On 9/11/2024 6:22 PM, Johannes Berg wrote:
On Wed, 2024-09-11 at 18:21 +0530, Karthikeyan Periyasamy wrote:
On 9/11/2024 3:12 PM, Johannes Berg wrote:
On Wed, 2024-09-04 at 16:59 +0530, Karthikeyan Periyasamy wrote:
Currently, wiphy_verify_combinations() fails for the multi-radio per wiphy
due to the condition check on global interface combination that DFS only
works on one channel.
As it should, really.
In a multi-radio scenario, global interface
combination encompasses the capabilities of all radio combinations, so it
supports more than one channel with DFS.
No, that's not correct, it doesn't.
But the attribute comment section clearly says the Global iface
combination encompass all radio combination as below
* @NL80211_ATTR_WIPHY_INTERFACE_COMBINATIONS: Nested attribute listing *
the
* supported interface combinations for all radios combined. In each
* nested item, it contains attributes defined in
* &enum nl80211_if_combination_attrs.
Is my understanding incorrect ?
I guess it depends on how you interpret "combined". It must be something
that can actually be done *regardless* of radio assignment, to be
compatible with older userspace.
So if you think "combined" == "superset of all radios" then your
understanding is incorrect. You need to think "combined" == "what the
device can do without caring about radio assignment".
The current implementation of radio specific advertisement global iface
combination (NL80211_ATTR_WIPHY_INTERFACE_COMBINATIONS) expects the
superset of all radios, wherever the radio idx is -1 from caller of
cfg80211_iter_combinations().
Is it incorrect ?
--
Karthikeyan Periyasamy
--
கார்த்திகேயன் பெரியசாமி
