Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Message-ID: <f3ace6f0-12fb-8e97-5115-5511b4ffb223@kernel.org>
Date:   Thu, 22 Dec 2022 16:19:08 +0100
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101
 Thunderbird/102.6.0
Subject: Re: [RFC PATCH 5/7] bus: stm32_sys_bus: add support for STM32MP15 and
 STM32MP13 system bus
Content-Language: en-US
To:     Gatien CHEVALLIER <gatien.chevallier@foss.st.com>,
        alexandre.torgue@foss.st.com, robh+dt@kernel.org,
        Oleksii_Moisieiev@epam.com, linus.walleij@linaro.org,
        gregkh@linuxfoundation.org
Cc:     linux-stm32@st-md-mailman.stormreply.com,
        linux-arm-kernel@lists.infradead.org, linux-kernel@vger.kernel.org,
        loic.pallardy@st.com, devicetree@vger.kernel.org,
        mark.rutland@arm.com, arnd@arndb.de
References: <20221221173055.11719-1-gatien.chevallier@foss.st.com>
 <20221221173055.11719-6-gatien.chevallier@foss.st.com>
 <ed8aa4d3-9570-eb72-a7d4-3b690adc3a6d@kernel.org>
 <e2086a52-4db8-e247-43de-c78ffa479634@foss.st.com>
From:   Krzysztof Kozlowski <krzk@kernel.org>
In-Reply-To: <e2086a52-4db8-e247-43de-c78ffa479634@foss.st.com>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit
Precedence: bulk

On 22/12/2022 15:30, Gatien CHEVALLIER wrote:
>>
>>> +	if (IS_ERR(mmio))
>>> +		return PTR_ERR(mmio);
>>> +
>>> +	pdata->sys_bus_base = mmio;
>>> +
>>> +	mdata = (struct stm32_sys_bus_match_data *)of_device_get_match_data(&pdev->dev);
>>
>> Why do you need the cast?
> 
> I do not :) TBH, mdata is not useful at all. Changing to directly assign 
> to pdata->pconf, that is now const there is no reason to modify it.

mdata should be const and then no need for cast.

> 
>>
>>> +	if (!mdata)
>>
>> Can you explain the case when this can actually happen? If it can, you
>> have bug in ID table.
> 
> No I cannot as the driver is probed. It is only a sanity check, I can 
> remove it for V3. However the function can return NULL... Would you 
> prefer an explicit check on NULL or a simple removal?

I propose to drop it. This simply cannot happen.

> 
>>
>>> +		return -EINVAL;
>>> +

Best regards,
Krzysztof