[llvm] [AMDGPU] Replace zext pattern from reg_sequence to cvt_u32_u16 (PR #208045)

[Guo Chen via llvm-commits]

================
@@ -14,8 +14,7 @@ define i64 @srl_bv_v4i16_extract_last(i16 %a, i16 %b, i16 %c, i16 %d) {
 ; GFX1250:       ; %bb.0:
 ; GFX1250-NEXT:    s_wait_loadcnt_dscnt 0x0
 ; GFX1250-NEXT:    s_wait_kmcnt 0x0
-; GFX1250-NEXT:    v_mov_b16_e32 v0.h, 0
-; GFX1250-NEXT:    v_mov_b16_e32 v0.l, v3.l
+; GFX1250-NEXT:    v_cvt_u32_u16_e32 v0, v3.l
----------------
broxigarchen wrote:

example to shows code quality improvement

https://github.com/llvm/llvm-project/pull/208045