Re: [lwip-users] httpd slow response

[address@hidden]

Bill Auerbach wrote:
> I see - tcp_output isn't a "send now" - I thought it meant that (if not, why
> do a tcp_write *and* a tcp_output?).  So with tcp_output, Nagle delays are
> not circumvented.  The contrib example doesn't do tcp_output - so is it
> optional?  Why?

I guess this is not really a good example, then: tcp_output is called 
from tcp_in.c after calling the receive callback, so in this case, there 
is no difference. But in general, tcp_enqueue really only enqueues and 
does *not* output, so if you don't call it, data might never get sent.

tcp_outupt *was* a "send now - everything you have, even if it is a very 
small packet" and was called by the sequential APIs only if there was 
enough data was enqueued (that was the nagle algorithm - only for 
netconn/socket API). However, this lead to bad throughput with the raw 
API since it did not implement the nagle algorithm.

With 1.3.0, nagle was moved to tcp_output so that raw API users also can 
use it - and it defaults to *on* (must be explicitly disabled if you 
don't want it - like on all other platforms).