This is the mail archive of the 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: Dealing with C++98/11 ABI incompatibilities

On 07/03/2012 03:18 PM, Jason Merrill wrote:
2) Object layout changes to std::list and std::basic_string.  For these
types, there is no way to both retain backward compatibility with older
C++98 code and conform to the C++11 standard.  The best we can hope for
is to allow old code to coexist with new code so long as they don't try
to touch the same string/list objects.

As above, one solution to this would be to change the linkage name of
the new versions so that they don't clash with the old versions.  But
that isn't enough in this case, as it wouldn't affect the linkage name
of a class like

struct Wrap { std::string s; };

so we would need some way to cause the name decoration to propagate to
other containing/derived classes.

I raised this issue on the ABI list, and someone pointed out to me that incomplete classes make quiet propagation impossible; any full coexistence solution will require changes to affected user types. There seem to be two options:

1) No propagation, add a warning to suggest that users decorate their types appropriately.
2) Propagate to class definitions, give an error if there was an earlier forward declaration without the appropriate tag.

I lean toward #1; people who rebuild everything regularly won't be affected by it, and people who need coexistence can use the warning flag to help them adjust.


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