Following the consolidation and cleanup of CPU capacity in [1], this serie
reworks how the scheduler gets the pressures on CPUs. We need to take into
account all pressures applied by cpufreq on the compute capacity of a CPU
for dozens of ms or more and not only cpufreq cooling device or HW
mitigiations. we split the pressure applied on CPU's capacity in 2 parts:
- one from cpufreq and freq_qos
- one from HW high freq mitigiation.

The next step will be to add a dedicated interface for long standing
capping of the CPU capacity (i.e. for seconds or more) like the
scaling_max_freq of cpufreq sysfs. The latter is already taken into
account by this serie but as a temporary pressure which is not always the
best choice when we know that it will happen for seconds or more.

[1] https://lore.kernel.org/lkml/20231211104855.558096-1-vincent.guittot@linaro.org/

Vincent Guittot (4):
  cpufreq: Add a cpufreq pressure feedback for the scheduler
  sched: Take cpufreq feedback into account
  thermal/cpufreq: Remove arch_update_thermal_pressure()
  sched: Rename arch_update_thermal_pressure into
    arch_update_hw_pressure

 arch/arm/include/asm/topology.h               |  6 +--
 arch/arm64/include/asm/topology.h             |  6 +--
 drivers/base/arch_topology.c                  | 26 ++++-----
 drivers/cpufreq/cpufreq.c                     | 48 +++++++++++++++++
 drivers/cpufreq/qcom-cpufreq-hw.c             |  4 +-
 drivers/thermal/cpufreq_cooling.c             |  3 --
 include/linux/arch_topology.h                 |  8 +--
 include/linux/cpufreq.h                       | 10 ++++
 include/linux/sched/topology.h                |  8 +--
 .../{thermal_pressure.h => hw_pressure.h}     | 14 ++---
 include/trace/events/sched.h                  |  2 +-
 init/Kconfig                                  | 12 ++---
 kernel/sched/core.c                           |  8 +--
 kernel/sched/fair.c                           | 53 ++++++++++---------
 kernel/sched/pelt.c                           | 18 +++----
 kernel/sched/pelt.h                           | 16 +++---
 kernel/sched/sched.h                          |  4 +-
 17 files changed, 152 insertions(+), 94 deletions(-)
 rename include/trace/events/{thermal_pressure.h => hw_pressure.h} (55%)

-- 
2.34.1

Hi Vincent,

I've been waiting for this feature, thanks!


On 12/12/23 14:27, Vincent Guittot wrote:
> Following the consolidation and cleanup of CPU capacity in [1], this serie
> reworks how the scheduler gets the pressures on CPUs. We need to take into
> account all pressures applied by cpufreq on the compute capacity of a CPU
> for dozens of ms or more and not only cpufreq cooling device or HW
> mitigiations. we split the pressure applied on CPU's capacity in 2 parts:
> - one from cpufreq and freq_qos
> - one from HW high freq mitigiation.
> 
> The next step will be to add a dedicated interface for long standing
> capping of the CPU capacity (i.e. for seconds or more) like the
> scaling_max_freq of cpufreq sysfs. The latter is already taken into
> account by this serie but as a temporary pressure which is not always the
> best choice when we know that it will happen for seconds or more.
> 
> [1] https://lore.kernel.org/lkml/20231211104855.558096-1-vincent.guittot@linaro.org/
> 
> Vincent Guittot (4):
>    cpufreq: Add a cpufreq pressure feedback for the scheduler
>    sched: Take cpufreq feedback into account
>    thermal/cpufreq: Remove arch_update_thermal_pressure()
>    sched: Rename arch_update_thermal_pressure into
>      arch_update_hw_pressure
> 
>   arch/arm/include/asm/topology.h               |  6 +--
>   arch/arm64/include/asm/topology.h             |  6 +--
>   drivers/base/arch_topology.c                  | 26 ++++-----
>   drivers/cpufreq/cpufreq.c                     | 48 +++++++++++++++++
>   drivers/cpufreq/qcom-cpufreq-hw.c             |  4 +-
>   drivers/thermal/cpufreq_cooling.c             |  3 --
>   include/linux/arch_topology.h                 |  8 +--
>   include/linux/cpufreq.h                       | 10 ++++
>   include/linux/sched/topology.h                |  8 +--
>   .../{thermal_pressure.h => hw_pressure.h}     | 14 ++---
>   include/trace/events/sched.h                  |  2 +-
>   init/Kconfig                                  | 12 ++---
>   kernel/sched/core.c                           |  8 +--
>   kernel/sched/fair.c                           | 53 ++++++++++---------
>   kernel/sched/pelt.c                           | 18 +++----
>   kernel/sched/pelt.h                           | 16 +++---
>   kernel/sched/sched.h                          |  4 +-
>   17 files changed, 152 insertions(+), 94 deletions(-)
>   rename include/trace/events/{thermal_pressure.h => hw_pressure.h} (55%)
> 

I would like to test it, but something worries me. Why there is 0/5 in
this subject and only 4 patches?

Could you tell me your base branch that I can apply this, please?

Regards,
Lukasz

On Thu, 14 Dec 2023 at 09:21, Lukasz Luba <lukasz.luba@arm.com> wrote:
>
> Hi Vincent,
>
> I've been waiting for this feature, thanks!
>
>
> On 12/12/23 14:27, Vincent Guittot wrote:
> > Following the consolidation and cleanup of CPU capacity in [1], this serie
> > reworks how the scheduler gets the pressures on CPUs. We need to take into
> > account all pressures applied by cpufreq on the compute capacity of a CPU
> > for dozens of ms or more and not only cpufreq cooling device or HW
> > mitigiations. we split the pressure applied on CPU's capacity in 2 parts:
> > - one from cpufreq and freq_qos
> > - one from HW high freq mitigiation.
> >
> > The next step will be to add a dedicated interface for long standing
> > capping of the CPU capacity (i.e. for seconds or more) like the
> > scaling_max_freq of cpufreq sysfs. The latter is already taken into
> > account by this serie but as a temporary pressure which is not always the
> > best choice when we know that it will happen for seconds or more.
> >
> > [1] https://lore.kernel.org/lkml/20231211104855.558096-1-vincent.guittot@linaro.org/
> >
> > Vincent Guittot (4):
> >    cpufreq: Add a cpufreq pressure feedback for the scheduler
> >    sched: Take cpufreq feedback into account
> >    thermal/cpufreq: Remove arch_update_thermal_pressure()
> >    sched: Rename arch_update_thermal_pressure into
> >      arch_update_hw_pressure
> >
> >   arch/arm/include/asm/topology.h               |  6 +--
> >   arch/arm64/include/asm/topology.h             |  6 +--
> >   drivers/base/arch_topology.c                  | 26 ++++-----
> >   drivers/cpufreq/cpufreq.c                     | 48 +++++++++++++++++
> >   drivers/cpufreq/qcom-cpufreq-hw.c             |  4 +-
> >   drivers/thermal/cpufreq_cooling.c             |  3 --
> >   include/linux/arch_topology.h                 |  8 +--
> >   include/linux/cpufreq.h                       | 10 ++++
> >   include/linux/sched/topology.h                |  8 +--
> >   .../{thermal_pressure.h => hw_pressure.h}     | 14 ++---
> >   include/trace/events/sched.h                  |  2 +-
> >   init/Kconfig                                  | 12 ++---
> >   kernel/sched/core.c                           |  8 +--
> >   kernel/sched/fair.c                           | 53 ++++++++++---------
> >   kernel/sched/pelt.c                           | 18 +++----
> >   kernel/sched/pelt.h                           | 16 +++---
> >   kernel/sched/sched.h                          |  4 +-
> >   17 files changed, 152 insertions(+), 94 deletions(-)
> >   rename include/trace/events/{thermal_pressure.h => hw_pressure.h} (55%)
> >
>
> I would like to test it, but something worries me. Why there is 0/5 in
> this subject and only 4 patches?

I removed a patch from the series but copied/pasted the cover letter
subject without noticing the /5 instead of /4

>
> Could you tell me your base branch that I can apply this, please?

It applies on top of tip/sched/core + [1]
and you can find it here:
https://git.linaro.org/people/vincent.guittot/kernel.git/log/?h=sched/system-pressure

>
> Regards,
> Lukasz

On 12/14/23 08:29, Vincent Guittot wrote:
> On Thu, 14 Dec 2023 at 09:21, Lukasz Luba <lukasz.luba@arm.com> wrote:
>>
>> Hi Vincent,
>>
>> I've been waiting for this feature, thanks!
>>
>>
>> On 12/12/23 14:27, Vincent Guittot wrote:
>>> Following the consolidation and cleanup of CPU capacity in [1], this serie
>>> reworks how the scheduler gets the pressures on CPUs. We need to take into
>>> account all pressures applied by cpufreq on the compute capacity of a CPU
>>> for dozens of ms or more and not only cpufreq cooling device or HW
>>> mitigiations. we split the pressure applied on CPU's capacity in 2 parts:
>>> - one from cpufreq and freq_qos
>>> - one from HW high freq mitigiation.
>>>
>>> The next step will be to add a dedicated interface for long standing
>>> capping of the CPU capacity (i.e. for seconds or more) like the
>>> scaling_max_freq of cpufreq sysfs. The latter is already taken into
>>> account by this serie but as a temporary pressure which is not always the
>>> best choice when we know that it will happen for seconds or more.
>>>
>>> [1] https://lore.kernel.org/lkml/20231211104855.558096-1-vincent.guittot@linaro.org/
>>>
>>> Vincent Guittot (4):
>>>     cpufreq: Add a cpufreq pressure feedback for the scheduler
>>>     sched: Take cpufreq feedback into account
>>>     thermal/cpufreq: Remove arch_update_thermal_pressure()
>>>     sched: Rename arch_update_thermal_pressure into
>>>       arch_update_hw_pressure
>>>
>>>    arch/arm/include/asm/topology.h               |  6 +--
>>>    arch/arm64/include/asm/topology.h             |  6 +--
>>>    drivers/base/arch_topology.c                  | 26 ++++-----
>>>    drivers/cpufreq/cpufreq.c                     | 48 +++++++++++++++++
>>>    drivers/cpufreq/qcom-cpufreq-hw.c             |  4 +-
>>>    drivers/thermal/cpufreq_cooling.c             |  3 --
>>>    include/linux/arch_topology.h                 |  8 +--
>>>    include/linux/cpufreq.h                       | 10 ++++
>>>    include/linux/sched/topology.h                |  8 +--
>>>    .../{thermal_pressure.h => hw_pressure.h}     | 14 ++---
>>>    include/trace/events/sched.h                  |  2 +-
>>>    init/Kconfig                                  | 12 ++---
>>>    kernel/sched/core.c                           |  8 +--
>>>    kernel/sched/fair.c                           | 53 ++++++++++---------
>>>    kernel/sched/pelt.c                           | 18 +++----
>>>    kernel/sched/pelt.h                           | 16 +++---
>>>    kernel/sched/sched.h                          |  4 +-
>>>    17 files changed, 152 insertions(+), 94 deletions(-)
>>>    rename include/trace/events/{thermal_pressure.h => hw_pressure.h} (55%)
>>>
>>
>> I would like to test it, but something worries me. Why there is 0/5 in
>> this subject and only 4 patches?
> 
> I removed a patch from the series but copied/pasted the cover letter
> subject without noticing the /5 instead of /4

OK

> 
>>
>> Could you tell me your base branch that I can apply this, please?
> 
> It applies on top of tip/sched/core + [1]
> and you can find it here:
> https://git.linaro.org/people/vincent.guittot/kernel.git/log/?h=sched/system-pressure

Thanks for the info and handy link.

On 12/14/23 08:32, Lukasz Luba wrote:
> 
> 
> On 12/14/23 08:29, Vincent Guittot wrote:
>> On Thu, 14 Dec 2023 at 09:21, Lukasz Luba <lukasz.luba@arm.com> wrote:
>>>
>>> Hi Vincent,
>>>
>>> I've been waiting for this feature, thanks!
>>>
>>>
>>> On 12/12/23 14:27, Vincent Guittot wrote:
>>>> Following the consolidation and cleanup of CPU capacity in [1], this 
>>>> serie
>>>> reworks how the scheduler gets the pressures on CPUs. We need to 
>>>> take into
>>>> account all pressures applied by cpufreq on the compute capacity of 
>>>> a CPU
>>>> for dozens of ms or more and not only cpufreq cooling device or HW
>>>> mitigiations. we split the pressure applied on CPU's capacity in 2 
>>>> parts:
>>>> - one from cpufreq and freq_qos
>>>> - one from HW high freq mitigiation.
>>>>
>>>> The next step will be to add a dedicated interface for long standing
>>>> capping of the CPU capacity (i.e. for seconds or more) like the
>>>> scaling_max_freq of cpufreq sysfs. The latter is already taken into
>>>> account by this serie but as a temporary pressure which is not 
>>>> always the
>>>> best choice when we know that it will happen for seconds or more.
>>>>
>>>> [1] 
>>>> https://lore.kernel.org/lkml/20231211104855.558096-1-vincent.guittot@linaro.org/
>>>>
>>>> Vincent Guittot (4):
>>>>     cpufreq: Add a cpufreq pressure feedback for the scheduler
>>>>     sched: Take cpufreq feedback into account
>>>>     thermal/cpufreq: Remove arch_update_thermal_pressure()
>>>>     sched: Rename arch_update_thermal_pressure into
>>>>       arch_update_hw_pressure
>>>>
>>>>    arch/arm/include/asm/topology.h               |  6 +--
>>>>    arch/arm64/include/asm/topology.h             |  6 +--
>>>>    drivers/base/arch_topology.c                  | 26 ++++-----
>>>>    drivers/cpufreq/cpufreq.c                     | 48 +++++++++++++++++
>>>>    drivers/cpufreq/qcom-cpufreq-hw.c             |  4 +-
>>>>    drivers/thermal/cpufreq_cooling.c             |  3 --
>>>>    include/linux/arch_topology.h                 |  8 +--
>>>>    include/linux/cpufreq.h                       | 10 ++++
>>>>    include/linux/sched/topology.h                |  8 +--
>>>>    .../{thermal_pressure.h => hw_pressure.h}     | 14 ++---
>>>>    include/trace/events/sched.h                  |  2 +-
>>>>    init/Kconfig                                  | 12 ++---
>>>>    kernel/sched/core.c                           |  8 +--
>>>>    kernel/sched/fair.c                           | 53 
>>>> ++++++++++---------
>>>>    kernel/sched/pelt.c                           | 18 +++----
>>>>    kernel/sched/pelt.h                           | 16 +++---
>>>>    kernel/sched/sched.h                          |  4 +-
>>>>    17 files changed, 152 insertions(+), 94 deletions(-)
>>>>    rename include/trace/events/{thermal_pressure.h => hw_pressure.h} 
>>>> (55%)
>>>>
>>>
>>> I would like to test it, but something worries me. Why there is 0/5 in
>>> this subject and only 4 patches?
>>
>> I removed a patch from the series but copied/pasted the cover letter
>> subject without noticing the /5 instead of /4
> 
> OK
> 
>>
>>>
>>> Could you tell me your base branch that I can apply this, please?
>>
>> It applies on top of tip/sched/core + [1]
>> and you can find it here:
>> https://git.linaro.org/people/vincent.guittot/kernel.git/log/?h=sched/system-pressure
> 
> Thanks for the info and handy link.
> 

I've tested your patches with: DTPM/PowerCap + thermal gov + cpufreq
sysfs scaling_max_freq. It works fine all my cases (couldn't cause
any issues). If you like to test DTPM you will need 2 fixed pending
in Rafael's tree.

So, I'm looking for a your v2 to continue reviewing it.