Public SourceMage Discussion List

[Eric Sandall <eric AT sandall.us>]

Quoting Ondra Tomecka <otom7630 AT lucy.troja.mff.cuni.cz>:
<snip>
> There should probably also some distinction for higher cpu-specific
> optimization flags, as spells compiled with -msse2 may not work on
> pentium3 cpu, but both of them are marked as i686. Or do you think it
> would be better to compile spells specifically without those higher-level
> optimizations for this tarball cache?

They would be compiled without any higher optimizations (e.g. i386 would run 
on
any i386 or above).

> And don't forget that this has potential for a huge security hole, when
> some malicious user contributes tarball with modified binary. How are we
> supposed to detect such a thing, or prevent it at all? Not that I think
> such tarball cache is bad idea, it would save me quite a lot of time
> sometimes, but still this security concern is a big one.

If the QA Team wants to use their already generated tarballs for this then 
this
concern is probably moot. Unless you don't trust our QA Team? ;)

If the QA Team does pick this up then I would say that we will not accept any
caches from users/developers other than the "official" caches from the QA
process.

-sandalle

--
Eric Sandall                     |  Source Mage GNU/Linux Developer
eric AT sandall.us PGP: 0xA8EFDD61  |  http://www.sourcemage.org/
http://eric.sandall.us/          |  SysAdmin @ Inst. Shock Physics @ WSU
http://counter.li.org/  #196285  |  http://www.shock.wsu.edu/

----------------------------------------------------------------
This message was sent using IMP, the Internet Messaging Program.