Why enforcing nawk and not using the regular awk?

I do not think it is necessary to add it to /lib instead of the regular /usr/lib given there is "nothing vital about this lib"

why not adding in a single line?

BTW the LDFLAGS is not necessary here given this is already default

Useless inclusion

Same as above

Do we have daemons in /bin or /sbin that might (later on) depend on blacklistd, and could start before /usr is mounted?

I am just using what was there before - I don't know why it is currently specified as 'nawk'.

OK, I'll move it to /usr/lib. I was thinking that sshd did not have a requirement for /usr/lib, but checking shows that it does require /usr/lib.

I looked at a few different Makefiles that use LIBADD, and the ones already in the tree do it both ways - listing on a single line, and listing on multiple lines. I'd be happy to make it a single line.

As for the LDFLAGS settings, it doesn't compile on a system that doesn't already have it installed without setting LDFLAGS. Once the library is already installed, the LDFLAGS isn't needed. Until the library is installed, it needs LDFLAGS to be set, to find the library file in the OBJDIR.

Since the introduction of this will be a new feature, not setting LDFLAGS would break everyone's build.

I will fix this.

I'll smash the LIBADD lines into one.

Same response for why LDFLAGS needs to be set.

It came from rS138061. This one is the only case of nawk in etc/ (vs. 31 awk). It probably makes sense to switch it in a separate commit.

Note that since sshd itself is in /usr/sbin /usr must be mounted for it to start anyhow :-)

I'd just set the variable unconditionally.

FWIW ${.CURDIR:H:H} results in much neater paths than
${.CURDIR}/../..

It's not my exact preference either, but there's the existence proof of both "_rtld-elf" and "_tftp-proxy" in the same list of variables. So it's already being done. (Perhaps it should be undone, but that would be a different set of patches.)

I was pretty careful to make sure that I wasn't adding something that did not match the existing code.

I'm torn about doing this. The code above (setting _LIBRARIES) ends the same way, with a trailing backslash, and the obligatory blank line. As does the code below my addition (appending to _LIBRARIES for MK_OFED). Another trailing backslash and followed by a blank line.

So, violate the style of the surrounding code, or leave the blank line and match the style? I left the blank line to match the existing style.

fwiw there is no need to comment blacklistd

Fixed in my local repository, will be fixed when I post the next review update / commit the code.