Message-ID: <690dec31-6de6-efe4-c1b1-f434ff1a771f@linaro.org>
Date:   Fri, 3 Mar 2023 11:12:38 +0100
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101
 Thunderbird/102.8.0
Subject: Re: [PATCH 01/16] dt-bindings: PCI: Rename Exynos PCIe binding to
 Samsung PCIe
Content-Language: en-US
To:     Shradha Todi <shradha.t@samsung.com>, lpieralisi@kernel.org,
        kw@linux.com, robh@kernel.org, bhelgaas@google.com,
        krzysztof.kozlowski+dt@linaro.org, alim.akhtar@samsung.com,
        jingoohan1@gmail.com, Sergey.Semin@baikalelectronics.ru,
        lukas.bulwahn@gmail.com, hongxing.zhu@nxp.com, tglx@linutronix.de,
        m.szyprowski@samsung.com, jh80.chung@samsung.co,
        pankaj.dubey@samsung.com
Cc:     linux-pci@vger.kernel.org, devicetree@vger.kernel.org,
        linux-arm-kernel@lists.infradead.org,
        linux-samsung-soc@vger.kernel.org, linux-kernel@vger.kernel.org
References: <20230214121333.1837-1-shradha.t@samsung.com>
 <CGME20230214121404epcas5p3bfa6af0151b7f319d418f7c0dbed7c5a@epcas5p3.samsung.com>
 <20230214121333.1837-2-shradha.t@samsung.com>
 <b40cafa1-396f-e6cd-3240-bc879d5f2c8b@linaro.org>
 <139501d94d06$2e159850$8a40c8f0$@samsung.com>
From:   Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
In-Reply-To: <139501d94d06$2e159850$8a40c8f0$@samsung.com>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit
Precedence: bulk

On 02/03/2023 13:54, Shradha Todi wrote:
> 
> 
>> -----Original Message-----
>> From: Krzysztof Kozlowski [mailto:krzysztof.kozlowski@linaro.org]
>> Sent: 16 February 2023 16:24
>> To: Shradha Todi <shradha.t@samsung.com>; lpieralisi@kernel.org;
>> kw@linux.com; robh@kernel.org; bhelgaas@google.com;
>> krzysztof.kozlowski+dt@linaro.org; alim.akhtar@samsung.com;
>> jingoohan1@gmail.com; Sergey.Semin@baikalelectronics.ru;
>> lukas.bulwahn@gmail.com; hongxing.zhu@nxp.com; tglx@linutronix.de;
>> m.szyprowski@samsung.com; jh80.chung@samsung.co;
>> pankaj.dubey@samsung.com
>> Cc: linux-pci@vger.kernel.org; devicetree@vger.kernel.org; linux-arm-
>> kernel@lists.infradead.org; linux-samsung-soc@vger.kernel.org; linux-
>> kernel@vger.kernel.org
>> Subject: Re: [PATCH 01/16] dt-bindings: PCI: Rename Exynos PCIe binding to
>> Samsung PCIe
>>
>> On 14/02/2023 13:13, Shradha Todi wrote:
>>> The current DT bindings is being used for Exynos5433 SoC only.
>>> In order to extend this binding for all SoCs manufactured by Samsung
>>> using DWC PCIe controller, renaming this file to a more generic name.
>>
>> Thank you for your patch. There is something to discuss/improve.
>>
>>>
>>> Signed-off-by: Shradha Todi <shradha.t@samsung.com>
>>> ---
>>>  .../pci/{samsung,exynos-pcie.yaml => samsung,pcie.yaml}     | 6 +++---
>>>  1 file changed, 3 insertions(+), 3 deletions(-)  rename
>>> Documentation/devicetree/bindings/pci/{samsung,exynos-pcie.yaml =>
>>> samsung,pcie.yaml} (93%)
>>>
>>> diff --git
>>> a/Documentation/devicetree/bindings/pci/samsung,exynos-pcie.yaml
>>> b/Documentation/devicetree/bindings/pci/samsung,pcie.yaml
>>> similarity index 93%
>>> rename from
>>> Documentation/devicetree/bindings/pci/samsung,exynos-pcie.yaml
>>> rename to Documentation/devicetree/bindings/pci/samsung,pcie.yaml
>>> index f20ed7e709f7..6cd36d9ccba0 100644
>>> --- a/Documentation/devicetree/bindings/pci/samsung,exynos-pcie.yaml
>>> +++ b/Documentation/devicetree/bindings/pci/samsung,pcie.yaml
>>
>> We keep the name rather tied to compatible, not generic. There are no other
>> compatibles here, so I don't think we should rename it.
>>
> 
> Our intention to rename was to have a common name for Samsung manufactured SoCs having PCIe controller.
> Though this change may not be a blocker for us but we feel it will be good to have a common name as this file will
> not have bindings only for Exynos series of SoC (Samsung Sys. LSI designed) but also custom ASICs such as FSD / ARTPEC SoC (Samsung Foundry designed). 
> We hope we are not breaking any ABI as such in this patch.

There is no FSD/Artpec added here, so renaming just for "rename" is a
no. If you add new hardware here, this could have sense, depending on
the hardware. But since we pretty often expect the first compatible to
be the name of the file, why renaming at all?

Best regards,
Krzysztof