Steve Grubb wrote:
> On Friday 26 May 2006 17:33, Paul Moore wrote:
>>For the curious this is the change between the lspp.29 patch and the one
>>attached to this mail, from the include/net/netlabel.h file:
> 
> Thanks for the patch update. lspp.31 is building, but I can confirm its 
> working.
> 

Thanks for adding it to lspp.31, sorry about the x86_64 problems with
the earlier patch.

I've uncovered some problems while running my poor-man's stress test on
the x86_64 kernels but that shouldn't get in the way of normal usage.
I'll post an update once I have resolved the issue.

> But looking at the patch...I have some questions. (Normally, I would put 
> comments inline with the patch, but this patch is too big for that.)
> 
> 1) How is errors sent back to user space when a metlink command does not work?

All of the NetLabel commands sent over netlink are ACK'd to some extent;
either through an explicit ACK message, and error message, or a data
message.  Check out the header files in net/netlabel to get a better
idea of the "protocol".

> 2) Are there any capabilities checks needed to issue commands to the 
> subsystem? Or can any user send commands?

Not yet, although there will need to be for obvious reasons.
Comments/suggestions/examples/etc are always welcome.

> 3) How carefully is the data from user space being checked before using it? 
> Usually it needs to be carefully checked and then used.

I think I'm doing a pretty through job of checking the commands from
user-space before acting on them, however, I haven't yet done any exotic
corner case testing so I'm sure there is a field or two waiting to panic
my test system ;)

Is this comment/question based on something you saw in the patch or just
a general concern about kernel-user interaction?  If there is something
specific please let me know.

> 4) Do you need to use NLMSG_ALIGN macro when checking sizes of things?

>From my understanding of netlink the start of the data portion needs to
be aligned but I think that is it ... however, I could very easily be
wrong here - anyone?

> 5) Should you be using NETLINK_CB(skb).pid instead of nl_hdr->nlmsg_pid?

I don't know, should I?  I haven't found much *good* documentation on
netlink so what I have in the patch is cobbled together from some old
Linux Journal article as well as a few other pieces of existing kernel
code.  As a result I'm sure there are a few dumb things, I just don't
know what they are, yet.

If you, or anyone else reading this, knows for certain please shoot me
some mail.

> Also, is there a user space tool that you are testing the netlink interface 
> with ?

Yep, I've posted a link to this a few times but it may have gotten
overlooked, here it is:

http://free.linux.hp.com/~pmoore/projects/linux_cipso/netlabel_tools-05032006.tar.gz

It's still a little crude but it works.

-- 
paul moore
linux security @ hp

--
redhat-lspp mailing list
[email protected]
https://www.redhat.com/mailman/listinfo/redhat-lspp

Reply via email to