Re: + watchdog-fix-possible-soft-lockup-warning-at-bootup-v2.patch added to -mm tree

Re: + watchdog-fix-possible-soft-lockup-warning-at-bootup-v2.patch added to -mm tree

[Thomas Gleixner]

akpm@linux-foundation.org writes:

> ------------------------------------------------------
> From: Waiman Long <longman@redhat.com>
> Subject: watchdog: Fix possible soft lockup warning at bootup

Completely empty changelog without any justification for this change.

> Link: http://lkml.kernel.org/r/20200103151032.19590-1-longman@redhat.com
> Signed-off-by: Waiman Long <longman@redhat.com>
> Cc: Ingo Molnar <mingo@kernel.org>
> Cc: Thomas Gleixner <tglx@linutronix.de>
> Cc: Mike Rapoport <rppt@linux.ibm.com>
> Cc: Kees Cook <keescook@chromium.org>
> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
> ---
>
>  kernel/watchdog.c |    4 +---
>  1 file changed, 1 insertion(+), 3 deletions(-)
>
> --- a/kernel/watchdog.c~watchdog-fix-possible-soft-lockup-warning-at-bootup-v2
> +++ a/kernel/watchdog.c
> @@ -496,9 +496,7 @@ static void watchdog_enable(unsigned int
>  		      HRTIMER_MODE_REL_PINNED_HARD);
>  
>  	/* Initialize timestamp */
> -	if (system_state != SYSTEM_BOOTING)
> -		__touch_watchdog();
> -
> +	__touch_watchdog();
>  	/* Enable the perf event */
>  	if (watchdog_enabled & NMI_WATCHDOG_ENABLED)
>  		watchdog_nmi_enable(cpu);
> _
>
> Patches currently in -mm which might be from longman@redhat.com are
>
> watchdog-fix-possible-soft-lockup-warning-at-bootup.patch
> watchdog-fix-possible-soft-lockup-warning-at-bootup-v2.patch

Please drop both. The initial one just papers over timer interrupt loss
and weakens debugging. That V2 thing is just fixing up the wreckage
introduced in the initial one.

Thanks,

        tglx

Re: + watchdog-fix-possible-soft-lockup-warning-at-bootup-v2.patch added to -mm tree

[Andrew Morton]

On Thu, 16 Jan 2020 12:55:49 +0100 Thomas Gleixner <tglx@linutronix.de> wrote:

> akpm@linux-foundation.org writes:
> 
> > ------------------------------------------------------
> > From: Waiman Long <longman@redhat.com>
> > Subject: watchdog: Fix possible soft lockup warning at bootup
> 
> Completely empty changelog without any justification for this change.

Well, this is the v1->v2 delta.  It's removing an unnecessary change
from the v1 patch.  Normally that's covered in the main patch's
changelog but removing an unneeded change isn't changeloggable.

> > Link: http://lkml.kernel.org/r/20200103151032.19590-1-longman@redhat.com
> > Signed-off-by: Waiman Long <longman@redhat.com>
> > Cc: Ingo Molnar <mingo@kernel.org>
> > Cc: Thomas Gleixner <tglx@linutronix.de>
> > Cc: Mike Rapoport <rppt@linux.ibm.com>
> > Cc: Kees Cook <keescook@chromium.org>
> > Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
> > ---
> >
> >  kernel/watchdog.c |    4 +---
> >  1 file changed, 1 insertion(+), 3 deletions(-)
> >
> > --- a/kernel/watchdog.c~watchdog-fix-possible-soft-lockup-warning-at-bootup-v2
> > +++ a/kernel/watchdog.c
> > @@ -496,9 +496,7 @@ static void watchdog_enable(unsigned int
> >  		      HRTIMER_MODE_REL_PINNED_HARD);
> >  
> >  	/* Initialize timestamp */
> > -	if (system_state != SYSTEM_BOOTING)
> > -		__touch_watchdog();
> > -
> > +	__touch_watchdog();
> >  	/* Enable the perf event */
> >  	if (watchdog_enabled & NMI_WATCHDOG_ENABLED)
> >  		watchdog_nmi_enable(cpu);
> > _
> >
> > Patches currently in -mm which might be from longman@redhat.com are
> >
> > watchdog-fix-possible-soft-lockup-warning-at-bootup.patch
> > watchdog-fix-possible-soft-lockup-warning-at-bootup-v2.patch
> 
> Please drop both. The initial one just papers over timer interrupt loss
> and weakens debugging. That V2 thing is just fixing up the wreckage
> introduced in the initial one.

Not really - it's removing a change which was deemed unneeded.

But sure, dropped.