RE: Possible IRA improvements for irregular register architectures

2010-01-11 Thread Ian Bolton
-Original Message- > From: gcc-ow...@gcc.gnu.org [mailto:gcc-ow...@gcc.gnu.org] On Behalf Of > Ian Bolton > Sent: 04 January 2010 14:19 > To: gcc@gcc.gnu.org > Subject: RE: Possible IRA improvements for irregular register > architectures > > Happy New Year! > >

RE: Possible IRA improvements for irregular register architectures

2010-01-04 Thread Ian Bolton
> -Original Message- > From: gcc-ow...@gcc.gnu.org [mailto:gcc-ow...@gcc.gnu.org] On Behalf Of > Ian Bolton > Sent: 18 December 2009 15:34 > To: gcc@gcc.gnu.org > Subject: Possible IRA improvements for irregular register architectures > > Let's assume I have two

Possible IRA improvements for irregular register architectures

2009-12-18 Thread Ian Bolton
Let's assume I have two sub-classes of ALL_REGS: BOTTOM_REGS (c0-c15) and TOP_CREGS (c16-c31). Let's also assume I have two main types of instruction: A-type Instructions, which can use ALL 32 registers, and B-type Instructions, which can only use the 16 BOTTOM_REGS. IRA will correctly calculate