This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: BINFO_OFFSET
- To: mark at codesourcery dot com
- Subject: Re: BINFO_OFFSET
- From: kenner at vlsi1 dot ultra dot nyu dot edu (Richard Kenner)
- Date: Sun, 20 Feb 00 15:00:13 EST
- Cc: gcc at gcc dot gnu dot org
In the C++ front-end, BINFO_OFFSET always has type sizetype, to the
best of my knowledge. However, we often need to subtract one
BINFO_OFFSET from another, and that often yields negative values. (In
other words, we ask the question: how far is it from this sub-object
to this other sub-object?)
That sounds like a common enough idiom to justify a size_diffop, which
takes two operands that are either both sizetype or both bitsize type
and returns an object of ssizetype or sbitsizetype.