TDLS: Update the comments related to TPK derivation

Update these comments based on IEEE Std 802.11-2016 to get rid of the
already resolved TODO comment regarding duplicated N_KEY use. The
implementation does not need any changes since it was already following
the fixed version in the current standard.

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>

src/rsn_supp/tdls.c

@@ -394,8 +394,9 @@ static void wpa_tdls_generate_tpk(struct wpa_tdls_peer *peer,
 	size_t len[2];
 	u8 data[3 * ETH_ALEN];
 
-	/* IEEE Std 802.11z-2010 8.5.9.1:
-	 * TPK-Key-Input = SHA-256(min(SNonce, ANonce) || max(SNonce, ANonce))
+	/* IEEE Std 802.11-2016 12.7.9.2:
+	 * TPK-Key-Input = Hash(min(SNonce, ANonce) || max(SNonce, ANonce))
+	 * Hash = SHA-256 for TDLS
 	 */
 	len[0] = WPA_NONCE_LEN;
 	len[1] = WPA_NONCE_LEN;
@@ -413,11 +414,8 @@ static void wpa_tdls_generate_tpk(struct wpa_tdls_peer *peer,
 			key_input, SHA256_MAC_LEN);
 
 	/*
-	 * TPK-Key-Data = KDF-N_KEY(TPK-Key-Input, "TDLS PMK",
-	 *	min(MAC_I, MAC_R) || max(MAC_I, MAC_R) || BSSID || N_KEY)
-	 * TODO: is N_KEY really included in KDF Context and if so, in which
-	 * presentation format (little endian 16-bit?) is it used? It gets
-	 * added by the KDF anyway..
+	 * TPK = KDF-Hash-Length(TPK-Key-Input, "TDLS PMK",
+	 *	min(MAC_I, MAC_R) || max(MAC_I, MAC_R) || BSSID)
 	 */
 
 	if (os_memcmp(own_addr, peer->addr, ETH_ALEN) < 0) {