This is the mail archive of the gcc@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: ppc glibc 2.3 failure


On Monday, October 7, 2002, at 04:37  PM, Aldy Hernandez wrote:

This should not happen.  As the comments in
rs6000_ra_ever_killed indicate, its expectation is that code (in
particular,
copies into LR) added in the prolog have a REG_MAYBE_DEAD note, to
prevent
this problem.  I can't test it, but try this:

    if (save_LR_around_toc_setup)
      rs6000_maybe_dead (
      emit_move_insn (gen_rtx_REG (Pmode, LINK_REGISTER_REGNUM),
                      gen_rtx_REG (Pmode, 11)));
Yup, that fixes it.

Is this the correct solution?
Yes, I think so.  I ran into the same problem on Darwin and found
the REG_MAYBE_DEAD notes as a way to identify code added in the prolog.
I wouldn't object to a more elegant identification, but I don't know
of anything else that works.

If so, I can test out the patch overnight.
Thanks.


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