[cfe-commits] Patch for review

[Chris Lattner]

On Feb 17, 2009, at 2:23 AM, Ben Laurie wrote:

> Don't include alloca.h if it doesn't exist.

Ideally the analyzer test cases wouldn't use the system assert.  Why  
not have an analyzer-specific assert.h that "is defined right"?

-Chris

>
>
> Index: test/Analysis/stack-addr-ps.c
> ===================================================================
> --- test/Analysis/stack-addr-ps.c       (revision 64650)
> +++ test/Analysis/stack-addr-ps.c       (working copy)
> @@ -1,9 +1,12 @@
> -// RUN: clang -analyze -checker-simple -verify %s &&
> -// RUN: clang -analyze -checker-cfref -analyzer-store-basic -verify  
> %s &&
> -// RUN: clang -analyze -checker-cfref -analyzer-store-region - 
> verify %s
> +// RUN: clang -analyze -checker-simple -verify -I../../../include  
> %s &&
> +// RUN: clang -analyze -checker-cfref -analyzer-store-basic -verify
> -I../../../include %s &&
> +// RUN: clang -analyze -checker-cfref -analyzer-store-region -verify
> -I../../../include %s
>
> #include <stdlib.h>
> -#include <alloca.h>
> +#include <llvm/Config/config.h>
> +#ifdef HAVE_ALLOCA_H
> +# include <alloca.h>
> +#endif
>
> int* f1() {
>   int x = 0;
> _______________________________________________
> cfe-commits mailing list
> cfe-commits at cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits