Problem building RTEMS(m68k)

[Sudheesh J - Atinav Incorporated]

Sorry, I missed out that in my last mail. When I tried to verify the target
tool set, the invalid assembler option error(as mentioned in the
documentation) occured. I had installed the target tools from RPMs. Right
now I'm trying the other way, that's using sources+ rtems patches.
    Do I need to apply the patches even if I installed the target tools from
RPMS? I had the idea that the RPMS contain tools that are already patched.
regards
Sudheesh.

----- Original Message -----
From: "Joel Sherrill" <joel.sherrill at OARcorp.com>
To: "Sudheesh J" <sudheesh at ATINAV.COM>
Cc: <rtems-users at oarcorp.com>
Sent: Monday, October 28, 2002 6:48 PM
Subject: Re: Problem building RTEMS(m68k)


>
>
> > Sudheesh J wrote:
> >
> > Hi,
> >         I'm new to both Makefiles and RTEMS. I found the documentation
> > on RTEMS site very helpful. I followed the steps, though had some
> > glitches in btw.
> > 1) While installing the m68k-gdb, got a failed dependency,
> > libncurses.so.4. I had libncurses.so.5 and forced the installation
> > using --nodeps.
> > 2) The listing of /opt/rtems/bin after the rpm installations was
> > missing the c preprocessor, m68k-rtems-cpp.
> >
> > 3) The confgure command,
> > ../rtems-ss-20021007/configure --target=m68k-rtems
> > --enable-rtemsbsp=efi332 \
> > --disable-itron --enable-cxx --enable-tests --enable-docs
> > --enable-networking
> > Went alright, but make produced the following output:( relevant parts
> > only)
> >
> > Making all in cpu
> > gmake[5]: Entering directory
> > `/root/rtems/build_m68k/m68k-rtems/c/efi332/exec/score/cpu'
> > Making all in m68k
> > gmake[6]: Entering directory
> > `/root/rtems/build_m68k/m68k-rtems/c/efi332/exec/score/cpu/m68k'mkdir
> > o-optimize
> > gcc --pipe -isystem ../../../../../.././efi332/lib/include
> > -O4 -fomit-frame-pointer
> > -g -Wall -mcpu32  -o o-optimize/cpu.o -c
> >
../../../../../../../../rtems-ss-20021007/c/src/../../cpukit/score/cpu/m68k/
cpu.c
> > In file included from
> > ../../../../../../efi332/lib/include/rtems/score/cpu.h:23,
> >                  from
> > ../../../../../../efi332/lib/include/rtems/system.h:138,
> >                  from
> >
../../../../../../../../rtems-ss-20021007/c/src/../../cpukit/score/cpu/m68k/
cpu.c:14:
> > ../../../../../../efi332/lib/include/rtems/score/m68k.h:228:2: #error
> > "Unsupported CPU model
> > -- are you sure you're running a 68k compiler?"
> > cc1: Invalid option `cpu32'
>
>
> ^^^^^^^  There's the big hint.  Now go back a few lines and notice the
> name
> of the compiler.  It is your native one not a m68k one.  Walk through
> the items
> in Chapter 5 -- "Building RTEMS" in the Getting Started manual.
>
>
http://www.oarcorp.com/rtems/doc-current/share/rtems/html/started/started000
24.html
>
> >
../../../../../../../../rtems-ss-20021007/c/src/../../cpukit/score/cpu/m68k/
cpu.c:
> > In function `_CPU_ISR_install_raw_handler':
> >
../../../../../../../../rtems-ss-20021007/c/src/../../cpukit/score/cpu/m68k/
cpu.c:110:
> > warning: assignment makes pointer from integer without a cast
> > {standard input}: Assembler messages:
> > {standard input}:344: Error: invalid character '.' in mnemonic
> > gmake[6]: *** [o-optimize/cpu.o] Error 1
> > gmake[6]: Leaving directory
> > `/root/rtems/build_m68k/m68k-rtems/c/efi332/exec/score/cpu/m68k'
> > gmake[5]: *** [all-recursive] Error 1
> > gmake[5]: Leaving directory
> > `/root/rtems/build_m68k/m68k-rtems/c/efi332/exec/score/cpu'
> > gmake[4]: *** [all-recursive] Error 1
> > gmake[4]: Leaving directory
> > `/root/rtems/build_m68k/m68k-rtems/c/efi332/exec/score'
> > gmake[3]: *** [all-recursive] Error 1
> > gmake[3]: Leaving directory
> > `/root/rtems/build_m68k/m68k-rtems/c/efi332/exec'
> > gmake[2]: *** [all-recursive] Error 1
> > gmake[2]: Leaving directory
> > `/root/rtems/build_m68k/m68k-rtems/c/efi332'
> > gmake[1]: *** [all-recursive] Error 1
> > gmake[1]: Leaving directory `/root/rtems/build_m68k/m68k-rtems/c'
> > make: *** [all-recursive] Error 1
> >
> > I had exported /opt/rtems/bin in PATH, but why the script still
> >  complaining abt a 68k compiler?
> > Is the missing m68k preprocessor causing the problem?  Or is it
> > because of a wrong configuration assumption that the -mcpu32 was used?
> >         Expecting your comments,
> >     Sudheesh
>
> --
> Joel Sherrill, Ph.D.             Director of Research & Development
> joel at OARcorp.com                 On-Line Applications Research
> Ask me about RTEMS: a free RTOS  Huntsville AL 35805
> Support Available                (256) 722-9985