push docs: inaccurate description of omitted destination?

It seems to me that push's manpage contains a couple of inaccurate (or at least confusingly incomplete) statements about omitting the destination part of the refspec.

First, the refspec section of the manpage has this to say:

    If `git push [<repository>]` without any `<refspec>` argument is set to
    update some ref at the destination with `<src>` with
    `remote.<repository>.push` configuration variable, `:<dst>` part can
    be omitted--such a push will update a ref that `<src>` normally updates
    without any `<refspec>` on the command line.  Otherwise, missing
    `:<dst>` means to update the same ref as the `<src>`.

Reading that, I'd think that, if I haven't configured remote.<repository>.push in a way that involves <src>, omitting <dst> will update the remote ref with the same name. But push.default is also consulted before falling back entirely to using the remote ref with the same name:

    $ git rev-parse --symbolic-full-name HEAD
    refs/heads/topic
    $ git rev-parse --symbolic-full-name @{upstream}
    refs/remotes/origin/master
    $ git -c push.default=upstream push -n origin topic ;# unmentioned case
    To ../rem
       09cc638..58f7f72  topic -> master

    $ git branch --unset-upstream topic
    $ git -c push.default=upstream push -n origin topic ;# fallback case
    To ../rem
     * [new branch]      topic -> topic

Second, the example section says

    `git push origin master`::
       Find a ref that matches `master` in the source repository
       (most likely, it would find `refs/heads/master`), and update
       the same ref (e.g. `refs/heads/master`) in `origin` repository
       with it.  If `master` did not exist remotely, it would be
       created.

Perhaps I'm misreading that, but I'd interpret that as saying the remote ref with the same name will always be updated, but that doesn't match the refspec description of an omitted destination shown above or the push.default-dependent behavior demonstrated in the example above.

Kyle Meyer <kyle@kyleam.com> writes:

... if such a push would update the ref at the remote with the same name, yes. Otherwise no. And that is determined by what push mode is being used (e.g. are we using the matching mode? The current mode? Something else?).

Yes. The push.default configuration participates in the selection of the ref at the remote that gets updated, given a refspec element with a missing :<dst>. And you do not have to write :<dst> to name the ref that a refspec with missing :<dst> would update anyway.

That part is random listing of examples with vanilla configuration, so depending on what setting you personally have, the behaviour may be different.