[llvm] [AMDGPU] Revert "Preliminary patch for divergence driven instruction selection. Operands Folding 1." (PR #71710)

[Jay Foad via llvm-commits]

================
@@ -2591,11 +2591,10 @@ define amdgpu_kernel void @v_test_v2i16_x_add_neg32_neg32(ptr addrspace(1) %out,
 ; SI-GISEL-NEXT:    s_waitcnt lgkmcnt(0)
 ; SI-GISEL-NEXT:    s_mov_b64 s[4:5], s[2:3]
 ; SI-GISEL-NEXT:    buffer_load_dword v2, v[0:1], s[4:7], 0 addr64
-; SI-GISEL-NEXT:    s_movk_i32 s2, 0xffe0
 ; SI-GISEL-NEXT:    s_waitcnt vmcnt(0)
 ; SI-GISEL-NEXT:    v_lshrrev_b32_e32 v3, 16, v2
-; SI-GISEL-NEXT:    v_add_i32_e32 v2, vcc, s2, v2
-; SI-GISEL-NEXT:    v_add_i32_e32 v3, vcc, s2, v3
+; SI-GISEL-NEXT:    v_add_i32_e32 v2, vcc, 0xffffffe0, v2
----------------
jayfoad wrote:

Folding improvement here.

https://github.com/llvm/llvm-project/pull/71710