Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1496100 > unrolled thread

Re: [GIT] Networking

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  Re: [GIT] Networking Linus Torvalds <torvalds@linux-foundation.org> - 2016-10-06 00:40 +0200
    Re: [GIT] Networking Pablo Neira Ayuso <pablo@netfilter.org> - 2016-10-06 02:20 +0200
      Re: [GIT] Networking David Miller <davem@davemloft.net> - 2016-10-06 02:30 +0200
    Re: [GIT] Networking Stephen Rothwell <sfr@canb.auug.org.au> - 2016-10-06 03:00 +0200
      Re: [GIT] Networking Linus Torvalds <torvalds@linux-foundation.org> - 2016-10-06 04:20 +0200
        Re: [GIT] Networking David Miller <davem@davemloft.net> - 2016-10-06 05:00 +0200
          Re: [GIT] Networking Stephen Rothwell <sfr@canb.auug.org.au> - 2016-10-06 07:10 +0200
        Re: [GIT] Networking Stephen Rothwell <sfr@canb.auug.org.au> - 2016-10-06 05:00 +0200

#1496100 — Re: [GIT] Networking

On Wed, Oct 5, 2016 at 3:29 PM, Stephen Rothwell <sfr@canb.auug.org.au> wrote:
>
> I have been carrying the following merge fix patch (for the merge of
> the net-next tree with Linus' tree) for a while now which seems to have
> got missed:

Ugh. It doesn't seem to be a merge error, because that double iph
assignment came from the original patch that introduced this function:
commit ddc8b6027ad0 ("netfilter: introduce nft_set_pktinfo_{ipv4,
ipv6}_validate()").

So I wouldn't call it a merge error - it just looks like a bug in the
network layer. So I'm not going to apply your patch even though it
looks plausible to me, simply because it's outside my area of
expertise.

David? Pablo?

           Linus

[toc] | [next] | [standalone]

#1496123

On Wed, Oct 05, 2016 at 03:37:17PM -0700, Linus Torvalds wrote:
> On Wed, Oct 5, 2016 at 3:29 PM, Stephen Rothwell <sfr@canb.auug.org.au> wrote:
> >
> > I have been carrying the following merge fix patch (for the merge of
> > the net-next tree with Linus' tree) for a while now which seems to have
> > got missed:
> 
> Ugh. It doesn't seem to be a merge error, because that double iph
> assignment came from the original patch that introduced this function:
> commit ddc8b6027ad0 ("netfilter: introduce nft_set_pktinfo_{ipv4,
> ipv6}_validate()").
> 
> So I wouldn't call it a merge error - it just looks like a bug in the
> network layer. So I'm not going to apply your patch even though it
> looks plausible to me, simply because it's outside my area of
> expertise.
> 
> David? Pablo?

This looks good, please take it so we speed up things.

Acked-by: Pablo Neira Ayuso <pablo@netfilter.org>

Thanks!

P.S: Sorry for not addressing this any sooner, traveling overhead,
conferente and unstable wifi connection has been a problem here.

[toc] | [prev] | [next] | [standalone]

#1496125

From: Pablo Neira Ayuso <pablo@netfilter.org>
Date: Thu, 6 Oct 2016 02:09:45 +0200

> On Wed, Oct 05, 2016 at 03:37:17PM -0700, Linus Torvalds wrote:
>> On Wed, Oct 5, 2016 at 3:29 PM, Stephen Rothwell <sfr@canb.auug.org.au> wrote:
>> >
>> > I have been carrying the following merge fix patch (for the merge of
>> > the net-next tree with Linus' tree) for a while now which seems to have
>> > got missed:
>> 
>> Ugh. It doesn't seem to be a merge error, because that double iph
>> assignment came from the original patch that introduced this function:
>> commit ddc8b6027ad0 ("netfilter: introduce nft_set_pktinfo_{ipv4,
>> ipv6}_validate()").
>> 
>> So I wouldn't call it a merge error - it just looks like a bug in the
>> network layer. So I'm not going to apply your patch even though it
>> looks plausible to me, simply because it's outside my area of
>> expertise.
>> 
>> David? Pablo?
> 
> This looks good, please take it so we speed up things.
> 
> Acked-by: Pablo Neira Ayuso <pablo@netfilter.org>

Applied.

[toc] | [prev] | [next] | [standalone]

#1496130

Hi Linus,

On Wed, 5 Oct 2016 15:37:17 -0700 Linus Torvalds <torvalds@linux-foundation.org> wrote:
>
> On Wed, Oct 5, 2016 at 3:29 PM, Stephen Rothwell <sfr@canb.auug.org.au> wrote:
> >
> > I have been carrying the following merge fix patch (for the merge of
> > the net-next tree with Linus' tree) for a while now which seems to have
> > got missed:  
> 
> Ugh. It doesn't seem to be a merge error, because that double iph
> assignment came from the original patch that introduced this function:
> commit ddc8b6027ad0 ("netfilter: introduce nft_set_pktinfo_{ipv4,
> ipv6}_validate()").

Except that commit effectively moved that function from
net/netfilter/nf_tables_netdev.c to
include/net/netfilter/nf_tables_ipv4.h while commit c73c24849011
("netfilter: nf_tables_netdev: remove redundant ip_hdr assignment")
removed the assignment in the original file (and has been in your tree
since v4.8-rc7) and that is where I originally actually got a conflict.

> So I wouldn't call it a merge error - it just looks like a bug in the
> network layer. So I'm not going to apply your patch even though it
> looks plausible to me, simply because it's outside my area of
> expertise.

no worries.
-- 
Cheers,
Stephen Rothwell

[toc] | [prev] | [next] | [standalone]

#1496140

On Wed, Oct 5, 2016 at 5:52 PM, Stephen Rothwell <sfr@canb.auug.org.au> wrote:
>
> Except that commit effectively moved that function from
> net/netfilter/nf_tables_netdev.c to
> include/net/netfilter/nf_tables_ipv4.h while commit c73c24849011
> ("netfilter: nf_tables_netdev: remove redundant ip_hdr assignment")
> removed the assignment in the original file (and has been in your tree
> since v4.8-rc7) and that is where I originally actually got a conflict.

Oh, interesting. Why didn't I get the conflict there then?

I'm guessing (but too lazy to actually look up the history), that
David ended up doing that merge and that ends up being why I never saw
a conflict.

               Linus

[toc] | [prev] | [next] | [standalone]

#1496147

From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Thu, 6 Oct 2016 13:51:52 +1100

> On Wed, 5 Oct 2016 19:14:21 -0700 Linus Torvalds <torvalds@linux-foundation.org> wrote:
>>
>> On Wed, Oct 5, 2016 at 5:52 PM, Stephen Rothwell <sfr@canb.auug.org.au> wrote:
>> >
>> > Except that commit effectively moved that function from
>> > net/netfilter/nf_tables_netdev.c to
>> > include/net/netfilter/nf_tables_ipv4.h while commit c73c24849011
>> > ("netfilter: nf_tables_netdev: remove redundant ip_hdr assignment")
>> > removed the assignment in the original file (and has been in your tree
>> > since v4.8-rc7) and that is where I originally actually got a conflict.  
>> 
>> Oh, interesting. Why didn't I get the conflict there then?
>> 
>> I'm guessing (but too lazy to actually look up the history), that
>> David ended up doing that merge and that ends up being why I never saw
>> a conflict.
> 
> Yeah, commit b50afd203a5e ("Merge
> git://git.kernel.org/pub/scm/linux/kernel/git/davem/net") actually
> merges v4.8 into the net-next tree.

Yes, this is where the change got lost.

I have all of the fixups queued up in my net tree and will send in a pull
request later.

[toc] | [prev] | [next] | [standalone]

#1496159

Hi Dave,

On Wed, 05 Oct 2016 22:56:12 -0400 (EDT) David Miller <davem@davemloft.net> wrote:
>
> Yes, this is where the change got lost.

No worries.

> I have all of the fixups queued up in my net tree and will send in a pull
> request later.

Thanks.

-- 
Cheers,
Stephen Rothwell

[toc] | [prev] | [next] | [standalone]

#1496148

Hi Linus,

On Wed, 5 Oct 2016 19:14:21 -0700 Linus Torvalds <torvalds@linux-foundation.org> wrote:
>
> On Wed, Oct 5, 2016 at 5:52 PM, Stephen Rothwell <sfr@canb.auug.org.au> wrote:
> >
> > Except that commit effectively moved that function from
> > net/netfilter/nf_tables_netdev.c to
> > include/net/netfilter/nf_tables_ipv4.h while commit c73c24849011
> > ("netfilter: nf_tables_netdev: remove redundant ip_hdr assignment")
> > removed the assignment in the original file (and has been in your tree
> > since v4.8-rc7) and that is where I originally actually got a conflict.  
> 
> Oh, interesting. Why didn't I get the conflict there then?
> 
> I'm guessing (but too lazy to actually look up the history), that
> David ended up doing that merge and that ends up being why I never saw
> a conflict.

Yeah, commit b50afd203a5e ("Merge
git://git.kernel.org/pub/scm/linux/kernel/git/davem/net") actually
merges v4.8 into the net-next tree.

-- 
Cheers,
Stephen Rothwell

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web