[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#708070: Dear maintainers,




Thank you so much for quick reply.

I think you are absolutely right for Debian Stable.
And I don't think I was right even in Debian Testing
because I do not know what is done in the trinity test.
I think I understand why the conservative approach is needed.

I was merely curious about what makes the differences
between x32-specific security bugs and non-x32-specific ones.
I believe the x32 binary support is just an option not different from the other kernel options;
and any kernel option increases the threatening level in security.
But we do not turned off all the kernel options; even though we all know that bugs are everywhere.
I can totally agree with you if the reason why x32-option is turned off is
that it can't be built as a module, rarely used, or hard to test.
But if not, I just thought x32-specific feature can be tested in Sid or Testing like the other features
because those stages aim to find out bugs.

Anyhow I really appreciate that you inform me of the current situation and the progress.
If you have any hyperlink to knowledge base about trinity test, please let me know. ( I googled it for a while bug found nothing. )

Thank you.



Regards,
Hae-woo Park




2013/9/22 Ben Hutchings <ben@decadent.org.uk>
On Sun, 2013-09-22 at 21:32 +0900, Hae-woo Park wrote:
>
> AFAIK, there are many x32 packages in amd64 repository.
>
> However, they are unusable due to that the kernel packages turned off
> the related option.
> Actually I do not know what the trinity test is; however,
> IMHO, the option could be turned on for user tests, at least on
> Jessie.

I don't think you understand my concern.  If there is any security issue
specific to x32 syscalls then we make all the amd64 installations
vulnerable, not just those where the administrator actually wants to
support x32.

> Is there any progress in this issue ?

I think Matthias Klose was going to try running trinity on x32.

Ben.

--
Ben Hutchings
compatible: Gracefully accepts erroneous data from any source


Reply to: