[Lldb-commits] [lldb] r240606 - Handle (or at least don't crash) trying to get the encoding for a bunch of new builtin types in clang trunk

[Enrico Granata]

> On Jun 30, 2015, at 2:58 AM, Pavel Labath <labath at google.com> wrote:
> 
> I would prefer if we were going in the direction of decreasing the number of warnings rather than increasing it

Hard to disagree with that one.

> (right now, this is the only warning that I get for my build config). Having the default clause in there also sounds reasonable to me, so how about we find a way to suppress this warning without changing semantics?
> option a: cast the switched value to int: this should get rid of the warning in this case..

If anything, this feels like a hack-around the warning. Given that there is at least a better option, I’d say “nay” personally

> option b: compile with -Wno-covered-switch-default: I personally think I could live just fine without this warning

This would be a good one, IMHO

> option c: ???
> 

None from me. But if anyone has other ideas, now is a good time to share them

> pl
> 
> 
> On 29 June 2015 at 23:47, Enrico Granata <egranata at apple.com <mailto:egranata at apple.com>> wrote:
> 
>> On Jun 29, 2015, at 11:58 AM, Ed Maste <emaste at freebsd.org <mailto:emaste at freebsd.org>> wrote:
>> 
>> On 29 June 2015 at 14:47, Enrico Granata <egranata at apple.com <mailto:egranata at apple.com>> wrote:
>>> 
>>>> I can see arguments for and against removing the default, but I think
>>>> removing it will cause a new warning to appear if Clang adds other new
>>>> kinds in the future.
>>> 
>>> Yes. And I am slightly more worried about the silent failure when that happens rather than the warning per se - which is why I am hesitant to just say “yes please remove"
>> 
>> I mean the warning will prompt us to go and add any new missing enum
>> values
> 
> Until we’re busy and forget to do it and some builds end up with bugs because of silent fall-through at runtime :-)
> 
>> if this happens -- especially if we enable -Werror :)
> 
> Which I suspect we’re not doing anytime soon
> 
>> 
>> I don't feel too strongly one way or the other,
> 
> Likewise. I have a slight preference for keeping the default in there, but only so slight
> 
>> but it seems we have
>> removed the defaults in all of these cases in the past.
> 
> 
> Thanks,
> - Enrico
> 📩 egranata@.com ☎️ 27683
> 
> 
> _______________________________________________
> lldb-commits mailing list
> lldb-commits at cs.uiuc.edu <mailto:lldb-commits at cs.uiuc.edu>
> http://lists.cs.uiuc.edu/mailman/listinfo/lldb-commits <http://lists.cs.uiuc.edu/mailman/listinfo/lldb-commits>
> 
> 


Thanks,
- Enrico
📩 egranata@.com ☎️ 27683

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/lldb-commits/attachments/20150630/b145d34c/attachment.html>