[issue4753] Faster opcode dispatch on gcc

[Skip Montanaro]

Skip Montanaro <skip at pobox.com> added the comment:

Antoine> I fear that with a configure option, disabled by default, the
    Antoine> code will get very poor testing and perhaps get broken in some
    Antoine> subtle way without anyone noticing.

That can be fixed by enabling that option on the buildbots where it is
presumed to help.  I see more slowdown on PPC than people are reporting as
speedups on Intel.  Is there something I can do to help debug the problem?
It doesn't appear the Apple version of gcc supports the -fno-crossjumping
flag.  If I dump the assembler code for ceval.c will that help others debug
the problem?

Skip

_______________________________________
Python tracker <report at bugs.python.org>
<http://bugs.python.org/issue4753>
_______________________________________