This is the mail archive of the gcc-patches@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: [PATCH] Fix PR27529, a step towards fixing PR27039

Ian Lance Taylor writes:
 > Andrew Haley <aph@redhat.com> writes:
 > 
 > > Me either.  However, there is some remarkably similar code in
 > > fold_cond_expr_with_comparison().
 > 
 > That code is also bogus in the current regime and should be
 > represented using different tree codes or using some sort of flag on
 > the tree.  (LVALUE_EXPR, maybe?)

I'm not disputing that, really.  I know my patch is a kludge.

 > It does seem like a new tree code is the right approach.

I think so, yes.  I think it's useful semantics to have a type
conversion that can be folded in only limited ways, such as

  (type1)(type1) -> (type1)

 > For the moment, if it helps, I'll approve your patch to go onto the
 > 4.2 branch, when it is created.  The patch shouldn't go into
 > mainline.

OK.

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