Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.33 as permitted sender) client-ip=23.128.96.33;
Date:   Wed, 22 Nov 2023 13:18:33 +0100
From:   Johan Hovold <johan@kernel.org>
To:     Bjorn Andersson <quic_bjorande@quicinc.com>
Cc:     Bjorn Andersson <andersson@kernel.org>,
        Konrad Dybcio <konrad.dybcio@linaro.org>,
        Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
        Rob Herring <robh+dt@kernel.org>,
        Krzysztof Kozlowski <krzysztof.kozlowski+dt@linaro.org>,
        Conor Dooley <conor+dt@kernel.org>,
        Wesley Cheng <quic_wcheng@quicinc.com>,
        Thinh Nguyen <Thinh.Nguyen@synopsys.com>,
        Felipe Balbi <balbi@kernel.org>,
        Philipp Zabel <p.zabel@pengutronix.de>,
        linux-arm-msm@vger.kernel.org, linux-usb@vger.kernel.org,
        devicetree@vger.kernel.org, linux-kernel@vger.kernel.org,
        Krishna Kurapati PSSNV <quic_kriskura@quicinc.com>
Subject: Re: [PATCH 06/12] usb: dwc3: qcom: Add dwc3 core reference in driver
 state
Message-ID: <ZV3xmW0fDWY5-6qZ@hovoldconsulting.com>
References: <20231016-dwc3-refactor-v1-0-ab4a84165470@quicinc.com>
 <20231016-dwc3-refactor-v1-6-ab4a84165470@quicinc.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20231016-dwc3-refactor-v1-6-ab4a84165470@quicinc.com>
Precedence: bulk

On Mon, Oct 16, 2023 at 08:11:14PM -0700, Bjorn Andersson wrote:
> In the coming changes the Qualcomm DWC3 glue will be able to either
> manage the DWC3 core as a child platform_device, or directly instantiate
> it within its own context.
> 
> Introduce a reference to the dwc3 core state and make the driver
> reference the dwc3 core either the child device or this new reference.
> 
> As the new member isn't assigned, and qcom->dwc_dev is assigned in all
> current cases, the change should have no functional impact.
> 
> Signed-off-by: Bjorn Andersson <quic_bjorande@quicinc.com>
> ---
>  drivers/usb/dwc3/dwc3-qcom.c | 100 +++++++++++++++++++++++++++++++++++--------
>  1 file changed, 83 insertions(+), 17 deletions(-)
> 
> diff --git a/drivers/usb/dwc3/dwc3-qcom.c b/drivers/usb/dwc3/dwc3-qcom.c
> index 7c810712d246..901e5050363b 100644
> --- a/drivers/usb/dwc3/dwc3-qcom.c
> +++ b/drivers/usb/dwc3/dwc3-qcom.c
> @@ -67,7 +67,8 @@ struct dwc3_acpi_pdata {
>  struct dwc3_qcom {
>  	struct device		*dev;
>  	void __iomem		*qscratch_base;
> -	struct platform_device	*dwc_dev;
> +	struct platform_device	*dwc_dev; /* only used when core is separate device */
> +	struct dwc3		*dwc; /* not used when core is separate device */

Hmm. This quickly become really messy and hard to maintain. It may be
fine as an intermediate step as part of this series, but why can't you
do the conversion fully so that the Qualcomm glue driver never registers
a core platform device? Is it just about where the core driver looks for
DT properties?

Johan