Factor out ip6_deletefraghdr()

[kp]

Summary

It'll be used by the pf fragment handling code.

[glebius]

Thanks. Patch looks good to me, but see comments. We probably need a nod from a IPv6 expert. I'll send link to ae@.

sys/netinet6/ip6_output.c

1215

Comment should end in dot.

1217

Comment should end in dot.

1218

Can you please use bcopy() here. And you don't need first caddr_t cast. The second cast should be changed to char *.

1223

Please capitalize comment and end it in dot.

1224

Let's add third argument 'wait' to ip6_deletefraghdr(). The argument will be passed directly to m_split().

sys/netinet6/ip6_var.h

391

Need single whitespace between 'mbuf' and '*'.

[kp]

Most of the remarks apply to the previous code as well.
I have a minor preference for the patch to be a straight move of the existing code. If you feel strongly we should to take the opportunity to do style cleanups I will of course do so.
I could also do the cleanups in a separate patch.

[ae]

I have no objections against it, the code looks similar to previous implementation.

[glebius]

Damn f*ckbrikator doesn't allow me to close the revision, since I don't own it.

Kristof, looks like you will need to manually close all your revisions as I commit them. Or we can just leave some trash in this "pretty" software.

[kp]

Submitted as r278828 - head/sys/netinet6

[rpaulo]

In D1764#17, @glebius wrote:

Damn f*ckbrikator doesn't allow me to close the revision, since I don't own it.

Kristof, looks like you will need to manually close all your revisions as I commit them. Or we can just leave some trash in this "pretty" software.

You need to use the full URL in your commit log in order to trigger the script that closes revisions. That's how it was setup for FreeBSD and is unrelated to the Phabricator software. I agree it shouldn't be possible to close revisions that you don't own, otherwise it would be extremely easy to screw up the system.