Re: [PATCH] fix iptables on systems with discontiguous processor ids

[Harald Welte <laforge@netfilter.org>]

[-- Attachment #1: Type: text/plain, Size: 1712 bytes --]

On Mon, Oct 10, 2005 at 02:15:18PM -0700, David S. Miller wrote:
> From: Harald Welte <laforge@netfilter.org>
> Date: Mon, 10 Oct 2005 18:41:41 +0200
> 
> > This is my proposed patch for the problem you've described.  Please test
> > and submit.  If it works, I'll also prepare a patch for {arp,ip6}_tables.
> 
> Ebtables needs it too.  "git grep num_possible_cpus" shows all
> of the users in the whole tree, %80 of which are the netfilter
> cases we're fixing here :-)

mh. I'm not very familiar with the ebtables specific changes, and
apparently it differs quite a bit in counter handling.

Anyway, I've tried to come up with a patch that also includes ebtables.

Bart: Could you please comment on it? If you have missed the previous
thread: some systems have smp processor id's "0 2" but no "1", e.g.
num_possible_cpus() is 2 (and we allocate two rulesets), but then we try
to index it with '2', e.g. the non-existing 3rd copy.

Unfortunately I don't have a SMP box during travel, so I can't test
whether it at least still works on 'normal' smp boxes.  So all I can say
is that it looks right, and it compiles with CONFIG_SMP.

Please, could anybody test this on their SMP boxes ASAP?   I won't be
having access to a SMP machine for testing until 2005-10-18, and
presumably we don't want to delay this fix.

-- 
- Harald Welte <laforge@netfilter.org>                 http://netfilter.org/
============================================================================
  "Fragmentation is like classful addressing -- an interesting early
   architectural error that shows how much experimentation was going
   on while IP was being designed."                    -- Paul Vixie

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]