[PATCH v3 7/8] ap: allow configuring UHR DBE

[Raja Mani]

On 4/28/2026 1:13 PM, Johannes Berg wrote:
> On Mon, 2026-04-27 at 17:03 +0530, Raja Mani wrote:
>>
>>> --- a/hostapd/config_file.c
>>> +++ b/hostapd/config_file.c
>>> @@ -4997,6 +4997,11 @@ static int hostapd_config_fill(struct hostapd_config *conf,
>>>    		conf->ieee80211bn = atoi(pos);
>>>    	} else if (os_strcmp(buf, "require_uhr") == 0) {
>>>    		conf->require_uhr = atoi(pos);
>>> +	} else if (os_strcmp(buf, "dbe_bandwidth") == 0) {
>>> +		conf->dbe_bandwidth = atoi(pos);
>>
>> Since this is copied directly into the UHR operation element, would it
>> be worth adding a maximum limit check here to prevent accidental
>> modification of other bits?
> 
> Aside, you this "question" format makes you sound like an LLM... You're

:-)

> actually free to have an opinion. Personally, I find it really annoying
> to just be asked questions as a review.

Next time i shoot straightaway. :-)

> 
> I don't see most of the bandwidth settings doing that, so dunno, is it
> worth doing? I'm sure there are worse ways to shoot yourself in the foot
> with hostapd configuration.

Agree, few cases missed out such validations. thought nice to have that
for newer additions at least. don't have strong push.

> 
> It's also used internally and translated to a real chandef, which will
> almost certainly utterly fail if this value is wrong.
> 
> johannes