hostapd: add 'driver' command support (under ANDROID flag)
Arend van Spriel
Mon Oct 6 02:40:15 PDT 2014
On 06-10-14 11:29, Arend van Spriel wrote:
> On 06-10-14 11:11, Garcia, Jeremie wrote:
>> Hello Jouni,
>> Thanks for answering.
>> I understand your comment.
>> In order to remain vendor agnostic, I initially planned to use the vendor extension functionality ("vendor" command).
>> However, this feature is available from kernel 3.14. However I would need to deal with older kernels.
>> So to remain backward compatible with older kernels, I wanted to use the DRIVER command instead.
> The vendor command does not make it driver/vendor agnostic. It is
> similar to the driver command, but using a differently transport layer,
> ie. nl80211. To get vendor agnostic you would have to get new NL80211
> commands in place that can replace the DRIVER commands and somehow
> backport that to older kernels (eq. using backports framework).
Support for backports has been added to android build environment by
Intel, which can be found on github .
>> Best regards
>> -----Original Message-----
>> From: hostap-bounces at lists.shmoo.com [mailto:hostap-bounces at lists.shmoo.com] On Behalf Of Jouni Malinen
>> Sent: Saturday, October 04, 2014 5:37 PM
>> To: hostap at lists.shmoo.com
>> Subject: Re: hostapd: add 'driver' command support (under ANDROID flag)
>> On Wed, Oct 01, 2014 at 02:25:22PM +0000, Garcia, Jeremie wrote:
>>> Please find attached 2 patches to make hostapd support "driver" command in an Android context.
>>> It is a mirror of what has been done for wpa_supplicant.
>> Why would this be needed in hostapd? That driver command in wpa_supplicant is a good example of how things should not be done and unless there is a strong justification for adding that to hostapd, I'd rather not do that.
>> The proper approach for adding new control interface functionality is by adding commands that can be implemented in driver independent way.
> HostAP mailing list
> HostAP at lists.shmoo.com
More information about the Hostap