> On Tue, 2005-11-08 at 10:39 -0700, Christopher Nelson wrote: > > > Yeah... I was thinking about that after I wrote it, and I > can see how > > that is the case. Does the implementation simply have a > user-slidable > > window? Something comparable to a seek, where I can now > access that > > memory from byte x to x+n where n is the window size? Or is it > > automated in some fashion? > > In EROS, it goes like this: > > I'm holding a capability to a memory region that you provided. > > I insert this capability into a known slot in my address > space structures. It is now mapped at a known location in my > address space. > > I now reference it in the usual way using load/store instructions.
How is that different from: -- Thread A -- share_id=share_memory(pointer_to_buffer, buffer_size); send(endpoint, share_id); -- Thread B -- share_id=recv(endpoint); if is_acceptable(share_size(share_id)) pointer=accept_memory(share_id); else reject_memory(share_id); _______________________________________________ L4-hurd mailing list [email protected] http://lists.gnu.org/mailman/listinfo/l4-hurd
