linux-kernel - Re: [PATCH] usb: gadget: u_ether: Don't warn in gether_setup_name

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Date:   Wed, 11 Jan 2023 12:24:14 +0100
From:   Andrzej Pietrasiewicz <andrzej.p@...labora.com>
To:     Jon Hunter <jonathanh@...dia.com>,
        Greg Kroah-Hartman <gregkh@...uxfoundation.org>
Cc:     linux-usb@...r.kernel.org, linux-kernel@...r.kernel.org,
        linux-tegra@...r.kernel.org
Subject: Re: [PATCH] usb: gadget: u_ether: Don't warn in
 gether_setup_name_default()

Hi,

W dniu 11.01.2023 o 11:35, Jon Hunter pisze:
> 
> On 10/01/2023 16:31, Andrzej Pietrasiewicz wrote:
>> Hi,
>>
>> Hasn't there been a similar patch already?
> 
> There could be, but I was not aware. Do you happen to have a link to it?

Nope, I only rely on my memory, which can obviously be faulty.

> 
>> W dniu 6.01.2023 o 17:17, Jon Hunter pisze:
>>> The function gether_setup_name_default() is called by various USB
>>> ethernet gadget drivers. This function always generates the MAC address
>>> for the ethernet gadget device and always prints a warning when
>>> generating the MAC address. Given that these messages are expected, make
>>> these prints informational instead of warnings.
>>>
>>> Signed-off-by: Jon Hunter <jonathanh@...dia.com>
>>> ---
>>>   drivers/usb/gadget/function/u_ether.c | 4 ++--
>>>   1 file changed, 2 insertions(+), 2 deletions(-)
>>>
>>> diff --git a/drivers/usb/gadget/function/u_ether.c 
>>> b/drivers/usb/gadget/function/u_ether.c
>>> index 8f12f3f8f6ee..c19d66cd1446 100644
>>> --- a/drivers/usb/gadget/function/u_ether.c
>>> +++ b/drivers/usb/gadget/function/u_ether.c
>>> @@ -845,13 +845,13 @@ struct net_device *gether_setup_name_default(const char 
>>> *netname)
>>>       snprintf(net->name, sizeof(net->name), "%s%%d", netname);
>>>       eth_random_addr(dev->dev_mac);
>>> -    pr_warn("using random %s ethernet address\n", "self");
>>> +    pr_info("using random %s ethernet address\n", "self");
>>
>> As far as I can tell this function is called by all Ethernet gadgets,
>> and using random Ethernet addresses is the default behavior for all of them,
>> including legacy gadgets. Why to inform about the default situation happening?
>> So in fact maybe it would be better to eliminate the pr_warn() altogether,
>> instead of replacing it with pr_info()? If the user does not care to
>> explicitly set some non-default address(es), why would she care to know
>> that a randomly selected address has been chosen? Note that learning
>> _what_ specific address has been chosen is perfectly doable without
>> these pr_info() calls.
> 
> 
> That would be fine with me. This print has been there for a long time and so I 
> figured people wanted some sort of message. I would be happy to remove.

It has. Maybe it's been a mistake?

A properly working driver should be silent. In case of Ethernet gadgets
using random MAC addresses is a proper (and default) behavior.

I think it's worth trying to just remove pr_warn(). Or, at least,
change to pr_debug(), because these messages look more like
debugging messages no regular user should be interested in.

There's also the maintainer who can have his say.

@Greg?

As a side note, there's quite a lot of cleanup needed in gadgets' messaging.

Regards,

Andrzej

> 
> Jon
>