netdev - Re: [PATCH] net: lan743x: LAN743X selects FIXED

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

Date:   Mon, 27 Feb 2023 09:37:07 +0100
From:   "Arnd Bergmann" <arnd@...db.de>
To:     "Tom Rix" <trix@...hat.com>,
        "Simon Horman" <simon.horman@...igine.com>,
        "Andrew Lunn" <andrew@...n.ch>
Cc:     "David S . Miller" <davem@...emloft.net>,
        "Eric Dumazet" <edumazet@...gle.com>,
        "Jakub Kicinski" <kuba@...nel.org>,
        "Paolo Abeni" <pabeni@...hat.com>, steen.hegelund@...rochip.com,
        Netdev <netdev@...r.kernel.org>, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] net: lan743x: LAN743X selects FIXED_PHY to resolve a link error

On Sun, Feb 26, 2023, at 16:15, Tom Rix wrote:
> On 2/21/23 8:20 AM, Simon Horman wrote:
>> On Tue, Feb 21, 2023 at 03:29:39AM +0100, Andrew Lunn wrote:
>>> On Mon, Feb 20, 2023 at 07:46:13AM +0100, Simon Horman wrote:
>>>>
>>>> LAN743X=y and FIXED_PHY=m does indeed produce the problem that Tom
>>>> describes. And his patch does appear to resolve the problem.
>>> O.K. So the commit message needs updating to describe the actual
>>> problem.
>> Yes, that would be a good improvement.
>>
>> Perhaps a fixes tag too?
>>
>>>> Unfortunately your proposed solution seems to run foul of a complex
>>>> dependency situation.
>>> I was never any good at Kconfig. Arnd is the expert at solving
>>> problems like this.
>>>
>>> You want either everything built in, or FIXED_PHY built in and LAN743X
>>> modular, or both modular.
>> I _think_ the patch, which uses select FIXED_PHY for LAN743X,
>> achieves that.
>>
>> I CCed Arnd in case he has any input. Though I think I read
>> in an recent email from him that he is out most of this week.

FWIW, the original patch looks good to me,

Reviewed-by: Arnd Bergmann <arnd@...db.de>

I'm not sure if the #else path in include/linux/phy_fixed.h
is actually helpful, as it does not avoid a link failure but
just makes it less common. Maybe we should just drop that
from the header and require all users of fixed_phy to
'select' that symbol even when they are built-in?

       Arnd