linux-kernel - Re: [PATCH 1/3] net: stmmac: dwmac-rk: Use DELAY_ENABLE macro for RK3328

Open Source and information security mailing list archives

Message-ID: <624f2474-9a39-46a3-a6e5-f9966471bf3d@kwiboo.se>
Date: Fri, 7 Mar 2025 00:28:42 +0100
From: Jonas Karlman <jonas@...boo.se>
To: Andrew Lunn <andrew@...n.ch>
Cc: Heiko Stuebner <heiko@...ech.de>, Andrew Lunn <andrew+netdev@...n.ch>,
 "David S . Miller" <davem@...emloft.net>, Eric Dumazet
 <edumazet@...gle.com>, Jakub Kicinski <kuba@...nel.org>,
 Paolo Abeni <pabeni@...hat.com>, Maxime Coquelin
 <mcoquelin.stm32@...il.com>, Alexandre Torgue
 <alexandre.torgue@...s.st.com>, Wadim Egorov <w.egorov@...tec.de>,
 netdev@...r.kernel.org, linux-rockchip@...ts.infradead.org,
 linux-arm-kernel@...ts.infradead.org, linux-kernel@...r.kernel.org,
 linux-stm32@...md-mailman.stormreply.com
Subject: Re: [PATCH 1/3] net: stmmac: dwmac-rk: Use DELAY_ENABLE macro for
 RK3328

Hi Andrew,

On 2025-03-06 23:25, Andrew Lunn wrote:
> On Thu, Mar 06, 2025 at 08:38:52PM +0000, Jonas Karlman wrote:
>> Support for Rockchip RK3328 GMAC and addition of the DELAY_ENABLE macro
>> was merged in the same merge window. This resulted in RK3328 not being
>> converted to use the new DELAY_ENABLE macro.
>>
>> Change to use the DELAY_ENABLE macro to help disable MAC delay when
>> RGMII_ID/RXID/TXID is used.
>>
>> Fixes: eaf70ad14cbb ("net: stmmac: dwmac-rk: Add handling for RGMII_ID/RXID/TXID")
> 
> Please add a description of the broken behaviour. How would i know i
> need this fix? What would i see?

Based on my layman testing I have not seen any real broken behaviour
with current enablement of a zero rx/tx MAC delay for RGMII_ID/RXID/TXID.

The driver ops is called with a rx/tx_delay=0 for RGMII_ID/RXID/TXID
modes, what the MAC does with enable=true and rx/tx_delay=0 is unclear
to me.

> 
> We also need to be careful with backwards compatibility. Is there the
> potential for double bugs cancelling each other out? A board which has
> the wrong phy-mode in DT, but because of this bug, the wrong register
> is written and it actually works because of reset defaults?

To my knowledge this should have very limited effect, however I am no
network expert and after doing very basic testing on several different
rk3328/rk3566/rk3568/rk3588 I could not see any real affect with/without
this change.

The use of Fixes-tag was more to have a reference to the commit that
first should have used the DELAY_ENABLE macro.

Regards,
Jonas

> 
>     Andrew
> 
> ---
> pw-bot: cr
> 
> _______________________________________________
> Linux-rockchip mailing list
> Linux-rockchip@...ts.infradead.org
> http://lists.infradead.org/mailman/listinfo/linux-rockchip

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