This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [PATH] PR 5126, fix for m68k PIC optimization bug
Richard Henderson <rth@redhat.com> writes:
|> On Mon, Dec 17, 2001 at 03:32:45AM +0100, Roman Zippel wrote:
|> > Hmm, this exists already on the gcc-2_95 branch, but somewhere got lost.
|> > [...searching...] This is the patch where it was changed:
|> > http://gcc.gnu.org/ml/gcc-patches/1999-09n/msg01147.html
|>
|> That would suggest that there was something wrong with using
|> regs_ever_live here. I hope Andreas can remember something
|> about this, since there is nothing accompanying the patch.
It was around this time when the pic register has been made fixed with
-fpic. Thus regs_ever_live[PIC_REGNUM] became meaningless. All these
variants of m68k_output_function_prologue and
m68k_output_function_epilogue must check
current_function_uses_pic_offset_table if they want to support -fpic in
the same way as the default definition.
Not making the pic register a fixed register does not work, IIRC this is
because references to the pic register can be created during reload when a
CONST_DOUBLE needs to be moved to the constant pool, and this is the first
time a pic reference is needed.
Andreas.
--
Andreas Schwab "And now for something
Andreas.Schwab@suse.de completely different."
SuSE Labs, SuSE GmbH, Schanzäckerstr. 10, D-90443 Nürnberg
Key fingerprint = 58CA 54C7 6D53 942B 1756 01D3 44D5 214B 8276 4ED5