Linux-Development-Sys Digest #839, Volume #7 Wed, 10 May 00 03:13:08 EDT
Contents:
Re: Linux behaving like Windows ("Calixto Melean")
Re: Linux behaving like Windows ([EMAIL PROTECTED])
ipfwadm -- How do you set multiple internet IP over a local network ? ("Rene
Chartrand")
Re: Select() in Linux 2.2.5 - threadsafe? ("Mark Graybill")
Re: Two really easy (I'm sure) questions ("Mark Graybill")
Re: Accesing a module within the kernel? ("duan")
Re: pid vs. thread question (Eric Taylor)
Re: Linux behaving like Windows (Eric Taylor)
Re: Two really easy (I'm sure) questions ("Mark Graybill")
Re: kernel manpages ("Dmitry A. Fedorov")
Re: Two really easy (I'm sure) questions ("Mark Graybill")
Re: TCP-IP Stack Question ("Mark Graybill")
Re: What cockheads you are (Charles Blackburn)
Re: What cockheads you are (nilesh patel)
Re: Determining amount of physical RAM from a driver? (nilesh patel)
Re: Two really easy (I'm sure) questions ("Mark Graybill")
----------------------------------------------------------------------------
From: "Calixto Melean" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.misc,comp.os.linux.setup
Subject: Re: Linux behaving like Windows
Date: Tue, 9 May 2000 21:48:26 -0400
thanks. It was a bad fan.
"Sam E. Trenholme" <[EMAIL PROTECTED]> wrote in message
news:8fa58e$1mn$[EMAIL PROTECTED]...
> >So, I am up to the point where I need to compile my modules, after
entering
> >"make modules", it starts compiling for about 3 or 4 minutes and then the
> >machine hangs, no mouse no keyboard, frozen!, I wait for about 30min and
> >nothing happens.
>
> This sounds like a hardware problem. What kind of motherboard and
> processor do you have, have you checked your RAM, etc?
>
> I suspect bad RAM myself.
>
> - Sam
>
> --
> Go to http://www.hoohahrecords.com/rap for information on the Bohemian RAP
CD
> Go to http://samiam.org/cgi-bin/mailme to get my email address
------------------------------
From: [EMAIL PROTECTED]
Crossposted-To: comp.os.linux.misc,comp.os.linux.setup
Subject: Re: Linux behaving like Windows
Date: 9 May 2000 19:04:36 -0700
In article <8fa58e$1mn$[EMAIL PROTECTED]>, [EMAIL PROTECTED]
says...
>
>>So, I am up to the point where I need to compile my modules, after entering
>>"make modules", it starts compiling for about 3 or 4 minutes and then the
>>machine hangs, no mouse no keyboard, frozen!, I wait for about 30min and
>>nothing happens.
>
>This sounds like a hardware problem. What kind of motherboard and
>processor do you have, have you checked your RAM, etc?
>
>I suspect bad RAM myself.
>
I also had a case, where my linux PC will suddenly freeze after
running for some time (even when I am not using it). It turned out
to be a bad RAM. After replacing the RAM, the hangs went away.
nasser
------------------------------
From: "Rene Chartrand" <[EMAIL PROTECTED]>
Subject: ipfwadm -- How do you set multiple internet IP over a local network ?
Date: Tue, 9 May 2000 23:07:47 -0400
ipfwadm -- How do you set multiple internet IP over a local network ?
I have Linux 2.0.36 (Pentium 233) working as transparent proxy with firewall
"ipfwadm 2.30" doing masquerading over a local network 192.168.10.0/24 and
internet. Also from my provider, allocated IP are :
(example)
206.175.34.96
206.175.34.127
Netmask : 255.255.255.224
The local network (mostly windows) is on eth0 .
The ISDN router is using .97, my web address .99 and the broadcast .127.
on eth1 )
Everything working perfectly since one year. I spent a lot of time ( it
worth it ) but I want to add more IP ...
I want my visitors going on 206.175.34.99 on linux ( for mail, web, ftp and
telnet ) and some others (known friends) on 206.175.34.100 on NT 4.0 for
VPN access (and all others port like ftp,telnet,http,smtp...). The ISDN
router is with linux because of firewall, security purpose and money.
I setup .100 as eth1:0 and I want to redirect all ports coming from eth1:0
to a NT server on my local network. First, I read a lot of HowTo. Then I
tried on my own network, and most of the time done nothing of what I want.
# in this example, that line alone is working well as transparent proxy
ipfwadm -F -a accept -m -S 192.168.10.0/24 -D 0.0.0.0/0
# then I tried to add this one
ipfwadm -F -a accept -m -b -P tcp -W eth1:0 -S 0.0.0.0/0 -D 192.168.10.9
ipfwadm -F -a accept -m -b -P udp -W eth1:0 -S 0.0.0.0/0 -D 192.168.10.9
# done nothing so I tried that one
ipfwadm -O -a accept -S 0.0.0.0/0 -D 192.168.10.9 -P tcp -W eth1:0
ipfwadm -O -a accept -S 0.0.0.0/0 -D 192.168.10.9 -P udp -W eth1:0
ipfwadm -I -a accept -S 0.0.0.0/0 -D 192.168.10.9 -P tcp -W eth1:0
ipfwadm -I -a accept -S 0.0.0.0/0 -D 192.168.10.9 -P udp -W eth1:0
# no !
Is anybody succeed with "ipfwadm 2.30" to re-route an IP (with all his
ports) trough linux over a NT server ? Or is there some other known solution
?
Hope I forget nothing
renec
------------------------------
Reply-To: "Mark Graybill" <[EMAIL PROTECTED]>
From: "Mark Graybill" <[EMAIL PROTECTED]>
Subject: Re: Select() in Linux 2.2.5 - threadsafe?
Date: Wed, 10 May 2000 03:48:14 GMT
Kaz Kylheku wrote in message ...
>On Wed, 19 Apr 2000 05:16:14 GMT, Mark Graybill <[EMAIL PROTECTED]> wrote:
>>So in actuality, I guess Kaz's statement is logically correct (my thought
>>thread that was working on this was preempted until after I posted the
last
>>message.) :)
>
>I don't know about logically; it's just a statement of fact. This is how
>it works.
My point was that a process is an encapsulation of a main thread and it's
resources. When additional threads are created, they run within the
boundaries of that encapsulation - hence the difference between threads and
processes.
The perspective you describe is correct, although it implies threads and
processes are two distinct yet peer execution entities with different
characteristics, which is a little misleading. I know you understand as I
do what are threads and process - I was simply trying to clarify their
relationship and differences. Threads do not run outside the boundaries of
a process, but a process has no units of execution themselves - that's the
thread's job.
Best,
Mark
------------------------------
Reply-To: "Mark Graybill" <[EMAIL PROTECTED]>
From: "Mark Graybill" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development,comp.os.linux.development.apps
Subject: Re: Two really easy (I'm sure) questions
Date: Wed, 10 May 2000 04:11:15 GMT
Eric wrote in message <[EMAIL PROTECTED]>...
>Mark Graybill wrote:
>I haven't followed the entire thread, so forgive me if I make a remark
>that was made before.
>void main() wouldn't be usable if you are interested in the succes of
>you're program. Catching a program's return value (0 on success, else
>failure) does require a return value.
No apology necessary - it's all in fun anyway. :)
Whether I am spawning child processes from a parent process or a script, if
I wish to capture the exit code, yes, I must use int return type. But if I
don't, I won't. My point was the resistance to the declaration that we
should never use void return type for main, as a stock axiom - that it was
bad practice. My point is, why is it bad practice?
In order to have a bad practice, there must be a reason that satisfies a
goal. For instance, it is bad practice to replace C language semantics with
#define ... supposedly to make code more readable. The goal is to write
code that is easily maintained and enhanced. The more the native C language
is replaced by #define, the more the next programmer has to learn and
understand about the code in order to maintain it. It's just common sense.
This costs time and resources for a company, or, IOW money. Also,
increasing the complexity of code is problematic in that it provides more
room for human error.
Another bad practice (C++), is to fail to declare destructors as virtual.
If the class is subclassed, this is easily missed, and the sub-class
destructor may not get called if it's super class is not virtual. Since
destructors are usually used for cleaning up and freeing resources, it
affords the opportunity for resource leaks.
ANSI makes no provision prohibiting the use of void main(), and in my 11
years programming in C, I have never met a compiler that didn't like it (my
earlier projects either had the program called by a script, so I used int
main(), or there was no main() function, such as device drivers or file
systems - so I may have run into the problem if I used it more.) Today, if
a compiler chokes on void main(), then it is not fully ANSI compliant.
We are all smart enough to know when we should use void or int, and when we
should not. :)
Cheers,
Mark
------------------------------
From: "duan" <[EMAIL PROTECTED]>
Subject: Re: Accesing a module within the kernel?
Date: Wed, 10 May 2000 11:53:47 +0800
in init_module you should fill the kernel support structure with the code
you have defined.
then you can use the module defined code by access the structure.
analyze the device driver code you can get more
Jpc <[EMAIL PROTECTED]> wrote in message news:[EMAIL PROTECTED]...
> Hi
>
> I just wrote a module and I want to use its functions from the kernel
> (e.g by modifying the kernel sources). I found out that I can use
> request_module() to load the module (with kmod) but i'm unable to figure
> out how to access it.
>
> Can anyone help me?
>
> Thanks in advance.
>
> J.p.
>
> papatata AT hotpop DOT com
------------------------------
From: Eric Taylor <[EMAIL PROTECTED]>
Subject: Re: pid vs. thread question
Date: Tue, 09 May 2000 21:41:34 -0700
> > Is there any other kernel support for threads,
> > is pthreads really just a posix library magic trick that
> > makes it look like linux has a kernel supported thread
> > model?
>
> As much as I know, in Linux there is a implementation
> for threads with the clone system call, which is used
> by the pthread calls!
>
I did a "man clone" and found this:
CLONE_PID
If CLONE_PID is set, the child process
is created with the same process ID
as its parent process.
BUGS
As of version 2.1.97 of the kernel, the CLONE_PID
flag should not be used, since other parts of the kernel and
most system software still assume that process IDs are unique.
I would guess that this is really an evolving of linux into
a true threaded kernel (i.e. full support for user process threading).
Given the BUGS comment, I think this might be the first step
towards a future pid/tid combo.
-e
------------------------------
From: Eric Taylor <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.misc,comp.os.linux.setup
Subject: Re: Linux behaving like Windows
Date: Tue, 09 May 2000 21:52:16 -0700
Calixto Melean wrote:
>
> thanks. It was a bad fan.
>
> "Sam E. Trenholme" <[EMAIL PROTECTED]> wrote in message
> news:8fa58e$1mn$[EMAIL PROTECTED]...
> > >So, I am up to the point where I need to compile my modules, after
> entering
> > >"make modules", it starts compiling for about 3 or 4 minutes and then the
> > >machine hangs, no mouse no keyboard, frozen!, I wait for about 30min and
> > >nothing happens.
> >
> > This sounds like a hardware problem. What kind of motherboard and
> > processor do you have, have you checked your RAM, etc?
> >
> > I suspect bad RAM myself.
> >
> > - Sam
Speaking of bad ram, is there a decent ram tester diag
around?
I'm forever suspicious of my ram. I even bought ecc ram
at 40% more, and it did make my system quite stable.
When ram drops a bit only once an hour or so,
you need a LOOOOONG run to detect this. Testing memory
would be a good use of the cpu when it is idle.
Hint Hint Hint....
-e
-e
------------------------------
Reply-To: "Mark Graybill" <[EMAIL PROTECTED]>
From: "Mark Graybill" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development,comp.os.linux.development.apps
Subject: Re: Two really easy (I'm sure) questions
Date: Wed, 10 May 2000 04:50:58 GMT
Nate Eldredge wrote in message <[EMAIL PROTECTED]>...
>[EMAIL PROTECTED] (Kaz Kylheku) writes:
>You're right. I meant to say that on most common systems this will
>actually work. The original poster seemed decidedly pragmatic, and I
>was pointing out that "you can't have an exit code" is not a valid
>objection to `void main' if you don't care about standards or
>portability, because it can still be done.
>
>It all comes down to, if you only care about what actually works, then
>fine, do what works, and standards be damned. But don't come crying
>later when it stops working.
ANSI provides for the use of void main() (makes no provisions against it),
and therefore compilers that do not support it are not fully ANSI compliant.
Unless I see ANSI declaring otherwise, it is simply dogma.
I have also not observed 'undefined behavior' by any program that I've used
void main() in. This is nonsense. It is simply a matter of opinion, not
standard.
All these 'standards' are mentioned, but they are a mystery - what are they?
If a college professor says 'don't do something', is it a standard?
Portability? I don't know where this comes from. Why are such statements
made that are not ratified or qualified?
I've checked with some colleagues, who are double-decade veterans also, and
they don't know where your argument is coming from. Their experiences are
quite different from my own, so I figured one of them would have had a bad
experience with using void main() (none has.) We all agreed that we use int
main() predominantly, and rarely see a need for void main() (except for
discardable test programs - :), but we have also used void main() and feel
it is simply a incidental design decision.
Now, let's get back to the original argument. The original poster described
a simple program that was giving him difficulty. It was something like:
void main(void){cout << "Hello world";} It was superfluous to the context
of the thread to say it was bad practice to use void main(). Says who?
Who cares?
Now, I understand the desire and validity to be disciplined and to be tidy,
but I see so many programmers being anal retentive about the stupidest
things, and yet they could careless about attention to details like good
comments, or code performance, efficiency, readability and maintainability -
or teamwork.
I would urge all new programmers, once they have graduated, to ratify the
practices they were taught in the commercial world, and be brave enough to
learn additional new practices that serve useful purposes in general, or in
the industry they are serving, and let go of those that do not - or at least
explore the alternatives and understand why.
This argument, as well as many others, is not entirely black and white as
everyone participating thus far seems to believe. I agree with
predominantly practicing the use of int main() - you never know when you may
wish to return an exit code in a future modification; however it is a matter
of design, not standard. The real motivation behind my end of this argument
is to force exploration for deeper understanding, and to discourage the
blind use of hard-fast rules - especially for trivial things that present no
technological consequence. We don't need new programmers entering the work
force blinding following hard-fast rules they don't understand - and who
careless about understanding why. We need programmers who are willing to
expand - who will use void main() because it was their design decision at
the time, and because there is no technological consequence; instead of
using int main() all the time just because someone told them it was bad
practice to do otherwise.
This is all in the hope that perhaps a paradigm or two can be expanded, or
shifted. Contrary to the academic world most programmers learn in, the name
of the game in the real world is one of improving skills, productivity, and
teamwork, not competition. Actually, the complexity of most of my software
projects in recent years demands it.
Sigh. :)
-Mark
------------------------------
From: "Dmitry A. Fedorov" <[EMAIL PROTECTED]>
Subject: Re: kernel manpages
Date: Wed, 10 May 2000 06:05:53 +0000
Lindanne Metley wrote:
>
> Does anything like section 2 and 3 man pages exists for [important|all]
> kernel space calls?
kernel space calls described in section 9 (unmaintained now).
original site was:
ftp://ftp.picturel.com/pub/source/ man9-*
last version disappeared from here, I save it:
ftp://ftp.inp.nsk.su/export/fedorov/man9-19980213.tar.gz
------------------------------
Reply-To: "Mark Graybill" <[EMAIL PROTECTED]>
From: "Mark Graybill" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development,comp.os.linux.development.apps
Subject: Re: Two really easy (I'm sure) questions
Date: Wed, 10 May 2000 06:37:52 GMT
Kaz Kylheku wrote in message ...
>On Mon, 08 May 2000 05:40:23 GMT, Mark Graybill <[EMAIL PROTECTED]> wrote:
>
>When I saw your use of the 2<&1 I did a double take, and thought that it
>was a genuine typo, rather than intentional. I was unaware of this form
until
>you reported that it worked for you, so then my first reaction was ``is
this
>unique to GNU bash?''. I went to the online Single UNIX Spec draft and
looked
>it up, discovering that it is a standard feature, but for duplicating input
>streams.
>
>That it also works on an output stream seems to be a lucky coincidence,
since
>the spec doesn't say what should happen. It's possible that the construct
>may fail, so it should probably be avoided in scripts that are intended to
>be portable.
>
>>tell me why it is bad practice to use void main() instead of int main().
>
>I don't recall the question at all; I certainly *could* tell you. I've
>seen enough threads about main in comp.lang.c to be entirely disinterested
in
>getting into another one. ;) The subject is treated adequately in the C
>programming FAQ ( http://www.eskimo.com/~scs/C-faq/top.html ) in questions
>11.12 through 11.15.
I've read this before, and disagreed with it. My old copy of the standard,
all my books, and all my experiences with many compilers and many platforms,
aren't this rigid. However, in retrospect, I probably rarely used void
main() in the past.
I will research the current copies of references Steve mentions. If the
references are this rigid, then I stand corrected, and it is an obvious
indication of all the compilers out there that are slacking. Although C
(and IBM 360/370 assembly) were the main programming languages I learned
back in college and used later, I cannot remember that far back to discern
whether the use of int main() instead of void main() was taught to me then
or not. Perhaps the old K&R C allowed it, perhaps not. Perhaps I reacted
to "its bad practice" without an explanation in your original reply post.
If the standards do have a proviso requiring main to be of type int, then it
would have been appropriate to state that in the first place. If you
weren't sure, then perhaps we both learned something.
>Isn't it obvious that when you write a function that is called by code
which
>you don't control, you can't just pluck a type signature out of the air?
Huh? I'm not plucking a type signature out of the air. I've only used int
and void types of main because my references and compilers allowed it. I'm
not sure where I picked up the void type. I probably didn't use it much,
but I didn't recall any rules against it.
However, I have recently discovered that It is possible, that a poorly
written crt0 may have a stack problem with using any return type that is not
the size of int - including void main().
>Doesn't the type system of a programming language mean anything to you?
Yes, which is why I refrain from using C style type casting in my C++
programs, and why I seldom use void* anymore, etc. (i.e. cannot use ptr
arithmetic with void*, and it doesn't preserve virtual function tables.)
>I don't see the connection between a graduate student with a personality
>disorder and a spec-following engineer.
I was not making such a connection. Her problem was not one of a
personality flaw or even a neurosis - it was simply indicative of
educational institutions that make up arbitrary rules or practices - that
are sometimes contradictive to the goals of an employer.
If there is a specification stating that 2>&1 is the correct method, then
excellent - I didn't know. My original (and limited) understanding of the
difference between 2>&1 and 2<&1 is whether stderr was redirected to stdout
or visa versa. With that in mind, I could not see why one would be correct
and the other incorrect. If the serendiputous method of 2<&1 was an
unintention fluke of shell function (on many platforms), then I can see your
point and I agree; however, I didn't believe it was - rather I believed that
the shell purposely allowed either way.
>You seem to have a ``try and see if it works'' approach toward programming
>languages. This approach is valid only in uncharted territory, not in
>a territory which is governed by applicable standards.
You're a little arrogant to think you can speculate on my professional
style, since you haven't a clue where I am at in that regard, and since the
inflexible semantics of programming languages is not conducive to the "try
and see if it works" approach. But I am curious where you infer this
perspective? Certainly you don't think that 2<&1 / 2>&1 is programming, do
you? Or the argument of void main() vs. int main() is indicative of 20 years
of experience?
>If the specification for an electrical circuit calls for a resistor that
can
>sink 100 watts of power, it would be irresponsible to substitute a 50 watt
>resistor, right? Yet in software we do this kind of substitution all too
often
>because of the ``I tried this on my machine/OS/compiler and it works''
>mentality.
Yes, it is done too often, which is problematic and usually causes someone
to get 'bit' by a painful 'bug', usually after implementation.
>If you do have to break the rules in order to avoid some gross inefficiency
>(for instance) it's nice to *know* that you are breaking the rules and
which
>ones.
True - which is exactly what I asked for.
However, there are more differences between software engineering and all
other engineering disciplines than there are similarities. For instance, in
reference to your analogy, if I break the rules of physics in an electronic
circuit, I have disfunction (or a popped resister or capacitor - or worse,
like a fire); but if I break a rule like 2>&1, what are my consequences?
No, the hard disk won't fail, nor is a virus is launched - the system won't
even hang. If it doesn't work, I edit the script file (or modify my typing)
and correct the syntax. It's like those who drive with their lights on all
the time, and think others are risky if they don't do the same (there's many
like that here.) Contrary to your assumptions, my approach to software
engineering is quite disciplined, yet creatively inclinded.
>However, engineering specifications are not dogma. We should be thankful
for
>whenever they are available to guide us in a given area that would
otherwise be
>ruled by chaos.
The engineering specifications you speak of, and arbitrary programming rules
taught in school are two different things. In my engineering practice, I
adhere to standards and specifications like ANSI (although I may miss a nit
or two, like void main()), the various RFCs, CORBA, UML, and the software
development process. I even adhere to the unofficial standards that have
been set by experts in the industry (e.g. Meyers, Coplian, and Stroustrup.)
Using 2<&1 because it works is a completely unrelated issue (BTW, it works
in every shell I have - including the command processor in NT and OS/2.) It
has also worked with OSF1, AIX, HP-UX, Sun, SCO, and the discarded, yet
great OS, Warp for PPC.
Don't forget that engineering is a balance between discipline and
creativity. Rigitiy and discipline is needed, but it's not as concrete as
Ohm's law. Yes, there needs to be standards, and I am an advocate of
industry wide standards. This is why I refuse to engineer software that
uses platform or 3rd party libraries in place of the ANSI C libraries or
STL. This is why I design my software using the OO standard UML, and make
use of peer reviewed patterns - before I write one line of code; hence
following a disciplined software development process. The dogma I speak of
are rules that are taught that serve a purpose in academia, but not in
business. The ones that are arbitrary and intangible.
-Mark
------------------------------
Reply-To: "Mark Graybill" <[EMAIL PROTECTED]>
From: "Mark Graybill" <[EMAIL PROTECTED]>
Subject: Re: TCP-IP Stack Question
Date: Wed, 10 May 2000 06:39:48 GMT
Fabrizio C. wrote in message <10s21vf$ril$[EMAIL PROTECTED]>...
>Interesting, but let me understand something more.
>
>Is there a the part of the code-path that is NOT duplicated (the low-level
>HW driver ?)
>Is this creating any bottle-neck ?
I could only say for certain how threads work; however, unless I actually
look at the Linux source for the TCP stack, I can only speculate.
There should be something in the LDP on this.
-Mark
------------------------------
From: [EMAIL PROTECTED] (Charles Blackburn)
Subject: Re: What cockheads you are
Date: Tue, 9 May 2000 21:04:06 +0100
Reply-To: [EMAIL PROTECTED]
On Tue, 09 May 2000 18:11:47 GMT, Kaz Kylheku wrote:
>>You cocksuckers couldn't afford a fucking windows environment....poor cunts
>>probably couldn't afford a mouse or a 500meg hard drive. fuck it would be a
>>cunt to live like you low life scum, and you have the nerve to call yourself
>>fucking developers......SUCK MY DICK you low life fuckwits.
>Are we to understand that you are propositioning the (largely male) readership
>of the newsgroup to engage in fellatio?
Sounds like it.
>--
>#exclude <windows.h>
#ifdef __WINDOWS_H__
#error "Get a real OS
#endif
--
Charles Blackburn -=- Remove NOSPAM to email a reply.
Summerfield Technology Limited - SuSE Linux Reseller & Birmingham L.U.G sponsor
[EMAIL PROTECTED]
[EMAIL PROTECTED]
8:59pm up 1:32, 2 users, load average: 0.00, 0.00, 0.00
------------------------------
From: nilesh patel <[EMAIL PROTECTED]>
Subject: Re: What cockheads you are
Date: Wed, 10 May 2000 12:19:08 +0530
"www.homepagewizard.com" wrote:
> You cocksuckers couldn't afford a fucking windows environment....poor cunts
> probably couldn't afford a mouse or a 500meg hard drive. fuck it would be a
> cunt to live like you low life scum, and you have the nerve to call yourself
> fucking developers......SUCK MY DICK you low life fuckwits.
He / She ??
------------------------------
From: nilesh patel <[EMAIL PROTECTED]>
Subject: Re: Determining amount of physical RAM from a driver?
Date: Wed, 10 May 2000 12:21:40 +0530
Timur Tabi wrote:
high_mem - 0xC0000000
Try this.
Nilesh
> I would think that this is a FAQ, but I can't find the answer anywhere.
>
> I need to be able to determine, from within my driver's init_module
> function, the amount of physical RAM in the system. I don't want to
> know anything about virtual memory, I'm simply interested in one
> number: the amount of physical RAM installed on a system, in megabytes.
>
> Can anyone help me?
>
> --
> Timur Tabi
> Interactive Silicon - http://www.interactivesi.com
> Remove "nospam_" from my email address when replying
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
------------------------------
Reply-To: "Mark Graybill" <[EMAIL PROTECTED]>
From: "Mark Graybill" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development,comp.os.linux.development.apps
Subject: Re: Two really easy (I'm sure) questions
Date: Wed, 10 May 2000 06:43:31 GMT
I have to amend the previous post with the comment that I have recently
discovered that I may have placed too much faith in my references, and that
I should consult the new ANSI C standard.
Disregard my declarations that ANSI C made no restrictions on void main().
I will post again once I have purchased a copy of the current standard and
have read it.
BTW, I tried void main() on all the compilers I currently have in my lab,
and only one complained with a warning: IBM VisualAge for C++ for OS/2.
Obviously I have never used void main() during my OS/2 development days, or
I would have remembered.
Cheers,
Mark
------------------------------
** 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
******************************
