[PATCH] staging: rtl8192u: ieee80211: Silence sparse endianness warning

[PATCH] staging: rtl8192u: ieee80211: Silence sparse endianness warning

[Gaston Gonzalez]

Silence the following sparse warning:

drivers/staging/rtl8192u/ieee80211/ieee80211_softmac.c:663:32: warning: incorrect type in assignment (different base types)
drivers/staging/rtl8192u/ieee80211/ieee80211_softmac.c:663:32:    expected restricted __le16 [usertype] frame_ctl
drivers/staging/rtl8192u/ieee80211/ieee80211_softmac.c:663:32:    got int

Signed-off-by: Gaston Gonzalez <gascoar@gmail.com>
---
 drivers/staging/rtl8192u/ieee80211/ieee80211_softmac.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/staging/rtl8192u/ieee80211/ieee80211_softmac.c b/drivers/staging/rtl8192u/ieee80211/ieee80211_softmac.c
index d2e8b12..7dd8442 100644
--- a/drivers/staging/rtl8192u/ieee80211/ieee80211_softmac.c
+++ b/drivers/staging/rtl8192u/ieee80211/ieee80211_softmac.c
@@ -660,2 +660,2 @@ inline struct sk_buff *ieee80211_authentication_req(struct ieee80211_network *be
 	auth = (struct ieee80211_authentication *)
 		skb_put(skb, sizeof(struct ieee80211_authentication));

-	auth->header.frame_ctl = IEEE80211_STYPE_AUTH;
+	auth->header.frame_ctl = cpu_to_le16(IEEE80211_STYPE_AUTH);
 	if (challengelen) auth->header.frame_ctl |= IEEE80211_FCTL_WEP;

 	auth->header.duration_id = 0x013a; //FIXME
--
2.1.4

Re: [PATCH] staging: rtl8192u: ieee80211: Silence sparse endianness warning

[Dan Carpenter]

On Mon, May 25, 2015 at 12:00:42PM -0300, Gaston Gonzalez wrote:
> Silence the following sparse warning:
> 
> drivers/staging/rtl8192u/ieee80211/ieee80211_softmac.c:663:32: warning: incorrect type in assignment (different base types)
> drivers/staging/rtl8192u/ieee80211/ieee80211_softmac.c:663:32:    expected restricted __le16 [usertype] frame_ctl
> drivers/staging/rtl8192u/ieee80211/ieee80211_softmac.c:663:32:    got int
> 
> Signed-off-by: Gaston Gonzalez <gascoar@gmail.com>
> ---
>  drivers/staging/rtl8192u/ieee80211/ieee80211_softmac.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/drivers/staging/rtl8192u/ieee80211/ieee80211_softmac.c b/drivers/staging/rtl8192u/ieee80211/ieee80211_softmac.c
> index d2e8b12..7dd8442 100644
> --- a/drivers/staging/rtl8192u/ieee80211/ieee80211_softmac.c
> +++ b/drivers/staging/rtl8192u/ieee80211/ieee80211_softmac.c
> @@ -660,2 +660,2 @@ inline struct sk_buff *ieee80211_authentication_req(struct ieee80211_network *be
>  	auth = (struct ieee80211_authentication *)
>  		skb_put(skb, sizeof(struct ieee80211_authentication));
> 
> -	auth->header.frame_ctl = IEEE80211_STYPE_AUTH;
> +	auth->header.frame_ctl = cpu_to_le16(IEEE80211_STYPE_AUTH);
>  	if (challengelen) auth->header.frame_ctl |= IEEE80211_FCTL_WEP;
                          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
This is also wrong then.

regards,
dan carpenter

Re: [PATCH] staging: rtl8192u: ieee80211: Silence sparse endianness warning

[Gaston Gonzalez]

On 25/05/15 13:35, Dan Carpenter wrote:
> This is also wrong then.
>
> regards,
> dan carpenter
Hi Dan,

Yes, you are right. It is the next sparse warning in line for that file.
Including the fix for that, the patch would be as showed below.

There are similar endianness warnings for other variables in that file,
like the FIXME in last line of the patch. But keeping in mind the rule
'one thing per patch' I guess this should be fixed in other patche/s, right?

Regards,

Gaston

diff --git a/drivers/staging/rtl8192u/ieee80211/ieee80211_softmac.c
b/drivers/staging/rtl8192u/ieee80211/ieee80211_softmac.c
index d2e8b12..0477ba1 100644
--- a/drivers/staging/rtl8192u/ieee80211/ieee80211_softmac.c
+++ b/drivers/staging/rtl8192u/ieee80211/ieee80211_softmac.c
@@ -660,2 +660,2 @@ inline struct sk_buff
*ieee80211_authentication_req(struct ieee80211_network *be
     auth = (struct ieee80211_authentication *)
         skb_put(skb, sizeof(struct ieee80211_authentication));

-    auth->header.frame_ctl = IEEE80211_STYPE_AUTH;
-    if (challengelen) auth->header.frame_ctl |= IEEE80211_FCTL_WEP;
+    auth->header.frame_ctl = cpu_to_le16(IEEE80211_STYPE_AUTH);
+    if (challengelen)
+        auth->header.frame_ctl |= cpu_to_le16(IEEE80211_FCTL_WEP);

     auth->header.duration_id = 0x013a; //FIXME

--
2.1.4

Re: [PATCH] staging: rtl8192u: ieee80211: Silence sparse endianness warning

[Sudip Mukherjee]

On Mon, May 25, 2015 at 07:50:55PM -0300, Gaston Gonzalez wrote:
> On 25/05/15 13:35, Dan Carpenter wrote:
> > This is also wrong then.
> >
> > regards,
> > dan carpenter
> Hi Dan,
> 
> Yes, you are right. It is the next sparse warning in line for that file.
> Including the fix for that, the patch would be as showed below.
please send a proper patch with proper commit message.

regards
sudip
> 

Re: [PATCH] staging: rtl8192u: ieee80211: Silence sparse endianness warning

[Dan Carpenter]

Btw, the subject should say "fix endian bug".  "silence" means that
their is a warning and possible some messy code but no runtime bug.

On Mon, May 25, 2015 at 07:50:55PM -0300, Gaston Gonzalez wrote:
> On 25/05/15 13:35, Dan Carpenter wrote:
> > This is also wrong then.
> >
> > regards,
> > dan carpenter
> Hi Dan,
> 
> Yes, you are right. It is the next sparse warning in line for that file.
> Including the fix for that, the patch would be as showed below.
> 
> There are similar endianness warnings for other variables in that file,
> like the FIXME in last line of the patch. But keeping in mind the rule
> 'one thing per patch' I guess this should be fixed in other patche/s, right?

The one thing per patch rule is a bit fuzzy.  It depends on how you sell
it a bit.  I wouldn't mind if you fixed the whole function at once.  Or
even all then endian bugs in a file if the patch wasn't too complicated.

But this is also acceptable to fix one struct member so send your
proposed patch.

regards,
dan carpenter