1 Copyright 2002, 2005 Free Software Foundation, Inc.
3 This file is part of the GNU MP Library.
5 The GNU MP Library is free software; you can redistribute it and/or modify it
6 under the terms of the GNU Lesser General Public License as published by the
7 Free Software Foundation; either version 3 of the License, or (at your
8 option) any later version.
10 The GNU MP Library is distributed in the hope that it will be useful, but
11 WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
12 FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License
15 You should have received a copy of the GNU Lesser General Public License along
16 with the GNU MP Library. If not, see http://www.gnu.org/licenses/.
22 This directory contains assembly code for nails-enabled 21264. The code is not
25 For addmul_N, as N grows larger, we could make multiple loads together, then do
26 about 3.3 i/c. 10 cycles after the last load, we can increase to 4 i/c. This
27 would surely allow addmul_4 to run at 2 c/l, but the same should be possible
28 also for addmul_3 and perhaps even addmul_2.
32 Routine c/l unroll c/l unroll c/l i/c
33 mul_1 3.25 2.75 2.75 3.273
34 addmul_1 4.0 4 3.5 4 14 3.25 3.385
35 addmul_2 4.0 1 2.5 2 10 2.25 3.333
36 addmul_3 3.0 1 2.33 2 14 2 3.333
37 addmul_4 2.5 1 2.125 2 17 2 3.135
43 (The "best" column doesn't account for bookkeeping instructions and
44 thereby assumes infinite unrolling.)
52 5 addmul_3 + addmul_2 2.3998