Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Dl walleye 4.4 ab #1

Closed
wants to merge 5 commits into from

Conversation

balsini
Copy link

@balsini balsini commented Mar 27, 2018

Note that I also removed
aacf274e4b Revert "sched/cpufreq_schedutil: make worker kthread be SCHED_DEADLINE"

balsini added 5 commits March 27, 2018 17:42
This patch adds the possibility of getting the delivery of a SIGXCPU
signal whenever there is a runtime overrun. The request is done through
the sched_flags field within the sched_attr structure.

Forward port of https://lkml.org/lkml/2009/10/16/170

Tested-by: Mathieu Poirier <mathieu.poirier@linaro.org>
Signed-off-by: Juri Lelli <juri.lelli@gmail.com>
Signed-off-by: Claudio Scordino <claudio@evidence.eu.com>
Signed-off-by: Luca Abeni <luca.abeni@santannapisa.it>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Tommaso Cucinotta <tommaso.cucinotta@sssup.it>
Link:
http://lkml.kernel.org/r/1513077024-25461-1-git-send-email-claudio@evidence.eu.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
…T MERGE ]

By forcing the removal of the bandwidth check, SCHED_DEADLINE will be
able (also) to accept tasks also when the affinity is not equal to the
whole root domain.
…DO NOT MERGE ]

If the task calls a sched_getattr on its own tid with flags = 1, the dl
runqueue status is updated and the current runtime and absolute deadline
values are returned.
Use valid dl parameters for sugov kthread, in such a way that priority
inheritance does not cause problems.

Use its special flag to not count its bandwidth.
The optimization of updating task parameters without moving the task
improves the performance for rt tasks, but causes problems in the
bandwidth accounting for dl tasks.

More specifically, dl tasks need to be enqueued and dequeued to properly
update the bandwidth of the runqueue, so this workaround disables that
optimization for dl tasks.
@balsini balsini closed this Mar 28, 2018
@balsini balsini deleted the dl-walleye-4.4-ab branch March 28, 2018 14:31
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

1 participant