* Re: [PATCH 13/19] MIPS: Unify checks for sibling CPUs
2017-08-26 12:27 ` [PATCH 13/19] MIPS: Unify checks for sibling CPUs Ralf Baechle
@ 2017-08-26 12:49 ` Rafael J. Wysocki
2017-08-29 7:42 ` Daniel Lezcano
1 sibling, 0 replies; 3+ messages in thread
From: Rafael J. Wysocki @ 2017-08-26 12:49 UTC (permalink / raw)
To: Ralf Baechle; +Cc: Paul Burton, Daniel Lezcano, linux-pm, linux-mips
On Saturday, August 26, 2017 2:27:49 PM CEST Ralf Baechle wrote:
> Hi,
>
> Paul didn't cc the other maintainers. Tglx gave me his Ack on IRC so I
> now only still need one of a drivers/cpuidle/ maintainer.
ACK
> On Sat, Aug 12, 2017 at 07:49:37PM -0700, Paul Burton wrote:
> > Date: Sat, 12 Aug 2017 19:49:37 -0700
> > From: Paul Burton <paul.burton@imgtec.com>
> > To: linux-mips@linux-mips.org
> > CC: Ralf Baechle <ralf@linux-mips.org>, Paul Burton <paul.burton@imgtec.com>
> > Subject: [PATCH 13/19] MIPS: Unify checks for sibling CPUs
> > Content-Type: text/plain
> >
> > Up until now we have open-coded checks for whether CPUs are siblings,
> > with slight variations on whether we consider the package ID or not.
> >
> > This will only get more complex when we introduce cluster support, so in
> > preparation for that this patch introduces a cpus_are_siblings()
> > function which can be used to check whether or not 2 CPUs are siblings
> > in a consistent manner.
> >
> > By checking globalnumber with the VP ID masked out this also has the
> > neat side effect of being ready for multi-cluster systems already.
> >
> > Signed-off-by: Paul Burton <paul.burton@imgtec.com>
> > Cc: Ralf Baechle <ralf@linux-mips.org>
> > Cc: linux-mips@linux-mips.org
> > ---
> >
> > arch/mips/include/asm/cpu-info.h | 17 +++++++++++++++++
> > arch/mips/kernel/smp-cps.c | 8 ++++----
> > arch/mips/kernel/smp.c | 12 +++++-------
> > drivers/cpuidle/cpuidle-cps.c | 2 +-
> > drivers/irqchip/irq-mips-cpu.c | 2 +-
> > 5 files changed, 28 insertions(+), 13 deletions(-)
> >
> > diff --git a/arch/mips/include/asm/cpu-info.h b/arch/mips/include/asm/cpu-info.h
> > index 9ae927282b12..0c61bdc82a53 100644
> > --- a/arch/mips/include/asm/cpu-info.h
> > +++ b/arch/mips/include/asm/cpu-info.h
> > @@ -158,6 +158,23 @@ static inline unsigned int cpu_vpe_id(struct cpuinfo_mips *cpuinfo)
> > extern void cpu_set_core(struct cpuinfo_mips *cpuinfo, unsigned int core);
> > extern void cpu_set_vpe_id(struct cpuinfo_mips *cpuinfo, unsigned int vpe);
> >
> > +static inline bool cpus_are_siblings(int cpua, int cpub)
> > +{
> > + struct cpuinfo_mips *infoa = &cpu_data[cpua];
> > + struct cpuinfo_mips *infob = &cpu_data[cpub];
> > + unsigned int gnuma, gnumb;
> > +
> > + if (infoa->package != infob->package)
> > + return false;
> > +
> > + gnuma = infoa->globalnumber & ~MIPS_GLOBALNUMBER_VP;
> > + gnumb = infob->globalnumber & ~MIPS_GLOBALNUMBER_VP;
> > + if (gnuma != gnumb)
> > + return false;
> > +
> > + return true;
> > +}
> > +
> > static inline unsigned long cpu_asid_inc(void)
> > {
> > return 1 << CONFIG_MIPS_ASID_SHIFT;
> > diff --git a/arch/mips/kernel/smp-cps.c b/arch/mips/kernel/smp-cps.c
> > index 699459ed293b..8cc508809466 100644
> > --- a/arch/mips/kernel/smp-cps.c
> > +++ b/arch/mips/kernel/smp-cps.c
> > @@ -147,7 +147,7 @@ static void __init cps_prepare_cpus(unsigned int max_cpus)
> > cpu_has_dc_aliases ? "dcache aliasing" : "");
> >
> > for_each_present_cpu(c) {
> > - if (cpu_core(&cpu_data[c]))
> > + if (!cpus_are_siblings(smp_processor_id(), c))
> > set_cpu_present(c, false);
> > }
> > }
> > @@ -319,10 +319,10 @@ static void cps_boot_secondary(int cpu, struct task_struct *idle)
> > mips_cm_unlock_other();
> > }
> >
> > - if (core != cpu_core(¤t_cpu_data)) {
> > + if (!cpus_are_siblings(cpu, smp_processor_id())) {
> > /* Boot a VPE on another powered up core */
> > for (remote = 0; remote < NR_CPUS; remote++) {
> > - if (cpu_core(&cpu_data[remote]) != core)
> > + if (!cpus_are_siblings(cpu, remote))
> > continue;
> > if (cpu_online(remote))
> > break;
> > @@ -431,7 +431,7 @@ void play_dead(void)
> >
> > /* Look for another online VPE within the core */
> > for_each_online_cpu(cpu_death_sibling) {
> > - if (cpu_core(&cpu_data[cpu_death_sibling]) != core)
> > + if (!cpus_are_siblings(cpu, cpu_death_sibling))
> > continue;
> >
> > /*
> > diff --git a/arch/mips/kernel/smp.c b/arch/mips/kernel/smp.c
> > index a54e5857c227..4cc43892b959 100644
> > --- a/arch/mips/kernel/smp.c
> > +++ b/arch/mips/kernel/smp.c
> > @@ -96,8 +96,7 @@ static inline void set_cpu_sibling_map(int cpu)
> >
> > if (smp_num_siblings > 1) {
> > for_each_cpu(i, &cpu_sibling_setup_map) {
> > - if (cpu_data[cpu].package == cpu_data[i].package &&
> > - cpu_core(&cpu_data[cpu]) == cpu_core(&cpu_data[i])) {
> > + if (cpus_are_siblings(cpu, i)) {
> > cpumask_set_cpu(i, &cpu_sibling_map[cpu]);
> > cpumask_set_cpu(cpu, &cpu_sibling_map[i]);
> > }
> > @@ -134,8 +133,7 @@ void calculate_cpu_foreign_map(void)
> > for_each_online_cpu(i) {
> > core_present = 0;
> > for_each_cpu(k, &temp_foreign_map)
> > - if (cpu_data[i].package == cpu_data[k].package &&
> > - cpu_core(&cpu_data[i]) == cpu_core(&cpu_data[k]))
> > + if (cpus_are_siblings(i, k))
> > core_present = 1;
> > if (!core_present)
> > cpumask_set_cpu(i, &temp_foreign_map);
> > @@ -186,11 +184,11 @@ void mips_smp_send_ipi_mask(const struct cpumask *mask, unsigned int action)
> >
> > if (mips_cpc_present()) {
> > for_each_cpu(cpu, mask) {
> > - core = cpu_core(&cpu_data[cpu]);
> > -
> > - if (core == cpu_core(¤t_cpu_data))
> > + if (cpus_are_siblings(cpu, smp_processor_id()))
> > continue;
> >
> > + core = cpu_core(&cpu_data[cpu]);
> > +
> > while (!cpumask_test_cpu(cpu, &cpu_coherent_mask)) {
> > mips_cm_lock_other(core, 0);
> > mips_cpc_lock_other(core);
> > diff --git a/drivers/cpuidle/cpuidle-cps.c b/drivers/cpuidle/cpuidle-cps.c
> > index 6041b6104f3d..72b5e47286b4 100644
> > --- a/drivers/cpuidle/cpuidle-cps.c
> > +++ b/drivers/cpuidle/cpuidle-cps.c
> > @@ -37,7 +37,7 @@ static int cps_nc_enter(struct cpuidle_device *dev,
> > * TODO: don't treat core 0 specially, just prevent the final core
> > * TODO: remap interrupt affinity temporarily
> > */
> > - if (!cpu_core(&cpu_data[dev->cpu]) && (index > STATE_NC_WAIT))
> > + if (cpus_are_siblings(0, dev->cpu) && (index > STATE_NC_WAIT))
> > index = STATE_NC_WAIT;
> >
> > /* Select the appropriate cps_pm_state */
> > diff --git a/drivers/irqchip/irq-mips-cpu.c b/drivers/irqchip/irq-mips-cpu.c
> > index 14461cbfab2f..66f97fde13d8 100644
> > --- a/drivers/irqchip/irq-mips-cpu.c
> > +++ b/drivers/irqchip/irq-mips-cpu.c
> > @@ -101,7 +101,7 @@ static void mips_mt_send_ipi(struct irq_data *d, unsigned int cpu)
> > local_irq_save(flags);
> >
> > /* We can only send IPIs to VPEs within the local core */
> > - WARN_ON(cpu_data[cpu].core != current_cpu_data.core);
> > + WARN_ON(!cpus_are_siblings(smp_processor_id(), cpu));
> >
> > vpflags = dvpe();
> > settc(cpu_vpe_id(&cpu_data[cpu]));
>
^ permalink raw reply [flat|nested] 3+ messages in thread* Re: [PATCH 13/19] MIPS: Unify checks for sibling CPUs
2017-08-26 12:27 ` [PATCH 13/19] MIPS: Unify checks for sibling CPUs Ralf Baechle
2017-08-26 12:49 ` Rafael J. Wysocki
@ 2017-08-29 7:42 ` Daniel Lezcano
1 sibling, 0 replies; 3+ messages in thread
From: Daniel Lezcano @ 2017-08-29 7:42 UTC (permalink / raw)
To: Ralf Baechle, Paul Burton, Rafael J. Wysocki, linux-pm; +Cc: linux-mips
On 26/08/2017 14:27, Ralf Baechle wrote:
> Hi,
>
> Paul didn't cc the other maintainers. Tglx gave me his Ack on IRC so I
> now only still need one of a drivers/cpuidle/ maintainer.
>
> Thanks,
>
> Ralf
>
> On Sat, Aug 12, 2017 at 07:49:37PM -0700, Paul Burton wrote:
>> Date: Sat, 12 Aug 2017 19:49:37 -0700
>> From: Paul Burton <paul.burton@imgtec.com>
>> To: linux-mips@linux-mips.org
>> CC: Ralf Baechle <ralf@linux-mips.org>, Paul Burton <paul.burton@imgtec.com>
>> Subject: [PATCH 13/19] MIPS: Unify checks for sibling CPUs
>> Content-Type: text/plain
>>
>> Up until now we have open-coded checks for whether CPUs are siblings,
>> with slight variations on whether we consider the package ID or not.
>>
>> This will only get more complex when we introduce cluster support, so in
>> preparation for that this patch introduces a cpus_are_siblings()
>> function which can be used to check whether or not 2 CPUs are siblings
>> in a consistent manner.
>>
>> By checking globalnumber with the VP ID masked out this also has the
>> neat side effect of being ready for multi-cluster systems already.
>>
>> Signed-off-by: Paul Burton <paul.burton@imgtec.com>
>> Cc: Ralf Baechle <ralf@linux-mips.org>
>> Cc: linux-mips@linux-mips.org
>> ---
Acked-by: Daniel Lezcano <daniel.lezcano@linaro.org>
--
<http://www.linaro.org/> Linaro.org │ Open source software for ARM SoCs
Follow Linaro: <http://www.facebook.com/pages/Linaro> Facebook |
<http://twitter.com/#!/linaroorg> Twitter |
<http://www.linaro.org/linaro-blog/> Blog
^ permalink raw reply [flat|nested] 3+ messages in thread