Re: [PATCH] move tc_u32_mark into pkt_cls.h

["Catalin(ux aka Dino) BOIE" <util@xxxxxxxxxxxxxxx>]

> > It can die, but there are users that use it.
> > So, I think that as long as u32 will stay in kernel, we can let
> > mark_in_u32 live.
> > It's a small modification and I think it may stay.
>
> Everyone told you that it would only be a temporary solution and once
> the meta ematch works flawlessly there is no reason to keep it in the
> tree. In order to give everyone enough time to switch the removal should
> be announced on LARTC mailinglist and commited 2-3 release cycles later.

Yes, you are right, everybody told me. I don't deny.
And I will respect the decision that will be taken.
But I'm wonder what are the gains removing nfmark from u32?
~64 bytes of code and 128 bytes of sources?!
I thing that the gain being in is bigger than the gains resulting from 
removing it.

I will not ask another set of questions. If I'm wrong, let me know and I 
will send, as soon as I can, patches to remove it from kernel (with 3 
releases warning).

Thank you for your time!

---
Catalin(ux aka Dino) BOIE
catab at deuroconsult.ro
http://kernel.umbrella.ro/