Hello, So, this question is addressed mostly to Felix. But if anyone else has some input, feel free to add.
http://git.openwrt.org/?p=project/netifd.git;a=blob;f=bridge.c;h=8e6c9a61e6300837ff2f39f8f379f176c4f14aa7;hb=HEAD#l275 That bit of code seem to have been added ~2012. Via commit: http://git.openwrt.org/?p=project/netifd.git;a=commitdiff;h=a9fd870450e82f31cba681651656276dfc1c39ac The issue I'm having is that the toggle there, generates link state events on the device/interface, when it's being moved from bridge_A to bridge_B, because device_set_present(false + true) would eventually do a system_if_down() + system_if_up(). Disclaimer: This may have been addressed in another way in a more recent netifd version ; I am using an older version (with plans to upgrade asap, but that's still away for some time). So, then questions: 1) any idea if it is safe to remove that toggle ? 2) any input/idea on what was the initial issue ? I know it's a bit long from 2012, but maybe understanding the initial problem would help. The link state events caused by this toggle, cause other [of our] modules to react even though they should not. And we've worked it around netifd [for now], but I'm trying to see if there's a proper way for netifd to handle this, or if netifd should take this into consideration. Thanks Alex _______________________________________________ Lede-dev mailing list Lede-dev@lists.infradead.org http://lists.infradead.org/mailman/listinfo/lede-dev
