Re: [x86, locking/rwlocks, btrfs] INFO: rcu_sched self-detected stall on CPU

[Waiman Long]

On 10/04/2014 06:06 AM, Chuck Ebbert wrote:
> On Fri, 03 Oct 2014 23:27:58 -0400
> Waiman Long<waiman.long@xxxxxx>  wrote:
>
> > On 10/03/2014 09:33 AM, Fengguang Wu wrote:
> > > Hi Waiman,
> > >
> > > FYI, we noticed the below changes on commit
> > >
> > > bd01ec1a13f9a327950c8e3080096446c7804753 ("x86, locking/rwlocks: Enable qrwlocks on x86")
> > >
> > > +----------------------------------------------+------------+------------+
> > > |                                              | 70af2f8a4f | bd01ec1a13 |
> > > +----------------------------------------------+------------+------------+
> > > | boot_successes                               | 3          | 2          |
> > > | boot_failures                                | 7          | 13         |
> > > | BUG:kernel_test_crashed                      | 7          | 8          |
> > > | INFO:rcu_sched_self-detected_stall_on_CPU    | 0          | 4          |
> > > | RIP:intel_idle                               | 0          | 4          |
> > > | RIP:queue_write_lock_slowpath                | 0          | 4          |
> > > | RIP:queue_read_lock_slowpath                 | 0          | 4          |
> > > | RIP:sys_imageblit_sysimgblt                  | 0          | 2          |
> > > | RIP:default_send_IPI_mask_sequence_phys      | 0          | 1          |
> > > | RIP:memcpy                                   | 0          | 1          |
> > > | RIP:delay_tsc                                | 0          | 4          |
> > > | backtrace:cpu_startup_entry                  | 0          | 3          |
> > > | backtrace:do_fsync                           | 0          | 4          |
> > > | backtrace:SyS_fsync                          | 0          | 4          |
> > > | backtrace:normal_work_helper                 | 0          | 1          |
> > > | backtrace:vfs_write                          | 0          | 3          |
> > > | backtrace:SyS_write                          | 0          | 3          |
> > > | backtrace:do_sys_open                        | 0          | 4          |
> > > | backtrace:SyS_open                           | 0          | 4          |
> > > | backtrace:flush_to_ldisc                     | 0          | 1          |
> > > | RIP:cpu_startup_entry                        | 0          | 1          |
> > > | RIP:native_read_tsc                          | 0          | 2          |
> > > | RIP:rcu_eqs_exit_common                      | 0          | 1          |
> > > | INFO:rcu_sched_detected_stalls_on_CPUs/tasks | 0          | 1          |
> > > +----------------------------------------------+------------+------------+
> > The btrfs filesystem had problem using qrwlock. This was a known btrfs
> > problem in 3.16-rc1. The following patch by Chris should have fixed the
> > problem:
> >
> >   >  commit ea4ebde02e08558b020c4b61bb9a4c0fcf63028e
> >   >  Author: Chris Mason<clm@xxxxxx>
> >   >  Date:   Thu Jun 19 14:16:52 2014 -0700
> >   >
> >   >       Btrfs: fix deadlocks with trylock on tree nodes
> >
> > Was that patch included in your test?
> That patch went in 3.16-rc2, so it can be assumed it was included in
> the test kernel (3.16.0)

The problem should be gone in 3.16.0. I was asking because the 2 commits 
bd01ec1a13 and 70af2f8a4f are the two consecutive qrwlock patches. The 
first one adds the code while the second one enables its use in x86. So 
if you just compare these two commits, you will certainly see some 
regressions in the test.

-Longman

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/