devicetree.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Konrad Dybcio <konrad.dybcio@linaro.org>
To: neil.armstrong@linaro.org, Rob Herring <robh+dt@kernel.org>,
	Krzysztof Kozlowski <krzysztof.kozlowski+dt@linaro.org>,
	Bjorn Andersson <andersson@kernel.org>,
	Konrad Dybcio <konrad.dybcio@somainline.org>,
	Andy Gross <agross@kernel.org>
Cc: linux-kernel@vger.kernel.org, devicetree@vger.kernel.org,
	linux-arm-msm@vger.kernel.org, linux-i2c@vger.kernel.org
Subject: Re: [PATCH v2 3/6] soc: qcom: geni-se: add desc struct to specify clocks from device match data
Date: Tue, 29 Nov 2022 14:21:32 +0100	[thread overview]
Message-ID: <92b39c01-1283-0b7c-6ba9-3edd2d4de002@linaro.org> (raw)
In-Reply-To: <cbccbcbe-5006-cdc9-f3af-39c53d87b8a5@linaro.org>



On 29.11.2022 09:24, Neil Armstrong wrote:
> On 18/11/2022 10:06, Konrad Dybcio wrote:
>>
>>
>> On 18/11/2022 09:45, Neil Armstrong wrote:
>>> The I2C Master Hub is a stripped down version of the GENI Serial Engine
>>> QUP Wrapper Controller but only supporting I2C serial engines without
>>> DMA support.
>>>
>>> Prepare support for the I2C Master Hub variant by moving the required
>>> clocks list to a new desc struct then passing it through the compatible
>>> match data.
>>>
>>> Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
>>> Reviewed-by: Konrad Dybcio <konrad.dybcio@linaro.org>
>>> ---
>>>   drivers/soc/qcom/qcom-geni-se.c | 59 +++++++++++++++++++++++++++++++----------
>>>   1 file changed, 45 insertions(+), 14 deletions(-)
>>>
>>> diff --git a/drivers/soc/qcom/qcom-geni-se.c b/drivers/soc/qcom/qcom-geni-se.c
>>> index a0ceeede450f..ced2a2932eda 100644
>>> --- a/drivers/soc/qcom/qcom-geni-se.c
>>> +++ b/drivers/soc/qcom/qcom-geni-se.c
>>> @@ -81,19 +81,31 @@
>>>    */
>>>   #define MAX_CLK_PERF_LEVEL 32
>>> -#define NUM_AHB_CLKS 2
>>> +#define MAX_CLKS 2
>>>   /**
>>>    * struct geni_wrapper - Data structure to represent the QUP Wrapper Core
>>>    * @dev:        Device pointer of the QUP wrapper core
>>>    * @base:        Base address of this instance of QUP wrapper core
>>> - * @ahb_clks:        Handle to the primary & secondary AHB clocks
>>> + * @clks:        Handle to the primary & optional secondary AHB clocks
>>> + * @num_clks:        Count of clocks
>>>    * @to_core:        Core ICC path
>>>    */
>>>   struct geni_wrapper {
>>>       struct device *dev;
>>>       void __iomem *base;
>>> -    struct clk_bulk_data ahb_clks[NUM_AHB_CLKS];
>>> +    struct clk_bulk_data clks[MAX_CLKS];
>>> +    unsigned int num_clks;
>>> +};
>>> +
>>> +/**
>>> + * struct geni_se_desc - Data structure to represent the QUP Wrapper resources
>>> + * @clks:        Name of the primary & optional secondary AHB clocks
>>> + * @num_clks:        Count of clock names
>>> + */
>>> +struct geni_se_desc {
>>> +    unsigned int num_clks;
>>> +    const char * const *clks;
>>>   };
>>>   static const char * const icc_path_names[] = {"qup-core", "qup-config",
>>> @@ -496,8 +508,7 @@ static void geni_se_clks_off(struct geni_se *se)
>>>       struct geni_wrapper *wrapper = se->wrapper;
>>>       clk_disable_unprepare(se->clk);
>>> -    clk_bulk_disable_unprepare(ARRAY_SIZE(wrapper->ahb_clks),
>>> -                        wrapper->ahb_clks);
>>> +    clk_bulk_disable_unprepare(wrapper->num_clks, wrapper->clks);
>>>   }
>>>   /**
>>> @@ -528,15 +539,13 @@ static int geni_se_clks_on(struct geni_se *se)
>>>       int ret;
>>>       struct geni_wrapper *wrapper = se->wrapper;
>>> -    ret = clk_bulk_prepare_enable(ARRAY_SIZE(wrapper->ahb_clks),
>>> -                        wrapper->ahb_clks);
>>> +    ret = clk_bulk_prepare_enable(wrapper->num_clks, wrapper->clks);
>>>       if (ret)
>>>           return ret;
>>>       ret = clk_prepare_enable(se->clk);
>>>       if (ret)
>>> -        clk_bulk_disable_unprepare(ARRAY_SIZE(wrapper->ahb_clks),
>>> -                            wrapper->ahb_clks);
>>> +        clk_bulk_disable_unprepare(wrapper->num_clks, wrapper->clks);
>>>       return ret;
>>>   }
>>> @@ -887,11 +896,23 @@ static int geni_se_probe(struct platform_device *pdev)
>>>           return PTR_ERR(wrapper->base);
>>>       if (!has_acpi_companion(&pdev->dev)) {
>>> -        wrapper->ahb_clks[0].id = "m-ahb";
>>> -        wrapper->ahb_clks[1].id = "s-ahb";
>>> -        ret = devm_clk_bulk_get(dev, NUM_AHB_CLKS, wrapper->ahb_clks);
>>> +        const struct geni_se_desc *desc;
>>> +        int i;
>>> +
>>> +        desc = device_get_match_data(&pdev->dev);
>>> +        if (!desc)
>>> +            return -EINVAL;
>>> +
>>> +        wrapper->num_clks = min_t(unsigned int, desc->num_clks, MAX_CLKS);
>>> +        if (wrapper->num_clks < desc->num_clks)
>> This will never execute (except if somebody adding a third desc would make a mistake or not update MAX_CLKS), as wrapper->num_clks will only be < desc->num_clks if desc->num_clks > MAX_CLKS.
> 
> You're right, I did read too fast.
> 
>>
>> I was thinking about getting the number of actual clocks passed to the device in the DT, but I can't find a helper function for that, so it would probably require some kind of manual looping.. I guess we can drop this. Or leave it to save somebody pulling their hair out in an unlikely event. I guess I'm fine with both.
> 
> This would be:
> 
> of_count_phandle_with_args(dev->of_node, "clocks", "#clock-cells")
> 
> but ultimately if the number of clocks is lower than requested, it will fail
> in the call to devm_clk_bulk_get().
> 
> Would we warn if the DT clocks count is higher ? or simply fail if lower ?
Just "fail if lower" sounds good.

Konrad
> 
> Neil
> 
>>
>>
>>> +            dev_warn(dev, "too much clocks described in DT\n")
>> If you leave it, s/too much/Too many/
>>
>>
>> Konrad
>>> +
>>> +        for (i = 0; i < wrapper->num_clks; ++i)
>>> +            wrapper->clks[i].id = desc->clks[i];
>>> +
>>> +        ret = devm_clk_bulk_get(dev, wrapper->num_clks, wrapper->clks);
>>>           if (ret) {
>>> -            dev_err(dev, "Err getting AHB clks %d\n", ret);
>>> +            dev_err(dev, "Err getting clks %d\n", ret);
>>>               return ret;
>>>           }
>>>       }
>>> @@ -901,8 +922,18 @@ static int geni_se_probe(struct platform_device *pdev)
>>>       return devm_of_platform_populate(dev);
>>>   }
>>> +static const char * const qup_clks[] = {
>>> +    "m-ahb",
>>> +    "s-ahb",
>>> +};
>>> +
>>> +static const struct geni_se_desc qup_desc = {
>>> +    .clks = qup_clks,
>>> +    .num_clks = ARRAY_SIZE(qup_clks),
>>> +};
>>> +
>>>   static const struct of_device_id geni_se_dt_match[] = {
>>> -    { .compatible = "qcom,geni-se-qup", },
>>> +    { .compatible = "qcom,geni-se-qup", .data = &qup_desc },
>>>       {}
>>>   };
>>>   MODULE_DEVICE_TABLE(of, geni_se_dt_match);
>>>
> 

  reply	other threads:[~2022-11-29 13:21 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-11-18  8:45 [PATCH v2 0/6] soc: qcom: add support for the I2C Master Hub Neil Armstrong
2022-11-18  8:45 ` [PATCH v2 1/6] dt-bindings: qcom: geni-se: document I2C Master Hub wrapper variant Neil Armstrong
2022-11-18  9:08   ` Krzysztof Kozlowski
2022-11-18  8:45 ` [PATCH v2 2/6] dt-bindings: i2c: qcom-geni: document I2C Master Hub serial I2C engine Neil Armstrong
2022-11-18  9:08   ` Krzysztof Kozlowski
2022-11-18  8:45 ` [PATCH v2 3/6] soc: qcom: geni-se: add desc struct to specify clocks from device match data Neil Armstrong
2022-11-18  9:06   ` Konrad Dybcio
2022-11-29  8:24     ` Neil Armstrong
2022-11-29 13:21       ` Konrad Dybcio [this message]
2022-11-18  8:45 ` [PATCH v2 4/6] soc: qcom: geni-se: add support for I2C Master Hub wrapper variant Neil Armstrong
2022-11-18  8:45 ` [PATCH v2 5/6] i2c: qcom-geni: add desc struct to prepare support for I2C Master Hub variant Neil Armstrong
2022-11-18  8:45 ` [PATCH v2 6/6] i2c: qcom-geni: add " Neil Armstrong

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=92b39c01-1283-0b7c-6ba9-3edd2d4de002@linaro.org \
    --to=konrad.dybcio@linaro.org \
    --cc=agross@kernel.org \
    --cc=andersson@kernel.org \
    --cc=devicetree@vger.kernel.org \
    --cc=konrad.dybcio@somainline.org \
    --cc=krzysztof.kozlowski+dt@linaro.org \
    --cc=linux-arm-msm@vger.kernel.org \
    --cc=linux-i2c@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=neil.armstrong@linaro.org \
    --cc=robh+dt@kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).