That's what I expected to be the synchronization technique at first but I don't see the lock being taken anywhere on the watch thread's call stack (xenwatch_thread, otherend_changed, config_detach, xennet_xenbus_detach).
On Sun, Feb 7, 2016 at 1:51 AM, Manuel Bouyer <bou...@antioche.eu.org> wrote: > On Sat, Feb 06, 2016 at 10:04:47PM -0800, Rohan Desai wrote: > > I'll give it a shot. Is this sufficient for SMP systems? I thought spl* > > funcitons only raised the ipl for the local cpu. Isn't it possible for > > detach and event handling to be executed by different cpus? > > the kernel_lock should prevent it > > -- > Manuel Bouyer <bou...@antioche.eu.org> > NetBSD: 26 ans d'experience feront toujours la difference > -- >
