linux-kernel - Re: [paulmck-rcu:dev.2025.06.04a] [rcutorture] 9605367000: WARNING:at_kernel/rcu/rcutorture.c:#rcu_sync_torture_init

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <aEo8in1XPCBD+7Aj@xsang-OptiPlex-9020>
Date: Thu, 12 Jun 2025 10:33:46 +0800
From: Oliver Sang <oliver.sang@...el.com>
To: "Paul E. McKenney" <paulmck@...nel.org>
CC: <oe-lkp@...ts.linux.dev>, <lkp@...el.com>, <linux-kernel@...r.kernel.org>,
	<oliver.sang@...el.com>
Subject: Re: [paulmck-rcu:dev.2025.06.04a] [rcutorture]  9605367000:
 WARNING:at_kernel/rcu/rcutorture.c:#rcu_sync_torture_init_trivial[rcutorture]

hi, Paul,

On Tue, Jun 10, 2025 at 02:16:02AM -0700, Paul E. McKenney wrote:
> On Tue, Jun 10, 2025 at 03:16:52PM +0800, kernel test robot wrote:
> > 
> > 
> > Hello,
> > 
> > kernel test robot noticed "WARNING:at_kernel/rcu/rcutorture.c:#rcu_sync_torture_init_trivial[rcutorture]" on:

[...]

> > 
> > [ 1078.542757][ T1445] rcu_sync_torture_init_trivial: Non-zero onoff_interval (0) or shuffle_interval (3) breaks trivial RCU, resetting to zero
> 
> This is expected behavior.
> 
> You are specifying shuffle_interval=3, and this is not permitted for
> trivial RCU.  So rcutorture is complaining.  ;-)
> 

[...]

> > [ 1079.633479][ T1445] trivial-torture:--- Start of test: nreaders=1 nfakewriters=4 stat_interval=60 verbose=1 test_no_idle_hz=1 shuffle_interval=0 stutter=5 irqreader=1 fqs_duration=0 fqs_holdoff=0 fqs_stutter=3 test_boost=1/0 test_boost_interval=7 test_boost_duration=4 test_boost_holdoff=0 shutdown_secs=0 stall_cpu=0 stall_cpu_holdoff=10 stall_cpu_irqsoff=0 stall_cpu_block=0 stall_cpu_repeat=0 n_barrier_cbs=0 onoff_interval=0 onoff_holdoff=0 read_exit_delay=13 read_exit_burst=16 reader_flavor=1 nocbs_nthreads=0 nocbs_toggle=1000 test_nmis=0 preempt_duration=0 preempt_interval=1000 n_up_down=32
> 
> It then resets shuffle_interval to zero, which allows the test to proceed
> without error.
> 
> The problem is that non-zero shuffle_interval can cause a task to be
> migrated twice in a row without actually executing on the specified CPU
> between the two migrations.  This means that readers on the skipped CPU
> can extend across the entire trivial-RCU grace period, which of course
> breaks trivial RCU.
> 
> So please set shuffle_interval=0 when testing trivial RCU.

thanks a lot for these details! we updated test scripts per your education.

> 
> 							Thanx, Paul
> 
> > -- 
> > 0-DAY CI Kernel Test Service
> > https://github.com/intel/lkp-tests/wiki
> >