Received: from PACIFIC-CARRIER-ANNEX.MIT.EDU by po7.MIT.EDU (5.61/4.7) id AA09246; Mon, 1 Jan 96 08:14:47 EST Received: from SENATOR-BEDFELLOW.MIT.EDU by MIT.EDU with SMTP id AA23923; Mon, 1 Jan 96 08:14:39 EST Received: (from root@localhost) by senator-bedfellow.MIT.EDU (8.6.12/2.3JIK) id IAA04011 for Linux-Development-System-Dist@senator-bedfellow.mit.edu; Mon, 1 Jan 1996 08:13:29 -0500 Message-Id: <199601011313.IAA04011@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, 1 Jan 96 08:13:27 EST Subject: Linux-Development-System Digest #170 Linux-Development-System Digest #170, Volume #2 Mon, 1 Jan 96 08:13:27 EST Contents: Re: applying kernel patches (Brad Littlejohn) Would like a scanner driver... (Stephen S Edwards II) Re: WANTED: A secure kernel `unlink' function. (Wolfram Gloger) 1.3.x kernels (was Re: Elf Sluggish - interactive perf measure wanted.) (Fredrik Sandstrom) problems with dummy.o module on 1.3.35 (Hamish Moffatt) Re: applying kernel patches (Mark Shadley) Re: 1.3.50; why is bdflush reported to be huge by 'ps' (Michael Nelson) ditto 3200, ditto dash, and ftape/zftape (Andrew Kieschnick) Re: WANTED: A secure kernel `unlink' function. ("Mark W. Eichin") Re: WANTED: A secure kernel `unlink' function. ("Mark W. Eichin") Re: modprobe.c compile error (Peter Mutsaers) Re: Where are Kernel Change Summaries? (Russell Nelson) Re: problems with dummy.o module on 1.3.35 (Hamish Moffatt) Re: Windows NT 'stripe set' on MBR? (Gerhard Traeger) Re: WANTED: A secure kernel `unlink' function. (Scott Johnson) Re: modprobe.c compile error (Mark Shadley) kernel message : TIOLINUX (0/8/9) ioctl gone - use /dev/vcs . Anybody ? (R. Janssen) Re: applying kernel patches (Rob Janssen) ---------------------------------------------------------------------------- From: tyketto@corona.unomaha.edu (Brad Littlejohn) Subject: Re: applying kernel patches Date: 31 Dec 1995 23:16:42 GMT Jason Mcmullan (jmcc@m5.vi.ri.cmu.edu) wrote: : Eric Berkowitz (ueberkow@uxa.ecn.bgu.edu) wrote: : : I have been trying to keep up with the development of the 1.3.xx : : keernels and with the speed that new versions come out I have a : : question. : : : Do the kernel patches need to be applied sequestialy or are they : : compunded from a certain level so that I can skip some. That is, if I : : have kernel 1.3.50 do I need to apply patch 1.3.51 and the 1.3.52 or can : : I apply 1.3.52 directly? : : All patches must be applied sequentially, ie: : 1.3.48 -> 1.3.49 -> 1.3.50 -> 1.3.51 -> 1.3.52 -> ... [script deleted] Why do that, when it would be easier to grab the whole source? Yes, you could go from 50, apply the patch to get 51, and so on to 52, but why would you want to compile the kernel to have to compile it again? Seems like too much to do. Why not just jump straight from 1.3.48 to 1.3.52? You should be able to find the full source distribution at most of the kernel ftp sites.. Home site for it is ftp.cs.helsinki.fi:/pub/Linux/Software/Kernel/v1.3. BL. ------------------------------ From: largo@xroads.com (Stephen S Edwards II) Subject: Would like a scanner driver... Date: 1 Jan 1996 00:03:36 GMT ...for a Microtek Scanmaker IIHR flatbed scanner. Anyone here know of where I may find one? -- ______________________________________________________ If someone tells you that you have holes in your head, make sure your surface normals are all facing outward. ====================================================== largo@xroads.com ------------------------------ From: Gloger@lrz.uni-muenchen.de (Wolfram Gloger) Subject: Re: WANTED: A secure kernel `unlink' function. Date: 1 Jan 1996 01:03:16 GMT Reply-To: Gloger@lrz.uni-muenchen.de rfg@ccnet.com (Ronald F. Guilmette) writes: >I'd like to have patch for the kernel... a relatively simple one. (If any >of you weekend hackers happen to have some time on your hands, maybe you >would consider implementing this.) >In a nutshell, I'd like to have a kernel in which the `unlink' function >actually overwrites (e.g. with zeros) the file being unlinked. I have good news for you - for the ext2 filesystem, this is already implemented: look at `chattr' and the `s' mode. Regards, Wolfram. ------------------------------ From: Fredrik Sandstrom Subject: 1.3.x kernels (was Re: Elf Sluggish - interactive perf measure wanted.) Date: 31 Dec 1995 16:01:05 GMT Reply-To: fred@spider.compart.fi smchorse@ringer.cs.utsa.edu (Shawn McHorse) wrote: >In article <4bqhh8$l8d@sheckley.cnam.fr> Edmund Ronald writes: >Take a glance at the 1.3.x development kernels. A lot the new things Linus has >has put in offer speed, speed, and yet more speed. Everybody keeps saying that, and I seem to be the only one experiencing ENORMOUS slowdowns with 1.3.x kernels. At first I didn't notice anything - the BogoMips was slightly lower, but not by a significant amount. I was cd'ing around and tried to edit files and everything worked fine. Then I started X and it was fine too, until I started a screensaver, and GOD it was slow! Then I tried playing DOOM (svgalib) and that really made me want to puke. Ever seen DOOM on a 16 MHz 386? Well this was worse, and apart from that the music was really screwd up, but that's a bug in the sound driver, I suspect. >A new scheduler, the kswap >patch, the new fine-grained cache invalidation for the 486+, and many other >things make 1.2.13 look like a real slow-poke. Unfortunately for me, every >1.3.x kernel I've ever tried has crashed in less than an hour. I've never run a 1.3.x kernel that long... :) >And I'm running a 486DX2/66 with 8MB RAM, BTW. Me too! -- Fredrik Sandstrom, Finland fred@spider.compart.fi ------------------------------ From: s9506579@yallara.cs.rmit.edu.au (Hamish Moffatt) Subject: problems with dummy.o module on 1.3.35 Date: 1 Jan 1996 01:18:37 GMT Reply-To: s9506579@yallara.cs.rmit.edu.au I'm running kernel 1.3.35 and want to use the dummy.o module (/usr/src/linux/drivers/net/dummy.o). Compiled the kernel okay (as ELF) (actually have done several times, been running 1.3.35 for months), "make modules" okay, but insmod: [12:15pm] root@silly:/lib/modules/1.3.35/net# insmod dummy.o (main) ether_setup undefined (main) dev_kfree_skb undefined (main) dev_get undefined (main) register_netdev undefined (main) printk undefined (main) unregister_netdev undefined (main) kfree undefined Failed to load module ! Would prefer to stay with 1.3.35 because it lwared (LinWare) runs on it and everything up to 1.3.37, and I just read in c.o.l.n that 1.3.37 has lots of TCP/IP error messages. I've just ftped 1.3.52 though and will try that. Any help would be appreciated. Thanks! hamish -- hamish moffatt CS/CSE Y1 hamish@rising.com.au s9506579@yallara.cs.rmit.edu.au running: linux 1.3.35, linware 0.95, and lots of other klunky hardware ...... ------------------------------ From: shadcat@catcher.com (Mark Shadley) Subject: Re: applying kernel patches Date: 1 Jan 1996 02:35:28 GMT Reply-To: shadcat@catcher.com In article: <4c75kq$fn4@s-cwis.unomaha.edu>, Brad Littlejohn wrote: >Jason Mcmullan (jmcc@m5.vi.ri.cmu.edu) wrote: >: Eric Berkowitz (ueberkow@uxa.ecn.bgu.edu) wrote: [snip] >: : Do the kernel patches need to be applied sequestialy or are they >: : compunded from a certain level so that I can skip some. That is, if I >: : have kernel 1.3.50 do I need to apply patch 1.3.51 and the 1.3.52 or can >: : I apply 1.3.52 directly? >: >: All patches must be applied sequentially, ie: >: 1.3.48 -> 1.3.49 -> 1.3.50 -> 1.3.51 -> 1.3.52 -> ... > > [script deleted] > > Why do that, when it would be easier to grab the whole source? Yes, >you could go from 50, apply the patch to get 51, and so on to 52, but why >would you want to compile the kernel to have to compile it again? Seems like >too much to do. Why not just jump straight from 1.3.48 to 1.3.52? You don't have to compile each kernel. Just grabbing the patches is a time saver, and for some a money saver. I always patch, unless for some reason the patch doesn't work. I always make sure I have a tarred copy of my current source before applying patches, and apply all patches with script running to record any errors. Mark -- Mark Shadley pgp id:C596C1F9 - fp:B60A796CAF6769B9 0688C30B390A0A66 - on servers ------------------------------ From: nelson@seahunt.imat.com (Michael Nelson) Subject: Re: 1.3.50; why is bdflush reported to be huge by 'ps' Date: 1 Jan 1996 03:55:48 GMT Reply-To: nelson@seahunt.imat.com In article <9512319120.181350@drinkel.ow.org>, miquels@drinkel.ow.org (Miquel van Smoorenburg) writes: |> |> ... but please don't put it on other ftp sites, as this is not |> yet the "final" one. I need two or three more days. Actually, |> probably just an hour or two but I'm going off to the new years |> parties right now :) Have Fun!! I didn't (and won't) put it on any ftp sites. Michael -- Michael Nelson nelson@seahunt.imat.com San Francisco, CA ------------------------------ From: andrewk@sneezy.cc.utexas.edu (Andrew Kieschnick) Crossposted-To: comp.os.linux.hardware,comp.os.linux.misc Subject: ditto 3200, ditto dash, and ftape/zftape Date: 1 Jan 1996 01:57:19 GMT I previously stated that the ditto dash works fine as a second fdc, but I have discovered that this may not be entirely true... I also stated that zftape 1.03 didn't seem to work correctly with it, and this was definitely not true... As it turns out, both ftape and zftape don't work correctly with the ditto dash and ditto 3200 unless I boot dos first and perform some sort of tape related operation in dos before booting linux - obviously the dos/windows backup software I have is initializing something (probably something in the ditto dash, but I don't know) that ftape/zftape doesn't do.... Hence, the ditto dash works, but only if dos/windows backup software is used prior to booting linux... Sorry if I caused anyone any trouble... later, Andrew Kieschnick ------------------------------ From: "Mark W. Eichin" Subject: Re: WANTED: A secure kernel `unlink' function. Date: 31 Dec 1995 22:17:26 -0500 Reply-To: eichin@mit.edu | In a nutshell, I'd like to have a kernel in which the `unlink' function | actually overwrites (e.g. with zeros) the file being unlinked. If I'm not mistaken, ext2fs already supports this with the "secure" attribute on a file. It's a per-file attribute that causes the file to be overwritten with zeros on unlink. From the "chattr" man page: | When a file with the `s' attribute set is deleted, its | blocks are zeroed and written back to the disk. Your point about ftruncate is well taken, I don't know if it is already implemented. _Mark_ Linux-Activists Digest Maintainer ------------------------------ From: "Mark W. Eichin" Subject: Re: WANTED: A secure kernel `unlink' function. Date: 31 Dec 1995 22:21:05 -0500 Reply-To: eichin@mit.edu (on closer inspection, secure overwrites with pseudo random data, rather than zeros, and does handle truncate as well as delete... it originally used zeros but was improved to use random data back in early 1994 :-) _Mark_ Linux-Activists Digest Maintainer ------------------------------ Crossposted-To: comp.os.linux.misc From: plm@compi.hobby.nl (Peter Mutsaers) Subject: Re: modprobe.c compile error Date: Sun, 31 Dec 1995 13:10:07 GMT >> On 30 Dec 1995 05:15:18 GMT, shadcat@catcher.com (Mark Shadley) >> said: MS> In article: <4c1vt4$t92@ngriffin.itc.gu.edu.au>, MS> Tony Nugent wrote: >> From depmod/modprobe.c... >> >> static void verbose (const char *ctl, ...) >> { >> if (verbose_flag){ >> va_list list; >> va_start (list,ctl); >> vprintf (ctl,list); >> va_end (list) >> fflush(stdout); <-- line 22 >> } >> } MS> Notice something ';' missing from line 21? :) va_end used to be a macro that included the semicolon therefore in much code you find it without a ;. Alas this seems te have changed in gcc 2.7.2. -- ______________________________________________________________________ Peter Mutsaers | Bunnik (Ut), | "Quod licet bovis, plm@compi.hobby.nl | the Netherlands | non licet Jovi." ------------------------------ From: nelson@ns.crynwr.com (Russell Nelson) Crossposted-To: comp.os.linux.help,comp.os.linux.misc Subject: Re: Where are Kernel Change Summaries? Date: 01 Jan 1996 04:36:02 GMT In article <4c6n5l$opr@nntp.interaccess.com> solomon@interaccess.com (Paul J. Solomon) writes: > Do any summaries of Linux Kernel changes exist, preferably in summary > form: i.e. I don't care about which exact files were changed, but what > new functionality, bug fixes, etc. were added to a specific version of > a kernel. . -- -russ http://www.crynwr.com/~nelson Crynwr Software | Crynwr Software sells packet driver support | PGP ok 11 Grant St. | +1 315 268 1925 voice | Good government is a contradiction Potsdam, NY 13676 | +1 315 268 9201 FAX | in terms. Ask me for details. ------------------------------ From: s9506579@yallara.cs.rmit.edu.au (Hamish Moffatt) Subject: Re: problems with dummy.o module on 1.3.35 Date: 1 Jan 1996 05:44:22 GMT Reply-To: s9506579@yallara.cs.rmit.edu.au Hamish Moffatt (s9506579@yallara.cs.rmit.edu.au) wrote: : I'm running kernel 1.3.35 and want to use the dummy.o module : (/usr/src/linux/drivers/net/dummy.o). Compiled the kernel okay (as ELF) : (actually have done several times, been running 1.3.35 for months), : "make modules" okay, but insmod: : [12:15pm] root@silly:/lib/modules/1.3.35/net# insmod dummy.o : (main) ether_setup undefined : (main) dev_kfree_skb undefined [...] Same problem with 1.3.52. I did make config make dep; make clean make zImage make modules make modules_install insmod -o dummy0 /lib/modules/1.3.52/net/dummy.o thanks, hamish -- hamish moffatt CS/CSE Y1 hamish@rising.com.au s9506579@yallara.cs.rmit.edu.au running: linux 1.3.35, linware 0.95, and lots of other klunky hardware ...... ------------------------------ From: gerhard@gt-priv.m.eunet.de (Gerhard Traeger) Crossposted-To: comp.os.ms-windows.nt.setup.misc,comp.os.ms-windows.nt.admin.misc Subject: Re: Windows NT 'stripe set' on MBR? Date: 1 Jan 1996 08:02:37 GMT in his message <4c3t7l$3bd@thokk.cs.cornell.edu> mdw@cs.cornell.edu (Matt Welsh) writes: >Linux is also installed on this system, and we have found that when the >LILO boot loader is installed on the MBR of the boot drive, NT loses >information about the stripe sets. lilo makes a copy of the original MBR. so to start NT, first start a small program copying back NT's MBR. when stoping NT, restore the lilo MBR. of course this is a mess, but that's the penalty for using MS stuff:-> gerhar ------------------------------ From: johnsos@holmes.ece.orst.edu (Scott Johnson) Subject: Re: WANTED: A secure kernel `unlink' function. Date: 1 Jan 1996 06:56:09 GMT In article <4c6sgh$5qt@ccnet3.ccnet.com>, Ronald F. Guilmette wrote: >I'd like to have patch for the kernel... a relatively simple one. (If any >of you weekend hackers happen to have some time on your hands, maybe you >would consider implementing this.) > >In a nutshell, I'd like to have a kernel in which the `unlink' function >actually overwrites (e.g. with zeros) the file being unlinked. > >Seems to me that security-minded people who are operating multi-user >systems would appreciate this. > >Of course there are bound to be naysayers who will immediately scream >that this would make the system too slow... but I have a suggestion >which could possibly reduce that problem. > >Here's the idea.... > >Implement a kernel `unlink' function which either does or doesn't do >erasing depending upon the setting of some environment variable in >the calling process. Does the kernel have any concept of environment variables? I was under the impression (I could be wrong though), that they were a user-space thing. Even if it did, I would be hesitant about making a kernel function's behavior dependant on one. /sj/ Scott Johnson: S/W Engineer, Linux Kernel Hacker, and All-Around Computer Geek. Check out my new (but unfinished) Web page--http://www.ece.orst.edu/~johnsos -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- History shows again and again, how nature points out the folly of men... -Blue Oyster Cult ------------------------------ From: Mark Shadley Subject: Re: modprobe.c compile error Date: Mon, 1 Jan 1996 03:37:44 -0800 In article: , Peter Mutsaers wrote: >>> On 30 Dec 1995 05:15:18 GMT, shadcat@catcher.com (Mark Shadley) >>> said: > > MS> In article: <4c1vt4$t92@ngriffin.itc.gu.edu.au>, > MS> Tony Nugent wrote: > > >> From depmod/modprobe.c... > >> > >> static void verbose (const char *ctl, ...) > >> { > >> if (verbose_flag){ > >> va_list list; > >> va_start (list,ctl); > >> vprintf (ctl,list); > >> va_end (list) > >> fflush(stdout); <-- line 22 > >> } > >> } > > MS> Notice something ';' missing from line 21? :) > >va_end used to be a macro that included the semicolon therefore in >much code you find it without a ;. Alas this seems te have changed in >gcc 2.7.2. I compiled it back in June, long before 2.7.2. I think I was using 2.6.3 then. maybe 2.5.8. Mark Mark Shadley pgp id:C596C1F9 - fp:B60A796CAF6769B9 0688C30B390A0A66 - on servers ------------------------------ From: rene@via.nl (R. Janssen) Subject: kernel message : TIOLINUX (0/8/9) ioctl gone - use /dev/vcs . Anybody ? Date: Mon, 01 Jan 1996 11:47:19 GMT Reply-To: rene@via.nl Hi, I just wanted to know what this message means : kernel message : TIOLINUX (0/8/9) ioctl gone - use /dev/vcs . I get it when I startup midnight commander. Sometimes the virtual console hangs sometimes it doesnt. thanks, Rene Janssen. ------------------------------ From: rob@pe1chl.ampr.org (Rob Janssen) Subject: Re: applying kernel patches Reply-To: pe1chl@wab-tis.rabobank.nl Date: Mon, 1 Jan 1996 09:59:32 GMT In <4c75kq$fn4@s-cwis.unomaha.edu> tyketto@corona.unomaha.edu (Brad Littlejohn) writes: > [script deleted] > Why do that, when it would be easier to grab the whole source? Yes, >you could go from 50, apply the patch to get 51, and so on to 52, but why >would you want to compile the kernel to have to compile it again? Seems like >too much to do. Why not just jump straight from 1.3.48 to 1.3.52? Because not everyone has T1 access to the Internet. It takes a lot of time to FTP the entire kernel source over a modem connection, and a patch will be much quicker to get. Besides, you *don't need* to compile every step if you don't want to. Just apply a few patches and compile the end result, if that is what you want. Rob -- +------------------------------------+--------------------------------------+ | Rob Janssen rob@knoware.nl | BBS: +31-302870036 (2300-0730 local) | | AMPRnet: rob@pe1chl.ampr.org | AX.25 BBS: PE1CHL@PI8WNO.#UTR.NLD.EU | +------------------------------------+--------------------------------------+ ------------------------------ ** 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 ******************************