netdev - Re: [PATCH net 1/3] net: hns3: fix loopback test of serdes and phy is failed if duplex is half

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]

Message-ID: <1e1779f0-def2-4143-8a8c-a11ced956c2a@huawei.com>
Date: Fri, 19 Sep 2025 14:36:58 +0800
From: Jijie Shao <shaojijie@...wei.com>
To: Andrew Lunn <andrew@...n.ch>
CC: <shaojijie@...wei.com>, <davem@...emloft.net>, <edumazet@...gle.com>,
	<kuba@...nel.org>, <pabeni@...hat.com>, <andrew+netdev@...n.ch>,
	<horms@...nel.org>, <shenjian15@...wei.com>, <liuyonglong@...wei.com>,
	<chenhao418@...wei.com>, <lantao5@...wei.com>,
	<huangdonghua3@...artners.com>, <yangshuaisong@...artners.com>,
	<huangdengdui@...artners.com>, <jonathan.cameron@...wei.com>,
	<salil.mehta@...wei.com>, <netdev@...r.kernel.org>,
	<linux-kernel@...r.kernel.org>
Subject: Re: [PATCH net 1/3] net: hns3: fix loopback test of serdes and phy is
 failed if duplex is half


on 2025/9/18 0:58, Andrew Lunn wrote:
> On Wed, Sep 17, 2025 at 08:29:52PM +0800, Jijie Shao wrote:
>> If duplex setting is half, mac and phy can not transmit and receive data
>> at the same time.
> Lets think about the fundamentals of Ethernet, MII and half duplex.
>
> Is this specific to your MAC/PHY combination, or just generally true?
>
> Should this is solved in phylib, because it is true for every MAC/PHY
> combination? phylib returning -EINAL to phy_loopback() would seem like
> the correct thing to do.
>
> Is it specific to your PHY, but independent of the MAC?  Then the PHY
> should return -EINVAL in its set_loopback() method.

My idea is to prevent customers from seeing command failures,
so I switched from half-duplex to full-duplex for testing.
But it doesn't seem quite appropriate.


>
>> +	hdev->hw.mac.duplex_last = phydev->duplex;
>> +
>> +	ret = phy_set_bits(phydev, MII_BMCR, BMCR_FULLDPLX);
> A MAC driver should not be doing this. What if the PHY is C45 only?
> And Marvell PHYs need a soft reset before such an operation take
> effect.
>
>      Andrew

Thanks for the reminder.

Thanks，
Jijie Shao