On 03/23/2011 02:40 PM, Nathan Froyd wrote: > I can look into just how many places might need to be fixed up because > of this, but scattering a bunch of ifs all over the place seemed less > elegant than handling it all in the emit* functions.
Yeah, maybe. On the other hand, the other complication that's going on there is attempting to provide a "sensible" return value in the case of emitting a NULL_RTX, and doing different things for insns vs non-insns. Merely removing the "different things" is surely an improvement. Getting rid of the "sensible" probably cleans things up more, but probably interferes more with reload. r~
