[cfe-dev] Fix for Clang PR 8426, 8427, & 8433

[Zhanyong Wan (λx.x x)]

Hi Ted,

On Wed, Oct 27, 2010 at 5:55 PM, Ted Kremenek <kremenek at apple.com> wrote:
> Hi Zhanyong,
>
> Thanks for working on this.  I admit that I haven't had a chance to investigate those PRs myself.
>
> The changes look good, but I had two questions/comments.
>
> (1) Do we even want to be analyzing template definitions (as opposed to instantiations)?  It seems like we'd be doing a bunch of guess work.  It also seems like a performance issue if we analyze template code that is never used.

I agree that there's little point analyzing template definitions.  It
would be a quite invasive change to stop doing that though, so I'd
like to get this patch in before attempting that.

> (2) I'd prefer to keep the test cases in one file, for faster processing by 'make test'.  These can easily go into misc-ps-region-store.cpp.  Each test case just needs a comment indicating the PR.

Good point.  Done.

Please see the updated patch attached in this message.  It's also
viewable at  http://codereview.appspot.com/2716044/ as patch set 2.
Thanks,

>
> On Oct 27, 2010, at 4:45 PM, Zhanyong Wan (λx.x x) wrote:
>
>> Hi Ted,
>>
>> I have a fix for Clang static analyzer bugs 8426, 8427, & 8433 (they
>> turned out to be caused by the same thing).  The patch is attached.  I
>> also uploaded them to http://codereview.appspot.com/2716044/ for easy
>> viewing/reviewing.  Would you be able to review it?  Thanks.
>>
>> I have commit privilege, so I can commit it myself when you are OK
>> with it.  Thanks.
>>
>> --
>> Zhanyong
>> <pr8426.patch>
>
>



-- 
Zhanyong
-------------- next part --------------
A non-text attachment was scrubbed...
Name: pr8426.patch
Type: application/octet-stream
Size: 2654 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/cfe-dev/attachments/20101027/7f7f73c7/attachment.obj>