I'll look through the rest of the review later.

I should have marked this as "Request Changes" in the last message :/...

Wait... Lua? what Lua tests are you planning to add (for the bootloader)?

No blocking concerns.

In D26751#599053, @arichardson wrote:

In D26751#599033, @dim wrote:

Might as well go full -O0. :)

Running the tests at -O2 in QEMU amd64 took:
44.52 real 27.28 user 13.96 sys
Changing it to -O1 makes it
91.80 real 73.32 user 14.28 sys
and -O0 results in
109.46 real 90.08 user 14.38 sys

So we might as well use -O0 since it doesn't seem to make much of a run time difference but has a big impact on compile time :)

This LGTM.

LGTM!

The change is good from a technical standpoint, but some of the documentation could be a bit more polished.

Please add __FreeBSD__ around any test modifications; it makes it easier for folks to upstream the changes as we're 3 years out of date.

Please use WITH_GOOGLETEST_INTERNAL_TESTS or something similar and default to on, since it seems to be limited to that case.

This shouldn't be committed to head.

I was a bit surprised to learn that this code relies on the jq utility from ports, by the way:

$ grep -r jq sbin/route/tests/
sbin/route/tests/utils.subr:    gateway=$(echo $result | jq -r ${query}.gateway)
sbin/route/tests/basic.sh:      atf_set require.progs jq
sbin/route/tests/basic.sh:      atf_set require.progs jq

I wish we had a tool capable of ingesting json in base :/...

NAK. I need to add Googletest support to kyua. Now that kyua has been forked, this is straightforward.

Curious... is there a reason why this was added in addition to lib/libc/tests/sys/sendfile_test.c?

This should work, for sure.

Use UF_OFFLINE instead of SF_IMMUTABLE, as recommended by @asomers.

In D25561#565211, @asomers wrote:

Instead of clearing flags, which might not be possible in a higher securelevel, how about just using UF_OFFLINE instead of SF_IMMUTABLE? That's what some of the other test cases do. The test doesn't care what the flag actually is; it just needs to be something.

LGTM!

LGTM

The FEATURE call should be moved down.

I appreciate this update.

Thank you!

The change looks good functionally. My comments are mostly about the documentation/tests.

In D24306#535777, @andrew_tao173.riddles.org.uk wrote:

In D24306#535757, @delphij wrote:

does not mean we should remove these links.

We already removed them. In 1994. What you're doing is adding them back, which is not a harmless change.

Plus, this is not only affecting PostgreSQL

Do you have a non-hypothetical example of this?

In D21259#473874, @fernape wrote:

In D21259#473832, @ngie wrote:

@fernape: could you please take this revision so I can sign off on the change?

Yes.

I would like to hear comments from @mat and others about the change though.

This is a precondition to fixing bug 244229

In D24224#533296, @0mp wrote:

After a quick chat with some other developers it looks like we would actually want to deprecate sysexits.

In D24218#532777, @koobs wrote:

• Please confirm portlint and poudriere (Py 2.x and Py 3.x) pass (ive added bullet points for these in TEST PLAN section)

Update generated patch

In D23783#523422, @ngie wrote:

One thing that isn't clear is why this sanity check is needed. Is this being specifically for platform integration issue with PowerPC64, or is there another better reason for why this needs to be done given the API? If for the latter reason and the item is well-documented, please add it. If not, this doesn't seem to serve much value and might be better implemented as a platform-specific regression test.

Please test this on amd64 as well as a sanity check, second.

One thing that isn't clear is why this sanity check is needed. Is this being specifically for platform integration issue with PowerPC64, or is there another better reason for why this needs to be done given the API? If for the latter reason and the item is well-documented, please add it. If not, this doesn't seem to serve much value and might be better implemented as a platform-specific regression test.

In D22272#486919, @gjb wrote:

Has it been confirmed that the tests-dbg.txz file in fact contains files? I removed it from the list of EXTRA_DISTRIBUTIONS in r278486 because it resulted in an archive that did not have any content.

The changes are great!

Remove unnecessary GitHub boilerplate noted by @bdrewery and @Vladimir.Chukharev_gmail.com .

Oh wait... GitHub was required because the release tarball doesn't contain the tests.

Use CHEESESHOP again.

In D21654#482533, @bdrewery wrote:

In D21654#482501, @ngie wrote:

@bdrewery, @mat, @woodsb02: ping?

The poudriere test completeness is unclear to me. It may seem like a big ask but I'm sure you understand the bound version dependency problem between python ports. I think you did run it but your paste was cut off or something. Just confirm you ran as Ben requested and I'll +1 this.