On Tue, Jun 23, 2015 at 01:32:19PM +0200, Paolo Bonzini wrote: > > > On 23/06/2015 13:18, Daniel P. Berrange wrote: > > > For 2.5, however, I wonder if SuspendThread/ResumeThread is needed at > > > all now that cpu_exit doesn't have to undo block chaining anymore. Even > > > on POSIX platforms the signal might not be necessary anymore. > > > > If you don't have that signal / SuspendThread/ResumtThread requirement, > > That was independent of QEMU reinventing the wheel for mutexes/condvars. > > > might that enable QEMU to just depend on the winpthreads library that > > is provided by Mingw project, and not bother reinventing the wheel for > > thread library portabilty ? > > We can and should just reuse glib these days as much as we can (probably > not entirely because glib doesn't have detached threads). At least a > few years ago, winpthreads was much slower than native Win32, which is > why everyone reinvents the wheel.
Are you sure that was wrt the (new) winpthreads library maintained by Mingw64 team, and not the confusingly similar pthreads-win32 library ? Regards, Daniel -- |: http://berrange.com -o- http://www.flickr.com/photos/dberrange/ :| |: http://libvirt.org -o- http://virt-manager.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: http://entangle-photo.org -o- http://live.gnome.org/gtk-vnc :|
