Discussion:
dead code in netinet6 (MIP6?)
(too old to reply)
c***@sdf.org
2019-07-11 20:31:56 UTC
Permalink
hi folks,

what is this about? do we want to keep it? :-)

thanks.

--
Posted automagically by a mail2news gateway at muc.de e.V.
Please direct questions, flames, donations, etc. to news-***@muc.de
Robert Swindells
2019-07-11 20:43:23 UTC
Permalink
Post by c***@sdf.org
what is this about? do we want to keep it? :-)
Where ?

I have the MobileIPv6 implementation from KAME in my tree so can't do
a search to see if any of it has got committed already by mistake.

--
Posted automagically by a mail2news gateway at muc.de e.V.
Please direct questions, flames, donations, etc. to news-***@muc.de
Robert Swindells
2019-07-11 21:59:28 UTC
Permalink
Post by Robert Swindells
Post by c***@sdf.org
what is this about? do we want to keep it? :-)
Where ?
I have the MobileIPv6 implementation from KAME in my tree so can't do
a search to see if any of it has got committed already by mistake.
No, it wasn't committed.
When working on other kernels, we had a rule that if it's out of tree,
it doesn't exist. I think it's reasonable. It's hard to consider cases
you don't see.
???

I remembered that I have a checked out -8 tree that I could search.

There is some stuff in in6_src.c wrapped in MIP6 and/or NMIP that will
have been there since the IPv6 code was imported from KAME.

I have changed it slightly to work in -current.

It might be cleaner to remove it for now and add all the MIP6 code
in one go at some point in the future.

--
Posted automagically by a mail2news gateway at muc.de e.V.
Please direct questions, flames, donations, etc. to news-***@muc.de
c***@sdf.org
2019-07-11 22:31:29 UTC
Permalink
Post by Robert Swindells
Post by Robert Swindells
Post by c***@sdf.org
what is this about? do we want to keep it? :-)
Where ?
I have the MobileIPv6 implementation from KAME in my tree so can't do
a search to see if any of it has got committed already by mistake.
No, it wasn't committed.
When working on other kernels, we had a rule that if it's out of tree,
it doesn't exist. I think it's reasonable. It's hard to consider cases
you don't see.
???
I remembered that I have a checked out -8 tree that I could search.
There is some stuff in in6_src.c wrapped in MIP6 and/or NMIP that will
have been there since the IPv6 code was imported from KAME.
I have changed it slightly to work in -current.
It might be cleaner to remove it for now and add all the MIP6 code
in one go at some point in the future.
Anyway, no worries, just figured that dead code = code that should be
gone.

--
Posted automagically by a mail2news gateway at muc.de e.V.
Please direct questions, flames, donations, etc. to news-***@muc.de
Maxime Villard
2019-07-12 04:28:11 UTC
Permalink
Post by Robert Swindells
Post by Robert Swindells
Post by c***@sdf.org
what is this about? do we want to keep it? :-)
Where ?
I have the MobileIPv6 implementation from KAME in my tree so can't do
a search to see if any of it has got committed already by mistake.
No, it wasn't committed.
When working on other kernels, we had a rule that if it's out of tree,
it doesn't exist. I think it's reasonable. It's hard to consider cases
you don't see.
???
I remembered that I have a checked out -8 tree that I could search.
There is some stuff in in6_src.c wrapped in MIP6 and/or NMIP that will
have been there since the IPv6 code was imported from KAME.
I have changed it slightly to work in -current.
It might be cleaner to remove it for now and add all the MIP6 code
in one go at some point in the future.
Honestly, if it's to end up with +40K lines of poorly written, highly buggy and
likely vulnerable code developed by god knows whom twenty years ago, I don't
see any reason to add MIP6 support. SCTP is sufficient in terms of broken,
invasive stuff.

--
Posted automagically by a mail2news gateway at muc.de e.V.
Please direct questions, flames, donations, etc. to news-***@muc.de
c***@sdf.org
2019-07-11 21:39:47 UTC
Permalink
Post by Robert Swindells
Post by c***@sdf.org
what is this about? do we want to keep it? :-)
Where ?
I have the MobileIPv6 implementation from KAME in my tree so can't do
a search to see if any of it has got committed already by mistake.
No, it wasn't committed.

When working on other kernels, we had a rule that if it's out of tree,
it doesn't exist. I think it's reasonable. It's hard to consider cases
you don't see.

--
Posted automagically by a mail2news gateway at muc.de e.V.
Please direct questions, flames, donations, etc. to news-***@muc.de
Loading...