Re: [PATCH] rtx_costs vs. const_int

2015-07-06 Thread Jeff Law
On 06/14/2015 08:33 PM, Alan Modra wrote: This patch changes the targetm.rtx_costs interface to pass a mode parameter, and removes a redundant parameter. The reason for the change is that powerpc and other backends need the mode that a const_int is used in to properly determine the cost. For in

Re: [PATCH] rtx_costs vs. const_int

2015-06-15 Thread Alan Modra
On Mon, Jun 15, 2015 at 01:37:48PM -0500, Segher Boessenkool wrote: > > sub-rtx's in rtx_cost are > > summed in order rather than reverse order so that the mode from an > > earlier operand can be used for a later operand lacking a mode. This > > is for ZERO_EXTEND and similar codes where the sub-r

Re: [PATCH] rtx_costs vs. const_int

2015-06-15 Thread Segher Boessenkool
Hi Alan, On Mon, Jun 15, 2015 at 12:03:47PM +0930, Alan Modra wrote: > This patch changes the targetm.rtx_costs interface to pass a mode > parameter, and removes a redundant parameter. The reason for the > change is that powerpc and other backends need the mode that a > const_int is used in to pr