Hi Christoph, On Mon, 2011-12-19 at 14:55 +0100, "Christoph Fünfzig" wrote: > Hi Gerrit, > > thanks for your hints. > Syncing aspects sequentially, does sound expensive above a certain number.
yep. > What do you think about adding a pre-allocation of aspect store vectors > before starting OpenSG? unfortunately that will only delay the problems/crashes as some of the underlying data the sync manipulates will be touched by all syncing threads and as they execute exactly the same sequence of syncs they will interfere with each other. I just changed it so now the sync uses cheap atomic spinlocks to lock each container individually while the synchronization is running so that they run overlapping and each thread is delayed as shortly as possible. So far all tests including your testprogram work. Only for your testprogram I get only a black result in the parallel windows but this might be something else. kind regards gerrit ------------------------------------------------------------------------------ Learn Windows Azure Live! Tuesday, Dec 13, 2011 Microsoft is holding a special Learn Windows Azure training event for developers. It will provide a great way to learn Windows Azure and what it provides. You can attend the event by watching it streamed LIVE online. Learn more at http://p.sf.net/sfu/ms-windowsazure _______________________________________________ Opensg-users mailing list Opensg-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/opensg-users
