[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[lwip-users] Re: [lwip] IOCTL
From: |
Adam Dunkels |
Subject: |
[lwip-users] Re: [lwip] IOCTL |
Date: |
Thu, 09 Jan 2003 00:27:23 -0000 |
Hi John!
On Tuesday 29 January 2002 01.38, you wrote:
> I've been looking at lwip with the idea of using it
> to run tcp/ip across a bluetooth connection. Seems pretty
> doable, but there are BT features that aren't easily
> exposed through the standard socket style interface.
> (The primary issue is that Bluetooth doesn't expose
> a "party-line" interface like a ethernet wire. Instead,
> a bunch of work must be done to establish a connection
> prior to issueing the first network packet).
Interesting! We actually ran lwIP over Bluetooth some time ago. We didn't
implement anything of the BT stack, we just sent the IP packets using the HCI
and didn't bother with addressing, L2CAP and such. We only used it in a small
demo where interoperability wasn't an issue.
> So it seems like adding a ioctlsocket call would a reasonable
> compromise.
>
> I would like to guarantee order of execution (eg previously
> issued API calls execute prior to calling my ioctl). Is there
> a way to do this, or would I be best off creating a new message
> type and using api_msg_post?
I would suggest adding a new IOCTL message type and use the existing API
framework. Or even add new API functions for dealing with the BT interface.
Would it not be possible to do the link setup stuff in the network device
driver instead of the application, by the way?
/adam
--
Adam Dunkels <address@hidden>
http://www.sics.se/~adam
[This message was sent through the lwip discussion list.]
[Prev in Thread] |
Current Thread |
[Next in Thread] |
- [lwip-users] Re: [lwip] IOCTL,
Adam Dunkels <=