Re: [lwip-users] Intention of poll, tcp_output() in tcp_slowtmr()call.

[Larry Piggins]

--- David Aldrich <address@hidden>
wrote:
> Hi,
> 
> I found a thread called "Intention of
> poll,tcp_output() in
> tcp_slowtmr()call." started in January 2003 by
> Michael Portmann that
> describes a problem I have stumbled across.  Here is
> the original
> message:
> 
<snip>
> 

This issue caused me problems for an afternoon too, so
it at least indicates that people's impressions of
this function are often not aligned with its
intentions.

I have an object-oriented layer wrapped around the
callback API, so my solution was to replace tcp_tmr()
with my own version, call tcp_fasttmr() &
tcp_slowtmr()on intervals as I calculate them, and
trap the poll callbacks in the object layer as well,
generating callbacks to the application layer on
specified intervals, effectively separating the poll
callback function from the tcp_output() function.
Maybe there's a reason these are tied together, or
maybe it was just a convenient place to put the call,
but it didn't make much sense to me.

- Larry

__________________________________
Do you Yahoo!?
Yahoo! Search - Find what you’re looking for faster
http://search.yahoo.com