[Ns-bugs] [Bug 491] It is painful to enable all checksums

[code@nsnam.ece.gatech.edu]

http://www.nsnam.org/bugzilla/show_bug.cgi?id=491





--- Comment #3 from Tom Henderson <tomh at tomh.org>  2009-04-27 01:08:41 EDT ---
(In reply to comment #2)
> I got bit by this again today.  V4Ping creates an icmp header outside of the
> Icmpv4L4Protocol where there is a calcChecksum.  Checksums in V4Ping aren't
> turned on when you turn on checksums in icmp.
> 
> It's time to fix this.
> 
> I don't think that adding a calChecksum to InternetStackHelper is the right
> thing to do.  There is no law that says that a script must ever instantiate an
> internet stack helper.  We need something that turns on checksums irrespective
> of what level script one is using.  
> 
> Like I said, I think there should be a global enable-checksums option that
> trumps the individual checksum attributes and just turns on all checksums in
> the system.  All checksum enabling code should respect this global value.
> 
> I would like to fix this tomorrow to get emu-ping working.  How about adding a
> GlobalValue in the internet stack module?
> 


I would like to see this feature so that there is a single-line or almost
single-line way to control this behavior.

I do not have a strong opinion on whether it is a global value or a helper; I
see your argument but also could argue that global value is not the typical way
we solve these types of problems in our system, and that it should be trivial
to instantiate an Internet stack helper (or simply make it a static class
method) to get access to this capability.  


-- 
Configure bugmail: http://www.nsnam.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.