[PATCH 0/2] s390: Hotplug notifier transitions

[PATCH 0/2] s390: Hotplug notifier transitions

[Anna-Maria Gleixner]

The following patches fix missing error handling of hotplug notifier
transitions.

Thanks,
Anna-Maria

---
 perf_cpum_cf.c |    1 +
 perf_cpum_sf.c |    2 +-
 2 files changed, 2 insertions(+), 1 deletion(-)

[PATCH 1/2] s390/cpum_cf: Fix missing cpu hotplug notifier transition

[Anna-Maria Gleixner]

The cpumf_pmu_notfier() hotplug callback lacks handling of the
CPU_DOWN_FAILED case. That means, if CPU_DOWN_PREPARE failes, the PMC
of the CPU is not setup again.

Add handling for CPU_DOWN_FAILED transition to setup the PMC of the
CPU.

Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: linux-s390@vger.kernel.org
Signed-off-by: Anna-Maria Gleixner <anna-maria@linutronix.de>
---
 arch/s390/kernel/perf_cpum_cf.c |    1 +
 1 file changed, 1 insertion(+)

--- a/arch/s390/kernel/perf_cpum_cf.c
+++ b/arch/s390/kernel/perf_cpum_cf.c
@@ -670,6 +670,7 @@ static int cpumf_pmu_notifier(struct not
 
 	switch (action & ~CPU_TASKS_FROZEN) {
 	case CPU_ONLINE:
+	case CPU_DOWN_FAILED:
 		flags = PMC_INIT;
 		smp_call_function_single(cpu, setup_pmc_cpu, &flags, 1);
 		break;

[PATCH 2/2] s390/cpum_sf: Fix cpu hotplug notifier transitions

[Anna-Maria Gleixner]

The cpumf_pmu_notfier() hotplug callback lacks handling of the
CPU_DOWN_FAILED case. That means, if CPU_DOWN_PREPARE failes, the PMC
of the CPU is not setup again. Furthermore the CPU_ONLINE_FROZEN case
will never be processed because of masking the switch expression with
CPU_TASKS_FROZEN.

Add handling for CPU_DOWN_FAILED transition to setup the PMC of the
CPU. Remove CPU_ONLINE_FROZEN case.

Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: linux-s390@vger.kernel.org
Signed-off-by: Anna-Maria Gleixner <anna-maria@linutronix.de>
---
 arch/s390/kernel/perf_cpum_sf.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

--- a/arch/s390/kernel/perf_cpum_sf.c
+++ b/arch/s390/kernel/perf_cpum_sf.c
@@ -1518,7 +1518,7 @@ static int cpumf_pmu_notifier(struct not
 
 	switch (action & ~CPU_TASKS_FROZEN) {
 	case CPU_ONLINE:
-	case CPU_ONLINE_FROZEN:
+	case CPU_DOWN_FAILED:
 		flags = PMC_INIT;
 		smp_call_function_single(cpu, setup_pmc_cpu, &flags, 1);
 		break;

Re: [PATCH 2/2] s390/cpum_sf: Fix cpu hotplug notifier transitions

[Heiko Carstens]

On Fri, Mar 11, 2016 at 11:52:38AM +0100, Anna-Maria Gleixner wrote:
> The cpumf_pmu_notfier() hotplug callback lacks handling of the
> CPU_DOWN_FAILED case. That means, if CPU_DOWN_PREPARE failes, the PMC
> of the CPU is not setup again. Furthermore the CPU_ONLINE_FROZEN case
> will never be processed because of masking the switch expression with
> CPU_TASKS_FROZEN.
> 
> Add handling for CPU_DOWN_FAILED transition to setup the PMC of the
> CPU. Remove CPU_ONLINE_FROZEN case.
> 
> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
> Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
> Cc: linux-s390@vger.kernel.org
> Signed-off-by: Anna-Maria Gleixner <anna-maria@linutronix.de>
> ---
>  arch/s390/kernel/perf_cpum_sf.c |    2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> --- a/arch/s390/kernel/perf_cpum_sf.c
> +++ b/arch/s390/kernel/perf_cpum_sf.c
> @@ -1518,7 +1518,7 @@ static int cpumf_pmu_notifier(struct not
> 
>  	switch (action & ~CPU_TASKS_FROZEN) {
>  	case CPU_ONLINE:
> -	case CPU_ONLINE_FROZEN:
> +	case CPU_DOWN_FAILED:
>  		flags = PMC_INIT;
>  		smp_call_function_single(cpu, setup_pmc_cpu, &flags, 1);
>  		break;

This is a bit odd, but independent from your patch: setup_pmc_cpu() will
only deallocate buffers but not allocate them.

Looking at the code a bit further this seems to imply that sampling will
not work on cpus that were added later.

Hendrik, could you please have a look?

Re: [PATCH 2/2] s390/cpum_sf: Fix cpu hotplug notifier transitions

[Hendrik Brueckner]

On Mon, Mar 14, 2016 at 09:51:50AM +0100, Heiko Carstens wrote:
> On Fri, Mar 11, 2016 at 11:52:38AM +0100, Anna-Maria Gleixner wrote:
> > The cpumf_pmu_notfier() hotplug callback lacks handling of the
> > CPU_DOWN_FAILED case. That means, if CPU_DOWN_PREPARE failes, the PMC
> > of the CPU is not setup again. Furthermore the CPU_ONLINE_FROZEN case
> > will never be processed because of masking the switch expression with
> > CPU_TASKS_FROZEN.
> > 
> > Add handling for CPU_DOWN_FAILED transition to setup the PMC of the
> > CPU. Remove CPU_ONLINE_FROZEN case.
> > 
> > Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
> > Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
> > Cc: linux-s390@vger.kernel.org
> > Signed-off-by: Anna-Maria Gleixner <anna-maria@linutronix.de>
> > ---
> >  arch/s390/kernel/perf_cpum_sf.c |    2 +-
> >  1 file changed, 1 insertion(+), 1 deletion(-)
> > 
> > --- a/arch/s390/kernel/perf_cpum_sf.c
> > +++ b/arch/s390/kernel/perf_cpum_sf.c
> > @@ -1518,7 +1518,7 @@ static int cpumf_pmu_notifier(struct not
> > 
> >  	switch (action & ~CPU_TASKS_FROZEN) {
> >  	case CPU_ONLINE:
> > -	case CPU_ONLINE_FROZEN:
> > +	case CPU_DOWN_FAILED:

The change looks ok for me.

Acked-by: Hendrik Brueckner <brueckner@linux.vnet.ibm.com>

> >  		flags = PMC_INIT;
> >  		smp_call_function_single(cpu, setup_pmc_cpu, &flags, 1);
> >  		break;
> 
> This is a bit odd, but independent from your patch: setup_pmc_cpu() will
> only deallocate buffers but not allocate them.

That's right.  Allocation is done when an perf event is about to be scheduled
on a CPU.

> 
> Looking at the code a bit further this seems to imply that sampling will
> not work on cpus that were added later.

That might be a problem for task-based perf events.  I will have a closer
look into that issue.

Many thanks.

Kind regards,
  Hendrik

Re: [PATCH 0/2] s390: Hotplug notifier transitions

[Heiko Carstens]

On Fri, Mar 11, 2016 at 11:52:36AM +0100, Anna-Maria Gleixner wrote:
> The following patches fix missing error handling of hotplug notifier
> transitions.
> 
> Thanks,
> Anna-Maria
> 
> ---
>  perf_cpum_cf.c |    1 +
>  perf_cpum_sf.c |    2 +-
>  2 files changed, 2 insertions(+), 1 deletion(-)

Applied, thanks!