[PATCH] soc: ti: pruss: fix return check for of_device_get_match_data()

[PATCH] soc: ti: pruss: fix return check for of_device_get_match_data()

[Luo Meng]

of_device_get_match_data() maybe return NULL pointer which cannot
be checked with IS_ERR.

Fixes: ba59c9b43c86 ("soc: ti: pruss: support CORECLK_MUX and IEPCLK_MUX")
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Luo Meng <luomeng12@huawei.com>
---
 drivers/soc/ti/pruss.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/soc/ti/pruss.c b/drivers/soc/ti/pruss.c
index cc0b4ad7a3d3..582f48051c30 100644
--- a/drivers/soc/ti/pruss.c
+++ b/drivers/soc/ti/pruss.c
@@ -126,7 +126,7 @@ static int pruss_clk_init(struct pruss *pruss, struct device_node *cfg_node)
 	int ret = 0;
 
 	data = of_device_get_match_data(dev);
-	if (IS_ERR(data))
+	if (!data)
 		return -ENODEV;
 
 	clks_np = of_get_child_by_name(cfg_node, "clocks");
@@ -175,7 +175,7 @@ static int pruss_probe(struct platform_device *pdev)
 	const char *mem_names[PRUSS_MEM_MAX] = { "dram0", "dram1", "shrdram2" };
 
 	data = of_device_get_match_data(&pdev->dev);
-	if (IS_ERR(data)) {
+	if (!data) {
 		dev_err(dev, "missing private data\n");
 		return -ENODEV;
 	}
-- 
2.25.4


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

Re: [PATCH] soc: ti: pruss: fix return check for of_device_get_match_data()

[Grzegorz Jaszczyk]

Hi

On Tue, 10 Nov 2020 at 02:31, Luo Meng <luomeng12@huawei.com> wrote:
>
> of_device_get_match_data() maybe return NULL pointer which cannot
> be checked with IS_ERR.
>
> Fixes: ba59c9b43c86 ("soc: ti: pruss: support CORECLK_MUX and IEPCLK_MUX")
> Reported-by: Hulk Robot <hulkci@huawei.com>
> Signed-off-by: Luo Meng <luomeng12@huawei.com>
> ---
>  drivers/soc/ti/pruss.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/soc/ti/pruss.c b/drivers/soc/ti/pruss.c
> index cc0b4ad7a3d3..582f48051c30 100644
> --- a/drivers/soc/ti/pruss.c
> +++ b/drivers/soc/ti/pruss.c
> @@ -126,7 +126,7 @@ static int pruss_clk_init(struct pruss *pruss, struct device_node *cfg_node)
>         int ret = 0;
>
>         data = of_device_get_match_data(dev);
> -       if (IS_ERR(data))
> +       if (!data)
>                 return -ENODEV;
>
>         clks_np = of_get_child_by_name(cfg_node, "clocks");
> @@ -175,7 +175,7 @@ static int pruss_probe(struct platform_device *pdev)
>         const char *mem_names[PRUSS_MEM_MAX] = { "dram0", "dram1", "shrdram2" };
>
>         data = of_device_get_match_data(&pdev->dev);
> -       if (IS_ERR(data)) {
> +       if (!data) {

Thank you for your interest but with your patch the driver will not be
functional anymore for all devices without match data (e.g.
"ti,am3356-pruss") which is not what we want. The proper fix was
already posted end of October:
https://patchwork.kernel.org/project/linux-arm-kernel/patch/20201026144943.30821-1-grzegorz.jaszczyk@linaro.org/

Thank you,
Grzegorz

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

Re: [PATCH] soc: ti: pruss: fix return check for of_device_get_match_data()

[Lee Jones]

On Tue, 10 Nov 2020, Grzegorz Jaszczyk wrote:

> Hi
> 
> On Tue, 10 Nov 2020 at 02:31, Luo Meng <luomeng12@huawei.com> wrote:
> >
> > of_device_get_match_data() maybe return NULL pointer which cannot
> > be checked with IS_ERR.
> >
> > Fixes: ba59c9b43c86 ("soc: ti: pruss: support CORECLK_MUX and IEPCLK_MUX")
> > Reported-by: Hulk Robot <hulkci@huawei.com>
> > Signed-off-by: Luo Meng <luomeng12@huawei.com>
> > ---
> >  drivers/soc/ti/pruss.c | 4 ++--
> >  1 file changed, 2 insertions(+), 2 deletions(-)
> >
> > diff --git a/drivers/soc/ti/pruss.c b/drivers/soc/ti/pruss.c
> > index cc0b4ad7a3d3..582f48051c30 100644
> > --- a/drivers/soc/ti/pruss.c
> > +++ b/drivers/soc/ti/pruss.c
> > @@ -126,7 +126,7 @@ static int pruss_clk_init(struct pruss *pruss, struct device_node *cfg_node)
> >         int ret = 0;
> >
> >         data = of_device_get_match_data(dev);
> > -       if (IS_ERR(data))
> > +       if (!data)
> >                 return -ENODEV;

Also this would miss any errors, leading to dereference issues.

I think what you meant was IS_ERR_OR_NULL(), but this would still be
incorrect for the reasons Grzegorz mentions below.

> >         clks_np = of_get_child_by_name(cfg_node, "clocks");
> > @@ -175,7 +175,7 @@ static int pruss_probe(struct platform_device *pdev)
> >         const char *mem_names[PRUSS_MEM_MAX] = { "dram0", "dram1", "shrdram2" };
> >
> >         data = of_device_get_match_data(&pdev->dev);
> > -       if (IS_ERR(data)) {
> > +       if (!data) {
> 
> Thank you for your interest but with your patch the driver will not be
> functional anymore for all devices without match data (e.g.
> "ti,am3356-pruss") which is not what we want. The proper fix was
> already posted end of October:
> https://patchwork.kernel.org/project/linux-arm-kernel/patch/20201026144943.30821-1-grzegorz.jaszczyk@linaro.org/
> 
> Thank you,
> Grzegorz

-- 
Lee Jones [李琼斯]
Senior Technical Lead - Developer Services
Linaro.org │ Open source software for Arm SoCs
Follow Linaro: Facebook | Twitter | Blog

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

Re: [PATCH] soc: ti: pruss: fix return check for of_device_get_match_data()

[Suman Anna]

Hi Luo,

On 11/10/20 3:54 AM, Lee Jones wrote:
> On Tue, 10 Nov 2020, Grzegorz Jaszczyk wrote:
> 
>> Hi
>>
>> On Tue, 10 Nov 2020 at 02:31, Luo Meng <luomeng12@huawei.com> wrote:
>>>
>>> of_device_get_match_data() maybe return NULL pointer which cannot
>>> be checked with IS_ERR.
>>>
>>> Fixes: ba59c9b43c86 ("soc: ti: pruss: support CORECLK_MUX and IEPCLK_MUX")
>>> Reported-by: Hulk Robot <hulkci@huawei.com>
>>> Signed-off-by: Luo Meng <luomeng12@huawei.com>
>>> ---
>>>  drivers/soc/ti/pruss.c | 4 ++--
>>>  1 file changed, 2 insertions(+), 2 deletions(-)
>>>
>>> diff --git a/drivers/soc/ti/pruss.c b/drivers/soc/ti/pruss.c
>>> index cc0b4ad7a3d3..582f48051c30 100644
>>> --- a/drivers/soc/ti/pruss.c
>>> +++ b/drivers/soc/ti/pruss.c
>>> @@ -126,7 +126,7 @@ static int pruss_clk_init(struct pruss *pruss, struct device_node *cfg_node)
>>>         int ret = 0;
>>>
>>>         data = of_device_get_match_data(dev);
>>> -       if (IS_ERR(data))
>>> +       if (!data)
>>>                 return -ENODEV;
> 
> Also this would miss any errors, leading to dereference issues.
> 
> I think what you meant was IS_ERR_OR_NULL(), but this would still be
> incorrect for the reasons Grzegorz mentions below.
> 
>>>         clks_np = of_get_child_by_name(cfg_node, "clocks");
>>> @@ -175,7 +175,7 @@ static int pruss_probe(struct platform_device *pdev)
>>>         const char *mem_names[PRUSS_MEM_MAX] = { "dram0", "dram1", "shrdram2" };
>>>
>>>         data = of_device_get_match_data(&pdev->dev);
>>> -       if (IS_ERR(data)) {
>>> +       if (!data) {
>>
>> Thank you for your interest but with your patch the driver will not be
>> functional anymore for all devices without match data (e.g.
>> "ti,am3356-pruss") which is not what we want. The proper fix was
>> already posted end of October:
>> https://patchwork.kernel.org/project/linux-arm-kernel/patch/20201026144943.30821-1-grzegorz.jaszczyk@linaro.org/

Wei Yongjun from Huawei has posted an identical patch originally [1], which is
what the above patch fixes. Please use Greg's updated patch instead.

regards
Suman

[1]
https://patchwork.kernel.org/project/linux-arm-kernel/patch/20200915075732.100804-1-weiyongjun1@huawei.com/

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel