This is the mail archive of the gcc@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]

Re: demengler in libstdc++ is changed


On Thu, Nov 13, 2003 at 11:26:27PM +0100, Carlo Wood wrote:
> On Wed, Nov 12, 2003 at 05:52:07PM -0800, H. J. Lu wrote:
> > Is that a good idea? Not only there are 2 demanglers in gcc, but also
> > they have different outputs on a given input.
> 
> The two demanglers give different output on many, many
> symbols and types.  This additional difference won't make
> a difference.
> 
> I mean - just consider this for example (but there many,
> many more):
> 
> _Z1fVAszL_ZZNK1N1A1fEvE3foo_0E_i
> 
> libiberty:
> 
> f(int[ sizeof(N::A::f()::foo [# ])const1] volatile)
> 
> libstdc++:
> 
> f(int volatile [sizeof (N::A::f() const::foo)])
> 
> That space won't make a difference, completely independent
> of what is better readable.

Which output is correct? If they are equivalent, can we pick one
and stick to it? Having different outputs is very confusing.

BTW, is anyone working on fixing the demangler in libiberty? There
is a replacement in C++.


H.J.


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