cmd/compile/internal/arm: fix comparison & conditional branch for SSA on ARM

Progress on SSA for ARM. Still not complete. Now Fibonacci function compiles
and runs correctly.

The old backend swaps the operands for CMP instruction. This CL does the same
on SSA backend, and uses conditional branch accordingly.

Updates #15365.

Change-Id: I117e17feb22f03d936608bd232f76970e4bbe21a
Reviewed-on: https://go-review.googlesource.com/22187Reviewed-by: Keith Randall <khr@golang.org>

src/cmd/compile/internal/arm/ssa.go

@@ -91,8 +91,10 @@ func ssaGenValue(s *gc.SSAGenState, v *ssa.Value) {
 	case ssa.OpARMCMP:
 		p := gc.Prog(v.Op.Asm())
 		p.From.Type = obj.TYPE_REG
-		p.From.Reg = gc.SSARegNum(v.Args[0])
-		p.Reg = gc.SSARegNum(v.Args[1])
+		// Special layout in ARM assembly
+		// Comparing to x86, the operands of ARM's CMP are reversed.
+		p.From.Reg = gc.SSARegNum(v.Args[1])
+		p.Reg = gc.SSARegNum(v.Args[0])
 	case ssa.OpARMMOVWload:
 		p := gc.Prog(v.Op.Asm())
 		p.From.Type = obj.TYPE_MEM
@@ -142,7 +144,7 @@ func ssaGenBlock(s *gc.SSAGenState, b, next *ssa.Block) {
 	case ssa.BlockRet:
 		gc.Prog(obj.ARET)
 	case ssa.BlockARMLT:
-		p := gc.Prog(arm.ABGE)
+		p := gc.Prog(arm.ABLT)
 		p.To.Type = obj.TYPE_BRANCH
 		s.Branches = append(s.Branches, gc.Branch{p, b.Succs[0]})
 		p = gc.Prog(obj.AJMP)