3
0
mirror of https://git.kernel.org/pub/scm/network/wireless/iwd.git synced 2024-11-23 07:29:28 +01:00
Commit Graph

3772 Commits

Author SHA1 Message Date
Denis Kenzior
eeac3e8f40 adhoc: Convert to IWD_MODULE 2019-05-28 14:24:37 -05:00
Denis Kenzior
a9e3b3e906 scan: Convert to IWD_MODULE 2019-05-28 14:12:25 -05:00
Denis Kenzior
d74fcfd25c storage: Use L_TFR instead of TFR 2019-05-28 14:06:22 -05:00
Denis Kenzior
57748347b0 treewide: Use L_TFR macro 2019-05-24 11:52:40 -05:00
Denis Kenzior
865492df8c ft: Fix endianness issues
The use of mmpdu_association_response members marked as __le requires
the relevant endianness conversion.
2019-05-23 16:59:34 -05:00
Denis Kenzior
360ec1488d fils: Fix endianness issues
Using mmpdu_associate_response and mmpdu_authenticate structure
members marked __le requires the relevant endianness conversion.
2019-05-23 16:59:14 -05:00
Denis Kenzior
e53bd17992 owe: Fix endianness issues
Using mmpdu_associate_response structure members marked __le requires
the relevant endianness conversion.
2019-05-23 16:58:59 -05:00
Denis Kenzior
85bd019c85 sae: Fix endianness issues
The use of mmpdu_authentication and mmpdu_association_response members
marked as __le requires the relevant endianness conversion.
2019-05-23 16:56:45 -05:00
James Prestwood
77d66ebc54 auto-t: FILS-FT autotest 2019-05-23 15:28:24 -05:00
James Prestwood
628c079e55 station: netdev: allow FILS-FT AKMs
This adds some checks for the FT_OVER_FILS AKMs in station and netdev
allowing the FILS-FT AKMs to be selected during a connection.

Inside netdev_connect_event we actually have to skip parsing the IEs
because FILS itself takes care of this (needs to handle them specially)
2019-05-23 15:27:29 -05:00
James Prestwood
e583b1d243 fils: update to handle FILS-FT
FILS unfortunately is a special case when it comes to fast transition.
We have to process the FT IEs internally since we cannot trigger the
same initial mobility association code path (via netdev).
2019-05-23 15:26:34 -05:00
James Prestwood
5ca4fa9b9e wiphy: allow both FILS-FT AKMs to be chosen 2019-05-23 15:23:46 -05:00
James Prestwood
7f7a5c861d ft: update for larger MIC/FILS-FT 2019-05-23 15:23:46 -05:00
James Prestwood
1accf534dc ie: allow 24 byte FTE MIC
FT over FILS-SHA384 uses a 24 byte FT MIC rather than the 16 byte MIC
used for all other AKMs. This change allows both the FT builder/parser
to handle both lengths of MIC. The mic length is now passed directly
into ie_parse_fast_bss_transition and ie_build_fast_bss_transition
2019-05-23 15:23:08 -05:00
James Prestwood
4097a49669 eapol: add FILS-FT AKMs to eapol_start
This will prevent FILS-FT from starting the 4-way handshake as it
does for regular FILS
2019-05-22 16:29:23 -05:00
James Prestwood
5d2b995175 handshake: update FT derivation functions for FILS-FT
FILS-FT could derive a longer PMKR0/R1 key, as well as uses a special
xxkey that it derives during FILS.
2019-05-22 16:15:23 -05:00
James Prestwood
0e9ed03e60 handshake: update key getters for FILS-FT
FILS-FT is a special case with respect to the PTK keys. The KCK getter
was updated to handle both FT-FILS AKMs, by returning the offset in
the PTK to the special KCK generated during FILS. A getter for the KCK
length was added, which handles the SHA384 variant. The PTK size was
also updated since FILS-FT can generate an additional 56 bytes of PTK
2019-05-22 16:14:51 -05:00
James Prestwood
bc381bd8c3 crypto: update FT derivation functions to use sha384
FILS-FT requires the SHA384 KDF to derive the FT keys
2019-05-22 16:14:14 -05:00
James Prestwood
aafb3fa7ed handshake: add key for FILS-FT in handshake_state
FILS derives its own FT key, for use as xxkey during fast transition.
2019-05-22 16:14:01 -05:00
James Prestwood
e109e1b1cf crypto: pass xxkey length for crypto_derive_pmk_r0
FILS-FT has a special derivation for xxkey, and depending on AKM
the key may be 32 or 48 bytes long.
2019-05-22 16:13:57 -05:00
James Prestwood
4c32dd09f6 fils: add forgotten break 2019-05-22 16:13:08 -05:00
Denis Kenzior
03ff5ef7d0 device: Convert to using IWD_MODULE 2019-05-22 10:20:10 -05:00
Denis Kenzior
978e538f81 station: Convert to use IWD_MODULE 2019-05-22 10:20:10 -05:00
Denis Kenzior
23b278ef52 wsc: Convert to use IWD_MODULE 2019-05-22 10:20:10 -05:00
Denis Kenzior
4cfcb9c88d network: Convert to use IWD_MODULE 2019-05-22 10:20:10 -05:00
Denis Kenzior
800d57d095 knownnetworks: Convert to IWD_MODULE 2019-05-22 10:20:10 -05:00
Denis Kenzior
546c3c26d3 simauth: Convert to use IWD_MODULE 2019-05-22 10:20:08 -05:00
Denis Kenzior
b8f6899f40 blacklist: Convert to use IWD_MODULE 2019-05-22 09:58:04 -05:00
Denis Kenzior
a1d9c07f2f erp: Convert to using IWD_MODULE 2019-05-19 13:07:13 -05:00
Denis Kenzior
2386fa7938 main: Add IWD_MODULE macro 2019-05-19 13:05:06 -05:00
Denis Kenzior
8dced8b7f7 main: Update to the new genl api 2019-05-17 17:10:18 -05:00
Denis Kenzior
71ebc00fb7 tools: Update test-runner to the new genl api 2019-05-17 17:10:15 -05:00
Denis Kenzior
dea23bede7 tools: Update hwsim to the new genl api 2019-05-17 17:10:13 -05:00
Denis Kenzior
4e4285497d TODO: Add some OCE related tasks 2019-05-13 21:43:24 -05:00
James Prestwood
e5d36bbb33 auto-t: add AP test with no AP iftype support
Ensures IWD properly handles an adapter which does not support AP mode.
2019-05-13 16:38:23 -05:00
James Prestwood
f30c4bf578 auto-t: handle exceptions correctly for start_ap
The start_ap method was raising potential dbus errors before converting
them to an IWD error type. This is due to dbus.Set() not taking an error
handler. The only way to address this is to catch the error, convert it
and raise the converted error.
2019-05-13 16:38:23 -05:00
James Prestwood
d6eade2252 auto-t: add test for CCMP with no support
Using the new cipher_disable hwsim option we can disable CCMP support
but require it on the AP. This should result in a Not Supported error.
2019-05-13 16:38:23 -05:00
James Prestwood
f038c11205 test-runner: allow disabling of iftype/cipher
Two new hardware configuration keys were added:

[radX]
iftype_disable=station,ap,adhoc,p2p_client,p2p_go,mesh_point
cipher_disable=wep40,wep104,tkip,ccmp,bip

Any of the above values are supported and can be disabled.
2019-05-13 16:38:23 -05:00
James Prestwood
2d173e5f24 test-runner: add a radio parameters structure for creation
Support is coming to configure radios with a specific set of interface
type and cipher support, so the input to create_hwsim_radio is better
suited to use a parameter structure rather than adding more parameters.
2019-05-13 16:38:23 -05:00
James Prestwood
2d51622c0b test-runner: allow radio list subset to be configured
The radio_confs key was parsed in a way that required all radios
to be specified in the list. This isnt optimal, as you may want to
specially configure a certain radio, while keeping all the others
default.

This change reworks some logic and allows any radio to be specially
configured on its own.
2019-05-13 16:38:23 -05:00
James Prestwood
0e31b33631 hwsim: allow specific radio iftypes/ciphers
mac80211_hwsim now allows setting supported iftypes/ciphers. This patch
enables this support in hwsim. Specific iftypes/ciphers can now be
disabled via the command line when creating a radio:

Disable iftypes:
--iftype-disable station,ap,adhoc,p2p_client,p2p_go,mesh_point

Disable cipher types:
--cipher-disable ccmp,tkip,wep
2019-05-13 16:38:23 -05:00
James Prestwood
79dd04c13a test-runner: in native HW mode, check for AP capabilities
The test should be aborted if there are not enough radios that support
AP mode. The iftype attribute will now be parsed during the wiphy dump
and a flag is set on the wiphy so we know to skip this radio when
creating the hostapd instances. Since hostapd gets started first, it
will automatically choose all the radios it needs which support AP mode.
This leaves the remainder of the radios (potentially STA only) for IWD.
2019-05-13 14:13:34 -05:00
James Prestwood
7c7aafa812 test-runner: skip hostapd iface creation if native HW
In native hardware mode the interfaces already exist, so there is
no need for hostapd to create new ones.
2019-05-13 14:13:18 -05:00
James Prestwood
97ec50ce28 test-runner: fix radio index 0 deletion
In the PCI/USB passthrough changes the wiphy ID was changed to be an
unsigned integer, where id zero corresponded to an error when in native
hardware mode. Along with this, the radio ID for hwsim was changed to a
pre-increment (only in test-runner), so the radio IDs would start at 1.
The repercussions were not fully investigated, but if they were it would
have been seen that hwsim creates radios IDs starting at zero. This left
test-runner and hwsim with unsynchronized radio IDs, and radio zero
never got deleted after each test causing each successive test to
discover old radio IDs.
2019-05-13 14:11:54 -05:00
Andrew Zaborowski
08ec88671a manager: Also delete interfaces without an ifindex, cleanup 2019-05-13 14:10:49 -05:00
Andrew Zaborowski
0b109c52b7 netdev: Add missing initialization
ifaddr is not guaranteed to be initialized, I'm not sure why there was
no compiler warning.  Also replace a | with a || for boolean conditions
and merge the wiphy check with that line.
2019-05-13 14:10:49 -05:00
Andrew Zaborowski
b86af171f0 netdev: Don't warn on genl messages not matching a netdev
This is going to be a normal situation when we start using interfaces
without an ifindex.
2019-05-13 14:10:49 -05:00
Andrew Zaborowski
925095f835 netdev: Drop remaining whitelist/blacklist code 2019-05-13 14:10:49 -05:00
Marcel Holtmann
c923448bac Release 0.18 2019-05-11 21:32:54 +02:00
Marcel Holtmann
df6003d8c8 build: Add generated certificates to .gitignore 2019-05-11 21:32:17 +02:00