Re: [PATCH x86/entry: Force rcu_irq_enter() when in idle task

2020-06-16 Thread Andy Lutomirski
On Tue, Jun 16, 2020 at 7:30 AM Joel Fernandes wrote: > > On Tue, Jun 16, 2020 at 10:40:04AM +0200, Thomas Gleixner wrote: > > Joel, > > > > Joel Fernandes writes: > > > On Fri, Jun 12, 2020 at 03:55:00PM +0200, Thomas Gleixner wrote: > > >> This is also correct vs. user mode entries in NOHZ

Re: [PATCH x86/entry: Force rcu_irq_enter() when in idle task

2020-06-16 Thread Joel Fernandes
On Tue, Jun 16, 2020 at 10:40:04AM +0200, Thomas Gleixner wrote: > Joel, > > Joel Fernandes writes: > > On Fri, Jun 12, 2020 at 03:55:00PM +0200, Thomas Gleixner wrote: > >> This is also correct vs. user mode entries in NOHZ full scenarios because > >> user mode entries bring RCU out of EQS and

Re: [PATCH x86/entry: Force rcu_irq_enter() when in idle task

2020-06-16 Thread Thomas Gleixner
Joel, Joel Fernandes writes: > On Fri, Jun 12, 2020 at 03:55:00PM +0200, Thomas Gleixner wrote: >> This is also correct vs. user mode entries in NOHZ full scenarios because >> user mode entries bring RCU out of EQS and force the RCU irq nesting state > > I had to re-read this sentence a couple

Re: [PATCH x86/entry: Force rcu_irq_enter() when in idle task

2020-06-15 Thread Joel Fernandes
Hi Thomas, Paul, On Fri, Jun 12, 2020 at 03:55:00PM +0200, Thomas Gleixner wrote: > The idea of conditionally calling into rcu_irq_enter() only when RCU is > not watching turned out to be not completely thought through. > > Paul noticed occasional premature end of grace periods in RCU torture >

Re: [PATCH x86/entry: Force rcu_irq_enter() when in idle task

2020-06-12 Thread Paul E. McKenney
On Fri, Jun 12, 2020 at 09:34:56PM +0200, Thomas Gleixner wrote: > Thomas Gleixner writes: > > > "Paul E. McKenney" writes: > >> On Fri, Jun 12, 2020 at 10:49:53AM -0700, Paul E. McKenney wrote: > >>> I will therefore address this issue in a follow-on patch. > >> > >> I should add that -your-

Re: [PATCH x86/entry: Force rcu_irq_enter() when in idle task

2020-06-12 Thread Thomas Gleixner
Thomas Gleixner writes: > "Paul E. McKenney" writes: >> On Fri, Jun 12, 2020 at 10:49:53AM -0700, Paul E. McKenney wrote: >>> I will therefore address this issue in a follow-on patch. >> >> I should add that -your- patch from yesterday did -not- cause this >> problem, in case that is of

Re: [PATCH x86/entry: Force rcu_irq_enter() when in idle task

2020-06-12 Thread Andy Lutomirski
> On Jun 12, 2020, at 12:25 PM, Thomas Gleixner wrote: > > "Paul E. McKenney" writes: >>> On Fri, Jun 12, 2020 at 10:49:53AM -0700, Paul E. McKenney wrote: >>> I will therefore address this issue in a follow-on patch. >> >> I should add that -your- patch from yesterday did -not- cause this

Re: [PATCH x86/entry: Force rcu_irq_enter() when in idle task

2020-06-12 Thread Thomas Gleixner
"Paul E. McKenney" writes: > On Fri, Jun 12, 2020 at 10:49:53AM -0700, Paul E. McKenney wrote: >> I will therefore address this issue in a follow-on patch. > > I should add that -your- patch from yesterday did -not- cause this > problem, in case that is of interest. So I still can add it back

Re: [PATCH x86/entry: Force rcu_irq_enter() when in idle task

2020-06-12 Thread Paul E. McKenney
On Fri, Jun 12, 2020 at 10:49:53AM -0700, Paul E. McKenney wrote: > On Fri, Jun 12, 2020 at 03:55:00PM +0200, Thomas Gleixner wrote: > > The idea of conditionally calling into rcu_irq_enter() only when RCU is > > not watching turned out to be not completely thought through. > > > > Paul noticed

Re: [PATCH x86/entry: Force rcu_irq_enter() when in idle task

2020-06-12 Thread Paul E. McKenney
On Fri, Jun 12, 2020 at 03:55:00PM +0200, Thomas Gleixner wrote: > The idea of conditionally calling into rcu_irq_enter() only when RCU is > not watching turned out to be not completely thought through. > > Paul noticed occasional premature end of grace periods in RCU torture > testing. Bisection

Re: [PATCH x86/entry: Force rcu_irq_enter() when in idle task

2020-06-12 Thread Andy Lutomirski
On Fri, Jun 12, 2020 at 6:55 AM Thomas Gleixner wrote: > > The idea of conditionally calling into rcu_irq_enter() only when RCU is > not watching turned out to be not completely thought through. > > Paul noticed occasional premature end of grace periods in RCU torture > testing. Bisection led to

Re: [PATCH x86/entry: Force rcu_irq_enter() when in idle task

2020-06-12 Thread Thomas Gleixner
Frederic Weisbecker writes: > On Fri, Jun 12, 2020 at 03:55:00PM +0200, Thomas Gleixner wrote: >> The idea of conditionally calling into rcu_irq_enter() only when RCU is >> not watching turned out to be not completely thought through. >> >> Paul noticed occasional premature end of grace periods

Re: [PATCH x86/entry: Force rcu_irq_enter() when in idle task

2020-06-12 Thread Frederic Weisbecker
On Fri, Jun 12, 2020 at 03:55:00PM +0200, Thomas Gleixner wrote: > The idea of conditionally calling into rcu_irq_enter() only when RCU is > not watching turned out to be not completely thought through. > > Paul noticed occasional premature end of grace periods in RCU torture > testing. Bisection

[PATCH x86/entry: Force rcu_irq_enter() when in idle task

2020-06-12 Thread Thomas Gleixner
The idea of conditionally calling into rcu_irq_enter() only when RCU is not watching turned out to be not completely thought through. Paul noticed occasional premature end of grace periods in RCU torture testing. Bisection led to the commit which made the invocation of rcu_irq_enter() conditional