On the 2GHz and and on the 5GHZ band only the extpa_gain setting from
the 5GHz band was checked. this patch makes it check the property from
the correct band.
Signed-off-by: Hauke Mehrtens <[email protected]>
---
drivers/net/wireless/b43/tables_nphy.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/wireless/b43/tables_nphy.c b/drivers/net/wireless/b43/tables_nphy.c
index 25d1cbd..b2f0d24 100644
--- a/drivers/net/wireless/b43/tables_nphy.c
+++ b/drivers/net/wireless/b43/tables_nphy.c
@@ -3728,7 +3728,7 @@ const u32 *b43_nphy_get_tx_gain_table(struct b43_wldev *dev)
switch (phy->rev) {
case 6:
case 5:
- if (sprom->fem.ghz5.extpa_gain == 3)
+ if (sprom->fem.ghz2.extpa_gain == 3)
return b43_ntab_tx_gain_epa_rev3_hi_pwr_2g;
/* fall through */
case 4:
--
2.1.4
> On the 2GHz and and on the 5GHZ band only the extpa_gain setting from
> the 5GHz band was checked. this patch makes it check the property from
> the correct band.
>
> Signed-off-by: Hauke Mehrtens <[email protected]>
Thanks, applied to wireless-drivers.git.
Kalle Valo
On 07/30/2015 08:54 AM, Rafał Miłecki wrote:
> On 30 July 2015 at 07:40, Kalle Valo <[email protected]> wrote:
>> Hauke Mehrtens <[email protected]> writes:
>>
>>> On the 2GHz and and on the 5GHZ band only the extpa_gain setting from
>>> the 5GHz band was checked. this patch makes it check the property from
>>> the correct band.
>>>
>>> Signed-off-by: Hauke Mehrtens <[email protected]>
>>
>> Is this a regression? If yes, please add a Fixes line.
>>
>> Should this go to 4.2?
>
> Not a regression. It's like this for a loooong time.
>
> AFAIK PCI(e) cards don't have external power amplifier, so this code
> is most likely used for some fancy SoCs only. There are 3 known SoCs
> with N-PHY rev 5 or 6: BCM4717A1, BCM4718A1 & BCM4716B0.
>
> So we may consider taking it for 4.2 without any "Fixes: " line but I
> wouldn't care too much about Cc-ing stable.
>
> Hauke: do you agree?
>
Yes, I do not have a device I know of which is affected by this error, I
just saw this while looking at code close by. Just put this into the
next Linux version and not to stable.
Hauke
On 30 July 2015 at 07:40, Kalle Valo <[email protected]> wrote:
> Hauke Mehrtens <[email protected]> writes:
>
>> On the 2GHz and and on the 5GHZ band only the extpa_gain setting from
>> the 5GHz band was checked. this patch makes it check the property from
>> the correct band.
>>
>> Signed-off-by: Hauke Mehrtens <[email protected]>
>
> Is this a regression? If yes, please add a Fixes line.
>
> Should this go to 4.2?
Not a regression. It's like this for a loooong time.
AFAIK PCI(e) cards don't have external power amplifier, so this code
is most likely used for some fancy SoCs only. There are 3 known SoCs
with N-PHY rev 5 or 6: BCM4717A1, BCM4718A1 & BCM4716B0.
So we may consider taking it for 4.2 without any "Fixes: " line but I
wouldn't care too much about Cc-ing stable.
Hauke: do you agree?
--
Rafał
On 29 July 2015 at 23:36, Hauke Mehrtens <[email protected]> wrote:
> On the 2GHz and and on the 5GHZ band only the extpa_gain setting from
> the 5GHz band was checked. this patch makes it check the property from
> the correct band.
Looks OK, thanks.
Hauke Mehrtens <[email protected]> writes:
> On the 2GHz and and on the 5GHZ band only the extpa_gain setting from
> the 5GHz band was checked. this patch makes it check the property from
> the correct band.
>
> Signed-off-by: Hauke Mehrtens <[email protected]>
Is this a regression? If yes, please add a Fixes line.
Should this go to 4.2?
--
Kalle Valo
Hauke Mehrtens <[email protected]> writes:
> On 07/30/2015 08:54 AM, Rafał Miłecki wrote:
>> On 30 July 2015 at 07:40, Kalle Valo <[email protected]> wrote:
>>> Hauke Mehrtens <[email protected]> writes:
>>>
>>>> On the 2GHz and and on the 5GHZ band only the extpa_gain setting from
>>>> the 5GHz band was checked. this patch makes it check the property from
>>>> the correct band.
>>>>
>>>> Signed-off-by: Hauke Mehrtens <[email protected]>
>>>
>>> Is this a regression? If yes, please add a Fixes line.
>>>
>>> Should this go to 4.2?
>>
>> Not a regression. It's like this for a loooong time.
>>
>> AFAIK PCI(e) cards don't have external power amplifier, so this code
>> is most likely used for some fancy SoCs only. There are 3 known SoCs
>> with N-PHY rev 5 or 6: BCM4717A1, BCM4718A1 & BCM4716B0.
>>
>> So we may consider taking it for 4.2 without any "Fixes: " line but I
>> wouldn't care too much about Cc-ing stable.
>>
>> Hauke: do you agree?
>
> Yes, I do not have a device I know of which is affected by this error, I
> just saw this while looking at code close by. Just put this into the
> next Linux version and not to stable.
Ok, I'll send this to 4.2.
--
Kalle Valo