Transform ImagePreReader into PreReadFile.

chrome/app/file_pre_reader_win.h

+// Copyright (c) 2012 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.
+
+// This file defines a function to pre-read a file in order to avoid hard page
+// faults when it is subsequently used.
+
+#ifndef CHROME_APP_FILE_PRE_READER_WIN_H_
+#define CHROME_APP_FILE_PRE_READER_WIN_H_
+
+#include "base/basictypes.h"

[grt (UTC plus 2), 2015/10/29 19:15:07]

unused

[fdoray, 2015/10/30 21:08:54]

Done.

+
+class FilePreReader {

[grt (UTC plus 2), 2015/10/29 19:15:07]

make this a free function rather than a class static
(https://google.github.io/styleguide/cppguide.html#Nonmember,_Static_Member,_a...).

[fdoray, 2015/10/30 21:08:54]

Done.

+ public:
+  // Reads the file passed in and touches pages to avoid hard page faults when
+  // the file is used. The step size indicates the number of bytes to skip after
+  // each touched byte.

[grt (UTC plus 2), 2015/10/29 19:15:07]

how should the caller pick a value for |step_size|? can you explain this a bit
in the comment?

[fdoray, 2015/10/30 21:08:54]

Done. Eventually, we should remove the parameter and always use 4 MB. But first,
I want to make a Canary experiment to make sure that this is the best value.

+  //
+  // This function checks the Windows version to determine which pre-reading
+  // mechanism to use.
+  static bool PreReadImage(const wchar_t* file_path, size_t step_size);

[grt (UTC plus 2), 2015/10/29 19:15:07]

why "Image" in the name when this is file_pre_reader? could this be PreReadFile?

[fdoray, 2015/10/30 21:08:54]

Done.

+};
+
+#endif  // CHROME_APP_FILE_PRE_READER_WIN_H_