From 9cf98dacb37fbb63c4427fd31f3a1ac05d47785b Mon Sep 17 00:00:00 2001
From: Justin Stoecker <justoeck@microsoft.com>
Date: Mon, 3 Oct 2022 16:03:29 -0700
Subject: [PATCH] Enable command list reuse for Xbox (#13173)

Removes the workaround introduced in #12063, which disabled DML command
list reuse for Xbox builds.

The ID3D12CommandList created in FusedGraphKernel takes points to an
ID3D12CommandAllocator that is local to the `BuildReusableCommandList`
function. On PC it would seem the command list is keeping the command
allocator alive, but this is highly suspect logic that definitely
doesn't work on Xbox. I find no documentation indicating this logic
should work (a section on [reference
counting](https://learn.microsoft.com/en-us/windows/win32/direct3d12/recording-command-lists-and-bundles#reference-counting)
makes it clear command lists take no refs on D3D objects passed as args
to its APIs; however, it's unclear if this also applies to its
construction).

A second (small) change is constructing the command list straight into
`ID3D12GraphicsCommandList` and removing an unnecessary QI.
---
 .../DmlExecutionProvider/src/FusedGraphKernel.cpp   | 13 +++++--------
 .../DmlExecutionProvider/src/GraphDescBuilder.cpp   |  5 -----
 2 files changed, 5 insertions(+), 13 deletions(-)

diff --git a/onnxruntime/core/providers/dml/DmlExecutionProvider/src/FusedGraphKernel.cpp b/onnxruntime/core/providers/dml/DmlExecutionProvider/src/FusedGraphKernel.cpp
index 3c6d1bcf4a..f1ef0ed1df 100644
--- a/onnxruntime/core/providers/dml/DmlExecutionProvider/src/FusedGraphKernel.cpp
+++ b/onnxruntime/core/providers/dml/DmlExecutionProvider/src/FusedGraphKernel.cpp
@@ -331,21 +331,17 @@ namespace Dml
 
             ORT_THROW_IF_FAILED(device->CreateBindingTable(&bindingTableDesc, IID_PPV_ARGS(&m_bindingTable)));
 
-            ComPtr<ID3D12CommandAllocator> allocator;
             ORT_THROW_IF_FAILED(d3dDevice->CreateCommandAllocator(
                 m_provider->GetCommandListTypeForQueue(),
-                IID_GRAPHICS_PPV_ARGS(allocator.ReleaseAndGetAddressOf())));
+                IID_GRAPHICS_PPV_ARGS(m_commandAllocator.ReleaseAndGetAddressOf())));
 
-            ComPtr<ID3D12CommandList> commandList;
             ORT_THROW_IF_FAILED(d3dDevice->CreateCommandList(
                 0,
                 m_provider->GetCommandListTypeForQueue(),
-                allocator.Get(),
+                m_commandAllocator.Get(),
                 nullptr,
-                IID_GRAPHICS_PPV_ARGS(commandList.ReleaseAndGetAddressOf())));
+                IID_GRAPHICS_PPV_ARGS(m_graphicsCommandList.ReleaseAndGetAddressOf())));
             
-            ORT_THROW_IF_FAILED(commandList.As(&m_graphicsCommandList));
-
             if (m_persistentResource)
             {
                 DML_BINDING_DESC persistentResourceBindingDesc =
@@ -359,7 +355,7 @@ namespace Dml
             ComPtr<IDMLCommandRecorder> recorder;
             ORT_THROW_IF_FAILED(device->CreateCommandRecorder(IID_PPV_ARGS(recorder.GetAddressOf())));
 
-            recorder->RecordDispatch(commandList.Get(), m_compiledExecutionPlanOperator.Get(), m_bindingTable.Get());
+            recorder->RecordDispatch(m_graphicsCommandList.Get(), m_compiledExecutionPlanOperator.Get(), m_bindingTable.Get());
 
             ORT_THROW_IF_FAILED(m_graphicsCommandList->Close());
         }
@@ -496,6 +492,7 @@ namespace Dml
 
         // Re-usable command list, supporting descriptor heap, and DML binding table to update that heap.
         ComPtr<ID3D12GraphicsCommandList> m_graphicsCommandList;
+        ComPtr<ID3D12CommandAllocator> m_commandAllocator;
         ComPtr<ID3D12DescriptorHeap> m_heap;
         ComPtr<IDMLBindingTable> m_bindingTable;
         std::optional<DML_BUFFER_BINDING> m_persistentResourceBinding;
diff --git a/onnxruntime/core/providers/dml/DmlExecutionProvider/src/GraphDescBuilder.cpp b/onnxruntime/core/providers/dml/DmlExecutionProvider/src/GraphDescBuilder.cpp
index fcc63fae94..cadcf0c8d0 100644
--- a/onnxruntime/core/providers/dml/DmlExecutionProvider/src/GraphDescBuilder.cpp
+++ b/onnxruntime/core/providers/dml/DmlExecutionProvider/src/GraphDescBuilder.cpp
@@ -95,11 +95,6 @@ namespace Dml::GraphDescBuilder
             reuseCommandList = true;
         }
 
-#ifdef _GAMING_XBOX
-        // #40265989
-        reuseCommandList = false;
-#endif
-
         auto constantCpuGraphInputGetter = [&transferredInitializerMap](const std::string& argName)
         {
             ComPtr<OnnxTensorWrapper> tensorWrapper;