Messages in this thread | | | Date | Thu, 19 May 2022 09:10:20 -0700 | From | "Paul E. McKenney" <> | Subject | Re: [PATCH 03/21] rcu: Add a note about noinstr VS unsafe eqs functions |
| |
On Thu, May 19, 2022 at 05:35:27PM +0200, Frederic Weisbecker wrote: > On Thu, May 19, 2022 at 08:26:11AM -0700, Paul E. McKenney wrote: > > On Thu, May 19, 2022 at 04:54:38PM +0200, Peter Zijlstra wrote: > > > On Tue, May 03, 2022 at 12:00:33PM +0200, Frederic Weisbecker wrote: > > > > Some RCU functions enter or exit into/from RCU idle mode while using > > > > trace-able and lockdep-aware IRQs (un-)masking. This could be easily > > > > solved with using raw versions of local_irq_*() but we would then > > > > lose some precious debugging informations. > > > > > > > > Another possible way to solve this may consist in using rude RCU-tasks > > > > in lockdep and irqsoff tracing. > > > > > > > > In any case and until this get solved, those RCU functions can't get > > > > tagged as noinstr even though they should. > > > > > > FWIW, this also means RCU_NONIDLE() is broken. Luckily there aren't many > > > users left. > > > > More to the point, rcu_irq_enter_irqson() and rcu_irq_exit_irqson() need > > the same help that rcu_idle_enter() and rcu_idle_exit() got earlier this > > week [1], independent of RCU_NONIDLE(). > > > > Unless Frederic beats me to it, I will queue something up. > > It's in the patchset I just posted (02/21).
Even better!!! Thank you!
Thanx, Paul
| |