[<prev] [next>] [day] [month] [year] [list]
Message-ID: <AANLkTi=cNhtBaOFYMEoN8K0UZXyO+AEbPYR+SwGzskpe@mail.gmail.com>
Date: Wed, 9 Feb 2011 11:30:19 +0100
From: Mickael Chazaux <mickael.chazaux@...il.com>
To: netdev@...r.kernel.org, buytenh@...tstofly.org
Subject: Hardware bridging for Distributed Switch Architecture (DSA)
Hi,
I'm interested in working on hardware bridging support for the
Distributed Switch Architecture (DSA).
I have a few questions about this patch [1] :
"(While the patch in itself works, the implementation is ugly, and I
don't see it being merged upstream in this form any time soon"
Why this? What needs to be done to clean it up, and have it in shape?
"While conceptually I think this is the way this should work (i.e. no
workflow changes for the system administrator) "
I have worked a bit on this subject too and concluded to the same idea.
" the code shouldn't be hooking directly into the bridge code at all.
Ideally, it should just be using a netlink interface providing this info:
- bridge port group creation/destruction
- adding/removing an interface to/from a bridge port group
- changing the STP state of an interface in a bridge port group"
Doesn't brctl already uses netlink to configure the bridge? Could you expand
a bit more on this?
I like this approach of automatically hardware-bridge ports if possible when
adding them to a bridge. Do you mean putting a specific API in netlink
for doing
the hardware bridging stuff ?
Regards,
Mickael
[1] http://patchwork.ozlabs.org/patch/16578/
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Powered by blists - more mailing lists