[cfe-commits] [Patch] -Wduplicate-enum which fixes PR6343

[Richard Trieu]

On Wed, Jul 18, 2012 at 5:41 PM, Ted Kremenek <kremenek at apple.com> wrote:

> Hi Richard,
>
> I think this checking is quadratic, n(n+1)/2.
>
Yes it is.  It does perform a check between each enum in the worst case.


>  This might be fairly expensive on a large set of enums.  Do you have any
> performance numbers?
>
I haven't run any performance tests.  I will try running some when I get
the chance.


>  I agree that it is a good checking, but the way it is implemented is
> likely to be noticeably slow on some (important) cases.
>
How many elements in an enum would be in these important cases?

>
> Ted
>
> On Jul 18, 2012, at 4:46 PM, Richard Trieu <rtrieu at google.com> wrote:
>
> http://llvm.org/bugs/show_bug.cgi?id=6343
>
> Warn on enum elements that are assigned values already in use.  This is
> based on the misconception that elements not given a value will be given
> the next smallest, unused value.  Instead, elements are assigned 1 more
> than the previous value.  Some example code this warning will catch:
>
> enum { A, B, C, Aref = A, count };
> Both B and count will have value 1.
>
> enum { A, B, C, D = -1, E, F };
> A = E = 0
> B = F = 1
>
> This warning found one such issue in LLVM that was fixed in r160465.
> <duplicate-enum.patch>_______________________________________________
> cfe-commits mailing list
> cfe-commits at cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/cfe-commits/attachments/20120718/c9c6f9e9/attachment.html>