I don't quite understand the need, want, or mechanism for this. For high performance or VR systems where context switch overhead would be too much (and I don't believe this, since they tend to use separate acceleration chips for real-time work), the answer would not be to move from userspace to the kernel -- kernel threads are like any other and have context switch penalties.
If you have real performance issues with Wayland's architecture, let's discuss that and try to solve it. There's probably a lot there that's slow, but I doubt context switches are the bottleneck. On Wed, Dec 30, 2015, 11:56 AM <[email protected]> wrote: > Hi. > > On 29.12.15 20:09, Dimitri Nüscheler wrote: > > Hello everyone > > > > I sometimes wonder where people talk about concept level and philosophy. > > At least Wayland has a big philosophy part - and it uses it to explain > > itself in contrast to X. > > I'm not much involved into it, but I think I understand some vital parts. > > So far I understand it as a "graphics buffer and input redirection > > protocol" and that's a very concept level perspective. The routers in > > this "redirection network" are obviously the compositors. > > I can't comment on Wayland itself. Philosophy does not have a monopoly > on concepts. Abstract mathematics has concepts too. This is what I > prefer because mathematical definitions are precise. Philosophical > definitions are not, so reasoning with them does no good. > _______________________________________________ > wayland-devel mailing list > [email protected] > http://lists.freedesktop.org/mailman/listinfo/wayland-devel >
_______________________________________________ wayland-devel mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/wayland-devel
