Wireshark · Wireshark-dev: Re: [Wireshark-dev] Annoying deviances from the 802.11-2016 spec in the names of cipher suites

[Graham Bloice <graham.bloice@xxxxxxxxxxxxx>]

On Thu, 29 Nov 2018 at 15:50, Richard Sharpe <realrichardsharpe@xxxxxxxxx> wrote:

Hi folks,

I notice that the names used in the code for the 802.11 dissector
deviate in annoying ways from IEEE 802.11-2016.

For example:

static const value_string ieee80211_rsn_cipher_vals[] = {
  {0, "NONE"},     // Spec says: Use group cipher suite
  {1, "WEP (40-bit)"},  // Spec says: WEP-40
  {2, "TKIP"},
  {3, "AES (OCB)"},  // Spec says: Reserved
  {4, "AES (CCM)"},  // Spec says: GCMP-128
  {5, "WEP (104-bit)"}, // Spec says: WEP-104
  {6, "BIP (128)"},   // Spec says: BIP-CMAC-128
  {7, "Group addressed traffic not allowed"},
  {8, "GCMP (128)" }, // Spec says: GCMP-128
  {9, "GCMP (256)" }, // Spec says: GCMP-256
  {10, "CCMP (256)" }, // Spec says: CCMP-256
  {11, "BIP (GMAC-128)" }, // Spec says: BIP-GMAC-128
  {12, "BIP (GMAC-256)" }, // Spec says: BIP-GMAC-256
  {13, "BIP (CMAC-256)" }, // Spec says: BIP-CMAC-256
  {0, NULL}
};

I think we should conform to the spec as far as possible.

Does anyone have any objections to me fixing these?

Is the spec actually correct?  i.e. the entries for 4 & 8 are both GCMP-128 according to your list.

--

Graham Bloice