alexfh added a comment. In https://reviews.llvm.org/D38171#925929, @xazax.hun wrote:
> In https://reviews.llvm.org/D38171#909346, @leanil wrote: > > > In https://reviews.llvm.org/D38171#901427, @xazax.hun wrote: > > > > > One problem to think about when we add all clang-diagnostic as "first or > > > second" class citizen, `checkes=*` might now enable all the warnings > > > which make no sense and might be surprising to the users. What do you > > > think? > > > > > > This is a good point. Should I insert ",-clang-diagnostic*" after any > > (positive) * ? > > > @alexfh do you have some thoughts on this? I don't think this particular point is a large concern. As I mentioned multiple times, enabling all checks is almost never useful due to many checks overlapping or producing conflicting advice. The only place I can think of, where -checks=* is useful is in combination with -list-checks, where the presence of clang-diagnostic- entries would be desired anyway. https://reviews.llvm.org/D38171 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits