Re: [lwip-users] Assert after dropped TX packet

[Kieran Mansley]

On Thu, 2011-10-20 at 15:14 -0500, Dittrich, Matthew wrote:
> Any hints would be greatly appreciated... Sorry for the long post, I
> hope you made it through the whole thing! This ended up much longer
> than I intended.

I think chaining the pbufs as you are in the received callback is fine.
Once lwIP has passed them to the receive callback it has no further use
for them (until they are recycled by you calling pbuf_free()) and so it
doesn't care how you use them.

The assertions that you mentioned concern pcbs (i.e. connection state)
rather than pbufs (packet state).  My guess is that you've hit the
common fault of failing to ensure that there is only a single thread
active in the lwIP core at any one time.  Perhaps you can give us some
more details about how the driver interfaces to lwIP, particularly to
pass up received packets, and how you protect against this being
concurrent with the application processing the stack.

Kieran