From a89a14ef5da44684a16b204e7a70460cc8c4922a Mon Sep 17 00:00:00 2001 From: Thomas Voss Date: Fri, 21 Jun 2024 23:36:36 +0200 Subject: Basic constant folding implementation --- vendor/gmp-6.3.0/mpn/x86_64/zen2/gmp-mparam.h | 276 ++++++++++++++++++++++++++ 1 file changed, 276 insertions(+) create mode 100644 vendor/gmp-6.3.0/mpn/x86_64/zen2/gmp-mparam.h (limited to 'vendor/gmp-6.3.0/mpn/x86_64/zen2') diff --git a/vendor/gmp-6.3.0/mpn/x86_64/zen2/gmp-mparam.h b/vendor/gmp-6.3.0/mpn/x86_64/zen2/gmp-mparam.h new file mode 100644 index 0000000..3748c5f --- /dev/null +++ b/vendor/gmp-6.3.0/mpn/x86_64/zen2/gmp-mparam.h @@ -0,0 +1,276 @@ +/* AMD Zen2 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 + +/* Disable use of slow functions. FIXME: We should disable lib inclusion. */ +#undef HAVE_NATIVE_mpn_mul_2 +#undef HAVE_NATIVE_mpn_addmul_2 + +/* 3600-4400 MHz Matisse */ +/* FFT tuning limit = 703,392,483 */ +/* Generated by tuneup.c, 2019-10-19, 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 4 +#define MOD_1U_TO_MOD_1_1_THRESHOLD 3 +#define MOD_1_1_TO_MOD_1_2_THRESHOLD 8 +#define MOD_1_2_TO_MOD_1_4_THRESHOLD 27 +#define PREINV_MOD_1_TO_MOD_1_THRESHOLD 1 +#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 13 +#define DIVEXACT_1_THRESHOLD 0 /* always (native) */ +#define BMOD_1_TO_MOD_1_THRESHOLD 22 + +#define DIV_1_VS_MUL_1_PERCENT 385 + +#define MUL_TOOM22_THRESHOLD 19 +#define MUL_TOOM33_THRESHOLD 125 +#define MUL_TOOM44_THRESHOLD 196 +#define MUL_TOOM6H_THRESHOLD 276 +#define MUL_TOOM8H_THRESHOLD 369 + +#define MUL_TOOM32_TO_TOOM43_THRESHOLD 121 +#define MUL_TOOM32_TO_TOOM53_THRESHOLD 138 +#define MUL_TOOM42_TO_TOOM53_THRESHOLD 129 +#define MUL_TOOM42_TO_TOOM63_THRESHOLD 132 +#define MUL_TOOM43_TO_TOOM54_THRESHOLD 185 + +#define SQR_BASECASE_THRESHOLD 0 /* always (native) */ +#define SQR_TOOM2_THRESHOLD 30 +#define SQR_TOOM3_THRESHOLD 117 +#define SQR_TOOM4_THRESHOLD 315 +#define SQR_TOOM6_THRESHOLD 446 +#define SQR_TOOM8_THRESHOLD 527 + +#define MULMID_TOOM42_THRESHOLD 38 + +#define MULMOD_BNM1_THRESHOLD 14 +#define SQRMOD_BNM1_THRESHOLD 20 + +#define MUL_FFT_MODF_THRESHOLD 436 /* k = 5 */ +#define MUL_FFT_TABLE3 \ + { { 436, 5}, { 25, 6}, { 25, 7}, { 13, 6}, \ + { 27, 7}, { 15, 6}, { 31, 7}, { 25, 8}, \ + { 13, 7}, { 28, 8}, { 15, 7}, { 32, 8}, \ + { 17, 7}, { 35, 8}, { 19, 7}, { 39, 8}, \ + { 27, 9}, { 15, 8}, { 35, 9}, { 19, 8}, \ + { 41, 9}, { 23, 8}, { 49, 9}, { 27,10}, \ + { 15, 9}, { 31, 8}, { 63, 9}, { 39,10}, \ + { 23, 9}, { 51,11}, { 15,10}, { 31, 9}, \ + { 71,10}, { 39, 9}, { 83,10}, { 47, 9}, \ + { 95,10}, { 55,11}, { 31,10}, { 79,11}, \ + { 47,10}, { 95,12}, { 31,11}, { 63,10}, \ + { 135,11}, { 79,10}, { 159,11}, { 95,10}, \ + { 191,12}, { 63,11}, { 127,10}, { 255, 9}, \ + { 511,11}, { 143,10}, { 287, 9}, { 575,11}, \ + { 159,12}, { 95,11}, { 191,13}, { 63,12}, \ + { 127,11}, { 255,10}, { 511,11}, { 271,10}, \ + { 543, 9}, { 1087,11}, { 287,10}, { 575,12}, \ + { 159,11}, { 319,10}, { 639,11}, { 335,10}, \ + { 671,11}, { 351,10}, { 703,11}, { 367,12}, \ + { 191,11}, { 383,10}, { 767,11}, { 415,10}, \ + { 831,12}, { 223,11}, { 447,13}, { 127,12}, \ + { 255,11}, { 543,10}, { 1087,12}, { 287,11}, \ + { 575,10}, { 1151,11}, { 607,10}, { 1215,12}, \ + { 319,11}, { 639,10}, { 1279,11}, { 671,10}, \ + { 1343,12}, { 351,11}, { 703,10}, { 1407,11}, \ + { 735,13}, { 191,12}, { 383,11}, { 767,10}, \ + { 1535,11}, { 799,12}, { 415,11}, { 831,10}, \ + { 1663,12}, { 447,11}, { 895,12}, { 479,14}, \ + { 127,13}, { 255,12}, { 543,11}, { 1087,10}, \ + { 2175,12}, { 575,11}, { 1151,12}, { 607,11}, \ + { 1215,10}, { 2431,13}, { 319,12}, { 639,11}, \ + { 1279,12}, { 671,11}, { 1343,10}, { 2687,12}, \ + { 703,11}, { 1471,10}, { 2943,13}, { 383,12}, \ + { 767,11}, { 1535,12}, { 799,11}, { 1599,12}, \ + { 831,11}, { 1663,13}, { 447,12}, { 959,11}, \ + { 1919,10}, { 3839,14}, { 255,13}, { 511,12}, \ + { 1087,11}, { 2175,13}, { 575,12}, { 1215,11}, \ + { 2431,13}, { 639,12}, { 1343,11}, { 2687,13}, \ + { 703,12}, { 1471,11}, { 2943,14}, { 383,13}, \ + { 767,12}, { 1599,11}, { 3199,13}, { 831,12}, \ + { 1727,13}, { 895,12}, { 1791,13}, { 959,12}, \ + { 1919,11}, { 3839,14}, { 511,13}, { 1087,12}, \ + { 2175,13}, { 1215,12}, { 2431,14}, { 639,13}, \ + { 1343,12}, { 2687,13}, { 1471,12}, { 2943,11}, \ + { 5887,14}, { 767,13}, { 1599,12}, { 3199,13}, \ + { 1727,12}, { 3455,14}, { 895,13}, { 1919,12}, \ + { 3839,11}, { 7679,15}, { 511,14}, { 1023,13}, \ + { 2175,14}, { 1151,13}, { 2431,12}, { 4863,14}, \ + { 1279,13}, { 2687,14}, { 1407,13}, { 2943,12}, \ + { 5887,15}, { 767,14}, { 1535,13}, { 3199,14}, \ + { 1663,13}, { 3455,12}, { 6911,14}, { 1919,13}, \ + { 3839,16}, { 511,15}, { 1023,14}, { 2175,13}, \ + { 4479,12}, { 8959,14}, { 2431,13}, { 4863,15}, \ + { 1279,14}, { 2943,13}, { 5887,12}, { 11775,15}, \ + { 1535,14}, { 3455,15}, { 1791,14}, { 3839,13}, \ + { 7679,14}, { 3967,16}, { 1023,15}, { 2047,14}, \ + { 4479,15}, { 2303,14}, { 4863,15}, { 2559,14}, \ + { 5247,15}, { 2815,14}, { 5887,16}, { 1535,15}, \ + { 3327,14}, { 6911,15}, { 3839,14}, { 7679,13}, \ + { 15359,17}, { 1023,16}, { 2047,15}, { 4351,14}, \ + { 8959,15}, { 4863,16}, { 2559,15}, { 5887,14}, \ + { 11775,16}, { 3071,15}, { 6911,16}, { 3583,15}, \ + { 7679,14}, { 15359,15}, { 7935,17}, { 2047,16}, \ + { 4095,15}, { 8959,16}, { 4607,15}, { 9983,14}, \ + { 19967,16}, { 5631,15}, { 11775,17}, { 3071,16}, \ + { 7679,15}, { 15871,18}, { 2047,17}, { 4095,16}, \ + { 9727,15}, { 19967,17}, { 5119,16}, { 65536,17}, \ + { 131072,18}, { 262144,19}, { 524288,20}, {1048576,21}, \ + {2097152,22}, {4194304,23}, {8388608,24} } +#define MUL_FFT_TABLE3_SIZE 275 +#define MUL_FFT_THRESHOLD 4736 + +#define SQR_FFT_MODF_THRESHOLD 396 /* k = 5 */ +#define SQR_FFT_TABLE3 \ + { { 396, 5}, { 25, 6}, { 25, 7}, { 13, 6}, \ + { 27, 7}, { 15, 6}, { 31, 7}, { 25, 8}, \ + { 13, 7}, { 28, 8}, { 15, 7}, { 32, 8}, \ + { 17, 7}, { 35, 8}, { 19, 7}, { 39, 8}, \ + { 27, 9}, { 15, 8}, { 35, 9}, { 19, 8}, \ + { 41, 9}, { 23, 8}, { 47, 9}, { 27,10}, \ + { 15, 9}, { 39,10}, { 23, 9}, { 51,11}, \ + { 15,10}, { 31, 9}, { 67,10}, { 39, 9}, \ + { 79,10}, { 55,11}, { 31,10}, { 79,11}, \ + { 47,10}, { 95,12}, { 31,11}, { 63,10}, \ + { 135,11}, { 79,10}, { 159,11}, { 95,12}, \ + { 63,11}, { 127,10}, { 255, 9}, { 511,11}, \ + { 143,10}, { 287, 9}, { 575,10}, { 303,11}, \ + { 159,12}, { 95,13}, { 63,12}, { 127,11}, \ + { 255,10}, { 511,11}, { 271,10}, { 543,11}, \ + { 287,10}, { 575,11}, { 303,12}, { 159,11}, \ + { 319,10}, { 639,11}, { 335,10}, { 671, 9}, \ + { 1343,11}, { 351,10}, { 703,11}, { 367,10}, \ + { 735,11}, { 383,10}, { 767,11}, { 415,10}, \ + { 831,12}, { 223,11}, { 447,13}, { 127,12}, \ + { 255,11}, { 511,10}, { 1023,11}, { 543,10}, \ + { 1087,12}, { 287,11}, { 575,10}, { 1151,11}, \ + { 607,10}, { 1215,12}, { 319,11}, { 639,10}, \ + { 1279,11}, { 671,10}, { 1343,12}, { 351,11}, \ + { 703,10}, { 1407,11}, { 735,10}, { 1471,12}, \ + { 383,11}, { 767,10}, { 1535,11}, { 799,12}, \ + { 415,11}, { 831,10}, { 1663,12}, { 447,11}, \ + { 895,12}, { 479,11}, { 959,14}, { 127,12}, \ + { 511,11}, { 1023,12}, { 543,11}, { 1087,10}, \ + { 2175,12}, { 575,11}, { 1151,12}, { 607,11}, \ + { 1215,10}, { 2431,12}, { 639,11}, { 1279,12}, \ + { 671,11}, { 1343,10}, { 2687,12}, { 703,11}, \ + { 1407,12}, { 735,11}, { 1471,10}, { 2943,13}, \ + { 383,12}, { 767,11}, { 1535,12}, { 799,11}, \ + { 1599,12}, { 831,11}, { 1663,13}, { 447,12}, \ + { 959,11}, { 1919,10}, { 3839,13}, { 511,12}, \ + { 1087,11}, { 2175,13}, { 575,12}, { 1215,11}, \ + { 2431,13}, { 639,12}, { 1343,11}, { 2687,13}, \ + { 703,12}, { 1471,11}, { 2943,14}, { 383,13}, \ + { 767,12}, { 1599,13}, { 831,12}, { 1727,11}, \ + { 3455,13}, { 959,12}, { 1919,11}, { 3839,14}, \ + { 511,13}, { 1023,12}, { 2047,13}, { 1087,12}, \ + { 2175,13}, { 1215,12}, { 2431,11}, { 4863,14}, \ + { 639,13}, { 1343,12}, { 2687,13}, { 1471,12}, \ + { 2943,11}, { 5887,14}, { 767,13}, { 1599,12}, \ + { 3199,13}, { 1727,12}, { 3455,14}, { 895,13}, \ + { 1919,12}, { 3839,15}, { 511,14}, { 1023,13}, \ + { 2175,14}, { 1151,13}, { 2431,12}, { 4863,14}, \ + { 1279,13}, { 2687,14}, { 1407,13}, { 2943,12}, \ + { 5887,15}, { 767,14}, { 1535,13}, { 3199,14}, \ + { 1663,13}, { 3455,12}, { 6911,14}, { 1919,13}, \ + { 3839,12}, { 7679,16}, { 511,15}, { 1023,14}, \ + { 2175,13}, { 4479,14}, { 2431,13}, { 4863,15}, \ + { 1279,14}, { 2943,13}, { 5887,12}, { 11775,15}, \ + { 1535,14}, { 3455,13}, { 6911,15}, { 1791,14}, \ + { 3839,13}, { 7679,14}, { 3967,16}, { 1023,15}, \ + { 2047,14}, { 4479,15}, { 2303,14}, { 4863,15}, \ + { 2559,14}, { 5247,15}, { 2815,14}, { 5887,13}, \ + { 11775,16}, { 1535,15}, { 3071,14}, { 6143,15}, \ + { 3327,14}, { 6911,15}, { 3839,14}, { 7679,17}, \ + { 1023,16}, { 2047,15}, { 4095,14}, { 8191,15}, \ + { 4351,14}, { 8959,15}, { 4863,16}, { 2559,15}, \ + { 5887,14}, { 11775,16}, { 3071,15}, { 6911,16}, \ + { 3583,15}, { 7679,14}, { 15359,15}, { 7935,17}, \ + { 2047,16}, { 4095,15}, { 8959,16}, { 4607,15}, \ + { 9983,14}, { 19967,16}, { 5119,15}, { 10239,16}, \ + { 5631,15}, { 11775,17}, { 3071,16}, { 7679,15}, \ + { 15359,18}, { 2047,17}, { 4095,16}, { 9727,15}, \ + { 19967,17}, { 5119,16}, { 65536,17}, { 131072,18}, \ + { 262144,19}, { 524288,20}, {1048576,21}, {2097152,22}, \ + {4194304,23}, {8388608,24} } +#define SQR_FFT_TABLE3_SIZE 282 +#define SQR_FFT_THRESHOLD 3264 + +#define MULLO_BASECASE_THRESHOLD 0 /* always */ +#define MULLO_DC_THRESHOLD 57 +#define MULLO_MUL_N_THRESHOLD 8907 +#define SQRLO_BASECASE_THRESHOLD 8 +#define SQRLO_DC_THRESHOLD 0 /* never mpn_sqrlo_basecase */ +#define SQRLO_SQR_THRESHOLD 6440 + +#define DC_DIV_QR_THRESHOLD 43 +#define DC_DIVAPPR_Q_THRESHOLD 154 +#define DC_BDIV_QR_THRESHOLD 46 +#define DC_BDIV_Q_THRESHOLD 93 + +#define INV_MULMOD_BNM1_THRESHOLD 36 +#define INV_NEWTON_THRESHOLD 141 +#define INV_APPR_THRESHOLD 149 + +#define BINV_NEWTON_THRESHOLD 264 +#define REDC_1_TO_REDC_N_THRESHOLD 47 + +#define MU_DIV_QR_THRESHOLD 1470 +#define MU_DIVAPPR_Q_THRESHOLD 1528 +#define MUPI_DIV_QR_THRESHOLD 47 +#define MU_BDIV_QR_THRESHOLD 1187 +#define MU_BDIV_Q_THRESHOLD 1589 + +#define POWM_SEC_TABLE 3,22,194,579 + +#define GET_STR_DC_THRESHOLD 12 +#define GET_STR_PRECOMPUTE_THRESHOLD 19 +#define SET_STR_DC_THRESHOLD 195 +#define SET_STR_PRECOMPUTE_THRESHOLD 1752 + +#define FAC_DSC_THRESHOLD 345 +#define FAC_ODD_THRESHOLD 0 /* always */ + +#define MATRIX22_STRASSEN_THRESHOLD 24 +#define HGCD2_DIV1_METHOD 1 /* 11.29% faster than 3 */ +#define HGCD_THRESHOLD 89 +#define HGCD_APPR_THRESHOLD 96 +#define HGCD_REDUCE_THRESHOLD 2681 +#define GCD_DC_THRESHOLD 465 +#define GCDEXT_DC_THRESHOLD 233 +#define JACOBI_BASE_METHOD 1 /* 25.56% faster than 4 */ + +/* Tuneup completed successfully, took 294200 seconds */ -- cgit v1.2.3