Re: Call for testers: nsock-engines experimental branch

[Henri Doreau <henri.doreau () greenbone net>]

2011/7/9 Luis Martín García <luis.martingarcia () gmail com>:
> [...]
> I tested nmap and nping and, as I said, they work as expected.

Hi Luis,

thank you very much for testing.

> I'm not sure this is relevant but I've noticed that Nping finishes a bit
> later when I run it with the new nsock than with the old. I attach the
> output of three different runs for each version. Files named newX.txt
> correspond to the new nsock, files named oldX.txt correspond to the old.

Great that you noticed that. I'm not entirely sure yet but I think
that this is due to a non optimized (yet) aspect of the epoll engine,
according to my strace based investigations.

Functions from nsock_core.c call engine->iod_modify() every time they
want to set or clear an event (socket_. This is not optimal as the
engine->iod_modify() function would be able to handle everything in a
single call...
For select this is not really a problem as it simply toggles a bit in
a fd_set: a very cheap operation. For epoll it's more expensive as
epoll_ctl is called...

I'll work on improving this.

Thanks again.


-- 
Henri Doreau |  Greenbone Networks GmbH  |  http://www.greenbone.net
Neuer Graben 17, 49074 Osnabrueck, Germany | AG Osnabrueck, HR B 202460
Executive Directors: Lukas Grunwald, Dr. Jan-Oliver Wagner
_______________________________________________
Sent through the nmap-dev mailing list
http://cgi.insecure.org/mailman/listinfo/nmap-dev
Archived at http://seclists.org/nmap-dev/