[Mlir-commits] [mlir] [mlir][Vector] Remove Vector{Load|Store}ToMemrefLoadLowering (PR #121454)

[Andrzej Warzyński]

================
@@ -3,10 +3,9 @@
   func.func @warp_extract(%arg0: index, %arg1: memref<1024x1024xf32>, %arg2: index, %arg3: vector<1xf32>) {
     %c0 = arith.constant 0 : index
     gpu.warp_execute_on_lane_0(%arg0)[32] {
-      // CHECK:%[[val:[0-9]+]] = llvm.extractelement
       // CHECK:%[[base:[0-9]+]] = llvm.extractvalue
       // CHECK:%[[ptr:[0-9]+]] = llvm.getelementptr %[[base]]
-      // CHECK:llvm.store %[[val]], %[[ptr]]
+      // CHECK:llvm.store %arg9, %[[ptr]] {alignment = 4 : i64} : vector<1xf32>, !llvm.ptr
----------------
banach-space wrote:

>  Should I take care of that in a separate PR, or just here?

I'd reduce the overall PR noise and remove it here. Please add a relevant note in the summary.

https://github.com/llvm/llvm-project/pull/121454