On 2008-11-21, Michael Schnell <[EMAIL PROTECTED]> wrote:
> How did you test the latency ?
I monitor the Ethernet traffic using a second computer running
wireshark (which uses libpcap) and then I look at the
difference between the timestamps on the packet going to the
NIOS board and the reply packet it sends back.
> Did you try to test the latency the system imposes onto a user
> thread with simpler tasks than Ethernet packets (e.g. an
> interrupt from a dedicated timer driver) ?
No. My guess is that the latency is due to delays in the user
thread being woken.
> If the latency is mainly imposed by the network stack, a great
> improvement could be achieved by a "zero-copy-stack".
The delay due to copying network packets should be constant.
When the board is running uClinux it sometimes responds in
1.8ms, and sometimes takes 8ms. When running eCos (which
doesn't have zero-copy network stack), the response is always
between 1.7 and 2.0ms. So, my conclusion is that the extra
time (up to 6ms) is probably the time it takes for the
scheduler to wake up the user task.
> I read that same is available for Linux and is used in routers
> that without same could not be done using Linux. I don't know
> if a zero-copy-stack is easily available and what is necessary
> to port it to uCLinux-NIOS.
--
Grant Edwards grante Yow! ... If I had heart
at failure right now,
visi.com I couldn't be a more
fortunate man!!
_______________________________________________
uClinux-dev mailing list
uClinux-dev@uclinux.org
http://mailman.uclinux.org/mailman/listinfo/uclinux-dev
This message was resent by uclinux-dev@uclinux.org
To unsubscribe see:
http://mailman.uclinux.org/mailman/options/uclinux-dev
