On Sat, Feb 03, 2018 at 09:40:56PM +0100, Toke Høiland-Jørgensen wrote:
> This adds support for source-specific IPv6 routes to Bird core. This is
> based on Dean Luga's original patch, with the review comments addressed.
> Sadr support is added to network address parsing in confbase.Y and to
> the kernel protocol on Linux.

Hi

Thanks, i will look at it in details later.

> A couple of issues remain with this patch:
> 
> - Had to change the default addr_type in rt_setup() to NET_SADR_IP6 to
>   get the kernel 'learn' mechanism to work.

This seems like a bug in 'learn' in 2.0.x, i will fix that.

> - Not sure if the logic in net_route_sadr_ip6() is right.

First, it should have a comment about what is the expected behavior,
whether it should be longest-dst-then-src or longest-src-then-dst, or
some other behavior. I would expect it to be consistent with Linux
kernel, but i am not sure what is used there.

The code mixes net_addr_ip6 and net_addr_sadr_ip6 during lookups, which
likely trigger asserts in fib_find?

> - The config parser barfs on sadr route literals in the static protocol
>   with an "Integer expression expected" error message.

Perhaps it created more Bison grammar conflicts? I will check that later.

> - There's no way to mix source-specific and non-source-specific routes
>   (i.e., sadr tables cannot be connected to non-sadr tables).

This is OK for now.

-- 
Elen sila lumenn' omentielvo

Ondrej 'Santiago' Zajicek (email: santi...@crfreenet.org)
OpenPGP encrypted e-mails preferred (KeyID 0x11DEADC3, wwwkeys.pgp.net)
"To err is human -- to blame it on a computer is even more so."

Attachment: signature.asc
Description: PGP signature

Reply via email to