This is the mail archive of the
mailing list for the GCC project.
Re: [PATCH] Fix PR middle-end/28690, indexed load/store performance + reload bug
How about if we back out the change to swap_commutative_operands_p
and instead modify commutative_operand_precedence to return slightly
different values for REGs versus REGs with REG_POINTER set?
The problem is that REG and MEM, to pick one, have the same precedence.
See my other message for what is really needed for this to work (and
to dump swap_commutative_operands_with_target). This change will have
to be SPEC-tested on x86, I can take care of that.