[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
rtl-digest V1 #62
rtl-digest Tuesday, May 29 2001 Volume 01 : Number 062
----------------------------------------------------------------------
Date: Fri, 25 May 2001 09:38:39 +0200
From: s186392 <s186392@studenti.ing.unipi.it>
Subject: [rtl] RTLinux on=?ISO-8859-1?Q?=03Asus?= CUV4X-D with wich kernel?!
Dear friends,
I have an installation trouble with my dual processor board (Asus CUV4X-D, 2
Pentium 3 1 Ghz ). I tried to install 2.4.0, 2.4.1 and 2.4.2 linux kernel with
3.0 RTlinux kernel and I faced some difficulties:
- -the kernel patch for 2.4.0 is wrong;
- -if I use 2.4.1 patch, that's ok with that kernel, but, while building
RTLinux, I receive the message: 'skew clock...the building process may be
uncomplete'. So, if I go on, the modules won't be built. Obviously, I already
disabled APM Bios. Is there something else to disable? Shall I use a different
combination of linux/rtlinux kernels?
- -- [rtl] ---
To unsubscribe:
echo "unsubscribe rtl" | mail majordomo@rtlinux.org OR
echo "unsubscribe rtl <Your_email>" | mail majordomo@rtlinux.org
- --
For more information on Real-Time Linux see:
http://www.rtlinux.org/rtlinux/
------------------------------
Date: Fri, 25 May 2001 12:07:11 +0530 (IST)
From: Anumula Venkata Naga Seshaiah <anumula@iitg.ernet.in>
Subject: Re: [rtl] rtlinux and Telnet
Hello Chandu,
I checked it out after i saw u'r mail. When u insert a module from
TELNET the output will be printed on the machine console u are contacting.
Not on the machine u are working. I have same experience previously.
When u insert a module from X server i am not getting output there itself
though i am working on the machine itself ( Not thro' TELNET ) , but the
ouput is coming on the textmode windows. As i am new to RTLinux i don't
the reason exactly.
Bye
Venkat
________________________________________________________
On Mon, 21 May 2001, chandu Reddy wrote:
> Hi
>
> i am working with rtlinux module programming.
> i faced probles while inserting module thru TELNET
> at that time i was working as a root.
> But i am able to compile the code and got the object code ( .o) file
> for the module thru TELNET.
>
> whether it is possible to insert module thru TELNET or not
> If it is Possible Can any one Help me to sort out this Problem
> in this regard.
>
> Thanks
> Chandu
>
>
>
>
>
> These are my personal intersets.
>
> -- [rtl] ---
> To unsubscribe:
> echo "unsubscribe rtl" | mail majordomo@rtlinux.org OR
> echo "unsubscribe rtl <Your_email>" | mail majordomo@rtlinux.org
> --
> For more information on Real-Time Linux see:
> http://www.rtlinux.org/rtlinux/
>
- -- [rtl] ---
To unsubscribe:
echo "unsubscribe rtl" | mail majordomo@rtlinux.org OR
echo "unsubscribe rtl <Your_email>" | mail majordomo@rtlinux.org
- --
For more information on Real-Time Linux see:
http://www.rtlinux.org/rtlinux/
------------------------------
Date: Fri, 25 May 2001 14:47:19 +0200 (CEST)
From: =?iso-8859-1?q?marco=20zoncu?= <mark_zonk@yahoo.com>
Subject: [rtl] Asus CUV4X-D & RTLinux
Dear friends,
I have an installation trouble with my dual processor
board (Asus CUV4X-D, 2 Pentium 3 1 Ghz ). I tried to
install 2.4.0, 2.4.1 and 2.4.2 linux kernel with 3.0
RTlinux kernel and I faced some difficulties:
- -the kernel patch for 2.4.0 is wrong;
- -if I use 2.4.1 patch, that's ok with that kernel,
but, while building RTLinux, I receive the message:
'skew clock...the building process may be uncomplete'.
So, if I go on, the modules won't be built. Obviously,
I already disabled APM Bios. Is there something else
to disable? Shall I use a different combination of
linux/rtlinux kernels?
Thank U all
Marco
______________________________________________________________________
Do You Yahoo!?
Il tuo indirizzo gratis e per sempre @yahoo.it su http://mail.yahoo.it
- -- [rtl] ---
To unsubscribe:
echo "unsubscribe rtl" | mail majordomo@rtlinux.org OR
echo "unsubscribe rtl <Your_email>" | mail majordomo@rtlinux.org
- --
For more information on Real-Time Linux see:
http://www.rtlinux.org/rtlinux/
------------------------------
Date: Fri, 25 May 2001 08:56:15 -0400
From: "Heinz Haeberle" <Heinz.Haeberle@gmx.net>
Subject: Re: [rtl] RTLinux: C++ static members != good ?
I sent an answer already. Got lost somehow.
Anyhow, the reason for this behavior is, that a static member is not just a
type definition, it is actually a real memory area. Therefor you have to
create a real variable therefore. I personally dont like that behavior, but
I think it was the only possible solution for C++. Again the reason is it
belongs to the class rather then to one of the objects.
What you have to add this to _one_ of the .cpp files:
int A::a;
or
int A::a=0;
if you want to initialize it
- ----- Original Message -----
From: "Evan Harvey" <poncholio@hotmail.com>
To: <rtl@rtlinux.org>
Sent: Wednesday, May 23, 2001 10:48 AM
Subject: [rtl] RTLinux: C++ static members != good ?
> I have modified the hello.cpp example that came with RTLinux. I moved
class
> A to a.h and then moved class B declarations to b.h and implementation to
> b.cpp.
>
> I can compile and run the modules no problem.
>
> my problem comes when I try to make
> int a; //from class A
> static int a;
>
> when I load the module I get an error like
>
> error in b.o : unresolved symbol _1A.a
>
> Any clue why this is? Can I not use static members with RTLinux? What is
> an alternative to static?
>
> Thank you for your help.
>
> -Evan
> poncholio@hotmail.com
>
> _________________________________________________________________
> Get your FREE download of MSN Explorer at http://explorer.msn.com
>
> -- [rtl] ---
> To unsubscribe:
> echo "unsubscribe rtl" | mail majordomo@rtlinux.org OR
> echo "unsubscribe rtl <Your_email>" | mail majordomo@rtlinux.org
> --
> For more information on Real-Time Linux see:
> http://www.rtlinux.org/rtlinux/
>
- -- [rtl] ---
To unsubscribe:
echo "unsubscribe rtl" | mail majordomo@rtlinux.org OR
echo "unsubscribe rtl <Your_email>" | mail majordomo@rtlinux.org
- --
For more information on Real-Time Linux see:
http://www.rtlinux.org/rtlinux/
------------------------------
Date: Fri, 25 May 2001 08:56:15 -0400
From: "Heinz Haeberle" <Heinz.Haeberle@gmx.net>
Subject: Re: [rtl] RTLinux: C++ static members != good ?
I sent an answer already. Got lost somehow.
Anyhow, the reason for this behavior is, that a static member is not just a
type definition, it is actually a real memory area. Therefor you have to
create a real variable therefore. I personally dont like that behavior, but
I think it was the only possible solution for C++. Again the reason is it
belongs to the class rather then to one of the objects.
What you have to add this to _one_ of the .cpp files:
int A::a;
or
int A::a=0;
if you want to initialize it
- ----- Original Message -----
From: "Evan Harvey" <poncholio@hotmail.com>
To: <rtl@rtlinux.org>
Sent: Wednesday, May 23, 2001 10:48 AM
Subject: [rtl] RTLinux: C++ static members != good ?
> I have modified the hello.cpp example that came with RTLinux. I moved
class
> A to a.h and then moved class B declarations to b.h and implementation to
> b.cpp.
>
> I can compile and run the modules no problem.
>
> my problem comes when I try to make
> int a; //from class A
> static int a;
>
> when I load the module I get an error like
>
> error in b.o : unresolved symbol _1A.a
>
> Any clue why this is? Can I not use static members with RTLinux? What is
> an alternative to static?
>
> Thank you for your help.
>
> -Evan
> poncholio@hotmail.com
>
> _________________________________________________________________
> Get your FREE download of MSN Explorer at http://explorer.msn.com
>
> -- [rtl] ---
> To unsubscribe:
> echo "unsubscribe rtl" | mail majordomo@rtlinux.org OR
> echo "unsubscribe rtl <Your_email>" | mail majordomo@rtlinux.org
> --
> For more information on Real-Time Linux see:
> http://www.rtlinux.org/rtlinux/
>
- -- [rtl] ---
To unsubscribe:
echo "unsubscribe rtl" | mail majordomo@rtlinux.org OR
echo "unsubscribe rtl <Your_email>" | mail majordomo@rtlinux.org
- --
For more information on Real-Time Linux see:
http://www.rtlinux.org/rtlinux/
------------------------------
Date: Fri, 25 May 2001 14:54:28 +0200
From: Linus Gasser <ineiti@gmx.net>
Subject: Re: [rtl] rtlinux and Telnet
On Friday 25 May 2001 08:37, you wrote:
> Hello Chandu,
>
> I checked it out after i saw u'r mail. When u insert a module from
> TELNET the output will be printed on the machine console u are contacting.
> Not on the machine u are working. I have same experience previously.
> When u insert a module from X server i am not getting output there itself
> though i am working on the machine itself ( Not thro' TELNET ) , but the
> ouput is coming on the textmode windows. As i am new to RTLinux i don't
> the reason exactly.
>
> Bye
> Venkat
> ________________________________________________________
>
> On Mon, 21 May 2001, chandu Reddy wrote:
> > Hi
> >
> > i am working with rtlinux module programming.
> > i faced probles while inserting module thru TELNET
> > at that time i was working as a root.
> > But i am able to compile the code and got the object code ( .o) file
> > for the module thru TELNET.
> >
> > whether it is possible to insert module thru TELNET or not
> > If it is Possible Can any one Help me to sort out this Problem
> > in this regard.
> >
> > Thanks
> > Chandu
Hi,
does it help if you let
tail -f /var/log/messages
run in a second session?
Ineiti
- -- [rtl] ---
To unsubscribe:
echo "unsubscribe rtl" | mail majordomo@rtlinux.org OR
echo "unsubscribe rtl <Your_email>" | mail majordomo@rtlinux.org
- --
For more information on Real-Time Linux see:
http://www.rtlinux.org/rtlinux/
------------------------------
Date: Fri, 25 May 2001 10:38:43 -0400
From: "Stephen D. Cohen" <scohen@xybion.com>
Subject: RE: [rtl] unresolved symbol __udivdi3 and __umoddi3
Gang,
> You need to link in the libgcc.a which is located at:
>
> /usr/lib/gcc-lib/i386-redhat-linux/egcs-2.91.66/libgcc.a
>
> It could be a slightly different directory depending on
> your linux
> install. These are some internal math rountines used by gcc.
The easy (and portable) way to find libgcc is to simply ask gcc
where it is. A command like:
gcc -print-libgcc-file-name
will print the location of the libgcc file. You can also use this
directly in your makefile with something like:
LIBGCC := $(shell $(CC) -print-libgcc-file-name)
and then adding ${LIBGCC} to the end of the ld line in your make
file.
> Now, I see you are including libm.a -- you will probably see
> multiple defines, since libm.a needs libc.a -- functions like
> perror. And
> you cannot bring in libc.a into the kernel for a variety of
> reasons. If
> you need math functionality, others may have solution to that ...
I have never had a problem using the math library without libc. In
fact, I have never successfully linked against libc at all for an RTLinux
module. Hmmm....
Regards,
Steve Cohen
- -- [rtl] ---
To unsubscribe:
echo "unsubscribe rtl" | mail majordomo@rtlinux.org OR
echo "unsubscribe rtl <Your_email>" | mail majordomo@rtlinux.org
- --
For more information on Real-Time Linux see:
http://www.rtlinux.org/rtlinux/
------------------------------
Date: Fri, 25 May 2001 10:38:43 -0400
From: "Stephen D. Cohen" <scohen@xybion.com>
Subject: RE: [rtl] unresolved symbol __udivdi3 and __umoddi3
Gang,
> You need to link in the libgcc.a which is located at:
>
> /usr/lib/gcc-lib/i386-redhat-linux/egcs-2.91.66/libgcc.a
>
> It could be a slightly different directory depending on
> your linux
> install. These are some internal math rountines used by gcc.
The easy (and portable) way to find libgcc is to simply ask gcc
where it is. A command like:
gcc -print-libgcc-file-name
will print the location of the libgcc file. You can also use this
directly in your makefile with something like:
LIBGCC := $(shell $(CC) -print-libgcc-file-name)
and then adding ${LIBGCC} to the end of the ld line in your make
file.
> Now, I see you are including libm.a -- you will probably see
> multiple defines, since libm.a needs libc.a -- functions like
> perror. And
> you cannot bring in libc.a into the kernel for a variety of
> reasons. If
> you need math functionality, others may have solution to that ...
I have never had a problem using the math library without libc. In
fact, I have never successfully linked against libc at all for an RTLinux
module. Hmmm....
Regards,
Steve Cohen
- -- [rtl] ---
To unsubscribe:
echo "unsubscribe rtl" | mail majordomo@rtlinux.org OR
echo "unsubscribe rtl <Your_email>" | mail majordomo@rtlinux.org
- --
For more information on Real-Time Linux see:
http://www.rtlinux.org/rtlinux/
------------------------------
Date: Fri, 25 May 2001 11:16:34 -0400
From: Claudia Fontes <Claudia.Fontes@comrise.com>
Subject: [rtl] Embedded Software: real-time Linux required
This message is in MIME format. Since your mail reader does not understand
this format, some or all of this message may not be legible.
- ------_=_NextPart_001_01C0E52D.AF38ACF0
Content-Type: text/plain;
charset="iso-8859-1"
Hi, I have a job opening in Irvine, CA with Telcordia (Bellcore). They are
looking for Embedded Software Engineer: (2 positions) Experience with
real-time Linux required. VxWorks experience helpful. Software development
experience on Power PC and Pentium III processors desired. C/C++ programming
experience required. We need extremely experienced professionals. The
project is for a major airline who is introducing internet access in
airplanes. Please, let me know ASAP if you know someone that would be
interested in this opportunity. The salary is open. The length of the
project is 6-12 months. The candidate MUST be a US Citizen, and be able to
start ASAP. we are sending resumes on Tuesday morning. Please, forward your
resume to claudia.fontes@comrise.com <mailto:claudia.fontes@comrise.com>
and/or phil@comrise.com <mailto:phil@comrise.com> ASAP! Please write
Embedded Irvine, CA in the subject line.
Thanks a lot for your help.
Claudia Fontes
Recruiting Manager
Comrise Technology, Inc.
1301 Highway 36
Concord Center
Hazlet, NJ 07730
Tel: 1-800-862-2330 ext. 234
Tel: (732) 739-2330 ext. 234
Fax: (732) 739-1996
claudia@comrise.com
http://www.comrise.com <http://www.comrise.com/>
- ------_=_NextPart_001_01C0E52D.AF38ACF0
Content-Type: text/html;
charset="iso-8859-1"
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=iso-8859-1">
<META content="MSHTML 5.00.3103.1000" name=GENERATOR></HEAD>
<BODY>
<DIV><SPAN class=093110915-25052001><FONT face=Arial size=2>Hi, I have a job
opening in Irvine, CA with Telcordia (Bellcore). They are looking for Embedded
Software Engineer: (2 positions) Experience with real-time Linux required.
VxWorks experience helpful. Software development experience on Power PC and
Pentium III processors desired. C/C++ programming experience required. We need
extremely experienced professionals. The project is for a major airline who is
introducing internet access in airplanes. Please, let me know ASAP if you
know someone that would be interested in this opportunity. The salary is open.
The length of the project is 6-12 months. The candidate MUST be a US Citizen,
and be able to start ASAP. we are sending resumes on Tuesday morning. Please,
forward your resume to <A
href="mailto:claudia.fontes@comrise.com";>claudia.fontes@comrise.com</A> and/or
<A href="mailto:phil@comrise.com";>phil@comrise.com</A> ASAP! Please write
Embedded Irvine, CA in the subject line.</FONT></SPAN></DIV>
<DIV><SPAN class=093110915-25052001></SPAN> </DIV>
<DIV><SPAN class=093110915-25052001><FONT face=Arial size=2>Thanks a lot for
your help.</FONT></DIV>
<P></SPAN><FONT face=Tahoma size=1>Claudia Fontes <BR>Recruiting
Manager<BR>Comrise Technology, Inc. <BR>1301 Highway 36 <BR>Concord Center
<BR>Hazlet, NJ 07730 <BR>Tel: 1-800-862-2330 ext. 234 <BR>Tel: (732)
739-2330 ext. 234 <BR>Fax: (732) 739-1996 <BR>claudia@comrise.com <BR><U><FONT
color=#0000ff><A href="http://www.comrise.com/";
target=_blank>http://www.comrise.com</A></FONT></U> </FONT></P>
<DIV> </DIV></BODY></HTML>
- ------_=_NextPart_001_01C0E52D.AF38ACF0--
- -- [rtl] ---
To unsubscribe:
echo "unsubscribe rtl" | mail majordomo@rtlinux.org OR
echo "unsubscribe rtl <Your_email>" | mail majordomo@rtlinux.org
- --
For more information on Real-Time Linux see:
http://www.rtlinux.org/rtlinux/
------------------------------
Date: Fri, 25 May 2001 11:56:03 -0500
From: Havard Gullbekk <hgull@mail.ukans.edu>
Subject: RE: [rtl] unresolved symbol __udivdi3 and __umoddi3
Hi Chris,
I tried linking in the libgcc.a like this:
ahrs.o: ahrs.c
$(CC) ${CFLAGS} -c -o ahrs_tmp.o ahrs.c
$(LD) -r -static ahrs_tmp.o -o ahrs.o -L/usr/lib -lm -lc
- -L/usr/lib/gcc-lib/i486-suse-linux/2.95.2 -lgcc
rm -f ahrs_tmp.o
Now my system freezes when I insert ahrs.o What did I do wrong?
(I don't get any error messages)
Havard
>===== Original Message From "Christopher D. Carothers" <chrisc@cs.rpi.edu>
=====
> Hi Havard --
>
> You need to link in the libgcc.a which is located at:
>
>/usr/lib/gcc-lib/i386-redhat-linux/egcs-2.91.66/libgcc.a
>
> It could be a slightly different directory depending on your linux
>install. These are some internal math rountines used by gcc.
>
> Now, I see you are including libm.a -- you will probably see
>multiple defines, since libm.a needs libc.a -- functions like perror. And
>you cannot bring in libc.a into the kernel for a variety of reasons. If
>you need math functionality, others may have solution to that ...
>
> Hope that helps,
> Chris
>
- -- [rtl] ---
To unsubscribe:
echo "unsubscribe rtl" | mail majordomo@rtlinux.org OR
echo "unsubscribe rtl <Your_email>" | mail majordomo@rtlinux.org
- --
For more information on Real-Time Linux see:
http://www.rtlinux.org/rtlinux/
------------------------------
Date: Fri, 25 May 2001 08:00:47 -1000 (HST)
From: Tony Denault <denault@hawaii.edu>
Subject: Re: [rtl] unresolved symbol __udivdi3 and __umoddi3
On Thu, 24 May 2001, Christopher D. Carothers wrote:
> Hi Havard --
>
> You need to link in the libgcc.a which is located at:
> /usr/lib/gcc-lib/i386-redhat-linux/egcs-2.91.66/libgcc.a
>
> > When I insert ahrs.o as a module (insmod ahrs.o) I get these errors:
> >
> > ahrs.o: unresolved symbol __udivdi3
> > ahrs.o: unresolved symbol __umoddi3
> >
> > I have no idea where the __udivdi3 and __umoddi3 are comming from and
> > what their purpose is. So any suggestions are welcome.
I had a similar problem but with __divdi3(). Linking in the libgcc.a
resolved the problem because the compiler is generating a reference to
this library.
I found out the __divdi3() is needed to do 64 bit division. Later I got
rid of my 64bit division, and didn't need to reference libgcc.a.
I suspect you are doing a divid & mod on a long long type.
Are you doing math on hrtime_t? hrtime_t is a long long.
To check, you can us the -S option in gcc to produce the assemble code
intermediate file. Then you can see were and why these function are needed
by the compiler.
Tony
/-----------------------------------------------------------------------------\
| Tony Denault | Internet: denault@irtf.ifa.hawaii.edu |
| NASA IRTF, Institute of Astronomy | Phone: (808) 974-4206 |
| 1175 Manono St., Bldg 393 | Fax: (808) 974-4207 |
| Hilo, Hawaii 96720 | |
\-----------------------------------------------------------------------------/
- -- [rtl] ---
To unsubscribe:
echo "unsubscribe rtl" | mail majordomo@rtlinux.org OR
echo "unsubscribe rtl <Your_email>" | mail majordomo@rtlinux.org
- --
For more information on Real-Time Linux see:
http://www.rtlinux.org/rtlinux/
------------------------------
Date: Sun, 27 May 2001 09:55:16 +0100
From: "chandu Reddy" <chandureddy@hclt.com>
Subject: Re: [rtl] rtl linux debuggung
Hi
fine, my debugging work going smoothly.
and i faced some problems while working with rtl debugging
i have some doubts regarding this........
i am working thru telnet.
1) when i am debugging thru telnet..., there may be chance that some other
person also working at the same time.
this i tried as my self logged in as two different instances of
oot( two windows).
whether it is possible that two person make use of same rtl_debug.o module
for their
or i need a separate instance of debugging module inerted into the
kernel with two different sets
one from 10,11, 12 and 20, 21, 22,23
i tried both of these but i am not able to succed. whether kernel
supports this?
Could any one help me to solve these problems..
i think Mr Thomas is Can.
Thanks
Chandu
These are my Personal openions/interests.
- -- [rtl] ---
To unsubscribe:
echo "unsubscribe rtl" | mail majordomo@rtlinux.org OR
echo "unsubscribe rtl <Your_email>" | mail majordomo@rtlinux.org
- --
For more information on Real-Time Linux see:
http://www.rtlinux.org/rtlinux/
------------------------------
Date: Mon, 28 May 2001 09:06:01 +0100
From: "chandu Reddy" <chandureddy@hclt.com>
Subject: Re: [rtl] rtl linux debuggung
Hi
All
sorry for my last mail, which was sent in middle of composition.
Ok
fine, my debugging work going smoothly.
and i faced some problems while working with rtl debugging
i have some doubts regarding this........
i m working thru telnet.
1) when i am debugging thru telnet..., there may be chance that some other
person also working at the same time.
this i tried like, my self logged in as two different instances of
root( two windows).
whether it is possible that two person make use of same rtl_debug.o module
which is currently loaded into the kernel.
or we need a separate instance of debugging module( with some other
name)
inerted into the kernel which may using( let us consider) two different
sets of fifos
one from 10,11, 12 and 20, 21, 22.
i tried both of these but i am not able to succeed.
whether kernel supports this type of functionality ?
or
this is because of any mistakes i made while trying like this.
Could any one help me to solve these problems..
2) second problem is like thuis
i have two c source files ( * . c) and one header file.
i developed a module with two source files( *.c )
i defined some functions in one file and calling the same in another
file.
while compiling , its giving error that " unresolved symbols
functionnmae()
Could any one help me to resolve these issues...
Thanks
Chandu
These are my Personal openions/interests.
- -- [rtl] ---
To unsubscribe:
echo "unsubscribe rtl" | mail majordomo@rtlinux.org OR
echo "unsubscribe rtl <Your_email>" | mail majordomo@rtlinux.org
- --
For more information on Real-Time Linux see:
http://www.rtlinux.org/rtlinux/
------------------------------
Date: Mon, 28 May 2001 07:58:06 +0200
From: Suessmilch Bernd <Bernd.Suessmilch@SWAROVSKI.COM>
Subject: RE: [rtl] uresolved symbol __udivdi3 and __umoddi3
Hello Havard,
I think the simplest solution for your problem is not posted so far:
just tell the compiler to use the inline code of the math-functions.
You can do this by adding the "-O2" flag to your CFLAGS.
The inline math-functions are faster than the one in libm, but less
error checking is performed. You have to check the returned values
for "NaN" to detect errors.
A further drawback of the inline functions is that gdb has difficulties
to identify the right source lines when debugging.
Bernd
* - * - * - * - * - * - * - * - * - * -
SWAROVSKI & CO
Dipl.-Inf. Bernd Suessmilch
Elektronik - Entwicklung 2
Tel. ++43 - +5224 - 500 DW 3671
Fax ++43 - +5224 - 571 90
Mailto: Bernd.Suessmilch@swarovski.com
- -- [rtl] ---
To unsubscribe:
echo "unsubscribe rtl" | mail majordomo@rtlinux.org OR
echo "unsubscribe rtl <Your_email>" | mail majordomo@rtlinux.org
- --
For more information on Real-Time Linux see:
http://www.rtlinux.org/rtlinux/
------------------------------
Date: Mon, 28 May 2001 09:22:16 +0200
From: Jan Albiez <albiez@fzi.de>
Subject: Re: [rtl] RTLinux: C++ static members != good ?
On Fri, May 25, 2001 at 08:56:15AM -0400, Heinz Haeberle wrote:
> What you have to add this to _one_ of the .cpp files:
> int A::a;
> or
> int A::a=0;
> if you want to initialize it
This is exactly one of the thing which just don't work at our place. I mean
the compiler doesn't cry or something, but the value is just not there. The
definition/declaration is a bit more complex, but you just reach nothing
over the pointer:
in .h file:
typdef tDescription char*;
class tTest {
static const tDescription[] desc;
}
in .C file:
tDecription tTest::desc[] = { "one", two" };
in Linux space this works quite well.
blue skies
Jan
- --
- -- Jan Christian Albiez --
- -- FZI -- Haid-und-Neu-Str. 10-14 -- 76131 Karlsruhe -- +49 721 9654 206 --
- -- segmentation violation in module reality.o ...
... please shutdown your universe and reboot --
- -- [rtl] ---
To unsubscribe:
echo "unsubscribe rtl" | mail majordomo@rtlinux.org OR
echo "unsubscribe rtl <Your_email>" | mail majordomo@rtlinux.org
- --
For more information on Real-Time Linux see:
http://www.rtlinux.org/rtlinux/
------------------------------
Date: Mon, 28 May 2001 14:21:19 +0100
From: "Goran Bervar" <goran.bervar@adapti.si>
Subject: [rtl] mbuf, rtl FIFOs and synchronisation
Hello,
Pardon my ignorance but I have some questions regarding both shared memory
and real-time FIFOs. Please note that all checking is omitted from examples
for simplicity.
1) Using shared memory allocated by mbuf on RT side and Linux user space to
share some struct, say my_struct, is there any locking involved? For
example, if I have the same code in module and application as follows, both
called in a loop:
- ---------------------------------------
volatile char * shm1;
...
volatile my_struct* pstruct = (volatile my_struct*)
mbuff_alloc("myinterface",sizeof(my_struct));
....
pstruct = internal_struct; /*just for example, copy the whole structure at
once in both directions*/
...
internal_struct = pstruct;
- ---------------------------------------
Is there any locking involved when writing form one side to another or is it
possible that either task gets "uncompletly written" structure because of
the task switch? I am asking because all examples I've seen don't implement
any semaphores or mutexes. If I do need to implement locking I'd be more
than grateful for a simple example.
2) Using FIFOs, I suppose there is a locking declared by opening type
argument and RT FIFO mechanisms. But in example:
- ------------------------------------------
/* rt side */
int ch_in, ch_out;
/* in init_module */
ch_in = rtf_create(1, 4000);
ch_out = rtf_create(2, 4000);
/* in RT task function, msg is some message structure */
rtf_get (1, &msg, sizeof(msg))) == sizeof(msg));
...
rtf_put (2, &msg, sizeof(msg));
/* application side */
int fd0 = open("/dev/rtf1", O_RDONLY);
int fd1 = open("/dev/rtf2", O_WRONLY);
....
write(fd1, &msg, sizeof(msg));
...
n = read(fd0, buf, BUFSIZE - 1);
buf[n] = 0;
- --------------------------------------------
Do I understand correctly if I think both read and rtf_get functions may
fail on incomplete data and it is possible that I have to retry reading
input later ? Is therefore line
rtf_get (1, &msg, sizeof(msg))) == sizeof(msg));
more or less unusuable as I have to be sure data is complete? I can not
imagine rtl_get/read will block execution of thread to wait for complete
data but on the other hand it is possible that FIFO's data counter is not
invalidated before the last byte is written with write/rtl_put. Is this the
case with RT FIFOs?
Thx,
goran
- -- [rtl] ---
To unsubscribe:
echo "unsubscribe rtl" | mail majordomo@rtlinux.org OR
echo "unsubscribe rtl <Your_email>" | mail majordomo@rtlinux.org
- --
For more information on Real-Time Linux see:
http://www.rtlinux.org/rtlinux/
------------------------------
Date: Mon, 28 May 2001 14:20:20 +0530
From: "Srinivasa A. Shikaripura" <srinivasa.shikaripura@wipro.com>
Subject: Re: [rtl] Asus CUV4X-D & RTLinux
hi,
marco zoncu wrote:
>
> Dear friends,
> I have an installation trouble with my dual processor
> board (Asus CUV4X-D, 2 Pentium 3 1 Ghz ). I tried to
> install 2.4.0, 2.4.1 and 2.4.2 linux kernel with 3.0
> RTlinux kernel and I faced some difficulties:
>
> -the kernel patch for 2.4.0 is wrong;
>
> -if I use 2.4.1 patch, that's ok with that kernel,
> but, while building RTLinux, I receive the message:
> 'skew clock...the building process may be uncomplete'.
For this problem, I think the date on the machine is
incorrect. Just 'touch' all the files to be re-compiled.
Like:
$ find . -name '*[cphs]' -exec touch {} \;
$ find . -name '[mM]akefile' -exec touch {} \;
then check if it still complains of clock skew...
> So, if I go on, the modules won't be built. Obviously,
> I already disabled APM Bios. Is there something else
> to disable? Shall I use a different combination of
> linux/rtlinux kernels?
>
> Thank U all
>
> Marco
>
> ______________________________________________________________________
> Do You Yahoo!?
> Il tuo indirizzo gratis e per sempre @yahoo.it su http://mail.yahoo.it
>
> -- [rtl] ---
> To unsubscribe:
> echo "unsubscribe rtl" | mail majordomo@rtlinux.org OR
> echo "unsubscribe rtl <Your_email>" | mail majordomo@rtlinux.org
> --
> For more information on Real-Time Linux see:
> http://www.rtlinux.org/rtlinux/
- -- [rtl] ---
To unsubscribe:
echo "unsubscribe rtl" | mail majordomo@rtlinux.org OR
echo "unsubscribe rtl <Your_email>" | mail majordomo@rtlinux.org
- --
For more information on Real-Time Linux see:
http://www.rtlinux.org/rtlinux/
------------------------------
Date: Mon, 28 May 2001 09:29:51 +0200
From: "Jonas Persson" <jonas.persson@comsys.se>
Subject: SV: [rtl] unresolved symbol __udivdi3 and __umoddi3
I have also had problems with __udivdi3. I got rid of the problem by typecasting to float, the math processor can do 64 bit division.
Regards,
Jonas
______________________________________________________________________
Jonas Persson E-mail: jonas.persson@comsys.se
Project Manager, M.Sc C.Sc&E http://www.comsys.se/
Comsys AB Phone: +46-(0)46-286 3051
Forskningsbyn IDEON Fax: +46-(0)46-286 3505
S-223 70 Lund, Sweden
- -----Ursprungligt meddelande-----
Fran: owner-rtl@fsmlabs.com [mailto:owner-rtl@fsmlabs.com]For Tony
Denault
Skickat: den 25 maj 2001 20:01
Till: rtl@fsmlabs.com
Amne: Re: [rtl] unresolved symbol __udivdi3 and __umoddi3
On Thu, 24 May 2001, Christopher D. Carothers wrote:
> Hi Havard --
>
> You need to link in the libgcc.a which is located at:
> /usr/lib/gcc-lib/i386-redhat-linux/egcs-2.91.66/libgcc.a
>
> > When I insert ahrs.o as a module (insmod ahrs.o) I get these errors:
> >
> > ahrs.o: unresolved symbol __udivdi3
> > ahrs.o: unresolved symbol __umoddi3
> >
> > I have no idea where the __udivdi3 and __umoddi3 are comming from and
> > what their purpose is. So any suggestions are welcome.
I had a similar problem but with __divdi3(). Linking in the libgcc.a
resolved the problem because the compiler is generating a reference to
this library.
I found out the __divdi3() is needed to do 64 bit division. Later I got
rid of my 64bit division, and didn't need to reference libgcc.a.
I suspect you are doing a divid & mod on a long long type.
Are you doing math on hrtime_t? hrtime_t is a long long.
To check, you can us the -S option in gcc to produce the assemble code
intermediate file. Then you can see were and why these function are needed
by the compiler.
Tony
/-----------------------------------------------------------------------------\
| Tony Denault | Internet: denault@irtf.ifa.hawaii.edu |
| NASA IRTF, Institute of Astronomy | Phone: (808) 974-4206 |
| 1175 Manono St., Bldg 393 | Fax: (808) 974-4207 |
| Hilo, Hawaii 96720 | |
\-----------------------------------------------------------------------------/
- -- [rtl] ---
To unsubscribe:
echo "unsubscribe rtl" | mail majordomo@rtlinux.org OR
echo "unsubscribe rtl <Your_email>" | mail majordomo@rtlinux.org
- --
For more information on Real-Time Linux see:
http://www.rtlinux.org/rtlinux/
- -- [rtl] ---
To unsubscribe:
echo "unsubscribe rtl" | mail majordomo@rtlinux.org OR
echo "unsubscribe rtl <Your_email>" | mail majordomo@rtlinux.org
- --
For more information on Real-Time Linux see:
http://www.rtlinux.org/rtlinux/
------------------------------
Date: Mon, 28 May 2001 16:53:35 +0200
From: Suessmilch Bernd <Bernd.Suessmilch@SWAROVSKI.COM>
Subject: RE: [rtl] uresolved symbol __udivdi3 and __umoddi3
Hello Havard,
it is me again. I think, my previous answer to your question was
precipitant.
Compiling with "-O2" may not solve your problem, but here is
a solution that works:
Use the compiler flags "-static -r -nostartfile -nostdlib -lgcc".
The explanation:
1. __divdi3 and __udivdi3 is used to compute "/" with operands
of type "long long" or "unsigned long long", respectively.
The are defined in libgcc.a. Therefore, you have to use "-lgcc"
2. When linking a program against libgcc, gcc automatically adds
a function for program startup. A kernel module must not linked
with this startup function. Therefore you need the flag "-nostartfile".
3. You cannot use libc in kernel space. Therefore you need the flag
"-nostdlib".
4. Kernel modules have to be relocatable. Therefore you need the "-r" flag.
Sorry for my previous misleading posting.
Regards,
Bernd
- -- [rtl] ---
To unsubscribe:
echo "unsubscribe rtl" | mail majordomo@rtlinux.org OR
echo "unsubscribe rtl <Your_email>" | mail majordomo@rtlinux.org
- --
For more information on Real-Time Linux see:
http://www.rtlinux.org/rtlinux/
------------------------------
Date: Mon, 28 May 2001 14:20:20 +0530
From: "Srinivasa A. Shikaripura" <srinivasa.shikaripura@wipro.com>
Subject: Re: [rtl] Asus CUV4X-D & RTLinux
hi,
marco zoncu wrote:
>
> Dear friends,
> I have an installation trouble with my dual processor
> board (Asus CUV4X-D, 2 Pentium 3 1 Ghz ). I tried to
> install 2.4.0, 2.4.1 and 2.4.2 linux kernel with 3.0
> RTlinux kernel and I faced some difficulties:
>
> -the kernel patch for 2.4.0 is wrong;
>
> -if I use 2.4.1 patch, that's ok with that kernel,
> but, while building RTLinux, I receive the message:
> 'skew clock...the building process may be uncomplete'.
For this problem, I think the date on the machine is
incorrect. Just 'touch' all the files to be re-compiled.
Like:
$ find . -name '*[cphs]' -exec touch {} \;
$ find . -name '[mM]akefile' -exec touch {} \;
then check if it still complains of clock skew...
> So, if I go on, the modules won't be built. Obviously,
> I already disabled APM Bios. Is there something else
> to disable? Shall I use a different combination of
> linux/rtlinux kernels?
>
> Thank U all
>
> Marco
>
> ______________________________________________________________________
> Do You Yahoo!?
> Il tuo indirizzo gratis e per sempre @yahoo.it su http://mail.yahoo.it
>
> -- [rtl] ---
> To unsubscribe:
> echo "unsubscribe rtl" | mail majordomo@rtlinux.org OR
> echo "unsubscribe rtl <Your_email>" | mail majordomo@rtlinux.org
> --
> For more information on Real-Time Linux see:
> http://www.rtlinux.org/rtlinux/
- -- [rtl] ---
To unsubscribe:
echo "unsubscribe rtl" | mail majordomo@rtlinux.org OR
echo "unsubscribe rtl <Your_email>" | mail majordomo@rtlinux.org
- --
For more information on Real-Time Linux see:
http://www.rtlinux.org/rtlinux/
------------------------------
Date: Mon, 28 May 2001 14:28:13 +0300
From: "Christos Tranoris" <tranoris@ee.upatras.gr>
Subject: [rtl] cpp example
Hello=20
In the distibution
the directory examples/cpp
has only a file rtl_cpp.c
which I cannot compile.
Also the rtl_cpp.o does not exist anywhere.
Should be also there a hello.c ?
Am I missing something?
thank you=20
Christos
- -- [rtl] ---
To unsubscribe:
echo "unsubscribe rtl" | mail majordomo@rtlinux.org OR
echo "unsubscribe rtl <Your_email>" | mail majordomo@rtlinux.org
- --
For more information on Real-Time Linux see:
http://www.rtlinux.org/rtlinux/
------------------------------
Date: Mon, 28 May 2001 21:10:00 -0400
From: "Heinz Haeberle" <Heinz.Haeberle@gmx.net>
Subject: Re: [rtl] RTLinux: C++ static members != good ?
well the only clue I have is that there must be some initialisation code,
which is not executed. Like the static objects have to be initialized too. I
remember from VxWorks, that there was something which called those init code
when you loaded the code. It has to be done only for cpp code. If you make
the initialisation in the init code. What's gonna happen then?
Heinz
> On Fri, May 25, 2001 at 08:56:15AM -0400, Heinz Haeberle wrote:
> > What you have to add this to _one_ of the .cpp files:
> > int A::a;
> > or
> > int A::a=0;
> > if you want to initialize it
>
> This is exactly one of the thing which just don't work at our place. I
mean
> the compiler doesn't cry or something, but the value is just not there.
The
> definition/declaration is a bit more complex, but you just reach nothing
> over the pointer:
>
> in .h file:
>
> typdef tDescription char*;
>
> class tTest {
>
> static const tDescription[] desc;
> }
>
> in .C file:
>
> tDecription tTest::desc[] = { "one", two" };
>
>
> in Linux space this works quite well.
- -- [rtl] ---
To unsubscribe:
echo "unsubscribe rtl" | mail majordomo@rtlinux.org OR
echo "unsubscribe rtl <Your_email>" | mail majordomo@rtlinux.org
- --
For more information on Real-Time Linux see:
http://www.rtlinux.org/rtlinux/
------------------------------
Date: Tue, 29 May 2001 03:34:33 -0600
From: Victor Yodaiken <yodaiken@fsmlabs.com>
Subject: Re: [rtl] Bug???
On Thu, May 24, 2001 at 12:00:29AM +0800, George Doukas wrote:
> I wrote this simple module for RTLinux...
>
> #include <rtl.h>
> #include <time.h>
> #include <rtl_sched.h>
> #include <rtl_sync.h>
>
> pthread_t thread;
>
> void * thread_code(void *t) {
> int i,j;
>
> for (i=0;i<10;i++) {
> for(j=0;j<10;j++)
> rtl_printf("-");
> }
>
> pthread_suspend_np(thread);
> return (void *)0;
> }
>
> int init_module(void) {
> pthread_create(&thread,NULL,thread_code,0);
> return 0;
> }
>
> void cleanup_module(void) {
> pthread_delete_np(thread);
> }
>
> There was no problem runing it.
>
> But when I changed the second for loop...
> for (i=0;i<10;i++) {
> for(j=0;j<1000;j++)
> rtl_printf("-");
> }
> I got this message:
> "Aiee, killing interrupt handler
> Kernel panic:Attempted to kill the idle task!
> In interrupt handler - not syncing."
>
> Why?
> Is there a maximum limit to the time that a RT process is runing?
What version of RTLinux and what compile options?
- -- [rtl] ---
To unsubscribe:
echo "unsubscribe rtl" | mail majordomo@rtlinux.org OR
echo "unsubscribe rtl <Your_email>" | mail majordomo@rtlinux.org
- --
For more information on Real-Time Linux see:
http://www.rtlinux.org/rtlinux/
------------------------------
Date: Tue, 29 May 2001 03:34:33 -0600
From: Victor Yodaiken <yodaiken@fsmlabs.com>
Subject: Re: [rtl] Bug???
On Thu, May 24, 2001 at 12:00:29AM +0800, George Doukas wrote:
> I wrote this simple module for RTLinux...
>
> #include <rtl.h>
> #include <time.h>
> #include <rtl_sched.h>
> #include <rtl_sync.h>
>
> pthread_t thread;
>
> void * thread_code(void *t) {
> int i,j;
>
> for (i=0;i<10;i++) {
> for(j=0;j<10;j++)
> rtl_printf("-");
> }
>
> pthread_suspend_np(thread);
> return (void *)0;
> }
>
> int init_module(void) {
> pthread_create(&thread,NULL,thread_code,0);
> return 0;
> }
>
> void cleanup_module(void) {
> pthread_delete_np(thread);
> }
>
> There was no problem runing it.
>
> But when I changed the second for loop...
> for (i=0;i<10;i++) {
> for(j=0;j<1000;j++)
> rtl_printf("-");
> }
> I got this message:
> "Aiee, killing interrupt handler
> Kernel panic:Attempted to kill the idle task!
> In interrupt handler - not syncing."
>
> Why?
> Is there a maximum limit to the time that a RT process is runing?
What version of RTLinux and what compile options?
- -- [rtl] ---
To unsubscribe:
echo "unsubscribe rtl" | mail majordomo@rtlinux.org OR
echo "unsubscribe rtl <Your_email>" | mail majordomo@rtlinux.org
- --
For more information on Real-Time Linux see:
http://www.rtlinux.org/rtlinux/
------------------------------
End of rtl-digest V1 #62
************************