Synching trees for BT 3.0 and a possible compat-bluetooth

Synching trees for BT 3.0 and a possible compat-bluetooth

[Luis R. Rodriguez]

I forgot to poke about BT 3.0 synching at the summit. To work on
bluetooth 3.0 we'll need some synching of trees between
wireless-testing and Marcel's bluetooth tree. Also a compat-bluetooth
might be in order to accomplish the same as we do with
compat-wireless, only it would seem logical to consider tying to just
merge this with compat-wireless to get the features required for
bluetooth 3.0; to do this having one tree with both would be nice with
the added benefit of all the same compat-2.6.3x.[ch] files being
reused. Synching of the trees would be required only when a feature on
one tree is not yet available on the latest rc kernel (provided marcel
rebases on Linus' latest rc as well, not sure), this will probably be
important during initial implementation, not sure how much divergence
we should expect after this goes through a release through Linus.

So any thoughts on synching bluetooth and wireless-testing other than
doing so on a private tree for development purposes?

  Luis

Re: Synching trees for BT 3.0 and a possible compat-bluetooth

[Marcel Holtmann]

Hi Luis,

> I forgot to poke about BT 3.0 synching at the summit. To work on
> bluetooth 3.0 we'll need some synching of trees between
> wireless-testing and Marcel's bluetooth tree. Also a compat-bluetooth
> might be in order to accomplish the same as we do with
> compat-wireless, only it would seem logical to consider tying to just
> merge this with compat-wireless to get the features required for
> bluetooth 3.0; to do this having one tree with both would be nice with
> the added benefit of all the same compat-2.6.3x.[ch] files being
> reused. Synching of the trees would be required only when a feature on
> one tree is not yet available on the latest rc kernel (provided marcel
> rebases on Linus' latest rc as well, not sure), this will probably be
> important during initial implementation, not sure how much divergence
> we should expect after this goes through a release through Linus.
> 
> So any thoughts on synching bluetooth and wireless-testing other than
> doing so on a private tree for development purposes?

we don't need this right now. For the initial development of the WiFi
AMP, we will use the fact that you can create HCI RAW devices from a
driver. These are skipped by BlueZ and only show up in hciconfig. That
should be enough for testing.

Regards

Marcel

Re: Synching trees for BT 3.0 and a possible compat-bluetooth

[Luis R. Rodriguez]

On Wed, Sep 23, 2009 at 4:27 PM, Marcel Holtmann <marcel@holtmann.org> wrote:
> Hi Luis,
>
>> I forgot to poke about BT 3.0 synching at the summit. To work on
>> bluetooth 3.0 we'll need some synching of trees between
>> wireless-testing and Marcel's bluetooth tree. Also a compat-bluetooth
>> might be in order to accomplish the same as we do with
>> compat-wireless, only it would seem logical to consider tying to just
>> merge this with compat-wireless to get the features required for
>> bluetooth 3.0; to do this having one tree with both would be nice with
>> the added benefit of all the same compat-2.6.3x.[ch] files being
>> reused. Synching of the trees would be required only when a feature on
>> one tree is not yet available on the latest rc kernel (provided marcel
>> rebases on Linus' latest rc as well, not sure), this will probably be
>> important during initial implementation, not sure how much divergence
>> we should expect after this goes through a release through Linus.
>>
>> So any thoughts on synching bluetooth and wireless-testing other than
>> doing so on a private tree for development purposes?
>
> we don't need this right now. For the initial development of the WiFi
> AMP, we will use the fact that you can create HCI RAW devices from a
> driver. These are skipped by BlueZ and only show up in hciconfig. That
> should be enough for testing.

Thanks Marcel, when do you expect we would need this?

  Luis