Re: [lwip-devel] Re: [task #7040] Work on tcp_enqueue

[Alain M.]

Hi Kieran,

Kieran Mansley escreveu:
> > The third stage would be to find a way to prevent pbuf queues for 
> > multiple small writes: at least one MAC I used was not able to do 
> > scatter-gather, which means although I had a DMA-enabled MAC, I had to 
> > use memcpy to get the packets in one piece -> large, unnecessary 
> > performance hole! This could however require the exact opposite of the 
> > previous point (not splitting up tcp_write and tcp_enqueue).
>
> So we really need to know which is the bigger problem for people using
> lwIP.  I want to avoid going down one road to solve one problem and then
> finding out we've made things worse for another problem.

My suggestion would be to create a compile time option:
a) keep messages n separate pbufs
b) allways concatenate if possible.

There is a real trick behind that: if you have a 32 bit arch like ARM, 
it can be a lot faster if there exists a function concat_data or 
concat_pbufs that by default is memcpy, but is optimyzable. I have a 
book that shows up to 5x speed increment for such functions (without 
even using asm)...

Alain