From 39897272516cd1620fd4dc7935e26abe17563ce8 Mon Sep 17 00:00:00 2001 From: lkcl Date: Sun, 24 Apr 2022 03:58:26 +0100 Subject: [PATCH] --- openpower/sv/biginteger/analysis.mdwn | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/openpower/sv/biginteger/analysis.mdwn b/openpower/sv/biginteger/analysis.mdwn index bda0ced55..bcaa0cd26 100644 --- a/openpower/sv/biginteger/analysis.mdwn +++ b/openpower/sv/biginteger/analysis.mdwn @@ -3,8 +3,8 @@ **DRAFT SVP64** * Revision 0.0: 21apr2022 -* Revision 0.1: 22apr2022 removal of msubed because sv.madded and sv.subfe works -* Revision 0.2: 22apr2022 128/64 scalar divide, investigate Goldschmidt +* Revision 0.01: 22apr2022 removal of msubed because sv.madded and sv.subfe works +* Revision 0.02: 22apr2022 128/64 scalar divide, investigate Goldschmidt This page covers an analysis of big integer operations, to work out optimal Scalar Instructions to propose be submitted to @@ -363,4 +363,5 @@ to 1-bit per loop, a simple compare, shift, and subtract). If the simplest approach were deployed then the completion time for the 128/64 scalar divide would be a whopping 128 cycles. To be workable an alternative algorithm is required, and one of the fastest appears to be -Goldschmidt Division. +Goldschmidt Division. Whilst typically deployed for Floating Point, +there is no reason why it should not be adapted to Fixed Point. -- 2.30.2