1
0

i686_define_bug.patch 1.2 KB

1234567891011121314151617181920212223
  1. Fix for compiling glibc with -march=i686 specified.
  2. The patch below found at: http://permalink.gmane.org/gmane.linux.lfs.devel/9758,
  3. a nice discussion (with a comment of the glibc maintainer) can be found here:
  4. http://old.nabble.com/-RFC-PATCH--glibc-doesn%27t-build-with-%22gcc--march%3Di686%22-td17442608.html
  5. To me, this simply shows once more how fucked up things can be. E.g. the (related) bug report here:
  6. http://bugs.gentoo.org/show_bug.cgi?id=201815. There are lots of duplicates to it, many people are
  7. affected, one can find really elaborate reports. And the responsible gentoo developers are too
  8. elite to even point to the actual patch fixing the issue.
  9. diff -Nur glibc-2.16.0.orig/sysdeps/unix/sysv/linux/i386/sysdep.h glibc-2.16.0/sysdeps/unix/sysv/linux/i386/sysdep.h
  10. --- glibc-2.16.0.orig/sysdeps/unix/sysv/linux/i386/sysdep.h 2012-06-30 21:12:34.000000000 +0200
  11. +++ glibc-2.16.0/sysdeps/unix/sysv/linux/i386/sysdep.h 2012-08-01 17:56:05.000000000 +0200
  12. @@ -28,6 +28,10 @@
  13. #include <dl-sysdep.h>
  14. #include <tls.h>
  15. +#if defined __i686 && defined __ASSEMBLER__
  16. +#undef __i686
  17. +#define __i686 __i686
  18. +#endif
  19. /* For Linux we can use the system call table in the header file
  20. /usr/include/asm/unistd.h