Latest compat-drivers

YanBo dreamfly281 at gmail.com
Wed May 8 13:47:37 EDT 2013


I just compiled with the compact-wireless just follow the Kalo's posted steps.
A little different is I'm not use CM but another android release base
one kernel 3.4

In the process of compile the cfg80211.ko and mac8011.ko  there is a
error reported
/var/tmp/output/net/wireless/nl80211.c: In function
'nl80211_crit_protocol_start':
/var/tmp/output/net/wireless/nl80211.c:8297:35: error: 'struct
genl_info' has no member named 'snd_portid'

apply below patch to fix this problem:

diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c
index e91cefe..f6a1b2e 100644
--- a/net/wireless/nl80211.c
+++ b/net/wireless/nl80211.c
@@ -8294,7 +8294,7 @@ static int nl80211_crit_protocol_start(struct
sk_buff *skb,

        ret = rdev_crit_proto_start(rdev, wdev, proto, duration);
        if (!ret)
-               rdev->crit_proto_nlportid = info->snd_portid;
+               rdev->crit_proto_nlportid = genl_info_snd_portid(info);

        return ret;
 }


On Wed, Apr 17, 2013 at 12:49 AM, Kalle Valo <kvalo at qca.qualcomm.com> wrote:
> Eugene Krasnikov <k.eugene.e at gmail.com> writes:
>
>> Yeah... finally i managed to build latest mac80211 based on your
>> instructions. But i still had to apply 3 patches from local folder.
>
> Too bad.
>
>>> https://github.com/KrasnikovEugene/wcn36xx/pull/7
>> It's merged but i still suggest to remove
>> DEFINE_DYNAMIC_DEBUG_METADATA completely. I'll send it out shortly so
>> you can test patch with maco.
>
> I'm perfectly fine with that.
>
> Actually I don't even like dynamic debug infrastructure, for me it feels
> next to useless. In practise either it's all debug messages enabled or
> all disabled, nothing in-between.
>
> Actually I was planning to propose a more suitable debug infrastructure
> following what ath6kl does with enum ATH6K_DEBUG_MASK:
>
> https://github.com/kvalo/ath6kl/blob/master/drivers/net/wireless/ath/ath6kl/debug.h
>
> But we can talk about that later.
>
>>> Maybe it also fixed that issue?
>>
>>  I guess it depends on kernel that is used in CM.
>
> Yeah, most likely.
>
> --
> Kalle Valo
>
> _______________________________________________
> wcn36xx mailing list
> wcn36xx at lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/wcn36xx



More information about the wcn36xx mailing list