[Mlir-commits] [mlir] [mlir][scf] Add parallelLoopUnrollByFactors() (PR #163806)

[llvmlistbot at llvm.org]

================
@@ -0,0 +1,148 @@
+// RUN: mlir-opt %s -test-parallel-loop-unrolling='unroll-factors=1,2' -split-input-file | FileCheck %s
+// RUN: mlir-opt %s -test-parallel-loop-unrolling='unroll-factors=1,2 loop-depth=1' -split-input-file | FileCheck %s --check-prefix CHECK-UNROLL-INNER
+// RUN: mlir-opt %s -test-parallel-loop-unrolling='unroll-factors=3,1' -split-input-file | FileCheck %s --check-prefix CHECK-UNROLL-BY-3
+
+func.func @unroll_simple_parallel_loop(%arg0: memref<1x16x12xf32>, %arg1: memref<1x16x12xf32>) {
+  %c12 = arith.constant 12 : index
+  %c16 = arith.constant 16 : index
+  %c0 = arith.constant 0 : index
+  %c1 = arith.constant 1 : index
+  scf.parallel (%arg2, %arg3, %arg4) = (%c0, %c0, %c0) to (%c1, %c16, %c12) step (%c1, %c1, %c1) {
+    %read = memref.load %arg0[%arg2, %arg3, %arg4] : memref<1x16x12xf32>
+    memref.store %read, %arg1[%arg2, %arg3, %arg4] : memref<1x16x12xf32>
+    scf.reduce
+  }
+  return
+}
+
+// CHECK-LABEL:   func @unroll_simple_parallel_loop
+// CHECK-SAME:     ([[ARG0:%.*]]: memref<1x16x12xf32>, [[ARG1:%.*]]: memref<1x16x12xf32>)
+// CHECK:           scf.parallel ([[IV0:%.*]], [[IV1:%.*]], [[IV2:%.*]]) = (%c0, %c0, %c0) to (%c1, %c16, %c12) step (%c1, %c1{{.*}}, %c2)
+// CHECK:             [[LOADED1:%.*]] = memref.load [[ARG0]][[[IV0]], [[IV1]], [[IV2]]] : memref<1x16x12xf32>
+// CHECK:             memref.store [[LOADED1]], [[ARG1]][[[IV0]], [[IV1]], [[IV2]]] : memref<1x16x12xf32>
+// CHECK:             [[UNR_Y:%.*]] = affine.apply {{.*}}([[IV2]])
----------------
fabrizio-indirli wrote:

Changed `UNR_Y` to `UNR_IV2` and added the expressions int he `affine.apply`

https://github.com/llvm/llvm-project/pull/163806