• Replace VOP_UNMOUNT with dounmount
• Plunk vfs_ref before to avoid deadlock

Now also required for ports-mgmt/packagekits tests.

There is nothing to discuss:

In D53433#1224456, @michaelo wrote:

I don't know whether it can be compiled on 13.5 and run older minor versions.

This looks OK to me, but I'm a python hat wearer.

In D53433#1223612, @michaelo wrote:

I agree with you, but that you I think require a PEP describing this (e.g., https://peps.python.org/pep-0513/).

In D53433#1223585, @michaelo wrote:

There is a subtile difference where. The wheels are already there with PEP 517, distutils just requires one more step, but I expect distutils to disappear in the future. So basically either "make wheel" would be a noop or it would move/copy the *.whl to ${WRKDIR}/whl/

In D53433#1223130, @michaelo wrote:

As a plugin, yes?

It should be turned into a proper pull request to the Poudriere upstream repo. One thing that already catched my eye is that wheel generation happens from the host, not within the jail. Bryan might disagree with this.

Link your Poudriere changes in there, so it is possible to get an idea how this is going to work overall.

Then this change alone isn't really enough and I'm not even sure it is a right step forward.

I don't quite get the rationale. Can you provide an example how is it going to be used?

Sure.

In D53474#1221017, @alven wrote:

Seems like I've got it: no actual SONAME change, so no need to update the dependencies.

What does readelf -d /usr/local/lib/libz3.so | grep SONAME say for the updated port?

The release changed the patch version and according to pkg-plist the shared library version wasn't bumped. This usually means that there is no need to recompile consumers, thus no bumps needed.

Why bump consumers? There seems to be no need in that.

In D53072#1219958, @kiwi wrote:

Even if autoconf can discovering this, it is far better to force the detection of sssd.conf than using symbolic links to fix some hardcoded defaults.

Ping.

I have fixed the hang, but the filesystem still does not get unmounted.

• Fix hang by marking fdata dead

TEST_DEPENDS are not encoded into package's manifest, AFAIK? So the resulting package does not change.

This was committed in https://cgit.freebsd.org/ports/commit/?id=f0fec200720c337dc354d2ee17b5927bac74eb53

This LGTM.

I had to learn a bit about gmock to properly review this. LGTM, FWIW.

Updated the commit message, waiting for the approval.