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 for Android: free password hash cracker in your pocket
[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <Y3XyX24/kuafuqGH@feng-clx>
Date:   Thu, 17 Nov 2022 16:35:43 +0800
From:   Feng Tang <feng.tang@...el.com>
To:     "Sun, Yunying" <yunying.sun@...el.com>,
        "tglx@...utronix.de" <tglx@...utronix.de>,
        "peterz@...radead.org" <peterz@...radead.org>,
        <dave.hansen@...el.com>
CC:     "paulmck@...nel.org" <paulmck@...nel.org>,
        "longman@...hat.com" <longman@...hat.com>,
        "x86@...nel.org" <x86@...nel.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] clocksource: Print clocksource name when clocksource is
 tested unstable

On Thu, Nov 17, 2022 at 04:03:07PM +0800, Sun, Yunying wrote:
> Hi Feng,
> 
> For the "tsc fall back to hpet" issue, thanks a lot for pointing me to the fix proposal patch earlier:
>  https://lore.kernel.org/lkml/20221021062131.1826810-1-feng.tang@intel.com/ .
> 
> I shared this patch set to our customer yesterday. 
> Just got feedback that customer has verified the "tsc fall back to hpet" issue does not appear anymore with this patch set, after 24hrs stress test on 8 servers.
 
Thanks for the testing! So your case is a real case that current
'nr_online_nodes' check cannot cover, which Dave concerned about.

Thanks,
Feng

> -Yunying
> 
> -----Original Message-----
> From: Tang, Feng <feng.tang@...el.com> 
> Sent: Thursday, 17 November, 2022 14:47
> To: Sun, Yunying <yunying.sun@...el.com>
> Cc: tglx@...utronix.de; peterz@...radead.org; paulmck@...nel.org; longman@...hat.com; x86@...nel.org; linux-kernel@...r.kernel.org
> Subject: Re: [PATCH] clocksource: Print clocksource name when clocksource is tested unstable
> 
> On Wed, Nov 16, 2022 at 04:22:21PM +0800, Yunying Sun wrote:
> > Some "TSC fall back to HPET" cases are seen on systems that have more 
> > than 2 numa nodes. When this happens, in kernel log it has:
> 
> Hi Yunying,
> 
> If the system's physical sockests number <= 2, I would suggest you to try the patch https://lore.kernel.org/lkml/20221021062131.1826810-1-feng.tang@intel.com/,
> which is still under review and discussion.
> 
> Thanks,
> Feng
> 
> > clocksource: timekeeping watchdog on CPU168: hpet read-back delay of 
> > 4296200ns, attempt 4, marking unstable
> > 
> > The "hpet" here is misleading since it prints only the name of 
> > watchdog, where actually it's measuring the delay of 3 reads: wd-clocksource-wd.
> > 
> > Signed-off-by: Yunying Sun <yunying.sun@...el.com>
> > ---
> >  kernel/time/clocksource.c | 4 ++--
> >  1 file changed, 2 insertions(+), 2 deletions(-)
> > 
> > diff --git a/kernel/time/clocksource.c b/kernel/time/clocksource.c 
> > index 8058bec87ace..fac8c0d90e61 100644
> > --- a/kernel/time/clocksource.c
> > +++ b/kernel/time/clocksource.c
> > @@ -257,8 +257,8 @@ static enum wd_read_status cs_watchdog_read(struct clocksource *cs, u64 *csnow,
> >  			goto skip_test;
> >  	}
> >  
> > -	pr_warn("timekeeping watchdog on CPU%d: %s read-back delay of %lldns, attempt %d, marking unstable\n",
> > -		smp_processor_id(), watchdog->name, wd_delay, nretries);
> > +	pr_warn("timekeeping watchdog on CPU%d: wd-%s-wd read-back delay of %lldns, attempt %d, marking unstable\n",
> > +		smp_processor_id(), cs->name, wd_delay, nretries);
> >  	return WD_READ_UNSTABLE;
> >  
> >  skip_test:
> > --
> > 2.17.0
> > 

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ