[Mlir-commits] [mlir] [mlir][ArmNeon] Update `LowerContractionToSMMLAPattern` to support proper unrolling for k dimension (PR #88591)

[Andrzej Warzyński]

================
@@ -269,3 +269,80 @@ func.func @test_lower_vector_arm_neon_matvec(%lhs: vector<8x8xi8>, %rhs: vector<
   %res = vector.contract {indexing_maps = [affine_map<(d0, d1) -> (d0, d1)>,affine_map<(d0, d1) -> (d1)>, affine_map<(d0, d1) -> (d0)>], iterator_types = ["parallel", "reduction"], kind = #vector.kind<add>} %lhs_extsi, %rhs_extsi, %acc : vector<8x8xi32>, vector<8xi32> into vector<8xi32>
   return %res : vector<8xi32>
 }
+
+
+// -----
+
+// CHECK-LABEL:   func.func @test_lower_vector_arm_neon_k_unroll(
----------------
banach-space wrote:

[nit] Skip "test" and "lower". We know it's a test (it's in test directory) and we know that it's for lowering (based on the file name). 

https://github.com/llvm/llvm-project/pull/88591