Re: [RFC PATCH net-next] bonding: add a vlan+srcmac tx hashing option

[Jay Vosburgh]

Jarod Wilson <jarod@xxxxxxxxxx> wrote:

>On Thu, Jan 07, 2021 at 07:03:40PM -0500, Jarod Wilson wrote:
>> On Fri, Dec 18, 2020 at 04:18:59PM -0800, Jay Vosburgh wrote:
>> > Jarod Wilson <jarod@xxxxxxxxxx> wrote:
>> > 
>> > >This comes from an end-user request, where they're running multiple VMs on
>> > >hosts with bonded interfaces connected to some interest switch topologies,
>> > >where 802.3ad isn't an option. They're currently running a proprietary
>> > >solution that effectively achieves load-balancing of VMs and bandwidth
>> > >utilization improvements with a similar form of transmission algorithm.
>> > >
>> > >Basically, each VM has it's own vlan, so it always sends its traffic out
>> > >the same interface, unless that interface fails. Traffic gets split
>> > >between the interfaces, maintaining a consistent path, with failover still
>> > >available if an interface goes down.
>> > >
>> > >This has been rudimetarily tested to provide similar results, suitable for
>> > >them to use to move off their current proprietary solution.
>> > >
>> > >Still on the TODO list, if these even looks sane to begin with, is
>> > >fleshing out Documentation/networking/bonding.rst.
>> > 
>> > 	I'm sure you're aware, but any final submission will also need
>> > to include netlink and iproute2 support.
>> 
>> I believe everything for netlink support is already included, but I'll
>> double-check that before submitting something for inclusion consideration.
>
>I'm not certain if what you actually meant was that I'd have to patch
>iproute2 as well, which I've definitely stumbled onto today, but it's a
>2-line patch, and everything seems to be working fine with it:

	Yes, that's what I meant.

>$ sudo ip link set bond0 type bond xmit_hash_policy 5

	Does the above work with the text label (presumably "vlansrc")
as well as the number, and does "ip link add test type bond help" print
the correct text for XMIT_HASH_POLICY?

	-J

>$ ip -d link show bond0
>11: bond0: <BROADCAST,MULTICAST,MASTER> mtu 1500 qdisc noop state DOWN mode DEFAULT group default qlen 1000
>    link/ether ce:85:5e:24:ce:90 brd ff:ff:ff:ff:ff:ff promiscuity 0 minmtu 68 maxmtu 65535
>    bond mode balance-xor miimon 0 updelay 0 downdelay 0 peer_notify_delay 0 use_carrier 1 arp_interval 0 arp_validate none arp_all_targets any primary_reselect always fail_over_mac none xmit_hash_policy vlansrc resend_igmp 1 num_grat_arp 1 all_slaves_active 0 min_links 0 lp_interval 1 packets_per_slave 1 lacp_rate slow ad_select stable tlb_dynamic_lb 1 addrgenmode eui64 numtxqueues 16 numrxqueues 16 gso_max_size 65536 gso_max_segs 65535
>$ grep Hash /proc/net/bonding/bond0
>Transmit Hash Policy: vlansrc (5)
>
>Nothing bad seems to happen on an older kernel if one tries to set the new
>hash, you just get told that it's an invalid argument.
>
>I *think* this is all ready for submission then, so I'll get both the kernel
>and iproute2 patches out soon.
>
>-- 
>Jarod Wilson
>jarod@xxxxxxxxxx

---
	-Jay Vosburgh, jay.vosburgh@xxxxxxxxxxxxx