Implement profiler log writing at shutdown

chrome/browser/task_profiler/task_profiler_data_serializer.cc

Index: chrome/browser/task_profiler/task_profiler_data_serializer.cc
diff --git a/chrome/browser/task_profiler/task_profiler_data_serializer.cc b/chrome/browser/task_profiler/task_profiler_data_serializer.cc
new file mode 100644
index 0000000000000000000000000000000000000000..243a60b59b4013c45f1b064f4b02f6967608b025
--- /dev/null
+++ b/chrome/browser/task_profiler/task_profiler_data_serializer.cc
@@ -0,0 +1,50 @@
+// Copyright (c) 2011 The Chromium Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style license that can be
+// found in the LICENSE file.
+
+#include "chrome/browser/task_profiler/task_profiler_data_serializer.h"
+
+#include "base/file_path.h"
+#include "base/file_util.h"
+#include "base/json/json_value_serializer.h"
+#include "base/time.h"
+#include "base/tracked_objects.h"
+#include "content/public/common/content_client.h"
+#include "googleurl/src/gurl.h"
+
+namespace task_profiler {
+
+bool TaskProfilerDataSerializer::WriteToFile(const FilePath &path) {
+  std::string output;
+  JSONStringValueSerializer serializer(&output);
+  serializer.set_pretty_print(true);
+
+  scoped_ptr<base::DictionaryValue> root(new DictionaryValue());
+
+  base::ListValue* snapshot_list = new ListValue();
+  base::DictionaryValue* shutdown_snapshot = new DictionaryValue();
+  base::ListValue* per_process_data = new ListValue();
+
+  root->SetInteger("version", 1);
+  root->SetString("userAgent", content::GetUserAgent(GURL()));
+
+  // TODO(rlarocque): We should collect data from other processes as they shut
+  // down, then add that data to the 'per_process_data' array here.

[jar (doing other things), 2012/01/10 20:51:35]

This is a big challenge :-/.  Most processes shut down rather dramatically, with
very little warning.  Perhaps you can cause the shutdown hooks in the browser
process to "go slow" and first request profile info.

Historically, this has been a hassle, when we wanted to get histograms relating
to page load.  We gathered the data... but then (in our rush for speed) killed
teh process at shutdown before we had a chance to upload "the last page loaded"
data.

You may be able to ignore this "rush" constraint since the user has asked for
the dump... and may have to "pay" with a slightly slower shutdown.

[ramant (doing other things), 2012/01/31 01:12:44]

Hi rlarocque,
  I had done collecting of data from child processes. Can work on getting data
from child processes when we are shutting down.

a small question: any updates on the status of this patch?

thanks,
raman

+  base::DictionaryValue* this_process_data =
+      tracked_objects::ThreadData::ToValue(false);
+  per_process_data->Append(this_process_data);
+
+  shutdown_snapshot->SetInteger(
+      "timestamp",
+      (base::Time::Now() - base::Time::UnixEpoch()).InSeconds());
+  shutdown_snapshot->Set("data", per_process_data);
+  snapshot_list->Append(shutdown_snapshot);
+  root->Set("snapshots", snapshot_list);
+
+  serializer.Serialize(*root);
+  int data_size = static_cast<int>(output.size());
+
+  return data_size == file_util::WriteFile(path, output.data(), data_size);
+}
+
+}  // namespace task_profiler