Linux-Misc Digest #334, Volume #19 Sat, 6 Mar 99 15:13:11 EST
Contents:
Re: More bad news for NT (Harry)
Re: Best Free Unix? (why FreeBSD?) (Justin Murdock)
Re: Can Linux use 36-bit Xeon addressing? (Robert Krawitz)
Re: Oh..me so dumb..me need help big time! (Dan Srebnick)
Re: OS with a seamless object model (Roger Espel Llima)
Re: Public license question (jik-)
ADI monitor help ([EMAIL PROTECTED])
Re: Can Linux use 36-bit Xeon addressing? (Robert Krawitz)
Re: Beowulf Anyone? (Oscar Stiffelman)
Re: Can Linux use 36-bit Xeon addressing? (Robert Krawitz)
Re: Best Free Unix? (why FreeBSD?) (Roger Espel Llima)
Re: equivalent of edit? (Marco Tephlant)
Re: Public license question (Barry Margolin)
mSQL Installation ([EMAIL PROTECTED])
Re: Help: Newbie doesn't know where to start with GNOME! (Ivan Bilenkey)
----------------------------------------------------------------------------
From: Harry <[EMAIL PROTECTED]>
Subject: Re: More bad news for NT
Date: Mon, 01 Mar 1999 03:45:41 -0500
> Not me. I started with JCL. <
I used to dream of JCL. Started with binary machine code. <
------------------------------
From: Justin Murdock <[EMAIL PROTECTED]>
Crossposted-To:
comp.unix.questions,comp.unix.advocacy,comp.unix.misc,comp.unix.bsd.freebsd.misc
Subject: Re: Best Free Unix? (why FreeBSD?)
Date: 05 Mar 1999 15:05:52 +0000
>>>>> "DrBoom" == DrBoom <[EMAIL PROTECTED]> writes:
DrBoom> C Lamb wrote:
>> Remember, all the world isn't the US, we don't get free local
>> calls.
DrBoom> No offense (lotta that going around -- strange for this thread
DrBoom> :-) ), but most(?) newsreaders have an option to download just
DrBoom> the headers if it worries you that much. You can then
DrBoom> disconnect and select the articles you want to read,
DrBoom> reconnect, and get them.
We also have minimum call charges...
--
"This 'telephone' has too many shortcomings to be seriously considered as a
means of communication." --Western Union internal memo,1876
"[We] are too possessive for this CERN hypertext thing to catch on" --me,1990
------------------------------
From: Robert Krawitz <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux,comp.os.linux.hardware
Subject: Re: Can Linux use 36-bit Xeon addressing?
Date: 06 Mar 1999 14:00:13 -0500
Johan Kullstam <[EMAIL PROTECTED]> writes:
> Robert Krawitz <[EMAIL PROTECTED]> writes:
> > That interpretation is simply incorrect. Virtual addresses (which are
> > the only kind that normal instructions ever deal with in protected
> > mode) are 32 bits wide, just as in all x86 processors from the 80386
> > on up. The processor (in hardware, by referring to the page tables)
> > translates these virtual addresses into physical memory addresses.
> > It's immaterial how wide the physical address bus is. The physical
> > address bus could be 20 bits wide (not that I'd care to use such a
> > machine), or 32 bits wide, or 40 bits wide. The kernel sets up the
> > mapping between virtual addresses and physical memory; the processor
> > actually performs the mapping in hardware, and the user code never
> > knows the difference.
>
> riddle me this, how can i have more than 4 GB of data accessible from
> one program?
You can't (via direct memory addressing), but suppose I want to have
several dozen memory-hungry processes all running concurrently? This
is not normally the case for desktop systems, and is seldom a problem
for http servers and such, but for heavy commercial data processing,
with parallel databases, it's a perfectly normal state of affairs.
Folks, the issue is not the single-process VM limit, it's the total
amount of memory available to all processes in the system!
--
Robert Krawitz <[EMAIL PROTECTED]> http://www.tiac.net/users/rlk/
Tall Clubs International -- http://www.tall.org/ or 1-888-IM-TALL-2
Member of the League for Programming Freedom -- mail [EMAIL PROTECTED]
"Linux doesn't dictate how I work, I dictate how Linux works."
--Eric Crampton
------------------------------
From: Dan Srebnick <[EMAIL PROTECTED]>
Subject: Re: Oh..me so dumb..me need help big time!
Date: 6 Mar 1999 09:07:07 -0500
K Lee <[EMAIL PROTECTED]> wrote:
> So, is there anyway to resolve this? I know, I removed glibc! That was
> soooo stupid! But is there a fix for this? I'd appreciate any input.
> Thank you very much in advance.
Boot off your rescue disk. Mount your root partition somewhere, I believe
Redhat give you a tmp directory to mount partitions. You can then copy
the libc*.so files from the live/lib directory of your cdrom to the
tmp/yourmountpoint/lib directory, and that should be good enough to start.
Then once you reboot, you can try a rpm -i glibc to put it all back as it
was.
------------------------------
From: [EMAIL PROTECTED] (Roger Espel Llima)
Crossposted-To: gnu.misc.discuss,comp.os.linux.advocacy
Subject: Re: OS with a seamless object model
Date: 6 Mar 1999 00:27:34 GMT
In article <[EMAIL PROTECTED]>,
Craig Kelley <[EMAIL PROTECTED]> wrote:
>Just because you rename a file to 'persistent object' doesn't change
>the fact that a meaningful token must name it and that that token
>cannot be used by others.
That's a bit like saying that all variables must be global, because no
other thing could possibly exist...
(I know, I'm playing both sides of the argument ;)
--
:r .sig
------------------------------
From: jik- <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux,gnu.misc.discuss
Subject: Re: Public license question
Date: Fri, 05 Mar 1999 15:25:17 -0800
> From the previous discussions on this thread, it doesn't seem as though
> copyright law (as currently written) would consider the program to be a
> derivative work of readline if they were dynamically linked (regardless
> of the existance of a clone lib). Now, if RMS and others think this
> is not in the "spirit of copyright law", then they should lobby to get
> copyright law changed. But until that time, this loophole seems to be
> perfectly legal.
RMS is wrong about the 'spirit' of the copyright. Copyright was created
for the pourpose of inovation...and here is were it really goes away
from his ideas...
The thought was, that people would be more inclined to come up with new
ideas if they held certain ownerships of them. If they cold hold onto
these ideas as being thier own and noone elses for a term, they could
make money off of thier idea and so they would me more inclined to want
to. If you thought you could make a few bucks off of something you
thought up in your head, you might want to go through the pain in the
ass of actually publishing (or whatever) your new idea....much less so
if you knew someone would just come along with more resources and take
it away.
Then more people would see your idea and get new ideas of thier
own,...they in turn would publish thier new idea etc..etc...until you
get here, were we are today. Just because these new ideas were based on
your own, does not mean you now have ownership over everything someone
throught up based on your idea!!! NO....that is not, nor has it ever
been in the 'spirit' of IP. They can even site your work in thier own
to prove thier own ideas and were they came from.
That is the true nature of a copyright...progress through the free trade
of ideas. That means making money off of them too....
BUT, they did not want to go as far as to stifle progress....which is
why you have all these terms like 'fair use', and why copyright expires
after a time. Even patents run out.
The distribution of the library is a given,...IP laws apply. But wether
or not you can distribute code which happens to link dynamicly (not even
sharing code) would not in my mind constitute a derivative work,..only
fair use...which is ok.
I will have to find that gopher site again and read the parts on
software to be sure.
------------------------------
From: [EMAIL PROTECTED]
Subject: ADI monitor help
Date: Sat, 06 Mar 1999 18:30:55 GMT
I have a ADI microscan 5+ and am having trouble setting the screen
resolution. I have a matrox millenium G200 16MB SGRAM but the video
card has no trouble. It it my monitor. It is displayed way to big. I
cannot see even half the screen. It looks like 200x200. I love linux
and xwindows but this is making me not run linux. I have tried
everything. So is there some special tricks I can try with this
monitor. I have run xconfigurator a few hundred times trying
everything but no luck. Any help would be greatly appreciated.
Thanks in advance.
Dennis
------------------------------
From: Robert Krawitz <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux,comp.os.linux.hardware
Subject: Re: Can Linux use 36-bit Xeon addressing?
Date: 06 Mar 1999 14:05:28 -0500
"David A. Frantz" <[EMAIL PROTECTED]> writes:
> Again this depends on the application. If there is a need to address more
> than 4GB of memory then the system and libraries must support that.
If there is a need for a single process to directly address more than
4 GB of memory, this is true. However, a single application may
consist of many processes, each of which requires less memory than
this but in toto requires far more.
For examples of such applications, please see:
http://www.torrent.com/press/intelopenworld.htm
http://www.torrent.com/press/Rudin.htm
DISCLAIMER: I work for Torrent Systems, Inc., but this post reflects
my personal opinions and not those of Torrent Systems.
--
Robert Krawitz <[EMAIL PROTECTED]> http://www.tiac.net/users/rlk/
Tall Clubs International -- http://www.tall.org/ or 1-888-IM-TALL-2
Member of the League for Programming Freedom -- mail [EMAIL PROTECTED]
"Linux doesn't dictate how I work, I dictate how Linux works."
--Eric Crampton
------------------------------
From: Oscar Stiffelman <[EMAIL PROTECTED]>
Subject: Re: Beowulf Anyone?
Date: 6 Mar 1999 19:31:58 GMT
A.R.Allen wrote:
>
> Is anyone here working on constructing a Beowulf System using Red Hat and
> the Beowulf software?
>
>
We are building a 1000 node beowulf using Pentium III's and (normal)
RedHat (not the Beowulf software).
-- Oscar Stiffelman
Stanford University Computer Science Department
Genetic Programming, Inc.
================== Posted via SearchLinux ==================
http://www.searchlinux.com
------------------------------
From: Robert Krawitz <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux,comp.os.linux.hardware
Subject: Re: Can Linux use 36-bit Xeon addressing?
Date: 06 Mar 1999 14:28:42 -0500
(Responding to several posts in one response)
Johan Kullstam <[EMAIL PROTECTED]> writes:
>
> "David A. Frantz" <[EMAIL PROTECTED]> writes:
>
> > Johan Kullstam wrote in message ...
> > >Robert Krawitz <[EMAIL PROTECTED]> writes:
> > >
> > >> 2 GB RAM is a satisfactory virtual address space for a single process
> > >> for most purposes, but 1 or 2 GB RAM is not a satisfactory upper limit
> > >> on RAM today.
> > >
> > >but these are not `most purposes'. the big ram user will almost
> > >certainly need a shitload of ram for *one* process. otherwise, you'd
> > >just buy more machines and run one of the big (but not humongous)
> > >processes on each.
The first statement (by Johan Kullstam) is flat out wrong, and I've
posted some references to just such applications. The second
statement implicitly assumes that the overall work is embarrassingly
parallel or close to it.
[For those not familiar with the term, "embarrassingly parallel" means
that the data can be broken up into equal chunks with no reference to
any of the other chunks. Therefore, no communication of data is
required. For example, matrix addition is embarrassingly parallel.
It is done by adding the corresponding elements in each of the two
matrices to produce the sum. I'm not certain where this term was
coined, but it was in use at least 10 years ago, when I was at
Thinking Machines Corp.]
>
> i was responding to someone who thought that you might need more than
> 4 GB of RAM yet no single process needed more than 4 GB (in order to
> avoid `far' pointers). if you need the BIG RAM, it is because you
> have ONE task which requires it. 7 tasks needing 1 GB each would be
> better served running one after another in series on a smaller machine
> or spread over several smaller boxes.
Oh, really? What about if my job consists of applying several
processing operations in turn to a series of records? Suppose several
of those operations require random-access lookups against very large
lookup tables? If my system is an SMP or NUMA, I lose big time by
sequencing the processing steps, because I can only use one
processor. This kind of job can really take advantage of an SMP, but
it eats memory like a pig. If I'm using an RDBMS such as Oracle, the
database will chew memory also if I have to do big joins. And
databases such as Oracle and Informix and DB/2 are quite capable of
taking advantage of multiprocessing capability.
I'm really not interested in arguing whether or not QCD or finite
element analysis should be done on Intel boxes. What I'm talking
about is something very different.
--
Robert Krawitz <[EMAIL PROTECTED]> http://www.tiac.net/users/rlk/
Tall Clubs International -- http://www.tall.org/ or 1-888-IM-TALL-2
Member of the League for Programming Freedom -- mail [EMAIL PROTECTED]
"Linux doesn't dictate how I work, I dictate how Linux works."
--Eric Crampton
------------------------------
From: [EMAIL PROTECTED] (Roger Espel Llima)
Crossposted-To: gnu.misc.discuss,comp.unix.advocacy,comp.unix.bsd.freebsd.misc
Subject: Re: Best Free Unix? (why FreeBSD?)
Date: 6 Mar 1999 00:37:55 GMT
In article <7blp98$jjs$[EMAIL PROTECTED]>,
C Lamb <[EMAIL PROTECTED]> wrote:
>Alexander Viro ([EMAIL PROTECTED]) wrote:
>: In article <nYjD2.948$[EMAIL PROTECTED]>,
>: Steven Michael ROBBINS <[EMAIL PROTECTED]> wrote:
>: >
>: >Dollar
>: >
>: >The origin of the almighty dollar is in what is now the Czech Republic.
>: >In 1519, a silver mine near the town of Joachimstal (literally "Joachim's
>: >valley," from the German Tal, meaning valley) began minting a silver coin
>: ^^^^^^ dale.
>
>Congratulations, Mr Viro. You win our star prize for most useless
>correction of the week, by replacing a word with its synonym!
The point was that "dale" and "Tal" are actually cognates.
--
Roger Espel Llima, [EMAIL PROTECTED]
http://www.eleves.ens.fr:8080/home/espel/index.html
------------------------------
From: Marco Tephlant <[EMAIL PROTECTED]>
Subject: Re: equivalent of edit?
Date: Sat, 06 Mar 1999 19:47:22 +0000
Reply-To: [EMAIL PROTECTED]
Dan Nguyen wrote:
> Mr. Tinkertrain <[EMAIL PROTECTED]> wrote:
> : use pico or vi. in x11 use emacs.
>
> Pico blows. Use vi or emacs.
>
> --
> Dan Nguyen | There is only one happiness in
> [EMAIL PROTECTED] | life, to love and be loved.
> http://www.cse.msu.edu/~nguyend7 | -George Sand
Pico rules and is nice like pine.
--
Marco
------------------------------
From: Barry Margolin <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux,gnu.misc.discuss
Subject: Re: Public license question
Date: Sat, 06 Mar 1999 00:54:03 GMT
In article <[EMAIL PROTECTED]>,
NF Stevens <[EMAIL PROTECTED]> wrote:
>Let's take a slightly more tenuous link between the GPL work and another
>program. What if I wrote a proprietory program which acts as a wrapper
>to gcc, allowing you to specify source files and compilation options
>using a gui. My program uses the options supplied and then exec's gcc.
>To me this seems little different from the dynamic linking case. It
>further fulfils the condition that my program would be useless without
>gcc; yet I very much doubt anyone could claim that it was a "derived
>work" of gcc.
Based on what RMS has said about plug-ins, I think he *would*.
--
Barry Margolin, [EMAIL PROTECTED]
GTE Internetworking, Powered by BBN, Burlington, MA
*** DON'T SEND TECHNICAL QUESTIONS DIRECTLY TO ME, post them to newsgroups.
Please DON'T copy followups to me -- I'll assume it wasn't posted to the group.
------------------------------
From: [EMAIL PROTECTED]
Subject: mSQL Installation
Date: Sat, 06 Mar 1999 18:51:01 GMT
While I am an experieced computer user, I am new to Linux. I am trying to
install mSQL on my web site that is hosted by a company using
linux servers. When I try to install, I keep getting messages as such:
./makegen/makegen: /bin/uname: Permission denied
./makegen/makegen: test: =: unary operator expected
./makegen/makegen: test: =: unary operator expected
This mysterious direcotry seems to be the root of all my problems. If I ignore
all the error messages (because the install procedure can't find/access this
directory) and try to run the make install command, I don't seem to be getting
any output in my directory (surprise?).
Does anyone know what this directory is? do I have to have it? what is suppose
to be in it?
Any help would be appreciated.
============= Posted via Deja News, The Discussion Network ============
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
------------------------------
Date: Sat, 06 Mar 1999 19:45:14 +0000
From: Ivan Bilenkey <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.linux,comp.os.linux.help,comp.os.linux.questions,comp.os.linux.setup,comp.os.linux.x
Subject: Re: Help: Newbie doesn't know where to start with GNOME!
Plissken wrote:
> On Fri, 05 Mar 1999 21:08:05 GMT, [EMAIL PROTECTED] (WARGY) wrote:
>
> >
> >I've just installed Slackware 3.6 and have X running fine. I now want
> >to install that cool desktop thing called GNOME to help me feel at
> >home in Linux/X. I've been to www.gnome.org but I was gob-smacked
> >completely when it came to installation. Don't get me wrong but there
> >were so many files there and I didn't have a clue what I was meant to
> >download.
> >
> >Also, does anyone know why these Linux programs/utilities are so
> >complicated to install? I would've thought it was just one file that
> >did installation automatically... ok it's Linux but at least let us
> >Microsoft users try the alternative with a bit of ease. :-)
> >
> >Anyway, could some please guide me with the installation of GNOME (or
> >even KDE, if that's any better)? Thanks.
> >
> >WARGY
>
> Warg,
>
> I've tried to install GNOME as well, with no luck (I'm also a newbie),
> BUT, KDE installed without a hitch. I just followed the instructions
> from KDE's install guide and I was up and running. With GNOME, I just
> got tons of failed dependencies. I then installed with "--nodeps"
> which allowed the RPMs to install, but never found a file called
> .Xclients, .xinitrc, or any of the others listed on the GNOME
> install-guide, to modify. When I just tried running "exec
> gnome-session" on it, I got booted to a login prompt. Seems weird to
> me. I have X running fine with AfterStep and the other default Redhat
> WMs, and got KDE to install, but GNOME is a no-go.
>
> Plissken
To install gnome 1.0 :
This is what I've done:
1. get all rpm packages except those marked 'devel' (I assume you' you're
not interested in developing)
1.5 put them all in a new temporary dir (eg. mkdir gnome)
2. # rpm -ivh glib-1.2.0-1.i386.rpm
3. # rpm -ivh gtk+-1.2.0-1.i386.rpm
4. # rpm -Uvh ORBit*
5. # rpm -Uvh au*
6. # rpm -Uvh esou*
7. # rpm -Uvh gnome-libs-1.0.1-1.i386.rpm
8. # rpm -Uvh gu*
9. # rpm -Uvh lib*
10. # rpm -Uvh gnom*
11. # rpm -Uvh x*
12. # rpm -Uvh GX*
13. # rpm -Uvh co*
14. # rpm -Uvh e*
15. # rpm -Uvh f*
16. # rpm -Uvh ...
well I don't have to type it all - you can find out yourself what depends
on what.
basically you have to use -ivh with glib and gtk+ because some other
programs depend on
their previous release that you have installed
As for the .Xclients if you don't have one - make it.
Pu the following into .Xclients:
exec /usr/local/bin/afterstep >~/.AfterStep-errors 2>&1 # only if your
afterstep is 1.6
gnome-session # or exec gnome-session - I'm not sure
And finally don't forget to make it executable:
chmod u+x .Xclients
Ivan.
------------------------------
** 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: [EMAIL PROTECTED]
You can send mail to the entire list (and comp.os.linux.misc) via:
Internet: [EMAIL PROTECTED]
Linux may be obtained via one of these FTP sites:
ftp.funet.fi pub/Linux
tsx-11.mit.edu pub/linux
sunsite.unc.edu pub/Linux
End of Linux-Misc Digest
******************************
