From: Kalle Valo <kalle.valo@nokia.com>
To: "Johannes Berg" <johannes@sipsolutions.net>
Cc: linux-wireless@vger.kernel.org, Dan Williams <dcbw@redhat.com>
Subject: Re: [PATCH v6 0/3] mac80211 dynamic power save
Date: Sun, 28 Dec 2008 14:54:33 +0200 [thread overview]
Message-ID: <871vvso3xy.fsf@nokia.com> (raw)
In-Reply-To: <1230069346.31228.51.camel@johannes> (ext Johannes Berg's message of "Tue\, 23 Dec 2008 22\:55\:46 +0100")
Johannes Berg <johannes@sipsolutions.net> writes:
> On Thu, 2008-12-18 at 23:35 +0200, Kalle Valo wrote:
>> Here is v5 of my dynamic powersave patches. I have tested them with
>> stlc45xx on Nokia N810.
>
>> Power save related features which I'll implement later:
>> o disable power save when software scanning
>> o move dynamic_ps_timeout to hw.conf so that drivers can use it in
>> case they need to
>> o beacon period (wakeup period) to the drivers, can be set from user
>> space but otherwise is bss's DTIM value
>
> Now that these are in, and more things for user interface are on the
> way, we should probably think about implementing the policy in
> userspace.
Good idea. Also we should consider the interface to the user space,
what parameters we want to provide to user space etc.
> Should that be part of NM?
I think it should be in NetworkManager, and also Intel's Connection
Manager might want to control power save parameters. And in Nokia we
want to control power save parameters from wlancond, our own Wireless
Extension dbus wrapper.
> It could take into account things like the system power state
> (AC/battery) for the decision, at least by default?
Actually I'm not that enthuastic about AC/battery mode detection, I
would like to save power even in AC mode. In my opinion we should have
more dynamic logic based on current state, for example based on amount
of transfered traffic, QoS, user activity (eg. is the display blanked)
and what not. But I guess AC/battery detection is a good start.
As for the user space interface, here's some ideas I have been
thinking:
Wakeup period (iwconfig wlan0 power period 2), sleep timeout (iwconfig
wlan0 power timeout 500m) and enable/disable of power save (iwconfig
wlan0 power on) are enough for now. If needed, we can add more
parameters later.
Wakeup period should be the DTIM value of the BSS by default and set
by mac80211, but user space can change the value whenever it wants.
Power save should be disabled by default in mac80211 to avoid all the
interoperability problems (there's enough of those, especially with
old APs), but it would be nice to have an easy way for the user to
enable power save from NetworkManager.
--
Kalle Valo
prev parent reply other threads:[~2008-12-28 12:55 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-12-18 21:35 [PATCH v6 0/3] mac80211 dynamic power save Kalle Valo
2008-12-18 21:35 ` [PATCH v6 1/3] mac80211: enable IEEE80211_CONF_PS only when associated Kalle Valo
2008-12-18 21:35 ` [PATCH v6 2/3] mac80211: track master queue status Kalle Valo
2008-12-18 21:35 ` [PATCH v6 3/3] mac80211: implement dynamic power save Kalle Valo
2008-12-19 6:17 ` Vivek Natarajan
2008-12-19 6:21 ` Vivek Natarajan
2008-12-19 7:37 ` Kalle Valo
2008-12-19 7:45 ` Kalle Valo
2008-12-19 9:21 ` Ivo Van Doorn
2008-12-23 21:55 ` [PATCH v6 0/3] mac80211 " Johannes Berg
2008-12-28 12:54 ` Kalle Valo [this message]
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=871vvso3xy.fsf@nokia.com \
--to=kalle.valo@nokia.com \
--cc=dcbw@redhat.com \
--cc=johannes@sipsolutions.net \
--cc=linux-wireless@vger.kernel.org \
/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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).