| 1 |
Nikos Chantziaras <realnc@×××××.com> wrote: |
| 2 |
> |
| 3 |
> Well, if you're running a local process that is trying to attack you, |
| 4 |
> you've been compromised already, imo. |
| 5 |
|
| 6 |
By your definition, you are compromised if you surf to the |
| 7 |
wrong webpage with enabled javascript. |
| 8 |
|
| 9 |
While this is arguably true, I would distinguish between various |
| 10 |
degree of compromise and would prefer if nevertheless such webpages |
| 11 |
would not be able to e.g. read the secret keys of a running gnupg |
| 12 |
process. |
| 13 |
|
| 14 |
> So, unless you're running some kind of server that offers execution time |
| 15 |
> to clients |
| 16 |
|
| 17 |
... or use your browser with not always disabled javascript ... |
| 18 |
|
| 19 |
> and the few packages that run untrusted code. |
| 20 |
|
| 21 |
You misunderstand: For the packages which run the code, the mitigations |
| 22 |
like retpoline do not help much. It is the packages which _somehow_ |
| 23 |
react (or can be called) by such a code which need the protection by |
| 24 |
retpoline built-in. |
| 25 |
And this is an awful lot of packages since it includes also all libraries |
| 26 |
which are possibly used by these packages, language interpreters used by |
| 27 |
these packages, etc. If in doubt, I would re-emerge the |
| 28 |
full -e @world with corresponding compiler switches enabled. |
| 29 |
|
| 30 |
Of course, rebuilding @world without changing your C*FLAGS before |
| 31 |
would be pointless. |
Archives