jeltz/hostap
1
0
Fork 0
Code Issues Pull requests Projects Releases Wiki Activity
569 commits 3 branches 0 tags 27 MiB
Commit graph

569 commits

This branch
This branch
All branches
Author SHA1 Message Date
Jouni Malinen
93ea8ed034 Partial fix for test_wpa build 2009-01-13 19:37:22 +02:00
Jouni Malinen
0b8695bb2b Renamed shadowed variable 2009-01-13 19:31:51 +02:00
Jouni Malinen
61e64e7b7f Removed dead code 2009-01-13 19:15:42 +02:00
Jouni Malinen
c77bb61aee Fixed copy-paste error in retry configuration (incorrect ioctl used) 2009-01-13 19:14:38 +02:00
Jouni Malinen
25ba219f3e Removed unused set_regulatory_domain driver function 2009-01-13 19:12:25 +02:00
Jouni Malinen
f49275c92f Fixed build without CONFIG_IEEE80211N 2009-01-12 22:08:53 +02:00
Jouni Malinen
4cdde5caf7 Fixed a typo in printf -> wpa_printf changes 2009-01-12 22:06:51 +02:00
Jouni Malinen
24bb3d9788 Fixed build with wpa_printf disabled 2009-01-12 22:06:06 +02:00
Jouni Malinen
a9843f614f Removed hw_features.o from build if hostapd MLME is not used 2009-01-12 22:03:58 +02:00
Jouni Malinen
7fd46d466f Added build option for disabling control interface (CONFIG_NO_CTRL_IFACE) 2009-01-12 21:44:57 +02:00
Jouni Malinen
30b32314eb Added build option for removing VLAN support (CONFIG_NO_VLAN) 2009-01-12 21:39:19 +02:00
Jouni Malinen
71f04b3cf9 Fixed CONFIG_NO_HOSTAPD_LOGGER build 2009-01-12 21:21:31 +02:00
Jouni Malinen
27750f29d6 Comment out RADIUS configuration parsing if RADIUS client support is 
disabled in the build.
 2009-01-12 16:49:39 +02:00
Jouni Malinen
10656fc23a Replaced printf() calls with wpa_printf() 2009-01-12 16:49:04 +02:00
Jouni Malinen
815d2189d9 Remove TLS-PRF from hostapd build if it is not needed 2009-01-12 15:20:12 +02:00
Jouni Malinen
03ba2cb4c6 Added CONFIG_NO_AES_EXTRAS for hostapd 
This allows unneeded AES routines to be removed from the build to reduce
binary size.
 2009-01-12 15:15:35 +02:00
Jouni Malinen
bc521fac13 Fixed couple of build failures with minimal configuration 2009-01-12 15:08:33 +02:00
Kel Modderman
e0f6d3d9b5 wpa_gui-qt4: silence compile warning + wpsTab enable/disable 
Fix compile warning by renaming gridLayout to wpsGridLayout in
wpagui.ui, it seems to be used internally.

/usr/bin/uic-qt4 eventhistory.ui -o .ui/ui_eventhistory.h
/usr/bin/uic-qt4 scanresults.ui -o .ui/ui_scanresults.h
/usr/bin/uic-qt4 wpagui.ui -o .ui/ui_wpagui.h
Warning: name gridLayout is already used
/usr/bin/uic-qt4 userdatarequest.ui -o .ui/ui_userdatarequest.h
/usr/bin/uic-qt4 networkconfig.ui -o .ui/ui_networkconfig.h

Also disable WPS tab if the running supplicant doesn't support that,
just as the WPS action menu is.

Signed-off-by: Kel Modderman <kel@otaku42.de>
 2009-01-11 21:44:07 +02:00
Jouni Malinen
3904625396 Silenced some sparse warnings 2009-01-11 10:42:07 +02:00
Jouni Malinen
b7fc385071 Fixed build with 802.11n disabled 2009-01-10 21:36:41 +02:00
Jouni Malinen
47dc21c70e Include SIM simular for EAP-SIM in Windows build 
USIM simulator for EAP-AKA was already included, but the SIM version was
forgotten from the build script.
 2009-01-09 21:45:31 +02:00
Jouni Malinen
5fa30f32da Cleaned up main() by splitting it into number of helper functions 2009-01-09 18:04:57 +02:00
Jouni Malinen
b5b969e9c7 Moved driver callback functions from hostapd.c to a new file 2009-01-09 17:27:03 +02:00
Jouni Malinen
81897f4c88 Moved TKIP countermeasures from hostapd.c to its own file 2009-01-09 17:12:33 +02:00
Jouni Malinen
9302c5e19c Cleanup: move related functions closer to eachother 2009-01-09 17:06:54 +02:00
Jouni Malinen
3313cea0d6 Fixed TKIP countermeasueres for drivers that do not use hostapd MLME 
The Michael MIC reporting functions have to be included even when using
driver MLME.
 2009-01-09 17:00:29 +02:00
Jouni Malinen
649d8890b1 Add wrapper functions for IEEE 802.11 driver calls into driver.h 
These functions allow the driver_*.c wrappers not to include
ieee802_11.h.
 2009-01-09 16:55:59 +02:00
Jouni Malinen
b83cbb3b3b Use a wrapper function for ieee802_1x_receive() to avoid ieee802_1x.h 
Driver wrappers should not need to include ieee802_1x.h, so use a
wrapper function declared in driver.h.
 2009-01-09 16:33:59 +02:00
Jouni Malinen
33eb3c4ec4 Moved disassociation processing away from driver_*.c 2009-01-09 16:27:30 +02:00
Jouni Malinen
940a0ce970 Moved association check from driver_*.c into ieee802_1x_receive() 2009-01-09 16:17:20 +02:00
Jouni Malinen
602996f8db Moved WPA setup etc. association processing away from driver_*.c 
This is all details that should not need to be handled in driver_*.c.
 2009-01-09 16:13:07 +02:00
Jouni Malinen
214021f585 Move RX-from-unknown-STA processing away from driver_*.c 
This cleans up the driver wrapper interface by getting rid of sta_info.h
dependency in all drivers that use MLME implementation in hostapd
(driver_hostap.c and driver_nl80211.c).
 2009-01-09 15:44:45 +02:00
Jouni Malinen
8607f4c31f Moved TX status processing for STA entries away from driver_*.c 
Driver wrappers should not need to know about this level of core hostapd
details.
 2009-01-09 15:33:36 +02:00
Jouni Malinen
76e2592190 Share the same radiotap helper implementation 2009-01-09 15:22:19 +02:00
Jouni Malinen
d200828b41 Moved rfc1042_header into driver_*.c that use it 
No need to share such a small buffer and make hostapd.[ch] more complex
with it.
 2009-01-09 15:06:48 +02:00
Jouni Malinen
a2f517fb93 Fixed a typo in a comment 2009-01-09 15:01:50 +02:00
Jouni Malinen
8383da8009 Fixed eapol_test build after RADIUS API changes (const) 2009-01-09 13:27:56 +02:00
Jouni Malinen
8d06da09a4 Move generic definitions away from hostapd.h 
This is an initial step in trying to make hostapd.h not needed to be
included in so many files.
 2009-01-08 20:14:05 +02:00
Jouni Malinen
bfddd95c9e Split hostapd/driver.h into two files 
driver.h contains the definitions needed in driver wrapper
implementations (driver_*.c) and driver_i.h contains the definitions
that are used in core hostapd code to interact with the driver wrappers.
 2009-01-08 20:02:56 +02:00
Jouni Malinen
f88bd28836 Added support for removing RADIUS accounting and RADIUS in general 
CONFIG_NO_ACCOUNTING=y and CONFIG_NO_RADIUS=y build options can now be
used to remove RADIUS support from the hostapd build.
 2009-01-08 19:15:25 +02:00
Jouni Malinen
9c584c06bd Remove MLME code from build if none of the enabled drivers need it 2009-01-08 18:40:14 +02:00
Jouni Malinen
bb305cbdcc Replaced printf() with wpa_printf() 2009-01-08 16:57:04 +02:00
Jouni Malinen
03018d3e7c Added CONFIG_NO_STDOUT_DEBUG for hostapd 
This is similar to one in wpa_supplicant and can be used to reduce the
binary size by removing debugging messages.
 2009-01-08 16:47:04 +02:00
Jouni Malinen
7d02e64157 Mark shared secret const in RADIUS client routines 2009-01-08 16:41:47 +02:00
Jouni Malinen
1c6e69ccda Moved documentation from developer.txt into source code files 
Use Doxygen comments for functions to replace the old text file that was
not up-to-date anymore.
 2009-01-08 16:33:00 +02:00
Jouni Malinen
72f24de82c Create PNG icon before running qmake to avoid resource warning 2009-01-06 20:22:52 +02:00
Jouni Malinen
6f78f2fb3b Preparations for 0.6.7 release 2009-01-06 20:11:15 +02:00
Eygene Ryabinkin
dfa141b1b0 wpa_cli: Embed command usage into the wpa_cli_commands array 
It is better to have command description around the command definitions:
the chances for omitting usage, forgotting to change/delete them, etc.
are smaller than for the separate usage statement. [Bug 290]

Signed-off-by: Eygene Ryabinkin <rea-fbsd@codelabs.ru>
 2009-01-05 21:33:34 +02:00
Eygene Ryabinkin
40fd868c09 wpa_cli: Improve sensitive command detection for readline history 
I had added flags to the every command description: just now the only
meaningful flag tells that this command has sensitive arguments and it
shouldn't be written to the disk.  I rewrote the logics for the search
for the sensitive commands: special procedure is now loops over all
commands and tries to see if command has sensitive data. [Bug 289]

Signed-off-by: Eygene Ryabinkin <rea-fbsd@codelabs.ru>
 2009-01-05 21:07:54 +02:00
Eygene Ryabinkin
413653e839 wpa_cli: fix readline history cleaning 
First of all, the history had not been written to the disk, since almost
all commands were cleaned up due to the error in the history cleaning:
the return value of the last os_strncasecmp() call was not compared to
zero, but was rather used as is.  So the condition was almost always
true and most commands were removed from the history.

The second problem was that the evaluation of the potentially sensitive
commands was started at the entry number 1, instead of very first entry.
 2009-01-05 20:48:45 +02:00