Re: [lwip-devel] again... Zero Copy Ethernet interface

[Alain M.]

address@hidden escreveu:
> >   
> Nice list. I guess the hook method is quite a nice start for a generic 
> solution: I can't see how we can easily cover all of the above.

Actualy I have one suggestion that might covver all those and future 
ones too...

Create a set of fuctions to be implemented in each port, to allocate, 
release and delete pbufs.
- There should be flags or different functions fot TX/RX
- there should be provision of the case where a TX pbuf is relased after 
DMA is completed
- there should be provision for reentrancy in all this because most will 
be happening inside IRQs.
- maybe a few parameters can be available even if they are not used in 
the refference implementation

The the standard implementation is just what happens today using 
malloc() and can be a reasonable example to make a first working version...

> In addition to the hooks we could/should provide:
> a) an example or template implementation of a relocatable pool to 
> allocate memory from a specific predefined region,
> b) a flag if a pbuf is allocated for RX or for TX and
> c) an example or ready-to-use routine to be used in low_level_output() 
> that can copy a pbuf to new pbufs if the originals cannot be used
>
> The seldom cases where an RX pbuf is reused for TX can be solved by c).
>
> Although this is not a complete solution for zero-copy, it would allow 
> many users for zero-copy DMA which currently have to change the lwIP 
> core code to get that.

I like more the modular approach, that is because those things are 
getting too complex, too fast.

If over time o suitable set of pbuf functions can be found, it would be 
easy to port. Something like AllocTxPbuf, AllocRxPbuf, FreeTxPbuf... and 
so on.

Alain