Re: [PATCH] media: atomisp: silence "dubious: !x | !y" warning

[Luc Van Oostenryck]

On Sat, Apr 17, 2021 at 09:31:32PM +, David Laight wrote:
> > > Upon running sparse, "warning: dubious: !x | !y" is brought to notice
> > > for this file.  Logical and bitwise OR are basically the same in this
> > > context so it doesn't cause a runtime bug.  But let's change it to
> > > logical OR to make it cleaner and silence the Sparse warning.
> 
> The old code is very likely to by slightly more efficient.
> 
> It may not matter here, but it might in a really hot path.
> 
> Since !x | !y and !x || !y always have the same value
> why is sparse complaining at all.

They both will have the same value here and any half-decent
compiler know that and thus generate the same code, so no
worries about efficiency.

Sparse complains because the programmer's intention is not clear.
Was a boolean context or a bitwise context that was meant?
Maybe '||' was meant and the RHS had to be short cut?
Maybe what was meant was '~x | ~y'?

-- Luc
___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel

Re: [PATCH] staging: rtl8192e, rtl8192u: use correct notation to define pointer

[Luc Van Oostenryck]

On Tue, Oct 27, 2020 at 12:23:03PM +0100, Greg KH wrote:
> 
> Are you sure these changes are correct?  This isn't just a list of
> structures after this at the end of the structure?
> 
> Please look at commit 5979afa2c4d1 ("staging: Replace zero-length array
> with flexible-array member") which made most of these flexible arrays.
> 
> This is not a pointer, it really is an array, I think sparse is really
> wrong here, be careful.

Sparse's warning is not about changing the definition of this member
as if it was the argument of a function. It's about how can you use
an array of structure when this structure has a flexible member.
It's a recent warning, added purposely to catch this.
See 
https://lore.kernel.org/r/CAHk-=wgjz05ap8vqdzwdwjvwvtzioytc6cnnb8gneqzenfm...@mail.gmail.com

-- Luc
___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel

[PATCH] staging: ks7010: fix ks_wlan_start_xmit()'s return type

[Luc Van Oostenryck]

The method ndo_start_xmit() is defined as returning an 'netdev_tx_t',
which is a typedef for an enum type, but the implementation in this
driver returns an 'int'.

Fix this by returning 'netdev_tx_t' in this driver too and
usind 'NETDEV_TX_OK' instead of 0 accordingly.

Signed-off-by: Luc Van Oostenryck 
---
 drivers/staging/ks7010/ks_wlan_net.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/staging/ks7010/ks_wlan_net.c 
b/drivers/staging/ks7010/ks_wlan_net.c
index 211dd4a11cac..334ae1ded684 100644
--- a/drivers/staging/ks7010/ks_wlan_net.c
+++ b/drivers/staging/ks7010/ks_wlan_net.c
@@ -46,7 +46,7 @@ struct wep_key {
  */
 static int ks_wlan_open(struct net_device *dev);
 static void ks_wlan_tx_timeout(struct net_device *dev, unsigned int txqueue);
-static int ks_wlan_start_xmit(struct sk_buff *skb, struct net_device *dev);
+static netdev_tx_t ks_wlan_start_xmit(struct sk_buff *skb, struct net_device 
*dev);
 static int ks_wlan_close(struct net_device *dev);
 static void ks_wlan_set_rx_mode(struct net_device *dev);
 static struct net_device_stats *ks_wlan_get_stats(struct net_device *dev);
@@ -2511,7 +2511,7 @@ void ks_wlan_tx_timeout(struct net_device *dev, unsigned 
int txqueue)
 }
 
 static
-int ks_wlan_start_xmit(struct sk_buff *skb, struct net_device *dev)
+netdev_tx_t ks_wlan_start_xmit(struct sk_buff *skb, struct net_device *dev)
 {
struct ks_wlan_private *priv = netdev_priv(dev);
int ret;
-- 
2.27.0

___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel

[PATCH] staging/wilc1000: let wilc_mac_xmit() to NETDEV_TX_OK

[Luc Van Oostenryck]

The method ndo_start_xmit() is defined as returning an 'netdev_tx_t',
which is a typedef for an enum type defining 'NETDEV_TX_OK' but this
driver returns '0' instead of 'NETDEV_TX_OK'.

Fix this by returning ''NETDEV_TX_OK' instead of 0.

Signed-off-by: Luc Van Oostenryck 
---
 drivers/staging/wilc1000/netdev.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/staging/wilc1000/netdev.c 
b/drivers/staging/wilc1000/netdev.c
index fda0ab97b02c..be3ae5486f44 100644
--- a/drivers/staging/wilc1000/netdev.c
+++ b/drivers/staging/wilc1000/netdev.c
@@ -678,14 +678,14 @@ netdev_tx_t wilc_mac_xmit(struct sk_buff *skb, struct 
net_device *ndev)
 
if (skb->dev != ndev) {
netdev_err(ndev, "Packet not destined to this device\n");
-   return 0;
+   return NETDEV_TX_OK;
}
 
tx_data = kmalloc(sizeof(*tx_data), GFP_ATOMIC);
if (!tx_data) {
dev_kfree_skb(skb);
netif_wake_queue(ndev);
-   return 0;
+   return NETDEV_TX_OK;
}
 
tx_data->buff = skb->data;
@@ -710,7 +710,7 @@ netdev_tx_t wilc_mac_xmit(struct sk_buff *skb, struct 
net_device *ndev)
srcu_read_unlock(>srcu, srcu_idx);
}
 
-   return 0;
+   return NETDEV_TX_OK;
 }
 
 static int wilc_mac_close(struct net_device *ndev)
-- 
2.27.0

___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel

Re: [PATCH] Removing ununsed variable int lo, hi, int data and int i from comedi/drivers/dt2814.c.

[Luc Van Oostenryck]

On Sun, May 24, 2020 at 07:02:20PM +0530, GAURAV PATHAK wrote:
> 
> For few drivers, I am getting:
> 
> CC [M]  drivers/staging/vt6656/key.o
>   CHECK   drivers/staging/vt6656/rf.c
> /usr/lib/gcc/x86_64-linux-gnu/7/include/stddef.h:417:9: warning: preprocessor 
> token offsetof redefined
> ./include/linux/stddef.h:17:9: this was the original definition
> 
> but I guess these are false positive, as #undef offsetof is already present 
> in ./include/linux/stddef.h.

But there is no #undef in /usr/lib/gcc/x86_64-linux-gnu/7/include/stddef.h
which seems to be sometimes included after Linux's include/linux/stddef.h

-- Luc 
___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel

Re: [PATCH] Removing ununsed variable int lo, hi, int data and int i from comedi/drivers/dt2814.c.

[Luc Van Oostenryck]

On Sun, May 24, 2020 at 06:19:22PM +0530, GAURAV PATHAK wrote:
> Thanks a lot for your reply.
> I am trying to run sparse on drivers/staging directory,
> but I am not getting any useful warnings.
> Steps that I did:
> 
> - Referred https://www.kernel.org/doc/man-pages/linux-next.html to get
> linux-next latest tag next-20200522.
> - After that executed following commands:
>   - make mrproper; make clean; make distclean
>   - make defconfig
>   - make menuconfig (Enabled all Staging Drivers with M)
>   - make modules_prepare
>   - make C=1 M=drivers/staging (Also tried individual directories
> inside staging directory e.g. make C=1 
> M=drivers/staging/comedi/drivers)
> 
> I am not getting any warning, I have attached the output in text format.
> 
> Any pointers for finding warning using sparse in drivers/staging
> directory will be helpful. I really appretiate your guidance in this
> regard.

Well it seems that most staging drivers have no sparse warnings at all.
I only see somes in the following staging drivers:
drivers/staging/isdn/
drivers/staging/kpc2000/
drivers/staging/uwb/
drivers/staging/vc04_services/
drivers/staging/wfx/
drivers/staging/wusbcore/

Your commands seems to be good but I would advice you to use 'C=2'
instead of 'C=1' in order for sparse to effectively run on each
input file even when the corresponding .o file already exists.

The simplest to use is:
- make allmodconfig
- make -j8 drivers/staging/
This will just compile all the files without running sparse.
So all warnings will be from the compiler.
The you can run:
- make C=2 drivers/staging/
which will only run sparse (and will thus be much faster).
 
Best regards,
-- Luc
___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel

Re: [PATCH] Removing ununsed variable int lo, hi, int data and int i from comedi/drivers/dt2814.c.

[Luc Van Oostenryck]

Hi,

On Sun, May 24, 2020 at 05:06:18PM +0530, Gaurav Pathak wrote:
> Silence following sparse warning:
> drivers/staging/comedi/drivers/dt2814.c: In function ‘dt2814_interrupt’:
> drivers/staging/comedi/drivers/dt2814.c:193:6: warning: variable ‘data’ set 
> but not used [-Wunused-but-set-variable]
>   int data;
>   ^~~~
> drivers/staging/comedi/drivers/dt2814.c: In function ‘dt2814_attach’:
> drivers/staging/comedi/drivers/dt2814.c:232:6: warning: variable ‘i’ set but 
> not used [-Wunused-but-set-variable]
>   int i;
>   ^

These warnings are not from sparse but simply from the compiler.
 
>  static irqreturn_t dt2814_interrupt(int irq, void *d)
>  {
> - int lo, hi;
>   struct comedi_device *dev = d;
>   struct dt2814_private *devpriv = dev->private;
>   struct comedi_subdevice *s = dev->read_subdev;
> - int data;
>  
>   if (!dev->attached) {
>   dev_err(dev->class_dev, "spurious interrupt\n");
>   return IRQ_HANDLED;
>   }
>  
> - hi = inb(dev->iobase + DT2814_DATA);
> - lo = inb(dev->iobase + DT2814_DATA);
> -
> - data = (hi << 4) | (lo >> 4);

OK, 'data' is unused but are these 2 'inb(dev->iobase + DT2814_DATA)'
needed or not? I would guess that they're needed but I don't know
this hardware.

> @@ -241,8 +233,6 @@ static int dt2814_attach(struct comedi_device *dev, 
> struct comedi_devconfig *it)
>   dev_err(dev->class_dev, "reset error (fatal)\n");
>   return -EIO;
>   }
> - i = inb(dev->iobase + DT2814_DATA);
> - i = inb(dev->iobase + DT2814_DATA);

Same here.

-- Luc
___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel

Re: [PATCH] staging: wilc1000: fix cast to restricted __le32

[Luc Van Oostenryck]

On Mon, Jan 07, 2019 at 03:54:37PM +0300, Dan Carpenter wrote:
> I'm sort of surprised that the zero day bot didn't complain about this.
> It probably should.
> 
> The endianness checking is not turned on by default in Sparse.  You have
> to do "make C=1 CF=-D__CHECK_ENDIAN__" to turn it on.

Bitwise (and thus endian) checking is enabled by default in the kernel
since 05de97003 ("linux/types.h: enable endian checks for all sparse builds")
included in v4.10-rc1. Since then they make more than half of sparse
warnings (in a x86-64 allyesconfig).

Best regards,
-- Luc
___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel

Re: [PATCH] binder: fix sparse warnings on locking context

[Luc Van Oostenryck]

On Mon, Dec 03, 2018 at 12:24:54PM -0800, Todd Kjos wrote:
> Add __acquire()/__release() annnotations to fix warnings
> in sparse context checking
> 
> There is one case where the warning was due to a lack of
> a "default:" case in a switch statement where a lock was
> being released in each of the cases, so the default
> case was added.
> 
> Signed-off-by: Todd Kjos 
> ---
>  drivers/android/binder.c   | 43 +-
>  drivers/android/binder_alloc.c |  1 +
>  2 files changed, 43 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/android/binder.c b/drivers/android/binder.c
> index 9f1000d2a40c7..9f2059d24ae2d 100644
> --- a/drivers/android/binder.c
> +++ b/drivers/android/binder.c
> @@ -751,12 +757,16 @@ _binder_node_unlock(struct binder_node *node, int line)
>  #define binder_node_inner_lock(node) _binder_node_inner_lock(node, __LINE__)
>  static void
>  _binder_node_inner_lock(struct binder_node *node, int line)
> + __acquires(>lock) __acquires(>proc->inner_lock)
>  {
>   binder_debug(BINDER_DEBUG_SPINLOCKS,
>"%s: line=%d\n", __func__, line);
>   spin_lock(>lock);
>   if (node->proc)
>   binder_inner_proc_lock(node->proc);
> + else
> + /* annotation for sparse */
> + __acquire(>proc->inner_lock);

This one is questionnable because:
1) if !node->proc, then '>proc->inner_lock' is not acquired
   since it doesn't even exist.
2) OTOH, the function can't have the annotation 100% right because
   it semantics allows unbalanced locking depending on node->proc
   being null or not.
But I see very well the intent and maybe it's a right solution.
I dunno.

Same for most of the following ones.


Best regards,
-- Luc Van Oostenryck
___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel

[PATCH v2] staging/wilc1000: fix wilc_mac_xmit()'s return type

[Luc Van Oostenryck]

The method ndo_start_xmit() is defined as returning an 'netdev_tx_t',
which is a typedef for an enum type, but the implementation in this
driver returns an 'int'.

Fix this by returning 'netdev_tx_t' in this driver too
and accordingly change 'return 0;' by 'return NETDEV_TX_OK;'

Signed-off-by: Luc Van Oostenryck <luc.vanoostenr...@gmail.com>
---
 drivers/staging/wilc1000/linux_wlan.c | 8 
 drivers/staging/wilc1000/wilc_wlan.h  | 2 +-
 2 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/staging/wilc1000/linux_wlan.c 
b/drivers/staging/wilc1000/linux_wlan.c
index 38a83bd31..d95b9750b 100644
--- a/drivers/staging/wilc1000/linux_wlan.c
+++ b/drivers/staging/wilc1000/linux_wlan.c
@@ -941,7 +941,7 @@ static void linux_wlan_tx_complete(void *priv, int status)
kfree(pv_data);
 }
 
-int wilc_mac_xmit(struct sk_buff *skb, struct net_device *ndev)
+netdev_tx_t wilc_mac_xmit(struct sk_buff *skb, struct net_device *ndev)
 {
struct wilc_vif *vif;
struct tx_complete_data *tx_data = NULL;
@@ -956,14 +956,14 @@ int wilc_mac_xmit(struct sk_buff *skb, struct net_device 
*ndev)
 
if (skb->dev != ndev) {
netdev_err(ndev, "Packet not destined to this device\n");
-   return 0;
+   return NETDEV_TX_OK;
}
 
tx_data = kmalloc(sizeof(*tx_data), GFP_ATOMIC);
if (!tx_data) {
dev_kfree_skb(skb);
netif_wake_queue(ndev);
-   return 0;
+   return NETDEV_TX_OK;
}
 
tx_data->buff = skb->data;
@@ -994,7 +994,7 @@ int wilc_mac_xmit(struct sk_buff *skb, struct net_device 
*ndev)
netif_stop_queue(wilc->vif[1]->ndev);
}
 
-   return 0;
+   return NETDEV_TX_OK;
 }
 
 static int wilc_mac_close(struct net_device *ndev)
diff --git a/drivers/staging/wilc1000/wilc_wlan.h 
b/drivers/staging/wilc1000/wilc_wlan.h
index fa157a67b..d3b5a6b38 100644
--- a/drivers/staging/wilc1000/wilc_wlan.h
+++ b/drivers/staging/wilc1000/wilc_wlan.h
@@ -298,7 +298,7 @@ void wilc_chip_sleep_manually(struct wilc *wilc);
 
 void wilc_enable_tcp_ack_filter(bool value);
 int wilc_wlan_get_num_conn_ifcs(struct wilc *wilc);
-int wilc_mac_xmit(struct sk_buff *skb, struct net_device *dev);
+netdev_tx_t wilc_mac_xmit(struct sk_buff *skb, struct net_device *dev);
 
 void WILC_WFI_p2p_rx(struct net_device *dev, u8 *buff, u32 size);
 void host_wakeup_notify(struct wilc *wilc);
-- 
2.17.0

___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel

[PATCH] staging/wilc1000: fix wilc_mac_xmit()'s return type

[Luc Van Oostenryck]

The method ndo_start_xmit() is defined as returning an 'netdev_tx_t',
which is a typedef for an enum type, but the implementation in this
driver returns an 'int'.

Fix this by returning 'netdev_tx_t' in this driver too.

Signed-off-by: Luc Van Oostenryck <luc.vanoostenr...@gmail.com>
---
 drivers/staging/wilc1000/linux_wlan.c | 2 +-
 drivers/staging/wilc1000/wilc_wlan.h  | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/staging/wilc1000/linux_wlan.c 
b/drivers/staging/wilc1000/linux_wlan.c
index 38a83bd31..e06aaed5e 100644
--- a/drivers/staging/wilc1000/linux_wlan.c
+++ b/drivers/staging/wilc1000/linux_wlan.c
@@ -941,7 +941,7 @@ static void linux_wlan_tx_complete(void *priv, int status)
kfree(pv_data);
 }
 
-int wilc_mac_xmit(struct sk_buff *skb, struct net_device *ndev)
+netdev_tx_t wilc_mac_xmit(struct sk_buff *skb, struct net_device *ndev)
 {
struct wilc_vif *vif;
struct tx_complete_data *tx_data = NULL;
diff --git a/drivers/staging/wilc1000/wilc_wlan.h 
b/drivers/staging/wilc1000/wilc_wlan.h
index fa157a67b..d3b5a6b38 100644
--- a/drivers/staging/wilc1000/wilc_wlan.h
+++ b/drivers/staging/wilc1000/wilc_wlan.h
@@ -298,7 +298,7 @@ void wilc_chip_sleep_manually(struct wilc *wilc);
 
 void wilc_enable_tcp_ack_filter(bool value);
 int wilc_wlan_get_num_conn_ifcs(struct wilc *wilc);
-int wilc_mac_xmit(struct sk_buff *skb, struct net_device *dev);
+netdev_tx_t wilc_mac_xmit(struct sk_buff *skb, struct net_device *dev);
 
 void WILC_WFI_p2p_rx(struct net_device *dev, u8 *buff, u32 size);
 void host_wakeup_notify(struct wilc *wilc);
-- 
2.17.0

___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel

[PATCH] staging: wlan-ng: fix p80211knetdev_hard_start_xmit()'s return type

[Luc Van Oostenryck]

The method ndo_start_xmit() is defined as returning an 'netdev_tx_t',
which is a typedef for an enum type, but the implementation in this
driver returns an 'int'.

Fix this by returning 'netdev_tx_t' in this driver too.

Signed-off-by: Luc Van Oostenryck <luc.vanoostenr...@gmail.com>
---
 drivers/staging/wlan-ng/p80211netdev.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/staging/wlan-ng/p80211netdev.c 
b/drivers/staging/wlan-ng/p80211netdev.c
index ec9cc00ee..710ce8394 100644
--- a/drivers/staging/wlan-ng/p80211netdev.c
+++ b/drivers/staging/wlan-ng/p80211netdev.c
@@ -95,7 +95,7 @@
 static int p80211knetdev_init(struct net_device *netdev);
 static int p80211knetdev_open(struct net_device *netdev);
 static int p80211knetdev_stop(struct net_device *netdev);
-static int p80211knetdev_hard_start_xmit(struct sk_buff *skb,
+static netdev_tx_t p80211knetdev_hard_start_xmit(struct sk_buff *skb,
 struct net_device *netdev);
 static void p80211knetdev_set_multicast_list(struct net_device *dev);
 static int p80211knetdev_do_ioctl(struct net_device *dev, struct ifreq *ifr,
@@ -321,7 +321,7 @@ static void p80211netdev_rx_bh(unsigned long arg)
  * zero on success, non-zero on failure.
  *
  */
-static int p80211knetdev_hard_start_xmit(struct sk_buff *skb,
+static netdev_tx_t p80211knetdev_hard_start_xmit(struct sk_buff *skb,
 struct net_device *netdev)
 {
int result = 0;
-- 
2.17.0

___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel

[PATCH] staging: rtl8723bs: fix rtw_cfg80211_monitor_if_xmit_entry()'s return type

[Luc Van Oostenryck]

The method ndo_start_xmit() is defined as returning an 'netdev_tx_t',
which is a typedef for an enum type, but the implementation in this
driver returns an 'int'.

Fix this by returning 'netdev_tx_t' in this driver too.

Signed-off-by: Luc Van Oostenryck <luc.vanoostenr...@gmail.com>
---
 drivers/staging/rtl8723bs/os_dep/ioctl_cfg80211.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/staging/rtl8723bs/os_dep/ioctl_cfg80211.c 
b/drivers/staging/rtl8723bs/os_dep/ioctl_cfg80211.c
index 46bc2e512..75d8a16f5 100644
--- a/drivers/staging/rtl8723bs/os_dep/ioctl_cfg80211.c
+++ b/drivers/staging/rtl8723bs/os_dep/ioctl_cfg80211.c
@@ -2472,7 +2472,7 @@ static int rtw_cfg80211_monitor_if_close(struct 
net_device *ndev)
return ret;
 }
 
-static int rtw_cfg80211_monitor_if_xmit_entry(struct sk_buff *skb, struct 
net_device *ndev)
+static netdev_tx_t rtw_cfg80211_monitor_if_xmit_entry(struct sk_buff *skb, 
struct net_device *ndev)
 {
int ret = 0;
int rtap_len;
-- 
2.17.0

___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel

[PATCH] staging: ks7010: fix ks_wlan_start_xmit()'s return type

[Luc Van Oostenryck]

The method ndo_start_xmit() is defined as returning an 'netdev_tx_t',
which is a typedef for an enum type, but the implementation in this
driver returns an 'int'.

Fix this by returning 'netdev_tx_t' in this driver too.

Signed-off-by: Luc Van Oostenryck <luc.vanoostenr...@gmail.com>
---
 drivers/staging/ks7010/ks_wlan_net.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/staging/ks7010/ks_wlan_net.c 
b/drivers/staging/ks7010/ks_wlan_net.c
index 9078e13b0..57412caac 100644
--- a/drivers/staging/ks7010/ks_wlan_net.c
+++ b/drivers/staging/ks7010/ks_wlan_net.c
@@ -71,7 +71,7 @@ static const struct iw_handler_def ks_wlan_handler_def;
  */
 static int ks_wlan_open(struct net_device *dev);
 static void ks_wlan_tx_timeout(struct net_device *dev);
-static int ks_wlan_start_xmit(struct sk_buff *skb, struct net_device *dev);
+static netdev_tx_t ks_wlan_start_xmit(struct sk_buff *skb, struct net_device 
*dev);
 static int ks_wlan_close(struct net_device *dev);
 static void ks_wlan_set_multicast_list(struct net_device *dev);
 static struct net_device_stats *ks_wlan_get_stats(struct net_device *dev);
@@ -2772,7 +2772,7 @@ void ks_wlan_tx_timeout(struct net_device *dev)
 }
 
 static
-int ks_wlan_start_xmit(struct sk_buff *skb, struct net_device *dev)
+netdev_tx_t ks_wlan_start_xmit(struct sk_buff *skb, struct net_device *dev)
 {
struct ks_wlan_private *priv = netdev_priv(dev);
int ret;
-- 
2.17.0

___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel

[PATCH] staging: visornic: fix visornic_xmit()'s return type

[Luc Van Oostenryck]

The method ndo_start_xmit() is defined as returning an 'netdev_tx_t',
which is a typedef for an enum type, but the implementation in this
driver returns an 'int'.

Fix this by returning 'netdev_tx_t' in this driver too.

Signed-off-by: Luc Van Oostenryck <luc.vanoostenr...@gmail.com>
---
 drivers/staging/unisys/visornic/visornic_main.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/staging/unisys/visornic/visornic_main.c 
b/drivers/staging/unisys/visornic/visornic_main.c
index 92dceb557..f734e835b 100644
--- a/drivers/staging/unisys/visornic/visornic_main.c
+++ b/drivers/staging/unisys/visornic/visornic_main.c
@@ -849,7 +849,7 @@ static bool vnic_hit_low_watermark(struct visornic_devdata 
*devdata,
  *
  * Return: NETDEV_TX_OK.
  */
-static int visornic_xmit(struct sk_buff *skb, struct net_device *netdev)
+static netdev_tx_t visornic_xmit(struct sk_buff *skb, struct net_device 
*netdev)
 {
struct visornic_devdata *devdata;
int len, firstfraglen, padlen;
-- 
2.17.0

___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel

[PATCH] staging: gdm724x: fix gdm_lte_tx()'s return type

[Luc Van Oostenryck]

The method ndo_start_xmit() is defined as returning an 'netdev_tx_t',
which is a typedef for an enum type, but the implementation in this
driver returns an 'int'.

Fix this by returning 'netdev_tx_t' in this driver too.

Signed-off-by: Luc Van Oostenryck <luc.vanoostenr...@gmail.com>
---
 drivers/staging/gdm724x/gdm_lte.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/staging/gdm724x/gdm_lte.c 
b/drivers/staging/gdm724x/gdm_lte.c
index 4f3c51830..1f8a33f04 100644
--- a/drivers/staging/gdm724x/gdm_lte.c
+++ b/drivers/staging/gdm724x/gdm_lte.c
@@ -408,7 +408,7 @@ static s32 gdm_lte_tx_nic_type(struct net_device *dev, 
struct sk_buff *skb)
return nic_type;
 }
 
-static int gdm_lte_tx(struct sk_buff *skb, struct net_device *dev)
+static netdev_tx_t gdm_lte_tx(struct sk_buff *skb, struct net_device *dev)
 {
struct nic *nic = netdev_priv(dev);
u32 nic_type;
-- 
2.17.0

___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel

[PATCH] hv_netsvc: fix netvsc_start_xmit()'s return type

[Luc Van Oostenryck]

The method ndo_start_xmit() is defined as returning an 'netdev_tx_t',
which is a typedef for an enum type, but the implementation in this
driver returns an 'int'.

Fix this by returning 'netdev_tx_t' in this driver too.

Signed-off-by: Luc Van Oostenryck <luc.vanoostenr...@gmail.com>
---
 drivers/net/hyperv/netvsc_drv.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/hyperv/netvsc_drv.c b/drivers/net/hyperv/netvsc_drv.c
index ecc84954c..d402fdb08 100644
--- a/drivers/net/hyperv/netvsc_drv.c
+++ b/drivers/net/hyperv/netvsc_drv.c
@@ -508,7 +508,7 @@ static int netvsc_vf_xmit(struct net_device *net, struct 
net_device *vf_netdev,
return rc;
 }
 
-static int netvsc_start_xmit(struct sk_buff *skb, struct net_device *net)
+static netdev_tx_t netvsc_start_xmit(struct sk_buff *skb, struct net_device 
*net)
 {
struct net_device_context *net_device_ctx = netdev_priv(net);
struct hv_netvsc_packet *packet = NULL;
-- 
2.17.0

___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel

[PATCH] vboxvideo: fix mode_valid's return type

[Luc Van Oostenryck]

The method struct drm_connector_helper_funcs::mode_valid is defined
as returning an 'enum drm_mode_status' but the driver implementation
for this method uses an 'int' for it.

Fix this by using 'enum drm_mode_status' in the driver too.

Signed-off-by: Luc Van Oostenryck <luc.vanoostenr...@gmail.com>
---
 drivers/staging/vboxvideo/vbox_mode.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/staging/vboxvideo/vbox_mode.c 
b/drivers/staging/vboxvideo/vbox_mode.c
index b265fe924..5c7ea2378 100644
--- a/drivers/staging/vboxvideo/vbox_mode.c
+++ b/drivers/staging/vboxvideo/vbox_mode.c
@@ -573,7 +573,7 @@ static int vbox_get_modes(struct drm_connector *connector)
return num_modes;
 }
 
-static int vbox_mode_valid(struct drm_connector *connector,
+static enum drm_mode_status vbox_mode_valid(struct drm_connector *connector,
   struct drm_display_mode *mode)
 {
return MODE_OK;
-- 
2.17.0

___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel

Re: [PATCH 1/1] staging: ks7010: Fix cast to restricted __le16 in ks_wlan_net.c

[Luc Van Oostenryck]

On Sun, Jul 2, 2017 at 10:49 PM, Janusz Lisiecki
<janusz.lisie...@gmail.com> wrote:
> W dniu 2017-07-02 o 21:38, Luc Van Oostenryck pisze:
>
>> On Sun, Jul 2, 2017 at 4:27 PM, Janusz Lisiecki
>> <janusz.lisie...@gmail.com> wrote:
>>>
>>> This patch fixes the following Sparse warnings in ks_wlan_net.c:
>>> drivers/staging/ks7010/ks_wlan_net.c:1359:24: warning: cast to restricted
>>> __le16
>>> Both sides of assignment are u16 so (as 'ap' is local_ap_t type and
>>> 'capability' member,
>>> have the same as local 'capabilities' type of u16) 'le16_to_cpu' is not
>>> needed
>>
>> It could be that it's ap->capability's type that is wrong (not
>> annotated with __le16).
>> Isn't it?
>>
>> Is ap->capability supposed to hold a little-endian value or a native
>> order value?
>>
>> -- Luc
>
> As I see in ks_hostif.c all assignments to link_ap_info_t->capability threat
> this value as native order (i.e get_ap_information, get_current_ap). As this
> is not a structure which comes from HW we can do the way you suggested.
> Still, as all other places in code threats this as native order value I
> decided to change only one place than many other around to fix Sparse
> warning.

Fine, but then please put this explanation in the commit message.
In others words, be very clear that the change is because ap->capability is in
native order and thus the conversion le16_to_cpu() is wrong and must be removed.

-- Luc
___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel

Re: [PATCH 1/1] staging: ks7010: Fix cast to restricted __le16 in ks_wlan_net.c

[Luc Van Oostenryck]

On Sun, Jul 2, 2017 at 4:27 PM, Janusz Lisiecki
 wrote:
> This patch fixes the following Sparse warnings in ks_wlan_net.c:
> drivers/staging/ks7010/ks_wlan_net.c:1359:24: warning: cast to restricted 
> __le16
> Both sides of assignment are u16 so (as 'ap' is local_ap_t type and 
> 'capability' member,
> have the same as local 'capabilities' type of u16) 'le16_to_cpu' is not needed

It could be that it's ap->capability's type that is wrong (not
annotated with __le16).
Isn't it?

Is ap->capability supposed to hold a little-endian value or a native
order value?

-- Luc
___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel