[Python-Dev] Performance (non)optimization: 31-bit ints in pointers
Tue, 13 Aug 2002 06:50:59 -0500
On Mon, Aug 12, 2002 at 08:44:46PM -0400, Tim Peters wrote:
> [Jeff Epler]
> > (due to alignment requirements on all common machines, all valid
> > pointers-to-struct have 0 in their low bit)
> Not so on word-addressed machines, though, or on machines using low-order
> pointer bits for their own notion of tag bits.
Of course, "all common machines" simply means "x86 machines".