Re: readutmp: load libsystemd with dlopen instead of linking

Simon Josefsson via Gnulib discussion list <[email protected]> Thu, 16 Jul 2026 09:17:45 +0200

Newsgroups	gmane.comp.lib.gnulib.bugs
Message-ID	<[email protected]>

Bruno Haible via Gnulib discussion list <[email protected]> writes:

> I really want to hear convincing arguments, before installing additional
> complexity into readutmp.c. Your syslogd argument is not convincing for me,
> so far.

Agreed!  The argument is a strawman and needs to be qualified before
becoming a justifiable argument to add complexity.

I do see how people coming from the
minimize-the-dependencies-in-my-container background reach for dlopen()
as a way to solve that problem.

It solve the problem in several cases, but I wonder if that is a
sustainable long-term approach.  Taking it to the extreme, essentially
all shared library calls should be converted into dlopen, with graceful
fallback when the library isn't available, and that seems just weird and
somehow exposing a deeper problem.

Dependency bloat is a real complex problem too, though.  We thought we
solved this with shared libraries, but this problem seems to come back
in a different form today, leading to the dlopen() approach.  Maybe we
will go back to static linking (most of the Go ecosystem).  We know how
to re-compile large sets of binaries easily today, and doing so has
other QA/compiler advantages, so maybe the argument for shared libraries
is weaker today than it used to be.

/Simon

signature.asc (application/pgp-signature, 1.2 KB)