1rtl_to_x86:
2* recognise alub(X,X,sub,1,lt,L1,L2,P) and turn it into 'dec',
3  this might improve the reduction test code slightly (X is
4  the pseudo for FCALLS)
5* recognise alu(Z,X,add,Y) and turn it into 'lea'.
6* rewrite tailcalls as parallel assignments before regalloc
7
8x86:
9* Use separate constructors for real regs (x86_reg) and pseudos (x86_temp).
10
11Frame:
12* drop tailcall rewrite
13
14Registers:
15* make the 2 regs now reserved for frame's tailcall rewrite available for arg passing
16
17Optimizations:
18* replace jcc cc,L1; jmp L0; L1: with jcc <not cc> L0; L1: (length:len/2)
19* Kill move X,X insns, either in frame or finalise
20* Instruction scheduling module
21* We can now choose to not have HP in %esi. However, this currently loses
22  performance due to (a) repeated moves to/from P_HP(P), and (b) spills of
23  the temp that contains a copy of P_HP(P). Both of these problems should be
24  fixed, and then, if we don't have any noticeable performance degradation, we
25  should permanently change to a non-reserved HP strategy.
26
27Loader:
28
29Assembler:
30
31Encode:
32