[llvm-commits] [llvm-gcc-4.2] r71287 - in /llvm-gcc-4.2/trunk: build_gcc gcc/ChangeLog.apple gcc/config/arm/arm.c gcc/cp/mangle.c gcc/objc/ChangeLog.apple gcc/objc/objc-act.c gcc/testsuite/ChangeLog.apple gcc/testsuite/g++.apple/anon-1.C gcc/testsuite/gcc.apple/weak.c gcc/testsuite/objc.dg/property-16.m gcc/tree-eh.c gcc/version.c

[Mike Stump]

On May 11, 2009, at 12:18 AM, Duncan Sands wrote:
> isn't that exactly what ExternalWeakLinkage is about?  A variable that
> ends up with WeakLinkage in LLVM is always defined.

I think you didn't catch my example in my previous email where I  
detailed how I think the above isn't true.  I think this gap comes  
from trying to treat defined at static link time the same as defined  
at runtime; they are different concepts and you have to treat them  
differently.

Now, I must admit, I didn't run the testcase I'm thinking about though  
the system to see if it behaves as I expected, but I think I have it  
right.