socket options and setsockopt.c src location

Posted on 2003-02-20
Medium Priority
Last Modified: 2013-12-09
I am having trouble finding the source code in the openBSD source tree for for following functions:

socket.c (if it exists)

Also, I am going to need to add a new socket option and have been unable to locate the source code locations that pertain to this (ie: where is the select structure that process the hex codes for socket options found in socket.h? where is the code located for individual socket options? where would code for a new socket option go?)

Question by:jellis613
Welcome to Experts Exchange

Add your voice to the tech community where 5M+ people just like you are talking about what matters.

  • Help others & share knowledge
  • Earn cash & points
  • Learn & ask questions
  • 3

Accepted Solution

Crossley earned 500 total points
ID: 8004829
Yes it certainly does exist.  But firstly (and apologies if this sounds like a stupid question) did you load the source code (seperate CD) when you installed OpenBSD?  If not it's easy enough to do subsequently (ask if you need help to do so).

Anyhow, if you do have the source installed you should be able to find the paths to these commands using 'locate socket.c' (look for hits in the /usr/src directories) - or if that doesn't work try 'find /usr/src -name socket.c -print'.

For example on a machine with the 3.2 build, you should find getsockopt.c at /usr/src/lib/libc_r/uthread/uthread_getsockopt.c - which begins:

#include <sys/types.h>
#include <sys/socket.h>
#include <pthread.h>
#include "pthread_private.h"

getsockopt(int fd, int level, int optname, void *optval, socklen_t *optlen)
        int             ret;

        if ((ret = _FD_LOCK(fd, FD_RDWR, NULL)) == 0) {
                ret = _thread_sys_getsockopt(fd, level, optname, optval, optlen)


It sounds like you're attempting an interesting (and ambitious) project - might I ask what you're trying to achieve?  I presume you're fully familar with 'TCPIP protocols - the implementations' by Stevens - if not I'd get a copy quick!

Rgds & good luck, M

Author Comment

ID: 8004988
Thanks, It seems that the code that we are looking for is not in upper level code like socket.c, setsockopt, etc...we are going to need to make some changes at the kernel level to achieve our goal.

What we are trying to do is set up an ip-filter at the kernel level of openBSD. To do this we must implement a new socket option which takes a list of ip-addresses for which to drop UDP (not TCP) packets. We would pass this list of addresses through the setsockopt function.

We are right now unable to locate:
- where socket options are implemented in the kernel code (and where we could add in a new option)
- an example of a place where a packet drop/send decision are made at the kernel level
- the selection mechanism that sorts through socket option calls and implements the desired socket option (like for those that are listed in socket.h)


Expert Comment

ID: 8011130
Thanks for the extra info - interesting.

I think if I were you I would be trying to find a way of achieving this that didn't involve hacking the kernel though...  you could certainly trivially achieve the filtering effect you describe via OpenBSD's built in kernel-level packet filter ('pf').  If you aren't already familiar with it, I would strongly recommend reading the manpage.  In fact, if you have a look at the source for pf you will see how it links into the kernel to decide how to drop/send a packet.

I presume you have a compelling reason to want to set the 'deny' list via a socket option (compatability with an existing application?  Remote control?).  One thing you might want to consider is an auxiliary program that listens on a different socket on the same port, and talks appropriately to the PF daemon.

Rgds, M

Expert Comment

ID: 8011238
Further to the above you can control pf (e.g. to add/delete filter rules) via ioctls to the /dev/pf device.  See http://www.openbsd.org/cgi-bin/man.cgi?query=pf&apropos=0&sektion=0&manpath=OpenBSD+Current&arch=i386&format=html  for details.

The normal means of generating these ioctls is with the userland pfctl command, but you could equally well do it with a program of course.

If you are definate about trying to do your job via a new socket option, then IMHO I think you should leave the actual filtering to pf, and make the objective of your patching activities with socket.c etc to generate the necessary ioctl calls....


Featured Post

Industry Leaders: We Want Your Opinion!

We value your feedback.

Take our survey and automatically be enter to win anyone of the following:
Yeti Cooler, Amazon eGift Card, and Movie eGift Card!

Question has a verified solution.

If you are experiencing a similar issue, please ask a related question

In tuning file systems on the Solaris Operating System, changing some parameters of a file system usually destroys the data on it. For instance, changing the cache segment block size in the volume of a T3 requires that you delete the existing volu…
Introduction Regular patching is part of a system administrator's tasks. However, many patches require that the system be in single-user mode before they can be installed. A cluster patch in particular can take quite a while to apply if the machine…
Learn several ways to interact with files and get file information from the bash shell. ls lists the contents of a directory: Using the -a flag displays hidden files: Using the -l flag formats the output in a long list: The file command gives us mor…
Learn how to find files with the shell using the find and locate commands. Use locate to find a needle in a haystack.: With locate, check if the file still exists.: Use find to get the actual location of the file.:
Suggested Courses

777 members asked questions and received personalized solutions in the past 7 days.

Join the community of 500,000 technology professionals and ask your questions.

Join & Ask a Question