Re: [PATCH v4] clocksource: Scale the max retry number of watchdog read according to CPU numbers

[Feng Tang]

On Tue, Feb 20, 2024 at 09:47:11AM -0800, Paul E. McKenney wrote:
> On Tue, Feb 20, 2024 at 11:43:02PM +0800, Feng Tang wrote:
> > There was a bug on one 8-socket server that the TSC is wrongly marked
> > as 'unstable' and disabled during boot time (reproduce rate is about
> > every 120 rounds of reboot tests), with log:
> > 
> >     clocksource: timekeeping watchdog on CPU227: wd-tsc-wd excessive read-back delay of 153560ns vs. limit of 125000ns,
> >     wd-wd read-back delay only 11440ns, attempt 3, marking tsc unstable
> >     tsc: Marking TSC unstable due to clocksource watchdog
> >     TSC found unstable after boot, most likely due to broken BIOS. Use 'tsc=unstable'.
> >     sched_clock: Marking unstable (119294969739, 159204297)<-(125446229205, -5992055152)
> >     clocksource: Checking clocksource tsc synchronization from CPU 319 to CPUs 0,99,136,180,210,542,601,896.
> >     clocksource: Switched to clocksource hpet
> > 
> > The reason is for platform with lots of CPU, there are sporadic big or
> > huge read latency of read watchog/clocksource during boot or when system
> > is under stress work load, and the frequency and maximum value of the
> > latency goes up with the increasing of CPU numbers. Current code already
> > has logic to detect and filter such high latency case by reading 3 times
> > of watchdog, and check the 2 deltas. Due to the randomness of the
> > latency, there is a low possibility situation that the first delta
> > (latency) is big, but the second delta is small and looks valid, which
> > can escape from the check, and there is a 'max_cswd_read_retries' for
> > retrying that check covering this case, whose default value is only 2
> > and may be not enough for machines with huge number of CPUs.
> > 
> > So scale and enlarge the max retry number according to CPU number to
> > better filter those latency noise for large systems, which has been
> > verified fine in 4 days reboot test on the 8-socket machine.
> > 
> > Also as suggested by Thomas, remove parameter 'max_cswd_read_retries'
> > which was originally introduced to cover this.
> > 
> > Signed-off-by: Feng Tang <feng.tang@xxxxxxxxx>
> > Tested-by: Jin Wang <jin1.wang@xxxxxxxxx>
> > Tested-by: Paul E. McKenney <paulmck@xxxxxxxxxx>
> > Reviewed-by: Waiman Long <longman@xxxxxxxxxx>
> > ---
> >  
> > Hi Paul, Waiman,
> > 
> > I keep your 'Tested-by' and 'Reviewed-by' tag for v3, as I think the
> > core logic of the patch isn't changed. Please let me know if you
> > think otherwise. thanks!
> 
> I retested, and all went well, so please keep my Tested-by.
 
Thanks for the testing, again!

> One nit below...
> 
> 							Thanx, Paul
> 
> > Changelog:
> > 
> >     since v3:
> >       * Remove clocksource's module parameter 'max_cswd_read_retries' (Thomas)
> >       * Use "ilog4" instead of ilog2 for max retry calculation, and
> >         may be adjusted later (Paul)
> > 
> >     since v2:
> >       * Fix the unexported symbol of helper function being used by
> >         kernel module issue (Waiman)
> > 
> >     since v1:
> >       * Add santity check for user input value of 'max_cswd_read_retries'
> >         and a helper function for getting max retry nubmer (Paul)
> >       * Apply the same logic to watchdog test code (Waiman)
> > 
> > 
> >  Documentation/admin-guide/kernel-parameters.txt  |  6 ------
> >  include/linux/clocksource.h                      |  1 -
> >  kernel/time/clocksource-wdtest.c                 | 13 +++++++------
> >  kernel/time/clocksource.c                        | 16 +++++++++++-----
> >  .../testing/selftests/rcutorture/bin/torture.sh  |  2 +-
> >  5 files changed, 19 insertions(+), 19 deletions(-)
> > 
> > diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt
> > index 31b3a25680d0..763e96dcf8b1 100644
> > --- a/Documentation/admin-guide/kernel-parameters.txt
> > +++ b/Documentation/admin-guide/kernel-parameters.txt
> > @@ -679,12 +679,6 @@
> >  			loops can be debugged more effectively on production
> >  			systems.
> >  
> > -	clocksource.max_cswd_read_retries= [KNL]
> > -			Number of clocksource_watchdog() retries due to
> > -			external delays before the clock will be marked
> > -			unstable.  Defaults to two retries, that is,
> > -			three attempts to read the clock under test.
> > -
> >  	clocksource.verify_n_cpus= [KNL]
> >  			Limit the number of CPUs checked for clocksources
> >  			marked with CLOCK_SOURCE_VERIFY_PERCPU that
> > diff --git a/include/linux/clocksource.h b/include/linux/clocksource.h
> > index 1d42d4b17327..b93f18270b5c 100644
> > --- a/include/linux/clocksource.h
> > +++ b/include/linux/clocksource.h
> > @@ -291,7 +291,6 @@ static inline void timer_probe(void) {}
> >  #define TIMER_ACPI_DECLARE(name, table_id, fn)		\
> >  	ACPI_DECLARE_PROBE_ENTRY(timer, name, table_id, 0, NULL, 0, fn)
> >  
> > -extern ulong max_cswd_read_retries;
> >  void clocksource_verify_percpu(struct clocksource *cs);
> >  
> >  #endif /* _LINUX_CLOCKSOURCE_H */
> > diff --git a/kernel/time/clocksource-wdtest.c b/kernel/time/clocksource-wdtest.c
> > index df922f49d171..d1025f956fab 100644
> > --- a/kernel/time/clocksource-wdtest.c
> > +++ b/kernel/time/clocksource-wdtest.c
> > @@ -105,7 +105,7 @@ static int wdtest_func(void *arg)
> >  {
> >  	unsigned long j1, j2;
> >  	char *s;
> > -	int i;
> > +	int i, max_retries;
> >  
> >  	schedule_timeout_uninterruptible(holdoff * HZ);
> >  
> > @@ -139,18 +139,19 @@ static int wdtest_func(void *arg)
> >  	WARN_ON_ONCE(time_before(j2, j1 + NSEC_PER_USEC));
> >  
> >  	/* Verify tsc-like stability with various numbers of errors injected. */
> > -	for (i = 0; i <= max_cswd_read_retries + 1; i++) {
> > -		if (i <= 1 && i < max_cswd_read_retries)
> > +	max_retries = ilog2(num_online_cpus()) / 2 + 1;
> 
> Please pull this into a function so that the two calculations of
> max_retries are automatically in synchronization with each other.

Will do, thanks.

- Feng