Add support for drivers to indicate STA connection when user space SME
(e.g., hostapd) is not used for MLD AP. Add new parameters in struct
station_info to provide (re)association link ID, station's MLD address
and (Re)Association Response IEs in cfg80211_new_sta().
Signed-off-by: Veerendranath Jakkam <[email protected]>
---
include/net/cfg80211.h | 22 ++++++++++++++++++++++
net/wireless/nl80211.c | 16 ++++++++++++++++
2 files changed, 38 insertions(+)
diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h
index 03d4f4deadae..15619e0f5f0b 100644
--- a/include/net/cfg80211.h
+++ b/include/net/cfg80211.h
@@ -1882,6 +1882,22 @@ struct cfg80211_tid_stats {
* received packet with an FCS error matches the peer MAC address.
* @airtime_link_metric: mesh airtime link metric.
* @connected_to_as: true if mesh STA has a path to authentication server
+ * @mlo_params_valid: Indicates @assoc_link_id and @mld_addr fields are valid.
+ * Drivers use this only in cfg80211_new_sta() calls when AP MLD's MLME/SME
+ * is offload to driver. Drivers won't fill this information in
+ * cfg80211_del_sta_sinfo(), get_station() and dump_station() callbacks.
+ * @assoc_link_id: Indicates link ID of the AP MLD link on which the station
+ * association happened.
+ * @mld_addr: MLD address if the station is an MLD. Otherwise, set to all zeros.
+ * @assoc_resp_ies: IEs from (Re)Association Response.
+ * This is used only when in AP mode with drivers that do not use user
+ * space MLME/SME implementation. The information is provided only for the
+ * cfg80211_new_sta() calls to notify user space of the IEs. Drivers won't
+ * fill this information in cfg80211_del_sta_sinfo(), get_station() and
+ * dump_station() callbacks. Userspace needs this information to determine
+ * the other affiliated link stations information when MLD station is
+ * connected.
+ * @assoc_resp_ies_len: Length of @assoc_resp_ies buffer in octets.
*/
struct station_info {
u64 filled;
@@ -1941,6 +1957,12 @@ struct station_info {
u32 airtime_link_metric;
u8 connected_to_as;
+
+ bool mlo_params_valid;
+ u8 assoc_link_id;
+ u8 mld_addr[ETH_ALEN] __aligned(2);
+ const u8 *assoc_resp_ies;
+ size_t assoc_resp_ies_len;
};
/**
diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c
index 33a82ecab9d5..5a20bf8b85d6 100644
--- a/net/wireless/nl80211.c
+++ b/net/wireless/nl80211.c
@@ -6527,6 +6527,22 @@ static int nl80211_send_station(struct sk_buff *msg, u32 cmd, u32 portid,
sinfo->assoc_req_ies))
goto nla_put_failure;
+ if (sinfo->assoc_resp_ies_len &&
+ nla_put(msg, NL80211_ATTR_RESP_IE, sinfo->assoc_resp_ies_len,
+ sinfo->assoc_resp_ies))
+ goto nla_put_failure;
+
+ if (sinfo->mlo_params_valid) {
+ if (nla_put_u8(msg, NL80211_ATTR_MLO_LINK_ID,
+ sinfo->assoc_link_id))
+ goto nla_put_failure;
+
+ if (!is_zero_ether_addr(sinfo->mld_addr) &&
+ nla_put(msg, NL80211_ATTR_MLD_ADDR, ETH_ALEN,
+ sinfo->mld_addr))
+ goto nla_put_failure;
+ }
+
cfg80211_sinfo_release_content(sinfo);
genlmsg_end(msg, hdr);
return 0;
--
2.25.1
Hi,
> + * @mlo_params_valid: Indicates @assoc_link_id and @mld_addr fields are valid.
> + * Drivers use this only in cfg80211_new_sta() calls when AP MLD's MLME/SME
> + * is offload to driver. Drivers won't fill this information in
> + * cfg80211_del_sta_sinfo(), get_station() and dump_station() callbacks.
>
> + * @mld_addr: MLD address if the station is an MLD. Otherwise, set to all zeros.
> + if (sinfo->mlo_params_valid) {
> + if (nla_put_u8(msg, NL80211_ATTR_MLO_LINK_ID,
> + sinfo->assoc_link_id))
> + goto nla_put_failure;
> +
> + if (!is_zero_ether_addr(sinfo->mld_addr) &&
> + nla_put(msg, NL80211_ATTR_MLD_ADDR, ETH_ALEN,
> + sinfo->mld_addr))
>
It'd be invalid, but now you could have mlo_params_valid == true &&
is_zero_ether_addr(sinfo->mld_addr), which would lead to a very strange
situation for userspace, it would see a link ID but no MLD address.
With the documented requirement that
mlo_params_valid == (mld_addr is valid)
wouldn't it make sense to just remove mlo_params_valid, and lift the
is_zero_ether_addr() check to the outer condition?
johannes
Thanks Johannes for reviewing the patches.
On 1/18/2023 9:11 PM, Johannes Berg wrote:
> Hi,
>
>> + * @mlo_params_valid: Indicates @assoc_link_id and @mld_addr fields are valid.
>> + * Drivers use this only in cfg80211_new_sta() calls when AP MLD's MLME/SME
>> + * is offload to driver. Drivers won't fill this information in
>> + * cfg80211_del_sta_sinfo(), get_station() and dump_station() callbacks.
>>
>> + * @mld_addr: MLD address if the station is an MLD. Otherwise, set to all zeros.
>
>> + if (sinfo->mlo_params_valid) {
>> + if (nla_put_u8(msg, NL80211_ATTR_MLO_LINK_ID,
>> + sinfo->assoc_link_id))
>> + goto nla_put_failure;
>> +
>> + if (!is_zero_ether_addr(sinfo->mld_addr) &&
>> + nla_put(msg, NL80211_ATTR_MLD_ADDR, ETH_ALEN,
>> + sinfo->mld_addr))
>>
>
> It'd be invalid, but now you could have mlo_params_valid == true &&
> is_zero_ether_addr(sinfo->mld_addr), which would lead to a very strange
> situation for userspace, it would see a link ID but no MLD address.
Only link ID and no MLD address also a valid combination. It indicates
the connected STA is non-MLD. User space needs the link ID information
to determine "STA connected to which affiliated AP of the AP MLD".
>
> With the documented requirement that
> mlo_params_valid == (mld_addr is valid)
No, I think there is some misinterpretation of the documentation here.
Let me submit new patch-set after adding more detailed text in the
documentation.
Actually what I mean, cfg80211 can consider parsing both the
"assoc_link_id" and "mld_addr" fields when "mlo_params_valid" is true.
"assoc_link_id" will be set to link ID of the affiliated AP on which STA
(MLD/non-MLD) completed association. "mld_addr" will be set to MLD
address of the STA if applicable (i.e., MLD STA) . Otherwise, will be
set to zeros (i.e., non-MLD STA).
--
veeru