[PATCH] mac80211: fix the multiple warnings for budding chipsets without ampdu_action defined

[PATCH] mac80211: fix the multiple warnings for budding chipsets without ampdu_action defined

[Chaitanya]

[-- Attachment #1: Type: text/plain, Size: 770 bytes --]



Warn once should suffice.

Eg: our driver doesn't support the ampdu_action yet, so
seeing a lot of warnings, hence the check for any
new drivers to avoid messing the kernel log.


Signed-off-by: T Krushna Chaitanya <chaitanyatk@posedge.com>
--- 
net/mac80211/agg-tx.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/net/mac80211/agg-tx.c b/net/mac80211/agg-tx.c
index 4152ed1..eb9df22 100644
--- a/net/mac80211/agg-tx.c
+++ b/net/mac80211/agg-tx.c
@@ -445,7 +445,7 @@ int ieee80211_start_tx_ba_session(struct ieee80211_sta *pubsta, u16 tid,
 
 	trace_api_start_tx_ba_session(pubsta, tid);
 
-	if (WARN_ON(!local->ops->ampdu_action))
+	if (WARN_ON_ONCE(!local->ops->ampdu_action))
 		return -EINVAL;
 
 	if ((tid >= IEEE80211_NUM_TIDS) ||





[-- Attachment #2: Attached Message Part --]
[-- Type: text/plain, Size: 0 bytes --]

Re: [PATCH] mac80211: fix the multiple warnings for budding chipsets without ampdu_action defined

[Johannes Berg]

On Fri, 2012-12-07 at 12:19 +0530, Chaitanya wrote:
> 
> Warn once should suffice.
> 
> Eg: our driver doesn't support the ampdu_action yet, so
> seeing a lot of warnings, hence the check for any

Yay, this worked. Applied with a changed commit log.

johannes

Re: [PATCH] mac80211: fix the multiple warnings for budding chipsets without ampdu_action defined

[Chaitanya]

Great to hear that johannes.
Finally after exhausting all options used the sendmail and it worked :-)

On Friday 07 December 2012 01:44 PM, Johannes Berg wrote:
> On Fri, 2012-12-07 at 12:19 +0530, Chaitanya wrote:
>> Warn once should suffice.
>>
>> Eg: our driver doesn't support the ampdu_action yet, so
>> seeing a lot of warnings, hence the check for any
> Yay, this worked. Applied with a changed commit log.
>
> johannes
>

[PATCH] mac80211: fix the multiple warnings for budding chipsets without ampdu_action defined

[Chaitanya]

Warn once should suffice.

Eg: our driver doesn't support the ampdu_action yet, so
seeing a lot of warnings, hence the check for any
new drivers to avoid messing the kernel log.

Changed the mail client to thunder bird to avoid whitespace issues.
Changed the subject and also the signing off to use offical mail.
Changed the setting for tabs (width=8, noexpandtab) and whitespaces.

Signed-off-by: T Krishna Chaitanya <chaitanyatk@posedge.com>
---
net/mac80211/agg-tx.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/net/mac80211/agg-tx.c b/net/mac80211/agg-tx.c
index 4152ed1..eb9df22 100644
--- a/net/mac80211/agg-tx.c
+++ b/net/mac80211/agg-tx.c
@@ -445,7 +445,7 @@ int ieee80211_start_tx_ba_session(struct ieee80211_sta *pubsta, u16 tid,
 
     trace_api_start_tx_ba_session(pubsta, tid);
 
-    if (WARN_ON(!local->ops->ampdu_action))
+    if (WARN_ON_ONCE(!local->ops->ampdu_action))
         return -EINVAL;
 
     if ((tid >= IEEE80211_NUM_TIDS) ||

Re: [PATCH] mac80211: fix the multiple warnings for budding chipsets without ampdu_action defined

[Stanislaw Gruszka]

On Thu, Dec 06, 2012 at 01:46:27PM +0530, Chaitanya wrote:
> Changed the mail client to thunder bird to avoid whitespace issues.
> Changed the subject and also the signing off to use offical mail.
> Changed the setting for tabs (width=8, noexpandtab) and whitespaces.

Don't need to mess up the change log with that.

> Signed-off-by: T Krishna Chaitanya <chaitanyatk@posedge.com>
> ---
> net/mac80211/agg-tx.c |    2 +-
>  1 files changed, 1 insertions(+), 1 deletions(-)
> 
> diff --git a/net/mac80211/agg-tx.c b/net/mac80211/agg-tx.c
> index 4152ed1..eb9df22 100644
> --- a/net/mac80211/agg-tx.c
> +++ b/net/mac80211/agg-tx.c
> @@ -445,7 +445,7 @@ int ieee80211_start_tx_ba_session(struct ieee80211_sta *pubsta, u16 tid,
>  
>      trace_api_start_tx_ba_session(pubsta, tid);
>  
> -    if (WARN_ON(!local->ops->ampdu_action))
^^^^^^^
Patch is still malformed (spaces instead of one tab). Before next
posting to mailing list, post patch to yourself and try to apply
it. Once that success, continue with mailing list posting.

Thanks
Stanislaw  

Re: [PATCH] mac80211: fix the multiple warnings for budding chipsets without ampdu_action defined

[Chaitanya]

I have tested the patch by applying it myself and it worked fine.
and also in My local editor (gedit) it shows tabs.

Need to check the issue and revert.

On Thursday 06 December 2012 02:21 PM, Stanislaw Gruszka wrote:
> On Thu, Dec 06, 2012 at 01:46:27PM +0530, Chaitanya wrote:
>> Changed the mail client to thunder bird to avoid whitespace issues.
>> Changed the subject and also the signing off to use offical mail.
>> Changed the setting for tabs (width=8, noexpandtab) and whitespaces.
> Don't need to mess up the change log with that.
>
>> Signed-off-by: T Krishna Chaitanya <chaitanyatk@posedge.com>
>> ---
>> net/mac80211/agg-tx.c |    2 +-
>>  1 files changed, 1 insertions(+), 1 deletions(-)
>>
>> diff --git a/net/mac80211/agg-tx.c b/net/mac80211/agg-tx.c
>> index 4152ed1..eb9df22 100644
>> --- a/net/mac80211/agg-tx.c
>> +++ b/net/mac80211/agg-tx.c
>> @@ -445,7 +445,7 @@ int ieee80211_start_tx_ba_session(struct ieee80211_sta *pubsta, u16 tid,
>>  
>>      trace_api_start_tx_ba_session(pubsta, tid);
>>  
>> -    if (WARN_ON(!local->ops->ampdu_action))
> ^^^^^^^
> Patch is still malformed (spaces instead of one tab). Before next
> posting to mailing list, post patch to yourself and try to apply
> it. Once that success, continue with mailing list posting.
>
> Thanks
> Stanislaw  

Re: [PATCH] mac80211: fix the multiple warnings for budding chipsets without ampdu_action defined

[Johannes Berg]

On Thu, 2012-12-06 at 15:07 +0530, Chaitanya wrote:
> I have tested the patch by applying it myself and it worked fine.
> and also in My local editor (gedit) it shows tabs.
> 
> Need to check the issue and revert.

Try after it made it to you from the list -- still broken here as well.

johannes

[PATCH] mac80211: fix the multiple warnings for budding chipsets without ampdu_action defined

[Chaitanya]

Warn once should suffice.

Eg: our driver doesn't support the ampdu_action yet, so
seeing a lot of warnings, hence the check for any
new drivers to avoid messing the kernel log.

Changed the mail client to thunder bird to avoid whitespace issues.
Changed the subject and also the signing off to use offical mail.

Signed-off-by: T Krushna Chaitanya <chaitanyatk@posedge.com>
---
net/mac80211/agg-tx.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/net/mac80211/agg-tx.c b/net/mac80211/agg-tx.c
index 4152ed1..eb9df22 100644
--- a/net/mac80211/agg-tx.c
+++ b/net/mac80211/agg-tx.c
@@ -445,7 +445,7 @@ int ieee80211_start_tx_ba_session(struct ieee80211_sta *pubsta, u16 tid,
 
     trace_api_start_tx_ba_session(pubsta, tid);
 
-    if (WARN_ON(!local->ops->ampdu_action))
+    if (WARN_ON_ONCE(!local->ops->ampdu_action))
         return -EINVAL;
 
     if ((tid >= IEEE80211_NUM_TIDS) ||

Re: [PATCH] mac80211: fix the multiple warnings for budding chipsets without ampdu_action defined

[Johannes Berg]

On Wed, 2012-12-05 at 17:57 +0530, Chaitanya wrote:

> --- a/net/mac80211/agg-tx.c
> +++ b/net/mac80211/agg-tx.c
> @@ -445,7 +445,7 @@ int ieee80211_start_tx_ba_session(struct ieee80211_sta *pubsta, u16 tid,
>  
>      trace_api_start_tx_ba_session(pubsta, tid);
>  
> -    if (WARN_ON(!local->ops->ampdu_action))
> +    if (WARN_ON_ONCE(!local->ops->ampdu_action))
>          return -EINVAL;

Hm, no tabs it seems, so this doesn't apply either.

I'd just fix it myself for this simple a thing, but you'll probably
eventually want to submit more complicated patches ...

johannes

Re: [PATCH] mac80211: fix the multiple warnings for budding chipsets without ampdu_action defined

[Chaitanya]

Hmm..yes true i need to fix my environment.
But i just tested this patch manually and it worked.

Let me check this once again, will fix this and send it again.

Sorry for the trouble caused.


On Wednesday 05 December 2012 06:00 PM, Johannes Berg wrote:
> Hm, no tabs it seems, so this doesn't apply either.
>
> I'd just fix it myself for this simple a thing, but you'll probably
> eventually want to submit more complicated patches ...