In D28906#647082, @arrowd wrote:

When preparing a bootstrap it is advised to build with all OPTIONS set to OFF to save space and build time. Would you mind rerolling it again? If you don't want to do this, let's just commit it.

It looks great now! Since this changes core libc functions, can you perform a buildworld on a system with the modified libc installed to confirm everything works fine?

• Fix incorrect path

In D28882#647110, @debdrup wrote:

The main advantage of inetd is that it starts a daemon only when a request comes in.

For rsyncd, smbd, prometheus_sysctl_exporter, et al., if the client machines have a weekly rsync job, or if prometheus only probes every 5 minutes, or only the occational connection via SMB is made - it makes more sense to only start them when they're needed.

In D28892#647126, @nyan wrote:

In D28892#646718, @ak wrote:

Also, I just noticed that in half of return statements we don't use parentheses, which is violation of style(7).

Yes, but it's a difference of style between FreeBSD and NetBSD.
So I don't change it.

Updating the patch to improve some details.

In D28892#646718, @ak wrote:

Also, I just noticed that in half of return statements we don't use parentheses, which is violation of style(7).

In vop_stddeallocate:

• Use zero_region for zero-filling
• Replace MAXPHYS with maxphys

In D28901#647089, @rlibby wrote:

I guess this case is intended for when a single-ish writer is outpacing the buf daemon?

Kind of, yes.

In D28882#647067, @lwhsu wrote:

Just being curious, what's the benefit of running rsyncd through inetd instead of using its standalone daemon?

Add missing .

In D28852#646943, @dim wrote:

Though I dislike this all getting so complex, I guess there is no good alternative way to get the required information. At some point using an external toolchain will become the default, and then it'd better work... :)

I would really like if there was a way to not always have to pass the resource dir option, though. It seems to be unconditonallly passed now? And I don't think we always want to embed rpaths into executables?

use MAX().

Yes that should also be removed. Will update

I forgot to accept this, didn't I.

Once these two minor things have been fixed (which can be done before commiting pushing), it looks good to me.

That's a very good point, unfortunately I don't have a clear answer.
I'm aware of the problem with grep and this patch definitely not resolves it.

I guess this case is intended for when a single-ish writer is outpacing the buf daemon?

When preparing a bootstrap it is advised to build with all OPTIONS set to OFF to save space and build time. Would you mind rerolling it again? If you don't want to do this, let's just commit it.

Can we boot on OrangePi Zero Plus now?

In D28831#646750, @royger wrote:

Do you have a git repo I can pick this patch from? So that it has the proper Author field. Also in FreeBSD we tend to use the 'Sponsored by:' tag to credit the company that paid for the work (if any):

https://cgit.freebsd.org/src/commit/?id=0eaa97f0e8629bcf678ff4de2678e4cba00f1c91