--- Ben Greear <greearb@candelatech.com> wrote:

quoted

Danial Thom wrote:

quoted

--- Ben Greear <greearb@candelatech.com>

wrote:

quoted

quoted

Danial Thom wrote:

quoted

I think the concensus is that 2.6 has made

trade

quoted

offs that lower raw throughput, which is

what

quoted

quoted

a

quoted

networking device needs. So as a router or
network appliance, 2.6 seems less

suitable.

quoted

A

quoted

quoted

raw

quoted

bridging test on a 2.0Ghz operton system:

FreeBSD 4.9: Drops no packets at 900K pps
Linux 2.4.24: Starts dropping packets at

350K

quoted

quoted

pps

quoted

Linux 2.6.12: Starts dropping packets at

100K

quoted

quoted

pps

I ran some quick tests using kernel 2.6.11,

1ms

quoted

quoted

tick (HZ=1000), SMP kernel.
Hardware is P-IV 3.0Ghz + HT on a new
SuperMicro motherboard with 64/133Mhz
PCI-X bus.  NIC is dual Intel pro/1000. 

Kernel

quoted

quoted

is close to stock 2.6.11.

quoted

What GigE adapters did you use? Clearly

every

quoted

quoted

driver is going to be different. My

experience is

quoted

that a 3.4Ghz P4 is about the performance

of

quoted

a

quoted

2.0Ghz Opteron. I have to try your tuning

script

quoted

tomorrow.

Intel pro/1000, as I mentioned.  I haven't
tried any other
NIC that comes close in performance to the
e1000.

quoted

If your test is still set up, try compiling
something large while doing the test. The

drops

quoted

go through the roof in my tests.

Installing RH9 on the box now to try some
tests...

Disk access always robs networking, in my
experience, so
I am not supprised you see bad ntwk

performance

quoted

while
compiling.

Ben

It would be useful if there were some way to
find
out "what" is getting "robbed". If networking
has
priority, then what is keeping it from getting
back to processing the rx interrupts? 

Ah, the e1000 has built-in interrupt
moderation.
I can't get into my lab until tomorrow
afternoon,
but if you get a chance try setting ITR in
e1000_main.c to something larger, like 20K. and
see if it makes a difference. At 200K pps that
would cause an interrupt every 10 packets,
which
may allow the routine to grab back the cpu more
often.


Danial