diff mbox series

[bug,report] usb: typec: ucsi: extract code to read PD caps

Message ID a3a2799c-04b3-4b8a-b808-5a118b330619@moroto.mountain (mailing list archive)
State New, archived
Headers show
Series [bug,report] usb: typec: ucsi: extract code to read PD caps | expand

Commit Message

Dan Carpenter April 11, 2024, 1:20 p.m. UTC
Hello Dmitry Baryshkov,

Commit ad4bc68bef3e ("usb: typec: ucsi: extract code to read PD
caps") from Mar 29, 2024 (linux-next), leads to the following Smatch
static checker warning:

	drivers/usb/typec/ucsi/ucsi.c:689 ucsi_get_pd_caps()
	warn: passing zero to 'ERR_PTR'

drivers/usb/typec/ucsi/ucsi.c
    680 static struct usb_power_delivery_capabilities *ucsi_get_pd_caps(struct ucsi_connector *con,
    681                                                                 enum typec_role role,
    682                                                                 bool is_partner)
    683 {
    684         struct usb_power_delivery_capabilities_desc pd_caps;
    685         int ret;
    686 
    687         ret = ucsi_get_pdos(con, role, is_partner, pd_caps.pdo);
    688         if (ret <= 0)
--> 689                 return ERR_PTR(ret);

This is returns NULL if ucsi_get_pdos() returns zero.  It's really hard
to say if this is intentional or not...  Originally, we treated error
codes and zero the same but we didn't report errors back to the user,
the code just continued silently.  Now it's reporting errors but just
continuing along if ucsi_get_pdos() returns zero.

My instinct says that we should modify ucsi_get_pdos() to not return
zero but I don't know the code...

Or, if returning NULL is intentional then it would be nice to add a
comment.

    690 
    691         if (ret < PDO_MAX_OBJECTS)
    692                 pd_caps.pdo[ret] = 0;
    693 
    694         pd_caps.role = role;
    695 
    696         return usb_power_delivery_register_capabilities(is_partner ? con->partner_pd : con->pd,
    697                                                         &pd_caps);
    698 }

regards,
dan carpenter

Comments

Dmitry Baryshkov April 11, 2024, 8:25 p.m. UTC | #1
Hello Dan,

On Thu, Apr 11, 2024 at 04:20:22PM +0300, Dan Carpenter wrote:
> Hello Dmitry Baryshkov,
> 
> Commit ad4bc68bef3e ("usb: typec: ucsi: extract code to read PD
> caps") from Mar 29, 2024 (linux-next), leads to the following Smatch
> static checker warning:
> 
> 	drivers/usb/typec/ucsi/ucsi.c:689 ucsi_get_pd_caps()
> 	warn: passing zero to 'ERR_PTR'
> 
> drivers/usb/typec/ucsi/ucsi.c
>     680 static struct usb_power_delivery_capabilities *ucsi_get_pd_caps(struct ucsi_connector *con,
>     681                                                                 enum typec_role role,
>     682                                                                 bool is_partner)
>     683 {
>     684         struct usb_power_delivery_capabilities_desc pd_caps;
>     685         int ret;
>     686 
>     687         ret = ucsi_get_pdos(con, role, is_partner, pd_caps.pdo);
>     688         if (ret <= 0)
> --> 689                 return ERR_PTR(ret);
> 
> This is returns NULL if ucsi_get_pdos() returns zero.  It's really hard
> to say if this is intentional or not...  Originally, we treated error
> codes and zero the same but we didn't report errors back to the user,
> the code just continued silently.  Now it's reporting errors but just
> continuing along if ucsi_get_pdos() returns zero.

The code is correct. The calling function checks that the result of
ucsi_get_pd_caps() is an ERR code and if that's not the case, assigns
PD capabilites to the storage at the connector. If ucsi_get_pdos()
returns 0, there are no PD objects, nothing to create capabilites for.
Thus the function correctly returns NULL.

If you think it is better to be explicit, I can send either an explicit
fixup `if (!ret) return NULL;` or just a comment that we should return
NULL if there are no PDOs.

> 
> My instinct says that we should modify ucsi_get_pdos() to not return
> zero but I don't know the code...
> 
> diff --git a/drivers/usb/typec/ucsi/ucsi.c b/drivers/usb/typec/ucsi/ucsi.c
> index 8abde8db6bcf..427b7610a074 100644
> --- a/drivers/usb/typec/ucsi/ucsi.c
> +++ b/drivers/usb/typec/ucsi/ucsi.c
> @@ -650,6 +650,8 @@ static int ucsi_get_pdos(struct ucsi_connector *con, enum typec_role role,
>  		return ret;
>  
>  	num_pdos = ret / sizeof(u32); /* number of bytes to 32-bit PDOs */
> +	if (num_pdos == 0)
> +		return -EINVAL;
>  	if (num_pdos < UCSI_MAX_PDOS)
>  		return num_pdos;
>  
> Or, if returning NULL is intentional then it would be nice to add a
> comment.
> 
>     690 
>     691         if (ret < PDO_MAX_OBJECTS)
>     692                 pd_caps.pdo[ret] = 0;
>     693 
>     694         pd_caps.role = role;
>     695 
>     696         return usb_power_delivery_register_capabilities(is_partner ? con->partner_pd : con->pd,
>     697                                                         &pd_caps);
>     698 }
> 
> regards,
> dan carpenter
Dan Carpenter April 14, 2024, 11:52 a.m. UTC | #2
On Thu, Apr 11, 2024 at 11:25:50PM +0300, Dmitry Baryshkov wrote:
> Hello Dan,
> 
> On Thu, Apr 11, 2024 at 04:20:22PM +0300, Dan Carpenter wrote:
> > Hello Dmitry Baryshkov,
> > 
> > Commit ad4bc68bef3e ("usb: typec: ucsi: extract code to read PD
> > caps") from Mar 29, 2024 (linux-next), leads to the following Smatch
> > static checker warning:
> > 
> > 	drivers/usb/typec/ucsi/ucsi.c:689 ucsi_get_pd_caps()
> > 	warn: passing zero to 'ERR_PTR'
> > 
> > drivers/usb/typec/ucsi/ucsi.c
> >     680 static struct usb_power_delivery_capabilities *ucsi_get_pd_caps(struct ucsi_connector *con,
> >     681                                                                 enum typec_role role,
> >     682                                                                 bool is_partner)
> >     683 {
> >     684         struct usb_power_delivery_capabilities_desc pd_caps;
> >     685         int ret;
> >     686 
> >     687         ret = ucsi_get_pdos(con, role, is_partner, pd_caps.pdo);
> >     688         if (ret <= 0)
> > --> 689                 return ERR_PTR(ret);
> > 
> > This is returns NULL if ucsi_get_pdos() returns zero.  It's really hard
> > to say if this is intentional or not...  Originally, we treated error
> > codes and zero the same but we didn't report errors back to the user,
> > the code just continued silently.  Now it's reporting errors but just
> > continuing along if ucsi_get_pdos() returns zero.
> 
> The code is correct. The calling function checks that the result of
> ucsi_get_pd_caps() is an ERR code and if that's not the case, assigns
> PD capabilites to the storage at the connector. If ucsi_get_pdos()
> returns 0, there are no PD objects, nothing to create capabilites for.
> Thus the function correctly returns NULL.
> 
> If you think it is better to be explicit, I can send either an explicit
> fixup `if (!ret) return NULL;` or just a comment that we should return
> NULL if there are no PDOs.
> 

Just add a comment.  This static checker rule is going to have false
positives, and I did think it might be a false positive, but it just
wasn't totally obvious.

regards,
dan carpenter
diff mbox series

Patch

diff --git a/drivers/usb/typec/ucsi/ucsi.c b/drivers/usb/typec/ucsi/ucsi.c
index 8abde8db6bcf..427b7610a074 100644
--- a/drivers/usb/typec/ucsi/ucsi.c
+++ b/drivers/usb/typec/ucsi/ucsi.c
@@ -650,6 +650,8 @@  static int ucsi_get_pdos(struct ucsi_connector *con, enum typec_role role,
 		return ret;
 
 	num_pdos = ret / sizeof(u32); /* number of bytes to 32-bit PDOs */
+	if (num_pdos == 0)
+		return -EINVAL;
 	if (num_pdos < UCSI_MAX_PDOS)
 		return num_pdos;