Address @kevans 's comment

Plenty of other file systems have the same check. If you're convinced that it's unnecessary I would suggest you post a patch that removes it from all of them, not just tarfs, and circulate it more widely.

Use netlink/route/interface.h

Works here on a custom host at -CURRENT.

Maps every IFLAF_VF_* presence bit to the stable public IFCONFIG_VF_F_* flags

Removed now unneeded ifr_nv_* churn, use IFLAF_VF_* numbers directly (and in subordinate reviews), add some clarity around extensions.

I was thinking of rcorder(8) for kernel. Of course run at compilation time, not an startup.

I'm not taking this task :) I just want to make kernel -Werror-assing-enum clean.

I mean that the backing file's permissions have already been checked since it has been successfully opened at that point, and the calling process's privilege to mount is checked elsewhere. Isn't that already checked in sys/kern/vfs_mount.c:vfs_domount()?

In D58776#1350103, @melifaro wrote:

LGTM (modulo removing nvlist helpers); left some comments in the diff.

Could you elaborate on the " This allows drivers to add information without expanding a central enumeration." a bit?
If the only concern is central enumeration, then it can be addressed differently - by having per-driver attributes.
Linux does it the following way: inside IFLA_LINKINFO attribute it stores IFLA_INFO_KIND attribute with the driver type and IFLA_INFO_DATA where any customer driver-specific structures can be stored.
I should have implemented something like that but never managed to (there are some bits and pieces in if_vlan.c that can serve as a simple example for the GET requests)

Implement per-arch hooks akin to mdthread; no arch ifdefs in common code

I noticed when I was doing the most recent update that this section is very stale. Sebastian -- upstream maintainer for 10+ years -- suggested this wording when I asked him.

The num namespaces changes should be separate

LGTM (modulo removing nvlist helpers); left some comments in the diff.

Constify key_checksockaddrs() args.
Remove clamping.
Assert that key_checksockaddrs() pass for helpers.
Change the socket address family in ipcomp6_nonexp_encapcheck().

In D58586#1349779, @kib wrote:

Ping?

I'm not sure - defaults tend to change and hardcoding those in netstat / rtsock seem somewhat fragile. Let's maybe take a step back - is there any measurable performance difference after the change?