Re: [PATCH RFC rcu/x86] Provide early rcu_cpu_starting() callback

On Fri, May 25, 2018 at 03:19:33PM -0400, Steven Rostedt wrote: > On Fri, 25 May 2018 12:18:02 -0700 > "Paul E. McKenney" wrote: > > > [ paulmck: Add !SMP support, per 0day Test Robot report. ] > > Somehow I read this as: > > [ paulmck: Add NR_CPUS=0 support ] >

Re: [PATCH RFC rcu/x86] Provide early rcu_cpu_starting() callback

On Fri, May 25, 2018 at 03:19:33PM -0400, Steven Rostedt wrote: > On Fri, 25 May 2018 12:18:02 -0700 > "Paul E. McKenney" wrote: > > > [ paulmck: Add !SMP support, per 0day Test Robot report. ] > > Somehow I read this as: > > [ paulmck: Add NR_CPUS=0 support ] > > no idea why. ;-) ;-) ;-)

Re: [PATCH RFC rcu/x86] Provide early rcu_cpu_starting() callback

On Fri, 25 May 2018 12:18:02 -0700 "Paul E. McKenney" wrote: > [ paulmck: Add !SMP support, per 0day Test Robot report. ] Somehow I read this as: [ paulmck: Add NR_CPUS=0 support ] no idea why. -- Steve

Re: [PATCH RFC rcu/x86] Provide early rcu_cpu_starting() callback

On Fri, 25 May 2018 12:18:02 -0700 "Paul E. McKenney" wrote: > [ paulmck: Add !SMP support, per 0day Test Robot report. ] Somehow I read this as: [ paulmck: Add NR_CPUS=0 support ] no idea why. -- Steve

[PATCH RFC rcu/x86] Provide early rcu_cpu_starting() callback

The x86/mtrr code does horrific things because hardware. It uses stop_machine_from_inactive_cpu(), which does a wakeup (of the stopper thread on another CPU), which uses RCU, all before the CPU is onlined. RCU complains about this, because wakeups use RCU and RCU does (rightfully) not consider

[PATCH RFC rcu/x86] Provide early rcu_cpu_starting() callback

The x86/mtrr code does horrific things because hardware. It uses stop_machine_from_inactive_cpu(), which does a wakeup (of the stopper thread on another CPU), which uses RCU, all before the CPU is onlined. RCU complains about this, because wakeups use RCU and RCU does (rightfully) not consider