Richard Henderson <[email protected]> writes: > On 04/29/2014 08:41 AM, Alex Bennée wrote: >> >> Richard Henderson <[email protected]> writes: >>> On 04/29/2014 04:25 AM, Alex Bennée wrote: >>>>> + *pc = (*pc & ~0x3fffffc) | reloc_pc24_val(pc, target); >>>> >>>> Should 0x3fffffc be a #define'd mask? Can the PC ever actually be >>>> non-word aligned? >>> <snip> >> Ahh ok that makes sense. I was thrown because the masking pattern seems >> to occur all around the ppc tcg code: >> <snip> > Well, its true that it's going to be replicated between the disassembler and > the two ppc backends. I'm slightly surprised that it appears more than twice > for each ppc backend, but I suppose that just means there's more room to tidy > up. > > I do think that's out of scope for this patch set though.
Agreed. -- Alex Bennée
