Received-SPF: pass (google.com: domain of linux-wireless+bounces-5980-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:4601:e00::3 as permitted sender) client-ip=2604:1380:4601:e00::3;
Message-ID: <4987737e-df85-452c-85a6-9032fb620dd5@quicinc.com>
Date: Mon, 8 Apr 2024 12:42:33 -0700
Precedence: bulk
MIME-Version: 1.0
User-Agent: Mozilla Thunderbird
Subject: Re: [PATCH] wifi: cfg80211: Clear mlo_links info when STA disconnect
Content-Language: en-US
To: Johannes Berg <johannes@sipsolutions.net>,
        Xin Deng
	<quic_deng@quicinc.com>, <ath12k@lists.infradead.org>
CC: <linux-wireless@vger.kernel.org>
References: <20240327032448.15098-1-quic_deng@quicinc.com>
 <e599d1b1c7c6e32856eb5374cfaa20965b49b44c.camel@sipsolutions.net>
From: Jeff Johnson <quic_jjohnson@quicinc.com>
In-Reply-To: <e599d1b1c7c6e32856eb5374cfaa20965b49b44c.camel@sipsolutions.net>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: 7bit

On 4/8/2024 11:41 AM, Johannes Berg wrote:
> On Tue, 2024-03-26 at 20:24 -0700, Xin Deng wrote:
>> wdev->valid_links is not cleared when upper layer disconnect from a
>> wdev->AP MLD. It has been observed that this would prevent offchannel
>> operations like remain-on-channel which would be needed for user space
>> operations with Public Action frame.
> 
> I agree that's a problem, we shouldn't leave the valid_links set.
> 
>> Signed-off-by: Xin Deng <quic_deng@quicinc.com>
>> ---
>>  net/wireless/sme.c | 1 +
>>  1 file changed, 1 insertion(+)
>>
>> diff --git a/net/wireless/sme.c b/net/wireless/sme.c
>> index 82e3ce42206c..86e837f37f8c 100644
>> --- a/net/wireless/sme.c
>> +++ b/net/wireless/sme.c
>> @@ -492,6 +492,7 @@ void cfg80211_wdev_release_link_bsses(struct wireless_dev *wdev, u16 link_mask)
>>  				 &wdev->links[link].client.current_bss->pub);
>>  		wdev->links[link].client.current_bss = NULL;
>>  	}
>> +	wdev->valid_links = 0;
>>
> 
> But this is (very obviously, even with only the limited context shown!)
> in the completely wrong place.

Concur, so NAK my Reviewed-by:

That function is potentially only removing a subset of links, so it should not
be setting the valid_links to 0 since not all of the links may have been removed.

/jeff