[ET-VK][ez] Enable max_pool2d.default (#14044)

max_pool2d_with_indices is already implemented; this diff enables max_pool2d as well by just re-using the same implementation.

Differential Revision: [D81513446](https://our.internmc.facebook.com/intern/diff/D81513446/)

Author: pytorchbot

backends/vulkan/op_registry.py

@@ -434,6 +434,7 @@ def try_find_keepdim_arg(node: torch.fx.Node) -> bool:
 @update_features(
     [
         exir_ops.edge.aten.avg_pool2d.default,
+        exir_ops.edge.aten.max_pool2d.default,
         exir_ops.edge.aten.max_pool2d_with_indices.default,
     ]
 )

backends/vulkan/runtime/graph/ops/glsl/max_pool2d.glsl

@@ -24,6 +24,8 @@ ${layout_declare_ubo(B, "ivec2", "kernel_size", "ivec2", "stride", "ivec2", "pad
 
 layout(local_size_x_id = 0, local_size_y_id = 1, local_size_z_id = 2) in;
 
+${layout_declare_spec_const(C, "int", "write_indices", "1")}
+
 void main() {
   const ivec3 pos = ivec3(gl_GlobalInvocationID);
 
@@ -55,5 +57,7 @@ void main() {
   }
 
   imageStore(t_out, pos, out_texel);
-  imageStore(t_idx, pos, idx_texel);
+  if (write_indices > 0) {
+    imageStore(t_idx, pos, idx_texel);
+  }
 }

backends/vulkan/runtime/graph/ops/impl/Pool.cpp

@@ -76,8 +76,19 @@ void add_max_pool2d_node(
     const ValueRef dilation,
     const ValueRef ceil_mode,
     const ValueRef out) {
-  const auto out_val = graph.get_value_list(out);
-  const ValueRef out_tensor = out_val->at(0);
+  ValueRef out_tensor = out;
+  // Placeholder tensor to fill binding slot for indices tensor in case we are
+  // computing max_pool2d instead of max_pool2d_with_indices.
+  TmpTensor tmp_indices_tensor =
+      TmpTensor(&graph, {}, graph.dtype_of(in), graph.storage_type_of(in));
+  ValueRef indices_tensor = tmp_indices_tensor.vref;
+  int32_t write_indices = 0;
+  if (graph.val_is_value_list(out)) {
+    const auto out_val = graph.get_value_list(out);
+    out_tensor = out_val->at(0);
+    indices_tensor = out_val->at(1);
+    write_indices = 1;
+  }
 
   check_pool2d_args(graph, in, out_tensor);
 
@@ -98,7 +109,7 @@ void add_max_pool2d_node(
       default_pick_global_wg_size,
       default_pick_local_wg_size,
       // Inputs and Outputs
-      {{{out_val->at(0), out_val->at(1)}, vkapi::kWrite}, {in, vkapi::kRead}},
+      {{{out_tensor, indices_tensor}, vkapi::kWrite}, {in, vkapi::kRead}},
       // Shader params buffers
       {
           graph.logical_limits_ubo(out_tensor),
@@ -108,7 +119,7 @@ void add_max_pool2d_node(
       // Push Constants
       {},
       // Specialization Constants
-      {},
+      {write_indices},
       // Resize Args
       {kernel_size, stride, padding, dilation, ceil_mode},
       // Resizing Logic
@@ -203,6 +214,7 @@ void avg_pool2d(ComputeGraph& graph, const std::vector<ValueRef>& args) {
 REGISTER_OPERATORS {
   VK_REGISTER_OP(aten.avg_pool2d.default, avg_pool2d);
   VK_REGISTER_OP(aten.max_pool2d_with_indices.default, max_pool2d);
+  VK_REGISTER_OP(aten.max_pool2d.default, max_pool2d);
 }
 
 } // namespace vkcompute

backends/vulkan/test/op_tests/cases.py

@@ -270,11 +270,13 @@ def get_avg_pool2d_inputs():
     return test_suite
 
 
-@register_test_suite("aten.max_pool2d_with_indices.default")
+@register_test_suite(
+    ["aten.max_pool2d_with_indices.default", "aten.max_pool2d.default"]
+)
 def get_max_pool2d_inputs():
     test_suite = VkTestSuite(
         [
-            ((S, M1, M2), [2, 2], [1, 1], [0, 0], [1, 1]),
+            ((1, 7, 89, 77), [2, 2], [1, 1], [0, 0], [1, 1]),
         ]
     )
     return test_suite