df: Import tests from NetBSD

[jlduran]

Summary

Adapt NetBSD's tests to run on FreeBSD, rather than directly using the
tests under contrib, as the differences in spacing are significant.

It uses a mocked getmntinfo(3) implementation to ensure consistent file
system enumeration during testing, generating an h_df binary that is
used for testing.

Test Plan

$ kyua test
t_df:hflag  ->  passed  [0.021s]
t_df:normal  ->  passed  [0.010s]

[jlduran]

I see some potential changes to df in GH #1906, let's start adding tests.

[jlduran]

NetBSD has fixed the output of their tests to match POSIX behavior (The percentage value shall be expressed as a positive integer, with any fractional result causing it to be rounded to the next highest integer) (lib/59811).

It is now our turn to fix it. Commit the output as is, for better comparison.

[jlduran]

• Keep the same NetBSD file name, we'll branch off our tests on another file.

[ngie]

bin/df/tests/getmntinfo.c

58

• style(9) bug.
• clarity around order of operations.

90–91

121–122

[jlduran]

Thank you!
I wanted as few diffs with NetBSD as possible, but using nitems() is a good suggestion; I'll probably submit it to GNATS.

[ngie]

In D54016#1243704, @jlduran wrote:

Thank you!
I wanted as few diffs with NetBSD as possible, but using nitems() is a good suggestion; I'll probably submit it to GNATS.

Ah, ok! If it's code that comes from NetBSD, we should use __arraycount(..) instead of nitems(..). __arraycount(..) is defined in libnetbsd.

[des]

What's the point of copying the NetBSD tests if we end up having to change everything about them, and then end up adding our own test script anyway as you do in D54265?

[jlduran]

In D54016#1243870, @des wrote:

What's the point of copying the NetBSD tests if we end up having to change everything about them, and then end up adding our own test script anyway as you do in D54265?

The spacing between the columns is different. Also, getmntinfo.c is somewhat different on FreeBSD.
I tried to #ifdef this test, but it soon became obvious we should branch out (it became unreadable).

[des]

In D54016#1243879, @jlduran wrote:

In D54016#1243870, @des wrote:

What's the point of copying the NetBSD tests if we end up having to change everything about them, and then end up adding our own test script anyway as you do in D54265?

I tried to #ifdef this test, but it soon became obvious we should branch out (it became unreadable).

Yes. Just write your own.

[jlduran]

In D54016#1243888, @des wrote:

In D54016#1243879, @jlduran wrote:

In D54016#1243870, @des wrote:

What's the point of copying the NetBSD tests if we end up having to change everything about them, and then end up adding our own test script anyway as you do in D54265?

I tried to #ifdef this test, but it soon became obvious we should branch out (it became unreadable).

Yes. Just write your own.

My name will also be added to the license header upon commit. It is in my best interest to keep these tests aligned as much as possible with NetBSD.