[cfe-commits] r79941 - in /cfe/trunk: lib/Analysis/CFG.cpp test/SemaTemplate/fun-template-def.cpp

[Douglas Gregor]

On Aug 25, 2009, at 10:40 AM, Chris Lattner wrote:

>
> On Aug 25, 2009, at 8:30 AM, Douglas Gregor wrote:
>
>>>
>>> Ok point taken.  Here's another way to look at it: is it  
>>> worthwhile (in terms of compile time) to check all uninstantiated  
>>> templates?  It would be better to only do these "heavier" semantic  
>>> checks on inline functions and templates that are actually used in  
>>> a translation unit rather than in every translation unit that they  
>>> are parsed in.  This applies equally well to C inline functions as  
>>> it does to all the fun C++ stuff, it's just that C++ will suffer  
>>> more if we don't do something like this.
>>>
>>> Thoughts?
>>
>>
>> I find these checks useful (as a programmer), as much in  
>> uninstantiated templates and inline functions as anywhere else.  
>> However, I suggest that we delay this discussion until we can  
>> quantify the actual compile-time cost on, e.g., <algorithm>,  
>> <numeric>, or <iostream>.
>
> I don't disagree about either point. :)  I'm just saying that one  
> way to reduce the cost is to only check each inline function in  
> translation units that use them.  This doesn't help library authors,  
> but does help application authors who (presumably) use every inline  
> function at least once in their program.

Oh, I agree with you. But we're talking about a tradeoff between early  
detection of problems and performance, but I don't know what we're  
gaining on the performance end.

> Deferring it is fine with me though :)


It's in my calendar to revisit at the point when I think we'll be able  
to parse <algorithm> well enough to measure something :)

	- Doug