This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.
Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
---|---|---|
Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |
Other format: | [Raw text] |
I presume you're not going to try and deal with cases where the allocno's mode requires > 2 registers (most obvious example is XFmode in general registers on x86, but there are others). I suspect that the gain for handling > 2 hard regs is so small that it's not worth the effort.This introduces a new structure, ira_object_t, which is split off from ira_allocno_t. Objects are used to track information related to conflicts. There is at the moment a 1:1 correspondence between objects and allocnos, but the plan is to introduce 2 objects for DImode allocnos with a suitable cover class.
Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
---|---|---|
Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |