820-no-mips-empic-relocs.patch 2.2 KB

1234567891011121314151617181920212223242526272829303132333435363738394041424344454647484950515253545556575859
  1. From: cgd at broadcom dot com
  2. To: gcc-patches at gcc dot gnu dot org
  3. Cc: mark at codesourcery dot com
  4. Date: 13 Jun 2004 22:51:30 -0700
  5. Subject: [trunk + 3.4-branch RFA] don't use empic relocs for mips-linuxeh
  6. This patch changes mips-linux to avoid using embedded-pic relocs for
  7. its eh data. (Support for generating these for new code is removed in
  8. current binutils srcs.)
  9. Relating to this, previously, mips-linux and mips64-linux would use
  10. different representations for their EH data (even for mips64-linux o32
  11. abi), due to the mips64-linux n32/64 BFDs not supporting the
  12. embedded-pic relocs. This was a bug.
  13. For more explanation, see the thread of the URL quoted in the comment
  14. in linux.h.
  15. Tested the same w/ sources of about a week ago for c/c++ for
  16. mips-linux (native) before/after. Also verified .o compatibility
  17. before/after just to be sure.
  18. I'd like this approved for the branch as well, so 3.4.1 will work
  19. nicely w/ the next major binutils release.
  20. thanks,
  21. chris
  22. 2004-06-13 Chris Demetriou <cgd@broadcom.com>
  23. * config/mips/linux.h (ASM_PREFERRED_EH_DATA_FORMAT): Redefine
  24. to return DW_EH_PE_absptr.
  25. Index: config/mips/linux.h
  26. ===================================================================
  27. RCS file: /cvs/gcc/gcc/gcc/config/mips/linux.h,v
  28. retrieving revision 1.77
  29. diff -u -p -r1.77 linux.h
  30. --- gcc/gcc/config/mips/linux.h 19 Feb 2004 22:07:51 -0000 1.77
  31. +++ gcc/gcc/config/mips/linux.h 14 Jun 2004 05:49:51 -0000
  32. @@ -170,10 +170,11 @@ Boston, MA 02111-1307, USA. */
  33. #undef FUNCTION_NAME_ALREADY_DECLARED
  34. #define FUNCTION_NAME_ALREADY_DECLARED 1
  35. -#define ASM_PREFERRED_EH_DATA_FORMAT(CODE, GLOBAL) \
  36. - (flag_pic \
  37. - ? ((GLOBAL) ? DW_EH_PE_indirect : 0) | DW_EH_PE_pcrel | DW_EH_PE_sdata4\
  38. - : DW_EH_PE_absptr)
  39. +/* If possible, we should attempt to use GP-relative relocs for this
  40. + (see <a href="http://sources.redhat.com/ml/binutils/2004-05/msg00227.html">http://sources.redhat.com/ml/binutils/2004-05/msg00227.html</a>).
  41. + However, until that is implement, this just uses standard, absolute
  42. + references. */
  43. +#define ASM_PREFERRED_EH_DATA_FORMAT(CODE, GLOBAL) DW_EH_PE_absptr
  44. /* The glibc _mcount stub will save $v0 for us. Don't mess with saving
  45. it, since ASM_OUTPUT_REG_PUSH/ASM_OUTPUT_REG_POP do not work in the