Ian Romanick <[EMAIL PROTECTED]> writes: > Please forgive my lack of understanding of the relevent issues. It seems to > me that this issue (or a closely related issue) has been brought up in the > past, and was brutally slain by Linus. Admittedly, a lot has changed in the > Linux world in 2+ years, but could someone explain how this proposal relates > to this earlier request: > > http://www.mail-archive.com/linux-kernel@vger.rutgers.edu/1999-month-12/msg02616.html
This is talking about tread private storage, where you have multiple threads that "share" memory apart from a special bit that can only be accessed from each thread (so then it isn't really threads it's processes with massive amounts of shared memory). The current discussion is about thread local memory, Ie. part of the overall shared memory that is (excluding bugs) only accessed from the current thread (but is in the memory space of all threads). -- # James Antill -- [EMAIL PROTECTED] :0: * ^From: .*james@and\.org /dev/null _______________________________________________________________ Hundreds of nodes, one monster rendering program. Now that’s a super model! Visit http://clustering.foundries.sf.net/ _______________________________________________ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel