openssh: Diff between upstream 9.9p1 and crypto/openssh

[emaste]

Summary

This is a diff between upstream and crypto/openssh for the in-progress 9.9p1 update

[emaste]

auth.c

525

It looks like upstream un-style(9)'d this from our submission. Will drop to reduce diffs.

configure.ac

1059

This bug was reported upstream but has not been addressed

log.c

449–450

Rebase leftovers, will remove

ssh_config.5

551–552

Leftover after upstream adopted our default, will remove

[jlduran]

If it helps, the diffs for blocklistd are OK.
I keep a local diff for just blocklistd: (https://github.com/jlduran/openssh-portable), just in case we decide to upstream our diffs.
EDIT: Just to clarify, upstream our blocklistd diffs to zoulasc/blocklist, just like NetBSD and Debian.

[kevans]

configure.ac

3333

too bad we don't have some kind of pkgconfig for base that we could use here instead of patching it in

pathnames.h

129

We already have a -D for this in lib/libssh/Makefile (though I find it a little silly... I would've done ${LOCALBASE:U/usr/local}/bin/ssh-askpass instead of just leaving it undefined), maybe we should drop this and define _PATH_XAUTH as well to drop that one?

Both of those should probably be done in secure/ssh.mk instead of in libssh directly.

regress/unittests/sshkey/test_sshkey.c

85

Does anyone actually run these tests? I don't see where this file would even be built today.

servconf.c

280

The related bits for this seem like they would be good to upstream

[emaste]

pathnames.h

129

Note that the changes in this review are what's already in the tree and are the result of all of our changes to date; I agree with cleanups like this, but they can happen independent of the 9.9p1 update

regress/unittests/sshkey/test_sshkey.c

85

Yeah I'm almost certain this has never actually been run. This came in from 1f474190fc280 in a tree-wide sweep of /usr/local.

servconf.c

280

Indeed.