Re: [PATCH][RFC v4] timekeeping: ignore the bogus sleep time if pm_trace is enabled

[Chen Yu]

Hi Oliver,
On Thu, Aug 18, 2016 at 12:36:51PM +0200, Oliver Neukum wrote:
> On Thu, 2016-08-18 at 18:43 +0800, Chen Yu wrote:
> > Previously we encountered some memory overflow issues due to
> > the bogus sleep time brought by inconsistent rtc, which is
> > triggered when pm_trace is enabled, please refer to:
> > https://patchwork.kernel.org/patch/9286365/
> > It's improper in the first place to call __timekeeping_inject_sleeptime()
> > in case that pm_trace is enabled simply because that "hash" time value
> > will wreckage the timekeeping subsystem.
> 
> Hi,
> 
> do you know since when this bug exists?
> 
I think as Xunlei mentioned, the memory overflow issue should exist since
we changed timespec to timespec64 in timekeeping_debug, which should be
in 3.17. But the bogus sleep time caused by pm_trace should always be there as
long as we use rtc for sleep compensation.
> >  /**
> > @@ -1662,6 +1668,12 @@ void timekeeping_resume(void)
> >  	} else if (timespec64_compare(&ts_new, &timekeeping_suspend_time) > 0) {
> >  		ts_delta = timespec64_sub(ts_new, timekeeping_suspend_time);
> >  		sleeptime_injected = true;
> > +		/*
> > +		 * If rtc is used as persist clock thus it
> > +		 * would be bogus when pm_trace is enabled.
> > +		 */
> > +		if (!persistent_clock_is_usable())
> > +			sleeptime_injected = false;
> >  	}
> >  
> >  	if (sleeptime_injected)
> 
> How about
> 
> sleeptime_injected = persistent_clock_is_usable();
> 
OK, this is simpler, will do in next version.

thanks,
Yu
>