This is the mail archive of the gcc-bugs@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]

2.95.2


    Hi !


I downloaded gcc-2.95.2 in sometime February and compiled it. 
I have been using it ever since, and do not remember any problem.
This weekend a got a program mc-4.1.35-A10, which is a sibling of the
well known midnight commander. 

    The compilation of this package revealed a very strange bug in gcc.
As it turned out, if I use any optimization, the preprocessor puts a very long
line of garbage in just one of the programs. If  optimization is not used
everything is OK.


==========================
gcc -c -I..  -I../../src/../vfs -I../../src/.. -I.. -DBINDIR=\""/usr/local/bin/"\" -DLIBDIR=\""/usr/local/lib/mc/"\" -DICONDIR=\""/usr/local/share/icons/mc/"\"
 -DLOCALEDIR=\""/usr/local/share/locale/"\"  -DREGEX_MALLOC -DHAVE_CONFIG_H -Wall -O2 -mcpu=k6 -malign-loops=2 -malign-jumps=2 -malign-functions=2 -fno-strength-reduce -Ino/include ../../src/mountlist.c
 ../../src/mountlist.c:105: parse error before `__extension__'
 make[1]: *** [mountlist.o] Error 1
 make[1]: Leaving directory `/l2/src/src/mc-4.1.35-A10/Compile1/src'
 make: *** [all] Error 1
  
=======================


I might be using strange or unnecessery flags, but they are not affecting
the problem. Only the -O what counts.

I think I attached all the necessary files to test this out. Or you might 
download the  whole package from its home site. I attached the 
README file as well.
 
 My system :
     kernel 2.2.16
     glibc 2.1.3

 yours truly
 

Holluby Istvan Budapest
(holluby@interware.hu)

mountlist.tar.bz2


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]