[llvm] [VectorCombine] isExtractExtractCheap - specify the extract/insert shuffle mask to improve shuffle costs (PR #114780)

[David Green via llvm-commits]

================
@@ -688,9 +688,9 @@ define i32 @load_multiple_extracts_with_constant_idx(ptr %x) {
 define i32 @load_multiple_extracts_with_constant_idx_profitable(ptr %x) {
 ; CHECK-LABEL: @load_multiple_extracts_with_constant_idx_profitable(
 ; CHECK-NEXT:    [[LV:%.*]] = load <8 x i32>, ptr [[X:%.*]], align 16
-; CHECK-NEXT:    [[E_0:%.*]] = extractelement <8 x i32> [[LV]], i32 0
-; CHECK-NEXT:    [[E_1:%.*]] = extractelement <8 x i32> [[LV]], i32 6
-; CHECK-NEXT:    [[RES:%.*]] = add i32 [[E_0]], [[E_1]]
+; CHECK-NEXT:    [[SHIFT:%.*]] = shufflevector <8 x i32> [[LV]], <8 x i32> poison, <8 x i32> <i32 6, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison>
----------------
davemgreen wrote:

What about the cost of VectorOpCost, if it will be a smaller vector that gets used for the binop? Will lowering that prevent this triggering elsewhere?

And do you know how this transform "may enable further optimization"? Maybe that would make this worth while anyway.

https://github.com/llvm/llvm-project/pull/114780