linux-kernel - Re: [PATCH] clocksource: Improve cs_watchdog

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: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20211110124821.GZ641268@paulmck-ThinkPad-P17-Gen-1>
Date:   Wed, 10 Nov 2021 04:48:21 -0800
From:   "Paul E. McKenney" <paulmck@...nel.org>
To:     Heiner Kallweit <hkallweit1@...il.com>
Cc:     John Stultz <john.stultz@...aro.org>,
        Thomas Gleixner <tglx@...utronix.de>,
        Stephen Boyd <sboyd@...nel.org>,
        Feng Tang <feng.tang@...el.com>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] clocksource: Improve cs_watchdog_read()

On Tue, Nov 09, 2021 at 09:55:08PM +0100, Heiner Kallweit wrote:
> If max_cswd_read_retries is set to 0 or 1 then the current warning
> behavior doesn't seem to make too much sense to me.
> If set to 0, then we'd warn with each watchdog run.
> If set to 1, then we'd warn at the first retry, even though the commit
> description of db3a34e17433 states that one retry is expected behavior.
> If printing a message at all in this case, then it should be debug
> level.

The behavior for max_cswd_read_retries==1 is exactly what you want when
you are checking to see whether or not your system would retry at all
for the duration of a given run.

The behavior for max_cswd_read_retries==0 is exactly what you want when
you are testing the ability to print that message on a system that will
not do a retry in a reasonable period of time.

Or am I missing something here?

> Whilst being at it, move declaration of wd_end and wd_delta into the
> loop and remove not needed braces.

I am OK with moving those two variables into the "for" loop.

I am personally OK removing the braces, but if I remember correctly,
my upstream maintainer asked that I add them due to the statement being
split across two lines.

							Thanx, Paul

> Signed-off-by: Heiner Kallweit <hkallweit1@...il.com>
> ---
>  kernel/time/clocksource.c | 6 +++---
>  1 file changed, 3 insertions(+), 3 deletions(-)
> 
> diff --git a/kernel/time/clocksource.c b/kernel/time/clocksource.c
> index f29d1a524..8c0be9c02 100644
> --- a/kernel/time/clocksource.c
> +++ b/kernel/time/clocksource.c
> @@ -208,10 +208,11 @@ module_param(verify_n_cpus, int, 0644);
>  static bool cs_watchdog_read(struct clocksource *cs, u64 *csnow, u64 *wdnow)
>  {
>  	unsigned int nretries;
> -	u64 wd_end, wd_delta;
>  	int64_t wd_delay;
>  
>  	for (nretries = 0; nretries <= max_cswd_read_retries; nretries++) {
> +		u64 wd_end, wd_delta;
> +
>  		local_irq_disable();
>  		*wdnow = watchdog->read(watchdog);
>  		*csnow = cs->read(cs);
> @@ -222,10 +223,9 @@ static bool cs_watchdog_read(struct clocksource *cs, u64 *csnow, u64 *wdnow)
>  		wd_delay = clocksource_cyc2ns(wd_delta, watchdog->mult,
>  					      watchdog->shift);
>  		if (wd_delay <= WATCHDOG_MAX_SKEW) {
> -			if (nretries > 1 || nretries >= max_cswd_read_retries) {
> +			if (nretries > 1)
>  				pr_warn("timekeeping watchdog on CPU%d: %s retried %d times before success\n",
>  					smp_processor_id(), watchdog->name, nretries);
> -			}
>  			return true;
>  		}
>  	}
> -- 
> 2.33.1
>