Linux-Development-Sys Digest #46, Volume #8 Wed, 26 Jul 00 03:13:17 EDT
Contents:
Re: Async IO in Linux... (Andreas Jaeger)
Re: read() question (Rudi Sluijtman)
Re: [ANN] Beta testing of CW for Linux on Intel and PowerPC (Kevin Holbrook)
Re: reentrant - explanation ("Norman Black")
Re: Can i get a MAC address ? (Bob Hauck)
kernel 2.2.12 - get_tty_driver() doesn't exist?? ("A derelict Engineer")
How to prevent a process from paging? (Stefan Ludwig)
Re: where is "clear" source?? ("A derelict Engineer")
Re: Can i get a MAC address ? (Wolfgang Denk)
Re: [ANN] Beta testing of CW for Linux on Intel and PowerPC (Bruce Hoult)
TCP/IP (Amit Utreja)
interrupt question (naveen gogineni)
Re: request_irq problem (naveen gogineni)
kdevelop/gide for win32? ("Jeff D. Hamann")
kernel compiling and LILO.conf ([EMAIL PROTECTED])
Re: Baud rate v.s. Port emulator (Bob Hauck)
Re: A good IDE (Jake Kesinger)
RFI: POSIX 1b IPC Status (Edgar Hilton)
Re: [ANN] Beta testing of CW for Linux on Intel and PowerPC (MWRon)
Re: [ANN] Beta testing of CW for Linux on Intel and PowerPC ([EMAIL PROTECTED])
Re: kernel compiling and LILO.conf (Davide Bianchi)
Re: kdevelop/gide for win32? (Marco van de Voort)
----------------------------------------------------------------------------
From: Andreas Jaeger <[EMAIL PROTECTED]>
Subject: Re: Async IO in Linux...
Date: 25 Jul 2000 21:29:17 +0200
>>>>> Kevin Beard writes:
> The glibc version is 2.1.3.� Here's there complete program as an attachment.
> Ignore the AIX stuff in it.
I've forwarded the program with a comment to the other glibc
developers and CC'ed you. This looks like a bug in glibc, the sig
argument is NULL in your case and therefore glibc segfaults. As a
quick hack, add a proper sig argument to lio_listio.
Andreas
--
Andreas Jaeger
SuSE Labs [EMAIL PROTECTED]
private [EMAIL PROTECTED]
------------------------------
Subject: Re: read() question
From: Rudi Sluijtman <[EMAIL PROTECTED]>
Date: Tue, 25 Jul 2000 19:40:09 GMT
Fro-Man <[EMAIL PROTECTED]> writes:
> When I am reading from stdin with read like such:
>
> rbytes = read( 0, buf, BUF_SIZE );
>
> It will not return until I hit enter. If I read() from a normal file, I
> do not have a problem with it reading only upto the next new line.
>
> I have been remote whenever I have tried this, but I would not assume a
> telnet session would wait until it finds a new line to send data.
>
> Is it possibley something some sort of buffering with the terminal?
Yes, a terminal in canonical mode buffers, and adds limited editing
capabilities. You can find more information about this in the
(g)libc.info files:
node "Low-Level Terminal Interface:" -> node "Canonical or Not"
Here is also information about how to change the terminal mode.
Have fun,
Rudi Sluijtman,
------------------------------
From: Kevin Holbrook <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.powerpc
Subject: Re: [ANN] Beta testing of CW for Linux on Intel and PowerPC
Date: Tue, 25 Jul 2000 15:10:50 -0500
I've suggested that we use the announcement newsgroup in the future, but there
are still several problems with that approach.
1) We really are looking for developers to test the product, not neophytes.
2) We really want PowerPC oriented folks.
Just putting that information in the post doesn't always help hit the desired
target audience.
I do agree that spam suxs, I get enough of it myself from posting in newsgroups
like these.
We didn't intend anything as spam, and those of us working on the product are
pushing so that we work better with the Linux community.
Personally I'd like to see dedicated newsgroups which would be the ideal channel
and only those interested ever see the posts.
That won't happen until a million or more people request it from Metrowerks on a
daily basis. <hint hint wink wink>
Also people should keep in mind when they jump all over a commercial company
that it doesn't represent the Linux community in a favorable manner.
It's easy to take the negativity and have the company say "look they don't want
our product" or "there is no positive market there".
Support for Linux commercially is a good thing, IMHO.
I need a job to pay the bills. Personally I'd rather be working at a company
that has Linux products.
So when you have the urge to pounce, stop and think, maybe just encouraging them
to use another avenue would be best for the community in general.
<Speaking for myself, and not the company.>
--
Kevin Holbrook
Developer
Metrowerks, a Motorola Company
------------------------------
From: "Norman Black" <[EMAIL PROTECTED]>
Subject: Re: reentrant - explanation
Date: Tue, 25 Jul 2000 13:23:15 -0700
Reply-To: "Norman Black" <[EMAIL PROTECTED]>
> Like a semaphore I suppose.
Yes a semaphore with at most a count of one. Better is a mutex, or a
spinlock.
--
Norman Black
Stony Brook Software
the reply, fubar => ix.netcom
"John Reynolds" <[EMAIL PROTECTED]> wrote in message
news:8lkag7$n5d$[EMAIL PROTECTED]...
> In article <8libgv$voj$[EMAIL PROTECTED]>,
> "Norman Black" <[EMAIL PROTECTED]> wrote:
> > Re-entrancy usually refers to the following.
>
> > All other data is
> > shared, unless by algorithm only one thread will possibly access it.
> > This problem can be resolved by using an exclusion object to protect
> > the thread
> > shared data from multiple access.
>
> Like a semaphore I suppose.
>
> >
> > In sum, a re-entrant procedure, means that two or more threads can
> safely
> > execute the procedure at the same time.
>
> Thanks!
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
------------------------------
From: [EMAIL PROTECTED] (Bob Hauck)
Subject: Re: Can i get a MAC address ?
Reply-To: hauck[at]codem{dot}com
Date: Tue, 25 Jul 2000 20:37:24 GMT
On 24 Jul 2000 20:13:36 -0400, Michael Meissner <[EMAIL PROTECTED]> wrote:
>Nowadays, a reason why people might want to change the MAC address
>is to...
...make the 68000 compiler work again since it is node-locked to the
MAC address of my old ethernet card.
Sorry, I'm just real annoyed at a certain vendor's licensing today.
--
-| Bob Hauck
-| Codem Systems, Inc.
-| http://www.codem.com/
------------------------------
From: "A derelict Engineer" <[EMAIL PROTECTED]>
Subject: kernel 2.2.12 - get_tty_driver() doesn't exist??
Date: Tue, 25 Jul 2000 13:47:22 -0700
In my kernel 2.0.36 module, I used the following code to get the tty driver
data for a tty driver, so I could link myself to it:
kdev_t tdev ;
struct tty_driver *dptr ;
tdev = (kdev_t) MKDEV(majorIdx, minorIdx) ;
dptr = (struct tty_driver *) get_tty_driver(tdev);
if (!dptr) {
return -ENODEV ;
However, in kernel 2.2.12, I get "unresolved symbol" when I try to access
that function. Is there some equivalent method for accomplishing this task
in the new kernel??
------------------------------
From: Stefan Ludwig <[EMAIL PROTECTED]>
Subject: How to prevent a process from paging?
Date: Tue, 25 Jul 2000 23:41:42 +0200
Reply-To: [EMAIL PROTECTED]
Is there a way to prevent a running process from using the swap
partition/file?
I need a way to force a process to keep all of its pages in memory.
Telling a process not to page for its lifetime when it is starting is
easy - I just need to do this later, whe it is already running...
There is already a kernel driver running that I could modify to do this,
if it is easier to do directly from "inside the kernel".
Thanks.
Stefan
P.S.: Modifying the program code is not an option.
------------------------------
From: "A derelict Engineer" <[EMAIL PROTECTED]>
Subject: Re: where is "clear" source??
Date: Tue, 25 Jul 2000 15:29:21 -0700
hmmm... I guess that'll work as long as the terminal is vt-compatible...
most are, I suppose...
Thanks...
Zirong Wang <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Hi dan,
>
> try printf(stdout, "\033\143");
>
> A derelict Engineer wrote:
> >
> > Can anyone tell me where the source code for the "clear" utility is
> > included?? I'd like to find out how to execute a clear_screen()
function,
> > without having to link curses, which I don't otherwise use... I've been
> > using system("clear") up to now, and that works adequately, but I'd
rather
> > learn how to do it directly...
> >
> > Thanks!!
>
>
> ---------------------------------------------------------------------
> Zirong Wang Oce Industries 1, Rue J. Lemoine 94015 Creteil France
------------------------------
From: Wolfgang Denk <[EMAIL PROTECTED]>
Subject: Re: Can i get a MAC address ?
Date: Tue, 25 Jul 2000 21:19:18 GMT
[EMAIL PROTECTED] (Rick Ellis) writes:
>In article <[EMAIL PROTECTED]>,
>>Well, that's how it's *supposed* to work. Unfortunatelly, I've seen at least
>>two cases where cheap-ass NIC manufacturers put the same MAC on two or more
>>different NIC's.
>Intel shipped a lot of cards with the same MAC address. Loads
>of fun!
Any Sun workstation with more than one network interface will, have
the same MAC address on all interfaces by default. At least under
Solaris. It's called a feature.
Wolfgang
--
Software Engineering: Embedded and Realtime Systems, Embedded Linux
Phone: (+49)-8142-4596-87 Fax: (+49)-8142-4596-88 Web: www.denx.de
The universe is all a spin-off of the Big Bang.
------------------------------
From: Bruce Hoult <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.powerpc
Subject: Re: [ANN] Beta testing of CW for Linux on Intel and PowerPC
Date: Wed, 26 Jul 2000 10:59:56 +1200
In article <[EMAIL PROTECTED]>, Kevin Holbrook
<[EMAIL PROTECTED]> wrote:
> I've suggested that we use the announcement newsgroup in the future, but
> there are still several problems with that approach.
>
> 1) We really are looking for developers to test the product, not
> neophytes.
> 2) We really want PowerPC oriented folks.
I suspect you'll find that all the complaints have come from readers in
comp.os.linux.development.system, not in comp.os.linux.powerpc.
-- Bruce
------------------------------
From: Amit Utreja <[EMAIL PROTECTED]>
Subject: TCP/IP
Date: Tue, 25 Jul 2000 19:02:01 -0400
Hi
I want to trace TCP/IP header on Linux Machine, to extract both TCP and
IP header from the packet received by the kernel. How should i proceed?
Is there any source or software available to read TCP/IP headers ?
Thanks in advance.
-Amit Utreja
------------------------------
From: naveen gogineni <[EMAIL PROTECTED]>
Subject: interrupt question
Date: Tue, 25 Jul 2000 18:51:25 -0400
HI,
I am writing a program to raise a certain interrupt and am running
into a segmentation fault. The interrupt handler is registered as shown
and loaded as a kernel module.
==========================
#include <asm/signal.h>
#include <stdio.h>
#include <linux/kernel.h>
#include <linux/module.h>
#include <linux/modversions.h>
#define INTERRUPT 49
void handler(int irq,void *dev_id, struct pt_regs *regs) {
printk("hello %d\n",irq);
}
int init_module() {
free_irq(INTERRUPT,NULL);
request_irq(INTERRUPT, handler, SA_INTERRUPT, "my_handler",NULL);
return 0;
}
====================
After loading the above module I checked the /proc/interrupts file to
make sure if everything was ok.
The interrupt raiser itself is in another module.
========================
#include <linux/kernel.h>
#include <linux/module.h>
#include <linux/modversions.h>
#include <pthread.h>
void raise_interrupt();
int init_module() {
printk("raising interrupt...");
raise_interrupt();
printk("done\n");
}
void raise_interrupt() {
__asm__("int $0x31");
}
=====================
When loading this module into the kernel I am getting a segmentation
fault. Could anyone give me some info as to why this occurs ? Thanks.
naveen
------------------------------
From: naveen gogineni <[EMAIL PROTECTED]>
Subject: Re: request_irq problem
Date: Tue, 25 Jul 2000 18:52:34 -0400
What is the interrupt number that you are trying to handle ?
Daniel Etzold wrote:
> My system hangs whenever the INTERRUPT occurs.
> Where is the problem?
>
> [...]
> ret = request_irq(INTERRUPT, irq_handler, 0, "myirq", NULL);
> [...]
>
> void irq_handler(int irq, void *dev_id, struct pt_regs *regs)
> {
> // do nothing
> }
>
> ------------------------------------------------
> Daniel Etzold
>
> ~
> � �
> /V\
> // \\
> /( )\
> ^`~'^
------------------------------
From: "Jeff D. Hamann" <[EMAIL PROTECTED]>
Subject: kdevelop/gide for win32?
Date: Tue, 25 Jul 2000 17:23:36 -0700
Is there a port of gIDE or KDevelop for Win32 (or something even close that
isn't borland of ms)?
===========================================================
Jeff D. Hamann
2601 NE Jack London Street #27
Corvallis, Oregon 97330 USA
541-740-5988
[EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED]
Crossposted-To: comp.os.linux.setup
Subject: kernel compiling and LILO.conf
Date: Tue, 25 Jul 2000 23:22:53 +1000
Hi,
2 questions
1. I am adding some code to the ELF section of the kernel and am just
exploring some options about recompilaion of the kernel. I've heard that
you can have sections of the kernel as modules and just compile and add
them to a running kernel. is this possible with the ELF bianry format
code? Would there be any issues involved with running executables when I
unload the ELF module from the kernel?
2. All the discussions I see say religiously " make a emergeny boot
floppy disk with the old kernel image" when compiling and running a new
kernel. I would like to know if it is posible just to use the lilo.conf
file and specify two boot options. ne for the old kernel and one for the
new kernel. That way if the new kernel stuffs up then I can jsut reboot
and use the old kernel by pressing TAB at the boot prompt.
eg.
A sample section from a lilo.conf file... would this work?
image=/boot/vmlinuz-2.0.36
label=linux
root=/dev/hda2
read-only
image=/boot/vmlinuz-2.0.1
label=old-kernel
root=/dev/hda2
read-only
other=/dev/hda1
label=win
thanks in advance
dan.
ps. could you post replies to my email as well please
([EMAIL PROTECTED])
pps. If this is bad ettiquette asking people to do this then please
advise me so. cheers.
------------------------------
From: [EMAIL PROTECTED] (Bob Hauck)
Subject: Re: Baud rate v.s. Port emulator
Reply-To: bobh{at}haucks{dot}org
Date: Wed, 26 Jul 2000 01:43:59 GMT
On 25 Jul 2000 09:51:44 GMT, [EMAIL PROTECTED]
<[EMAIL PROTECTED]> wrote:
> Could anyone tell me where to find source codes of port
> emulators which transmit/receive data according to their
> baud-rate settings ?
Just to clarify, I interpret this to mean that you want a "soft uart".
Something that emulates what the serial port hardware does, only that
uses just a simple digital IO port. All the bit sampling and baud rate
generation being done in software. Is this correct?
The last time I saw something like that was in Embedded Systems
Journal. Wish I could give a specific date, but it was a while ago. I
know that microcontroller people have been known to do this sort of
thing, in assembler of course. It isn't very efficient on cycles, but
if you save a buck each on a production run of a million...
--
-| Bob Hauck
-| To Whom You Are Speaking
-| http://www.haucks.org/
------------------------------
From: [EMAIL PROTECTED] (Jake Kesinger)
Crossposted-To: comp.os.linux.development.apps,comp.os.linux.hardware
Subject: Re: A good IDE
Date: 26 Jul 2000 01:30:37 GMT
cLIeNUX user ([EMAIL PROTECTED]) wrote:
: The "IDE" feature I use the most is "+", which is supported by Pico and
: the "most" pager. When gcc says
: gcc:238:You_lose.c: HAHAHAHHAHAHAHAA You call that C?? HAHAHHA
: I then do
: pico +238 You_lose.c
: and I'm Integrated and Developing in my Environment.
: If "best" means "get the most work done", you just need to learn to use
: what you've already got.
Well, emacs can do that too. Even better, you can do M-x compile,
and C-x ` (that's a backquote) takes you to the next error in the
output, opening files as necessary.
==Jake
------------------------------
From: Edgar Hilton <[EMAIL PROTECTED]>
Subject: RFI: POSIX 1b IPC Status
Date: Tue, 25 Jul 2000 22:45:37 -0400
Does anybody know what the present status of POSIX style IPC for Linux?
I am namely interested in shared memory, semaphores, and message
queues. There is a document describing the POSIX interface and Linux,
but it seems that the latest version is dated 1998...
Any information on the subject matter would be much appreciated.
-- Edgar
======================================================
Edgar F. Hilton email: [EMAIL PROTECTED]
office#: (804) 924-6233 fax#: (804) 982-2246
http://www.people.virginia.edu/~efh4v
------------------------------
From: [EMAIL PROTECTED] (MWRon)
Crossposted-To: comp.os.linux.powerpc
Subject: Re: [ANN] Beta testing of CW for Linux on Intel and PowerPC
Date: Tue, 25 Jul 2000 22:33:43 -0400
In article <[EMAIL PROTECTED]>, Kevin Holbrook
<[EMAIL PROTECTED]> wrote:
>I've suggested that we use the announcement newsgroup in the future, but there
>are still several problems with that approach.
I am aware of this now and will do so in the future.
>2) We really want PowerPC oriented folks.
We really want both Intel and PowerPC oriented folks. Please intel Linux
users drop me a line and sign on for beta testing :)
>Personally I'd like to see dedicated newsgroups which would be the ideal
channel
>and only those interested ever see the posts.
We got it and I plan to post in the Announce group, there is now a
codewarrior.linux newsgroup. Ask your ISP to add it (and the entire
codewarrior.* hierarchy) to your service.
Ron
--
Metrowerks, a Motorola Company - Ron Liechty
"Software Starts Here" - [EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: [ANN] Beta testing of CW for Linux on Intel and PowerPC
Crossposted-To: comp.os.linux.powerpc
Date: Wed, 26 Jul 2000 04:34:20 GMT
In comp.os.linux.development.system MWRon <[EMAIL PROTECTED]> wrote:
> We got it and I plan to post in the Announce group, there is now a
> codewarrior.linux newsgroup. Ask your ISP to add it (and the entire
> codewarrior.* hierarchy) to your service.
As we continue to drift 1k miles off-topic, I feel obliged to point
out that poluting the hierarchy is an even worse solution than most.
--
Matt Gauthier <[EMAIL PROTECTED]>
------------------------------
From: [EMAIL PROTECTED] (Davide Bianchi)
Crossposted-To: comp.os.linux.setup
Subject: Re: kernel compiling and LILO.conf
Date: Wed, 26 Jul 2000 06:18:50 GMT
Reply-To: [EMAIL PROTECTED]
On Tue, 25 Jul 2000 23:22:53 +1000, [EMAIL PROTECTED] wrote:
>Hi,
>
> 2 questions
>
> 1. I am adding some code to the ELF section of the kernel and am just
>exploring some options about recompilaion of the kernel. I've heard that
>you can have sections of the kernel as modules and just compile and add
>them to a running kernel. is this possible with the ELF bianry format
>code?
I don't think so. AFAIK, the binaries support is not "modular", only
device drivers and filesystem support is modular. But I can be wrong,
so check during the compilation.
> 2. All the discussions I see say religiously " make a emergeny boot
>floppy disk with the old kernel image" when compiling and running a new
>kernel. I would like to know if it is posible just to use the lilo.conf
>file and specify two boot options. ne for the old kernel and one for the
>new kernel. That way if the new kernel stuffs up then I can jsut reboot
>and use the old kernel by pressing TAB at the boot prompt.
Yes, you can configure LILO with multiple kernel, but if something
goes VEEERY WRONG, you can be unable to access the / partition, this
will rendere unusable the system, *unless* you have an old-good
boot disk that allow you to re-enter the system and fix the problem.
>eg.
> A sample section from a lilo.conf file... would this work?
<ZAP>
Look fine to me.
Davide
------------------------------
From: [EMAIL PROTECTED] (Marco van de Voort)
Subject: Re: kdevelop/gide for win32?
Date: 26 Jul 2000 06:55:12 GMT
>Is there a port of gIDE or KDevelop for Win32 (or something even close that
>isn't borland of ms)?
Work-in-progress : Lazarus.freepascal.org
------------------------------
** 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
******************************
