On 07/13/2011 07:42 AM, Matthias Bolte wrote:
2011/7/9 Eric Blake <eblake(a)redhat.com>:
> The compiler might optimize based on our declaration that something
> is unused.
Can this actually happen? The unused marker only says that something
_might_ be unused. I don't think that a compiler can optimize
something based on this when it cannot actually prove that it is
really unused.
Hmm, given gcc's documentation that it is a 'might' be unused, then
yeah, gcc shouldn't do premature optimizations on the caller side. But
better safe than sorry.
> Putting that declaration in the header risks getting
> out of sync with the actual implementation, so it belongs better
> only in the .c files. We were mostly compliant, and a new syntax
> check will help us in the future.
This is a valid point.
Consistency is a good argument, even if the argument for (lack of)
compiler optimizations is weak in this case :)
ACK.
I've now applied 25, 26, and 28. Expect a v3 later today which fixes
the fallout comments on the remaining patches.
--
Eric Blake eblake(a)redhat.com +1-801-349-2682
Libvirt virtualization library
http://libvirt.org