[mlir][gpu] NFC: Change gpu.launch_func ops to custom format.

[csigg]

Summary

This should fix the reason for the failures after ec7780ebdab480139596c3cb08ee77d7035457b3. I will roll forward in a separate change.

mlir/test/Conversion/SPIRVToLLVM/lower-host-to-llvm-calls.mlir

gpu.module @foo {
     gpu.func @bar(%arg0: memref<6xi32>) kernel attributes {spv.entry_point_abi = {local_size = dense<1> : vector<3xi32>}} {
       gpu.return
     }
   }
 
   func @main() {
     %buffer = alloc() : memref<6xi32>
     %one = constant 1 : index
-    "gpu.launch_func"(%one, %one, %one,
-                      %one, %one, %one,
-                      %buffer) {kernel = @foo::@bar} : (index, index, index, index, index, index, memref<6xi32>) -> ()
+    gpu.launch_func @foo::@bar blocks in (%one, %one, %one)
+        threads in (%one, %one, %one) args(%buffer : memref<6xi32>)
     return
   }
 }

mlir/test/mlir-spirv-cpu-runner/double.mlir

func @main() {
     %four = constant 4 : i32
     %zero = constant 0 : i32
     %input_casted = memref_cast %input : memref<6xi32> to memref<?xi32>
     %output_casted = memref_cast %output : memref<6xi32> to memref<?xi32>
     call @fillI32Buffer(%input_casted, %four) : (memref<?xi32>, i32) -> ()
     call @fillI32Buffer(%output_casted, %zero) : (memref<?xi32>, i32) -> ()
 
     %one = constant 1 : index
-    "gpu.launch_func"(%one, %one, %one,
-                      %one, %one, %one,
-                      %input, %output) { kernel = @kernels::@double }
-        : (index, index, index, index, index, index, memref<6xi32>, memref<6xi32>) -> ()
+    gpu.launch_func @kernels::@double
+        blocks in (%one, %one, %one) threads in (%one, %one, %one)
+        args(%input : memref<6xi32>, %output : memref<6xi32>)
     %result = memref_cast %output : memref<6xi32> to memref<*xi32>
     call @print_memref_i32(%result) : (memref<*xi32>) -> ()
     return
   }
 
   func @fillI32Buffer(%arg0 : memref<?xi32>, %arg1 : i32)
   func @print_memref_i32(%ptr : memref<*xi32>)
 }

mlir/test/mlir-spirv-cpu-runner/simple_add.mlir

func @main() {
     %input1_casted = memref_cast %input1 : memref<3xf32> to memref<?xf32>
     %input2_casted = memref_cast %input2 : memref<3x3xf32> to memref<?x?xf32>
     %output_casted = memref_cast %output : memref<3x3x3xf32> to memref<?x?x?xf32>
     call @fillF32Buffer1D(%input1_casted, %3) : (memref<?xf32>, f32) -> ()
     call @fillF32Buffer2D(%input2_casted, %4) : (memref<?x?xf32>, f32) -> ()
     call @fillF32Buffer3D(%output_casted, %0) : (memref<?x?x?xf32>, f32) -> ()
 
     %one = constant 1 : index
-    "gpu.launch_func"(%one, %one, %one,
-                      %one, %one, %one,
-                      %input1, %input2, %output) { kernel = @kernels::@sum }
-        : (index, index, index, index, index, index, memref<3xf32>, memref<3x3xf32>, memref<3x3x3xf32>) -> ()
+    gpu.launch_func @kernels::@sum
+        blocks in (%one, %one, %one) threads in (%one, %one, %one)
+        args(%input1 : memref<3xf32>, %input2 : memref<3x3xf32>, %output : memref<3x3x3xf32>)
     %result = memref_cast %output : memref<3x3x3xf32> to memref<*xf32>
     call @print_memref_f32(%result) : (memref<*xf32>) -> ()
     return
   }
   func @fillF32Buffer1D(%arg0 : memref<?xf32>, %arg1 : f32)
   func @fillF32Buffer2D(%arg0 : memref<?x?xf32>, %arg1 : f32)
   func @fillF32Buffer3D(%arg0 : memref<?x?x?xf32>, %arg1 : f32)
   func @print_memref_f32(%arg0 : memref<*xf32>)
 }