[llvm] [AMDGPU] Vectorize more 16 bit shuffles (PR #90648)

[Matt Arsenault via llvm-commits]

================
@@ -503,8 +503,8 @@ define void @shuffle() {
 ; GFX9-10-SIZE-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %v8i8_8 = shufflevector <8 x i8> undef, <8 x i8> undef, <8 x i32> <i32 2, i32 3, i32 0, i32 1, i32 2, i32 3, i32 0, i32 1>
 ; GFX9-10-SIZE-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %v16i8_16 = shufflevector <16 x i8> undef, <16 x i8> undef, <16 x i32> <i32 2, i32 3, i32 0, i32 1, i32 2, i32 3, i32 0, i32 1, i32 2, i32 3, i32 0, i32 1, i32 2, i32 3, i32 0, i32 1>
 ; GFX9-10-SIZE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v2i16_2 = shufflevector <2 x i16> undef, <2 x i16> undef, <2 x i32> <i32 1, i32 0>
-; GFX9-10-SIZE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %v4i16_4 = shufflevector <4 x i16> undef, <4 x i16> undef, <4 x i32> <i32 1, i32 3, i32 2, i32 0>
-; GFX9-10-SIZE-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %v8i16_8 = shufflevector <8 x i16> undef, <8 x i16> undef, <8 x i32> <i32 2, i32 3, i32 0, i32 1, i32 2, i32 3, i32 0, i32 1>
+; GFX9-10-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v4i16_4 = shufflevector <4 x i16> undef, <4 x i16> undef, <4 x i32> <i32 1, i32 3, i32 2, i32 0>
----------------
arsenm wrote:

Testing 2 undefs sources is a bit dubious - this is probably getting identified as a SK_PermuteSingleSrc, but should probably be treated as SK_PermuteTwoSrc. Should we fix these tests to stress the two cases? 

https://github.com/llvm/llvm-project/pull/90648