The site labels the problem PROVED; its remarks record Tao's bounds $\tau(n/m)/m\le G(n)\le\tau(n)$, displayed for any divisor $m$ of $n$ and true for $m>1$, answering the first question yes, and Erdős's 1982 remark that the divergence is trivial. The standing in the frontmatter derives from the claim pages under the parts divergence and asymptotic: the accepted full claim on Erdős and Tenenbaum's 1983 paper settles both questions, its Théorème 2 giving $\sum_{n\le x}G(n)=x\log x\,(1+o(1))$, and the accepted partial claim on Erdős's remark settles the first; the problem is solved with the value proved.