This is the mail archive of the gcc-patches@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]
Other format: [Raw text]

Re: [PATCH 0/9] Cleanups to i386 prologue/epilogue generation


On 08/03/2010 04:53 PM, Richard Henderson wrote:
> Tested on x86_64-linux.  There are no code generation changes within
> x86_64 cc1 after the patch series.  Final testing on i686-linux is
> yet to complete; I'll check on that tomorrow morning.

I committed this patch series after fixing a bug on i686.

  /* Special care must be taken for the normal return case of a function
     using eh_return: the eax and edx registers are marked as saved, but
     not restored along this path.  Adjust the save location to match.  */
  if (crtl->calls_eh_return && style != 2)
    frame.reg_save_offset -= 2 * UNITS_PER_WORD;

This hunk had gotten lost somehow from patch 7, ix86_expand_epilogue.
(Probably I did something stupid and synced things the wrong way between
machines during testing.) Re-adding it fixes a few exception failures
for 32-bit.

I also consulted with Jack Howarth about the darwin oddity.  He has filed
a radar bug report against the linker.  Since the resulting executables
work, he agreed to add the linker warning to the pruning if a linker fix
is not forthcoming.


r~


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