Ziaulhaque Qazi | 1 Jan 2012 15:12
Favicon

problem installing gcc


Hi All

Being a beginner of RTEMS in a virtual machine running SL5 (scientific Linux 5) , I am trying to install
rtems-4.10.2 for power pc as my target is mvme 2300 board.

I could install the binary utilities but at the time I tried to install gcc, it failed. I tried many times but
it failed always.

First time, it was declaring the gmp and mpfr not compatible. I installed the latest packages of the two but
still installing gcc failed. However this time the error report was changed and I found a number of
messages reporting something about mpfr.

I configure the gcc as follows:

../gcc-4.4.6/configure --target=powerpc-rtems4.1
0.2 --prefix=/usr/local/rtems/rtems-4.10.2 --disable-libstdcxx-pch --with-gmp= "
/usr/local/include/gmp.h" --with-mpfr= "/usr/local/include/mpfr.h" --with-gnu-as
 --with-gnu-ld --verbose --with-newlib --with-system-zlib --disable-win32-regist
ry --disable-nls --without-included-gettext --enable-version-specific-runtime-li
bs --enable-threads --enable-newlib-io-c99-formats --enable-languages="c,c++"

And I obtained following warnings during configuration:

configure: WARNING: you should use --build, --host, --target
configure: WARNING: invalid host type: /usr/local/include/gmp.h
configure: WARNING: you should use --build, --host, --target
configure: WARNING: invalid host type: /usr/local/include/mpfr.h
checking build system type... /usr/local/include/gmp.h
checking host system type... /usr/local/include/gmp.h
(Continue reading)

Ralf Corsepius | 1 Jan 2012 15:32
Favicon

Re: problem installing gcc

On 01/01/2012 03:12 PM, Ziaulhaque Qazi wrote:
>
> Hi All
>
> Being a beginner of RTEMS in a virtual machine running SL5 (scientific Linux 5) ,

Is this distro supposed to be RHEL5 compatible?

Then CentOS5 binary rpms are supposed to work.

> I am trying to install rtems-4.10.2 for power pc as my target is mvme 2300 board.
>
> I could install the binary utilities but at the time I tried to install gcc, it failed. I tried many times but
it failed always.
It's close to impossible to help you based on your description - More 
details please (e.g. yum's output).

Ralf
_______________________________________________
rtems-users mailing list
rtems-users@...
http://www.rtems.org/mailman/listinfo/rtems-users

Gedare Bloom | 2 Jan 2012 23:15
Favicon

Re: problem installing gcc

On Sun, Jan 1, 2012 at 9:12 AM, Ziaulhaque Qazi
<ziaulhaque.qazi@...> wrote:
>
> Hi All
>
> Being a beginner of RTEMS in a virtual machine running SL5 (scientific Linux 5) , I am trying to install
rtems-4.10.2 for power pc as my target is mvme 2300 board.
>
> I could install the binary utilities but at the time I tried to install gcc, it failed. I tried many times but
it failed always.
>
> First time, it was declaring the gmp and mpfr not compatible. I installed the latest packages of the two but
still installing gcc failed. However this time the error report was changed and I found a number of
messages reporting something about mpfr.
>
One thought is that you need the development versions of those
libraries to build gcc; something like gmp-devel or gmp-dev or similar
and the same for mpfr. If you search for gmp / mpfr you should find
them. Also I think you need to have libmpc-devel to compile current
gcc versions. You can also fetch the sources for gmp/mpfr/mpc and
place them in the root directory of your gcc (softlink with name "gmp"
"mpfr" or "mpc" to the source trees of each). Then building gcc with
it configured as just --with-mpfr and --with-gmp should compile those
libraries for you.

Or find suitable RPMs to use.

> I configure the gcc as follows:
>
> ../gcc-4.4.6/configure --target=powerpc-rtems4.1
(Continue reading)

Andrei Chichak | 3 Jan 2012 07:26
Picon

Re: MPC5634M starting point

Thanks Thomas and Joel,

I have moved my tools and RTEMS code version up to 4.10.2. I'd like to take a peek at the 5674F BSP to see what kind of changes were necessary to make it go together. Would I just take the CVS head of the libbsp/powerpc arm?

Is there anything weird that I would have to do to configure the ppc version of GCC to generate appropriate code? I will be using a e200z3 processor, whereas the 5674F uses an e200z7.

I'm probably doing something wrong, but right out of the box there is a problem with the mpc5566evb BSP, wherein when I compile a simple printf hello world test program, the compile bombs with:

c:/opt/rtems-4.10/bin/../lib/gcc/powerpc-rtems4.10/4.4.6/../../../../powerpc-rtems4.10/bin/ld.exe:linkcmds.mpc5566evb:3: warning: redeclaration of memory region `ROM'
c:/opt/rtems-4.10/bin/../lib/gcc/powerpc-rtems4.10/4.4.6/../../../../powerpc-rtems4.10/bin/ld.exe:linkcmds.mpc5566evb:4: warning: redeclaration of memory region `RAM'
c:/opt/rtems-4.10/bin/../lib/gcc/powerpc-rtems4.10/4.4.6/../../../../powerpc-rtems4.10/bin/ld.exe:linkcmds.mpc5566evb:5: warning: redeclaration of memory region `RAM_EXT'
c:/opt/rtems-4.10/bin/../lib/gcc/powerpc-rtems4.10/4.4.6/../../../../powerpc-rtems4.10/bin/ld.exe:linkcmds.mpc5566evb:6: warning: redeclaration of memory region `NIRVANA'
c:/opt/rtems-4.10/bin/../lib/gcc/powerpc-rtems4.10/4.4.6/../../../../powerpc-rtems4.10/bin/ld.exe:linkcmds.mpc5566evb:23: error: redefinition of memory region alias `REGION_TEXT'

It looks like the linker files need a bit of work.

Andrei

On 2011-December-30, at 2:27 PM, Thomas Doerfler wrote:

Andrei,

I am off site now, but my colleague Sebastian Huber is currently working
on the MPC55xx BSP, also adapting it to the MPC5674F, see

http://wiki.rtems.org/wiki/index.php/MPC5674FEVB

But you should upgrade to the latest git version of RTEMS to work with it.

wkr,

Thomas.


The mpc55xx BSP is a framework, including adaptions for various boards
and chips, therefore it won't compile for a target "mpc55xx".
Am 30.12.2011 21:38, schrieb Andrei Chichak:
Hey all,

A while back I got a hand from Chris to get started with the MC68332,
then Eric Norum helped me get a BSP going for the Axiom CML5282, again
68K based.

Now I am getting going on one of Freescale's MPC5634M StarterTRAK dev
board. It uses a PPC with hardware floating point.

I am currently using RTEMS 4.9.4 on a windows VM using MINGW/MSYS. I
didn't move up to 4.10 yet, since it wasn't stable when I was working on
that project.

To get the basics going I used:

cd powerpc
../rtems-4.9.4/configure --target=powerpc-rtems4.9 \
--disable-itron \
--enable-posix \
--enable-rtemsbsp='mpc55xx' \
--enable-networking \
--disable-tests \
--prefix /c/opt/src/rtems/powerpc

make all install

since I figured that a 55xx would be within a million miles. This is
basically the same command that I use to set up for the m68k based BSPs
(with appropriate changes of parameters of course).

For some odd reason it looks like the 55xx BSP wasn't installed
properly, since it didn't compile the libraries for a couple of hours.

It also looks like the distribution has the mpc55xx BSP and mpc55xxevb,
but it looks like portions are missing.

Is there someone using either the 5632M or a 55xx that can suggest a
better config command to get started?



Sorry that I'm not as concise and complete as you might hope, my first
language isn't German :-)

Andrei


_______________________________________________
rtems-users mailing list
rtems-users-WKV1R6NGPIzYtjvyW6yDsg@public.gmane.org
http://www.rtems.org/mailman/listinfo/rtems-users


--
--------------------------------------------
Embedded Brains GmbH
Thomas Doerfler           Obere Lagerstr. 30
D-82178 Puchheim          Germany
email: Thomas.Doerfler <at> embedded-brains.de
Phone: +49-89-18908079-2
Fax:   +49-89-18908079-9

_______________________________________________
rtems-users mailing list
rtems-users@...
http://www.rtems.org/mailman/listinfo/rtems-users
Ralf Corsepius | 3 Jan 2012 07:46
Favicon

Re: MPC5634M starting point

On 01/03/2012 07:26 AM, Andrei Chichak wrote:

> Is there anything weird that I would have to do to configure the ppc
> version of GCC to generate appropriate code? I will be using a e200z3
> processor, whereas the 5674F uses an e200z7.
>
> I'm probably doing something wrong, but right out of the box there is a
> problem with the mpc5566evb BSP, wherein when I compile a simple printf
> hello world test program, the compile bombs with:
>
> c:/opt/rtems-4.10/bin/../lib/gcc/powerpc-rtems4.10/4.4.6/../../../../powerpc-rtems4.10/bin/ld.exe:linkcmds.mpc5566evb:3:
> warning: redeclaration of memory region `ROM'
> c:/opt/rtems-4.10/bin/../lib/gcc/powerpc-rtems4.10/4.4.6/../../../../powerpc-rtems4.10/bin/ld.exe:linkcmds.mpc5566evb:4:
> warning: redeclaration of memory region `RAM'
> c:/opt/rtems-4.10/bin/../lib/gcc/powerpc-rtems4.10/4.4.6/../../../../powerpc-rtems4.10/bin/ld.exe:linkcmds.mpc5566evb:5:
> warning: redeclaration of memory region `RAM_EXT'
> c:/opt/rtems-4.10/bin/../lib/gcc/powerpc-rtems4.10/4.4.6/../../../../powerpc-rtems4.10/bin/ld.exe:linkcmds.mpc5566evb:6:
> warning: redeclaration of memory region `NIRVANA'
> c:/opt/rtems-4.10/bin/../lib/gcc/powerpc-rtems4.10/4.4.6/../../../../powerpc-rtems4.10/bin/ld.exe:linkcmds.mpc5566evb:23:
> error: redefinition of memory region alias `REGION_TEXT'
>
> It looks like the linker files need a bit of work.

Is this with CVS-HEAD?

Building CVS-HEAD with rtems-4.10 tools is not supported. You need to to 
use the rtems-4.11 tools.

Ralf

_______________________________________________
rtems-users mailing list
rtems-users@...
http://www.rtems.org/mailman/listinfo/rtems-users

Andrei Chichak | 3 Jan 2012 08:07
Picon

Re: MPC5634M starting point



Is this with CVS-HEAD?

No 4.10.2.  Straight out of the box. Latest version of the 4.10 tools. 

System config command looks like:

../rtems-4.10.2/configure --target=powerpc-rtems4.10 \
--disable-itron \
--enable-posix \
--enable-rtemsbsp='mpc5566evb' \
--enable-networking \
--disable-tests \
--prefix /c/opt/src/rtems/powerpc

make all install


Project config command looks like:

cd ../build
../src/configure --host=powerpc-rtems4.10 \
--prefix=/c/opt/src/rtems/powerpc \
--enable-rtemsbsp=mpc5566evb

code looks like:

#include <stdio.h>

int
main()
{
  printf ("Example program main\n");
  return 0;
}


Building CVS-HEAD with rtems-4.10 tools is not supported. You need to to use the rtems-4.11 tools.

Ralf


What I'm trying to do is use the latest supported base (4.10.2), create a new BSP for a new board from something that is within a million miles (done that twice now),  and get something to compile.

Thanks,
Andrei
_______________________________________________
rtems-users mailing list
rtems-users@...
http://www.rtems.org/mailman/listinfo/rtems-users
Sebastian Huber | 3 Jan 2012 10:18
Picon
Favicon

Re: MPC5634M starting point

Hello Andrei,

I have a lot of local changes for the MPC55XX and MPC56XX BSPs.  In particular 
the support for the MPC5643L.  It is now quite easy to integrate new MPC56XX 
variants.  The big problem currently is the the Git repository is not available 
yet and the CVS is read only.  You have to wait until we have the Git 
repository up and running.  Once this is available you can use the latest 
development version of RTEMS with the 4.11 tool chain.  Everything else is a 
waste of time considering the massive changes for these BSPs since RTEMS 4.10.2.

Kind regards,
	Sebastian

--

-- 
Sebastian Huber, embedded brains GmbH

Address : Obere Lagerstr. 30, D-82178 Puchheim, Germany
Phone   : +49 89 18 90 80 79-6
Fax     : +49 89 18 90 80 79-9
E-Mail  : sebastian.huber@...
PGP     : Public key available on request.

Diese Nachricht ist keine geschäftliche Mitteilung im Sinne des EHUG.
_______________________________________________
rtems-users mailing list
rtems-users@...
http://www.rtems.org/mailman/listinfo/rtems-users

Joel Sherrill | 11 Jan 2012 16:27
Gravatar

git conversion status

Hi,

Thanks to the holidays and different people taking
holiday at different times, the final sysadmin
work on the git conversion didn't get done until
this week.

It looks like everything is in place for git but
Chris is still on holiday until Jan 12. I expect
that once he gets home, he will be online shortly
after. Given that it is about 2am on Jan 12 in
Sidney as I write this, Amar and I would like to
delay going live until he gets a chance to review
things.

Sorry this has taken so long but we all know how
the holidays are and we all wanted to be 100%
confident that things were right.

--

-- 
Joel Sherrill, Ph.D.             Director of Research&  Development
joel.sherrill@...        On-Line Applications Research
Ask me about RTEMS: a free RTOS  Huntsville AL 35805
    Support Available             (256) 722-9985

_______________________________________________
rtems-users mailing list
rtems-users@...
http://www.rtems.org/mailman/listinfo/rtems-users

Dmitry Shapovalov | 12 Jan 2012 04:54
Picon

initial port to stm32f4-discovery (cortex-m4)

Hi All

I made the initial port of rtems on stm32f4-discovery with cortex-m4 cpu onboard. for now, it has only ugly, polling mode, usart driver(pins pc6,pc7). checked 'hello world' and posix thread examples. working well. took as an example lm3s69xx. I'm not sure that irq and clock-config parts are correct.
I will appreciate any help.

https://gitorious.org/rtems_stm324f

With Best Regard.

Dmitry Shapovalov

_______________________________________________
rtems-users mailing list
rtems-users@...
http://www.rtems.org/mailman/listinfo/rtems-users
Sebastian Huber | 12 Jan 2012 08:42
Picon
Favicon

Re: initial port to stm32f4-discovery (cortex-m4)

Hi,

once we have the Git repository available we can integrate this BSP.  The IRQ 
and clock support should be shared among all ARMv7-M based BSPs.  I will take 
care about this, but I wait until the Git repository is available.

On 01/12/2012 04:54 AM, Dmitry Shapovalov wrote:
> Hi All
>
> I made the initial port of rtems on stm32f4-discovery with cortex-m4 cpu
> onboard. for now, it has only ugly, polling mode, usart driver(pins pc6,pc7).
> checked 'hello world' and posix thread examples. working well. took as an
> example lm3s69xx. I'm not sure that irq and clock-config parts are correct.
> I will appreciate any help.
>
> https://gitorious.org/rtems_stm324f
>
> With Best Regard.
>
> Dmitry Shapovalov
>
>
>
> _______________________________________________
> rtems-users mailing list
> rtems-users <at> rtems.org
> http://www.rtems.org/mailman/listinfo/rtems-users

--

-- 
Sebastian Huber, embedded brains GmbH

Address : Obere Lagerstr. 30, D-82178 Puchheim, Germany
Phone   : +49 89 18 90 80 79-6
Fax     : +49 89 18 90 80 79-9
E-Mail  : sebastian.huber <at> embedded-brains.de
PGP     : Public key available on request.

Diese Nachricht ist keine geschäftliche Mitteilung im Sinne des EHUG.
_______________________________________________
rtems-users mailing list
rtems-users <at> rtems.org
http://www.rtems.org/mailman/listinfo/rtems-users

Gmane