D5029: Use seccomp for implementing a sandbox for kscreenlocker_greet

[Martin Gräßlin]

graesslin added inline comments.

INLINE COMMENTS

> broulik wrote in seccomp_test.cpp:95
> It might fail because KDE CI infrastructure does a redirect we don't follow instead of failing because we actually restricted it.

I'm pretty sure that it would fail properly for the IP variants in such a case.

> broulik wrote in greeterapp.cpp:148
> Yeah but it will lead to code in the form of
> 
>   return new Authenticator(Authenticator::AuthenticationMode::Delayed, this);
>   return new Authenticator(Authenticator::AuthenticationMode::Direct, this);
> 
> the latter being unreachable. Just being picky here, though, feel free to ignore.

That is totally fine. Having different return values in different paths is quite common. E.g. an early exit with a return nullptr.

REVISION DETAIL
  https://phabricator.kde.org/D5029

To: graesslin, #plasma
Cc: broulik, plasma-devel, progwolff, lesliezhai, ali-mohamed, jensreuterberg, abetts, sebas, apol
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/plasma-devel/attachments/20170312/b8bddd06/attachment.html>