[PATCH] Add warning capabilities in LLVM (backend part), Take 2

[Quentin Colombet]

On Dec 18, 2013, at 2:57 AM, Tobias Grosser <tobias at grosser.es> wrote:

> On 12/16/2013 06:41 PM, Quentin Colombet wrote:
>> Hi Tobias,
>> 
>> On Dec 14, 2013, at 1:20 AM, Tobias Grosser <tobias at grosser.es> wrote:
>> 
>>> On 12/12/2013 06:40 PM, Quentin Colombet wrote:
>>>>   - Suppress DiagnosticInfoOther class.
>>>>   - Add DK_FirstPluginKind and the related getNextAvailablePluginKind function.
>>>> 
>>>> Hi dblaikie, rengolin, chandlerc, echristo,
>>>> 
>>>> http://llvm-reviews.chandlerc.com/D2376
>>> 
>>> Hi Quentin,
>>> 
>>> I just wanted to say that I did a simple test with this in Polly (as a plugin), and it seems to work as expected.
>> I am glad it worked for you!
>> 
>>> 
>>> I found one minor issue:
>>> 
>>> - The return type of getNextAvailablePluginDiagnosticKind should be
>>>  DiagnosticKind, not int.
>> Well, technically, the returned type is not DiagnosticKind, as the returned value is not a valid value in the statically known DiagnosticKind enum.
> 
> True.
> 
> I added a missing namespace to the implementation of getNextAvailablePluginDiagnosticKind in r197556.
Good catch, thanks!

> 
> I also used this interface to implement the first diagnostic in Polly (r197558).
> 
> I also saw you started to submit patches to clang. For plugin authors, it would be amazing if a frontend could optionally register a callback that demangels function names. Like this, the plugin default printers could return something sensible.
Yes, we have to figure out how to do that.

Thanks,
-Quentin
> 
> Thanks again,
> Tobias
>