ethtool says that eth0 uses MII but twisted pair is used

networking ethtool

ethtool says that "Supported ports: [ TP MII ]" and "Port: MII". I guess that TP means twisted pair and Media Independent Interface. However physically I see that network card uses twisted pair cable, not this MII connector. So how can it be explained?


All modern Ethernet chips have the MAC built-in, and use a MII protocol over GMII, RGMII, or SerDes physical ports to communicate with the PHY device (even if it's built-in, this is usually true). If it's going to be a twisted pair interface, from the PHY the signal passes through a set of transformers and into the RJ-45 connection.

+-------+   +-------+   +-------+       +-------+    +-------+   +-------+
|       |   |       |   |       |       |       |    |       |   |       |
|Host   |<--+       |<--+       |<------+       |<---+       |<--+       |
|Ctrl   |   | FIFOs |   |  MAC  |  MII  |  PHY  |    |Xformer|   | RJ-45 |
|uC     +-->|       +-->|       +------>|       +--->|       +-->|       |
|       |   |       |   |       |       |       |    |       |   |       |
+-------+   +-------+   +-------+       +-------+    +-------+   +-------+

Related

Solve $X$ from $AXA^{-1}T = BXB^{-1}$
Integral $\int_{-\pi}^{\pi}\frac{\operatorname {d}\phi}{\sin\phi+4} $
Understanding / learning how to work with quotient spaces
Suppose $K \subseteq \mathbb{R}$ is compact and non-empty. Prove that $2K$ is compact
Maximum possible number of extrema of the function?
Neighborhood of $f^{-1}(K)$ where $K$ is compact and $f$ is proper
Question about Lebesgue Number Lemma proof
Definition of UFD and the fact that UFDs are integrally closed
Prove: $T \in L(Y,\ell^\infty(\mathbb K))$ is bounded $\implies$ $\forall n \in \mathbb N:g_n \in L(Y,\mathbb K)$ is bounded
Elementary proof that $\sum_{j=1}^{n} \prod_{k \neq j} \frac{1}{1+(a_j - a_k + i)^2} \in \mathbb{R}$ for $a_1, \dots, a_n \in \mathbb{R}$ distinct
if $f: \Omega \to \mathbb{C} \setminus \{0\}$ is holomorphic and $\Omega$ is simply connected, show there is a holomorphic function $g^2 = f$.
Theorem 15.2 of Munkres Topology