From 11da511c784eca003deb90c23570f0873954e0de Mon Sep 17 00:00:00 2001 From: Duncan Wilkie Date: Sat, 18 Nov 2023 06:11:09 -0600 Subject: Initial commit. --- gmp-6.3.0/mpn/x86_64/k8/gmp-mparam.h | 237 +++++++++++++++++++++++++++++++++++ 1 file changed, 237 insertions(+) create mode 100644 gmp-6.3.0/mpn/x86_64/k8/gmp-mparam.h (limited to 'gmp-6.3.0/mpn/x86_64/k8/gmp-mparam.h') diff --git a/gmp-6.3.0/mpn/x86_64/k8/gmp-mparam.h b/gmp-6.3.0/mpn/x86_64/k8/gmp-mparam.h new file mode 100644 index 0000000..d87cc3b --- /dev/null +++ b/gmp-6.3.0/mpn/x86_64/k8/gmp-mparam.h @@ -0,0 +1,237 @@ +/* AMD K8 gmp-mparam.h -- Compiler/machine parameter header file. + +Copyright 2019 Free Software Foundation, Inc. + +This file is part of the GNU MP Library. + +The GNU MP Library is free software; you can redistribute it and/or modify +it under the terms of either: + + * the GNU Lesser General Public License as published by the Free + Software Foundation; either version 3 of the License, or (at your + option) any later version. + +or + + * the GNU General Public License as published by the Free Software + Foundation; either version 2 of the License, or (at your option) any + later version. + +or both in parallel, as here. + +The GNU MP Library is distributed in the hope that it will be useful, but +WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY +or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License +for more details. + +You should have received copies of the GNU General Public License and the +GNU Lesser General Public License along with the GNU MP Library. If not, +see https://www.gnu.org/licenses/. */ + +#define GMP_LIMB_BITS 64 +#define GMP_LIMB_BYTES 8 + +#if 0 +#undef mpn_sublsh_n +#define mpn_sublsh_n(rp,up,vp,n,c) \ + (((rp) == (up)) ? mpn_submul_1 (rp, vp, n, CNST_LIMB(1) << (c)) \ + : MPN(mpn_sublsh_n)(rp,up,vp,n,c)) +#endif + +/* 2500 MHz K8 Brisbane */ +/* FFT tuning limit = 115,768,433 */ +/* Generated by tuneup.c, 2019-10-18, gcc 8.3 */ + +#define MOD_1_NORM_THRESHOLD 0 /* always */ +#define MOD_1_UNNORM_THRESHOLD 0 /* always */ +#define MOD_1N_TO_MOD_1_1_THRESHOLD 5 +#define MOD_1U_TO_MOD_1_1_THRESHOLD 2 +#define MOD_1_1_TO_MOD_1_2_THRESHOLD 14 +#define MOD_1_2_TO_MOD_1_4_THRESHOLD 35 +#define PREINV_MOD_1_TO_MOD_1_THRESHOLD 9 +#define USE_PREINV_DIVREM_1 1 /* native */ +#define DIV_QR_1_NORM_THRESHOLD 1 +#define DIV_QR_1_UNNORM_THRESHOLD MP_SIZE_T_MAX /* never */ +#define DIV_QR_2_PI2_THRESHOLD MP_SIZE_T_MAX /* never */ +#define DIVEXACT_1_THRESHOLD 0 /* always (native) */ +#define BMOD_1_TO_MOD_1_THRESHOLD 16 + +#define DIV_1_VS_MUL_1_PERCENT 309 + +#define MUL_TOOM22_THRESHOLD 28 +#define MUL_TOOM33_THRESHOLD 81 +#define MUL_TOOM44_THRESHOLD 232 +#define MUL_TOOM6H_THRESHOLD 324 +#define MUL_TOOM8H_THRESHOLD 478 + +#define MUL_TOOM32_TO_TOOM43_THRESHOLD 97 +#define MUL_TOOM32_TO_TOOM53_THRESHOLD 153 +#define MUL_TOOM42_TO_TOOM53_THRESHOLD 154 +#define MUL_TOOM42_TO_TOOM63_THRESHOLD 160 +#define MUL_TOOM43_TO_TOOM54_THRESHOLD 226 + +#define SQR_BASECASE_THRESHOLD 0 /* always (native) */ +#define SQR_TOOM2_THRESHOLD 34 +#define SQR_TOOM3_THRESHOLD 114 +#define SQR_TOOM4_THRESHOLD 336 +#define SQR_TOOM6_THRESHOLD 430 +#define SQR_TOOM8_THRESHOLD 0 /* always */ + +#define MULMID_TOOM42_THRESHOLD 36 + +#define MULMOD_BNM1_THRESHOLD 17 +#define SQRMOD_BNM1_THRESHOLD 19 + +#define MUL_FFT_MODF_THRESHOLD 654 /* k = 5 */ +#define MUL_FFT_TABLE3 \ + { { 654, 5}, { 21, 6}, { 11, 5}, { 23, 6}, \ + { 12, 5}, { 25, 6}, { 13, 5}, { 27, 6}, \ + { 27, 7}, { 14, 6}, { 29, 7}, { 15, 6}, \ + { 31, 7}, { 29, 8}, { 15, 7}, { 32, 8}, \ + { 17, 7}, { 37, 8}, { 19, 7}, { 39, 8}, \ + { 21, 7}, { 44, 8}, { 23, 7}, { 47, 8}, \ + { 25, 7}, { 51, 8}, { 31, 7}, { 63, 8}, \ + { 37, 9}, { 19, 8}, { 43, 9}, { 23, 8}, \ + { 53, 9}, { 27, 8}, { 57, 9}, { 31, 8}, \ + { 67, 9}, { 35, 8}, { 71, 9}, { 39, 8}, \ + { 81, 9}, { 43,10}, { 23, 9}, { 55, 8}, \ + { 111,10}, { 31, 9}, { 71,10}, { 39, 9}, \ + { 87,10}, { 47, 9}, { 99,10}, { 55, 9}, \ + { 111,11}, { 31,10}, { 63, 9}, { 131,10}, \ + { 71, 9}, { 147,10}, { 87,11}, { 47,10}, \ + { 111,11}, { 63,10}, { 143,11}, { 79,10}, \ + { 167,11}, { 95,10}, { 199,11}, { 111,12}, \ + { 63,11}, { 143,10}, { 287,11}, { 159,12}, \ + { 95,11}, { 191,10}, { 383,11}, { 207,10}, \ + { 415,13}, { 63,12}, { 127,11}, { 255,10}, \ + { 511,11}, { 271,10}, { 543,11}, { 287,12}, \ + { 159,11}, { 319,10}, { 639,11}, { 335,10}, \ + { 671,11}, { 351,12}, { 191,11}, { 415,12}, \ + { 223,11}, { 447,13}, { 127,12}, { 255,11}, \ + { 543,12}, { 287,11}, { 575,10}, { 1151,11}, \ + { 607,12}, { 319,11}, { 671,12}, { 351,11}, \ + { 703,13}, { 191,12}, { 383,11}, { 767,12}, \ + { 415,11}, { 831,12}, { 447,11}, { 895,12}, \ + { 479,14}, { 127,13}, { 255,12}, { 543,11}, \ + { 1087,12}, { 575,11}, { 1151,12}, { 607,13}, \ + { 319,12}, { 735,13}, { 383,12}, { 831,13}, \ + { 447,12}, { 959,14}, { 255,13}, { 511,12}, \ + { 1087,13}, { 575,12}, { 1215,13}, { 639,12}, \ + { 1279,13}, { 703,12}, { 1407,14}, { 383,13}, \ + { 767,12}, { 1535,13}, { 831,12}, { 1663,13}, \ + { 959,15}, { 255,14}, { 511,13}, { 1215,14}, \ + { 639,13}, { 1471,14}, { 767,13}, { 1663,14}, \ + { 895,13}, { 1855,15}, { 511,14}, { 1023,13}, \ + { 2047,14}, { 1151,13}, { 2367,14}, { 1407,15}, \ + { 767,14}, { 1791,16}, { 511,15}, { 1023,14}, \ + { 2303,15}, { 1279,14}, { 2687,15}, { 1535,14}, \ + { 3199,15}, { 1791,16}, { 1023,15}, { 2047,14}, \ + { 4223,15}, { 2303,14}, { 4735,15}, { 2559,16}, \ + { 1535,15}, { 3071,14}, { 6271,15}, { 3327,17}, \ + { 131072,18}, { 262144,19}, { 524288,20}, {1048576,21}, \ + {2097152,22}, {4194304,23}, {8388608,24} } +#define MUL_FFT_TABLE3_SIZE 183 +#define MUL_FFT_THRESHOLD 11520 + +#define SQR_FFT_MODF_THRESHOLD 540 /* k = 5 */ +#define SQR_FFT_TABLE3 \ + { { 540, 5}, { 21, 6}, { 11, 5}, { 23, 6}, \ + { 12, 5}, { 25, 6}, { 13, 5}, { 27, 6}, \ + { 16, 5}, { 33, 6}, { 29, 7}, { 15, 6}, \ + { 31, 7}, { 16, 6}, { 33, 7}, { 33, 8}, \ + { 17, 7}, { 37, 8}, { 19, 7}, { 39, 8}, \ + { 21, 7}, { 43, 8}, { 23, 7}, { 47, 8}, \ + { 25, 7}, { 51, 8}, { 29, 9}, { 15, 8}, \ + { 37, 9}, { 19, 8}, { 43, 9}, { 23, 8}, \ + { 51, 9}, { 27, 8}, { 55, 9}, { 31, 8}, \ + { 65, 9}, { 35, 8}, { 71, 9}, { 43,10}, \ + { 23, 9}, { 55,10}, { 31, 9}, { 71,10}, \ + { 39, 9}, { 83,10}, { 47, 9}, { 99,10}, \ + { 55, 9}, { 111,11}, { 31,10}, { 63, 9}, \ + { 127,10}, { 87,11}, { 47,10}, { 111,12}, \ + { 31,11}, { 63,10}, { 143,11}, { 79,10}, \ + { 167,11}, { 95,10}, { 191,11}, { 111,12}, \ + { 63,11}, { 127, 9}, { 511,11}, { 143,10}, \ + { 287, 9}, { 575,11}, { 159,12}, { 95,11}, \ + { 191,10}, { 383, 9}, { 767,11}, { 207,10}, \ + { 415,13}, { 63,12}, { 127,10}, { 511, 9}, \ + { 1023,11}, { 271,10}, { 543, 9}, { 1087,11}, \ + { 287,10}, { 575,12}, { 159,11}, { 319,10}, \ + { 639,11}, { 335,10}, { 671,11}, { 351,10}, \ + { 703,12}, { 191,11}, { 383,10}, { 767,11}, \ + { 415,10}, { 831,12}, { 223,11}, { 447,13}, \ + { 127,11}, { 511,10}, { 1023,11}, { 543,10}, \ + { 1087,12}, { 287,11}, { 575,10}, { 1151,11}, \ + { 607,12}, { 319,11}, { 639,10}, { 1279,11}, \ + { 671,12}, { 351,11}, { 703,13}, { 191,12}, \ + { 383,11}, { 767,12}, { 415,11}, { 831,12}, \ + { 447,11}, { 895,14}, { 127,12}, { 511,11}, \ + { 1023,12}, { 543,11}, { 1087,12}, { 575,11}, \ + { 1151,12}, { 607,11}, { 1215,13}, { 319,12}, \ + { 639,11}, { 1279,12}, { 671,11}, { 1343,12}, \ + { 703,11}, { 1407,12}, { 735,13}, { 383,12}, \ + { 767,11}, { 1535,12}, { 831,13}, { 447,12}, \ + { 959,13}, { 511,12}, { 1087,13}, { 575,12}, \ + { 1215,13}, { 639,12}, { 1343,13}, { 703,12}, \ + { 1407,14}, { 383,13}, { 767,12}, { 1535,13}, \ + { 831,12}, { 1663,13}, { 895,12}, { 1791,13}, \ + { 959,14}, { 511,13}, { 1215,14}, { 639,13}, \ + { 1471,14}, { 767,13}, { 1663,14}, { 895,13}, \ + { 1791,15}, { 511,14}, { 1023,13}, { 2111,14}, \ + { 1151,13}, { 2303,14}, { 1407,15}, { 767,14}, \ + { 1791,16}, { 511,15}, { 1023,14}, { 2303,15}, \ + { 1279,14}, { 2687,15}, { 1535,14}, { 3199,15}, \ + { 1791,16}, { 1023,15}, { 2047,14}, { 4223,15}, \ + { 2303,14}, { 4863,15}, { 2559,16}, { 1535,15}, \ + { 3071,14}, { 6271,15}, { 3327,17}, { 131072,18}, \ + { 262144,19}, { 524288,20}, {1048576,21}, {2097152,22}, \ + {4194304,23}, {8388608,24} } +#define SQR_FFT_TABLE3_SIZE 202 +#define SQR_FFT_THRESHOLD 7296 + +#define MULLO_BASECASE_THRESHOLD 0 /* always */ +#define MULLO_DC_THRESHOLD 61 +#define MULLO_MUL_N_THRESHOLD 22239 +#define SQRLO_BASECASE_THRESHOLD 8 +#define SQRLO_DC_THRESHOLD 0 /* never mpn_sqrlo_basecase */ +#define SQRLO_SQR_THRESHOLD 14281 + +#define DC_DIV_QR_THRESHOLD 47 +#define DC_DIVAPPR_Q_THRESHOLD 266 +#define DC_BDIV_QR_THRESHOLD 38 +#define DC_BDIV_Q_THRESHOLD 104 + +#define INV_MULMOD_BNM1_THRESHOLD 54 +#define INV_NEWTON_THRESHOLD 252 +#define INV_APPR_THRESHOLD 250 + +#define BINV_NEWTON_THRESHOLD 258 +#define REDC_1_TO_REDC_2_THRESHOLD 35 +#define REDC_2_TO_REDC_N_THRESHOLD 79 + +#define MU_DIV_QR_THRESHOLD 2089 +#define MU_DIVAPPR_Q_THRESHOLD 1895 +#define MUPI_DIV_QR_THRESHOLD 99 +#define MU_BDIV_QR_THRESHOLD 1787 +#define MU_BDIV_Q_THRESHOLD 1895 + +#define POWM_SEC_TABLE 1,16,194,960,2825 + +#define GET_STR_DC_THRESHOLD 16 +#define GET_STR_PRECOMPUTE_THRESHOLD 26 +#define SET_STR_DC_THRESHOLD 248 +#define SET_STR_PRECOMPUTE_THRESHOLD 1747 + +#define FAC_DSC_THRESHOLD 1240 +#define FAC_ODD_THRESHOLD 27 + +#define MATRIX22_STRASSEN_THRESHOLD 21 +#define HGCD2_DIV1_METHOD 3 /* 4.10% faster than 5 */ +#define HGCD_THRESHOLD 141 +#define HGCD_APPR_THRESHOLD 181 +#define HGCD_REDUCE_THRESHOLD 4633 +#define GCD_DC_THRESHOLD 622 +#define GCDEXT_DC_THRESHOLD 496 +#define JACOBI_BASE_METHOD 1 /* 0.97% faster than 3 */ + +/* Tuneup completed successfully, took 131832 seconds */ -- cgit v1.2.3