[csl-asm?] PR middle-end/11821: Tweak arm_rtx_costs_1

[Richard Earnshaw]

> Unfortunately, it looks like the special RTXen it generates to query the
> backend, PLUS (reg, MULT (reg, const_int)), etc..., aren't treated or
> handled specially by the ARM backend's arm_rtx_costs_1 at all.  The x86
> backend's ix86_rtx_costs, on the other hand, does check for these
> patterns, makes sure the constant multiplier is 2, 4 or 8 and returns
> the CPU's cost for using a single "lea" instruction.


What makes you say this?  I've just had a look at the code and init_expmed 
gets a cost of 1 for all of add_cost, shiftadd_cost[*] and 
shiftsub_cost[*].

So it looks to me as though the costs are about right.  It is certainly 
spotting this case (look at the way PLUS falls through into AND, IOR etc).

R.