diff options
author | Johannes Berg <johannes@sipsolutions.net> | 2009-04-16 17:04:25 +0200 |
---|---|---|
committer | John W. Linville <linville@tuxdriver.com> | 2009-04-22 16:57:16 -0400 |
commit | 955394c98c9cb79bdb3e6b479695af3a90ea0623 (patch) | |
tree | db4e8341606ecd2683ce1649ca3f0dcfd656c675 /include | |
parent | bbbdff9e00449928f228867076a07bdfecd3dca8 (diff) |
mac80211: document powersaving/beacon filter future
Document what mac80211 will do in the future to help save power.
We're not quite there yet, but a plan helps. Also, while at it,
fix the docs wrt. multicast traffic.
Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Reviewed-by: Kalle Valo <kalle.valo@iki.fi>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Diffstat (limited to 'include')
-rw-r--r-- | include/net/mac80211.h | 60 |
1 files changed, 50 insertions, 10 deletions
diff --git a/include/net/mac80211.h b/include/net/mac80211.h index a593bedcfed..52808bdcc6c 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -1109,11 +1109,9 @@ ieee80211_get_alt_retry_rate(const struct ieee80211_hw *hw, * need software support for parsing the TIM bitmap. This is also supported * by mac80211 by combining the %IEEE80211_HW_SUPPORTS_PS and * %IEEE80211_HW_PS_NULLFUNC_STACK flags. The hardware is of course still - * required to pass up beacons. Additionally, in this case, mac80211 will - * wake up the hardware when multicast traffic is announced in the beacon. - * - * FIXME: I don't think we can be fast enough in software when we want to - * receive multicast traffic? + * required to pass up beacons. The hardware is still required to handle + * waking up for multicast traffic; if it cannot the driver must handle that + * as best as it can, mac80211 is too slow. * * Dynamic powersave mode is an extension to normal powersave mode in which * the hardware stays awake for a user-specified period of time after sending @@ -1134,11 +1132,53 @@ ieee80211_get_alt_retry_rate(const struct ieee80211_hw *hw, * way the host will only receive beacons where some relevant information * (for example ERP protection or WMM settings) have changed. * - * Beacon filter support is informed with %IEEE80211_HW_BEACON_FILTER flag. - * The driver needs to enable beacon filter support whenever power save is - * enabled, that is %IEEE80211_CONF_PS is set. When power save is enabled, - * the stack will not check for beacon miss at all and the driver needs to - * notify about complete loss of beacons with ieee80211_beacon_loss(). + * Beacon filter support is advertised with the %IEEE80211_HW_BEACON_FILTER + * hardware capability. The driver needs to enable beacon filter support + * whenever power save is enabled, that is %IEEE80211_CONF_PS is set. When + * power save is enabled, the stack will not check for beacon loss and the + * driver needs to notify about loss of beacons with ieee80211_beacon_loss(). + * + * The time (or number of beacons missed) until the firmware notifies the + * driver of a beacon loss event (which in turn causes the driver to call + * ieee80211_beacon_loss()) should be configurable and will be controlled + * by mac80211 and the roaming algorithm in the future. + * + * Since there may be constantly changing information elements that nothing + * in the software stack cares about, we will, in the future, have mac80211 + * tell the driver which information elements are interesting in the sense + * that we want to see changes in them. This will include + * - a list of information element IDs + * - a list of OUIs for the vendor information element + * + * Ideally, the hardware would filter out any beacons without changes in the + * requested elements, but if it cannot support that it may, at the expense + * of some efficiency, filter out only a subset. For example, if the device + * doesn't support checking for OUIs it should pass up all changes in all + * vendor information elements. + * + * Note that change, for the sake of simplification, also includes information + * elements appearing or disappearing from the beacon. + * + * Some hardware supports an "ignore list" instead, just make sure nothing + * that was requested is on the ignore list, and include commonly changing + * information element IDs in the ignore list, for example 11 (BSS load) and + * the various vendor-assigned IEs with unknown contents (128, 129, 133-136, + * 149, 150, 155, 156, 173, 176, 178, 179, 219); for forward compatibility + * it could also include some currently unused IDs. + * + * + * In addition to these capabilities, hardware should support notifying the + * host of changes in the beacon RSSI. This is relevant to implement roaming + * when no traffic is flowing (when traffic is flowing we see the RSSI of + * the received data packets). This can consist in notifying the host when + * the RSSI changes significantly or when it drops below or rises above + * configurable thresholds. In the future these thresholds will also be + * configured by mac80211 (which gets them from userspace) to implement + * them as the roaming algorithm requires. + * + * If the hardware cannot implement this, the driver should ask it to + * periodically pass beacon frames to the host so that software can do the + * signal strength threshold checking. */ /** |