[v1] usb: dwc3: qcom: Don't leave BCR asserted

[PATCH] usb: dwc3: qcom: Don't leave BCR asserted

Posted by Krishna Kurapati 6 months, 2 weeks ago

Leaving the USB BCR asserted prevents the associated GDSC to turn on. This
blocks any subsequent attempts of probing the device, e.g. after a probe
deferral, with the following showing in the log:

[    1.332226] usb30_prim_gdsc status stuck at 'off'

Leave the BCR deasserted when exiting the driver to avoid this issue.

Signed-off-by: Krishna Kurapati <krishna.kurapati@oss.qualcomm.com>
---
 drivers/usb/dwc3/dwc3-qcom.c | 7 ++-----
 1 file changed, 2 insertions(+), 5 deletions(-)

diff --git a/drivers/usb/dwc3/dwc3-qcom.c b/drivers/usb/dwc3/dwc3-qcom.c
index 7334de85ad10..ca7e1c02773a 100644
--- a/drivers/usb/dwc3/dwc3-qcom.c
+++ b/drivers/usb/dwc3/dwc3-qcom.c
@@ -680,12 +680,12 @@ static int dwc3_qcom_probe(struct platform_device *pdev)
 	ret = reset_control_deassert(qcom->resets);
 	if (ret) {
 		dev_err(&pdev->dev, "failed to deassert resets, err=%d\n", ret);
-		goto reset_assert;
+		return ret;
 	}
 
 	ret = clk_bulk_prepare_enable(qcom->num_clocks, qcom->clks);
 	if (ret < 0)
-		goto reset_assert;
+		return ret;
 
 	r = platform_get_resource(pdev, IORESOURCE_MEM, 0);
 	if (!r) {
@@ -755,8 +755,6 @@ static int dwc3_qcom_probe(struct platform_device *pdev)
 	dwc3_core_remove(&qcom->dwc);
 clk_disable:
 	clk_bulk_disable_unprepare(qcom->num_clocks, qcom->clks);
-reset_assert:
-	reset_control_assert(qcom->resets);
 
 	return ret;
 }
@@ -771,7 +769,6 @@ static void dwc3_qcom_remove(struct platform_device *pdev)
 	clk_bulk_disable_unprepare(qcom->num_clocks, qcom->clks);
 
 	dwc3_qcom_interconnect_exit(qcom);
-	reset_control_assert(qcom->resets);
 }
 
 static int dwc3_qcom_pm_suspend(struct device *dev)
-- 
2.34.1

Re: [PATCH] usb: dwc3: qcom: Don't leave BCR asserted

Posted by Thinh Nguyen 6 months, 2 weeks ago

On Wed, Jun 04, 2025, Krishna Kurapati wrote:
> Leaving the USB BCR asserted prevents the associated GDSC to turn on. This
> blocks any subsequent attempts of probing the device, e.g. after a probe
> deferral, with the following showing in the log:
> 
> [    1.332226] usb30_prim_gdsc status stuck at 'off'
> 
> Leave the BCR deasserted when exiting the driver to avoid this issue.
> 
> Signed-off-by: Krishna Kurapati <krishna.kurapati@oss.qualcomm.com>

Is this a fix? Does this need to be backported to stable?

Either way,

Acked-by: Thinh Nguyen <Thinh.Nguyen@synopsys.com>

Thanks,
Thinh

> ---
>  drivers/usb/dwc3/dwc3-qcom.c | 7 ++-----
>  1 file changed, 2 insertions(+), 5 deletions(-)
> 
> diff --git a/drivers/usb/dwc3/dwc3-qcom.c b/drivers/usb/dwc3/dwc3-qcom.c
> index 7334de85ad10..ca7e1c02773a 100644
> --- a/drivers/usb/dwc3/dwc3-qcom.c
> +++ b/drivers/usb/dwc3/dwc3-qcom.c
> @@ -680,12 +680,12 @@ static int dwc3_qcom_probe(struct platform_device *pdev)
>  	ret = reset_control_deassert(qcom->resets);
>  	if (ret) {
>  		dev_err(&pdev->dev, "failed to deassert resets, err=%d\n", ret);
> -		goto reset_assert;
> +		return ret;
>  	}
>  
>  	ret = clk_bulk_prepare_enable(qcom->num_clocks, qcom->clks);
>  	if (ret < 0)
> -		goto reset_assert;
> +		return ret;
>  
>  	r = platform_get_resource(pdev, IORESOURCE_MEM, 0);
>  	if (!r) {
> @@ -755,8 +755,6 @@ static int dwc3_qcom_probe(struct platform_device *pdev)
>  	dwc3_core_remove(&qcom->dwc);
>  clk_disable:
>  	clk_bulk_disable_unprepare(qcom->num_clocks, qcom->clks);
> -reset_assert:
> -	reset_control_assert(qcom->resets);
>  
>  	return ret;
>  }
> @@ -771,7 +769,6 @@ static void dwc3_qcom_remove(struct platform_device *pdev)
>  	clk_bulk_disable_unprepare(qcom->num_clocks, qcom->clks);
>  
>  	dwc3_qcom_interconnect_exit(qcom);
> -	reset_control_assert(qcom->resets);
>  }
>  
>  static int dwc3_qcom_pm_suspend(struct device *dev)
> -- 
> 2.34.1
>

Re: [PATCH] usb: dwc3: qcom: Don't leave BCR asserted

Posted by Konrad Dybcio 6 months ago

On 6/6/25 2:11 AM, Thinh Nguyen wrote:
> On Wed, Jun 04, 2025, Krishna Kurapati wrote:
>> Leaving the USB BCR asserted prevents the associated GDSC to turn on. This
>> blocks any subsequent attempts of probing the device, e.g. after a probe
>> deferral, with the following showing in the log:
>>
>> [    1.332226] usb30_prim_gdsc status stuck at 'off'
>>
>> Leave the BCR deasserted when exiting the driver to avoid this issue.
>>
>> Signed-off-by: Krishna Kurapati <krishna.kurapati@oss.qualcomm.com>
> 
> Is this a fix? Does this need to be backported to stable?

yes and "might as well" - this file was re-made last cycle and this
patch won't apply without the recent 

e33ebb133a24 ("usb: dwc3: qcom: Use bulk clock API and devres")

Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>

Konrad

Re: [PATCH] usb: dwc3: qcom: Don't leave BCR asserted

Posted by Krishna Kurapati 6 months ago


On 6/18/2025 2:58 AM, Konrad Dybcio wrote:
> On 6/6/25 2:11 AM, Thinh Nguyen wrote:
>> On Wed, Jun 04, 2025, Krishna Kurapati wrote:
>>> Leaving the USB BCR asserted prevents the associated GDSC to turn on. This
>>> blocks any subsequent attempts of probing the device, e.g. after a probe
>>> deferral, with the following showing in the log:
>>>
>>> [    1.332226] usb30_prim_gdsc status stuck at 'off'
>>>
>>> Leave the BCR deasserted when exiting the driver to avoid this issue.
>>>
>>> Signed-off-by: Krishna Kurapati <krishna.kurapati@oss.qualcomm.com>
>>
>> Is this a fix? Does this need to be backported to stable?
> 
> yes and "might as well" - this file was re-made last cycle and this
> patch won't apply without the recent
> 
> e33ebb133a24 ("usb: dwc3: qcom: Use bulk clock API and devres")
> 

I sent the patch on top of usb-next. Perhaps when I made this patch, the 
above mentioned patch was already present.

Thanks for the review. Will send a v2 with cc'ing stable and adding a 
fixes tag.

Regards,
Krishna,

> Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
> 
> Konrad