[PATCH] usb: gadget: m66592-udc: Add check for clk_enable()

[PATCH] usb: gadget: m66592-udc: Add check for clk_enable()

[Mingwei Zheng]

Add check for the return value of clk_enable() to catch the potential
error.

Fixes: b4822e2317e8 ("usb: gadget: m66592-udc: Convert to use module_platform_driver()")
Signed-off-by: Mingwei Zheng <zmw12306@gmail.com>
Signed-off-by: Jiasheng Jiang <jiashengjiangcool@gmail.com>
---
 drivers/usb/gadget/udc/m66592-udc.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/drivers/usb/gadget/udc/m66592-udc.c b/drivers/usb/gadget/udc/m66592-udc.c
index a938b2af0944..bf408476a24c 100644
--- a/drivers/usb/gadget/udc/m66592-udc.c
+++ b/drivers/usb/gadget/udc/m66592-udc.c
@@ -1606,7 +1606,11 @@ static int m66592_probe(struct platform_device *pdev)
 			ret = PTR_ERR(m66592->clk);
 			goto clean_up2;
 		}
-		clk_enable(m66592->clk);
+		ret = clk_enable(m66592->clk);
+		if (ret) {
+			clk_put(m66592->clk);
+			goto clean_up2;
+		}
 	}
 
 	INIT_LIST_HEAD(&m66592->gadget.ep_list);
-- 
2.34.1

Re: [PATCH] usb: gadget: m66592-udc: Add check for clk_enable()

[Greg KH]

On Sun, Dec 15, 2024 at 03:53:58PM -0500, Mingwei Zheng wrote:
> Add check for the return value of clk_enable() to catch the potential
> error.
> 
> Fixes: b4822e2317e8 ("usb: gadget: m66592-udc: Convert to use module_platform_driver()")
> Signed-off-by: Mingwei Zheng <zmw12306@gmail.com>
> Signed-off-by: Jiasheng Jiang <jiashengjiangcool@gmail.com>

Why this order of signed-off-by lines?  Shouldn't yours be last?  Who
wrote this patch?

> ---
>  drivers/usb/gadget/udc/m66592-udc.c | 6 +++++-
>  1 file changed, 5 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/usb/gadget/udc/m66592-udc.c b/drivers/usb/gadget/udc/m66592-udc.c
> index a938b2af0944..bf408476a24c 100644
> --- a/drivers/usb/gadget/udc/m66592-udc.c
> +++ b/drivers/usb/gadget/udc/m66592-udc.c
> @@ -1606,7 +1606,11 @@ static int m66592_probe(struct platform_device *pdev)
>  			ret = PTR_ERR(m66592->clk);
>  			goto clean_up2;
>  		}
> -		clk_enable(m66592->clk);
> +		ret = clk_enable(m66592->clk);
> +		if (ret) {
> +			clk_put(m66592->clk);
> +			goto clean_up2;
> +		}

How did you find this and how was it tested?

thanks,

greg k-h

Re: [PATCH] usb: gadget: m66592-udc: Add check for clk_enable()

[Mingwei Zheng]

Hi Greg,

On Mon, Dec 16, 2024 at 2:56 AM Greg KH <gregkh@linuxfoundation.org> wrote:
>
> On Sun, Dec 15, 2024 at 03:53:58PM -0500, Mingwei Zheng wrote:
> > Add check for the return value of clk_enable() to catch the potential
> > error.
> >
> > Fixes: b4822e2317e8 ("usb: gadget: m66592-udc: Convert to use module_platform_driver()")
> > Signed-off-by: Mingwei Zheng <zmw12306@gmail.com>
> > Signed-off-by: Jiasheng Jiang <jiashengjiangcool@gmail.com>
>
> Why this order of signed-off-by lines?  Shouldn't yours be last?  Who
> wrote this patch?
>

I listed two names because both of us co-authored this patch.

> > ---
> >  drivers/usb/gadget/udc/m66592-udc.c | 6 +++++-
> >  1 file changed, 5 insertions(+), 1 deletion(-)
> >
> > diff --git a/drivers/usb/gadget/udc/m66592-udc.c b/drivers/usb/gadget/udc/m66592-udc.c
> > index a938b2af0944..bf408476a24c 100644
> > --- a/drivers/usb/gadget/udc/m66592-udc.c
> > +++ b/drivers/usb/gadget/udc/m66592-udc.c
> > @@ -1606,7 +1606,11 @@ static int m66592_probe(struct platform_device *pdev)
> >                       ret = PTR_ERR(m66592->clk);
> >                       goto clean_up2;
> >               }
> > -             clk_enable(m66592->clk);
> > +             ret = clk_enable(m66592->clk);
> > +             if (ret) {
> > +                     clk_put(m66592->clk);
> > +                     goto clean_up2;
> > +             }
>
> How did you find this and how was it tested?
>
> thanks,
>
> greg k-h

We found it through a static analysis tool. Additionally, we validated
the patch's correctness
using the built-in tests provided during the compilation process.
Please kindly let me know if you need further details or have any
questions. Thank you!

Best,
Mingwei

Re: [PATCH] usb: gadget: m66592-udc: Add check for clk_enable()

[Greg KH]

On Tue, Dec 17, 2024 at 10:06:26PM -0500, Mingwei Zheng wrote:
> Hi Greg,
> 
> On Mon, Dec 16, 2024 at 2:56 AM Greg KH <gregkh@linuxfoundation.org> wrote:
> >
> > On Sun, Dec 15, 2024 at 03:53:58PM -0500, Mingwei Zheng wrote:
> > > Add check for the return value of clk_enable() to catch the potential
> > > error.
> > >
> > > Fixes: b4822e2317e8 ("usb: gadget: m66592-udc: Convert to use module_platform_driver()")
> > > Signed-off-by: Mingwei Zheng <zmw12306@gmail.com>
> > > Signed-off-by: Jiasheng Jiang <jiashengjiangcool@gmail.com>
> >
> > Why this order of signed-off-by lines?  Shouldn't yours be last?  Who
> > wrote this patch?
> >
> 
> I listed two names because both of us co-authored this patch.
> 
> > > ---
> > >  drivers/usb/gadget/udc/m66592-udc.c | 6 +++++-
> > >  1 file changed, 5 insertions(+), 1 deletion(-)
> > >
> > > diff --git a/drivers/usb/gadget/udc/m66592-udc.c b/drivers/usb/gadget/udc/m66592-udc.c
> > > index a938b2af0944..bf408476a24c 100644
> > > --- a/drivers/usb/gadget/udc/m66592-udc.c
> > > +++ b/drivers/usb/gadget/udc/m66592-udc.c
> > > @@ -1606,7 +1606,11 @@ static int m66592_probe(struct platform_device *pdev)
> > >                       ret = PTR_ERR(m66592->clk);
> > >                       goto clean_up2;
> > >               }
> > > -             clk_enable(m66592->clk);
> > > +             ret = clk_enable(m66592->clk);
> > > +             if (ret) {
> > > +                     clk_put(m66592->clk);
> > > +                     goto clean_up2;
> > > +             }
> >
> > How did you find this and how was it tested?
> >
> > thanks,
> >
> > greg k-h
> 
> We found it through a static analysis tool.

Then you need to properly document that as our documentation says it is
required, right?

thanks,

greg k-h

[PATCH] usb: gadget: m66592-udc: Add check for clk_enable()

[Mingwei Zheng]

The APP-Miner reported the missing check.
Add check for the return value of clk_enable() to catch the potential
error.

Fixes: b4822e2317e8 ("usb: gadget: m66592-udc: Convert to use module_platform_driver()")
Signed-off-by: Mingwei Zheng <zmw12306@gmail.com>
Signed-off-by: Jiasheng Jiang <jiashengjiangcool@gmail.com>
---
 drivers/usb/gadget/udc/m66592-udc.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/drivers/usb/gadget/udc/m66592-udc.c b/drivers/usb/gadget/udc/m66592-udc.c
index a938b2af0944..bf408476a24c 100644
--- a/drivers/usb/gadget/udc/m66592-udc.c
+++ b/drivers/usb/gadget/udc/m66592-udc.c
@@ -1606,7 +1606,11 @@ static int m66592_probe(struct platform_device *pdev)
 			ret = PTR_ERR(m66592->clk);
 			goto clean_up2;
 		}
-		clk_enable(m66592->clk);
+		ret = clk_enable(m66592->clk);
+		if (ret) {
+			clk_put(m66592->clk);
+			goto clean_up2;
+		}
 	}
 
 	INIT_LIST_HEAD(&m66592->gadget.ep_list);
-- 
2.34.1

Re: [PATCH] usb: gadget: m66592-udc: Add check for clk_enable()

[Mingwei Zheng]

Hi Greg,

On Mon, Dec 23, 2024 at 12:46 PM Greg KH <gregkh@linuxfoundation.org> wrote:
>
> On Tue, Dec 17, 2024 at 10:06:26PM -0500, Mingwei Zheng wrote:
> > Hi Greg,
> >
> > On Mon, Dec 16, 2024 at 2:56 AM Greg KH <gregkh@linuxfoundation.org> wrote:
> > >
> > > On Sun, Dec 15, 2024 at 03:53:58PM -0500, Mingwei Zheng wrote:
> > > > Add check for the return value of clk_enable() to catch the potential
> > > > error.
> > > >
> > > > Fixes: b4822e2317e8 ("usb: gadget: m66592-udc: Convert to use module_platform_driver()")
> > > > Signed-off-by: Mingwei Zheng <zmw12306@gmail.com>
> > > > Signed-off-by: Jiasheng Jiang <jiashengjiangcool@gmail.com>
> > >
> > > Why this order of signed-off-by lines?  Shouldn't yours be last?  Who
> > > wrote this patch?
> > >
> >
> > I listed two names because both of us co-authored this patch.
> >
> > > > ---
> > > >  drivers/usb/gadget/udc/m66592-udc.c | 6 +++++-
> > > >  1 file changed, 5 insertions(+), 1 deletion(-)
> > > >
> > > > diff --git a/drivers/usb/gadget/udc/m66592-udc.c b/drivers/usb/gadget/udc/m66592-udc.c
> > > > index a938b2af0944..bf408476a24c 100644
> > > > --- a/drivers/usb/gadget/udc/m66592-udc.c
> > > > +++ b/drivers/usb/gadget/udc/m66592-udc.c
> > > > @@ -1606,7 +1606,11 @@ static int m66592_probe(struct platform_device *pdev)
> > > >                       ret = PTR_ERR(m66592->clk);
> > > >                       goto clean_up2;
> > > >               }
> > > > -             clk_enable(m66592->clk);
> > > > +             ret = clk_enable(m66592->clk);
> > > > +             if (ret) {
> > > > +                     clk_put(m66592->clk);
> > > > +                     goto clean_up2;
> > > > +             }
> > >
> > > How did you find this and how was it tested?
> > >
> > > thanks,
> > >
> > > greg k-h
> >
> > We found it through a static analysis tool.
>
> Then you need to properly document that as our documentation says it is
> required, right?
>
> thanks,
>
> greg k-h

   I send a PATCH v2 specifying the tool name in the commit msg for
your review. Thank you so much for your suggestion!

Best,
Mingwei

Re: [PATCH] usb: gadget: m66592-udc: Add check for clk_enable()

[Greg KH]

On Tue, Dec 24, 2024 at 03:22:17PM -0500, Mingwei Zheng wrote:
> The APP-Miner reported the missing check.

What does this mean?  Please explain it better as the documentation
states you must do when using random research tools.

> Add check for the return value of clk_enable() to catch the potential
> error.
> 
> Fixes: b4822e2317e8 ("usb: gadget: m66592-udc: Convert to use module_platform_driver()")

Why is this only a Fixes tag and not also a cc: stable?  And how did you
test it?

thanks,

greg k-h