1Copyright 1996, 2001 Free Software Foundation, Inc. 2 3This file is part of the GNU MP Library. 4 5The GNU MP Library is free software; you can redistribute it and/or modify 6it under the terms of either: 7 8 * the GNU Lesser General Public License as published by the Free 9 Software Foundation; either version 3 of the License, or (at your 10 option) any later version. 11 12or 13 14 * the GNU General Public License as published by the Free Software 15 Foundation; either version 2 of the License, or (at your option) any 16 later version. 17 18or both in parallel, as here. 19 20The GNU MP Library is distributed in the hope that it will be useful, but 21WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY 22or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License 23for more details. 24 25You should have received copies of the GNU General Public License and the 26GNU Lesser General Public License along with the GNU MP Library. If not, 27see https://www.gnu.org/licenses/. 28 29 30 31 32 33This directory contains mpn functions for various SPARC chips. Code that 34runs only on version 8 SPARC implementations, is in the v8 subdirectory. 35 36RELEVANT OPTIMIZATION ISSUES 37 38 Load and Store timing 39 40On most early SPARC implementations, the ST instructions takes multiple 41cycles, while a STD takes just a single cycle more than an ST. For the CPUs 42in SPARCstation I and II, the times are 3 and 4 cycles, respectively. 43Therefore, combining two ST instructions into a STD when possible is a 44significant optimization. 45 46Later SPARC implementations have single cycle ST. 47 48For SuperSPARC, we can perform just one memory instruction per cycle, even 49if up to two integer instructions can be executed in its pipeline. For 50programs that perform so many memory operations that there are not enough 51non-memory operations to issue in parallel with all memory operations, using 52LDD and STD when possible helps. 53 54UltraSPARC-1/2 has very slow integer multiplication. In the v9 subdirectory, 55we therefore use floating-point multiplication. 56 57STATUS 58 591. On a SuperSPARC, mpn_lshift and mpn_rshift run at 3 cycles/limb, or 2.5 60 cycles/limb asymptotically. We could optimize speed for special counts 61 by using ADDXCC. 62 632. On a SuperSPARC, mpn_add_n and mpn_sub_n runs at 2.5 cycles/limb, or 2 64 cycles/limb asymptotically. 65 663. mpn_mul_1 runs at what is believed to be optimal speed. 67 684. On SuperSPARC, mpn_addmul_1 and mpn_submul_1 could both be improved by a 69 cycle by avoiding one of the add instructions. See a29k/addmul_1. 70 71The speed of the code for other SPARC implementations is uncertain. 72