The remarks of the Theorem 4 in the paper "On the complexity of circuit satisfiability" claims that: if circuit satisfiability (CktSat) problem can be decided by deterministic circuits of $2^{o(n)}$ size, then ETH fails. However, there are no references and proofs for this statement in the paper, and it is hard for me to get proof of this statement.
In fact, I have tried to prove the following weaker statement but failed. "If $SAT \in P/poly$, then ETH fails."
QUESTION: How to prove that: if Circuit Satisfiability (CktSat) problem can be decided by deterministic circuits of $2^{o(n)}$ size, then ETH fails?
