looks good

We need to figure out how to get all of this upstream as well.

I would suggest describing this plan on freebsd-net@ and freebsd-arch@ mailing lists

And mbuf.9 change along with mbuf.h.

Ugly, but amazing. I would suggest committing the macro change first followed by the refactoring.

so as part of the rename abi_start will disappear I presume

Fine with me.

In D21473#473720, @grembo wrote:

That was actually what I called it at first (showconfig or dumpconfig), but then I realized that commands provided by freebsd-update prefer brevity over clarity (e.g. cron actually fetches updates from cron, upgrade fetches upgrades, but doesn't install them, IDS has no verb at all).

one little note, for future uploads can you include context (as described on https://wiki.freebsd.org/Phabricator) e.g.

git diff -U999999 other-branch > change.diff
git show -U999999 <commit-hash> > change.diff
svn diff --diff-cmd=diff -x -U999999 > change.diff

some quite long lines in here; we don't have a good established lua style yet

In D21775#475207, @kevans wrote:

but decided that was going too far when it's already out of PATH and disguised as bsdlua.

Let's make lua a build-tool :)

Note that my objection was specifically to the change that unconditionally disabled Capsicum in the Makefile, regardless of whether 9p support is included or not.

Was committed in rS331294

Thank you for the submission - I committed a portion of the new #defines in rS352551.

I'm happy with the config part so far, but probably use showconfig.
I haven't yet reviewed the updatesready.

Fine with me, although I'd prefer a different name for config_Components_src. This is not directly used for the config file as other config_-prefixed functions are. Maybe something like filter_components?

I agree with mark - this will be a nice cleanup and I think should be done, but will take some careful coordination with downstreams, and ports.

I'll let @mjg weigh in on the asm

For reference, https://pubs.opengroup.org/onlinepubs/9699919799/functions/open.html

I opened an upstream issue at https://github.com/jemalloc/jemalloc/issues/1629

For upstreaming a pull request against https://github.com/jemalloc/jemalloc presumably

I'm not an expert here but seems reasonable.

phabricator didn't count that for closing this review

Looks like the Differential Revision: string and the URL were on different lines.

I don't want to add the whole secteam yet...

Would be good to explain "why" in the commit message

This adds these cases:

use c99 initializers suggested by mjg

make linux_renameat pass through to linux_renameat2

Wait, this doesn't work.

upload proper diff

Looks fine to me but I defer to ports experts

Overall LGTM

In D20974#470171, @jchandra wrote:

SBSA watchdog should work on ThunderX2, the specific issue was due to the linux driver not handling the full spec.

In D20974#463006, @greg_unrelenting.technology wrote:

The original is https://github.com/NetBSD/src/blob/trunk/sys/arch/arm/acpi/sbsawdt_acpi.c — a true diff would show most things (except the defines) as different, it's better to just visually compare :)

Fine with me

Content seems fine, I think it's fine to commit after igor/mandoc -Tlint updates.

I think they should have the same default, so leave your change to src.opts.mk in place.