[llvm] [AMDGPU] Use reverse iteration in CodeGenPrepare (PR #145484)

[Matt Arsenault via llvm-commits]

================
@@ -1634,29 +1634,18 @@ define float @v_recip_sqrt_f32_ulp25_contract(float %x) {
 ; IR-IEEE-SDAG-LABEL: v_recip_sqrt_f32_ulp25_contract:
 ; IR-IEEE-SDAG:       ; %bb.0:
 ; IR-IEEE-SDAG-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
-; IR-IEEE-SDAG-NEXT:    s_mov_b32 s4, 0xf800000
-; IR-IEEE-SDAG-NEXT:    v_mul_f32_e32 v1, 0x4f800000, v0
+; IR-IEEE-SDAG-NEXT:    s_mov_b32 s4, 0x800000
----------------
arsenm wrote:

This looks better, so problem solved? 

https://github.com/llvm/llvm-project/pull/145484