Given the limit could potentially get further adjustments in the
future, add it to the log so it becomes obvious what the current
limit is w/o having to check the source first. This may also be
helpful for debugging complexity related issues on kernels that
backport from upstream.