LGTM, but I am NOT an src committer.

In D60368#1384760, @ngie wrote:

For the record: this is why code changes for utilities like this need tests written beforehand (@des has been leading by example in this area).

Thanks for the review!

Remove gnu directories and add non-tracked ones

Fix style.

Use one single if statement.

Reuse variable in error message.

Remove comments

In D59621#1372603, @christos wrote:

Also I don't hit the error your test plan shows on 15.0-RELEASE:

$ kldstat | grep cuse
 5    1 0xffffffff84148000     6710 cuse.ko
$ virtual_oss
virtual_oss: Could not connect to cuse module

Thanks for the review!

Use variable instead of hardcoded name.

In D32297#1367860, @ziaee wrote:

Currently the OS is shown in manpage header and footer. The patch would make it shown only in the footer, and make the header the same as on other OSes. Should we do this, or close this revision?

Thanks for the review!

Use PORTSDIR and default to /usr/ports if empty.

Try and fix leak issue.

Address feedback by ngie@

In D51683#1359560, @ngie wrote:

This change seems like it would cause further confusion.
The Linux [coreutils'] version of su uses -g for overriding the primary group and -G to specify the supplementary group: https://www.man7.org/linux/man-pages/man1/su.1.html .
In an ideal world, I honestly think it would be best to rename -c to -C for symmetry with other utilities, but I'm sure that would cause loads of teeth gnashing and complaining from folks since -c has meant "class" since 1997 (renaming the option would break UX for a lot of scripts).
-C seems appropriate here, if nothing else... but I'm not sure this change is necessary. What issue are you ultimately trying to solve and what lead you to this conclusion?

Thank you very much for the review!

Remove spurious string copy.

Ping! 🔔

Friendly ping! 🔔

Fast exit on unknown device