Re: RCU idle CPU detection is broken in linux-next

["Paul E. McKenney" <paulmck@linux.vnet.ibm.com>]

On Tue, Sep 25, 2012 at 01:41:18AM +0200, Frederic Weisbecker wrote:
> 2012/9/25 Frederic Weisbecker <fweisbec@gmail.com>:
> > 2012/9/25 Sasha Levin <levinsasha928@gmail.com>:
> >> On 09/25/2012 01:06 AM, Frederic Weisbecker wrote:
> >>> 2012/9/25 Sasha Levin <levinsasha928@gmail.com>:
> >>>> On 09/25/2012 12:47 AM, Sasha Levin wrote:
> >>>>>  - While I no longer see the warnings I've originally noticed, if I run with Paul's last debug patch I see the following warning:
> >>>>
> >>>> Correction: Original warnings are still there, they just got buried in the huge spew that was caused by additional debug warnings
> >>>> so I've missed them initially.
> >>>
> >>> Are they the same? Could you send me your dmesg?
> >>>
> >>> Thanks.
> >>>
> >>
> >> Log is attached, you can go directly to 168.703017 when the warnings begin.
> >
> > Thanks!
> >
> > So here is the first relevant warning:
> >
> > [  168.703017] ------------[ cut here ]------------
> > [  168.708117] WARNING: at kernel/rcutree.c:502 rcu_eqs_exit_common+0x4a/0x3a0()
> > [  168.710034] Pid: 7871, comm: trinity-child65 Tainted: G        W
> > 3.6.0-rc6-next-20120924-sasha-00030-g71f256c #5
> > [  168.710034] Call Trace:
> > [  168.710034]  <IRQ>  [<ffffffff811c737a>] ? rcu_eqs_exit_common+0x4a/0x3a0
> > [  168.710034]  [<ffffffff811078b6>] warn_slowpath_common+0x86/0xb0
> > [  168.710034]  [<ffffffff811079a5>] warn_slowpath_null+0x15/0x20
> > [  168.710034]  [<ffffffff811c737a>] rcu_eqs_exit_common+0x4a/0x3a0
> > [  168.710034]  [<ffffffff811c79cc>] rcu_eqs_exit+0x9c/0xb0
> > [  168.710034]  [<ffffffff811c7a4c>] rcu_user_exit+0x6c/0xd0
> > [  168.710034]  [<ffffffff8106eb1f>] do_general_protection+0x1f/0x170
> > [  168.710034]  [<ffffffff83a0e624>] ? restore_args+0x30/0x30
> > [  168.710034]  [<ffffffff83a0e875>] general_protection+0x25/0x30
> > [  168.710034]  [<ffffffff810a3f06>] ? native_read_msr_safe+0x6/0x20
> > [  168.710034]  [<ffffffff81a0b34b>] __rdmsr_safe_on_cpu+0x2b/0x50
> > [  168.710034]  [<ffffffff819ec971>] ? list_del+0x11/0x40
> > [  168.710034]  [<ffffffff811886dc>]
> > generic_smp_call_function_single_interrupt+0xec/0x120
> > [  168.710034]  [<ffffffff81151147>] ? account_system_vtime+0xd7/0x140
> > [  168.710034]  [<ffffffff81096f72>]
> > smp_call_function_single_interrupt+0x22/0x40
> > [  168.710034]  [<ffffffff83a0fe2f>] call_function_single_interrupt+0x6f/0x80
> > [  168.710034]  <EOI>  [<ffffffff83a0e5f4>] ? retint_restore_args+0x13/0x13
> > [  168.710034]  [<ffffffff811c7285>] ? rcu_user_enter+0x105/0x110
> > [  168.710034]  [<ffffffff8107e06d>] syscall_trace_leave+0xfd/0x150
> > [  168.710034]  [<ffffffff83a0f1ef>] int_check_syscall_exit_work+0x34/0x3d
> > [  168.710034] ---[ end trace fd408dd21b70b87c ]---
> >
> > This is an exception inside an interrupt, and the interrupt
> > interrupted RCU user mode.
> > And we have that nesting:
> >
> > rcu_irq_enter(); <--- irq entry
> > rcu_user_exit(); <--- exception entry
> >
> > And rcu_eqs_exit() doesn't handle that very well...
> 
> So either I should return immediately from rcu_user_exit() if
> we are in an interrupt, or we make rcu_user_exit() able to nest
> on rcu_irq_enter()   :)

Both of the two are eminently doable, with varying degrees of hackery.

What makes the most sense from an adaptive-idle viewpoint?

							Thanx, Paul