Re: [lwip-users] Re transmission problem?

[Simon Goldschmidt]

Phil Perryman <address@hidden> wrote:
> This yields a 6% increase in bandwidth over using the sys_check_timeouts
> method and having an interrupt which triggers every 10ms in order to
> generate a timestamp for the sys_now function.
> 
> Are there any disadvantages to using the faster method compared to using
> sys_check_timeouts? ie. is it less reliable?

It's not less reliable. The function we provide is meant to make it easy to set 
up a running system. Of course there's often room for improvement at the cost 
of less portable code. When not using our function, you have to make sure ALL 
timers get called at the defined interval, TCP timers are not the only timers 
you have to call.

Simon
>