Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Tue, 21 Jan 2020 18:32:23 +0900
From:   Mike Hommey <mh@glandium.org>
To:     linux-kernel@vger.kernel.org
Subject: Re: Uptime completely off after resume from suspend
Message-ID: <20200121093223.cvvlzlpadsyqiemb@glandium.org>
References: <20200110040603.3lxanijleog6tfl6@glandium.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20200110040603.3lxanijleog6tfl6@glandium.org>
User-Agent: NeoMutt/20180716
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Fri, Jan 10, 2020 at 01:06:03PM +0900, Mike Hommey wrote:
> Hi,
> 
> I have a one month old machine, and was surprised to notice an uptime of
> several hundreds of days in top. Since I had been suspending the machine
> on many occasions, I was wondering if that could be related, so I put
> the machine to sleep, came back 30 minutes later, and uptime had
> advanced 35 days!
> 
> The actual boot time, according to journald was a week ago. And the
> "timestamp" for the suspend exit in dmesg is 312967, which is about 86
> hours, which is about half, which seems about right considering the
> suspends.
> 
> The wall clock time and date is right too at the time of resume in the
> journald logs, assuming systemd-timesyncd doesn't synchronize with NTP
> before systemd logs "Started Suspend" after resuming.
> 
> This is all with 5.5rc4 + the patches from the Debian kernel packages,
> on a Threadripper 3970X, in case that matters.
> 
> Any hints where I should be looking to find out what's going wrong?

FWIW, it doesn't seem to have happened in 8 days of running 5.5rc6.

Cheers,

Mike