kernel.git/net/mac80211/sta_info.h, branch linux-rolling-stable

wifi: mac80211: add support to accumulate removed link statistics

2025-06-24T13:19:26Z

Currently, if a link gets removed in between for a station then directly accumulated data will fall down to sum of other active links. This will bring inconsistency in station dump statistics. For instance, let's take Tx packets - at t=0-> link-0:2 link-1:3 Tx packets => accumulated = 5 - at t=1-> link-0:4 link-1:6 Tx packets => accumulated = 10 let say at t=2, link-0 went down => link-0:0 link-1:7 => accumulated = 7 Here, suddenly accumulated Tx packets will come down to 7 from 10. This is showing inconsistency. Therefore, store link-0 data when it went down and add to accumulated Tx packet = 11. Hence, store the removed link statistics data in sta structure and add it in accumulated statistics for consistency. Signed-off-by: Sarika Sharma Link: https://patch.msgid.link/20250528054420.3050133-7-quic_sarishar@quicinc.com Signed-off-by: Johannes Berg

wifi: mac80211: add support towards MLO handling of station statistics

2025-06-24T13:19:26Z

Currently, in supporting API's to fill sinfo structure from sta structure, is mapped to fill the fields from sta->deflink. However, for multi-link (ML) station, sinfo structure should be filled from corresponding link_id. Therefore, add link_id as an additional argument in supporting API's for filling sinfo structure correctly. Link_id is set to -1 for non-ML station and corresponding link_id for ML stations. In supporting API's for filling sinfo structure, check for link_id, if link_id < 0, fill the sinfo structure from sta->deflink, otherwise fill from sta->link[link_id]. Current, changes are done at the deflink level i.e, pass -1 as link_id. Actual link_id will be added in subsequent patches to support station statistics for MLO. Signed-off-by: Sarika Sharma Link: https://patch.msgid.link/20250528054420.3050133-2-quic_sarishar@quicinc.com Signed-off-by: Johannes Berg

Revert "mac80211: Dynamically set CoDel parameters per station"

2025-04-23T13:22:11Z

This reverts commit 484a54c2e597dbc4ace79c1687022282905afba0. The CoDel parameter change essentially disables CoDel on slow stations, with some questionable assumptions, as Dave pointed out in [0]. Quoting from there: But here are my pithy comments as to why this part of mac80211 is so wrong... static void sta_update_codel_params(struct sta_info *sta, u32 thr) { - if (thr && thr < STA_SLOW_THRESHOLD * sta->local->num_sta) { 1) sta->local->num_sta is the number of associated, rather than active, stations. "Active" stations in the last 50ms or so, might have been a better thing to use, but as most people have far more than that associated, we end up with really lousy codel parameters, all the time. Mistake numero uno! 2) The STA_SLOW_THRESHOLD was completely arbitrary in 2016. - sta->cparams.target = MS2TIME(50); This, by itself, was probably not too bad. 30ms might have been better, at the time, when we were battling powersave etc, but 20ms was enough, really, to cover most scenarios, even where we had low rate 2Ghz multicast to cope with. Even then, codel has a hard time finding any sane drop rate at all, with a target this high. - sta->cparams.interval = MS2TIME(300); But this was horrible, a total mistake, that is leading to codel being completely ineffective in almost any scenario on clients or APS. 100ms, even 80ms, here, would be vastly better than this insanity. I'm seeing 5+seconds of delay accumulated in a bunch of otherwise happily fq-ing APs.... 100ms of observed jitter during a flow is enough. Certainly (in 2016) there were interactions with powersave that I did not understand, and still don't, but if you are transmitting in the first place, powersave shouldn't be a problemmmm..... - sta->cparams.ecn = false; At the time we were pretty nervous about ecn, I'm kind of sanguine about it now, and reliably indicating ecn seems better than turning it off for any reason. [...] In production, on p2p wireless, I've had 8ms and 80ms for target and interval for years now, and it works great. I think Dave's arguments above are basically sound on the face of it, and various experimentation with tighter CoDel parameters in the OpenWrt community have show promising results[1]. So I don't think there's any reason to keep this parameter fiddling; hence this revert. [0] https://lore.kernel.org/linux-wireless/CAA93jw6NJ2cmLmMauz0xAgC2MGbBq6n0ZiZzAdkK0u4b+O2yXg@mail.gmail.com/ [1] https://forum.openwrt.org/t/reducing-multiplexing-latencies-still-further-in-wifi/133605/130 Suggested-By: Dave Taht In-memory-of: Dave Taht Signed-off-by: Toke Høiland-Jørgensen Link: https://patch.msgid.link/20250403183930.197716-1-toke@toke.dk Signed-off-by: Johannes Berg

wifi: mac80211: add some support for RX OMI power saving

2025-01-13T14:26:43Z

In order to save power, it can be desirable to change the RX operating mode using OMI to reduce the bandwidth. As the handshake must be done in the HTC+ field, it cannot be done by mac80211 directly, so expose functions to the driver to request and finalize the necessary updates. Note that RX OMI really only changes what the peer (AP) will transmit to us, but in order to use it to actually save some power (by reducing the listen bandwidth) we also update rate scaling and then the channel context's mindef accordingly. The updates are split into two in order to sequence them correctly, when reducing bandwidth first reduce the rate scaling and thus TX, then send OMI, then reduce the listen bandwidth (chandef); when increasing bandwidth this is the other way around. This also requires tracking in different variables which part is applicable already. Signed-off-by: Johannes Berg Signed-off-by: Miri Korenblit Link: https://patch.msgid.link/20250101070249.2c1a1934bd73.I4e90fd503504e37f9eac5bdae62e3f07e7071275@changeid Signed-off-by: Johannes Berg

wifi: mac80211: add an option to filter a sta from being flushed

2025-01-13T12:53:04Z

Sometimes we might want to flush only part of the stations of a vif, for example only the TDLS ones. To allow this, add a do_not_flush_sta argument to __sta_info_flush, which in turn, will not flush this station. Signed-off-by: Miri Korenblit Link: https://patch.msgid.link/20241224192322.535e1fcca192.Icecf7f443bf98c9535ce8ec03b24d0d17dfbc28e@changeid Signed-off-by: Johannes Berg

wifi: mac80211, cfg80211: miscellaneous spelling fixes

2024-10-08T19:52:09Z

Correct spelling here and there as suggested by codespell. Signed-off-by: Dmitry Antipov Link: https://patch.msgid.link/20240913084919.118862-1-dmantipov@yandex.ru Signed-off-by: Johannes Berg

wifi: mac80211: add ieee80211_tdls_sta_link_id()

2024-06-26T08:22:04Z

We've open-coded this twice and will need it again, add ieee80211_tdls_sta_link_id() to get the one link ID for a TDLS STA. Reviewed-by: Miriam Rachel Korenblit Link: https://patch.msgid.link/20240612143707.9f8141ae1725.I343822bbba0ae08dedb2f54a0ce87f2ae5ebeb2b@changeid Signed-off-by: Johannes Berg

wifi: mac80211: add return docs for sta_info_flush()

2024-04-19T08:27:21Z

Use the Return: annotation instead of spelling out "Returns" in the documentation, for both sta_info_flush()/__sta_info_flush(). Signed-off-by: Johannes Berg

wifi: mac80211: track capability/opmode NSS separately

2024-03-04T13:25:27Z

We're currently tracking rx_nss for each station, and that is meant to be initialized to the capability NSS and later reduced by the operating mode notification NSS. However, we're mixing up capabilities and operating mode NSS in the same variable. This forces us to recalculate the NSS capability on operating mode notification RX, which is a bit strange; due to the previous fix I had to never keep rx_nss as zero, it also means that the capa is never taken into account properly. Fix all this by storing the capability value, that can be recalculated unconditionally whenever needed, and storing the operating mode notification NSS separately, taking it into account when assigning the final rx_nss value. Cc: stable@vger.kernel.org Fixes: dd6c064cfc3f ("wifi: mac80211: set station RX-NSS on reconfig") Reviewed-by: Miriam Rachel Korenblit Link: https://msgid.link/20240228120157.0e1c41924d1d.I0acaa234e0267227b7e3ef81a59117c8792116bc@changeid Signed-off-by: Johannes Berg

wifi: mac80211: flush only stations using requests links

2024-02-12T20:13:57Z

Whenever sta_flush() function is invoked, all STAs present in that interface are flushed. In case of MLO, it is desirable to only flush such STAs that are at least using a given link id as one of their links. Add support for this by making change in the __sta_info_flush API argument to accept a link ID. And then, only if the STA is using the given link as one of its links, it would be flushed. Signed-off-by: Aditya Kumar Singh Link: https://msgid.link/20240205162952.1697646-3-quic_adisi@quicinc.com [reword commit message, in particular this isn't about "active" links] Signed-off-by: Johannes Berg