Re: [PATCH] rcu: Allow to eliminate softirq processing from rcutree

2019-03-21 Thread Paul E. McKenney
On Thu, Mar 21, 2019 at 08:06:56AM -0400, Joel Fernandes wrote: > On Wed, Mar 20, 2019 at 12:28:35PM +0100, Sebastian Andrzej Siewior wrote: > > On 2019-03-19 20:26:13 [-0400], Joel Fernandes wrote: > > > > @@ -2769,19 +2782,121 @@ static void invoke_rcu_callbacks(struct > > > > rcu_data *rdp) > >

Re: [PATCH] rcu: Allow to eliminate softirq processing from rcutree

2019-03-21 Thread Joel Fernandes
On Wed, Mar 20, 2019 at 12:28:35PM +0100, Sebastian Andrzej Siewior wrote: > On 2019-03-19 20:26:13 [-0400], Joel Fernandes wrote: > > > @@ -2769,19 +2782,121 @@ static void invoke_rcu_callbacks(struct rcu_data > > > *rdp) > > > { > > > if (unlikely(!READ_ONCE(rcu_scheduler_fully_active))) > >

Re: [PATCH] rcu: Allow to eliminate softirq processing from rcutree

2019-03-20 Thread Paul E. McKenney
On Tue, Mar 19, 2019 at 08:26:13PM -0400, Joel Fernandes wrote: > Adding the r...@vger.kernel.org list as well, more comment below: > > On Fri, Mar 15, 2019 at 12:11:30PM +0100, Sebastian Andrzej Siewior wrote: > > From: "Paul E. McKenney" > > > > Running RCU out of softirq is a problem for some

Re: [PATCH] rcu: Allow to eliminate softirq processing from rcutree

2019-03-20 Thread Sebastian Andrzej Siewior
On 2019-03-19 20:26:13 [-0400], Joel Fernandes wrote: > > @@ -2769,19 +2782,121 @@ static void invoke_rcu_callbacks(struct rcu_data > > *rdp) > > { > > if (unlikely(!READ_ONCE(rcu_scheduler_fully_active))) > > return; > > - if (likely(!rcu_state.boost)) { > > - rcu_do_

Re: [PATCH] rcu: Allow to eliminate softirq processing from rcutree

2019-03-19 Thread Joel Fernandes
Adding the r...@vger.kernel.org list as well, more comment below: On Fri, Mar 15, 2019 at 12:11:30PM +0100, Sebastian Andrzej Siewior wrote: > From: "Paul E. McKenney" > > Running RCU out of softirq is a problem for some workloads that would > like to manage RCU core processing independently of

Re: [PATCH] rcu: Allow to eliminate softirq processing from rcutree

2019-03-18 Thread Paul E. McKenney
On Fri, Mar 15, 2019 at 12:11:30PM +0100, Sebastian Andrzej Siewior wrote: > From: "Paul E. McKenney" > > Running RCU out of softirq is a problem for some workloads that would > like to manage RCU core processing independently of other softirq work, > for example, setting kthread priority. > This

Re: [PATCH] rcu: Allow to eliminate softirq processing from rcutree

2019-03-15 Thread Sebastian Andrzej Siewior
On 2019-03-15 09:35:44 [-0400], Steven Rostedt wrote: > On Fri, 15 Mar 2019 12:11:30 +0100 > Sebastian Andrzej Siewior wrote: > > > +static void rcu_cpu_kthread_park(unsigned int cpu) > > +{ > > Should we add one of the trace_rcu_.. trace events here? If it is required and I'm told which one it

Re: [PATCH] rcu: Allow to eliminate softirq processing from rcutree

2019-03-15 Thread Steven Rostedt
On Fri, 15 Mar 2019 12:11:30 +0100 Sebastian Andrzej Siewior wrote: > +static void rcu_cpu_kthread_park(unsigned int cpu) > +{ Should we add one of the trace_rcu_.. trace events here? -- Steve > + per_cpu(rcu_cpu_kthread_status, cpu) = RCU_KTHREAD_OFFCPU; > +} > +

[PATCH] rcu: Allow to eliminate softirq processing from rcutree

2019-03-15 Thread Sebastian Andrzej Siewior
From: "Paul E. McKenney" Running RCU out of softirq is a problem for some workloads that would like to manage RCU core processing independently of other softirq work, for example, setting kthread priority. This commit therefore introduces the `rcunosoftirq' option which moves the RCU core work fr