From: lkcl <lkcl@web>
Date: Thu, 21 Apr 2022 07:23:46 +0000 (+0100)
Subject: (no commit message)
X-Git-Tag: opf_rfc_ls005_v1~2664
X-Git-Url: https://git.libre-soc.org/?a=commitdiff_plain;h=dcc555fd6c1d23b154681e30061fa840ae19aa65;p=libreriscv.git

---

diff --git a/openpower/sv/biginteger.mdwn b/openpower/sv/biginteger.mdwn
index ee2694751..bfd566cfd 100644
--- a/openpower/sv/biginteger.mdwn
+++ b/openpower/sv/biginteger.mdwn
@@ -41,6 +41,15 @@ ALU.
 
 ## Multiply
 
+Long-multiply, assuming an O(N^2) algorithm, is performed by summing
+NxN separate smaller multiplications together.  Karatsuba's algorithm
+reduces the number of small multiplies at the expense of increasing
+the number of additions.  Some algorithms follow the Vedic Multiply
+pattern by grouping together all multiplies of the same magnitude/power
+(same column) whilst others perform row-based multiplication: a single
+digit of B multiplies the entirety of A, summed a row at a time. This
+algorithm is the basis of the analysis below (Knuth's Algorithm M).
+
 Multiply is tricky: 64 bit operands actually produce a 128-bit result,
 which clearly cannot fit into an orthogonal register file.
 Most Scalar RISC ISAs have separate `mul-low-half` and `mul-hi-half`