Differential D50929

iwx: correct rssi calculations and reporting
AcceptedPublic
Actions

Authored by bz on Jun 19 2025, 1:20 AM.

Details

Reviewers

thj

Group Reviewers

wireless

Summary

Overall for the moment when we have to hard code a noise floor use -127.

iwx_rxmq_get_signal_strength(): return a signed value.
iwx_get_noise(): return an nunsigned value and correct maths; seems
c&p from iwm(4) which has this code compiled out in favor of a static
-96 nf.
in iwx_rx_mpdu_mq(): remove the "signal level normilization"; make the
local rssi an int8_t which is enough; if we have a signal normalize it
to the net80211 internal representation; if we have no value report the
noise floor which probably is better than conditionally reprting the
ressi and conditionally setting IEEE80211_R_RSSI.
Make sc_noise an int8_t as it is enough and initialize it during attach
as it will take a few beacons until we get a IWX_STATISTICS_NOTIFICATION
and have any value to use instead. This will get the reports of the
first round of scanning slightly more sensible.

Diff Detail

Repository

rG FreeBSD src repository

Lint

Lint Passed

Unit

No Test Coverage

Build Status

Buildable 64971
Build 61854: arc lint + arc unit

Event Timeline

bz created this revision.Jun 19 2025, 1:20 AM

Herald added a subscriber: imp. · View Herald TranscriptJun 19 2025, 1:20 AM

bz requested review of this revision.Jun 19 2025, 1:20 AM

Harbormaster completed remote builds in B64971: Diff 157293.Jun 19 2025, 1:20 AM

Note: this can go in w/o D50928 if deemed correct.

thj accepted this revision.Jun 19 2025, 11:34 AM

This revision is now accepted and ready to land.Jun 19 2025, 11:34 AM

bz mentioned this in D53780: iwx: clean up / document noise floor and RSSI fetching.Sun, Nov 16, 12:50 PM

hm, i wonder why iwx used -107 as its "nothing present" value. I wonder what iwlwifi is doing too.

like, -107dBm is a very low thermal noise floor; that's like for a 5MHz wide channel. (The noise floor gets higher as the channel width increases.)
But it's a damn sight better than -127dBm. :-)

also in my poking I found that I'm not seeing any empty airtime RSSI (it's always 0) so it was setting sc->sc_noise to -127. so we still need to clamp that too to a default noise floor.

Also, a lot of net80211 code currently expects nf and rssi to be passed in as parameters into functions, and that deserves to be cleaned up.

In D50929#1228064, @adrian wrote:

hm, i wonder why iwx used -107 as its "nothing present" value. I wonder what iwlwifi is doing too.

like, -107dBm is a very low thermal noise floor; that's like for a 5MHz wide channel. (The noise floor gets higher as the channel width increases.)
But it's a damn sight better than -127dBm. :-)

There is none. You get a signal value. All the values we ever used in iwm/... are arbitrary. I think normally we used to us -96.

In D50929#1228066, @bz wrote:

In D50929#1228064, @adrian wrote:

hm, i wonder why iwx used -107 as its "nothing present" value. I wonder what iwlwifi is doing too.

like, -107dBm is a very low thermal noise floor; that's like for a 5MHz wide channel. (The noise floor gets higher as the channel width increases.)
But it's a damn sight better than -127dBm. :-)

There is none. You get a signal value. All the values we ever used in iwm/... are arbitrary. I think normally we used to us -96.

-96 comes from ath(4), like a lot of net80211 and wifi history :-)

In D50929#1228065, @adrian wrote:

also in my poking I found that I'm not seeing any empty airtime RSSI (it's always 0) so it was setting sc->sc_noise to -127. so we still need to clamp that too to a default noise floor.

It's been a long time since I did this for iwx to compare to LinuxKPI; see the other open stack change which needs more cleanup and dealing as I want to break the ioctls only exactly once and the way this is done now won't help. But I believe once you are assoc you will see them.