Re: [PATCH RFC net-next 0/3] Multi-CPU DSA support

[Florian Fainelli <f.fainelli@gmail.com>]

On 4/11/2021 11:39 AM, Andrew Lunn wrote:
> On Sun, Apr 11, 2021 at 08:01:35PM +0200, Marek Behun wrote:
>> On Sat, 10 Apr 2021 15:34:46 +0200
>> Ansuel Smith <ansuelsmth@gmail.com> wrote:
>>
>>> Hi,
>>> this is a respin of the Marek series in hope that this time we can
>>> finally make some progress with dsa supporting multi-cpu port.
>>>
>>> This implementation is similar to the Marek series but with some tweaks.
>>> This adds support for multiple-cpu port but leave the driver the
>>> decision of the type of logic to use about assigning a CPU port to the
>>> various port. The driver can also provide no preference and the CPU port
>>> is decided using a round-robin way.
>>
>> In the last couple of months I have been giving some thought to this
>> problem, and came up with one important thing: if there are multiple
>> upstream ports, it would make a lot of sense to dynamically reallocate
>> them to each user port, based on which user port is actually used, and
>> at what speed.
>>
>> For example on Turris Omnia we have 2 CPU ports and 5 user ports. All
>> ports support at most 1 Gbps. Round-robin would assign:
>>   CPU port 0 - Port 0
>>   CPU port 1 - Port 1
>>   CPU port 0 - Port 2
>>   CPU port 1 - Port 3
>>   CPU port 0 - Port 4
>>
>> Now suppose that the user plugs ethernet cables only into ports 0 and 2,
>> with 1, 3 and 4 free:
>>   CPU port 0 - Port 0 (plugged)
>>   CPU port 1 - Port 1 (free)
>>   CPU port 0 - Port 2 (plugged)
>>   CPU port 1 - Port 3 (free)
>>   CPU port 0 - Port 4 (free)
>>
>> We end up in a situation where ports 0 and 2 share 1 Gbps bandwidth to
>> CPU, and the second CPU port is not used at all.
>>
>> A mechanism for automatic reassignment of CPU ports would be ideal here.
> 
> One thing you need to watch out for here source MAC addresses. I've
> not looked at the details, so this is more a heads up, it needs to be
> thought about.
> 
> DSA slaves get there MAC address from the master interface. For a
> single CPU port, all the slaves have the same MAC address. What
> happens when you have multiple CPU ports? Does the slave interface get
> the MAC address from its CPU port?

It seems to be addressed by this part of patch 2:

+	if (ether_addr_equal(dev->dev_addr, master->dev_addr))
+		eth_hw_addr_inherit(dev, cpu_dev);

although this could create an interesting set of issues if done fully
dynamically while the data path is active.

> What happens when a slave moves
> from one CPU interface to another CPU interface? Does its MAC address
> change. ARP is going to be unhappy for a while? Also, how is the
> switch deciding on which CPU port to use? Some switches are probably
> learning the MAC address being used by the interface and doing
> forwarding based on that. So you might need unique slave MAC
> addresses, and when a slave moves, it takes it MAC address with it,
> and you hope the switch learns about the move. But considered trapped
> frames as opposed to forwarded frames. So BPDU, IGMP, etc. Generally,
> you only have the choice to send such trapped frames to one CPU
> port. So potentially, such frames are going to ingress on the wrong
> port. Does this matter? What about multicast? How do you control what
> port that ingresses on? What about RX filters on the master
> interfaces? Could it be we added it to the wrong master?
> 
> For this series to make progress, we need to know what has been
> tested, and if all the more complex functionality works, not just
> basic pings.

Agreed.
-- 
Florian