[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <bap2qy5cqhdzphc7qeoqztp65g6jsmwpxkghpgagxpj4qdifts@ogj7nb34rv6f>
Date: Thu, 20 Feb 2025 12:48:14 +0200
From: Dmitry Baryshkov <dmitry.baryshkov@...aro.org>
To: Jagadeesh Kona <quic_jkona@...cinc.com>
Cc: Bjorn Andersson <andersson@...nel.org>,
Michael Turquette <mturquette@...libre.com>, Stephen Boyd <sboyd@...nel.org>, Rob Herring <robh@...nel.org>,
Krzysztof Kozlowski <krzk+dt@...nel.org>, Conor Dooley <conor+dt@...nel.org>,
Konrad Dybcio <konradybcio@...nel.org>, Ajit Pandey <quic_ajipan@...cinc.com>,
Imran Shaik <quic_imrashai@...cinc.com>, Taniya Das <quic_tdas@...cinc.com>,
Satya Priya Kakitapalli <quic_skakitap@...cinc.com>, linux-arm-msm@...r.kernel.org, linux-clk@...r.kernel.org,
devicetree@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH 3/5] clk: qcom: common: Attach clock power domains
conditionally
On Thu, Feb 20, 2025 at 12:43:42PM +0530, Jagadeesh Kona wrote:
>
>
> On 2/19/2025 5:27 PM, Dmitry Baryshkov wrote:
> > On Wed, Feb 19, 2025 at 05:06:11PM +0530, Jagadeesh Kona wrote:
> >>
> >>
> >> On 2/18/2025 10:48 PM, Dmitry Baryshkov wrote:
> >>> On Tue, Feb 18, 2025 at 07:56:48PM +0530, Jagadeesh Kona wrote:
> >>>> Attach clock power domains in qcom_cc_really_probe() only
> >>>> if the clock controller has not already attached to them.
> >>>
> >>> Squash this to the previous patch and call the new function. No need to
> >>> duplicate the code.
> >>>
> >>
> >> I tried calling the new function here instead of duplicating code, but that
> >> is leading to below warning since the desc passed to qcom_cc_really_probe()
> >> has a const qualifier and hence we cannot update desc->pd_list inside
> >> qcom_cc_really_probe().
> >>
> >> drivers/clk/qcom/common.c:305:33: WARNING : passing argument 2 of ‘qcom_cc_attach_pds’ discards ‘const’ qualifier from pointer target type [-Wdiscarded-qualifiers]
> >
> > It sounds like this can be fixed with a one-line patch.
> >
>
> Removing const qualifier to qcom_cc_really_probe() will fix this, but that requires changes in
> many other drivers which are currently passing const descriptor to it.
And this points out that the pd_list should not be a part of the
struct qcom_cc_desc. You are not using it in the code, so allocate that
memory on the fly, pass it to devm_pm_domain_attach_list() and then
forget about it.
>
> But I can squash this to my previous patch by updating my qcom_cc_attach_pds() function
> prototype as below and then calling that new function here
>
> -int qcom_cc_attach_pds(struct device *dev, struct qcom_cc_desc *desc)
> +int qcom_cc_attach_pds(struct device *dev, struct dev_pm_domain_list *pd_list)
>
> - ret = devm_pm_domain_attach_list(dev, NULL, &cc->pd_list);
> - if (ret < 0 && ret != -EEXIST)
> + ret = qcom_cc_attach_pds(dev, cc->pd_list);
> + if (ret)
>
> Thanks,
> Jagadeesh
>
> >>
> >> Thanks,
> >> Jagadeesh
> >>
> >>>>
> >>>> Signed-off-by: Jagadeesh Kona <quic_jkona@...cinc.com>
> >>>> ---
> >>>> drivers/clk/qcom/common.c | 9 ++++++---
> >>>> 1 file changed, 6 insertions(+), 3 deletions(-)
> >>>>
> >>>> diff --git a/drivers/clk/qcom/common.c b/drivers/clk/qcom/common.c
> >>>> index ec27f70b24bdec24edd2f6b3df0d766fc1cdcbf0..eb7e2a56d1d135f839fd9bd470ba6231ce775a8c 100644
> >>>> --- a/drivers/clk/qcom/common.c
> >>>> +++ b/drivers/clk/qcom/common.c
> >>>> @@ -300,9 +300,12 @@ int qcom_cc_really_probe(struct device *dev,
> >>>> if (!cc)
> >>>> return -ENOMEM;
> >>>>
> >>>> - ret = devm_pm_domain_attach_list(dev, NULL, &cc->pd_list);
> >>>> - if (ret < 0 && ret != -EEXIST)
> >>>> - return ret;
> >>>> + cc->pd_list = desc->pd_list;
> >>>> + if (!cc->pd_list) {
> >>>> + ret = devm_pm_domain_attach_list(dev, NULL, &cc->pd_list);
> >>>> + if (ret < 0 && ret != -EEXIST)
> >>>> + return ret;
> >>>> + }
> >>>>
> >>>> reset = &cc->reset;
> >>>> reset->rcdev.of_node = dev->of_node;
> >>>>
> >>>> --
> >>>> 2.34.1
> >>>>
> >>>
> >
--
With best wishes
Dmitry
Powered by blists - more mailing lists