[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
RE: [lwip-users] Which opt.h options effect TCP receive throughput?
From: |
Bill Auerbach |
Subject: |
RE: [lwip-users] Which opt.h options effect TCP receive throughput? |
Date: |
Wed, 23 Jan 2008 13:42:51 -0500 |
Thanks Kieran.
> There are I suppose a couple of theoretical limits:
> - your physical layer bandwidth minus the TCP/IP header overhead
> (likely to be a few percent)
>
> - if you are window limited: bandwidth = TCP_WND / round trip time.
> i.e. if you can increase the TCP_WND or decrease the round trip time you
> should get increased bandwidth.
Isn't 65535 the maximum value for TCP_WND? I got the nice warning added to
init.c when I went above this. I do not know how to decrease the round trip
time.
> You say that "98% of the processor is processing lwIP timers and
> checking for Ethernet packets". That seems like a lot, and if your CPU
> is saturated like that you're probably at the limit of your performance
> without offloading something like checksum calculation if you aren't
> already.
Sorry, I mean 98% of the processor is dedicated to supporting TCP/IP.
Eventually I have to process the incoming data (which will be a small
fraction of CPU time) but for now, nothing else is going except to receive a
2MB image and copy it to a contiguous memory space.
Bill