Fix various IPV6_CHECKSUM issues

[tuexen]

Summary

This patch fixes the following issues related to the IPPROTO_IPV6 level socket option IPV6_CHECKSUM:

• When setting the offset, don't allow negative values except -1, since they are not supported.
• When receiving a packet with an incorrect checksum, just discard it and don't send an ICMP message indicating a parameter problem.
• When computing the checksum for a packet to be sent, ensure you have enough bytes in the packet to write the checksum. The existing check was wrong (off by one error) and would result in writing one byte after the packet buffer.
• When computing the checksum for a received packet, ensure that you have enough bytes in the packet such that it contains the checksum. This fixes in particular syzbot, which was injecting an empty (plen = 0) packet, which resulted in a panic.

[emaste]

sys/netinet6/ip6_output.c

2224–2228

Can probably elide the optval >= 0

[bz]

This seems like four different issues and it's kind of hard to keep them apart in a single change. For the sake of having a readable history and easily seeing/understanding each problem, can you please split them up?

[tuexen]

In D19965#429164, @bz wrote:

This seems like four different issues and it's kind of hard to keep them apart in a single change. For the sake of having a readable history and easily seeing/understanding each problem, can you please split them up?

Split up in four reviews or committing them separately once approved?

[tuexen]

Address Ed's comment

[tuexen]

sys/netinet6/ip6_output.c

2224–2228

Fixed. Thanks for noting it.

[bz]

In D19965#429167, @tuexen wrote:

Split up in four reviews or committing them separately once approved?

For review already if possible. Means I need less brain cycles now to divide the four apart as well.

[tuexen]

In D19965#429179, @bz wrote:

In D19965#429167, @tuexen wrote:

Split up in four reviews or committing them separately once approved?

For review already if possible. Means I need less brain cycles now to divide the four apart as well.

Splitted up in:

1. D19966
2. D19967
3. D19968
4. D19969

I hope this speeds up the review process...

[tuexen]

All for reviews have been committed. Thanks a lot for the very quick reviews!