linux-kernel - Re: [PATCH 1/3] USB: dwc3: qcom: fix resource leaks on probe deferral

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <uof6ti3ula3gfijsucp5yjp6bxldjhmyek77xub5ybifrm3g3b@d3qd4jqusvix>
Date:   Fri, 17 Nov 2023 13:03:48 -0600
From:   Andrew Halaney <ahalaney@...hat.com>
To:     Johan Hovold <johan+linaro@...nel.org>
Cc:     Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        Andy Gross <agross@...nel.org>,
        Bjorn Andersson <andersson@...nel.org>,
        Konrad Dybcio <konrad.dybcio@...aro.org>,
        Thinh Nguyen <Thinh.Nguyen@...opsys.com>,
        Krishna Kurapati PSSNV <quic_kriskura@...cinc.com>,
        linux-arm-msm@...r.kernel.org, linux-usb@...r.kernel.org,
        linux-kernel@...r.kernel.org, stable@...r.kernel.org,
        Christophe JAILLET <christophe.jaillet@...adoo.fr>,
        Lee Jones <lee@...nel.org>
Subject: Re: [PATCH 1/3] USB: dwc3: qcom: fix resource leaks on probe deferral

On Fri, Nov 17, 2023 at 06:36:48PM +0100, Johan Hovold wrote:
> The driver needs to deregister and free the newly allocated dwc3 core
> platform device on ACPI probe errors (e.g. probe deferral) and on driver
> unbind but instead it leaked those resources while erroneously dropping
> a reference to the parent platform device which is still in use.
> 
> For OF probing the driver takes a reference to the dwc3 core platform
> device which has also always been leaked.
> 
> Fix the broken ACPI tear down and make sure to drop the dwc3 core
> reference for both OF and ACPI.
> 
> Fixes: 8fd95da2cfb5 ("usb: dwc3: qcom: Release the correct resources in dwc3_qcom_remove()")
> Fixes: 2bc02355f8ba ("usb: dwc3: qcom: Add support for booting with ACPI")
> Fixes: a4333c3a6ba9 ("usb: dwc3: Add Qualcomm DWC3 glue driver")
> Cc: stable@...r.kernel.org      # 4.18
> Cc: Christophe JAILLET <christophe.jaillet@...adoo.fr>
> Cc: Lee Jones <lee@...nel.org>
> Signed-off-by: Johan Hovold <johan+linaro@...nel.org>
> ---

I foobared the review of one of the patches listed as being fixed by
this, but for what it is worth I think this makes sense.

Hopefully my eye is better this time.

Acked-by: Andrew Halaney <ahalaney@...hat.com>

>  drivers/usb/dwc3/dwc3-qcom.c | 9 ++++++---
>  1 file changed, 6 insertions(+), 3 deletions(-)
> 
> diff --git a/drivers/usb/dwc3/dwc3-qcom.c b/drivers/usb/dwc3/dwc3-qcom.c
> index 3de43df6bbe8..00c3021b43ce 100644
> --- a/drivers/usb/dwc3/dwc3-qcom.c
> +++ b/drivers/usb/dwc3/dwc3-qcom.c
> @@ -758,6 +758,7 @@ static int dwc3_qcom_of_register_core(struct platform_device *pdev)
>  	if (!qcom->dwc3) {
>  		ret = -ENODEV;
>  		dev_err(dev, "failed to get dwc3 platform device\n");
> +		of_platform_depopulate(dev);
>  	}
>  
>  node_put:
> @@ -899,7 +900,7 @@ static int dwc3_qcom_probe(struct platform_device *pdev)
>  
>  	if (ret) {
>  		dev_err(dev, "failed to register DWC3 Core, err=%d\n", ret);
> -		goto depopulate;
> +		goto clk_disable;
>  	}
>  
>  	ret = dwc3_qcom_interconnect_init(qcom);
> @@ -934,7 +935,8 @@ static int dwc3_qcom_probe(struct platform_device *pdev)
>  	if (np)
>  		of_platform_depopulate(&pdev->dev);
>  	else
> -		platform_device_put(pdev);
> +		platform_device_del(qcom->dwc3);
> +	platform_device_put(qcom->dwc3);
>  clk_disable:
>  	for (i = qcom->num_clocks - 1; i >= 0; i--) {
>  		clk_disable_unprepare(qcom->clks[i]);
> @@ -957,7 +959,8 @@ static void dwc3_qcom_remove(struct platform_device *pdev)
>  	if (np)
>  		of_platform_depopulate(&pdev->dev);
>  	else
> -		platform_device_put(pdev);
> +		platform_device_del(qcom->dwc3);
> +	platform_device_put(qcom->dwc3);
>  
>  	for (i = qcom->num_clocks - 1; i >= 0; i--) {
>  		clk_disable_unprepare(qcom->clks[i]);
> -- 
> 2.41.0
> 
>