From: Maurizio Vitale (mav_at_[hidden])
Date: 2007-05-03 22:16:39
"Phil Endecott" <spam_from_boost_dev_at_[hidden]> writes:
> If I want to do 32x32->64, I need to break up the two operands into two
> 16-bit portions and do this calculation:
I just stumbled on this, which migh be of interest to you,
http://www.east.isi.edu/~jsuh/publ/sci01-jsuh-rev.pdf
I haven't read it in detail yet.
Regards,
Maurizio