| 1 |
aeriksson@××××××××.fm wrote: |
| 2 |
|
| 3 |
>>please try chpaxing or paxctling it: |
| 4 |
>> |
| 5 |
>>chpax -pemsrv or paxctl --help |
| 6 |
>> |
| 7 |
>>and then find out which protection needs to be disabled, reenable all |
| 8 |
>>the others and tell us what thing you had to turn on/off to make it work. |
| 9 |
>> |
| 10 |
>>i doubt its the hardened and i assume its some kind of memory access/io |
| 11 |
>>access that leads it to fail. |
| 12 |
>> |
| 13 |
> |
| 14 |
> |
| 15 |
> I recently hacked a bit on cpufreqd (unapplied) and discovered that |
| 16 |
> adding elements to the middle of its structures resulted in data |
| 17 |
> corruption. My (limited) investigations suggested that some users of |
| 18 |
> the structs assume a certain layout for memcpy operations. Adding my |
| 19 |
> stuff to the end of the structures fixed it for me. |
| 20 |
> |
| 21 |
> I'm clueless about pax and hardening, but maybe some over optimistic |
| 22 |
> parts of the code got trapped by the pax magic? |
| 23 |
> |
| 24 |
> /A |
| 25 |
> |
| 26 |
> |
| 27 |
> |
| 28 |
> -- |
| 29 |
> gentoo-hardened@g.o mailing list |
| 30 |
> |
| 31 |
yes, and the paxctl and chpax ops also REMOVE the ability to randomize |
| 32 |
with certain arguments, so keep on trying, my friends and tell me back |
| 33 |
what you got, |
| 34 |
|
| 35 |
TIA |
| 36 |
|
| 37 |
Alex |
| 38 |
|
| 39 |
-- |
| 40 |
Chaos is just a perceptional disorder of reality. |
| 41 |
|
| 42 |
-- |
| 43 |
gentoo-hardened@g.o mailing list |
Archives