[cfe-dev] Working on the rest of PR10063: destructors and the CFG are causing issues with -Wreturn-type

[Chandler Carruth]

On Mon, Sep 12, 2011 at 11:30 AM, Ted Kremenek <kremenek at apple.com> wrote:

> On Sep 12, 2011, at 2:22 AM, Chandler Carruth wrote:
>
> If this patch is OK, or after any performance tuning you'd like to see
> here, I think I also better understand the FIXME in the analysis file. Do
> you actually want to switch the CFG to have *no* successors for blocks which
> contain a no-return element? That seems like it would be a very clean way of
> modelling this and would simplify the analysis for -Wreturn-type
> significantly.
>
>
> One possibility is to create a special empty block, "noreturn", that all
> blocks ending with a noreturn destructor/call jump to.  That block would
> then have "Exit" as its successor.  -Wreturn-type could just walk the
> predecessors of Exit, and skip this block.
>

That would definitely work.

The other idea I had was to sink an 'isNoReturn' bit into the CFGBlock
itself and just mark blocks which appropriately. Then it would just walk the
predecessors and skip all blocks with the bit set. It would also prevent
having an extra block in the CFG. This might be no more costly than the
added CFGBlock as we'd need some way to distinguish it from any other
CFGBlock... I'd have to look at the implementation details of it though.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/cfe-dev/attachments/20110912/8f2f9bb0/attachment.html>