On Tue, 22 Oct 2019 at 09:51, David Ahern [off-list ref] wrote:
Hi David,
Thank you for the review!
On 10/21/19 12:47 PM, Taehee Yoo wrote:
quoted
Current code doesn't limit the number of nested devices.
Nested devices would be handled recursively and this needs huge stack
memory. So, unlimited nested devices could make stack overflow.
This patch adds upper_level and lower_level, they are common variables
and represent maximum lower/upper depth.
When upper/lower device is attached or dettached,
{lower/upper}_level are updated. and if maximum depth is bigger than 8,
attach routine fails and returns -EMLINK.
In addition, this patch converts recursive routine of
netdev_walk_all_{lower/upper} to iterator routine.
They were made recursive because of a particular setup. Did you verify
your changes did not break it? See commits starting with
5bb61cb5fd115bed1814f6b97417e0f397da3c79
I didn't change the actual logic of walking APIs.
These walking iterator APIs work as DFS.
So it doesn't break existing codes.
quoted
Test commands:
ip link add dummy0 type dummy
ip link add link dummy0 name vlan1 type vlan id 1
ip link set vlan1 up
for i in {2..55}
do
let A=$i-1
ip link add vlan$i link vlan$A type vlan id $i
done
ip link del dummy0
8 levels of nested vlan seems like complete nonsense. Why not just limit
that stacking and not mess with the rest which can affect real use cases?
VLAN, BONDING, TEAM, MACSEC, MACVLAN, IPVLAN, VIRT_WIFI, and VXLAN
These interface types can be nested and these could be combined.
team6
|
vlan5
|
team4
|
macvlan3
|
bond2
|
vlan1
|
dummy0
There are so many similar cases even they are not real use cases.
So I think generic code is needed.
Thank you
Taehee Yoo