On Thu, Feb 05, 2015 at 02:02:35PM -0500, Rik van Riel wrote: > On 02/05/2015 01:56 PM, Paul E. McKenney wrote: > > > The real danger is doing neither. > > > > On tick_nohz_full_cpu() CPUs, the exit-to-userspace code should invoke > > rcu_user_enter(), which sets some per-CPU state telling RCU to ignore > > that CPU, since it cannot possibly do host RCU read-side critical sections > > while running a guest. > > > > In contrast, a non-tick_nohz_full_cpu() CPU doesn't let RCU > > know that it is executing in a guest or in userspace. So the > > rcu_virt_note_context_switch() does the notification in that case. > > Looking at context_tracking.h, I see the > function context_tracking_cpu_is_enabled(). > > That looks like it should do the right thing > in this case. Right you are -- that same check is used to guard the context_tracking_user_enter() function's call to rcu_user_enter(). Not sure why it open-codes the check rather than invoking context_tracking_cpu_is_enabled(). Hmmm.... One reason is that the context_tracking_cpu_is_enabled() function isn't available in that context, according to my compiler. ;-) Thanx, Paul -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
