> Nitpick. It's accurate for IPv6 which is the only address family that > currently emits the new attribute. It's also accurate for IPv4. Other > address families (MPLS / PHONET / IPMR) can start making use of > strict_start_type in their policies, but it doesn't belong in this > series. > > Reviewed-by: Ido Schimmel <[email protected]>
Thanks. I also went through the two Sashiko runs on this series: https://netdev-ai.bots.linux.dev/sashiko/#/patchset/20260808005642.26901-1-sigefriedhyy%40gmail.com https://sashiko.dev/#/patchset/20260808005642.26901-1-sigefriedhyy%40gmail.com Most of the comments describe the intermediate state of a split series and are answered by later patches. The remaining one, the bare OSError caught around the RA sent in the selftest, does not seem worth a respin on its own; I will fix it if the series needs one more revision. Thanks, Yuyang
