[Webkit-unassigned] [Bug 133028] Build break of 2.4.1 under mingw32/msys

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=133028





--- Comment #21 from Alberto Garcia <berto at igalia.com>  2014-05-29 05:09:07 PST ---
(In reply to comment #20)
> > I actually think you should replace OS(WINDOWS) in
> > BinarySemaphore.h with PLATFORM(WINDOWS), does that work?
> I can change the header to use PLATFORM(WIN), yes, but then the
> semaphore will be a void, while the code seems to be meant to be
> used on windows and it compiles fine (after fixing the .cpp's file
> check).

Are you sure it would be void?

There are two implementation of BinarySemaphore:

 - One uses CreateEvent / WaitForSingleObject from the Windows API and
   goes in BinarySemaphoreWin.cpp.

 - One uses WebKit's own Mutex class which has implementations in
   ThreadingPthreads.cpp and ThreadingWin.cpp. This goes in
   BinarySemaphore.cpp

The OS(WINDOWS) condition in the header is used to choose between
those two, but you also have to use BinarySemaphore.cpp or
BinarySemaphoreWin.cpp depending on the case.

BinarySemaphoreWin.cpp is not even shipped in the webkitgtk tarballs,
but I think BinarySemaphore.cpp should work fine.

> More importantly, how do I test the change?

This is used by the WebProcess for IPC, just run the MiniBrowser and
browser a couple of websites and you will see the WebProcess calling
WTF::BinarySemaphore::wait() all the time.

But I wonder what's the implementation that you are using if you use
#if !OS(WINDOWS) in BinarySemaphore.cpp, because I don't see any
other.

-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.