viz: Convert a sync api in ServerGpuMemoryBufferManager into async.

components/viz/common/server_gpu_memory_buffer_manager.cc

Index: components/viz/common/server_gpu_memory_buffer_manager.cc
diff --git a/components/viz/common/server_gpu_memory_buffer_manager.cc b/components/viz/common/server_gpu_memory_buffer_manager.cc
index 23fe8942da670b8c84f1c0e9d4cbd04a6faad4fd..568f45ae8243677519cab2a42d18571a02104732 100644
--- a/components/viz/common/server_gpu_memory_buffer_manager.cc
+++ b/components/viz/common/server_gpu_memory_buffer_manager.cc
@@ -5,6 +5,7 @@
 #include "components/viz/common/server_gpu_memory_buffer_manager.h"
 
 #include "base/logging.h"
+#include "base/threading/thread_task_runner_handle.h"
 #include "gpu/ipc/client/gpu_memory_buffer_impl.h"
 #include "gpu/ipc/client/gpu_memory_buffer_impl_shared_memory.h"
 #include "gpu/ipc/common/gpu_memory_buffer_support.h"
@@ -18,36 +19,45 @@ ServerGpuMemoryBufferManager::ServerGpuMemoryBufferManager(
     : gpu_service_(gpu_service),
       client_id_(client_id),
       native_configurations_(gpu::GetNativeGpuMemoryBufferConfigurations()),
+      task_runner_(base::ThreadTaskRunnerHandle::Get()),
       weak_factory_(this) {}
 
 ServerGpuMemoryBufferManager::~ServerGpuMemoryBufferManager() {}
 
-gfx::GpuMemoryBufferHandle
-ServerGpuMemoryBufferManager::CreateGpuMemoryBufferHandle(
+void ServerGpuMemoryBufferManager::CreateGpuMemoryBufferHandle(
     gfx::GpuMemoryBufferId id,
     int client_id,
     const gfx::Size& size,
     gfx::BufferFormat format,
     gfx::BufferUsage usage,
-    gpu::SurfaceHandle surface_handle) {
-  DCHECK(CalledOnValidThread());
+    gpu::SurfaceHandle surface_handle,
+    base::OnceCallback<void(const gfx::GpuMemoryBufferHandle&)> callback) {
+  DCHECK(task_runner_->BelongsToCurrentThread());
   if (gpu::GetNativeGpuMemoryBufferType() != gfx::EMPTY_BUFFER) {
     const bool is_native = native_configurations_.find(std::make_pair(
                                format, usage)) != native_configurations_.end();
     if (is_native) {
-      gfx::GpuMemoryBufferHandle handle;
-      gpu_service_->CreateGpuMemoryBuffer(id, size, format, usage, client_id,
-                                          surface_handle, &handle);
-      if (!handle.is_null())
-        native_buffers_[client_id].insert(handle.id);
-      return handle;
+      pending_buffers_.insert(client_id);
+      gpu_service_->CreateGpuMemoryBuffer(
+          id, size, format, usage, client_id, surface_handle,
+          base::Bind(&ServerGpuMemoryBufferManager::OnGpuMemoryBufferCreated,
+                     weak_factory_.GetWeakPtr(), client_id,
+                     base::Passed(std::move(callback))));
+      return;
     }
   }
 
-  DCHECK(gpu::GpuMemoryBufferImplSharedMemory::IsUsageSupported(usage))
-      << static_cast<int>(usage);
-  return gpu::GpuMemoryBufferImplSharedMemory::CreateGpuMemoryBuffer(id, size,
-                                                                     format);
+  // Early out if we cannot fallback to shared memory buffer.
+  if (!gpu::GpuMemoryBufferImplSharedMemory::IsUsageSupported(usage) ||

[reveman, 2017/06/15 15:17:43]

Why can't we keep the DCHECK here now?

[sadrul, 2017/06/15 16:58:05]

I get crashes (with usage == SCANOUT) when running a chromeos build on linux
desktop with ozone-x11 backend. (which uses
https://cs.chromium.org/chromium/src/ui/ozone/common/stub_client_native_pixma...)

[reveman, 2017/06/15 17:46:58]

Can we instead avoid trying to create scanout buffers in this case? I think
that's preferred when we control the code and it's not exposed to the renderer
and the web platform.

+      !gpu::GpuMemoryBufferImplSharedMemory::IsSizeValidForFormat(size,
+                                                                  format)) {
+    std::move(callback).Run(gfx::GpuMemoryBufferHandle());
+    return;
+  }
+
+  std::move(callback).Run(
+      gpu::GpuMemoryBufferImplSharedMemory::CreateGpuMemoryBuffer(id, size,
+                                                                  format));
 }
 
 std::unique_ptr<gfx::GpuMemoryBuffer>
@@ -57,8 +67,25 @@ ServerGpuMemoryBufferManager::CreateGpuMemoryBuffer(
     gfx::BufferUsage usage,
     gpu::SurfaceHandle surface_handle) {
   gfx::GpuMemoryBufferId id(next_gpu_memory_id_++);
-  gfx::GpuMemoryBufferHandle handle = CreateGpuMemoryBufferHandle(
-      id, client_id_, size, format, usage, surface_handle);
+  gfx::GpuMemoryBufferHandle handle;
+  base::WaitableEvent wait_event(
+      base::WaitableEvent::ResetPolicy::MANUAL,
+      base::WaitableEvent::InitialState::NOT_SIGNALED);
+  // We block with a WaitableEvent until the callbacks are run. So using
+  // base::Unretained() is safe here.
+  auto reply_callback = base::BindOnce(
+      &ServerGpuMemoryBufferManager::OnGpuMemoryBufferCreatedWait,
+      base::Unretained(this), &handle, &wait_event);
+  auto alloc_callback =

[reveman, 2017/06/15 15:17:43]

nit: s/alloc/allocate/

[sadrul, 2017/06/15 16:58:05]

Done.

+      base::BindOnce(&ServerGpuMemoryBufferManager::CreateGpuMemoryBufferHandle,
+                     base::Unretained(this), id, client_id_, size, format,
+                     usage, surface_handle, std::move(reply_callback));
+  if (task_runner_->BelongsToCurrentThread())

[reveman, 2017/06/15 15:17:43]

Would it be better to always have this done by PostTask to avoid having a
different call stack depending on the thread it's called from?

[sadrul, 2017/06/15 16:58:05]

I have made the change to always PostTask(). The reason I ran immediately in
this case was for potential latency improvements. But I am not sure if that can
actually be an issue?

+    std::move(alloc_callback).Run();
+  else
+    task_runner_->PostTask(FROM_HERE, std::move(alloc_callback));
+  base::ThreadRestrictions::ScopedAllowWait allow_wait;
+  wait_event.Wait();
   if (handle.is_null())
     return nullptr;
   return gpu::GpuMemoryBufferImpl::CreateFromHandle(
@@ -70,7 +97,6 @@ ServerGpuMemoryBufferManager::CreateGpuMemoryBuffer(
 void ServerGpuMemoryBufferManager::SetDestructionSyncToken(
     gfx::GpuMemoryBuffer* buffer,
     const gpu::SyncToken& sync_token) {
-  DCHECK(CalledOnValidThread());
   static_cast<gpu::GpuMemoryBufferImpl*>(buffer)->set_destruction_sync_token(
       sync_token);
 }
@@ -79,17 +105,44 @@ void ServerGpuMemoryBufferManager::DestroyGpuMemoryBuffer(
     gfx::GpuMemoryBufferId id,
     int client_id,
     const gpu::SyncToken& sync_token) {
-  DCHECK(CalledOnValidThread());
+  DCHECK(task_runner_->BelongsToCurrentThread());
   if (native_buffers_[client_id].erase(id))
     gpu_service_->DestroyGpuMemoryBuffer(id, client_id, sync_token);
 }
 
 void ServerGpuMemoryBufferManager::DestroyAllGpuMemoryBufferForClient(
     int client_id) {
-  DCHECK(CalledOnValidThread());
+  DCHECK(task_runner_->BelongsToCurrentThread());
   for (gfx::GpuMemoryBufferId id : native_buffers_[client_id])
     gpu_service_->DestroyGpuMemoryBuffer(id, client_id, gpu::SyncToken());
   native_buffers_.erase(client_id);
+  pending_buffers_.erase(client_id);
+}
+
+void ServerGpuMemoryBufferManager::OnGpuMemoryBufferCreated(
+    int client_id,
+    base::OnceCallback<void(const gfx::GpuMemoryBufferHandle&)> callback,
+    const gfx::GpuMemoryBufferHandle& handle) {
+  DCHECK(task_runner_->BelongsToCurrentThread());
+  if (!pending_buffers_.count(client_id)) {

[reveman, 2017/06/15 15:17:43]

nit: use find instead

[sadrul, 2017/06/15 16:58:05]

Done.

+    // The client has been destroyed since the memory request was made.

[reveman, 2017/06/15 15:17:43]

s/memory request/allocation request/ ?

[sadrul, 2017/06/15 16:58:05]

Done.

+    if (!handle.is_null())
+      gpu_service_->DestroyGpuMemoryBuffer(handle.id, client_id,
+                                           gpu::SyncToken());
+    std::move(callback).Run(gfx::GpuMemoryBufferHandle());
+    return;
+  }
+  if (!handle.is_null())
+    native_buffers_[client_id].insert(handle.id);
+  std::move(callback).Run(handle);
+}
+
+void ServerGpuMemoryBufferManager::OnGpuMemoryBufferCreatedWait(
+    gfx::GpuMemoryBufferHandle* handle,
+    base::WaitableEvent* wait_event,
+    const gfx::GpuMemoryBufferHandle& allocated_handle) {
+  *handle = allocated_handle;
+  wait_event->Signal();
 }
 
 }  // namespace viz