Re: [PATCH -next v3 00/13] wifi: cfg80211: consolidate cookie assignment for async ops
Johannes Berg <[email protected]> Sun, 02 Aug 2026 20:07:15 +0200
| Newsgroups | dev.linux.lists.brcm80211,org.kernel.vger.linux-wireless |
|---|---|
| Message-ID | <[email protected]> |
On Fri, 2026-07-31 at 14:34 +0200, Arend van Spriel wrote: > NL80211_ATTR_COOKIE is a userspace-visible u64 that correlates an async > nl80211 operation with its completion event. Three cfg80211 ops currently > delegate cookie generation to the driver: remain_on_channel, mgmt_tx, and > probe_peer. This produces inconsistent strategies across drivers: some us= e > an incrementing counter, some use get_random_u32(), some use a CID cast t= o > u64, and brcmfmac's mgmt_tx always assigned zero - which is arguably brok= en > since cfg80211 treats zero as an invalid cookie. Thanks for cleaning this up, I've applied it. Incidentally, this also simplified some changes that I'm slowly prototyping in this area - I want to queue ethernet frames on the TXQs, initially at least for client mode, in order to build A-MSDUs at TXQ dequeue rather than enqueue, and to be able to more easily move frames from one place to another during SMD transition. I guess it'll be a while before I'll post _all_ of that though, might post some prep patches soon with some "further thoughts" in the cover letter. johannes