[PATCH] ath10k: unregister and remove frag_threshold callback

[PATCH] ath10k: unregister and remove frag_threshold callback

[Peter Oh]

Setting fragmentation threshold has not been supported by
any of firmware versions, hence unregister the callback and
remove the function.

Signed-off-by: Peter Oh <poh@qca.qualcomm.com>
---
 drivers/net/wireless/ath/ath10k/mac.c | 24 ------------------------
 1 file changed, 24 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/mac.c b/drivers/net/wireless/ath/ath10k/mac.c
index c400567..604a4bc 100644
--- a/drivers/net/wireless/ath/ath10k/mac.c
+++ b/drivers/net/wireless/ath/ath10k/mac.c
@@ -3989,29 +3989,6 @@ static int ath10k_set_rts_threshold(struct ieee80211_hw *hw, u32 value)
 	return ret;
 }
 
-static int ath10k_set_frag_threshold(struct ieee80211_hw *hw, u32 value)
-{
-	struct ath10k *ar = hw->priv;
-	struct ath10k_vif *arvif;
-	int ret = 0;
-
-	mutex_lock(&ar->conf_mutex);
-	list_for_each_entry(arvif, &ar->arvifs, list) {
-		ath10k_dbg(ar, ATH10K_DBG_MAC, "mac vdev %d fragmentation threshold %d\n",
-			   arvif->vdev_id, value);
-
-		ret = ath10k_mac_set_frag(arvif, value);
-		if (ret) {
-			ath10k_warn(ar, "failed to set fragmentation threshold for vdev %d: %d\n",
-				    arvif->vdev_id, ret);
-			break;
-		}
-	}
-	mutex_unlock(&ar->conf_mutex);
-
-	return ret;
-}
-
 static void ath10k_flush(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 			 u32 queues, bool drop)
 {
@@ -4655,7 +4632,6 @@ static const struct ieee80211_ops ath10k_ops = {
 	.remain_on_channel		= ath10k_remain_on_channel,
 	.cancel_remain_on_channel	= ath10k_cancel_remain_on_channel,
 	.set_rts_threshold		= ath10k_set_rts_threshold,
-	.set_frag_threshold		= ath10k_set_frag_threshold,
 	.flush				= ath10k_flush,
 	.tx_last_beacon			= ath10k_tx_last_beacon,
 	.set_antenna			= ath10k_set_antenna,
-- 
1.9.1

[PATCH] ath10k: unregister and remove frag_threshold callback

[Peter Oh]

Setting fragmentation threshold has not been supported by
any of firmware versions, hence unregister the callback and
remove the function.

Signed-off-by: Peter Oh <poh@qca.qualcomm.com>
---
 drivers/net/wireless/ath/ath10k/mac.c | 24 ------------------------
 1 file changed, 24 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/mac.c b/drivers/net/wireless/ath/ath10k/mac.c
index c400567..604a4bc 100644
--- a/drivers/net/wireless/ath/ath10k/mac.c
+++ b/drivers/net/wireless/ath/ath10k/mac.c
@@ -3989,29 +3989,6 @@ static int ath10k_set_rts_threshold(struct ieee80211_hw *hw, u32 value)
 	return ret;
 }
 
-static int ath10k_set_frag_threshold(struct ieee80211_hw *hw, u32 value)
-{
-	struct ath10k *ar = hw->priv;
-	struct ath10k_vif *arvif;
-	int ret = 0;
-
-	mutex_lock(&ar->conf_mutex);
-	list_for_each_entry(arvif, &ar->arvifs, list) {
-		ath10k_dbg(ar, ATH10K_DBG_MAC, "mac vdev %d fragmentation threshold %d\n",
-			   arvif->vdev_id, value);
-
-		ret = ath10k_mac_set_frag(arvif, value);
-		if (ret) {
-			ath10k_warn(ar, "failed to set fragmentation threshold for vdev %d: %d\n",
-				    arvif->vdev_id, ret);
-			break;
-		}
-	}
-	mutex_unlock(&ar->conf_mutex);
-
-	return ret;
-}
-
 static void ath10k_flush(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 			 u32 queues, bool drop)
 {
@@ -4655,7 +4632,6 @@ static const struct ieee80211_ops ath10k_ops = {
 	.remain_on_channel		= ath10k_remain_on_channel,
 	.cancel_remain_on_channel	= ath10k_cancel_remain_on_channel,
 	.set_rts_threshold		= ath10k_set_rts_threshold,
-	.set_frag_threshold		= ath10k_set_frag_threshold,
 	.flush				= ath10k_flush,
 	.tx_last_beacon			= ath10k_tx_last_beacon,
 	.set_antenna			= ath10k_set_antenna,
-- 
1.9.1


_______________________________________________
ath10k mailing list
ath10k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath10k

Re: [PATCH] ath10k: unregister and remove frag_threshold callback

[Ben Greear]

On 12/05/2014 09:16 AM, Peter Oh wrote:
> Setting fragmentation threshold has not been supported by
> any of firmware versions, hence unregister the callback and
> remove the function.

How hard is it to just fix the firmware instead?

Thanks,
Ben

-- 
Ben Greear <greearb@candelatech.com>
Candela Technologies Inc  http://www.candelatech.com

Re: [PATCH] ath10k: unregister and remove frag_threshold callback

[Ben Greear]

On 12/05/2014 09:16 AM, Peter Oh wrote:
> Setting fragmentation threshold has not been supported by
> any of firmware versions, hence unregister the callback and
> remove the function.

How hard is it to just fix the firmware instead?

Thanks,
Ben

-- 
Ben Greear <greearb@candelatech.com>
Candela Technologies Inc  http://www.candelatech.com

_______________________________________________
ath10k mailing list
ath10k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath10k

Re: [PATCH] ath10k: unregister and remove frag_threshold callback

[poh]

On 12/05/2014 07:59 PM, Ben Greear wrote:
>
>
> On 12/05/2014 09:16 AM, Peter Oh wrote:
>> Setting fragmentation threshold has not been supported by
>> any of firmware versions, hence unregister the callback and
>> remove the function.
>
> How hard is it to just fix the firmware instead?
It's due to hardware limitation such as lack of buffers/RAM size 
limitation, hence it's very hard from sw perspective.
>
> Thanks,
> Ben
>

Re: [PATCH] ath10k: unregister and remove frag_threshold callback

[poh]

On 12/05/2014 07:59 PM, Ben Greear wrote:
>
>
> On 12/05/2014 09:16 AM, Peter Oh wrote:
>> Setting fragmentation threshold has not been supported by
>> any of firmware versions, hence unregister the callback and
>> remove the function.
>
> How hard is it to just fix the firmware instead?
It's due to hardware limitation such as lack of buffers/RAM size 
limitation, hence it's very hard from sw perspective.
>
> Thanks,
> Ben
>


_______________________________________________
ath10k mailing list
ath10k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath10k

Re: [PATCH] ath10k: unregister and remove frag_threshold callback

[Ben Greear]

On 12/10/2014 10:33 AM, poh wrote:
> 
> On 12/05/2014 07:59 PM, Ben Greear wrote:
>>
>>
>> On 12/05/2014 09:16 AM, Peter Oh wrote:
>>> Setting fragmentation threshold has not been supported by
>>> any of firmware versions, hence unregister the callback and
>>> remove the function.
>>
>> How hard is it to just fix the firmware instead?
> It's due to hardware limitation such as lack of buffers/RAM size limitation, hence it's very hard from sw perspective.

Out of curiosity, how much more RAM do you need?  I was able to squeeze somewhere around 80k out of the firmware
in my quest to support 64 station vdevs....

Thanks,
Ben

-- 
Ben Greear <greearb@candelatech.com>
Candela Technologies Inc  http://www.candelatech.com

Re: [PATCH] ath10k: unregister and remove frag_threshold callback

[Ben Greear]

On 12/10/2014 10:33 AM, poh wrote:
> 
> On 12/05/2014 07:59 PM, Ben Greear wrote:
>>
>>
>> On 12/05/2014 09:16 AM, Peter Oh wrote:
>>> Setting fragmentation threshold has not been supported by
>>> any of firmware versions, hence unregister the callback and
>>> remove the function.
>>
>> How hard is it to just fix the firmware instead?
> It's due to hardware limitation such as lack of buffers/RAM size limitation, hence it's very hard from sw perspective.

Out of curiosity, how much more RAM do you need?  I was able to squeeze somewhere around 80k out of the firmware
in my quest to support 64 station vdevs....

Thanks,
Ben

-- 
Ben Greear <greearb@candelatech.com>
Candela Technologies Inc  http://www.candelatech.com


_______________________________________________
ath10k mailing list
ath10k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath10k

Re: [PATCH] ath10k: unregister and remove frag_threshold callback

[Peter Oh]

On 12/10/2014 11:03 AM, Ben Greear wrote:
> On 12/10/2014 10:33 AM, poh wrote:
>> On 12/05/2014 07:59 PM, Ben Greear wrote:
>>>
>>> On 12/05/2014 09:16 AM, Peter Oh wrote:
>>>> Setting fragmentation threshold has not been supported by
>>>> any of firmware versions, hence unregister the callback and
>>>> remove the function.
>>> How hard is it to just fix the firmware instead?
>> It's due to hardware limitation such as lack of buffers/RAM size limitation, hence it's very hard from sw perspective.
> Out of curiosity, how much more RAM do you need?  I was able to squeeze somewhere around 80k out of the firmware
> in my quest to support 64 station vdevs....
I don't have exact number of memory since the answer is from our 
firmware team.
I assume it's not coming in near future to firmware.
> Thanks,
> Ben
>
Thanks,
Peter

Re: [PATCH] ath10k: unregister and remove frag_threshold callback

[Peter Oh]

On 12/10/2014 11:03 AM, Ben Greear wrote:
> On 12/10/2014 10:33 AM, poh wrote:
>> On 12/05/2014 07:59 PM, Ben Greear wrote:
>>>
>>> On 12/05/2014 09:16 AM, Peter Oh wrote:
>>>> Setting fragmentation threshold has not been supported by
>>>> any of firmware versions, hence unregister the callback and
>>>> remove the function.
>>> How hard is it to just fix the firmware instead?
>> It's due to hardware limitation such as lack of buffers/RAM size limitation, hence it's very hard from sw perspective.
> Out of curiosity, how much more RAM do you need?  I was able to squeeze somewhere around 80k out of the firmware
> in my quest to support 64 station vdevs....
I don't have exact number of memory since the answer is from our 
firmware team.
I assume it's not coming in near future to firmware.
> Thanks,
> Ben
>
Thanks,
Peter

_______________________________________________
ath10k mailing list
ath10k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath10k

Re: [PATCH] ath10k: unregister and remove frag_threshold callback

[Kalle Valo]

Peter Oh <poh@qca.qualcomm.com> writes:

> Setting fragmentation threshold has not been supported by
> any of firmware versions, hence unregister the callback and
> remove the function.
>
> Signed-off-by: Peter Oh <poh@qca.qualcomm.com>

Thanks, applied.

-- 
Kalle Valo

Re: [PATCH] ath10k: unregister and remove frag_threshold callback

[Kalle Valo]

Peter Oh <poh@qca.qualcomm.com> writes:

> Setting fragmentation threshold has not been supported by
> any of firmware versions, hence unregister the callback and
> remove the function.
>
> Signed-off-by: Peter Oh <poh@qca.qualcomm.com>

Thanks, applied.

-- 
Kalle Valo

_______________________________________________
ath10k mailing list
ath10k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath10k