On Tue, Jan 05, 2021 at 02:42:32PM +0100, Peter Zijlstra wrote: > On Wed, Dec 23, 2020 at 07:12:31AM -0800, Paul E. McKenney wrote: > > On Wed, Dec 23, 2020 at 02:09:37PM +0530, Neeraj Upadhyay wrote: > > > For the smp_call_function() optimization, where callbacks can run from > > > idle context, in commit 806f04e9fd2c ("rcu: Allow for smp_call_function() > > > running callbacks from idle"), an additional check is added in > > > rcu_is_cpu_rrupt_from_idle(), for dynticks_nmi_nesting value being 0, > > > for these smp_call_function() callbacks running from idle loop. > > > However, this commit missed updating a preexisting underflow check > > > of dynticks_nmi_nesting, which checks for a non zero positive value. > > > Fix this warning and while at it, read the counter only once. > > > > > > Signed-off-by: Neeraj Upadhyay <neeraju@xxxxxxxxxxxxxx> > > > --- > > > > > > Hi, > > > > > > I was not able to get this warning, with scftorture. > > > > > > RCU_LOCKDEP_WARN(__this_cpu_read(rcu_data.dynticks_nmi_nesting) <= 0, > > > "RCU dynticks_nmi_nesting counter underflow/zero!"); > > > > > > Not sure if idle loop smp_call_function() optimization is already present > > > in mainline? > > > > Now that you mention it, I don't see it. > > kernel/sched/idle.c:do_idle() calls flush_smp_call_function_from_idle(). > > (nothing x86 specific about it) Got it, thank you! The reason Neeraj was unable to trigger the problematic warning from scftorture is that its smp_call_function() handlers do not invoke rcu_is_cpu_rrupt_from_idle(). Without adding this to those handlers (which would be a good change to make), the only way to trigger this is for an expedited RCU grace period to IPI a CPU that goes idle while the IPI is in flight, which is not the easiest thing to make happen. > > > Another thing, which I am not sure of is, maybe lockdep gets disabled > > > in the idle loop contexts, where rcu_is_cpu_rrupt_from_idle() is called? > > > Was this the original intention, to keep the lockdep based > > > RCU_LOCKDEP_WARN(__this_cpu_read(rcu_data.dynticks_nmi_nesting) <= 0 > > > check separate from idle task context nesting value > > > WARN_ON_ONCE(!nesting && !is_idle_task(current)) check? > > > > An easy way to test lockdep is to create a pair of locks, acquire them > > in one order then release them both, and finally acquire them in the > > other order and then release them both. If lockdep is configured and > > enabled, it will complain. > > IIRC (and this is after not staring at the computer for 2 weeks) lockdep > should work just fine in idle, except of course that RCU will be stopped > so actually taking locks will scream bloody murder due to tracing etc.. Fair enough... And I hope that the time off went well for you and yours! > > The only reason I used RCU_LOCKDEP_WARN() was that people were complaining > > to me about idle-entry overhead back at that time. So without lockdep, > > there is zero overhead. Maybe people have become more tolerant of idle > > delays, or perhaps they are not so worried about an extra check of a > > cache-hot quantity. > > Not having checks also saves on $I and branches, in general I think > having checks depend on DEBUG features, esp. those we don't really > expect to trigger is still sane. OK, so should we convert the WARN_ON_ONCE() to RCU_LOCKDEP_WARN() while we are in the area? > > I am tempted to pull this in as is, given the current logical > > inconsistency in the checks. Thoughts? > > Patch looks ok, although I've seen compilers do CSE on > __this_cpu_read() (on x86). True, but the compilers might might have a harder time of this on other architectures. Thanx, Paul