On Wed, Nov 30, 2016 at 01:13:03PM -0600, Josh Poimboeuf wrote: > On Tue, Nov 29, 2016 at 12:32:59PM -0800, Paul E. McKenney wrote: > > On Tue, Nov 29, 2016 at 12:07:11PM -0800, Paul E. McKenney wrote: > > > On Tue, Nov 29, 2016 at 08:52:04PM +0100, Peter Zijlstra wrote: > > > > On Tue, Nov 29, 2016 at 11:39:35AM -0800, Paul E. McKenney wrote: > > > > > On Tue, Nov 29, 2016 at 06:10:38PM +0100, Peter Zijlstra wrote: > > > > > > > > > > It mostly works, most of the time, and that seems to be what Linus > > > > > > wants, since its really the best we can have given the constraints. > > > > > > But > > > > > > for debugging, when you have a UART, it totally blows. > > > > > > > > > > UART??? They still make those things??? ;-) > > > > > > > > Yes, most computer like devices actually have them, trouble is, most > > > > consumer devices don't have the pins exposed. Luckily most server class > > > > hardware still does. > > > > > > > > And they're absolutely _awesome_ for debugging; getting data out is a > > > > matter of trivial MMIO poll loops. Rock solid stuff. > > > > > > They very clearly need to bring the baud rate into the current millenium, > > > many tens of Mbaud at the -very- least. > > > > On a more practical note... > > > > Currently, cond_resched_rcu_qs() is not permitted to be invoked until > > after the scheduler has started. However, it appears that there is some > > kernel code that can loop for quite some time at runtime, but which also > > executes during early boot. So it would be good to make it so that > > cond_resched_rcu_qs() could be called at boot. > > > > One approach would be to check rcu_scheduler_active, but this isn't > > defined in normal Tiny RCU builds. I can expand Tiny RCU, or I can > > kludge the non-CONFIG_DEBUG_LOCK_ALLOC value of rcu_scheduler_active > > to false (with this latter being the current state). But it occurred > > to me that I could also condition on !is_idle_task(), given that idle > > tasks shouldn't ever be invoking the scheduler anyway. > > This question was probably intended for other folks, but I should point > out that idle tasks *do* invoke the scheduler. cpu_idle_loop() calls > schedule_preempt_disabled().
Good point. My next fallback is that idle loops should not be running for long periods of time within RCU_NONIDLE(). Does that work? Thanx, Paul > > So is the following a sensible approach, or should I look elsewhere? > > > > #define cond_resched_rcu_qs() \ > > do { \ > > if (!is_idle_task(current) && !cond_resched()) \ > > rcu_note_voluntary_context_switch(current); \ > > } while (0) > > > > Thanx, Paul > > > > -- > Josh >