This is the mail archive of the gcc-bugs@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

[Bug c++/86301] g++ does not mark declarations used in non-dependent types as referenced


https://gcc.gnu.org/bugzilla/show_bug.cgi?id=86301

--- Comment #5 from zhonghao at pku dot org.cn ---
(In reply to Jonathan Wakely from comment #4)
> If you tried the latest g++ why are you reporting bugs for version 8.0.1?

(In reply to Jonathan Wakely from comment #3)
> (In reply to zhonghao from comment #2)
> > (In reply to Jonathan Wakely from comment #1)
> > > (In reply to zhonghao from comment #0)
> > > > A previous version of clang++ produces similar error message for this code,
> > > > but was fixed.
> > > 
> > > No it didn't, the error shown in the Clang bug comes from GCC!
> > > 
> > > Please make more effort to understand what you are doing, OR I WILL BAN YOUR
> > > ACCOUNT.
> > 
> > Did you try the latest g++ and clang++.
> 
> Yes.
> 
> > They produce different results when
> > they compile the code. My clang is 7.0.0. It does not produce any error
> > messages for this code.
> 
> That doesn't mean the code is valid.

Is the code invalid? Do you mean that clang accepts a piece of invalid code?
That sounds a bug in clang? 

> 
> > Do the difference indicate bugs in g++?
> 
> No. See https://gcc.gnu.org/gcc-7/porting_to.html#hypothetical-instantiation

Is this code a corner case? Or, does it illustrate an undefined behavior?

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]