public inbox for netdev@vger.kernel.org 
 help / color / mirror / Atom feed
From: ebiederm@xmission•com (Eric W. Biederman)
To: Julian Anastasov <ja@ssi•bg>
Cc: David Miller <davem@davemloft•net>,
	netdev@vger•kernel.org, netfilter-devel@vger•kernel.org,
	Stephen Hemminger <stephen@networkplumber•org>,
	Juanjo Ciarlante <jjciarla@raiz•uncu.edu.ar>,
	Wensong Zhang <wensong@linux-vs•org>,
	Simon Horman <horms@verge•net.au>,
	Pablo Neira Ayuso <pablo@netfilter•org>,
	Patrick McHardy <kaber@trash•net>,
	Jozsef Kadlecsik <kadlec@blackhole•kfki.hu>,
	Jamal Hadi Salim <jhs@mojatatu•com>,
	Steffen Klassert <steffen.klassert@secunet•com>,
	Herbert Xu <herbert@gondor•apana.org.au>
Subject: Re: [PATCH net-next 00/15] Simplify netfilter and network namespaces
Date: Wed, 17 Jun 2015 15:55:25 -0500	[thread overview]
Message-ID: <87wpz25a2q.fsf@x220.int.ebiederm.org> (raw)
In-Reply-To: <alpine.LFD.2.11.1506172216440.2649@ja.home.ssi.bg> (Julian Anastasov's message of "Wed, 17 Jun 2015 22:38:26 +0300 (EEST)")

Julian Anastasov <ja@ssi•bg> writes:

> 	Hello,
>
> On Sun, 14 Jun 2015, Eric W. Biederman wrote:
>
>> This patshset roots out all of the very weird network namespace
>> computation logic (except for the code in ipvs) and fixes it.  I really
>> don't like how the code has been essentially guessing  which network
>> namespace to use.
>> 
>> Probably the worst guessing is in ipvs in the function skb_net. I have
>> some preliminary changes to fix ipvs but they are not quite ready yet.
>> Cleaning up ipvs enough that I can kill skb_net is on my short list.
>
> 	For IPVS skb_net is too complicated. One of
> the first things we do in hook handler is to check
> skb_dst, so even now dev_net(skb_dst(skb)->dev) should
> work.

A couple of things:
- You don't check skb_dst() first thing in all hooks.

- All hooks have state->net after my changes so it gets
  a lot easier.

- I am trying to avoid dev_net(skb_dst(skb)->dev) if I can
  because of my ulterior motive of allowing a destination
  network device in another network namespace.

>  sock_net is used for administration via netlink,
> skb_net is not used there (ip_vs_ctl.c).

Agreed.  sock_net is the right thing for the netlink sockets,
and while obscured by skb_sknet that works fine.

> 	As for removing/replacing skb_net, ip_vs_conn_net()
> is ok when cp is present,

Yes I have seen that.

> we have also svc->net,

And I have seen this.

>  otherwise
> we can store net into struct ip_vs_iphdr, it is filled by
> ip_vs_fill_iph_skb from small number of places when hook
> handler is entered.

Interesting I had not thought of changing ip_vs_iphdr.

I actually have a patch that passes state->net down to where it is
needed or uses ip_vs_conn_net() and it it looks ok, but the patch
is so busy with small tweaks that I need to break it up.

Further I am staring at things and trying to see if there is a way to
make the mass of parameters more readable/obvious.   Something like
the passing of struct nf_hook_state into the netfilter hooks.

Or in other words I want to make the change to ip_vs deadly boring,
obviously correct, and trivial to maintain, and I am not quite there
yet.

Eric

p.s.  I do have my patch that I can toss in your direction if you are
interested.

  reply	other threads:[~2015-06-17 20:55 UTC|newest]

Thread overview: 89+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-06-15  3:07 [PATCH net-next 00/15] Simplify netfilter and network namespaces Eric W. Biederman
2015-06-15  3:12 ` [PATCH net-next 01/15] netfilter: Kill unused copies of RCV_SKB_FAIL Eric W. Biederman
2015-06-15  3:13 ` [PATCH net-next 02/15] netfilter: Pass struct net into the netfilter hooks Eric W. Biederman
2015-06-15  3:13 ` [PATCH net-next 03/15] netfilter: Use nf_hook_state.net Eric W. Biederman
2015-06-15  3:13 ` [PATCH net-next 04/15] ebtables: Simplify the arguments to ebt_do_table Eric W. Biederman
2015-06-15  3:13 ` [PATCH net-next 05/15] inet netfilter: Remove hook from ip6t_do_table, arp_do_table, ipt_do_table Eric W. Biederman
2015-06-15  3:13 ` [PATCH net-next 06/15] inet netfilter: Prefer state->hook to ops->hooknum Eric W. Biederman
2015-06-15  3:13 ` [PATCH net-next 07/15] nftables: kill nft_pktinfo.ops Eric W. Biederman
2015-06-15  3:13 ` [PATCH net-next 08/15] tc: Simplify em_ipset_match Eric W. Biederman
2015-06-15  3:13 ` [PATCH net-next 09/15] x_tables: Pass struct net in xt_action_param Eric W. Biederman
2015-06-15  3:13 ` [PATCH net-next 10/15] x_tables: Use par->net instead of computing from the passed net devices Eric W. Biederman
2015-06-15  3:13 ` [PATCH net-next 11/15] nftables: Pass struct net in nft_pktinfo Eric W. Biederman
2015-06-15  3:13 ` [PATCH net-next 12/15] nf_tables: Use pkt->net instead of computing net from the passed net_devices Eric W. Biederman
2015-06-15  3:13 ` [PATCH net-next 13/15] nf_conntrack: Add a struct net parameter to l4_pkt_to_tuple Eric W. Biederman
2015-06-15  3:13 ` [PATCH net-next 14/15] ipv4: Pass struct net into ip_defrag and ip_check_defrag Eric W. Biederman
2015-06-15  3:13 ` [PATCH net-next 15/15] ipv6: Pass struct net into nf_ct_frag6_gather Eric W. Biederman
2015-06-15  7:06 ` [PATCH net-next 00/15] Simplify netfilter and network namespaces Pablo Neira Ayuso
2015-06-15 15:06   ` Eric W. Biederman
2015-06-15 15:20     ` Pablo Neira Ayuso
2015-06-16  0:10 ` David Miller
2015-06-16  0:26   ` Eric W. Biederman
2015-06-16  2:14     ` David Miller
2015-06-16 10:32     ` Pablo Neira Ayuso
2015-06-16 21:00       ` Eric W. Biederman
2015-06-17 15:09 ` [PATCH net-next 00/43] Simplify netfilter and network namespaces (take 2) Eric W. Biederman
2015-06-17 15:28   ` [PATCH net-next 01/43] netfilter: Kill unused copies of RCV_SKB_FAIL Eric W. Biederman
2015-06-17 15:28   ` [PATCH net-next 02/43] netfilter: Pass struct net into the netfilter hooks Eric W. Biederman
2015-06-17 15:28   ` [PATCH net-next 03/43] netfilter: Use nf_hook_state.net Eric W. Biederman
2015-06-17 15:28   ` [PATCH net-next 04/43] ebtables: Simplify the arguments to ebt_do_table Eric W. Biederman
2015-06-17 15:28   ` [PATCH net-next 05/43] inet netfilter: Remove hook from ip6t_do_table, arp_do_table, ipt_do_table Eric W. Biederman
2015-06-17 15:28   ` [PATCH net-next 06/43] inet netfilter: Prefer state->hook to ops->hooknum Eric W. Biederman
2015-06-17 15:28   ` [PATCH net-next 07/43] nftables: kill nft_pktinfo.ops Eric W. Biederman
2015-06-17 15:28   ` [PATCH net-next 08/43] tc: Simplify em_ipset_match Eric W. Biederman
2015-06-17 15:28   ` [PATCH net-next 09/43] x_tables: Pass struct net in xt_action_param Eric W. Biederman
2015-06-17 15:28   ` [PATCH net-next 10/43] x_tables: Use par->net instead of computing from the passed net devices Eric W. Biederman
2015-06-17 15:28   ` [PATCH net-next 11/43] nftables: Pass struct net in nft_pktinfo Eric W. Biederman
2015-06-17 15:28   ` [PATCH net-next 12/43] nf_tables: Use pkt->net instead of computing net from the passed net_devices Eric W. Biederman
2015-06-17 15:28   ` [PATCH net-next 13/43] nf_conntrack: Add a struct net parameter to l4_pkt_to_tuple Eric W. Biederman
2015-06-17 15:28   ` [PATCH net-next 14/43] ipv4: Pass struct net into ip_defrag and ip_check_defrag Eric W. Biederman
2015-06-17 15:28   ` [PATCH net-next 15/43] ipv6: Pass struct net into nf_ct_frag6_gather Eric W. Biederman
2015-06-17 15:28   ` [PATCH net-next 16/43] net: include missing headers in net/net_namespace.h Eric W. Biederman
2015-06-17 15:28   ` [PATCH net-next 17/43] netfilter: use forward declaration instead of including linux/proc_fs.h Eric W. Biederman
2015-06-17 15:28   ` [PATCH net-next 18/43] netfilter: don't pull include/linux/netfilter.h from netns headers Eric W. Biederman
2015-06-17 15:28   ` [PATCH net-next 19/43] ipvs: Read hooknum from state rather than ops->hooknum Eric W. Biederman
2015-06-17 15:28   ` [PATCH net-next 20/43] netfilter: Pass priv instead of nf_hook_ops to netfilter hooks Eric W. Biederman
2015-06-17 15:28   ` [PATCH net-next 21/43] netfilter: Add a network namespace Kconfig conflict Eric W. Biederman
2015-06-17 15:28   ` [PATCH net-next 22/43] netfilter: Add a struct net parameter to nf_register_hook[s] Eric W. Biederman
2015-06-17 15:28   ` [PATCH net-next 23/43] netfilter: Add a struct net parameter to nf_unregister_hook[s] Eric W. Biederman
2015-06-17 15:28   ` [PATCH net-next 24/43] netfilter: Make the netfilter hooks per network namespace Eric W. Biederman
2015-06-17 15:28   ` [PATCH net-next 25/43] netfilter: Make nf_hook_ops just a parameter structure Eric W. Biederman
2015-06-17 15:28   ` [PATCH net-next 26/43] netfitler: Remove spurios included of netfilter.h Eric W. Biederman
2015-06-17 15:28   ` [PATCH net-next 27/43] x_tables: Add magical hook registration in the common case Eric W. Biederman
2015-06-17 15:28   ` [PATCH net-next 28/43] x_tables: Where possible convert to the new hook registration method Eric W. Biederman
2015-06-17 15:28   ` [PATCH net-next 29/43] x_tables: Kill xt_[un]hook_link Eric W. Biederman
2015-06-17 15:28   ` [PATCH net-next 30/43] x_tables: Update ip?table_nat to register their hooks in all network namespaces Eric W. Biederman
2015-06-17 15:28   ` [PATCH net-next 31/43] netfilter: nf_tables: adapt it to pernet hooks Eric W. Biederman
2015-06-17 15:28   ` [PATCH net-next 32/43] netfilter: ipt_CLUSTERIP: adapt it to support " Eric W. Biederman
2015-06-17 15:28   ` [PATCH net-next 33/43] netfilter: ebtables: adapt the filter and nat table to " Eric W. Biederman
2015-06-17 15:28   ` [PATCH net-next 34/43] netfilter: bridge: adapt it " Eric W. Biederman
2015-06-17 15:28   ` [PATCH net-next 35/43] ipvs: Register netfilter hooks in all network namespaces Eric W. Biederman
2015-06-17 15:28   ` [PATCH net-next 36/43] netfilter: nf_conntract: " Eric W. Biederman
2015-06-17 15:28   ` [PATCH net-next 37/43] netfilter: nf_defrag: " Eric W. Biederman
2015-06-17 15:28   ` [PATCH net-next 38/43] netfilter: synproxy: " Eric W. Biederman
2015-06-17 15:28   ` [PATCH net-next 39/43] selinux: adapt it to pernet hooks Eric W. Biederman
2015-06-17 15:28   ` [PATCH net-next 40/43] smack: " Eric W. Biederman
2015-06-17 15:28   ` [PATCH net-next 41/43] netfilter: Remove the network namespace Kconfig conflict Eric W. Biederman
2015-06-17 15:28   ` [PATCH net-next 42/43] netfilter bridge: Make the sysctl knobs per network namespace Eric W. Biederman
2015-06-17 15:28   ` [PATCH net-next 43/43] netfilter: Skip unnecessary calls to synchronize_net Eric W. Biederman
2015-06-17 17:20     ` Patrick McHardy
2015-06-17 20:32       ` Eric W. Biederman
2015-06-18 19:40   ` [PATCH net-next 00/43] Simplify netfilter and network namespaces (take 2) Pablo Neira Ayuso
2015-07-10 23:11   ` [PATCH -next 0/6] Per network namespace netfilter chains Eric W. Biederman
2015-07-10 23:12     ` [PATCH -next 1/6] netfilter: nf_queue: Don't recompute the hook_list head Eric W. Biederman
2015-07-10 23:13     ` [PATCH -next 2/6] netfilter: kill nf_hooks_active Eric W. Biederman
2015-07-10 23:13     ` [PATCH -next 3/6] netfilter: Simply the tests for enabling and disabling the ingress queue hook Eric W. Biederman
2015-07-10 23:14     ` [PATCH -next 4/6] netfilter: Factor out the hook list selection from nf_register_hook Eric W. Biederman
2015-07-10 23:15     ` [PATCH -next 5/6] netfilter: Per network namespace netfilter hooks Eric W. Biederman
2015-07-15 19:00       ` Pablo Neira Ayuso
2015-07-15 20:22         ` Eric W. Biederman
2015-07-10 23:15     ` [PATCH -next 6/6] netfilter: nftables: Only run the nftables chains in the proper netns Eric W. Biederman
2015-07-15 17:20     ` [PATCH -next 0/6] Per network namespace netfilter chains Pablo Neira Ayuso
2015-07-15 20:05       ` Eric W. Biederman
2015-07-16 11:01         ` Pablo Neira Ayuso
2015-06-17 19:38 ` [PATCH net-next 00/15] Simplify netfilter and network namespaces Julian Anastasov
2015-06-17 20:55   ` Eric W. Biederman [this message]
2015-06-17 22:01     ` Julian Anastasov
2015-06-18 14:45       ` Eric W. Biederman
2015-06-18 19:21         ` Julian Anastasov
2015-06-19 14:24           ` Eric W. Biederman

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=87wpz25a2q.fsf@x220.int.ebiederm.org \
    --to=ebiederm@xmission$(echo .)com \
    --cc=davem@davemloft$(echo .)net \
    --cc=herbert@gondor$(echo .)apana.org.au \
    --cc=horms@verge$(echo .)net.au \
    --cc=ja@ssi$(echo .)bg \
    --cc=jhs@mojatatu$(echo .)com \
    --cc=jjciarla@raiz$(echo .)uncu.edu.ar \
    --cc=kaber@trash$(echo .)net \
    --cc=kadlec@blackhole$(echo .)kfki.hu \
    --cc=netdev@vger$(echo .)kernel.org \
    --cc=netfilter-devel@vger$(echo .)kernel.org \
    --cc=pablo@netfilter$(echo .)org \
    --cc=steffen.klassert@secunet$(echo .)com \
    --cc=stephen@networkplumber$(echo .)org \
    --cc=wensong@linux-vs$(echo .)org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox