[v7,14/24] x86/resctrl: Allow resctrl_arch_rmid_read() to sleep

  MPAM's cache occupancy counters can take a little while to settle once
the monitor has been configured. The maximum settling time is described
to the driver via a firmware table. The value could be large enough
that it makes sense to sleep. To avoid exposing this to resctrl, it
should be hidden behind MPAM's resctrl_arch_rmid_read().

resctrl_arch_rmid_read() may be called via IPI meaning it is unable
to sleep. In this case resctrl_arch_rmid_read() should return an error
if it needs to sleep. This will only affect MPAM platforms where
the cache occupancy counter isn't available immediately, nohz_full is
in use, and there are no housekeeping CPUs in the necessary domain.

There are three callers of resctrl_arch_rmid_read():
__mon_event_count() and __check_limbo() are both called from a
non-migrateable context. mon_event_read() invokes __mon_event_count()
using smp_call_on_cpu(), which adds work to the target CPUs workqueue.
rdtgroup_mutex() is held, meaning this cannot race with the resctrl
cpuhp callback. __check_limbo() is invoked via schedule_delayed_work_on()
also adds work to a per-cpu workqueue.

The remaining call is add_rmid_to_limbo() which is called in response
to a user-space syscall that frees an RMID. This opportunistically
reads the LLC occupancy counter on the current domain to see if the
RMID is over the dirty threshold. This has to disable preemption to
avoid reading the wrong domain's value. Disabling pre-emption here
prevents resctrl_arch_rmid_read() from sleeping.

add_rmid_to_limbo() walks each domain, but only reads the counter
on one domain. If the system has more than one domain, the RMID will
always be added to the limbo list. If the RMIDs usage was not over the
threshold, it will be removed from the list when __check_limbo() runs.
Make this the default behaviour. Free RMIDs are always added to the
limbo list for each domain.

The user visible effect of this is that a clean RMID is not available
for re-allocation immediately after 'rmdir()' completes, this behaviour
was never portable as it never happened on a machine with multiple
domains.

Removing this path allows resctrl_arch_rmid_read() to sleep if its called
with interrupts unmasked. Document this is the expected behaviour, and
add a might_sleep() annotation to catch changes that won't work on arm64.

Tested-by: Shaopeng Tan <tan.shaopeng@fujitsu.com>
Tested-by: Peter Newman <peternewman@google.com>
Reviewed-by: Shaopeng Tan <tan.shaopeng@fujitsu.com>
Signed-off-by: James Morse <james.morse@arm.com>
---
The previous version allowed resctrl_arch_rmid_read() to be called on the
wrong CPUs, but now that this needs to take nohz_full and housekeeping into
account, its too complex.

Changes since v3:
 * Removed error handling for smp_call_function_any(), this can't race
   with the cpuhp callbacks as both hold rdtgroup_mutex.
 * Switched to the alternative of removing the counter read, this simplifies
   things dramatically.

Changes since v4:
 * Messed with capitalisation.
 * Removed some dead code now that entry->busy will never be zero in
   add_rmid_to_limbo().
 * Rephrased the comment above resctrl_arch_rmid_read_context_check().

Changes since v5:
 * Really rephrased the comment above resctrl_arch_rmid_read_context_check().

No changes since v6
---
 arch/x86/kernel/cpu/resctrl/monitor.c | 25 +++++--------------------
 include/linux/resctrl.h               | 23 ++++++++++++++++++++++-
 2 files changed, 27 insertions(+), 21 deletions(-)
  

Hi James,

On 10/25/2023 11:03 AM, James Morse wrote:
> MPAM's cache occupancy counters can take a little while to settle once
> the monitor has been configured. The maximum settling time is described
> to the driver via a firmware table. The value could be large enough
> that it makes sense to sleep. To avoid exposing this to resctrl, it
> should be hidden behind MPAM's resctrl_arch_rmid_read().
> 
> resctrl_arch_rmid_read() may be called via IPI meaning it is unable
> to sleep. In this case resctrl_arch_rmid_read() should return an error
> if it needs to sleep. This will only affect MPAM platforms where
> the cache occupancy counter isn't available immediately, nohz_full is
> in use, and there are no housekeeping CPUs in the necessary domain.
> 
> There are three callers of resctrl_arch_rmid_read():
> __mon_event_count() and __check_limbo() are both called from a
> non-migrateable context. mon_event_read() invokes __mon_event_count()
> using smp_call_on_cpu(), which adds work to the target CPUs workqueue.
> rdtgroup_mutex() is held, meaning this cannot race with the resctrl
> cpuhp callback. __check_limbo() is invoked via schedule_delayed_work_on()
> also adds work to a per-cpu workqueue.
> 
> The remaining call is add_rmid_to_limbo() which is called in response
> to a user-space syscall that frees an RMID. This opportunistically
> reads the LLC occupancy counter on the current domain to see if the
> RMID is over the dirty threshold. This has to disable preemption to
> avoid reading the wrong domain's value. Disabling pre-emption here
> prevents resctrl_arch_rmid_read() from sleeping.
> 
> add_rmid_to_limbo() walks each domain, but only reads the counter
> on one domain. If the system has more than one domain, the RMID will
> always be added to the limbo list. If the RMIDs usage was not over the
> threshold, it will be removed from the list when __check_limbo() runs.
> Make this the default behaviour. Free RMIDs are always added to the
> limbo list for each domain.
> 
> The user visible effect of this is that a clean RMID is not available
> for re-allocation immediately after 'rmdir()' completes, this behaviour
> was never portable as it never happened on a machine with multiple
> domains.
> 
> Removing this path allows resctrl_arch_rmid_read() to sleep if its called
> with interrupts unmasked. Document this is the expected behaviour, and
> add a might_sleep() annotation to catch changes that won't work on arm64.
> 
> Tested-by: Shaopeng Tan <tan.shaopeng@fujitsu.com>
> Tested-by: Peter Newman <peternewman@google.com>
> Reviewed-by: Shaopeng Tan <tan.shaopeng@fujitsu.com>
> Signed-off-by: James Morse <james.morse@arm.com>
> ---
> The previous version allowed resctrl_arch_rmid_read() to be called on the
> wrong CPUs, but now that this needs to take nohz_full and housekeeping into
> account, its too complex.
> 
> Changes since v3:
>  * Removed error handling for smp_call_function_any(), this can't race
>    with the cpuhp callbacks as both hold rdtgroup_mutex.
>  * Switched to the alternative of removing the counter read, this simplifies
>    things dramatically.
> 
> Changes since v4:
>  * Messed with capitalisation.
>  * Removed some dead code now that entry->busy will never be zero in
>    add_rmid_to_limbo().
>  * Rephrased the comment above resctrl_arch_rmid_read_context_check().
> 
> Changes since v5:
>  * Really rephrased the comment above resctrl_arch_rmid_read_context_check().
> 
> No changes since v6

If I trusted this I would not have taken the time to review this patch.

> ---

Reviewed-by: Reinette Chatre <reinette.chatre@intel.com>

Reinette
  

Hi James,

On 10/25/23 13:03, James Morse wrote:
> MPAM's cache occupancy counters can take a little while to settle once
> the monitor has been configured. The maximum settling time is described
> to the driver via a firmware table. The value could be large enough
> that it makes sense to sleep. To avoid exposing this to resctrl, it
> should be hidden behind MPAM's resctrl_arch_rmid_read().
> 
> resctrl_arch_rmid_read() may be called via IPI meaning it is unable
> to sleep. In this case resctrl_arch_rmid_read() should return an error
> if it needs to sleep. This will only affect MPAM platforms where
> the cache occupancy counter isn't available immediately, nohz_full is
> in use, and there are no housekeeping CPUs in the necessary domain.
> 
> There are three callers of resctrl_arch_rmid_read():
> __mon_event_count() and __check_limbo() are both called from a
> non-migrateable context. mon_event_read() invokes __mon_event_count()
> using smp_call_on_cpu(), which adds work to the target CPUs workqueue.
> rdtgroup_mutex() is held, meaning this cannot race with the resctrl
> cpuhp callback. __check_limbo() is invoked via schedule_delayed_work_on()
> also adds work to a per-cpu workqueue.
> 
> The remaining call is add_rmid_to_limbo() which is called in response
> to a user-space syscall that frees an RMID. This opportunistically
> reads the LLC occupancy counter on the current domain to see if the
> RMID is over the dirty threshold. This has to disable preemption to
> avoid reading the wrong domain's value. Disabling pre-emption here
> prevents resctrl_arch_rmid_read() from sleeping.

I dont know what did you mean by "This has to disable preemption to
avoid reading the wrong domain's value."

Who is disabling the preemption here? Is that specific to ARM?
Can you please make that clear? Or Am i missing something?

Thanks
Babu

> 
> add_rmid_to_limbo() walks each domain, but only reads the counter
> on one domain. If the system has more than one domain, the RMID will
> always be added to the limbo list. If the RMIDs usage was not over the
> threshold, it will be removed from the list when __check_limbo() runs.
> Make this the default behaviour. Free RMIDs are always added to the
> limbo list for each domain.
> 
> The user visible effect of this is that a clean RMID is not available
> for re-allocation immediately after 'rmdir()' completes, this behaviour
> was never portable as it never happened on a machine with multiple
> domains.
> 
> Removing this path allows resctrl_arch_rmid_read() to sleep if its called
> with interrupts unmasked. Document this is the expected behaviour, and
> add a might_sleep() annotation to catch changes that won't work on arm64.
> 
> Tested-by: Shaopeng Tan <tan.shaopeng@fujitsu.com>
> Tested-by: Peter Newman <peternewman@google.com>
> Reviewed-by: Shaopeng Tan <tan.shaopeng@fujitsu.com>
> Signed-off-by: James Morse <james.morse@arm.com>
> ---
> The previous version allowed resctrl_arch_rmid_read() to be called on the
> wrong CPUs, but now that this needs to take nohz_full and housekeeping into
> account, its too complex.
> 
> Changes since v3:
>  * Removed error handling for smp_call_function_any(), this can't race
>    with the cpuhp callbacks as both hold rdtgroup_mutex.
>  * Switched to the alternative of removing the counter read, this simplifies
>    things dramatically.
> 
> Changes since v4:
>  * Messed with capitalisation.
>  * Removed some dead code now that entry->busy will never be zero in
>    add_rmid_to_limbo().
>  * Rephrased the comment above resctrl_arch_rmid_read_context_check().
> 
> Changes since v5:
>  * Really rephrased the comment above resctrl_arch_rmid_read_context_check().
> 
> No changes since v6
> ---
>  arch/x86/kernel/cpu/resctrl/monitor.c | 25 +++++--------------------
>  include/linux/resctrl.h               | 23 ++++++++++++++++++++++-
>  2 files changed, 27 insertions(+), 21 deletions(-)
> 
> diff --git a/arch/x86/kernel/cpu/resctrl/monitor.c b/arch/x86/kernel/cpu/resctrl/monitor.c
> index fa3319021881..409817b0ae2c 100644
> --- a/arch/x86/kernel/cpu/resctrl/monitor.c
> +++ b/arch/x86/kernel/cpu/resctrl/monitor.c
> @@ -277,6 +277,8 @@ int resctrl_arch_rmid_read(struct rdt_resource *r, struct rdt_domain *d,
>  	u64 msr_val, chunks;
>  	int ret;
>  
> +	resctrl_arch_rmid_read_context_check();
> +
>  	if (!cpumask_test_cpu(smp_processor_id(), &d->cpu_mask))
>  		return -EINVAL;
>  
> @@ -455,8 +457,6 @@ static void add_rmid_to_limbo(struct rmid_entry *entry)
>  {
>  	struct rdt_resource *r = &rdt_resources_all[RDT_RESOURCE_L3].r_resctrl;
>  	struct rdt_domain *d;
> -	int cpu, err;
> -	u64 val = 0;
>  	u32 idx;
>  
>  	lockdep_assert_held(&rdtgroup_mutex);
> @@ -464,17 +464,7 @@ static void add_rmid_to_limbo(struct rmid_entry *entry)
>  	idx = resctrl_arch_rmid_idx_encode(entry->closid, entry->rmid);
>  
>  	entry->busy = 0;
> -	cpu = get_cpu();
>  	list_for_each_entry(d, &r->domains, list) {
> -		if (cpumask_test_cpu(cpu, &d->cpu_mask)) {
> -			err = resctrl_arch_rmid_read(r, d, entry->closid,
> -						     entry->rmid,
> -						     QOS_L3_OCCUP_EVENT_ID,
> -						     &val);
> -			if (err || val <= resctrl_rmid_realloc_threshold)
> -				continue;
> -		}
> -
>  		/*
>  		 * For the first limbo RMID in the domain,
>  		 * setup up the limbo worker.
> @@ -484,15 +474,10 @@ static void add_rmid_to_limbo(struct rmid_entry *entry)
>  		set_bit(idx, d->rmid_busy_llc);
>  		entry->busy++;
>  	}
> -	put_cpu();
>  
> -	if (entry->busy) {
> -		rmid_limbo_count++;
> -		if (IS_ENABLED(CONFIG_RESCTRL_RMID_DEPENDS_ON_CLOSID))
> -			closid_num_dirty_rmid[entry->closid]++;
> -	} else {
> -		list_add_tail(&entry->list, &rmid_free_lru);
> -	}
> +	rmid_limbo_count++;
> +	if (IS_ENABLED(CONFIG_RESCTRL_RMID_DEPENDS_ON_CLOSID))
> +		closid_num_dirty_rmid[entry->closid]++;
>  }
>  
>  void free_rmid(u32 closid, u32 rmid)
> diff --git a/include/linux/resctrl.h b/include/linux/resctrl.h
> index bd4ec22b5a96..8649fc84aac2 100644
> --- a/include/linux/resctrl.h
> +++ b/include/linux/resctrl.h
> @@ -236,7 +236,12 @@ void resctrl_offline_domain(struct rdt_resource *r, struct rdt_domain *d);
>   * @eventid:		eventid to read, e.g. L3 occupancy.
>   * @val:		result of the counter read in bytes.
>   *
> - * Call from process context on a CPU that belongs to domain @d.
> + * Some architectures need to sleep when first programming some of the counters.
> + * (specifically: arm64's MPAM cache occupancy counters can return 'not ready'
> + *  for a short period of time). Call from a non-migrateable process context on
> + * a CPU that belongs to domain @d. e.g. use smp_call_on_cpu() or
> + * schedule_work_on(). This function can be called with interrupts masked,
> + * e.g. using smp_call_function_any(), but may consistently return an error.
>   *
>   * Return:
>   * 0 on success, or -EIO, -EINVAL etc on error.
> @@ -245,6 +250,22 @@ int resctrl_arch_rmid_read(struct rdt_resource *r, struct rdt_domain *d,
>  			   u32 closid, u32 rmid, enum resctrl_event_id eventid,
>  			   u64 *val);
>  
> +/**
> + * resctrl_arch_rmid_read_context_check()  - warn about invalid contexts
> + *
> + * When built with CONFIG_DEBUG_ATOMIC_SLEEP generate a warning when
> + * resctrl_arch_rmid_read() is called with preemption disabled.
> + *
> + * The contract with resctrl_arch_rmid_read() is that if interrupts
> + * are unmasked, it can sleep. This allows NOHZ_FULL systems to use an
> + * IPI, (and fail if the call needed to sleep), while most of the time
> + * the work is scheduled, allowing the call to sleep.
> + */
> +static inline void resctrl_arch_rmid_read_context_check(void)
> +{
> +	if (!irqs_disabled())
> +		might_sleep();
> +}
>  
>  /**
>   * resctrl_arch_reset_rmid() - Reset any private state associated with rmid
  

Hi Reinette,

On 09/11/2023 17:47, Reinette Chatre wrote:
> On 10/25/2023 11:03 AM, James Morse wrote:
>> MPAM's cache occupancy counters can take a little while to settle once
>> the monitor has been configured. The maximum settling time is described
>> to the driver via a firmware table. The value could be large enough
>> that it makes sense to sleep. To avoid exposing this to resctrl, it
>> should be hidden behind MPAM's resctrl_arch_rmid_read().
>>
>> resctrl_arch_rmid_read() may be called via IPI meaning it is unable
>> to sleep. In this case resctrl_arch_rmid_read() should return an error
>> if it needs to sleep. This will only affect MPAM platforms where
>> the cache occupancy counter isn't available immediately, nohz_full is
>> in use, and there are no housekeeping CPUs in the necessary domain.
>>
>> There are three callers of resctrl_arch_rmid_read():
>> __mon_event_count() and __check_limbo() are both called from a
>> non-migrateable context. mon_event_read() invokes __mon_event_count()
>> using smp_call_on_cpu(), which adds work to the target CPUs workqueue.
>> rdtgroup_mutex() is held, meaning this cannot race with the resctrl
>> cpuhp callback. __check_limbo() is invoked via schedule_delayed_work_on()
>> also adds work to a per-cpu workqueue.
>>
>> The remaining call is add_rmid_to_limbo() which is called in response
>> to a user-space syscall that frees an RMID. This opportunistically
>> reads the LLC occupancy counter on the current domain to see if the
>> RMID is over the dirty threshold. This has to disable preemption to
>> avoid reading the wrong domain's value. Disabling pre-emption here
>> prevents resctrl_arch_rmid_read() from sleeping.
>>
>> add_rmid_to_limbo() walks each domain, but only reads the counter
>> on one domain. If the system has more than one domain, the RMID will
>> always be added to the limbo list. If the RMIDs usage was not over the
>> threshold, it will be removed from the list when __check_limbo() runs.
>> Make this the default behaviour. Free RMIDs are always added to the
>> limbo list for each domain.
>>
>> The user visible effect of this is that a clean RMID is not available
>> for re-allocation immediately after 'rmdir()' completes, this behaviour
>> was never portable as it never happened on a machine with multiple
>> domains.
>>
>> Removing this path allows resctrl_arch_rmid_read() to sleep if its called
>> with interrupts unmasked. Document this is the expected behaviour, and
>> add a might_sleep() annotation to catch changes that won't work on arm64.

[...]

>> Changes since v3:
>>  * Removed error handling for smp_call_function_any(), this can't race
>>    with the cpuhp callbacks as both hold rdtgroup_mutex.
>>  * Switched to the alternative of removing the counter read, this simplifies
>>    things dramatically.
>>
>> Changes since v4:
>>  * Messed with capitalisation.
>>  * Removed some dead code now that entry->busy will never be zero in
>>    add_rmid_to_limbo().
>>  * Rephrased the comment above resctrl_arch_rmid_read_context_check().
>>
>> Changes since v5:
>>  * Really rephrased the comment above resctrl_arch_rmid_read_context_check().
>>
>> No changes since v6
> 
> If I trusted this I would not have taken the time to review this patch.

I'm not sure what you want me to do from this comment ... but this 'no changes' annotation
doesn't work for either of us, so I'll remove them.


> Reviewed-by: Reinette Chatre <reinette.chatre@intel.com>


Thanks,

James
  

Hi Babu,

On 09/11/2023 20:42, Moger, Babu wrote:
> On 10/25/23 13:03, James Morse wrote:
>> MPAM's cache occupancy counters can take a little while to settle once
>> the monitor has been configured. The maximum settling time is described
>> to the driver via a firmware table. The value could be large enough
>> that it makes sense to sleep. To avoid exposing this to resctrl, it
>> should be hidden behind MPAM's resctrl_arch_rmid_read().
>>
>> resctrl_arch_rmid_read() may be called via IPI meaning it is unable
>> to sleep. In this case resctrl_arch_rmid_read() should return an error
>> if it needs to sleep. This will only affect MPAM platforms where
>> the cache occupancy counter isn't available immediately, nohz_full is
>> in use, and there are no housekeeping CPUs in the necessary domain.
>>
>> There are three callers of resctrl_arch_rmid_read():
>> __mon_event_count() and __check_limbo() are both called from a
>> non-migrateable context. mon_event_read() invokes __mon_event_count()
>> using smp_call_on_cpu(), which adds work to the target CPUs workqueue.
>> rdtgroup_mutex() is held, meaning this cannot race with the resctrl
>> cpuhp callback. __check_limbo() is invoked via schedule_delayed_work_on()
>> also adds work to a per-cpu workqueue.
>>
>> The remaining call is add_rmid_to_limbo() which is called in response
>> to a user-space syscall that frees an RMID. This opportunistically
>> reads the LLC occupancy counter on the current domain to see if the
>> RMID is over the dirty threshold. This has to disable preemption to
>> avoid reading the wrong domain's value. Disabling pre-emption here
>> prevents resctrl_arch_rmid_read() from sleeping.

> I dont know what did you mean by "This has to disable preemption to
> avoid reading the wrong domain's value."

Pre-emption lets this thread be scheduled out, and potentially scheduled back in on a
different CPU, possibly in a different domain. Any code with the concept of 'this domain'
has to to ensure it can't be migrated. Disabling pre-emption is the most common way of
doing that.

Disabling pre-emption also prevents the thread from sleeping, because it can't be
scheduled out.


> Who is disabling the preemption here? Is that specific to ARM?
> Can you please make that clear? Or Am i missing something?

add_rmid_to_limbo() is calling get_cpu(), which raises the pre-empt counter.
If it only wanted the CPU number it could have just called smp_processor_id() - but that
wouldn't be safe because the thread can be migrated, meaning the cpu number can change.

All this is to ensure that cpumask_test_cpu() and resctrl_arch_rmid_read() run on the same
CPU.


Thanks,

James

>> add_rmid_to_limbo() walks each domain, but only reads the counter
>> on one domain. If the system has more than one domain, the RMID will
>> always be added to the limbo list. If the RMIDs usage was not over the
>> threshold, it will be removed from the list when __check_limbo() runs.
>> Make this the default behaviour. Free RMIDs are always added to the
>> limbo list for each domain.
>>
>> The user visible effect of this is that a clean RMID is not available
>> for re-allocation immediately after 'rmdir()' completes, this behaviour
>> was never portable as it never happened on a machine with multiple
>> domains.
>>
>> Removing this path allows resctrl_arch_rmid_read() to sleep if its called
>> with interrupts unmasked. Document this is the expected behaviour, and
>> add a might_sleep() annotation to catch changes that won't work on arm64.


>> diff --git a/arch/x86/kernel/cpu/resctrl/monitor.c b/arch/x86/kernel/cpu/resctrl/monitor.c
>> index fa3319021881..409817b0ae2c 100644
>> --- a/arch/x86/kernel/cpu/resctrl/monitor.c
>> +++ b/arch/x86/kernel/cpu/resctrl/monitor.c
>> @@ -464,17 +464,7 @@ static void add_rmid_to_limbo(struct rmid_entry *entry)
>>  	idx = resctrl_arch_rmid_idx_encode(entry->closid, entry->rmid);
>>  
>>  	entry->busy = 0;
>> -	cpu = get_cpu();
>>  	list_for_each_entry(d, &r->domains, list) {
>> -		if (cpumask_test_cpu(cpu, &d->cpu_mask)) {
>> -			err = resctrl_arch_rmid_read(r, d, entry->closid,
>> -						     entry->rmid,
>> -						     QOS_L3_OCCUP_EVENT_ID,
>> -						     &val);
>> -			if (err || val <= resctrl_rmid_realloc_threshold)
>> -				continue;
>> -		}
>> -
>>  		/*
>>  		 * For the first limbo RMID in the domain,
>>  		 * setup up the limbo worker.
>> @@ -484,15 +474,10 @@ static void add_rmid_to_limbo(struct rmid_entry *entry)
>>  		set_bit(idx, d->rmid_busy_llc);
>>  		entry->busy++;
>>  	}
>> -	put_cpu();
>>  
>> -	if (entry->busy) {
>> -		rmid_limbo_count++;
>> -		if (IS_ENABLED(CONFIG_RESCTRL_RMID_DEPENDS_ON_CLOSID))
>> -			closid_num_dirty_rmid[entry->closid]++;
>> -	} else {
>> -		list_add_tail(&entry->list, &rmid_free_lru);
>> -	}
>> +	rmid_limbo_count++;
>> +	if (IS_ENABLED(CONFIG_RESCTRL_RMID_DEPENDS_ON_CLOSID))
>> +		closid_num_dirty_rmid[entry->closid]++;
>>  }
>>
  

Hi James,

On 12/14/2023 3:37 AM, James Morse wrote:
> On 09/11/2023 17:47, Reinette Chatre wrote:
>> On 10/25/2023 11:03 AM, James Morse wrote:

>>>
>>> No changes since v6
>>
>> If I trusted this I would not have taken the time to review this patch.
> 
> I'm not sure what you want me to do from this comment ... but this 'no changes' annotation
> doesn't work for either of us, so I'll remove them.
> 

The "no changes" annotation is valuable when it is accurate. When the patch
explicitly claims to have no changes since previous version when in fact there
are changes then it complicates the review process.

Reinette