[P4] Fwd: T4P4S

[Sándor Laki]

Ez pár napja jött.



-------- Továbbított üzenet --------
Tárgy: 	T4P4S
Dátum: 	Wed, 15 May 2019 17:48:39 -0400
Feladó: 	Nik Sultana <nsultana at seas.upenn.edu>
Címzett: 	info at p4.elte.hu



Hi, as part of a research project we're trying to find a way of running
P4 programs on CPU and maximising performance to carry out baseline
experiments.

To keep our research focussed we'd rather use an existing tool rather
than build it ourselves, and were wondering if T4P4S is the best way of
doing this.

We have Intel XL710 cards, already use DPDK in a different part of the
project, and our team has extensive experience using P4, but we haven't
got any practical experience of using T4P4S yet, so I thought I'd get in
touch.

The examples provided in the publications linked from http://p4.elte.hu/
are interested but they tend to be simple -- e.g., L2 or L3 forwarding
-- which don't seem to push P4 program syntax very hard, and this made
me unsure about T4P4S' suitability in our case.

To help me get a better understanding of T4P4S I'd like to ask a few 
quick questions:
1. Can T4P4S compile any P4 program to run over DPDK, or does it only 
support a supset of P4 program features?
2. Does T4P4S support P4 extern functions and/or extern objects? Given 
that these functions are typically written in something like C, is it 
straightforward to compile/link them together with the output generated 
by T4P4S?
3. What are T4P4S limitations?
4. Has T4P4S been used externally to ELTE before, as part of somebody 
else's research project?

Thanks for your time,
Nik Sultana


---
Ezt az e-mailt az Avast víruskereső szoftver átvizsgálta.
https://www.avast.com/antivirus
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://plc.inf.elte.hu/pipermail/p4/attachments/20190518/e201646c/attachment.html>