electron/patches/chromium/add_contentgpuclient_precreatemessageloop_callback.patch
Electron Bot 53033a66b1
chore: bump chromium to 7d4ac110045b7647fa8a81f5f0709 (master) (#24970)
* chore: bump chromium in DEPS to 760c905ccca34d8ed83c54f6a42c42a375792a6b

* chore: update patches

- Introduce WebContentsObserverList to encapsulate common behaviour
  https://chromium-review.googlesource.com/c/chromium/src/+/2343269
- DumpAccTree mac testing: refactor tree formatter logic to make it
suitable for nested attribute calls
  https://chromium-review.googlesource.com/c/chromium/src/+/2274545
- Fix FontPlatformDataMacTest.VariableOpticalSizingThreshold for Big Sur
  https://chromium-review.googlesource.com/c/chromium/src/+/2351299

* chore: bump chromium in DEPS to b04584161e07d4ac110045b7647fa8a81f5f0709

* chore: update patches

* [printing] Move printing::MarginType to //printing/mojom/print.mojom

https://chromium-review.googlesource.com/c/chromium/src/+/2341222

Co-authored-by: deepak1556 <hop2deep@gmail.com>
2020-08-14 18:51:28 -07:00

49 lines
2 KiB
Diff

From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
From: Milan Burda <milan.burda@gmail.com>
Date: Thu, 11 Apr 2019 14:49:20 +0200
Subject: add_contentgpuclient_precreatemessageloop_callback.patch
This adds a ContentGpuClient::PreCreateMessageLoop() callback.
Invoke in GpuMain after SetErrorMode, before starting the message loop.
Allows Electron to restore WER when ELECTRON_DEFAULT_ERROR_MODE is set.
This should be upstreamed.
diff --git a/content/gpu/gpu_main.cc b/content/gpu/gpu_main.cc
index ed134d1d40fb23808a4cc251be50a7fbd833b49b..5aaf2356cc2c0fec9a90f3201f1e16553c69a5e3 100644
--- a/content/gpu/gpu_main.cc
+++ b/content/gpu/gpu_main.cc
@@ -255,6 +255,10 @@ int GpuMain(const MainFunctionParams& parameters) {
// to the GpuProcessHost once the GpuServiceImpl has started.
viz::GpuServiceImpl::InstallPreInitializeLogHandler();
+ auto* client = GetContentClient()->gpu();
+ if (client)
+ client->PreCreateMessageLoop();
+
// We are experiencing what appear to be memory-stomp issues in the GPU
// process. These issues seem to be impacting the task executor and listeners
// registered to it. Create the task executor on the heap to guard against
@@ -389,7 +393,6 @@ int GpuMain(const MainFunctionParams& parameters) {
}
#endif
- auto* client = GetContentClient()->gpu();
if (client)
client->PostIOThreadCreated(gpu_process.io_task_runner());
diff --git a/content/public/gpu/content_gpu_client.h b/content/public/gpu/content_gpu_client.h
index 88e80bdaecbdd9000109f638c6d79d8102d0e537..a0c0f17e6642dcd2690c095b95ff2c2de9b6ae4f 100644
--- a/content/public/gpu/content_gpu_client.h
+++ b/content/public/gpu/content_gpu_client.h
@@ -30,6 +30,10 @@ class CONTENT_EXPORT ContentGpuClient {
public:
virtual ~ContentGpuClient() {}
+ // Allows the embedder to perform platform-specific initialization before
+ // creating the message loop.
+ virtual void PreCreateMessageLoop() {}
+
// Called during initialization once the GpuService has been initialized.
virtual void GpuServiceInitialized() {}