From: Digestifier To: Linux-Misc@senator-bedfellow.mit.edu Reply-To: Linux-Misc@senator-bedfellow.mit.edu Date: Thu, 2 Sep 93 21:13:11 EDT Subject: Linux-Misc Digest #86 Linux-Misc Digest #86, Volume #1 Thu, 2 Sep 93 21:13:11 EDT Contents: Windows Pop Quiz Re: NT versus Linux (Enrico Scotoni) ftp-rl (read line) for NET-2 linux? (Framstag) Re: Linux-compatible DikuMUD derivative (CircleMUD 2.10) (Naved A. Surve) Re: Low Cost SLS 1.03 on Diskette (Matt Welsh) minute patch to ncurses 1.8 (sorry!) (Zeyd M. Ben-Halim) Re: Low Cost SLS 1.03 on Diskette (Zhuo Er Lin) Re: Stacker-like Compression? (Warner Losh) Re: Checker, a gnu Purify subset (Warner Losh) Re: Checker, a gnu Purify subset (Todd Austin) Re: Windows Pop Quiz Re: NT versus Linux (Andy "Traff" Trafford) Re: Linux and Tcsh: Soooo Funny!! (A Joke) Re: Speed testing different versions of POV (David E. Wexelblat) Re: SLS considered harmful (wasRe: Bashing Peter MacDonald) (J. Michael Diehl) Re: ideas for FAQs and printed LDP books (J. Michael Diehl) RE: High speed modems & linux (Utz-Uwe Haus) Seyon thinks I'm permanently online (Brad Maxwell) ---------------------------------------------------------------------------- From: Enrico.Scotoni@p46.f520.n301.z2.fido.imp.com (Enrico Scotoni) Subject: Windows Pop Quiz Re: NT versus Linux Date: Wed, 01 Sep 1993 16:35:00 +0200 > I can't speak for the majority, but I personally like Windows > because the box, with documentation and all, is perfect for > propping my office door open. I have worried that someone > might steal it, but decided that in that case, it would be the > theive's problem to find a new use for it. This one is nice. It reminds me of the Dec Alpha PC with NT as OS that we have in our office: The only USEFUL application we found on that system was the CD-Player SW. But since everybody in the office was too lazy to switch CD's and nobody wanted to hear the same CD all day long, we changed back to the old radio and powered the Alpha down. Alpha PC with NT: The most expensive CD-Player in the world ;-) Enrico ------------------------------ From: orakel@rz.uni-ulm.de (Framstag) Subject: ftp-rl (read line) for NET-2 linux? Date: Thu, 2 Sep 1993 18:58:11 GMT I have a nice ftp-client called "ftp-rl" (rl stands for gnu readline), which won't compile anymore under NET-2 Linux In file included from ftp.c:33: /usr/include/netinet/ip.h:43: dev.h: No such file or directory /usr/include/netinet/ip.h:44: linux/sock_ioctl.h: No such file or directory /usr/include/netinet/ip.h:159: eth.h: No such file or directory Is there a newer version of this ftp-client, which will work with the new NET-2? \ Ulli 'Framstag' Horlacher \ psi%(0262)45050260103::orakel \ \ Student consultant VAX/VMS and networks \ orakel@dulruu51.bitnet \ \ Rechenzentrum Universitaet Ulm Germany \ orakel@rz.uni-ulm.de \ \ G=ulli;S=horlacher;OU=rz;P=uni-ulm;A=dbp;C=de \ rzmain::orakel \ \ "With mainframes you just can do nonsense, with PCs not even that." \ ------------------------------ From: surve_n@zebra.cs.jhu.edu (Naved A. Surve) Subject: Re: Linux-compatible DikuMUD derivative (CircleMUD 2.10) Date: Thu, 2 Sep 1993 19:28:19 GMT In article <1993Sep1.203538.359@knobel.knirsch.de> andreas@knobel.knirsch.de (Andreas Klemm) writes: >jelson@circle.cs.jhu.edu (Jeremy Elson) writes: > >>I just uploaded the latest version of CircleMUD (version 2.10) to >>tsx-11.mit.edu. Version 2.10 runs under Linux, which is why I'm >>posting to this newsgroup. > >Would be nice, if everybody could give a "one sentence description" >what the program xyz does or for what purposes it was made. > I totally agree, but to properly describe what CircleMUD is defies "one sentence description". Briefly, it is a multi-user network game. For more info, you probably would want to read the FAQ on MUDS and mudding, which is periodically posted to news.answers and rec.games.mud.announce Hope that helps -Naved -- =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= Naved A. Surve | The Johns Hopkins University surve_n@server.cs.jhu.edu | Baltimore, Maryland -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- ------------------------------ From: mdw@sunSITE.unc.edu (Matt Welsh) Subject: Re: Low Cost SLS 1.03 on Diskette Date: 2 Sep 1993 20:45:15 GMT In article , Andrew J. Piziali wrote: >Michael Johnston asks: Actually, Matt Welsh asks. The parenthetical remark at the top of the posting was added by myself, because I wanted to get feedback from the c.o.l.a readershuip. For the record, I support low-cost distributions of SLS etc. on diskette, and I think that Michael Johnston's (not to be confused with Michael Johnson's) effort is a great one. That being said, I wanted to see how the readers of c.o.l.a felt about the issue in general. I never meant to bash on Micahel's article or attempts: If I had, I wouldn't have approved the article. mdw -- Send submissions for comp.os.linux.announce to: linux-announce@tc.cornell.edu ------------------------------ From: zmbenhal@netcom.com (Zeyd M. Ben-Halim) Subject: minute patch to ncurses 1.8 (sorry!) Date: Thu, 2 Sep 1993 21:08:10 GMT *** lib_move.c~ Thu Sep 2 14:02:37 1993 --- lib_move.c Thu Sep 2 14:02:17 1993 *************** *** 12,18 **** #include "curses.priv.h" ! int wmove(WINDOW *win, int y, x) { #ifdef TRACE if (_tracing) --- 12,18 ---- #include "curses.priv.h" ! int wmove(WINDOW *win, int y, int x) { #ifdef TRACE if (_tracing) -- Zeyd M. Ben-Halim zmbenhal@netcom.com 10479 1/4 Santa Monica Blvd, LA, CA, 90025 (310) 470-0281 ------------------------------ From: umlin000@ccu.umanitoba.ca (Zhuo Er Lin) Subject: Re: Low Cost SLS 1.03 on Diskette Date: Thu, 2 Sep 1993 20:57:43 GMT In <263p41$4i9@samba.oit.unc.edu> mjohnsto@ditdah.morse.net (Michael R. Johnston) writes: I would say given their charges $1.25/disk is fair. That may be considered the cost of diskette and copying. I wanted upload Linux SLS series to local BBS to promote Linux, but considering it is not practical for a BBS user download 30 disk with 1 hour/day access time. May be I should try upload a working diskette demo. But I suggest someone make such a demo verison and put it in sunsite or tsx-11. So others can use it to promote Linux. For example, you can bring the diskette to local user group meeting, etc.... I believe a larger user base will give Linux a better future. So some will prefer pay the fees. Though this way they hard to keep track of the latest version. And some will prefer this rather than spend hours of time downloading files. Just my $0.02. -- ======================================================================== | Eric Lin Voice: (204) 783-2884 | | Computer Engineering FAX Modem: (204) 783-2884 | | University of Manitoba Internet: Umlin000@cc.Umanitoba.CA | ------------------------------ From: imp@boulder.parcplace.com (Warner Losh) Subject: Re: Stacker-like Compression? Date: Tue, 31 Aug 1993 06:05:12 GMT kfogel@colossus.cs.oberlin.edu (Karl Fogel) writes: >Hey, would it be difficult to add a feature to the kernel >whereby gunzip is automatically called on any compressed file when >access to it is requested (and gzip called when it's time to write?) I had this thought a while ago. There are lots of problems. The first one is that g*zip is not reentrant, which means that you could only have one stream open at any time, or worse. There are also lots of performance issues that would make it hard to do with any reasonable performance. Warner -- Warner Losh imp@boulder.parcplace.COM ParcPlace Boulder I've almost finished my brute force solution to subtlety. ------------------------------ From: imp@boulder.parcplace.com (Warner Losh) Subject: Re: Checker, a gnu Purify subset Date: Wed, 1 Sep 1993 17:13:44 GMT In article <262ajf$16ns@rs6000.bham.ac.uk> G.PICKERING@bham.ac.uk writes: >I do not wish to put a downer on what you have done because I think >yours is an outstanding product in its own right, but please be careful >in comparing, or making claims in respect to purify. If it offers >all that Purify does then that is ok, but if it is a subset then >say "Checker will be a free and GNU implementation of a subset >of the Purify(tm) product". Agreed. If you can detect things like void foo() { char *bar; *bar++ = '\0'; } then it will be purify like. Otherwise, it is just a very good malloc debugger. Purify does check for memory leaks, but it also checks for lots of other things, like parameters to functions that are overlapping, the above code, etc. Warner -- Warner Losh imp@boulder.parcplace.COM ParcPlace Boulder I've almost finished my brute force solution to subtlety. ------------------------------ From: austin@aura.cs.wisc.edu (Todd Austin) Subject: Re: Checker, a gnu Purify subset Date: 2 Sep 93 23:12:01 GMT In article <262ajf$16ns@rs6000.bham.ac.uk>, pickerig@eee.bham.ac.uk (Guy Pickering) writes: |> SBPM Marc GINGOLD (marc@david.saclay.cea.fr) wrote: |> : Checker V0.1e-7 (c) 1993 Tristan Gingold |> : Checker will be a free and GNU implementation of Purify(tm). |> |> While I am not connected with the Purify product (other than a user) |> I would like to point out that Purify is not just a replacement |> for malloc(). It works by inserting extra code into the object |> files and relinks them (it actually takes copies of the files first). |> In doing this it offers far more than the facilities you describe. |> |> I do not wish to put a downer on what you have done because I think |> yours is an outstanding product in its own right, but please be careful |> in comparing, or making claims in respect to purify. If it offers |> all that Purify does then that is ok, but if it is a subset then |> say "Checker will be a free and GNU implementation of a subset |> of the Purify(tm) product". |> |> Guy |> Agreed... Purify offers significantly more functionality, specifically: - it puts "red zone"'s around all (?) variables so when you reference immediately past the end of an array it will be immediately detected - it knows if you access unallocated storage, so to help in finding references to freed storage, it "ages" freed storage, keeping them in the heap free list for an extended period of time And like Checker, - it uses conservative collection to (try) to detect storage leaks, and it also attaches partial call stack information to allocations to aid in debugging Purify probably does more stuff, I've never actually seen the product, this is what they describe in the 1992 Usenix paper, "Purify: Fast Detection of Memory Leaks and Access Errors" The primary disadvantages of Purify is that its not portable (it rewrites object code) and it does *not* provide full coverage for memory access errors, e.g., its possible to dereference a pointer past the end of a variable, into the next without detection. Hopefully, I'll release Safe-C next month -- it solves both of the aforementioned problems. Its a portable package based on GCC which provides complete coverage for all memory access errors, both spatial and temporal (except in the presence of casts from non-pointer values to pointer values). The software (developed under Linux) is complete, I'm just holding out (not easy) on posting it until I have published a report on the methodology. When that occurs I'll make it available with the report. Be safe, -Todd "disjoint from Purify, Checker; but attached to Safe-C" Austin -- %% Todd Austin, austin@cs.wisc.edu %% Department of Computer Sciences, University of Wisconsin -- Madison %% 1210 West Dayton, Madison, WI 53706 ------------------------------ From: traff@panix.com (Andy "Traff" Trafford) Subject: Re: Windows Pop Quiz Re: NT versus Linux Date: 2 Sep 1993 19:18:37 -0400 In article <746952814.AA05966@gccs.fido.imp> Enrico.Scotoni@p46.f520.n301.z2.fido.imp.com (Enrico Scotoni) writes: > >Alpha PC with NT: The most expensive CD-Player in the world ;-) > >Enrico > Nowhere near the most expensive (buy yourself a high-end audio magazine sometime), but certainly the most bug-ridden :) Traff ------------------------------ From: paul@pbunyk.physics.sunysb.edu () Crossposted-To: comp.os.linux Subject: Re: Linux and Tcsh: Soooo Funny!! (A Joke) Date: 2 Sep 1993 22:44:24 GMT In article <1993Aug30.115438.1@vax.sonoma.edu>, burd@vax.sonoma.edu writes: |> In article <13038@obelix.icce.rug.nl>, karel@icce.rug.nl () writes: |> > Hi Linuxers: |> > |> > Just a semi-rancid joke to help you pass the day. You need the tcsh |> > to run this; it won't work for bash users. So, if you do use bash: |> > just startup a (temporary) tcsh and try it out. |> > |> > At the prompt, type: |> > %blow |> > The % is necessary. Then watch for the output.... |> > [Har har.. bangs his head against the wall whilst laughing madly.] |> |> he-he-he. I have a list of such jokes like this around here somewhere. If I |> find it I'll post them. |> |> Erik |> In some Unixes on 'make love' make answers "I don't know how to make love..." - unfortunately, GNU make adds some unnecessary punctuation. Sorry for non-linux posting... Paul ------------------------------ From: dwex@mtgzfs3.att.com (David E. Wexelblat) Crossposted-To: comp.graphics Subject: Re: Speed testing different versions of POV Date: 2 Sep 93 19:52:26 GMT In article <1993Sep2.121709.6788@mnemosyne.cs.du.edu> jepler@nyx.cs.du.edu (Jeff Epler) writes: > In article <263fob$4cv@panix.com> traff@panix.com (Andy "Traff" Trafford) writes: > > > >Background for Linux folks: There has been a lot of discussion in > >comp.graphics recently about 'Fastpov', which is a DOS version of POV > >compiled with the Watcom compiler. > > > > > >I decided to run off a few quick tests this morning to see how different > >versions of POV performed on my no-name 486/50 (8Mb ram). > >I used the standard POV benchmark test - ie rendering CHESS.POV to /dev/null > >with no antialiasing, no display, quality 9 etc etc etc. > > > >The results were as follows: > > > > > >POV version | Render time (320x200) | Chesstones (pixels/sec) > >------------------+-----------------------+------------------------- > >Standard DOS | 337 seconds | 189.91 > > | | > >Fast POV for DOS | 289 seconds | 221.45 > > | | > >POV for Linux | 281 seconds | 227.76 > > > > > >The Linux version of POV was compiled with standard cc - would gcc or > >any other compiler produce faster code? > > I think that on linux, cc and gcc are the same program. > > However, to get the maximum speed, '-m486' and '-O6' should be > specified. (The former gets better performance on 486s at the price > of executable size. Supposedly, the performance on a 386 is the > same. I'm a bit suspicious of this, but oh well.) This has been posted about a dozen times before, but... THERE IS NO DIFFERENCE BETWEEN -O2 and -On, FOR ANY n GREATER THAN 2! I would love to know where this piece of folklore came from. All it takes is reading a couple of files in the gcc to learn that there is absolutely no effect whatsoever. -O1 and -O2 are defined by gcc itself; any n greater than 2 is available to be defined by the various architecture support files. No such definitions exist for x86. Using -O6 by habit may lead to problems later if someone does define a -O3 optimization, and it turns out to not be a good general optimization. We did extensive testing of -m486 when building the very first binaries for XFree86. We needed to determine whether we had to build 386 and 486 binary kits. We ran tons of benchmarks, none of which showed any noticable difference in server performance. Outside of the kernel, you are not likely to find more complex code than the X server. Hence we use -m486 for all of our binary distributions. I would think the only place you might see a difference is a very old/cheap 386 with no cache. But then you deserve what you get anyhow. > > I'm not sure if specifying some of the -f flags on top of this (IE > -funroll-all-loops or the like.) will add more speed. In general, no. For specific pieces of code it can. > > (What was the default optimisation in the makefile? I think it had > -O2, but I'm not sure.) > > I assume that you had your linux machine (mostly) unloaded while you > executed the test? It might be nice to know what "time" had to say > about the POV process. This command would tell you how much of the > processor time was used for POV, discounting what other processes used. > (But if all you had going were some gettys and init, then it shouldn't > matter too much...) > > > -- > Jeff Epler jepler@nyx.cs.du.edu (Preferred) or bx304@cleveland.freenet.edu > ____ "Nuke the unborn gay whales" -- Never seen on a protest sign > \bi/ I have no time for petty theft, I have no time for sex, > \/ But I have time for what I like, And that is what is best. -- David Wexelblat (908) 957-5871 Fax: (908) 957-5305 AT&T Bell Laboratories, 200 Laurel Ave - 3F-428, Middletown, NJ 07748 XFree86 requests should be addressed to "Shining, flying, purple wolfhound, show me where you are." Yes, "Yours Is No Disgrace" ------------------------------ From: mdiehl@unm.edu (J. Michael Diehl) Subject: Re: SLS considered harmful (wasRe: Bashing Peter MacDonald) Date: 2 Sep 1993 18:07:14 -0600 In article <25u8er$5a1@samba.oit.unc.edu>, Matt Welsh wrote: > >On the side, I think that Peter's response is appropriate. I think he's >a good guy. If you don't like SLS, you should help to fix it or put out >your own release (it makes my job a lot easier if you just help SLS or >another exisiting release in some way... keeping up with dozens of separate >releases is starting to become difficult). If you have problems with >Peter's philsophy and design, nobody is forcing you to use SLS. And, Peter >shouldn't be held responsible for everything. > Ok, here is the the question that has always burned in my heart. I figured that as soon as I had a machine to run Linux on, I'd ask, but now seems as good a time as any. Just what is meant by a "distribution?" Why can't we just dload related programs as a "package?" We could get the boot disk, the root disk, and the compiler package, the X package, network package....etc. Then we could simply dload our favorite install script for each package and people would be supporting their own install-script-religion. ;^) This would be much simpler than fighting over entire directories of packages. Mike. ------------------------------ From: mdiehl@unm.edu (J. Michael Diehl) Subject: Re: ideas for FAQs and printed LDP books Date: 2 Sep 1993 18:23:19 -0600 In article <1993Aug31.005801.12876@hellgate.utah.edu>, Greg Alt wrote: >First, a suggestion about the FAQs... I installed Linux a couple weeks ago, >and the most confusing thing about all of the FAQs was trying to find out if >I had the most current one. It would be nice if every faq/infosheet/metafaq >had a last-revision date at the very top. Also, it would be nice if the one >main meta-faq had a list of all of the faqs with the most current dates for >each. This should be distributed in such a way that it is obvious that it >should be read first (e.g. on the ftp sites, give it a title like >"read_first_meta_faq" or something like that. YES YES YES YES YES. If no one else wants the job, I'd bet I could be persuaded If no one has volunteered in, say, 15 days, start sending my info on whatever package you maintain. J. Michael Diehl ;^) |*The 2nd Amendment is there in case the mdiehl@triton.unm.edu | Government forgets about the 1st! Mike.Diehl@f29.n301.z1 |*God is a good Physicist, and an even .fidonet.org | better Mathematician. al945@cwns9.ins.cwru.edu|*I'm just looking for the opportunity to (505) 299-2282 (voice) | be Politicly Incorrect! Can we impeach him yet? |*Protected by 18 USC 2511 and 18 USC 2703. PGP Key = 7C06F1 = A6 27 E1 1D 5F B2 F2 F1 12 E7 53 2D 85 A2 10 5D ------------------------------ From: root@pse.panic.bln.sub.org (Utz-Uwe Haus) Subject: RE: High speed modems & linux Date: Thu, 2 Sep 1993 18:00:42 GMT From root Thu Sep 2 19:57:41 1993 remote from pse Received: by pse.panic.bln.sub.org (Smail3.1.28.1 #6) id m0oYIug-0001oVC; Thu, 2 Sep 93 19:57 MET DST Message-Id: Date: Thu, 2 Sep 93 19:57 MET DST From: root@pse.panic.bln.sub.org (Utz-Uwe Haus) Subject: Re: High speed modems & Linux Newsgroups: comp.os.linux.misc References: <25blk2$d7m@agate.berkeley.edu> <25ml55$nbt@news.bu.edu> <1993Aug29.071036.2740@jonh.wimsey.bc.ca> <25ntjf$6rc@news.bu.edu> In comp.os.linux.misc you write: >In article <1993Aug29.071036.2740@jonh.wimsey.bc.ca> jhenders@jonh.wimsey.bc.ca (John Henders) writes: >> >> UUCP is quite fast for me under linux. Here's the result of uurate >>for the last 24 hours here. This is with a USR dual standard 14,400 >Have you done anything to customize your uucp configuration? The >best rate I've seen here so far is about 1100CPS, but on average it >is much worse. Well, although my UUCP isn't really reliable with all the hosts I'd like to connect to (random errors, usually missing the \000 after the Shere=... handshake-opening sequence - anybody got a clue what's happening ?) I get an average of 1500cps using Taylor 1.04 (self-compiled), USRobotics Sportster 14.400 and 19.200 bps serial line speed using the e or f protocol on mostly compressed newsbatches (c.o.l-hirarchy :))! Just my cent's worth of thought... Utz-Uwe Haus (root%pse@chamber.in-berlin.de, thats the reliable feed or root@pse.panic.bln.sub.org - might get here some day...) -- sig :) -- sig :) ------------------------------ From: bcmaxwel@infonode.ingr.com (Brad Maxwell) Subject: Seyon thinks I'm permanently online Date: Fri, 3 Sep 1993 00:53:40 GMT Hey there, I'm using Seyon on my Gateway 4DX-33 with a Supra 9600 baud fax modem. Seyon seems to compile and run fine, but I can't use it to dial cause it thinks I'm online already and aborts the dial. All the control panel lights are on (except for ring, of course) all the time the modem's on. Just thought somebody might've seen it before. Thanks, Brad ------------------------------ ** 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-Misc-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux.misc) via: Internet: Linux-Misc@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-Misc Digest ******************************