This is the mail archive of the
mailing list for the GCC project.
Re: [4.2 only] RFA: PR 33848: reload rematerialisation of labels
> But that's the whole idea. There should be no cases in which register
> allocation (viewed as a black box) changes the JUMP_LABEL.
What do you mean by 'should'? 'Cannot but I'm not sure', 'Cannot modulo
bugs'? Because Andrew's change precisely does that and was OKed by RTH.
> I'm struggling to understand what your objection is.
My objection is that your patch is not the minimal one to fix the problem.
More specifically, I don't see the need for changing the code when the jump
has no JUMP_LABEL.