Hello,
Thanks for helping me. I have some more questions related to parallel
usage of p2p and STA (station) interface.
I use a Nexus 5 device which has a BCM4339 [1] chip and it seems that
it's possible to transfer data in parallel using two connections: one
based on Wi-Fi direct(p2p, 2.4Ghz) and the other one based on a
connection to AP (wlan0, 5Ghz).
This is done by _time-sharing_ the wireless interface radio? From my
understanding the mac80211 layer have support for multiple virtual
interfaces and the switching between this interfaces is done inside
mac80211. Does the driver creates multiple virtual interfaces (one for
p2p and one for the connection to the AP) and the mac80211 manages the
time-sharing between them?
I am interested in understanding the code that decides how much the
radio should stay on a specific channel. I would be really grateful if
you could give me some details about the architecture.
[1] http://www.broadcom.com/press/release.php?id=s766879
Thanks,
Doru
On 09/08/2015 09:50 AM, Gucea Doru wrote:
> On Tue, Sep 8, 2015 at 12:01 AM, Arend van Spriel <[email protected]> wrote:
>
>> The BCM4339 chip does not rely on mac80211. Instead, there is an
>> 802.11 stack running on the device, ie. in firmware. The driver creates the
>> primary interface, ie. wlan0. Additional P2P interfaces are created by
>> wpa_supplicant.
>>
>
> Hm, BCM4339 uses the bcmdhd driver.
> So the bcmdhd driver is not a softmac[1] driver and the the 802.11
> processing is not done in software, right?
Sure it is done in software. Just not on the host, ie. in linux kernel,
but on the device.
Regards,
Arend
> [1] https://wireless.wiki.kernel.org/en/developers/documentation/glossary#softmac
>
> Regards,
> Doru
>
On 09/07/2015 10:41 PM, Gucea Doru wrote:
> Hello,
>
> Thanks for helping me. I have some more questions related to parallel
> usage of p2p and STA (station) interface.
>
> I use a Nexus 5 device which has a BCM4339 [1] chip and it seems that
> it's possible to transfer data in parallel using two connections: one
> based on Wi-Fi direct(p2p, 2.4Ghz) and the other one based on a
> connection to AP (wlan0, 5Ghz).
>
> This is done by _time-sharing_ the wireless interface radio? From my
> understanding the mac80211 layer have support for multiple virtual
> interfaces and the switching between this interfaces is done inside
> mac80211. Does the driver creates multiple virtual interfaces (one for
> p2p and one for the connection to the AP) and the mac80211 manages the
> time-sharing between them?
Nope. The BCM4339 chip does not rely on mac80211. Instead, there is an
802.11 stack running on the device, ie. in firmware. The driver creates
the primary interface, ie. wlan0. Additional P2P interfaces are created
by wpa_supplicant.
> I am interested in understanding the code that decides how much the
> radio should stay on a specific channel. I would be really grateful if
> you could give me some details about the architecture.
That code runs in firmware on the device and is proprietary. So I can
not disclose any details.
Regards,
Arend
> [1] http://www.broadcom.com/press/release.php?id=s766879
>
> Thanks,
> Doru
> --
> To unsubscribe from this list: send the line "unsubscribe linux-wireless" in
> the body of a message to [email protected]
> More majordomo info at http://vger.kernel.org/majordomo-info.html
>
On Tue, Sep 8, 2015 at 12:01 AM, Arend van Spriel <[email protected]> wrote:
> The BCM4339 chip does not rely on mac80211. Instead, there is an
> 802.11 stack running on the device, ie. in firmware. The driver creates the
> primary interface, ie. wlan0. Additional P2P interfaces are created by
> wpa_supplicant.
>
Hm, BCM4339 uses the bcmdhd driver.
So the bcmdhd driver is not a softmac[1] driver and the the 802.11
processing is not done in software, right?
[1] https://wireless.wiki.kernel.org/en/developers/documentation/glossary#softmac
Regards,
Doru