[cfe-dev] [PATCH 1/1] Analysis: CFG, add comment about NULL CFGBlock successors

Ted Kremenek kremenek at apple.com
Wed Apr 27 23:24:39 PDT 2011


committed in r130387.  Thanks!

On Apr 27, 2011, at 11:18 PM, Jiri Slaby wrote:

> CFGBlock successor may be NULL to preserve succs ordering rules. Note
> this to the CFGBlock documentation.
> 
> Signed-off-by: Jiri Slaby <jirislaby at gmail.com>
> ---
> include/clang/Analysis/CFG.h |    2 ++
> 1 files changed, 2 insertions(+), 0 deletions(-)
> 
> diff --git a/include/clang/Analysis/CFG.h b/include/clang/Analysis/CFG.h
> index 53ed705..ca46459 100644
> --- a/include/clang/Analysis/CFG.h
> +++ b/include/clang/Analysis/CFG.h
> @@ -264,6 +264,8 @@ public:
> ///     ? operator      LHS expression;  RHS expression
> ///     &&, ||          expression that uses result of && or ||, RHS
> ///
> +/// But note that any of that may be NULL in case of optimized-out edges.
> +///
> class CFGBlock {
>   class ElementList {
>     typedef BumpVector<CFGElement> ImplTy;
> -- 
> 1.7.4.2
> 
> 




More information about the cfe-dev mailing list