Re: [rcutorture] 3b745c8969: WARNING:at_mm/slab_common.c:#kmalloc_slab

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

[-- Attachment #1: Type: text/plain, Size: 1775 bytes --]

On Thu, Jul 26, 2018 at 04:34:36PM -0700, Joel Fernandes wrote:
> On Thu, Jul 26, 2018 at 04:53:25AM -0700, Paul E. McKenney wrote:
> > On Thu, Jul 26, 2018 at 04:50:15PM +0800, kernel test robot wrote:
> > > 
> > > FYI, we noticed the following commit (built with gcc-5):
> > > 
> > > commit: 3b745c8969c752601cb68c82a06735363563ab42 ("rcutorture: Make boost test more robust")
> > > https://git.kernel.org/cgit/linux/kernel/git/next/linux-next.git master
> > > 
> > > in testcase: boot
> > > 
> > > on test machine: qemu-system-x86_64 -enable-kvm -smp 2 -m 512M
> > > 
> > > caused below changes (please refer to attached dmesg/kmsg for entire log/backtrace):
> > 
> > Is this fixed by 4babd855fd61 ("rcutorture: Add support to detect
> > if boost kthread prio is too low")?  That could address the
> > rcu_torture_stats_print() failures, depending on exactly what they were.
> > (Yes, I should have reversed these two commits, but they are in -tip
> > now, so that ship has sailed.)
> > 
> > Joel, any other thoughts?
> 
> I ran the next tree myself and was not able to reproduce the issue with the
> same configuration. Although I don't have rcupdate.rcu_cpu_stall_timeout=100
> passed in like they do (which I can also try if you think its of significance
> here).
> 
> It seems from their logs that most Locking API self tests are failing. the
> Lock API test suite is run before rcutorture where rcutorture hasn't even
> started.
> 
> Also as per their rcutorture output, it appears the 'rtf' value is also
> non-zero (rcu_torture_free test). Which makes sense if the earlier memory
> allocation warnings are somehow related.

It could potentially be an OOM issue.  Something to keep an eye out for,
then.

							Thanx, Paul