Re: linux vs. win32: gdk_threads_enter behaving different

["Jonathan Winterflood" <jonathan winterflood gmail com>]

I too am rather intrigued by this problem...
I found this:
http://research.operationaldynamics.com/blogs/andrew/software/gnome-desktop/gtk-thread-awareness.html
and I'm wondering: is your gtk_main() surrounded by gdk_threads_enter/leave.
I imagine this missing would break in linux too, but just thinking...

I've yet to find the time to try multithreading gtk myself

Jonathan

On 6/15/07, Felix Kater <fkater googlemail com> wrote:
> Kevin DeKorte <kdekorte gmail com> wrote:
>
> > - From my limited GTK experience I have found that g_idle_add always
> > seems to work better than using gtk_threads_enter/leave.
>
> Yes, the reason why idle functions work in any case is most probably
> because they are running in the same thread as where gtk was init'ed.
> So, using idle functions probably means that you do not need
> gdk_thread_enter at all. So, what it is for then?
>
> The answer to this seems to be that gds_threads_enter is useful for
> linux only (where you *do* can sync access to gtk from different threads
> using gdk_threads_enter) on win32 you can't. However, this means that
> code relying on gds_threads_enter is not necessarily portable.
>
> Also, I'd like to understand why gdk_threads_enter() does not work on
> gtk/win32. So, maybe there is a solution with an own recursive lock
> instead of wrapping all into idle functions.
>
> Felix
> _______________________________________________
> gtk-app-devel-list mailing list
> gtk-app-devel-list gnome org
> http://mail.gnome.org/mailman/listinfo/gtk-app-devel-list



--
<Morpheus> linux, c'est une question de VI ou de MORE