From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Paul E. McKenney" Date: Fri, 28 Jul 2017 19:03:50 +0000 Subject: Re: RCU lockup issues when CONFIG_SOFTLOCKUP_DETECTOR=n - any one else seeing this? Message-Id: <20170728190349.GM3730@linux.vnet.ibm.com> List-Id: References: <20170727014214.GH3730@linux.vnet.ibm.com> <20170727143400.23e4d2b2@roar.ozlabs.ibm.com> <20170727124913.GL3730@linux.vnet.ibm.com> <20170727144903.000022a1@huawei.com> <20170727173923.000001b2@huawei.com> <20170727165245.GD3730@linux.vnet.ibm.com> <20170728084411.00001ddb@huawei.com> <20170728142403.0000122b@huawei.com> <20170728165529.GF3730@linux.vnet.ibm.com> <20170728182053.000072aa@huawei.com> In-Reply-To: <20170728182053.000072aa@huawei.com> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable To: linux-arm-kernel@lists.infradead.org On Fri, Jul 28, 2017 at 06:27:05PM +0100, Jonathan Cameron wrote: > On Fri, 28 Jul 2017 09:55:29 -0700 > "Paul E. McKenney" wrote: >=20 > > On Fri, Jul 28, 2017 at 02:24:03PM +0100, Jonathan Cameron wrote: > > > On Fri, 28 Jul 2017 08:44:11 +0100 > > > Jonathan Cameron wrote: =20 > >=20 > > [ . . . ] > >=20 > > > Ok. Some info. I disabled a few driver (usb and SAS) in the interes= t of having > > > fewer timer events. Issue became much easier to trigger (on some run= s before > > > I could get tracing up and running) > > >e > > > So logs are large enough that pastebin doesn't like them - please sho= et if =20 > > >>e another timer period is of interest. =20 > > >=20 > > > https://pastebin.com/iUZDfQGM for the timer trace. > > > https://pastebin.com/3w1F7amH for dmesg. =20 > > >=20 > > > The relevant timeout on the RCU stall detector was 8 seconds. Event = is > > > detected around 835. > > >=20 > > > It's a lot of logs, so I haven't identified a smoking gun yet but the= re > > > may well be one in there. =20 > >=20 > > The dmesg says: > >=20 > > rcu_preempt kthread starved for 2508 jiffies! g112 c111 f0x0 RCU_GP_WAI= T_FQS(3) ->state=3D0x1 > >=20 > > So I look for "rcu_preempt" timer events and find these: > >=20 > > rcu_preempt-9 [019] .... 827.579114: timer_init: timer=FFff8017d5= fc7da0 > > rcu_preempt-9 [019] d..1 827.579115: timer_start: timer=FFff8017d= 5fc7da0 function=3Dprocess_timeout=20 > >=20 > > Next look for "ffff8017d5fc7da0" and I don't find anything else. > It does show up off the bottom of what would fit in pastebin... >=20 > rcu_preempt-9 [001] d..1 837.681077: timer_cancel: timer=FFff8= 017d5fc7da0 > rcu_preempt-9 [001] .... 837.681086: timer_init: timer=FFff801= 7d5fc7da0 > rcu_preempt-9 [001] d..1 837.681087: timer_start: timer=FFff80= 17d5fc7da0 function=3Dprocess_timeout expiresB95101298 [timeout=3D1] cpu=3D= 1 idx=3D0 flags Odd. I would expect an expiration... And ten seconds is way longer than the requested one jiffy! > > The timeout was one jiffy, and more than a second later, no expiration. > > Is it possible that this event was lost? I am not seeing any sign of > > this is the trace. > >=20 > > I don't see any sign of CPU hotplug (and I test with lots of that in > > any case). > >=20 > > The last time we saw something like this it was a timer HW/driver probl= em, > > but it is a bit hard to imagine such a problem affecting both ARM64 > > and SPARC. ;-) > Could be different issues, both of which were hidden by that lockup detec= tor. >=20 > There is an errata work around for the timers on this particular board. > I'm only vaguely aware of it, so may be unconnected. >=20 > https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit= /drivers/clocksource/arm_arch_timer.c?h=3Dv4.13-rc2&id=BB42ca47401010fc0290= 1b5e8f79e40a26f208cb >=20 > Seems unlikely though! + we've not yet seen it on the other chips that > errata effects (not that that means much). If you can reproduce quickly, might be worth trying anyway... Thanx, Paul > Jonathan >=20 > >=20 > > Thomas, any debugging suggestions? > >=20 > > Thanx, Paul > >=20 >=20