[cfe-commits] r76637 - in /cfe/trunk/test/CodeGenObjC: property-agrr-getter.m protocols-lazy.m

[Fariborz Jahanian]

On Jul 21, 2009, at 2:11 PM, Mike Stump wrote:

> On Jul 21, 2009, at 1:54 PM, Fariborz Jahanian wrote:
>> On Jul 21, 2009, at 1:50 PM, Mike Stump wrote:
>>
>>> Author: mrs
>>> Date: Tue Jul 21 15:50:41 2009
>>> New Revision: 76637
>>>
>>> URL: http://llvm.org/viewvc/llvm-project?rev=76637&view=rev
>>> Log:
>>> Prep for new warning.
>>
>> Can you give more context to above statement.
>
> Sure, this is part of the set of changes that began with (quoted  
> from the first check-in comment in the series):
>
>> Prep for new warnings about control flow falling off the ends of
>> functions that return a value.  I was going to buffer the whole lot
>> up, but it should be easier to review if I check them in
>> incrementally.  Most of the forth coming changes either add a return
>> value, or make it impossible to return, or alter the return type.
>
> The idea is that we want to warn by default for control flow off the  
> ends of functions that return a value.  To do this, all the test  
> cases need to avoid tripping the new warning.  I'm doing this prep  
> work first, so that when the warning goes, there are 0 new failures  
> in the test suite.

Warning on by default will break (our) user code. Try this most  
trivial test case with gcc.

@interface I @end
@implementation I
- Meth {}
@end

-Wmost is OK though.

- Fariborz

>
>
> Real code will never trip the code, as real code would return a  
> value appropriate for the function (or it should be fixed).