netdev - Re: [PATCH 2/2] net: dwmac_socfpga: use the standard "ahb" reset

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: <c8ffee03-8a6b-1612-37ee-e5ec69853ab7@kernel.org>
Date: Thu, 13 Jul 2023 10:24:57 +0200
From: Krzysztof Kozlowski <krzk@...nel.org>
To: Jakub Kicinski <kuba@...nel.org>, Dinh Nguyen <dinguyen@...nel.org>
Cc: netdev@...r.kernel.org, davem@...emloft.net, edumazet@...gle.com,
 joabreu@...opsys.com, pabeni@...hat.com, robh+dt@...nel.org,
 krzysztof.kozlowskii+dt@...aro.org, conor+dt@...nel.org,
 devicetree@...r.kernel.org
Subject: Re: [PATCH 2/2] net: dwmac_socfpga: use the standard "ahb" reset

On 13/07/2023 02:08, Jakub Kicinski wrote:
> On Mon, 10 Jul 2023 16:13:13 -0500 Dinh Nguyen wrote:
>> -	dwmac->stmmac_ocp_rst = devm_reset_control_get_optional(dev, "stmmaceth-ocp");
>> -	if (IS_ERR(dwmac->stmmac_ocp_rst)) {
>> -		ret = PTR_ERR(dwmac->stmmac_ocp_rst);
>> -		dev_err(dev, "error getting reset control of ocp %d\n", ret);
>> -		goto err_remove_config_dt;
>> -	}
>> -
>> -	reset_control_deassert(dwmac->stmmac_ocp_rst);
> 
> Noob question, perhaps - what's the best practice for incompatible
> device tree changes?

They are an ABI break.

> Updating the in-tree definitions is good enough?

No, because this is an ABI so we expect:
1. old DTS
2. out-of-tree DTS
to work properly with new kernel (not broken by a change).

However for ABI breaks with scope limited to only one given platform, it
is the platform's maintainer choice to allow or not allow ABI breaks.
What we, Devicetree maintainers expect, is to mention and provide
rationale for the ABI break in the commit msg.

> Seems like we could quite easily continue to support "stmmaceth-ocp"
> but no point complicating the code if not required.

Best regards,
Krzysztof