[patch] Produce .weak_def_can_be_hidden for some linkonce_odr values

[Nick Lewycky]

Rafael Espíndola wrote:
> On 29 October 2013 00:33, Rafael Espíndola<rafael.espindola at gmail.com>  wrote:
>>>> How much do we lose
>>>> if you don't do this recomparison, is that tolerable? I realize that getting
>>>> a second run of globalopt at the end is going to be time consuming to prove
>>>> worthwhile, but I really think it's the right way to go.
>>>
>>> With a notaddrtaken attribute the run of GlobalOpt we have would be
>>> sufficient. This check is at the very end of the compilation.  I will
>>> benchmark a build without the GlobalStatus::analyzeGlobal and report
>>> as soon as the bootstrap finishes.
>>>
>>
>> The result is far more modest:
>>
>> $ nm build-patch2/bin/clang | grep ' T ' | wc
>>
>>     20850 62550 1763368
>
> The more restricted patch is attached. We can start with that if you want.

Nope that's fine, I'm convinced you had it right the first time. LGTM!

Nick