diff options
author | Torbjorn Granlund <tg@gmplib.org> | 2017-02-23 14:14:44 +0100 |
---|---|---|
committer | Torbjorn Granlund <tg@gmplib.org> | 2017-02-23 14:14:44 +0100 |
commit | 60a9637ee7c120dca79173432aa6c26ec659dd46 (patch) | |
tree | 88d7168933e180772303b5bebc552f8780a193c4 /mpn/ia64 | |
parent | 752510f977c2e4292ddb426f112410b27a162d70 (diff) | |
download | gmp-60a9637ee7c120dca79173432aa6c26ec659dd46.tar.gz |
Retune.
Diffstat (limited to 'mpn/ia64')
-rw-r--r-- | mpn/ia64/gmp-mparam.h | 225 |
1 files changed, 89 insertions, 136 deletions
diff --git a/mpn/ia64/gmp-mparam.h b/mpn/ia64/gmp-mparam.h index feb517ee7..c7b0f4f00 100644 --- a/mpn/ia64/gmp-mparam.h +++ b/mpn/ia64/gmp-mparam.h @@ -1,6 +1,6 @@ /* gmp-mparam.h -- Compiler/machine parameter header file. -Copyright 2000-2005, 2009-2011, 2014, 2015 Free Software Foundation, Inc. +Copyright 2017 Free Software Foundation, Inc. This file is part of the GNU MP Library. @@ -31,18 +31,18 @@ see https://www.gnu.org/licenses/. */ #define GMP_LIMB_BITS 64 #define GMP_LIMB_BYTES 8 -/* 900MHz Itanium2 (titanic.gmplib.org) */ -/* FFT tuning limit = 60 M */ -/* Generated by tuneup.c, 2015-10-11, gcc 4.4 */ +/* 900MHz Itanium2 (olympic.gmplib.org) */ +/* FFT tuning limit = 0.5 M */ +/* Generated by tuneup.c, 2017-02-19, gcc 4.2 */ #define MOD_1_1P_METHOD 2 #define MOD_1_NORM_THRESHOLD 0 /* always */ #define MOD_1_UNNORM_THRESHOLD 0 /* always */ -#define MOD_1N_TO_MOD_1_1_THRESHOLD 3 -#define MOD_1U_TO_MOD_1_1_THRESHOLD 5 -#define MOD_1_1_TO_MOD_1_2_THRESHOLD 26 +#define MOD_1N_TO_MOD_1_1_THRESHOLD 8 +#define MOD_1U_TO_MOD_1_1_THRESHOLD 6 +#define MOD_1_1_TO_MOD_1_2_THRESHOLD 18 #define MOD_1_2_TO_MOD_1_4_THRESHOLD 0 /* never mpn_mod_1s_2p */ -#define PREINV_MOD_1_TO_MOD_1_THRESHOLD 10 +#define PREINV_MOD_1_TO_MOD_1_THRESHOLD 13 #define USE_PREINV_DIVREM_1 1 /* native */ #define DIV_QR_1N_PI1_METHOD 1 #define DIV_QR_1_NORM_THRESHOLD MP_SIZE_T_MAX /* never */ @@ -51,156 +51,109 @@ see https://www.gnu.org/licenses/. */ #define DIVEXACT_1_THRESHOLD 0 /* always (native) */ #define BMOD_1_TO_MOD_1_THRESHOLD MP_SIZE_T_MAX /* never */ -#define DIV_1_VS_MUL_1_PERCENT 315 +#define DIV_1_VS_MUL_1_PERCENT 316 -#define MUL_TOOM22_THRESHOLD 39 -#define MUL_TOOM33_THRESHOLD 129 -#define MUL_TOOM44_THRESHOLD 214 -#define MUL_TOOM6H_THRESHOLD 318 -#define MUL_TOOM8H_THRESHOLD 430 +#define MUL_TOOM22_THRESHOLD 48 +#define MUL_TOOM33_THRESHOLD 90 +#define MUL_TOOM44_THRESHOLD 216 +#define MUL_TOOM6H_THRESHOLD 327 +#define MUL_TOOM8H_THRESHOLD 454 -#define MUL_TOOM32_TO_TOOM43_THRESHOLD 121 -#define MUL_TOOM32_TO_TOOM53_THRESHOLD 145 -#define MUL_TOOM42_TO_TOOM53_THRESHOLD 124 -#define MUL_TOOM42_TO_TOOM63_THRESHOLD 151 -#define MUL_TOOM43_TO_TOOM54_THRESHOLD 201 +#define MUL_TOOM32_TO_TOOM43_THRESHOLD 97 +#define MUL_TOOM32_TO_TOOM53_THRESHOLD 153 +#define MUL_TOOM42_TO_TOOM53_THRESHOLD 148 +#define MUL_TOOM42_TO_TOOM63_THRESHOLD 153 +#define MUL_TOOM43_TO_TOOM54_THRESHOLD 226 #define SQR_BASECASE_THRESHOLD 11 -#define SQR_TOOM2_THRESHOLD 85 -#define SQR_TOOM3_THRESHOLD 131 -#define SQR_TOOM4_THRESHOLD 494 -#define SQR_TOOM6_THRESHOLD 0 /* always */ -#define SQR_TOOM8_THRESHOLD 0 /* always */ +#define SQR_TOOM2_THRESHOLD 98 +#define SQR_TOOM3_THRESHOLD 140 +#define SQR_TOOM4_THRESHOLD 260 +#define SQR_TOOM6_THRESHOLD 354 +#define SQR_TOOM8_THRESHOLD 502 -#define MULMID_TOOM42_THRESHOLD 98 +#define MULMID_TOOM42_THRESHOLD 99 -#define MULMOD_BNM1_THRESHOLD 23 -#define SQRMOD_BNM1_THRESHOLD 25 +#define MULMOD_BNM1_THRESHOLD 25 +#define SQRMOD_BNM1_THRESHOLD 33 -#define MUL_FFT_MODF_THRESHOLD 500 /* k = 5 */ +#define MUL_FFT_MODF_THRESHOLD 840 /* k = 5 */ #define MUL_FFT_TABLE3 \ - { { 500, 5}, { 29, 6}, { 35, 7}, { 18, 6}, \ - { 37, 7}, { 19, 6}, { 39, 7}, { 33, 8}, \ - { 17, 7}, { 37, 8}, { 19, 7}, { 39, 8}, \ - { 21, 7}, { 43, 8}, { 37, 9}, { 19, 8}, \ + { { 840, 5}, { 31, 6}, { 16, 5}, { 33, 6}, \ + { 17, 5}, { 35, 6}, { 37, 7}, { 19, 6}, \ + { 43, 7}, { 44, 8}, { 23, 7}, { 47, 8}, \ + { 43, 9}, { 23, 8}, { 51, 9}, { 27, 8}, \ + { 57, 9}, { 31, 8}, { 65, 9}, { 35, 8}, \ + { 71, 9}, { 43,10}, { 23, 9}, { 59,10}, \ + { 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}, { 111,11}, { 63,10}, { 143,11}, \ + { 79,10}, { 167,11}, { 95,10}, { 191,11}, \ + { 111,12}, { 63,11}, { 175,12}, { 95,11}, \ + { 207,13}, { 8192,14}, { 16384,15}, { 32768,16}, \ + { 65536,17}, { 131072,18}, { 262144,19}, { 524288,20}, \ + {1048576,21}, {2097152,22}, {4194304,23}, {8388608,24} } +#define MUL_FFT_TABLE3_SIZE 60 +#define MUL_FFT_THRESHOLD 8832 + +#define SQR_FFT_MODF_THRESHOLD 765 /* k = 5 */ +#define SQR_FFT_TABLE3 \ + { { 765, 5}, { 35, 6}, { 43, 7}, { 43, 8}, \ { 43, 9}, { 23, 8}, { 49, 9}, { 27, 8}, \ - { 57, 9}, { 43,10}, { 23, 9}, { 59,10}, \ - { 31, 9}, { 71,10}, { 39, 9}, { 87,10}, \ - { 47, 9}, { 99,10}, { 55,11}, { 31,10}, \ - { 87,11}, { 47,10}, { 111,12}, { 31,11}, \ - { 63,10}, { 143,11}, { 79,10}, { 167,11}, \ - { 95,10}, { 191,11}, { 111,12}, { 63,11}, \ - { 143,10}, { 287,11}, { 159,12}, { 95,11}, \ - { 191,10}, { 399,11}, { 207,10}, { 431,13}, \ - { 63,12}, { 127,11}, { 271,10}, { 543,11}, \ - { 287,10}, { 575,11}, { 303,12}, { 159,11}, \ - { 335,10}, { 671,11}, { 367,12}, { 191,11}, \ - { 431,12}, { 223,13}, { 127,12}, { 255,11}, \ - { 543,12}, { 287,11}, { 607,12}, { 351,11}, \ - { 703,13}, { 191,12}, { 415,11}, { 863,12}, \ - { 447,14}, { 127,13}, { 255,12}, { 607,13}, \ - { 319,12}, { 735,13}, { 383,12}, { 863,13}, \ - { 447,12}, { 927,11}, { 1855,12}, { 959,14}, \ - { 255,13}, { 575,12}, { 1247,13}, { 703,14}, \ - { 383,13}, { 831,12}, { 1727,13}, { 895,12}, \ - { 1791,15}, { 255,14}, { 511,13}, { 1087,12}, \ - { 2175,13}, { 1151,14}, { 639,13}, { 1343,12}, \ - { 2687,13}, { 1471,14}, { 767,13}, { 1599,12}, \ - { 3199,13}, { 1663,14}, { 895,13}, { 1855,15}, \ - { 511,14}, { 1023,13}, { 2175,14}, { 1151,13}, \ - { 2431,14}, { 1279,13}, { 2687,14}, { 1407,15}, \ - { 767,14}, { 1663,13}, { 3455,14}, { 1791,16}, \ - { 511,15}, { 1023,14}, { 2431,15}, { 1279,14}, \ - { 2943,15}, { 1535,14}, { 3455,15}, { 1791,14}, \ + { 57, 9}, { 31, 8}, { 63, 9}, { 43,10}, \ + { 23, 9}, { 55,10}, { 31, 9}, { 71,10}, \ + { 39, 9}, { 83,10}, { 47, 9}, { 99,10}, \ + { 55,11}, { 31,10}, { 87,11}, { 47,10}, \ + { 111,12}, { 31,11}, { 63,10}, { 143,11}, \ + { 79,10}, { 167,11}, { 111,12}, { 63,11}, \ + { 159,12}, { 95,11}, { 207,13}, { 8192,14}, \ { 16384,15}, { 32768,16}, { 65536,17}, { 131072,18}, \ { 262144,19}, { 524288,20}, {1048576,21}, {2097152,22}, \ {4194304,23}, {8388608,24} } -#define MUL_FFT_TABLE3_SIZE 142 -#define MUL_FFT_THRESHOLD 5760 +#define SQR_FFT_TABLE3_SIZE 46 +#define SQR_FFT_THRESHOLD 6784 -#define SQR_FFT_MODF_THRESHOLD 464 /* k = 5 */ -#define SQR_FFT_TABLE3 \ - { { 464, 5}, { 29, 6}, { 15, 5}, { 31, 6}, \ - { 35, 7}, { 18, 6}, { 37, 7}, { 33, 8}, \ - { 17, 7}, { 37, 8}, { 19, 7}, { 40, 8}, \ - { 29, 9}, { 15, 8}, { 37, 9}, { 19, 8}, \ - { 43, 9}, { 23, 8}, { 49, 9}, { 27, 8}, \ - { 57, 9}, { 43,10}, { 23, 9}, { 55,10}, \ - { 31, 9}, { 71,10}, { 39, 9}, { 83,10}, \ - { 47, 9}, { 99,10}, { 55,11}, { 31,10}, \ - { 87,11}, { 47,10}, { 111,12}, { 31,11}, \ - { 63,10}, { 135,11}, { 79,10}, { 167,11}, \ - { 95,10}, { 191,11}, { 111,12}, { 63,11}, \ - { 143,10}, { 287,11}, { 159,10}, { 319,12}, \ - { 95,11}, { 191,10}, { 399,11}, { 207,10}, \ - { 415,13}, { 63,12}, { 127,11}, { 255,10}, \ - { 511,11}, { 271,10}, { 543,11}, { 287,10}, \ - { 575,11}, { 303,12}, { 159,11}, { 335,10}, \ - { 671,11}, { 367,10}, { 735,12}, { 191,11}, \ - { 399,10}, { 799,11}, { 431,12}, { 223,11}, \ - { 463,13}, { 127,12}, { 255,11}, { 543,12}, \ - { 287,11}, { 607,12}, { 319,11}, { 671,12}, \ - { 351,11}, { 735,13}, { 191,12}, { 415,11}, \ - { 863,12}, { 447,14}, { 127,13}, { 255,12}, \ - { 607,13}, { 319,12}, { 735,13}, { 383,12}, \ - { 863,13}, { 447,12}, { 959,14}, { 255,13}, \ - { 575,12}, { 1183,13}, { 703,14}, { 383,13}, \ - { 831,12}, { 1663,13}, { 895,12}, { 1791,15}, \ - { 255,14}, { 511,13}, { 1087,12}, { 2175,13}, \ - { 1151,14}, { 639,13}, { 1343,12}, { 2687,13}, \ - { 1407,14}, { 767,13}, { 1599,12}, { 3199,13}, \ - { 1663,14}, { 895,13}, { 1855,15}, { 511,14}, \ - { 1023,13}, { 2175,14}, { 1151,13}, { 2431,14}, \ - { 1279,13}, { 2687,14}, { 1407,15}, { 767,14}, \ - { 1663,13}, { 3327,14}, { 1791,16}, { 511,15}, \ - { 1023,14}, { 2431,15}, { 1279,14}, { 2943,15}, \ - { 1535,14}, { 3327,15}, { 1791,14}, { 16384,15}, \ - { 32768,16}, { 65536,17}, { 131072,18}, { 262144,19}, \ - { 524288,20}, {1048576,21}, {2097152,22}, {4194304,23}, \ - {8388608,24} } -#define SQR_FFT_TABLE3_SIZE 149 -#define SQR_FFT_THRESHOLD 4032 - -#define MULLO_BASECASE_THRESHOLD 36 +#define MULLO_BASECASE_THRESHOLD 47 #define MULLO_DC_THRESHOLD 0 /* never mpn_mullo_basecase */ -#define MULLO_MUL_N_THRESHOLD 11278 +#define MULLO_MUL_N_THRESHOLD 17561 #define SQRLO_BASECASE_THRESHOLD 0 /* always */ -#define SQRLO_DC_THRESHOLD 11 -#define SQRLO_SQR_THRESHOLD 7916 +#define SQRLO_DC_THRESHOLD 152 +#define SQRLO_SQR_THRESHOLD 13555 -#define DC_DIV_QR_THRESHOLD 59 -#define DC_DIVAPPR_Q_THRESHOLD 222 -#define DC_BDIV_QR_THRESHOLD 94 -#define DC_BDIV_Q_THRESHOLD 264 +#define DC_DIV_QR_THRESHOLD 76 +#define DC_DIVAPPR_Q_THRESHOLD 260 +#define DC_BDIV_QR_THRESHOLD 116 +#define DC_BDIV_Q_THRESHOLD 315 -#define INV_MULMOD_BNM1_THRESHOLD 78 -#define INV_NEWTON_THRESHOLD 133 -#define INV_APPR_THRESHOLD 133 +#define INV_MULMOD_BNM1_THRESHOLD 94 +#define INV_NEWTON_THRESHOLD 17 +#define INV_APPR_THRESHOLD 21 -#define BINV_NEWTON_THRESHOLD 250 +#define BINV_NEWTON_THRESHOLD 296 #define REDC_1_TO_REDC_2_THRESHOLD 0 /* always */ -#define REDC_2_TO_REDC_N_THRESHOLD 148 +#define REDC_2_TO_REDC_N_THRESHOLD 174 -#define MU_DIV_QR_THRESHOLD 1187 -#define MU_DIVAPPR_Q_THRESHOLD 1120 +#define MU_DIV_QR_THRESHOLD 1470 +#define MU_DIVAPPR_Q_THRESHOLD 1164 #define MUPI_DIV_QR_THRESHOLD 0 /* always */ -#define MU_BDIV_QR_THRESHOLD 1210 -#define MU_BDIV_Q_THRESHOLD 1470 +#define MU_BDIV_QR_THRESHOLD 1685 +#define MU_BDIV_Q_THRESHOLD 1787 -#define POWM_SEC_TABLE 1,34,321,1985 +#define POWM_SEC_TABLE 1,22,175,1896 #define GET_STR_DC_THRESHOLD 15 -#define GET_STR_PRECOMPUTE_THRESHOLD 21 -#define SET_STR_DC_THRESHOLD 1216 -#define SET_STR_PRECOMPUTE_THRESHOLD 3170 - -#define FAC_DSC_THRESHOLD 746 -#define FAC_ODD_THRESHOLD 0 /* always */ - -#define MATRIX22_STRASSEN_THRESHOLD 24 -#define HGCD_THRESHOLD 117 -#define HGCD_APPR_THRESHOLD 111 -#define HGCD_REDUCE_THRESHOLD 3389 -#define GCD_DC_THRESHOLD 460 -#define GCDEXT_DC_THRESHOLD 411 +#define GET_STR_PRECOMPUTE_THRESHOLD 33 +#define SET_STR_DC_THRESHOLD 1418 +#define SET_STR_PRECOMPUTE_THRESHOLD 4091 + +#define FAC_DSC_THRESHOLD 1547 +#define FAC_ODD_THRESHOLD 23 + +#define MATRIX22_STRASSEN_THRESHOLD 23 +#define HGCD_THRESHOLD 141 +#define HGCD_APPR_THRESHOLD 147 +#define HGCD_REDUCE_THRESHOLD 4455 +#define GCD_DC_THRESHOLD 710 +#define GCDEXT_DC_THRESHOLD 505 #define JACOBI_BASE_METHOD 2 |