[lwip-users] Which opt.h options effect TCP receive throughput?

[Bill Auerbach]

Hello,

 

I’m still benchmarking lwIP TCP (PPC Power Arch) while receiving 2MB files to the PPC.  I’m using raw api (call backs) and NO_SYS.  98% of the processor is processing lwIP timers and checking for Ethernet packets and the callbacks are doing the work.  I have not moved to Jumbo (Huge) frames yet – I want to get standard packets as fast as possible.

 

Interestingly, copying from the pbufs to contiguous memory is using about 10% of the bandwidth *including* using DMA to move the data (it was 40% with memcpy and 20% with an optimized memcpy).

 

Right now I’m at 464MbS for a 2MB transfer.  Finally the question of the subject line: Which lwIP options most effect receive throughput?  Is it just these, which I’ve set as follows:

 

#define TCP_WND                         (64*1024-1)

#define TCP_MSS                         (8*1024)

 

One thing I notice is there is an ACK sent for every 2 packets received.  Does this sound right?  (2874 received, 1437 sent).

 

I tried a direct point to point link (PC to PPC box) with no change in the results.  So the router is giving this burst everything – in fact I see no intervening packets in wireshark from the first to last TCP packet in my transfer (this surprised me actually).

 

What is the theoretical limit with TCP?  Is it the PC slowing things down, or am I simply close to “as good as it gets”?

 

Thanks,

Bill Auerbach