[ovs-dev] [PATCH] gre: Fix ICMP translation for path MTU discovery.

Ben Pfaff blp at nicira.com
Wed Apr 21 16:16:35 UTC 2010


On Tue, Apr 20, 2010 at 05:22:01PM -0400, Jesse Gross wrote:
> The translation of fragmentation-needed messages from outside the
> tunnel to inside didn't quite make the transition from the old
> GRE implementation to the new one intact.  This fixes a number of
> minor bugs in the implmentation.  The primary issues are with computing
> the tunnel header length and comparing the input vs. output values
> for tunnel parameters such as the key.

How thoroughly do you recommend that I should review this?  I looked it
over fairly quickly and didn't see anything that looked bad, but I could
spend some more time on it, if you think I should, and increase my
understanding.

I did notice one place where code was written like this:
        if ((flags & A) && (flags & B))
Usually I would write this as:
        if ((flags & (A|B)) == (A|B))
Maybe you did it that way because the "flags" and "A" and "B" are all
pretty long and would cause the line to wrap in an ugly way.  I suspect
that GCC generates the same code either way, so it probably makes sense
to leave it alone, but I did think it was at least worth pointing out.




More information about the dev mailing list