a directory. >The partition table as follows. > >Disk /dev/sda: 64 heads, 32 sectors, 514 cylinders >Units = cylinders of 2048 * 512 bytes > > Device Boot Begin Start End Blocks Id System > >/dev/sdb1 1 1 514 526320 83 Linux native > >?? But the "df -v" says it only 300 Mbytes available ?? > >The other case : > This new disk has one or more partitions but every one less than >300 Mbytes. > >?? The machine panic when I mount any partition to a directory. > >The messages as follows. > >Kernel panic: ExT2-fs panic (device 8/17): read_block_bitmap: >Cannot read block bitmap_block_group = 13,block_bitmap=106500. > >If Linux can only mount a file system less than 300 Mbytes ? >and, why it panic in the second case ???? Linux can definately mount filesystems larger than 300 Mbytes, see below for my df output, maybe the drive you're trying to mount is defective, I tried once to mount an old HD and got lot's of kernel panics after some days, It turned out that the HD was broken, Fortunately I didn't store too important stuff on it, although I lost about 20 Mb of stuff derived from the internet. It's probably not a bug in the kernel, I can't remember of hearing of some bug in the filesystem in kernel 1.3.25, but I can remember having some problems compiling some kernels around that number. my df-output: Filesystem 1024-blocks Used Available Capacity Mounted on /dev/hdb3 187008 72334 104695 41% / /dev/hdb4 368850 197728 152071 57% /usr /dev/hda4 52891 22209 27860 44% /usr/bin /dev/hdb1 150259 98803 43696 69% /usr/src /dev/hda1 184236 125292 58944 68% /dosc /dev/hda5 255732 186700 69032 73% /dosd /dev/hdb5 255732 56568 199164 22% /dose /dev/hdb6 255732 174092 81640 68% /dosf /dev/hdd 679276 679276 0 100% /cdrom the device "/dev/hdb4 is larger than 300 Mb so Linux can mount them. The cdrom is another kind of filesystem so I don't want to compare it to an ext2 filesystem. Servaas te Brake ------------------------------ From: John Timmers Subject: Sound and 1.3.45 don't work for me Date: Sun, 17 Dec 1995 20:44:56 -0800 Uh, has anybody else had any problems with sound; more specifically using Cyberphone and experiencing problems in transmitting (and receiving at times, also...)? I have the following arrangement here: 1. kernel 1.3.45 2. libc 5.2.18 3. ld.so 1.7.12 4. libm.so 5.0.5 5. binutils-2.6.0.2 6. SoundBlaster 16 (, I know, cheap - but it has worked well) 7. Pentium 75 mhz I had tried Cyberphone with earlier versions of the kernel (just lately, with 1.3.32 -- quite a stable release, I thought) and had some good luck with it. I just installed 1.3.45 a couple of days ago, and have re-compiled it about 5-6 times with different configurations for the soundcard, because I couldn't get it working so that I could converse with other people running Cyberphone. I haven't stumbled on a successful configuration yet. Basically, I have no problem playing *.au files locally; it seems that I can't input (?) sound... I have clipped a couple of lines from /var/log/messages (I load sound support as a module...) Dec 17 19:27:43 yaro kernel: Sound initialization started Dec 17 19:27:43 yaro kernel: snd2 at 0x220 irq 10 drq 1 Dec 17 19:27:43 yaro kernel: snd6 at 0x220 irq 10 drq 5 Dec 17 19:27:43 yaro kernel: Sound initialization complete then, two lines later...(after I've SLIP'ed into ISP - started Cyberphone...) Dec 17 19:50:35 yaro kernel: Sound: DMA (input) timed out - IRQ/DRQ config error? The sharp-eyed reader will see that I do not have FM or MIDI support; well, I did, but this was my last compile, and I had not includes support for that this time. I didn't have any luck with it, so tried it without :=) The settings I had with the earlier kernel (working configuration) have not changed for this kernel. I see that quite a few of the *.h and *.c sound files have changed from 1.3.32 -> 1.3.45, but am not experienced in C programming, so am unable to decipher what they all mean. If someone would provide a couple of hints here, as to what I might be able to do, it would be appreciated regards...john -- ...my other signature is illegible... ==== return email address is: john_timmers@mindlink.bc.ca ==== ------------------------------ From: ltliu@midnight.CS.Berkeley.EDU (Lok Tin Liu) Subject: kmalloc/mmap bug? Date: 17 Dec 1995 21:35:09 GMT In the old days, kmalloc() could allocate only ~4KB (one page). Now kmalloc() can allocate ~128KB. However, I noticed that kmalloc() still incremented the mem_map[] entry only for the first page. Therefore, when I tried to mmap the memory region allocated by kmalloc() to user space, only the page table entry for the first page was set up correctly. Is it a bug in the RMQUEUE and EXPAND macros in mm/swap.c in the 1.2.13 kernel? The code seems to be the same in the 1.3.30 kernel. Has it been fixed in the recent 1.3.x kernel? Thanks. Lok ------------------------------ From: jbuck@synopsys.com (Joe Buck) Subject: Re: Linux Security - C2 and beyond Date: 18 Dec 1995 04:53:19 GMT hm@seneca.han.de (Harald Milz) writes: >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. You're confused about how security certifications work. Anything in an OS that depends on the fact that the source is secret is called "security by obscurity" -- reliance on security by obscurity is highly frowned on by experts in the field. It will be assumed that the bad guys know how the system works. The fact that source is available has nothing to do with whether a system is secure. -- -- Joe Buck (not speaking for Synopsys, Inc) ------------------------------ ** 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 ******************************