Message-ID: <d73f0eec-6e02-e72f-79d7-2c56245f7651@linaro.org>
Date:   Tue, 24 Jan 2023 10:53:35 +0100
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101
 Thunderbird/102.7.0
Subject: Re: [PATCH 2/7] clk: qcom: Add Global Clock Controller driver for
 IPQ9574
Content-Language: en-US
To:     Devi Priya <quic_devipriy@quicinc.com>, agross@kernel.org,
        andersson@kernel.org, robh+dt@kernel.org,
        krzysztof.kozlowski+dt@linaro.org, mturquette@baylibre.com,
        sboyd@kernel.org, linus.walleij@linaro.org,
        catalin.marinas@arm.com, will@kernel.org, p.zabel@pengutronix.de,
        shawnguo@kernel.org, arnd@arndb.de, marcel.ziswiler@toradex.com,
        dmitry.baryshkov@linaro.org, nfraprado@collabora.com,
        broonie@kernel.org, tdas@codeaurora.org,
        linux-arm-msm@vger.kernel.org, devicetree@vger.kernel.org,
        linux-kernel@vger.kernel.org, linux-clk@vger.kernel.org,
        linux-gpio@vger.kernel.org, linux-arm-kernel@lists.infradead.org
Cc:     quic_srichara@quicinc.com, quic_gokulsri@quicinc.com,
        quic_sjaganat@quicinc.com, quic_kathirav@quicinc.com,
        quic_arajkuma@quicinc.com, quic_anusha@quicinc.com,
        quic_poovendh@quicinc.com
References: <20230110121316.24892-1-quic_devipriy@quicinc.com>
 <20230110121316.24892-3-quic_devipriy@quicinc.com>
 <de346d71-1fe7-e357-d220-d4468e4bb933@linaro.org>
 <afd2e5c8-fa5a-ac1f-4ede-4ab1f91c0d0d@quicinc.com>
 <9bdf757d-1fa0-106f-eb77-7f2a8593213f@linaro.org>
 <2852fc37-284f-6534-f163-45b37b153db1@quicinc.com>
From:   Konrad Dybcio <konrad.dybcio@linaro.org>
In-Reply-To: <2852fc37-284f-6534-f163-45b37b153db1@quicinc.com>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Precedence: bulk


On 24.01.2023 08:27, Devi Priya wrote:
> 
> 
> On 1/13/2023 7:39 PM, Konrad Dybcio wrote:
>>
>>
>> On 13.01.2023 14:21, Devi Priya wrote:
>>>
>>>
>>> On 1/10/2023 6:07 PM, Konrad Dybcio wrote:
>>>>
>>>>
>>>> On 10.01.2023 13:13, devi priya wrote:
>>>>> Add Global Clock Controller (GCC) driver for ipq9574 based devices
>>>>>
>>>>> Co-developed-by: Anusha Rao <quic_anusha@quicinc.com>
>>>>> Signed-off-by: Anusha Rao <quic_anusha@quicinc.com>
>>>>> Signed-off-by: devi priya <quic_devipriy@quicinc.com>
>>>>> ---
>> [...]
>>
>>>>> +static struct clk_branch gcc_blsp1_qup6_i2c_apps_clk = {
>>>>> +    .halt_reg = 0x07024,
>>>>> +    .clkr = {
>>>>> +        .enable_reg = 0x07024,
>>>>> +        .enable_mask = BIT(0),
>>>>> +        .hw.init = &(struct clk_init_data) {
>>>>> +            .name = "gcc_blsp1_qup6_i2c_apps_clk",
>>>>> +            .parent_hws = (const struct clk_hw *[]) {
>>>>> +                    &blsp1_qup6_i2c_apps_clk_src.clkr.hw },
>>>>> +            .num_parents = 1,
>>>>> +            .flags = CLK_SET_RATE_PARENT | CLK_IGNORE_UNUSED,
>>>> Sounds very much like a hack..
>>> Got it, will remove the clock entry as it is not being used in linux
>> I'm not sure removing it is the best option, somebody might have a
>> funky board where they use this particular QUP for I2C for whatever
>> reason and then the clock would have to be re-added..
> Sure, Understood
> This clock is used by the RPM component to communicate with PMIC and we
> would add the critical flag here
Okay, so this SoC is intended to ship with some RPM PMICs and
*always* with an I2C companion that's required for some basic
functionality, correct?

Otherwise, if it's just for wifi/multimedia/etc (like PM8008ij
on some newer devices), you should not make it critical and
simply rely on Linux keeping it alive like so:

consumer takes a regulator
the regulator does not go to sleep because it's consumed
the PMIC is active because a regulator on it is being used
the I2C bus is active because its child PMIC is used
the I2C clocks are alive because there's an active user

Konrad
> 
>> Thanks for addressing all of the review comments so thoroughly!
>>
>> Konrad
> 
> Best Regards,
> Devi Priya