Linux-Development-Sys Digest #8, Volume #7 Thu, 29 Jul 99 16:14:18 EDT
Contents:
Undeclared function ("fernando Ortega")
Re: help on gcc (Paul D. Smith)
Re: Writing shared libraries (Kevin Woodward)
Linux Journal - worth or not? (Bonn)
Re: problems with two network cards. Fundamental limitation? (Grant Edwards)
Re: help on gcc (Grant Edwards)
Re: Problem with glibc compilation (Primary libc)! (Andreas Jaeger)
Enabling core dumps for setuid programs ("Randy Martin")
Re: problems with two network cards. Fundamental limitation? (Rich)
Re: help on gcc (Torbjorn Tallroth)
switching switchbox makes mouse go crazy: any fix? (tom bergerson)
Re: Why ignore a theoretical possibility? (Grant Edwards)
Adaptec 7897 (or PERC2) driver? (bill davidsen)
Re: Writing shared libraries (Paul D. Smith)
can't probe ppp module in kernel 2.2.2.5-15 using loadable module (James Yu)
Re: Writing shared libraries (Graffiti)
Re: Linux Journal - worth or not? (Boudewijn Rempt)
----------------------------------------------------------------------------
From: "fernando Ortega" <[EMAIL PROTECTED]>
Subject: Undeclared function
Date: Thu, 29 Jul 1999 17:55:18 +0200
I am building a module with some functions that must be used for the kernel.
I introduce a call to these funstions in any point of the source code, I
also made and "file.h" defining these funstions as extern, when I try to
compile the kernel , I got a message like:
Undeclared function "........"
I�m missing something......WHAT?
Thanks in advance, Fernando Ortega.
------------------------------
From: [EMAIL PROTECTED] (Paul D. Smith)
Crossposted-To: comp.os.linux.development.apps,comp.os.linux.misc
Subject: Re: help on gcc
Date: 29 Jul 1999 12:30:05 -0400
Reply-To: [EMAIL PROTECTED]
%% grant@nowhere. (Grant Edwards) writes:
>> | man gcc?
>>
>> Or, maybe, info gcc. AFAIK, the man page is not maintained any more,
>> and the info page is more complete anyway.
ge> I hate info. I don't want to wander around some menu tree for
ge> hours looking for something. I just want to search for strings!
Have you tried... gasp!... reading the docs on info?
You want the "s" key.
Also see the very cool "i" key.
--
===============================================================================
Paul D. Smith <[EMAIL PROTECTED]> Network Management Development
"Please remain calm...I may be mad, but I am a professional." --Mad Scientist
===============================================================================
These are my opinions---Nortel Networks takes no responsibility for them.
------------------------------
From: [EMAIL PROTECTED] (Kevin Woodward)
Subject: Re: Writing shared libraries
Date: Thu, 29 Jul 1999 16:52:25 GMT
On 28 Jul 1999 13:52:57 -0400, Paul D. Smith wrote:
>%% [EMAIL PROTECTED] (Kevin Woodward) writes:
>
> >> No. The .a is a static library, in case you want to link =
statically
> >> (otherwise you don't need it), and the .so is a shared library.
>
> kw> Thanks for clear that up. Are there any reasons for using static
> kw> libs over shared libs.
>
>Sure. At least these reasons:
>
> 1) There is runtime overhead associated with shared libs; your apps
> will run slightly slower than with static (all other things being
> equal).
What percentage are we taking about here? Also, is the performance hit
limited to calls to the shared libs or does it include internal calls
inside the lib? Performance is very important in the system I'm
working on.
Kevin.
------------------------------
From: Bonn <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development.apps
Subject: Linux Journal - worth or not?
Date: Fri, 30 Jul 1999 01:11:17 +0800
Reply-To: [EMAIL PROTECTED]
i started using linux for a while. is it worthwhile to subscribe the
'Linux Journal'?
any comment?
regards
bonn
--
___________________________________________________________
|> Food is the first thing. Morals follow on. <|
|< Bread and Puppets >|
|< <|
|> and I wonder... <|
------------------------------
From: grant@nowhere. (Grant Edwards)
Subject: Re: problems with two network cards. Fundamental limitation?
Date: Thu, 29 Jul 1999 16:40:08 GMT
In article <[EMAIL PROTECTED]>, Steve Hier wrote:
>> : Is, by any chance, your DNS server on the side that goes down?
>>
>> Yes. But that shouldn't affect anything. I am telnetting directly
>> to the IP of the internal interface and hence no IP resolution is necessary.
>> And as I said, I can ping that (internal) IP alright.... just no telnet or
>> any other deamon connection.
>
>Ah -- but if your router is trying to do a reverse lookup for the IP
>that is connecting, you have to wait for that to time out -- which I've
>seen take up to 5 minutes on some systems.
>
>Question: When it does "dead" and you telnet, do you get a
>non-responsive
>connection? (as opposed to connection refused / etc.)
That (reverse-DNS timing out) sure sounds like the problem to
me. I recently ran into the same problem when doing some
consulting. Telnet into an SCO box would just "hang" for a
couple seconds. Turns out they had a Micro-Squish DNS server
that wasn't configured to handle reverse-DNS requests.
Where I'm at now, we have the opposite problem: reverse DNS
works for all addresses, but forward doesn't for the DHCP
allocated addresses. Another MS DNS server, I imagine.
--
Grant Edwards grante Yow! I've got a COUSIN
at who works in the GARMENT
visi.com DISTRICT...
------------------------------
From: grant@nowhere. (Grant Edwards)
Crossposted-To: comp.os.linux.development.apps,comp.os.linux.misc
Subject: Re: help on gcc
Date: Thu, 29 Jul 1999 15:59:02 GMT
In article <[EMAIL PROTECTED]>, Jan Andres wrote:
>In article <7nmons$l2$[EMAIL PROTECTED]>, Peter.vanHelden wrote:
>| jievis ([EMAIL PROTECTED]) wrote:
>| : Hi, All:
>| : When I try to compile a c files, I want gcc output a middle files in
>| : which the expression is computed and enumeration is also computed, the
>| : output files should be also .c format. So who can tell me the switcher?
>| : Thanx in advance
>|
>| man gcc?
>
>Or, maybe, info gcc. AFAIK, the man page is not maintained any more,
>and the info page is more complete anyway.
I hate info. I don't want to wander around some menu tree for
hours looking for something. I just want to search for strings!
Is there a way to convert a set of info files into a single
text file that can be searched?
--
Grant Edwards grante Yow! Wow! Look!! A stray
at meatball!! Let's interview
visi.com it!
------------------------------
From: Andreas Jaeger <[EMAIL PROTECTED]>
Subject: Re: Problem with glibc compilation (Primary libc)!
Date: 24 Jul 1999 08:36:35 +0200
>>>>> Ilya Bassine writes:
> [.../
> (All works fine till the moment when test-installation.pl.sh
> starts)
Have a look at the script itself and see what exactly goes wrong. The
script is scripts/test-installation.pl and compiles a simple "hello
world" program linked against all libraries. Do it yourself:
- Compile it, check ldd output.
- Compile it against all libraries (see what the script does). Does
the script still work ?
Andreas
--
Andreas Jaeger [EMAIL PROTECTED] [EMAIL PROTECTED]
for pgp-key finger [EMAIL PROTECTED]
------------------------------
From: "Randy Martin" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.misc,comp.os.linux.questions
Subject: Enabling core dumps for setuid programs
Date: Thu, 29 Jul 1999 13:08:29 -0400
Is there a way to enable core dumping for setuid programs? I know that it
is disabled by default
for security reasons, but i'm developing software on a test system and it
would be nice to
get core dumps for segmentation faults. I saw a blurb in the linux FAQ
about this, but it's not
clear on what exactly I need to change and how to rebuild the kernel.
Anyone know how to do
this?
Thanks,
Randy
------------------------------
From: [EMAIL PROTECTED] (Rich)
Subject: Re: problems with two network cards. Fundamental limitation?
Reply-To: [EMAIL PROTECTED]
Date: Thu, 29 Jul 1999 16:15:36 GMT
On 28 Jul 1999 16:11:06 GMT, Chetan Ahuja <[EMAIL PROTECTED]> wrote:
> Yes. But that shouldn't affect anything. I am telnetting directly
>to the IP of the internal interface and hence no IP resolution is necessary.
Ah, no.
The target box will attempt to perform a reverse-DNS lookup of any incoming
connections. If DNS is not accessible ( and you don't have your
hosts file set up with the local addresses ) then you can expect
to wait several minutes for the DNS lookup to time out, before
you get a login: prompt.
- Rich
------------------------------
Crossposted-To: comp.os.linux.development.apps,comp.os.linux.misc
Date: Thu, 29 Jul 1999 18:30:24 +0200
From: Torbjorn Tallroth <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Subject: Re: help on gcc
On Thu, 29 Jul 1999, Grant Edwards wrote:
> Is there a way to convert a set of info files into a single
> text file that can be searched?
info COMMAND > COMMAND.txt
--
tth
------------------------------
From: tom bergerson <[EMAIL PROTECTED]>
Subject: switching switchbox makes mouse go crazy: any fix?
Date: Thu, 29 Jul 1999 10:00:56 -0500
the problem is similar to the 04/99 post about killing the mouse.
i have a microsoft intellimouse attached to a box running stock redhat
6. i generally do not use gpm because i have had incompatibility
problems with it in the past in x, but in x i have XF86Config setup as
Section "Pointer"
Protocol "IMPS/2"
# Protocol "MouseSystems"
Device "/dev/mouse"
# Device "/dev/gpmdata"
ZAxisMapping 4 5
Buttons 3
.......
EndSection
you can see i have tried to route x through gpm but i had some problems
and no time to sort them out yet. anyway,
if i switch to one of my nt boxes and then switch back with x running,
the mouse bounces all over in the upper right quadrant of the screen.
when i tried the gpm to x setup i also noticed that if running gpm and i
switch to nt and back then the mouse cursor in the shell also displays
identical behavior in bash.
this means i cannot switch, in my normal configuration, without first
closing down x (and usually i just switch when someone wants me to ssh
something to a web server somewhere) so when i come back to the linux
box i have to restart x again which is very annoying.
is there something about losing the constant connection to the mouse
that causes the driver to fritz? is there any way to stop or correct
this behaviour? i am only just learning to program so there is no way i
can look at the code to come up with a fix. anyone else?
tom bergerson
------------------------------
From: grant@nowhere. (Grant Edwards)
Subject: Re: Why ignore a theoretical possibility?
Date: Thu, 29 Jul 1999 18:10:05 GMT
In article <59lnn7.76.ln@JuAnJuX>, *puntero_loco wrote:
>El Mon, 26 Jul 1999 13:00:33 +0200, Benedetto Proietti <[EMAIL PROTECTED]> escribi�:
>|I really thank you for the time dedicated to this idea.
>|I will appreciate very much any comment to these problems, please feel
>|free to answer to the newsgroup.
>
>I think your idea is **very** interesting, not only for the theory, but also
>because it may be complicated, but not imposible to make such a incremental
>compiler.
If you split your source code up into multiple files and use
"make" you get almost the same effect. Most of us have been
doing that for 15+ years.
>For example, Linux Kernel developers would thank a lot it, if you
>can adds modifications to the source and recompile it in a few seconds instead
>of in 10-20 minutes.
Odd. When I worked on the kernel, recompiles only took a few
seconds (OK, maybe 30?). If you use "make" only the modified
files get recompiled. Assuming you don't create programs with
100K lines of source code all in a single file, incremental
compile isn't a big advantage over "make".
>I must also accelerate drastically the development of any other huge programs or
>proyects.
Defining requirements, planning, testing, meeting, and
documenting are where the time goes. Besides, if it didn't take
a while to do builds somtimes, when would I read news?
--
Grant Edwards grante Yow! If I pull this SWITCH
at I'll be RITA HAYWORTH!! Or
visi.com a SCIENTOLOGIST!
------------------------------
From: [EMAIL PROTECTED] (bill davidsen)
Subject: Adaptec 7897 (or PERC2) driver?
Date: 29 Jul 1999 17:16:24 GMT
I just got a new Linux box from Dell, and while I'm waiting for them to
get back to me I'll ask here. Has anyone done a driver for the RAID
controller they use? The box is otherwise a nice little server,
2xXeon-500, 1GB RAM, 210GB disk, but if I can't run Linux it does me no
good.
I asked if this would run Linux before I ordered, the answer does not
appear to have been correct:-(
--
bill davidsen <[EMAIL PROTECTED]> CTO, TMR Associates, Inc
The Internet is not the fountain of youth, but some days it feels like
the fountain of immaturity.
------------------------------
From: [EMAIL PROTECTED] (Paul D. Smith)
Subject: Re: Writing shared libraries
Date: 29 Jul 1999 13:31:39 -0400
Reply-To: [EMAIL PROTECTED]
%% [EMAIL PROTECTED] (Kevin Woodward) writes:
>> 1) There is runtime overhead associated with shared libs; your apps
>> will run slightly slower than with static (all other things being
>> equal).
kw> What percentage are we taking about here?
It's hard to say. I've heard numbers ranging from 1% to 10% for ELF
shared libs. Note that's overhead per reference to the shared lib; your
app won't run 10% slower!
kw> Also, is the performance hit limited to calls to the shared libs
kw> or does it include internal calls inside the lib? Performance is
kw> very important in the system I'm working on.
My understanding is that it affects all calls referencing code in the
shared library, whether they originate from within the lib or not. The
thing is that the code in a shared library is generated as PIC (position
independant). That means all references in the shared lib are
essentially offsets. So, whenever you want to access data or functions
in the shared lib you have to do an extra calculation to add the offset
to whatever fixed value was used to load the shared lib in this run of
the app.
That has to be weighed against the potential speed increase you gain by
having more RAM available on the system due to more sharing of data
between apps.
Obviously, shared libs won't save you anything unless you have many
_different_ apps using the same code. Merely having lots of instances
of the _same_ app isn't going to benefit since all instances of the same
app use the same code image in memory anyway.
So, sometimes it makes sense to use shared libs, sometimes it doesn't.
You have to consider the applications you're writing and the environment
they'll run in to make the best choice.
--
===============================================================================
Paul D. Smith <[EMAIL PROTECTED]> Network Management Development
"Please remain calm...I may be mad, but I am a professional." --Mad Scientist
===============================================================================
These are my opinions---Nortel Networks takes no responsibility for them.
------------------------------
From: James Yu <[EMAIL PROTECTED]>
Subject: can't probe ppp module in kernel 2.2.2.5-15 using loadable module
Date: Thu, 29 Jul 1999 12:29:02 -0700
Hi,
I got the following error messages when I try to load the ppp into the
kernel (2.2.2.5-15) in RedHat 6.0 distributiton. I got the same error
when I use insmod command too. When I exam the source code "ppp.c" under
drivers/net directory, I find that the cli(), save_flags() and
restore_flags() are used all over. They are defined under
include/asm-i386/system.h file. Any idea about how I can make it work ?
Thanks.
[jyu@viper2 2.2.5-15]# modprobe ppp
/lib/modules/2.2.5-15/net/ppp.o: unresolved symbol __global_cli
/lib/modules/2.2.5-15/net/ppp.o: unresolved symbol __global_save_flags
/lib/modules/2.2.5-15/net/ppp.o: unresolved symbol
__global_restore_flags
James
------------------------------
From: Graffiti <[EMAIL PROTECTED]>
Subject: Re: Writing shared libraries
Date: 29 Jul 1999 11:44:49 -0700
In article <[EMAIL PROTECTED]>,
Paul D. Smith <[EMAIL PROTECTED]> wrote:
>%% [EMAIL PROTECTED] (Kevin Woodward) writes:
>
> >> 1) There is runtime overhead associated with shared libs; your apps
> >> will run slightly slower than with static (all other things being
> >> equal).
>
> kw> What percentage are we taking about here?
>
>It's hard to say. I've heard numbers ranging from 1% to 10% for ELF
>shared libs. Note that's overhead per reference to the shared lib; your
>app won't run 10% slower!
It's the startup time that's hit. The runtime overhead is lost in the
system noise. (Unless you're counting CPU cycles for each function
call, in which case you want to use asm anyways. *ducks incoming flames*)
>My understanding is that it affects all calls referencing code in the
>shared library, whether they originate from within the lib or not. The
>thing is that the code in a shared library is generated as PIC (position
>independant). That means all references in the shared lib are
>essentially offsets. So, whenever you want to access data or functions
>in the shared lib you have to do an extra calculation to add the offset
>to whatever fixed value was used to load the shared lib in this run of
>the app.
Yes, this is all resolved when the binary is first run and loaded into
the system's memory. It's not resolved at run-time while the app is
running (i.e. enters main()) unless you explicitly tell it to via
the dlsym() functions.
>Obviously, shared libs won't save you anything unless you have many
>_different_ apps using the same code. Merely having lots of instances
>of the _same_ app isn't going to benefit since all instances of the same
>app use the same code image in memory anyway.
Sure it will. You have a custom libfoo.so (2M) with the function bar().
bar() has a buffer overflow. The binary (baz) you distribute is 10M
static or 4M dynamic. You fix bar().
Option 1:
Make a static 10M baz available.
Option 2:
Make a 2M libfoo.so available.
Option 3:
Both.
-- DN
------------------------------
From: [EMAIL PROTECTED] (Boudewijn Rempt)
Crossposted-To: comp.os.linux.development.apps
Subject: Re: Linux Journal - worth or not?
Date: 29 Jul 1999 19:12:36 GMT
Reply-To: [EMAIL PROTECTED]
Bonn <[EMAIL PROTECTED]> wrote:
> i started using linux for a while. is it worthwhile to subscribe the
> 'Linux Journal'?
> any comment?
>
I've had a subscription for years myself, and while it's still a nice
read, it's becoming more a trade journal for those commercially interested
in Linux. Less kernel hacking, software development or using obscure
packages articles, and more articles about distributors and so on - the
technical content is still there, though. But why not look for yourself?
They're at http://www.linuxjournal.com with a full table of contents
and some on-line articles.
--
Boudewijn Rempt | http://www.xs4all.nl/~bsarempt
------------------------------
** 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
******************************
