[Click] RX Overruns with eepro100

[Eddie Kohler]

Hi Barry,

> How do you calculate the PCI max throughput?
> I would have thought that 32 bit * 33Mhz would be
> 1056 Mbps availble throughput on the PCI bus. 

I got the case wrong (read MB, wrote Mb); you are correct about the
theoretical maximum.

> The reason why I am persuing this is because I have
> notes recording that I was able to achieve 8000
> 1500byte packets per second on an older setup. The
> only difference was the version of Click and a
> different batch of eepro100 cards. I've replaced the
> old version of Click but can't reproduce the rates. 

Although the numbers were wrong, my advice from the earlier mail still
stands:

You could probably distinguish PCI overload from CPU overload by checking
out some of the cards' counters, in combination with the per-device
"cycles" information you get from compiling with
CLICK_DEVICE_STATS. For instance, if the PollDevices *always* succeed (the
incoming DMA rings are never empty), that's a good sign that the CPU is
overloaded. There was a thread on the list recently about
CLICK_DEVICE_STATS et al;
https://amsterdam.lcs.mit.edu/pipermail/click/2003-June/001986.html

Eddie