Update some Mac Bluetooth TODOs.

Update some Mac Bluetooth TODOs.

BUG=none
TEST=none
R=keybuk@chromium.org

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=283356

[Ilya Sherman, 2014-07-12 05:54:50]

keybuk for OWNERS review, others as FYI and extra eyes.

Unfortunately, all of these TODOs remain valid -- I've now verified them.

The BluetoothSocketMac instances are not freed until the user quits Chrome, even
if the Chrome app that spawned them has been closed.  I haven't yet spent time
investigating possible solutions.

For dynamically assigning PSMs and Channels, the situation is odd: The system
seems to do no error checking of the values passed in, so it'll happily assign a
PSM and Channel equal to 0.  For Channels, it will then assign unused channels,
if any are available, until the channel space is saturated.  Once that happens,
it'll keep assigning channels outside of the valid range.  For PSMs, it'll just
create service record after service record with a PSM of 0 (so, both invalid and
used for multiple services).  I spent some time investigating solutions, but
didn't find anything that looked promising...

[Ilya Sherman, 2014-07-12 05:56:19]

On 2014/07/12 05:54:50, Ilya Sherman wrote:
> For dynamically assigning PSMs and Channels, the situation is odd: The system
> seems to do no error checking of the values passed in, so it'll happily assign
a
> PSM and Channel equal to 0.  For Channels, it will then assign unused
channels,
> if any are available, until the channel space is saturated.  Once that
happens,
> it'll keep assigning channels outside of the valid range.  For PSMs, it'll
just
> create service record after service record with a PSM of 0 (so, both invalid
and
> used for multiple services).  I spent some time investigating solutions, but
> didn't find anything that looked promising...

Well, I suppose we could manually scan the SDP records present on the system,
and try to manually compensate for the system.  Not sure whether that's worth
the overhead, though...

[keybuk, 2014-07-15 20:36:52]

lgtm

you could have totally TBRd this ;-)

[Ilya Sherman, 2014-07-15 20:39:17]

On 2014/07/15 20:36:52, keybuk wrote:
> lgtm
> 
> you could have totally TBRd this ;-)

Sure, but there also wasn't any particular rush to land it :)

[Ilya Sherman, 2014-07-15 20:39:21]

The CQ bit was checked by isherman@chromium.org

[commit-bot: I haz the power, 2014-07-15 20:42:12]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/isherman@chromium.org/385333002/20001

[commit-bot: I haz the power, 2014-07-16 05:26:45]

Change committed as 283356