Sender: Larry Finger <larry.finger@gmail.com>
Message-ID: <dff29d82-9c4b-1933-c1c1-a3becf2a0f1f@lwfinger.net>
Date:   Fri, 24 Feb 2023 09:45:14 -0600
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101
 Thunderbird/102.8.0
Subject: Re: [PATCH] wifi: wext: warn about usage only once
Content-Language: en-US
To:     Johannes Berg <johannes@sipsolutions.net>,
        linux-wireless@vger.kernel.org
Cc:     Linus Torvalds <torvalds@linux-foundation.org>,
        netdev@vger.kernel.org,
        Nicolas Cavallari <Nicolas.Cavallari@green-communications.fr>,
        Johannes Berg <johannes.berg@intel.com>
References: <20230224135933.94104aeda1a0.Ie771c6a66d7d6c3cf67da5f3b0c66cea66fd514c@changeid>
From:   Larry Finger <Larry.Finger@lwfinger.net>
In-Reply-To: <20230224135933.94104aeda1a0.Ie771c6a66d7d6c3cf67da5f3b0c66cea66fd514c@changeid>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit
Precedence: bulk

On 2/24/23 06:59, Johannes Berg wrote:
> From: Johannes Berg <johannes.berg@intel.com>
> 
> Warn only once since the ratelimit parameters are still
> allowing too many messages to happen. This will no longer
> tell you all the different processes, but still gives a
> heads-up of sorts.
> 
> Also modify the message to note that wext stops working
> for future Wi-Fi 7 hardware, this is already implemented
> in commit 4ca69027691a ("wifi: wireless: deny wireless
> extensions on MLO-capable devices") and is maybe of more
> relevance to users than the fact that we'd like to have
> wireless extensions deprecated.
> 
> The issue with Wi-Fi 7 is that you can now have multiple
> connections to the same AP, so a whole bunch of things
> now become per link rather than per netdev, which can't
> really be handled in wireless extensions.
> 
> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
> ---
> Not really sure I see a better solution ...
> 
>   - tracking it per task would be nice in a way I guess,
>     but is also awful;
>   - adjusting the rate limit will lead us into an endless
>     bikeshedding discussion about the parameters;
>   - removing the warning will leave us with no indiciation
>     of what happens with Wi-Fi 7 hardware, although most of
>     the processes using them now (like Chrome browser??)
>     probably ignore failures from it
>   - trying to support a 30+ year old technology on modern
>     Wi-Fi 7 hardware will be "interesting" and lead to all
>     kinds of hacks there
> ---
>   net/wireless/wext-core.c | 4 ++--
>   1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/net/wireless/wext-core.c b/net/wireless/wext-core.c
> index 13a72b17248e..a125fd1fa134 100644
> --- a/net/wireless/wext-core.c
> +++ b/net/wireless/wext-core.c
> @@ -641,8 +641,8 @@ static void wireless_warn_cfg80211_wext(void)
>   {
>   	char name[sizeof(current->comm)];
>   
> -	pr_warn_ratelimited("warning: `%s' uses wireless extensions that are deprecated for modern drivers; use nl80211\n",
> -			    get_task_comm(name, current));
> +	pr_warn_once("warning: `%s' uses wireless extensions which will stop working for Wi-Fi 7 hardware; use nl80211\n",
> +		     get_task_comm(name, current));
>   }
>   #endif
>   

Johannes,

Although this patch will stop the log spamming that I see, it will not provide 
much information upstream for fixing the problems.

Even if this patch is applied to the kernel, I plan to keep my local, once per 
task, patch so that I can keep upstream informed, and perhaps get fixes applied 
there.

Larry