On Thu, Jun 14, 2012 at 1:05 PM, David Blaikie <[email protected]> wrote:

> On Thu, Jun 14, 2012 at 11:08 AM, Chandler Carruth <[email protected]>
> wrote:
> > The driver support for rewrite includes LGTM, commit away.
>
> r158463
>
> > Regarding the crash reporting, that looks too easy! ;]
>
> It does, doesn't it...
>
> > Does anything need to be changed in the reproduction script to make it
> use the include-rewritten
> > file instead of a fully preprocessed file?
>
> Ah, I hadn't actually tried the repro script file - I tend to just run
> the crashers myself (but of course then I don't get the extra flags,
> etc, from the repro).
>
> The script is generated based on the file name reported as the output
> file by the compilation command, as I understand/read it - so it's not
> referring to a non-existent file (& the file name remains the same in
> any case).
>
> >  Do we need to change the filename?
>

I should have said filename *extension*.


> It seems to work fine (reckon it'd be appropriate to have a test case
> for this that actually executes the crash report script file? that
> seems a little nasty, but with some value). The -cc1 command includes
> an explicit -x so I don't think there's any file extension based
> language detection (so it doesn't try to skip preprocessing, for
> example).


Wild. I was thinking the '.ii' extension might cause it to do the wrong
thing. I think we should still fix the extension to be the same as the
original file.
_______________________________________________
cfe-commits mailing list
[email protected]
http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits

Reply via email to