Page MenuHomeFreeBSD

net80211: add valid VHT MCS combinations and helper functions
ClosedPublic

Authored by adrian on Jan 23 2025, 2:38 AM.
Referenced Files
F135553882: D48608.id150165.diff
Mon, Nov 10, 6:41 PM
F135546030: D48608.id149900.diff
Mon, Nov 10, 5:09 PM
F135526642: D48608.id150004.diff
Mon, Nov 10, 1:32 PM
F135512044: D48608.id150150.diff
Mon, Nov 10, 10:54 AM
F135484373: D48608.id151524.diff
Mon, Nov 10, 5:51 AM
F135481861: D48608.id149848.diff
Mon, Nov 10, 5:26 AM
Unknown Object (File)
Thu, Nov 6, 2:28 AM
Unknown Object (File)
Wed, Nov 5, 3:51 PM

Details

Summary

Not all NSS / MCS / channel bandwidth combinations are valid.

This will be important for rate control and transmit rate
setup (eg static management, unicast, multicast rates.)

This is based on 802.11-2020 Section 21.5 (Parameters for VHT-MCSs.)

Diff Detail

Repository
rG FreeBSD src repository
Lint
Lint Skipped
Unit
Tests Skipped
Build Status
Buildable 62130
Build 59014: arc lint + arc unit

Event Timeline

I am fine with this either way. If you don't want to change this is accepted too.

sys/net80211/ieee80211_phy.c
680

Surprised you don't use a default here for future-proof-ness.

sys/net80211/ieee80211_phy.c
680

I want the compiler to complain, it's an enum, right? :)

bz added inline comments.
sys/net80211/ieee80211_phy.c
680

Do you ever expect more supported bw in VHT? Fine by me either way.

This revision is now accepted and ready to land.Feb 16 2025, 10:15 PM

Not for VHT, but who knows what we'll end up doing for EHT/HE, what we want to do for half/quarter rates, etc.

Right now ni_chw is the protocol width in the IE, not specifically whatever the PHY is configured as. All kinds of fun stuff may happen in the future!