[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Another problem with egcs-1.0.2-prerelease

Jacob W Janovetz <janovetz at ews.uiuc.edu> wrote:
> Did you do the:
> rm -f xlimits.h *.a
> make CFLAGS="-O2 -g" libgcc.a all install
> as Joel mentioned in the INSTRUCTIONS?
> Cheers,
> Jake

Yes, but that doesn't have anything to do with the fact that -m68040  
doesn't predefine any symbols, and that -mcpu32 -ansi predefines  
only __mcpu32__, and that a compiler to be used with RTEMS should  

Actually, I used the old `bit' script to build all the tools.  Which  
brings up another point.  Joel, you don't have to remove any fortran  
or objective-C source directories to build cross-compilers.  Just  
add LANGUAGES="c c++" to the bit script.  Or perhaps you've already  
done this and I'm just lagging behind.

gmake -w CC=gcc CFLAGS="-O2 -g -pipe -DTARGET_MEM_FUNCTIONS=1"  
LANGUAGES="c c++" all install
check_fatal $?  "**** FAILED **** step 1"

# Regenerate the limits.h now that we have that the include_next will work
cd gcc
rm -f xlimits.h *.a
gmake -w CC=gcc CFLAGS="-O2 -g -pipe -DTARGET_MEM_FUNCTIONS=1"  
LANGUAGES="c c++" libgcc.a all install
check_fatal $?  "**** FAILED **** step 2"

Eric Norum                                 eric at skatter.usask.ca
Saskatchewan Accelerator Laboratory        Phone: (306) 966-6308
University of Saskatchewan                 FAX:   (306) 966-6058
Saskatoon, Canada.