Re: [PATCH 2/2] rcu: Fix up timeouts for forcing the quiescent state

2015-09-08 Thread Paul E. McKenney
On Mon, Sep 07, 2015 at 05:21:02PM +0200, Petr Mladek wrote: > On Fri 2015-09-04 16:49:46, Paul E. McKenney wrote: > > On Fri, Sep 04, 2015 at 02:11:30PM +0200, Petr Mladek wrote: > > > The deadline to force the quiescent state (jiffies_force_qs) is currently > > > updated only when the previous

Re: [PATCH 2/2] rcu: Fix up timeouts for forcing the quiescent state

2015-09-08 Thread Paul E. McKenney
On Mon, Sep 07, 2015 at 05:21:02PM +0200, Petr Mladek wrote: > On Fri 2015-09-04 16:49:46, Paul E. McKenney wrote: > > On Fri, Sep 04, 2015 at 02:11:30PM +0200, Petr Mladek wrote: > > > The deadline to force the quiescent state (jiffies_force_qs) is currently > > > updated only when the previous

Re: [PATCH 2/2] rcu: Fix up timeouts for forcing the quiescent state

2015-09-07 Thread Petr Mladek
On Fri 2015-09-04 16:49:46, Paul E. McKenney wrote: > On Fri, Sep 04, 2015 at 02:11:30PM +0200, Petr Mladek wrote: > > The deadline to force the quiescent state (jiffies_force_qs) is currently > > updated only when the previous timeout passed. But the timeout used for > > wait_event() is always

Re: [PATCH 2/2] rcu: Fix up timeouts for forcing the quiescent state

2015-09-07 Thread Petr Mladek
On Fri 2015-09-04 16:49:46, Paul E. McKenney wrote: > On Fri, Sep 04, 2015 at 02:11:30PM +0200, Petr Mladek wrote: > > The deadline to force the quiescent state (jiffies_force_qs) is currently > > updated only when the previous timeout passed. But the timeout used for > > wait_event() is always

Re: [PATCH 2/2] rcu: Fix up timeouts for forcing the quiescent state

2015-09-04 Thread Paul E. McKenney
On Fri, Sep 04, 2015 at 02:11:30PM +0200, Petr Mladek wrote: > The deadline to force the quiescent state (jiffies_force_qs) is currently > updated only when the previous timeout passed. But the timeout used for > wait_event() is always the entire original timeout. This is strange. They tell me

[PATCH 2/2] rcu: Fix up timeouts for forcing the quiescent state

2015-09-04 Thread Petr Mladek
The deadline to force the quiescent state (jiffies_force_qs) is currently updated only when the previous timeout passed. But the timeout used for wait_event() is always the entire original timeout. This is strange. First, we might miss the deadline if we wait after a spurious wake up or after

[PATCH 2/2] rcu: Fix up timeouts for forcing the quiescent state

2015-09-04 Thread Petr Mladek
The deadline to force the quiescent state (jiffies_force_qs) is currently updated only when the previous timeout passed. But the timeout used for wait_event() is always the entire original timeout. This is strange. First, we might miss the deadline if we wait after a spurious wake up or after

Re: [PATCH 2/2] rcu: Fix up timeouts for forcing the quiescent state

2015-09-04 Thread Paul E. McKenney
On Fri, Sep 04, 2015 at 02:11:30PM +0200, Petr Mladek wrote: > The deadline to force the quiescent state (jiffies_force_qs) is currently > updated only when the previous timeout passed. But the timeout used for > wait_event() is always the entire original timeout. This is strange. They tell me