* Re: [PATCH v2 1/1] docs: scheduler: Start documenting the EEVDF scheduler
2024-07-13 14:04 ` [PATCH v2 1/1] " Carlos Bilbao
@ 2024-07-14 13:40 ` Sergio González Collado
2024-07-15 5:15 ` Naik, Avadhut
2024-07-16 16:45 ` Jonathan Corbet
2 siblings, 0 replies; 6+ messages in thread
From: Sergio González Collado @ 2024-07-14 13:40 UTC (permalink / raw)
To: Carlos Bilbao
Cc: corbet, peterz, rdunlap, bilbao, jembid, linux-kernel, linux-doc
On Sat, 13 Jul 2024 at 16:05, Carlos Bilbao
<carlos.bilbao.osdev@gmail.com> wrote:
>
> Add some documentation regarding the newly introduced scheduler EEVDF.
>
> Reviewed-by: Randy Dunlap <rdunlap@infradead.org>
> Tested-by: Randy Dunlap <rdunlap@infradead.org>
> Signed-off-by: Carlos Bilbao <carlos.bilbao.osdev@gmail.com>
> ---
> Documentation/scheduler/index.rst | 1 +
> Documentation/scheduler/sched-design-CFS.rst | 10 +++--
> Documentation/scheduler/sched-eevdf.rst | 44 ++++++++++++++++++++
> 3 files changed, 51 insertions(+), 4 deletions(-)
> create mode 100644 Documentation/scheduler/sched-eevdf.rst
>
> diff --git a/Documentation/scheduler/index.rst b/Documentation/scheduler/index.rst
> index 43bd8a145b7a..1f2942c4d14b 100644
> --- a/Documentation/scheduler/index.rst
> +++ b/Documentation/scheduler/index.rst
> @@ -12,6 +12,7 @@ Scheduler
> sched-bwc
> sched-deadline
> sched-design-CFS
> + sched-eevdf
> sched-domains
> sched-capacity
> sched-energy
> diff --git a/Documentation/scheduler/sched-design-CFS.rst b/Documentation/scheduler/sched-design-CFS.rst
> index bc1e507269c6..b703c6dcb3cd 100644
> --- a/Documentation/scheduler/sched-design-CFS.rst
> +++ b/Documentation/scheduler/sched-design-CFS.rst
> @@ -8,10 +8,12 @@ CFS Scheduler
> 1. OVERVIEW
> ============
>
> -CFS stands for "Completely Fair Scheduler," and is the new "desktop" process
> -scheduler implemented by Ingo Molnar and merged in Linux 2.6.23. It is the
> -replacement for the previous vanilla scheduler's SCHED_OTHER interactivity
> -code.
> +CFS stands for "Completely Fair Scheduler," and is the "desktop" process
> +scheduler implemented by Ingo Molnar and merged in Linux 2.6.23. When
> +originally merged, it was the replacement for the previous vanilla
> +scheduler's SCHED_OTHER interactivity code. Nowadays, CFS is making room
> +for EEVDF, for which documentation can be found in
> +:ref:`sched_design_EEVDF`.
>
> 80% of CFS's design can be summed up in a single sentence: CFS basically models
> an "ideal, precise multi-tasking CPU" on real hardware.
> diff --git a/Documentation/scheduler/sched-eevdf.rst b/Documentation/scheduler/sched-eevdf.rst
> new file mode 100644
> index 000000000000..019327da333a
> --- /dev/null
> +++ b/Documentation/scheduler/sched-eevdf.rst
> @@ -0,0 +1,44 @@
> +.. _sched_design_EEVDF:
> +
> +===============
> +EEVDF Scheduler
> +===============
> +
> +The "Earliest Eligible Virtual Deadline First" (EEVDF) was first introduced
> +in a scientific publication in 1995 [1]. The Linux kernel began
> +transitioning to EEVDF in version 6.6 (as a new option in 2024), moving
> +away from the earlier Completely Fair Scheduler (CFS) in favor of a version
> +of EEVDF proposed by Peter Zijlstra in 2023 [2-4]. More information
> +regarding CFS can be found in :ref:`sched_design_CFS`.
> +
> +Similarly to CFS, EEVDF aims to distribute CPU time equally among all
> +runnable tasks with the same priority. To do so, it assigns a virtual run
> +time to each task, creating a "lag" value that can be used to determine
> +whether a task has received its fair share of CPU time. In this way, a task
> +with a positive lag is owed CPU time, while a negative lag means the task
> +has exceeded its portion. EEVDF picks tasks with lag greater or equal to
> +zero and calculates a virtual deadline (VD) for each, selecting the task
> +with the earliest VD to execute next. It's important to note that this
> +allows latency-sensitive tasks with shorter time slices to be prioritized,
> +which helps with their responsiveness.
> +
> +There are ongoing discussions on how to manage lag, especially for sleeping
> +tasks; but at the time of writing EEVDF uses a "decaying" mechanism based
> +on virtual run time (VRT). This prevents tasks from exploiting the system
> +by sleeping briefly to reset their negative lag: when a task sleeps, it
> +remains on the run queue but marked for "deferred dequeue," allowing its
> +lag to decay over VRT. Hence, long-sleeping tasks eventually have their lag
> +reset. Finally, tasks can preempt others if their VD is earlier, and tasks
> +can request specific time slices using the new sched_setattr() system call,
> +which further facilitates the job of latency-sensitive applications.
> +
> +REFERENCES
> +==========
> +
> +[1] https://citeseerx.ist.psu.edu/document?repid=rep1&type=pdf&doi=805acf7726282721504c8f00575d91ebfd750564
> +
> +[2] https://lore.kernel.org/lkml/a79014e6-ea83-b316-1e12-2ae056bda6fa@linux.vnet.ibm.com/
> +
> +[3] https://lwn.net/Articles/969062/
> +
> +[4] https://lwn.net/Articles/925371/
> --
> 2.43.0
>
Reviewed-by: Sergio González Collado <sergio.collado@gmail.com>
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH v2 1/1] docs: scheduler: Start documenting the EEVDF scheduler
2024-07-13 14:04 ` [PATCH v2 1/1] " Carlos Bilbao
2024-07-14 13:40 ` Sergio González Collado
@ 2024-07-15 5:15 ` Naik, Avadhut
2024-07-16 16:45 ` Jonathan Corbet
2 siblings, 0 replies; 6+ messages in thread
From: Naik, Avadhut @ 2024-07-15 5:15 UTC (permalink / raw)
To: Carlos Bilbao, corbet, sergio.collado, peterz, rdunlap
Cc: bilbao, jembid, linux-kernel, linux-doc
On 7/13/2024 09:04, Carlos Bilbao wrote:
> Add some documentation regarding the newly introduced scheduler EEVDF.
>
> Reviewed-by: Randy Dunlap <rdunlap@infradead.org>
> Tested-by: Randy Dunlap <rdunlap@infradead.org>
> Signed-off-by: Carlos Bilbao <carlos.bilbao.osdev@gmail.com>
> ---
> Documentation/scheduler/index.rst | 1 +
> Documentation/scheduler/sched-design-CFS.rst | 10 +++--
> Documentation/scheduler/sched-eevdf.rst | 44 ++++++++++++++++++++
> 3 files changed, 51 insertions(+), 4 deletions(-)
> create mode 100644 Documentation/scheduler/sched-eevdf.rst
>
> diff --git a/Documentation/scheduler/index.rst b/Documentation/scheduler/index.rst
> index 43bd8a145b7a..1f2942c4d14b 100644
> --- a/Documentation/scheduler/index.rst
> +++ b/Documentation/scheduler/index.rst
> @@ -12,6 +12,7 @@ Scheduler
> sched-bwc
> sched-deadline
> sched-design-CFS
> + sched-eevdf
> sched-domains
> sched-capacity
> sched-energy
> diff --git a/Documentation/scheduler/sched-design-CFS.rst b/Documentation/scheduler/sched-design-CFS.rst
> index bc1e507269c6..b703c6dcb3cd 100644
> --- a/Documentation/scheduler/sched-design-CFS.rst
> +++ b/Documentation/scheduler/sched-design-CFS.rst
> @@ -8,10 +8,12 @@ CFS Scheduler
> 1. OVERVIEW
> ============
>
> -CFS stands for "Completely Fair Scheduler," and is the new "desktop" process
> -scheduler implemented by Ingo Molnar and merged in Linux 2.6.23. It is the
> -replacement for the previous vanilla scheduler's SCHED_OTHER interactivity
> -code.
> +CFS stands for "Completely Fair Scheduler," and is the "desktop" process
> +scheduler implemented by Ingo Molnar and merged in Linux 2.6.23. When
> +originally merged, it was the replacement for the previous vanilla
> +scheduler's SCHED_OTHER interactivity code. Nowadays, CFS is making room
> +for EEVDF, for which documentation can be found in
> +:ref:`sched_design_EEVDF`.
>
> 80% of CFS's design can be summed up in a single sentence: CFS basically models
> an "ideal, precise multi-tasking CPU" on real hardware.
> diff --git a/Documentation/scheduler/sched-eevdf.rst b/Documentation/scheduler/sched-eevdf.rst
> new file mode 100644
> index 000000000000..019327da333a
> --- /dev/null
> +++ b/Documentation/scheduler/sched-eevdf.rst
> @@ -0,0 +1,44 @@
> +.. _sched_design_EEVDF:
> +
> +===============
> +EEVDF Scheduler
> +===============
> +
> +The "Earliest Eligible Virtual Deadline First" (EEVDF) was first introduced
> +in a scientific publication in 1995 [1]. The Linux kernel began
> +transitioning to EEVDF in version 6.6 (as a new option in 2024), moving
> +away from the earlier Completely Fair Scheduler (CFS) in favor of a version
> +of EEVDF proposed by Peter Zijlstra in 2023 [2-4]. More information
> +regarding CFS can be found in :ref:`sched_design_CFS`.
> +
> +Similarly to CFS, EEVDF aims to distribute CPU time equally among all
> +runnable tasks with the same priority. To do so, it assigns a virtual run
> +time to each task, creating a "lag" value that can be used to determine
> +whether a task has received its fair share of CPU time. In this way, a task
> +with a positive lag is owed CPU time, while a negative lag means the task
> +has exceeded its portion. EEVDF picks tasks with lag greater or equal to
> +zero and calculates a virtual deadline (VD) for each, selecting the task
> +with the earliest VD to execute next. It's important to note that this
> +allows latency-sensitive tasks with shorter time slices to be prioritized,
> +which helps with their responsiveness.
> +
> +There are ongoing discussions on how to manage lag, especially for sleeping
> +tasks; but at the time of writing EEVDF uses a "decaying" mechanism based
> +on virtual run time (VRT). This prevents tasks from exploiting the system
> +by sleeping briefly to reset their negative lag: when a task sleeps, it
> +remains on the run queue but marked for "deferred dequeue," allowing its
> +lag to decay over VRT. Hence, long-sleeping tasks eventually have their lag
> +reset. Finally, tasks can preempt others if their VD is earlier, and tasks
> +can request specific time slices using the new sched_setattr() system call,
> +which further facilitates the job of latency-sensitive applications.
> +
> +REFERENCES
> +==========
> +
> +[1] https://citeseerx.ist.psu.edu/document?repid=rep1&type=pdf&doi=805acf7726282721504c8f00575d91ebfd750564
> +
> +[2] https://lore.kernel.org/lkml/a79014e6-ea83-b316-1e12-2ae056bda6fa@linux.vnet.ibm.com/
> +
> +[3] https://lwn.net/Articles/969062/
> +
> +[4] https://lwn.net/Articles/925371/
LGTM.
Reviewed-by: Avadhut Naik <avadhut.naik@amd.com>
--
Thanks,
Avadhut Naik
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH v2 1/1] docs: scheduler: Start documenting the EEVDF scheduler
2024-07-13 14:04 ` [PATCH v2 1/1] " Carlos Bilbao
2024-07-14 13:40 ` Sergio González Collado
2024-07-15 5:15 ` Naik, Avadhut
@ 2024-07-16 16:45 ` Jonathan Corbet
2024-07-20 0:07 ` Carlos Bilbao
2 siblings, 1 reply; 6+ messages in thread
From: Jonathan Corbet @ 2024-07-16 16:45 UTC (permalink / raw)
To: Carlos Bilbao, sergio.collado, peterz, rdunlap
Cc: bilbao, jembid, linux-kernel, linux-doc, Carlos Bilbao
Carlos Bilbao <carlos.bilbao.osdev@gmail.com> writes:
> Add some documentation regarding the newly introduced scheduler EEVDF.
A nit, but one I wish more people would make note of...
[...]
> --- a/Documentation/scheduler/sched-design-CFS.rst
> +++ b/Documentation/scheduler/sched-design-CFS.rst
> @@ -8,10 +8,12 @@ CFS Scheduler
> 1. OVERVIEW
> ============
>
> -CFS stands for "Completely Fair Scheduler," and is the new "desktop" process
> -scheduler implemented by Ingo Molnar and merged in Linux 2.6.23. It is the
> -replacement for the previous vanilla scheduler's SCHED_OTHER interactivity
> -code.
> +CFS stands for "Completely Fair Scheduler," and is the "desktop" process
> +scheduler implemented by Ingo Molnar and merged in Linux 2.6.23. When
> +originally merged, it was the replacement for the previous vanilla
> +scheduler's SCHED_OTHER interactivity code. Nowadays, CFS is making room
> +for EEVDF, for which documentation can be found in
> +:ref:`sched_design_EEVDF`.
If, here, you just say "can be found in
Documentation/scheduler/sched-eevdf.rst", the right cross-reference will
be created and ...
> 80% of CFS's design can be summed up in a single sentence: CFS basically models
> an "ideal, precise multi-tasking CPU" on real hardware.
> diff --git a/Documentation/scheduler/sched-eevdf.rst b/Documentation/scheduler/sched-eevdf.rst
> new file mode 100644
> index 000000000000..019327da333a
> --- /dev/null
> +++ b/Documentation/scheduler/sched-eevdf.rst
> @@ -0,0 +1,44 @@
> +.. _sched_design_EEVDF:
...you can take out this unnecessary label.
Thanks,
jon
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH v2 1/1] docs: scheduler: Start documenting the EEVDF scheduler
2024-07-16 16:45 ` Jonathan Corbet
@ 2024-07-20 0:07 ` Carlos Bilbao
0 siblings, 0 replies; 6+ messages in thread
From: Carlos Bilbao @ 2024-07-20 0:07 UTC (permalink / raw)
To: Jonathan Corbet, sergio.collado, peterz, rdunlap
Cc: bilbao, jembid, linux-kernel, linux-doc
On 7/16/24 11:45, Jonathan Corbet wrote:
> Carlos Bilbao <carlos.bilbao.osdev@gmail.com> writes:
>
>> Add some documentation regarding the newly introduced scheduler EEVDF.
> A nit, but one I wish more people would make note of...
>
> [...]
>
>> --- a/Documentation/scheduler/sched-design-CFS.rst
>> +++ b/Documentation/scheduler/sched-design-CFS.rst
>> @@ -8,10 +8,12 @@ CFS Scheduler
>> 1. OVERVIEW
>> ============
>>
>> -CFS stands for "Completely Fair Scheduler," and is the new "desktop" process
>> -scheduler implemented by Ingo Molnar and merged in Linux 2.6.23. It is the
>> -replacement for the previous vanilla scheduler's SCHED_OTHER interactivity
>> -code.
>> +CFS stands for "Completely Fair Scheduler," and is the "desktop" process
>> +scheduler implemented by Ingo Molnar and merged in Linux 2.6.23. When
>> +originally merged, it was the replacement for the previous vanilla
>> +scheduler's SCHED_OTHER interactivity code. Nowadays, CFS is making room
>> +for EEVDF, for which documentation can be found in
>> +:ref:`sched_design_EEVDF`.
> If, here, you just say "can be found in
> Documentation/scheduler/sched-eevdf.rst", the right cross-reference will
> be created and ...
>
>> 80% of CFS's design can be summed up in a single sentence: CFS basically models
>> an "ideal, precise multi-tasking CPU" on real hardware.
>> diff --git a/Documentation/scheduler/sched-eevdf.rst b/Documentation/scheduler/sched-eevdf.rst
>> new file mode 100644
>> index 000000000000..019327da333a
>> --- /dev/null
>> +++ b/Documentation/scheduler/sched-eevdf.rst
>> @@ -0,0 +1,44 @@
>> +.. _sched_design_EEVDF:
> ...you can take out this unnecessary label.
Love it! Labels are a source of warnings.
Sending v3 now.
>
> Thanks,
>
> jon
Thanks,
Carlos
^ permalink raw reply [flat|nested] 6+ messages in thread