The Following 10 Users Say Thank You to lizardo For This Useful Post: | ||
|
2011-08-12
, 13:57
|
Posts: 284 |
Thanked: 320 times |
Joined on May 2010
@ Peterborough, UK
|
#32
|
|
2011-08-12
, 14:02
|
Posts: 53 |
Thanked: 90 times |
Joined on Nov 2009
@ Manaus, Brazil
|
#33
|
The Following User Says Thank You to lizardo For This Useful Post: | ||
|
2011-08-12
, 17:01
|
Posts: 13 |
Thanked: 0 times |
Joined on Aug 2009
@ SC; USA
|
#34
|
|
2011-08-13
, 09:12
|
Moderator |
Posts: 5,320 |
Thanked: 4,464 times |
Joined on Oct 2009
|
#36
|
http://pastebin.com/x0LiTcDB
Looks like the WL1271_spi driver supports RFMON, but it's likely broken again ala wl1251. No support for live frame injection yet. Once kernels get up and running, mac80211 can be tossed on and then patched. This is high on my priority list along with FMTX.
If anybody has any project requests or wants to discuss something less "mainstream", please post up.
|
2011-08-13
, 17:04
|
Posts: 284 |
Thanked: 161 times |
Joined on Dec 2009
|
#37
|
|
2011-08-14
, 12:47
|
|
Posts: 1,030 |
Thanked: 792 times |
Joined on Jun 2009
|
#38
|
The Following User Says Thank You to hawaii For This Useful Post: | ||
|
2011-08-14
, 21:58
|
|
Posts: 560 |
Thanked: 423 times |
Joined on May 2010
@ Switzerland
|
#39
|
|
2011-08-15
, 05:09
|
|
Posts: 4,708 |
Thanked: 4,649 times |
Joined on Oct 2007
@ Bulgaria
|
#40
|
I few ideas for people who want to build custom kernels for N950 in order to disable aegis:
* If you want to disable Aegis, just disabling it on "make menuconfig" will actually make the system unbootable (from my own experience). This is because various commands used during initialization will begin to fail.
* My tip to "disable" Aegis is to actually neutralize it, i.e. make the userspace think it is enabled, but on kernel side it is not enforced.
For instance, aegis is not "sealed" by default. At some point during initialization it is sealed with a write to a /sys/... file. So you can modify the function which implements the writing on kernel side (write_enable() in security/aegis/validator/enforce.c) to actually not set the seal bit, so it remains unsealed. You may also need to disable the bit which prevents "unsigned" kernel modules from being loaded, given that you built the kernel yourself, the hashes have probably changed.
I was too lazy to verify what each of those bits in /sys/.../enforce means, so I simply added a:
memset(&valinfo, 0, sizeof(valinfo));
right before "return count" in write_enable().
Hope that helps,
Anderson Lizardo