Linux-Development-Sys Digest #585, Volume #6 Mon, 5 Apr 99 18:14:39 EDT
Contents:
Re: Problem compiling program with glibc2.0.6 (Juergen Heinzl)
Re: Clueless Users Are Bad for Debian - was Re: After Week 1 With Linux -- licking
wounds. (Jim Roberts)
Re: Programming tools for Linux/Unix: Editor, IDE, Frontend to GCC. (Stephan Schulz)
Re: CodeWarror for Linux (was: Re: Programming tools for ...) (Tim Triemstra)
Re: Help! sys_shmat (Kernelprogramming) (Andi Kleen)
Re: runnig more than one netscape window (Bill Anderson)
Re: Proposal: "Linux 2000 Platform" (Jeremy Crabtree)
----------------------------------------------------------------------------
From: [EMAIL PROTECTED] (Juergen Heinzl)
Subject: Re: Problem compiling program with glibc2.0.6
Date: Mon, 05 Apr 1999 17:37:34 GMT
In article <[EMAIL PROTECTED]>, Fabrice Bauthier wrote:
[...]
>Before installing the new kernel, i decided to install glibc2.0.6.
Since Andreas answered quite a lot already ...
[...]
>gcc -D_GNU_SOURCE -O3 -I. -I/usr/include -I/usr/i486-linuxlibc5/include
>-Wall -Wno-unused -DSYSV -c skill.c
>skill.c:71: warning: #warning Your header files are not
>standards-compliant!
>skill.c:75: warning: #warning Your header files are not
>standards-compliant!
>.....
You do not need -I/usr/include either and with the glibc the message
shows up, ignore it.
[...]
Regarding the ncurses library; since some older programmes
might want the ncurses-4.2 compiled against the libc5 you
are in trouble.
There is a pre1 version of the ncurses-5.0 library available
and it works fine so far. The advantage is there is no version
number clash 4.2 <-> 5.0 and it might be worth a try so. See
the File Watcher link on http://www.linuxhq.com/ for more.
Esp. if the bash uses the readline library and is linked with
the ncurses library you might not be able to boot anymore
else (assume you've /usr/i486-linuxlibc5/lib and /usr is
a separate partition ...).
Cheers,
Juergen
--
\ Real name : J�rgen Heinzl \ no flames /
\ EMail Private : [EMAIL PROTECTED] \ send money instead /
\ Phone Private : +44 181-332 0750 \ /
------------------------------
From: [EMAIL PROTECTED] (Jim Roberts)
Subject: Re: Clueless Users Are Bad for Debian - was Re: After Week 1 With Linux --
licking wounds.
Date: Sun, 04 Apr 1999 21:13:38 GMT
Got to agree with this. Those that are not willing to do the research
sould stay with a MS product. A product that has little choice or selection
availible within it.
Dejenews and other resources would answer 99% of the questions new users
have if they would just use their heads.
You should see some of the comments posted in the SUN/Solaris/SCO groups
since the MFG's have started to release free single user license for their
X86 Unix products.
One more rant, how many times do we have to hear about WINMODEMS.
I'm sure I'll take some flames here but.....
> As if to emphasise the points I made below, I just read this on
> slashdot......
>
> Clueless Users Are Bad For Debian
>
> Posted by Hemos on Wednesday March 31,
> @08:38AM EDT
> from the interesting-to-read dept.
> Helmholtz writes "I just got done reading a very
> interesting article about Debian and Clueless Users that
> appeared on
> the Debian Weekly News site. I think this article would be a very good
> thing for
> users of all distributions to read, as it touches upon what might become
> a very real problem.
> Now that the word "Linux" has been splashed around by such
> 'heavyweights'
> as CNN and NPR, everyone who want to be seen as a 'cool computer guy'
> is trying
> to get Linux up. This of course is done without any heed to the absolute
> requirement that some learning to occur. This is a concept that the
> Debian article holds at its core, I believe. "
>
> -------------------------------------------------------------------------------
>
> Sid Boyce wrote:
>>
>> John Myers wrote:
>> >
>> >
>> > Nor am I a "newbie" (except where linux is concerned). I have configured
>> > and maintained DOS, Windows, and NetWare systems for many years, and have
>> > at least a rudimentary understanding of C. After one very frustrating
>> > month with debian linux, I can't even get this box to send garbage to the
>> > printer... X-Windows is installed and working, but seems to be simply
>> > a "brighter" version of the command line...very impressive!
>> >
>> Perhaps you should get an easier distribution like RedHat or Suse where
>> printer configuration amongst other things is a cinch. If you think
>> Xwindows is simply that, you should see other people's desktops or go to
>> the KDE, GNOME, WX-Windows, Afterstep, fvwm2-95 or other homepages and
>> look at some screen shots, then rewrite that statement. It's getting so
>> crowded in that area that on a recent visit to our HQ in Sunnyvale,
>> Calif., I was sitting using a linux box and didn't know it for a couple
>> of days, nor did I recognise Linux on a Sparc box in our European HQ. If
>> I tell you that a MS Windows desktop looks very jaded by comparison, I'm
>> not kidding and you can check it out yourself.
>>
>> > I won't requote your entire message, but it did my heart good to read it.
>> > I have read this forum extensively, and as a result have been hesitant to
>> > submit my own basic questions. The standard response ...read the docs
>> > (/usr/doc/whatever/examples/examples/examples/what.the.#%&#$%!!!!), is
>> > somewhat insulting. I have read hundreds of pages of docs, FAQs, man pages
>> > and have found precious few answers to seemingly simple questions. The
>> > resultant conclusion from geekdom, of course, is that I must be a moron.
>> > I've been thought of and called worse, but this does not answer my
>> > questions, or help to get my system operating.
>> >
>> As suggested above, perhaps a distribution that does most of the work
>> for you is in order. I've seen many guys without your expansive
>> experience cope very well with Linux, many of them straight out of
>> DOS/Windows, so I don't know if that casts any aspertions your way. I
>> can tell you that recently a number of colleagues have installed either
>> RedHat or Suse and have commented on how easy the setup was, one told me
>> he had everything up and running including X, but had a bit of a tussle
>> with setting up PPP, but eventually got it going. There are some nice
>> tools for setting up PPP very easily if you know them. If you ask for
>> help and the question is specific, with pertinent details of what you
>> are seeing, then it's easy for anyone to help, too often we've seen a
>> "XXX don't work, I need help" appeal which leaves you quite cold, you
>> don't know anything of the environment or the specific problem, but you
>> are asked to conjure up a solution out of thin air. Personally, I've
>> offered considerable help to people over the years and my approach, once
>> I have something concrete to work on, is to tell the person exactly what
>> needs to be typed in to fix the problem and may be an example that's
>> very easy to follow. I have read many HOWTO's in my time and have found
>> that if followed to the letter logically, they are excellent, though a
>> few times when I've hit a dead end, a re-read of the HOWTO has often
>> showed that I missed some simple but critical point.
>>
>> > I was interested in a multi-tasking, multi-user operating system for use
>> > in my office. To accomplish real work. I am not interested in endlessly
>> > playing with a sophisticated "geek toy". I am willing to put in the time
>> > and effort required, so long as that investment remains reasonable and
>> > shows SOME return. To date debian. X-Windows, Samba have not met this
>> > requirement.
>> >
>> I'm not too au fait with debian, I've usually brewed my own or started
>> from a base system like slackware when a rogue motherboard wrecked my
>> filesystem. X-windows configuration tools are around that make the job
>> simplicity itself if you know just a few points gleaned from the manuals
>> that accompany your video card and monitor. I even have a friend who has
>> an old 14" monitor with no specs, but have done a little guess work
>> based on the age of the monitor and it works fine. You have perhaps not
>> heard of sambaconfig or kcmsamba (for KDE), linuxconf, or Webmin, though
>> for some of them you need X and for Webmin you need a browser also.
>>
>> > I have not given up on linux, yet, but if I have learned anything from
>> > this project it is that Mr. Gates needn't lose any sleep over the "linux
>> > threat"
>> >
>> > Still trying (for a while, anyway)
>> >
>> I find it difficult to see how you've become so confusticated (if not a
>> dictionary word, it's a colloquialism used here in the UK to signify
>> total loss of bearings).
>> On a daily basis, for WORK, with Linux ONLY, I use Netscape for mail
>> and Web browsing, x3270, Citrix Winframe client, acroread,
>> StarOffice/Applixware (for Word/Excel) and that's besides lots of other
>> things I do in the personal area. Whatever innovation the Corporation
>> has introduced, I've been able to fall in with on Linux.
>> As for Mr. Gates, we don't lose any sleep over him and we wish him many
>> peaceful nights with pleasant dreams, but that won't help him to produce
>> an OS that's not full of gremlins, nor applications that are not prone
>> to ad-infinitum attacks like Happy-99 and Melissa viruses, I'm sure the
>> guys that write them too have peaceful nights at the thought of a job
>> well done and wake up full of the joys of spring for a fresh set of
>> endeavours they know will also be successful and will make IT managers
>> more nervous than the FUD Gates&Co. throw at Linux. All it needs is for
>> some to do their homework on realising that Windows is not industrial
>> strength and the consequences will follow, or they can just take the
>> grief and losses as part of life's rich tapestry.
>>
>> > Thanks for listening
>> >
>> > John Myers
SNIP
--
Jim Roberts Never enough time!
[EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED] (Stephan Schulz)
Crossposted-To:
comp.os.linux.advocacy,comp.os.linux.development.apps,comp.os.linux.help,comp.unix.programmer
Subject: Re: Programming tools for Linux/Unix: Editor, IDE, Frontend to GCC.
Date: 5 Apr 1999 21:15:32 GMT
In article <7eajmo$[EMAIL PROTECTED]>,
Michael Davis <[EMAIL PROTECTED]> wrote:
>In article <7e0q1e$aef$[EMAIL PROTECTED]>,
>Stephan Schulz <[EMAIL PROTECTED]> wrote:
>>
>>If something wents wrong, it typically happens after 5 minutes of run
>>time, and on data strucures with millions of cells. Stepping through
>>this with a debugger is not pleasant. I'd rather print out most
>>important variables and invariants and use grep and emacs ediff mode
>>on the output to find out what's wrong. And sometimes only deep
>>meditation and sacrifices to the elder gods will help...
>
>For that kind of programming, you're right of course. I work on something
>much more mundane - accounting software for stockbrokers, but it's a big
>app... I find debuggers useful for finding out where my program segv's and
>stuff like that, but I agree that it's not the best tool when you have
>complicated data structures. For that stuff, I usually define a class
>member:
>void Dump( FILE * fp )
>and have a debug mode on my programs where I can switch on a feature
>to get it to print contents to a file - sounds like what you do.
Yes, that's somewhat similar. I have a variety of debug switches that
turn on protocolling for different parts of the program. However, the
more powerful and faster the prover becomes, the more difficult it is
to wade through all this output. So now I often just write some
additional functions that do check internal consistency and
assumptions (basically, assertions taking to a pretty high level) and
report on inconsistent or unexpected behaviour.
>So it depends, debuggers are great for some things, but not for
>everything, definitely not great for complicated data structures such
>as vectores of maps of vectors of ...
Right. I do use ddd if I have a lot of new code and something simple
happens, e.g. a seg fault, to find out which access caused it, and
where the wrong value was introduced. But this only works for simple
cases.
>>BTW, does someone else have these strange experience? Sometimes I am
>>looking for hours or days for a bug, without any apparent
>>progress. Then, when doing something completly unrelated (usually
>>sports), it goes BANG and I do not even need to check the code to know
>>what was wrong and how to fix it.
>
>Yes, that happens to me a lot (I don't watch sports though). This
>experience is well documented. Lots of scientists and engineers have
>this, and do things like take naps when they can't see a solution.
Oh, I don't watch sports at all (unless you count professional
wrestling as sports -- call me a pervert, will you?), but I do
sports. Swimming is good for developing new algorithms, aerobics is
good for finding bugs. Bicycling is good for getting killed by a
stupid car driver who does not know traffic rules ;-)
Bye,
Stephan
========================== It can be done! =================================
Please email me as [EMAIL PROTECTED] (Stephan Schulz)
============================================================================
------------------------------
From: Tim Triemstra <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.linux.advocacy,comp.os.linux.development.apps,comp.os.linux.help,comp.unix.programmer
Subject: Re: CodeWarror for Linux (was: Re: Programming tools for ...)
Date: Mon, 05 Apr 1999 16:25:05 -0400
Michael Powe wrote:
> >>>>> "Tim" == Tim Triemstra <[EMAIL PROTECTED]> writes:
>
> Tim> People that think using only GNU tools is somehow "right" or
> Tim> "morally sound" are simply living in a world different from
> Tim> most. Computers are tools, not moral or ethical decisions.
>
> This is just corporate-speak. "I'll work for anyone if the price is
> right -- I don't care what they want me to write." "The business of
> business is business -- not making moral or ethical decisions." How
> you use your computer is a moral decision; what software you put on it
> and use is a moral decision. To say otherwise is just silly.
This is fanatic speak, to say otherwise is just silly :) Take 1000
random people and ask them if the software they choose to use on their
computer is a moral decision. I'd bet you were the minority by a
percentage hard to even find on any meter. You claim it is moral
because all logic fails. Supporting free software is a good thing, as
opposed to putting work into something that cannot benefit anyone. It
is not "better" than using commercial software, it is just different. I
would like all things to be free. A free car may drive you from point A
to point B just as well as if it were a porche. In fact, the free car
may even be more reliable than the porche. That doesn't mean that it is
morally better than the porche. The only issue with morality and
computers has to do with what you would do with that money if you
weren't to spend it on the software. Spending money on the sick and
needy may make spending money on commercial software less moral, but
then what about that darn hardware that is keeping food from the mouths
of children. The only reason hardware costs money is that someone else
had to work hard to make the facilities to mine and forge that
hardware. Why does that deserve money morally but software doesn't - if
the same time and work is spent? One thing being good doesn't make the
other bad; I'd hate to live in your black and white world...
> Tim> And as such, noone says "find a good rock, you can hammer in
> Tim> a nail just fine and its free and natural! don't use that
> Tim> damn, corporate hammer!" why do people act like this for
> Tim> computer software? I'm a huge fan of free software. I think
> Tim> if something doesn't provide unique value worth marketing, it
> Tim> should be free, with nothing in between. But when I hear
>
> Hah! As though emacs, linux, FreeBSD, sendmail, Apache &c &c offered
> nothing "of unique value worth marketing."
At the time many of these things were started noone would have paid
money for them. They evolved and were created under free licenses.
They were only able to evolve because they were free in the beginning.
Can you show me an example of a top-notch, first release free product
that could have been sold BEFORE it was made free and put into the
public? GNOME and such were free from the beginning, before it was even
released. There were no real deadlines, etc. And, quite honestly, it
isn't worth a nickel yet...
The commercial market has the benefit of bringing products that are
useful immediately, rather than on the timeframe it takes for free
software to either catch up or fix the bugs. In the end, you may end up
with some great free software (as I said, I like it) but it is soooooo
silly to deny the benefits that commercial software can make. Look at
Java... Sure, you can get free software versions of tools, etc, but
people like you just say the commercial stuff sucks until it is free.
For instance, if Symantec's Java compiler were free software you'd say
"look at this great, fast, efficient piece of software." Since its
commercial you say "it is worthless and you're not very smart for
needing it." Be honest with yourself for once: value is value, if it
free or not software performs a tast and some of it is better than
others, and not all of the good stuff is free.
> Tim> Advocates would then say "Symantec is evil for making a
> Tim> superior product and not making it free" or "it wouldn't be
> Tim> so buggy if it was free." Both may be correct statements,
> Tim> but the fact is that if I use Symantec's IDE I can put better
> Tim> food on my table than if I use the command line and VI. We
>
> Doubtful. If you are as smart as you claim you are, which tools you
> use would make no difference in the amount of money you're paid for
> your work. However, if you don't actually know the internal processes
> that are masked by your IDE, you may well find yourself outgunned when
> confronted with the need to do some such by hand.
Show me where I claimed I'm "so" smart. Personal attacks just make you
look more like a zealot, oblivious to reason. That's so stupid its
funny. "Which tools you use would make no difference in the amount of
money you're paid for your work" - are a nimrod or do you just play one
on TV? Sorry, I didn't mean to be offensive, but how simplistic is
acceptable? I know how to do EVERYTHING inside my IDE, but because it
is a commercial piece of software I can't easily do some of the things
the IDE does without the IDE. So, why NOT use the IDE since I can be
more productive with it at time...
If your boss asks you to write a CORBA app; you write the entire thing
by hand, write your makefile and compile - then test and debug... I open
Visual Cafe, code a little IDL and let VCafe generate some code. I drag
a button onto a panel, draw a line to an event and write a little
interaction code. I click build and show my boss. A week later (being
generous, assuming you can write an ORB client in a week since Orbix is
commercial so you shouldn't use it) you have a prototype that may or may
not work right. You show your boss and he shows you your pink slip.
Yeah, right, tools have no bearing... You are paid to do a job on
schedule and sometimes the tools are necessary. You're right that alot
of IDE's are junk, and alot of the time, even the good ones aren't where
you need to spend you time. But don't tell me that a javac compile from
the command line is as fast as using the integrated compilers of some
commercial IDEs.
> Tim> IDE's are just another tool that may or may not be better for
> Tim> you. But people that say "don't use an IDE you wimp" and
> Tim> then say "Emacs IS an IDE" don't really pay attention to
> Tim> their own words, do they? If CodeWarrior doesn't compile
> Tim> faster, provide a better make facility or a better editor, I
> Tim> won't use it. If it does, why wouldn't I use it? I might go
> Tim> to computer hell? :)
>
> I didn't see anybody but you making the contradictory statements.
> After all, you say that you "love free software" as long as it's not
> particularly useful, but you prefer commercial software because you
> think that commercial software is, by definition, "really useful"
> software.
You didn't see other people I guess for the same reason you won't see me
do it again. Logic has no bearing on people that think moral value is
attributed to software types. Opinions like yours, poorly thought out
and closed-minded make people like myself not want to release our
software free - for fear that if I ever do release something commercial
I'll get rediculed. Better to just be another commercial vendor than
one that - heaven forbid - tries to sell software but still work with
the free software people...
I don't know where you get this. Its like because I said commercial
software can be useful I am anti free software. Software is software,
until you see the price tag or the license agreement, all you know is if
it is useful or not. Does it do its job well? If, after looking at the
price and license you decide it falls outside your resources or
distribution requirements, then don't use it. The original value of the
software is still there regardless of its commercial or free status.
And if that value is high enough, you should pay money for it.
As much as I like free software, without corporations and commercial
software we wouldn't have online banking, nearly the Internet
penetration, Java, or many other services that make computers worth
using. I'm just so freaking tired of this moral high ground BS people
are spouting. If I went to just free software I wouldn't be able to
track my finances the way I like, compile a Java program as fast as I
want etc. I use VI as my editor on Windows as well as Linux; I test my
web development against Apache because its a good product and provides
certain advantages. I use VCafe because it also provides certain
benefits. When I don't need those advantages, and would prefer VI's
advantages, I use VI. I highly resent you trying to make people who use
commercial software feel as though they don't belong around your
precious Linux or other free software. You only hurt yourself, and if
that attitude keeps up, people like Oracle will start to rethink their
contributions, or publisher will stop writing the books many of need to
learn about some free applications and OS's.
Guess what, alot of us that contribute to the free software movement
ALSO write commercial software. If you piss enough of us off with this
talk, I'd bet you see alot slower advancement of your precious holy
movement... I hope you can pick up all the slack yourself.
--
Tim Triemstra ... [EMAIL PROTECTED] ... Atlanta, GA USA
Home page: http://detlanta.com
------------------------------
From: Andi Kleen <[EMAIL PROTECTED]>
Subject: Re: Help! sys_shmat (Kernelprogramming)
Date: 05 Apr 1999 18:11:54 +0200
Guido Viehoff <[EMAIL PROTECTED]> writes:
> a friend and me are writing a program consisting of two
> sections: the first
> in kernel, which should limit the max. output for some
> TCP/UDP flows speci-
> fied by the other section in userspace. This second should
> calculate some
> values for the first. Exchange should be performed by SHARED
> MEMORY. The
> code (robin) is inserted in route.c:
> rth->u.dst.output = robin; /* previously:
> ip_output; */
> (changing ip_queue_xmit to robin in tcp_ipv4.c wouldn't
> affect UDP)
This looks very unclean.
A much better way would it be to write
a new queueing discipline, see net/sched/sch_teql.c (which BTW
already does round robin between devices) for an example. You would
use a firewalling module (that hooks in using the firewall.c
API) to mark the skbs based on their flow (skb->fwmark), then
schedule based on the mark in a queueing discipline. It is just
all already implemented by a wide range of modules, e.g. to limit
a particularly marked flow just use the sch_tbf (Token Bucket Filter)
module.
>
> source code:
> key_t shm_key;
> int shm_id, er_val;
> ulong *shm_addr;
> int shm_size = 256;
> int shm_flags = 0666 | IPC_CREAT;
> struct shmid_ds buffer;
>
> shm_key = 'RGC1'; /*ftok not av. in kernel?*/
> shm_id = sys_shmget(shm_key, shm_size, shm_flags);
> er_val = sys_shmat(shm_id, (char *) 0, 0, shm_addr);
A) You cannot pass directly kernel address structures to kernel calls,
they expect user space counters. If you want to do it, you have to
switch segments first with set_fs.
B) Networking runs in a bottom half (some kind of interrupt), which
can run at any time. System calls expect to run in a process context,
which run in cooperative multitasking. Calling them from a BH
breaks the locking completely.
I would better forget SYSV shm (it is a relict), better write a short
device driver that allows your user space program to mmap() a kernel
buffer. See Rubini's Linux Kernel Driver book for an example how to do
it (or e.g. the bttv.c driver in 2.2). If you just need reliable message
passing consider using netlink.
-Andi
--
This is like TV. I don't like TV.
------------------------------
From: Bill Anderson <[EMAIL PROTECTED]>
Subject: Re: runnig more than one netscape window
Date: Mon, 05 Apr 1999 17:18:54 +0000
PiX wrote:
>
> I have a problem with nestcape running more than
> one windows it seems that the kernel isn't capable
> of managing differnt server is ther a special
> thing to do (like multicasting??)
>
> thanks
>
> Pierre
Can you be more specific? I tend to have several Netscape windows open
at any given moment.
--
Bill Anderson Linux Administrator
MCS-Boise (ARC) [EMAIL PROTECTED]
My opinions are just that; _my_ opinions.
------------------------------
From: [EMAIL PROTECTED] (Jeremy Crabtree)
Crossposted-To: alt.os.linux,comp.os.linux.advocacy,comp.os.linux.misc
Subject: Re: Proposal: "Linux 2000 Platform"
Date: 5 Apr 1999 19:05:07 GMT
Reply-To: [EMAIL PROTECTED]
Peter Samuelson allegedly wrote:
>[Jeremy Crabtree <[EMAIL PROTECTED]>]
>> Okay...I was really hoping for a way to past from, say, tty1 (an
>> actual console, not an XTerm) to someplace in X.
>
>I thought about this some time ago.
So, I'm not the only crazy person out there to need/want to do this?!
cool! :)
>(I am quite adept at thinking of
>ways to implement features I am likely never to get around to acting
>on. Sometimes I think I am the ultimate hacker wannabe.)
I can identify with that.
>One way to do this would be to provide an API within gpm -- or possibly
>a helper app -- that would bridge the X selection and the gpm
>selection, keeping them in sync as one changed. I originally thought
>it would involve simply signalling gpm to open such-and-such an X
>display, and would be pretty simple. Then I started thinking and began
>to foresee problems:
[SNIP, problems]
Ah...but what if <dum dum duuum> you use a gpm repeater as your
pointer in X? How would that affect the problems?
>
>> Again, I was hoping for something that would work on a regular
>> console, and paste into X...a pipe dream, I know, but still...
>
>Not a pipe dream, more likely something on the lines of sockets...
Hrmm...maybe...
--
"Being myself a remarkably stupid fellow, I have had to unteach myself
the difficulties, and now beg to present to my fellow fools the parts
that are not hard" --Silvanus P. Thompson, from "Calculus Made Easy."
------------------------------
** 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.development.system) 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-Development-System Digest
******************************
