[cfe-commits] r161501 - in /cfe/trunk: include/clang/AST/Expr.h include/clang/Basic/DiagnosticGroups.td include/clang/Basic/DiagnosticSemaKinds.td lib/AST/Expr.cpp lib/Sema/SemaExpr.cpp lib/Sema/SemaOverload.cpp test/SemaCXX/constant-expression-c

[David Blaikie]

On Thu, Aug 9, 2012 at 11:08 AM, jahanian <fjahanian at apple.com> wrote:
>
> On Aug 9, 2012, at 11:00 AM, David Blaikie wrote:
>
>> On Thu, Aug 9, 2012 at 10:38 AM, jahanian <fjahanian at apple.com> wrote:
>>>
>>> On Aug 9, 2012, at 9:52 AM, David Dean wrote:
>>>
>>>> It looks like you just disabled the warning, wouldn't it be good to actually add it in as expected?
>>>> (and remove the one we've been ignoring?)
>>>
>>> Good idea, go ahead. However, it will be subject to future failures as warning texts tend to change.
>>> Original intention of tests were not meant to check specific warnings.
>>
>> What's it meant to check, then?
>
> Should have said that if original test was testing some diagnostics then makes sense to check the warnings.
> If intention of the test was to check for crash, code gen. etc. Then checking for warning is not needed.
> - fariborz

Ah, fair enough. Perhaps we could turn off all warnings on the
crash/codegen tests if that's easy to do mechanically (both easy to
classify and modify)

- David

>
>>
>>> - Fariborz
>>>
>>>>
>>>> On 9 Aug 2012, at 9:46 AM, jahanian <fjahanian at apple.com> wrote:
>>>>
>>>>>
>>>>> On Aug 9, 2012, at 9:45 AM, David Dean wrote:
>>>>>
>>>>>> After perusing the logs further, I've decided not to pester you with them.
>>>>>> We were, in fact, ignoring the missing diagnostic in nullptr1.C
>>>>>>
>>>>>> I'll update the tests.
>>>>>
>>>>> It is done.
>>>>> - fariborz
>>>>>
>>>>
>>>> -David
>>>>
>>>>
>>>
>