From: Lukasz Luba <l.luba@partner.samsung.com>
To: linux-kernel@vger.kernel.org, linux-pm@vger.kernel.org
Cc: b.zolnierkie@samsung.com, myungjoo.ham@samsung.com,
cw00.choi@samsung.com, kyungmin.park@samsung.com,
m.szyprowski@samsung.com, s.nawrocki@samsung.com,
tkjos@google.com, joel@joelfernandes.org, chris.diamand@arm.com,
mka@chromium.org, rostedt@goodmis.org, mingo@redhat.com,
Lukasz Luba <l.luba@partner.samsung.com>
Subject: [PATCH v3 0/7] drivers: devfreq: fix and optimize workqueue mechanism
Date: Tue, 12 Feb 2019 23:23:51 +0100 [thread overview]
Message-ID: <1550010238-24002-1-git-send-email-l.luba@partner.samsung.com> (raw)
In-Reply-To: CGME20190212222422eucas1p1624203db4db3e495035820dea542e23a@eucas1p1.samsung.com
This patch set changes workqueue related features in devfreq framework.
First patch switches to delayed work instead of deferred.
The second switches to regular system work and deletes custom 'devfreq'.
Using deferred work in this context might harm the system performance.
When the CPU enters idle, deferred work is not fired. The devfreq device's
utilization does not have to be connected with a particular CPU.
The drivers for GPUs, Network on Chip, cache L3 rely on devfreq governor.
They all are missing opportunity to check the HW state and react when
the deferred work is not fired.
A corner test case, when Dynamic Memory Controller is utilized by CPUs running
on full speed, might show x5 worse performance if the crucial CPU is in idle.
There was a discussion regarding v2 that the power usage because of waking up
an idle CPU would be too high. In my opinion it won't and fixing bug is more
important than < 1% more power used [1].
I have addressed also this issue. In this patch set there is a mechanism
which prevents from to frequent checks when there is no need.
When the device enters its lowest state (OPP) the framework sets polling
interval to 'polling_idle_ms'. In default Kconfig it is 500ms, so 2 times per
second.
It is tunable from the sysfs interface per device, thus driver developer can
experiment and choose best intervals for the system.
Changes:
v3:
- reordered first two patches
- added functionality to lower power consumption when the device is less busy;
there is a new polling interval enabled when device enters lowest frequency;
- added trace events to capture the behaviour of the system
v2:
- single patch split into two
- added cover letter
link for the previous version and discussion:
https://marc.info/?l=linux-pm&m=154989907416072&w=2
https://marc.info/?l=linux-pm&m=154904631226997&w=2
Regards,
Lukasz Luba
[1] https://marc.info/?l=linux-kernel&m=155000641622443&w=2
Lukasz Luba (7):
drivers: devfreq: change deferred work into delayed
drivers: devfreq: change devfreq workqueue mechanism
Kconfig: drivers: devfreq: add default idle polling
include: devfreq: add polling_idle_ms to 'profile'
drivers: devfreq: add longer polling interval in idle
trace: events: add devfreq trace event file
drivers: devfreq: add tracing for scheduling work
MAINTAINERS | 1 +
drivers/devfreq/Kconfig | 13 +++
drivers/devfreq/devfreq.c | 184 ++++++++++++++++++++++++------
drivers/devfreq/governor.h | 3 +-
drivers/devfreq/governor_simpleondemand.c | 6 +-
include/linux/devfreq.h | 6 +
include/trace/events/devfreq.h | 39 +++++++
7 files changed, 218 insertions(+), 34 deletions(-)
create mode 100644 include/trace/events/devfreq.h
--
2.7.4
next parent reply other threads:[~2019-02-12 22:24 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <CGME20190212222422eucas1p1624203db4db3e495035820dea542e23a@eucas1p1.samsung.com>
2019-02-12 22:23 ` Lukasz Luba [this message]
[not found] ` <CGME20190212222430eucas1p1ad7992e29d224790c1e20ef7442e62fe@eucas1p1.samsung.com>
2019-02-12 22:23 ` [PATCH v3 1/7] drivers: devfreq: change deferred work into delayed Lukasz Luba
2019-02-14 4:10 ` Chanwoo Choi
[not found] ` <CGME20190212222431eucas1p1697607e6536a90283cf7dad37fa74dbb@eucas1p1.samsung.com>
2019-02-12 22:23 ` [PATCH v3 2/7] drivers: devfreq: change devfreq workqueue mechanism Lukasz Luba
2019-02-14 4:11 ` Chanwoo Choi
[not found] ` <CGME20190212222433eucas1p264602d67a916c644c7eb5012932fc17a@eucas1p2.samsung.com>
2019-02-12 22:23 ` [PATCH v3 3/7] Kconfig: drivers: devfreq: add default idle polling Lukasz Luba
[not found] ` <CGME20190212222434eucas1p134dcdce827df19704c698fd6452b0a06@eucas1p1.samsung.com>
2019-02-12 22:23 ` [PATCH v3 4/7] include: devfreq: add polling_idle_ms to 'profile' Lukasz Luba
2019-02-14 4:51 ` Chanwoo Choi
[not found] ` <CGME20190212222436eucas1p21eebc80796406787a2ebf9a84ee5b868@eucas1p2.samsung.com>
2019-02-12 22:23 ` [PATCH v3 5/7] drivers: devfreq: add longer polling interval in idle Lukasz Luba
[not found] ` <CGME20190212222437eucas1p198db6fca1f1ba3056d93c57327dd48ed@eucas1p1.samsung.com>
2019-02-12 22:23 ` [PATCH v3 6/7] trace: events: add devfreq trace event file Lukasz Luba
2019-02-12 23:14 ` Steven Rostedt
2019-02-13 13:35 ` Lukasz Luba
2019-02-14 5:01 ` Chanwoo Choi
2019-02-13 13:56 ` Steven Rostedt
2019-02-13 14:37 ` Lukasz Luba
[not found] ` <CGME20190212222438eucas1p27e020c2b36f2e5a2188e4df6fb18488b@eucas1p2.samsung.com>
2019-02-12 22:23 ` [PATCH v3 7/7] drivers: devfreq: add tracing for scheduling work Lukasz Luba
2019-02-14 4:57 ` Chanwoo Choi
2019-02-13 0:18 ` [PATCH v3 0/7] drivers: devfreq: fix and optimize workqueue mechanism Chanwoo Choi
2019-02-13 11:14 ` Lukasz Luba
2019-02-13 14:52 ` Lukasz Luba
2019-02-14 0:41 ` Chanwoo Choi
[not found] ` <CGME20190212222436eucas1p21eebc80796406787a2ebf9a84ee5b868@epcms1p3>
2019-02-18 4:33 ` [PATCH v3 5/7] drivers: devfreq: add longer polling interval in idle MyungJoo Ham
2019-02-19 8:33 ` Lukasz Luba
[not found] ` <CGME20190212222436eucas1p21eebc80796406787a2ebf9a84ee5b868@epcms1p7>
2019-02-21 5:56 ` MyungJoo Ham
2019-02-22 16:03 ` Lukasz Luba
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1550010238-24002-1-git-send-email-l.luba@partner.samsung.com \
--to=l.luba@partner.samsung.com \
--cc=b.zolnierkie@samsung.com \
--cc=chris.diamand@arm.com \
--cc=cw00.choi@samsung.com \
--cc=joel@joelfernandes.org \
--cc=kyungmin.park@samsung.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-pm@vger.kernel.org \
--cc=m.szyprowski@samsung.com \
--cc=mingo@redhat.com \
--cc=mka@chromium.org \
--cc=myungjoo.ham@samsung.com \
--cc=rostedt@goodmis.org \
--cc=s.nawrocki@samsung.com \
--cc=tkjos@google.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.