You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I think there is no need to stall #2716 much longer.
I think we have to take a look at software-controlled LEDs spanning multiple LAN ports in conjunction with DSA, as this is currently not handled nicely upstream. I assume some other devices have lost sane handling with migrating to DSA also.
tldr;
Several devices in OpenWrt use 'br-lan' as LED-trigger in DSA, while swconfig provided a trigger for the switch, which in combination with a portmask aloowed to only fire on lan ports.
The text was updated successfully, but these errors were encountered:
Why where they removed? Because registering them as mdio and firing them with netdev doesn't show the offloaded traffic between lan ports on a switch and because they cause many interrupts to be fired on mdio which reduces performance massively.
Discussion was here (it spans over several posts): openwrt/openwrt#10829 (comment)
OpenWRT maintainers are waiting for this to be resolved upstream (Linux), so they don't have to maintain a solution downstream that belongs upstream. The solution would be to allow setting registers on the switch to enable and disable it from there (or something like that). Right now the switch handles blinking on packets itself. (in hardware/offloaded)
This was a first try to send in a patch upstream, I think: https://lwn.net/ml/linux-kernel/20220503151633.18760-1-ansuelsmth@gmail.com/
I think there is no need to stall #2716 much longer.
Originally posted by @blocktrron in #2716 (comment)
tldr;
Several devices in OpenWrt use 'br-lan' as LED-trigger in DSA, while swconfig provided a trigger for the switch, which in combination with a portmask aloowed to only fire on lan ports.
The text was updated successfully, but these errors were encountered: