On Fri, Apr 17, 2015 at 1:25 PM, Oliver Chang <[email protected]> wrote: > > On 17 April 2015 at 06:31, Aaron Ballman <[email protected]> wrote: >> >> I agree with Richard's comments about this being extended to apply to >> a list of sanitizers. I wish we would have done that for the >> no_sanitize_blah attributes. >> >> Also, why no C++-style attribute? > > > > You mean something like GCC<"no_sanitize_vtpr"> instead?
No, that would only be if GCC supported the attribute. I was talking about two spellings: GNU<"no_sanitize_vptr"> and CXX11<"clang", "no_sanitize_vptr"> > >> >> I think some more documentation would be nice. People aren't always >> familiar with what those vptr checks would look like, or why they >> might want to have them removed. What's the benefit to applying this >> attribute, how would you decide when it's appropriate to add it, etc. > > > Would that documentation belong here? Yes, I believe it does. ~Aaron > >> >> >> >> Why is the shell required for this test? > > > I forgot to delete this when I copied the file from another test. Removed. > > > -- > Cheers, > Oliver _______________________________________________ cfe-commits mailing list [email protected] http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
