This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: mutually-recursive types and an old puzzle
- From: kenner at vlsi1 dot ultra dot nyu dot edu (Richard Kenner)
- To: Joe dot Buck at synopsys dot com
- Cc: gcc at gcc dot gnu dot org
- Date: Wed, 30 Jun 04 12:45:14 EDT
- Subject: Re: mutually-recursive types and an old puzzle
Yes, but the question is still relevant, because my real reason for
bringing it up was to make sure that the GIMPLE representation is rich
enough. So we are talking both about Chris Torek's puzzle (I didn't
formulate it) *and* the IR level representation.
Sure, the type structure allows it. It always has. The problem is just that
the tree walking code can go into infinite loops or recursions at present
if the type dependence is "tight" enoug.