Hi Jeff, Johannes, >From your points we understand, we need a length variable and the td_bitmap value. As per the spec https://www.wi-fi.org/download.php?file=/sites/default/files/private/WPA3_Specification_v3.0.pdf( Table 4. Transition Disable KDE format), we do have variable length of the bitmap. So we could 1. Add 2 arguments 2. One would be for the length of td_bitmap. 3. The second argument would be an u8 bitmap array depending on length of bitmap. Accordingly checks can be added to indicate/ignore the indication of bitmap to upper layer. Also the driver can update these fields as length 0 and array NULL in case no value to be updated by the driver. On Mon, Aug 29, 2022 at 3:22 PM Johannes Berg <johannes@xxxxxxxxxxxxxxxx> wrote: > > On Fri, 2022-08-26 at 08:48 -0700, Jeff Johnson wrote: > > > > > @@ -3268,6 +3270,7 @@ enum nl80211_attrs { > > > > > > NL80211_ATTR_TX_HW_TIMESTAMP, > > > NL80211_ATTR_RX_HW_TIMESTAMP, > > > + NL80211_ATTR_TD_BITMAP, > > > > > > /* add attributes here, update the policy in nl80211.c */ > > > > Johannes, do you want the policy updated even thought this is > > driver->userspace and hence the policy is never applied to it? > > Yeah in a sense, it doesn't really matter... I think not updating is > fine, then it will likely even be rejected, at least in any new > commands. > > > > struct { > > > u8 bssid[ETH_ALEN]; > > > + s16 td_bitmap; > > > > I know you are using -1 as an indication that the bitmap is not used, > > but using signed with a bitmap seems strange since bitops can be > > affected by sign extension. Just something that set off my Spider-Sense. > > Yeah true ... maybe a separate validity bool would've been better? > dunno. > > johannes >
Attachment:
smime.p7s
Description: S/MIME Cryptographic Signature
