On Fri, Nov 18, 2005 at 08:12:32PM -0500, Richard Kenner wrote:
> What about the other cases where this would apply? I can't think of any
> cases where checking the lang hook directly is correct. Can you?
Deciding if one can replace one object with another, or one
pointer with another in an actu
> if (!is_gimple_reg_type (TREE_TYPE (param))
> ! || !tree_ssa_useless_type_conversion_1 (TREE_TYPE
(param),
> ! TREE_TYPE
(arg)))
Seems reasonable. These are the conversions we strip acr
On Fri, Nov 18, 2005 at 08:42:43AM -0500, Richard Kenner wrote:
> if (!is_gimple_reg_type (TREE_TYPE (param))
> ! || !tree_ssa_useless_type_conversion_1 (TREE_TYPE (param),
> ! TREE_TYPE (arg)))
Seems reasonable. Thes
Be careful, tree_ssa_useless_type_conversion_1 strips CV qualifiers
from pointers, f.i. See various different cleanup-patches I posted
long time ago.
Right, but the point is that whatever it does, at least that code (and
perhaps other current callers of the lang hook) have to duplicat
On 11/18/05, Richard Kenner <[EMAIL PROTECTED]> wrote:
> I sent email about this a few months ago (I can't find it since I'm having
> a problem getting a browser to work on gcc.gnu.org) and thought I'd raise
> it again since it would be good to get this into 4.1
>
> Currently, tail call detected is
