[Gmp-commit] /var/hg/gmp: Provide zen3 tuning tables.

mercurial at gmplib.org mercurial at gmplib.org
Fri Jan 1 19:24:37 UTC 2021


details:   /var/hg/gmp/rev/0178d0bbdd66
changeset: 18192:0178d0bbdd66
user:      Torbjorn Granlund <tg at gmplib.org>
date:      Fri Jan 01 20:24:27 2021 +0100
description:
Provide zen3 tuning tables.

diffstat:

 mpn/x86_64/zen3/gmp-mparam.h |  222 +++++++++++++++++++++++++++++++++++++++++++
 1 files changed, 222 insertions(+), 0 deletions(-)

diffs (226 lines):

diff -r 575ff753614f -r 0178d0bbdd66 mpn/x86_64/zen3/gmp-mparam.h
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/mpn/x86_64/zen3/gmp-mparam.h	Fri Jan 01 20:24:27 2021 +0100
@@ -0,0 +1,222 @@
+/* AMD Zen3 gmp-mparam.h -- Compiler/machine parameter header file.
+
+Copyright 2021 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
+
+/* 3800-4700 MHz Vermeer */
+/* FFT tuning limit = 10,000,000 */
+/* Generated by tuneup.c, 2021-01-01, gcc 9.3 */
+
+#define MOD_1_NORM_THRESHOLD                64
+#define MOD_1_UNNORM_THRESHOLD              85
+#define MOD_1N_TO_MOD_1_1_THRESHOLD          5
+#define MOD_1U_TO_MOD_1_1_THRESHOLD          6
+#define MOD_1_1_TO_MOD_1_2_THRESHOLD        11
+#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              9
+#define DIV_QR_1_UNNORM_THRESHOLD        MP_SIZE_T_MAX  /* never */
+#define DIV_QR_2_PI2_THRESHOLD              15
+#define DIVEXACT_1_THRESHOLD                 0  /* always (native) */
+#define BMOD_1_TO_MOD_1_THRESHOLD           18
+
+#define DIV_1_VS_MUL_1_PERCENT             398
+
+#define MUL_TOOM22_THRESHOLD                20
+#define MUL_TOOM33_THRESHOLD                89
+#define MUL_TOOM44_THRESHOLD               130
+#define MUL_TOOM6H_THRESHOLD               303
+#define MUL_TOOM8H_THRESHOLD               418
+
+#define MUL_TOOM32_TO_TOOM43_THRESHOLD      89
+#define MUL_TOOM32_TO_TOOM53_THRESHOLD     107
+#define MUL_TOOM42_TO_TOOM53_THRESHOLD      87
+#define MUL_TOOM42_TO_TOOM63_THRESHOLD      94
+#define MUL_TOOM43_TO_TOOM54_THRESHOLD     130
+
+#define SQR_BASECASE_THRESHOLD               0  /* always (native) */
+#define SQR_TOOM2_THRESHOLD                 32
+#define SQR_TOOM3_THRESHOLD                109
+#define SQR_TOOM4_THRESHOLD                336
+#define SQR_TOOM6_THRESHOLD                414
+#define SQR_TOOM8_THRESHOLD                592
+
+#define MULMID_TOOM42_THRESHOLD             38
+
+#define MULMOD_BNM1_THRESHOLD               14
+#define SQRMOD_BNM1_THRESHOLD               17
+
+#define MUL_FFT_MODF_THRESHOLD             332  /* k = 5 */
+#define MUL_FFT_TABLE3                                      \
+  { {    332, 5}, {     19, 6}, {     10, 5}, {     21, 6}, \
+    {     21, 7}, {     11, 6}, {     23, 7}, {     12, 6}, \
+    {     25, 7}, {     21, 8}, {     11, 7}, {     25, 8}, \
+    {     13, 7}, {     27, 8}, {     15, 7}, {     31, 8}, \
+    {     17, 7}, {     35, 8}, {     21, 9}, {     11, 8}, \
+    {     27, 9}, {     15, 8}, {     35, 9}, {     19, 8}, \
+    {     39, 9}, {     23, 8}, {     47, 9}, {     27,10}, \
+    {     15, 9}, {     31, 8}, {     63, 9}, {     35, 8}, \
+    {     73, 9}, {     39,10}, {     23, 9}, {     51,11}, \
+    {     15,10}, {     31, 9}, {     71,10}, {     39, 9}, \
+    {     79,10}, {     55,11}, {     31,10}, {     79,11}, \
+    {     47,10}, {     95,12}, {     31,11}, {     63,10}, \
+    {    127, 9}, {    255,10}, {    135,11}, {     79,10}, \
+    {    159, 9}, {    319, 8}, {    639,10}, {    175, 9}, \
+    {    351,11}, {     95,10}, {    191, 9}, {    383,10}, \
+    {    207,11}, {    111,10}, {    223,12}, {     63,11}, \
+    {    127,10}, {    255, 9}, {    511,10}, {    271, 9}, \
+    {    543,11}, {    143,10}, {    287, 9}, {    575,10}, \
+    {    303, 9}, {    607,11}, {    159,10}, {    319, 9}, \
+    {    639,11}, {    175,10}, {    351,12}, {     95,11}, \
+    {    191,10}, {    383,11}, {    207,10}, {    415,11}, \
+    {    223,13}, {     63,12}, {    127,11}, {    255,10}, \
+    {    511,11}, {    271,10}, {    543,11}, {    287,10}, \
+    {    575,11}, {    303,10}, {    607,12}, {    159,11}, \
+    {    319,10}, {    639,11}, {    351,10}, {    703,12}, \
+    {    191,11}, {    415,12}, {    223,11}, {    479,13}, \
+    {    127,12}, {    255,11}, {    543,10}, {   1087,12}, \
+    {    287,11}, {    575,10}, {   1151,11}, {    607,12}, \
+    {    319,11}, {    639,12}, {    351,11}, {    703,13}, \
+    {    191,12}, {    383,11}, {    767,12}, {    415,11}, \
+    {    831,10}, {   1663,12}, {    447,11}, {    895,12}, \
+    {    479,14}, {    127,13}, {    255,12}, {    511,11}, \
+    {   1023,12}, {    543,11}, {   1087,12}, {    575,11}, \
+    {   1151,12}, {    607,13}, {    319,12}, {    639,11}, \
+    {   1279,12}, {    671,11}, {   1343,10}, {   2687,12}, \
+    {    703,13}, {    383,12}, {    767,11}, {   1535,12}, \
+    {    831,11}, {   1663,13}, {    447,12}, {    895,11}, \
+    {   1791,12}, {    959,11}, {   1919,10}, {   3839,14}, \
+    {    255,13}, {    511,12}, {   1087,13}, {    575,12}, \
+    {   1151,11}, {   2303,12}, {   1215,11}, {   2431,13}, \
+    {    639,12}, {   1279, 8}, {  24063,10}, {   6399,11}, \
+    {   3327,13}, {    895,12}, {   1791,13}, {    959,12}, \
+    {   1919,11}, {   3839,15}, {    255,14}, {    511,13}, \
+    {   1087,12}, {   2175,13}, {   1215,12}, {   2431,14}, \
+    {  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 186
+#define MUL_FFT_THRESHOLD                 3264
+
+#define SQR_FFT_MODF_THRESHOLD             340  /* k = 5 */
+#define SQR_FFT_TABLE3                                      \
+  { {    340, 5}, {     19, 6}, {     10, 5}, {     21, 6}, \
+    {     11, 5}, {     23, 6}, {     25, 7}, {     13, 6}, \
+    {     27, 7}, {     25, 8}, {     13, 7}, {     28, 8}, \
+    {     15, 7}, {     31, 8}, {     21, 9}, {     11, 8}, \
+    {     27, 9}, {     15, 8}, {     33, 9}, {     19, 8}, \
+    {     43, 9}, {     23, 8}, {     47, 9}, {     27,10}, \
+    {     15, 9}, {     31, 8}, {     63, 9}, {     43,10}, \
+    {     23, 9}, {     47,11}, {     15,10}, {     31, 9}, \
+    {     67,10}, {     39, 9}, {     83,10}, {     47, 9}, \
+    {     95,10}, {     55,11}, {     31,10}, {     79,11}, \
+    {     47,10}, {     95,12}, {     31,11}, {     63,10}, \
+    {    127, 9}, {    255,10}, {    135,11}, {     79,10}, \
+    {    159, 9}, {    319,11}, {     95,10}, {    191, 9}, \
+    {    383,12}, {     63,11}, {    127,10}, {    255, 9}, \
+    {    511,10}, {    271,11}, {    143,10}, {    287, 9}, \
+    {    575,10}, {    303, 9}, {    607,11}, {    159,10}, \
+    {    319, 9}, {    639,12}, {     95,11}, {    191,10}, \
+    {    383,13}, {     63,12}, {    127,11}, {    255,10}, \
+    {    511,11}, {    271,10}, {    543,11}, {    287,10}, \
+    {    575,11}, {    303,10}, {    607,12}, {    159,11}, \
+    {    319,10}, {    639,11}, {    335,10}, {    671,11}, \
+    {    351,10}, {    703,12}, {    191,11}, {    383,10}, \
+    {    767,11}, {    415,12}, {    223,11}, {    447,13}, \
+    {    127,12}, {    255,11}, {    543,12}, {    287,11}, \
+    {    607,12}, {    319,11}, {    671,12}, {    351,11}, \
+    {    703,13}, {    191,12}, {    383,11}, {    767,12}, \
+    {    415,11}, {    831,12}, {    479,14}, {    127,13}, \
+    {    255,12}, {    511,11}, {   1023,12}, {    607,11}, \
+    {   1215,13}, {    319,12}, {    671,11}, {   1343,12}, \
+    {    735,13}, {    383,12}, {    799,11}, {   1599,10}, \
+    {   3199,12}, {    831,13}, {    447,12}, {    895,11}, \
+    {   1791,12}, {    959,11}, {   1919,13}, {    511,12}, \
+    {   1087,13}, {    575,12}, {   1215,13}, {    639,12}, \
+    {   1343,13}, {    703,12}, {   1407,14}, {    383,13}, \
+    {    767,12}, {   1599,13}, {    831,12}, {   1663,13}, \
+    {    895,12}, {   1791,13}, {    959,12}, {   1919,15}, \
+    {    255,14}, {    511,13}, {   1023, 9}, {  17919,10}, \
+    {   9727,12}, {   4096,13}, {   8192,14}, {  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 161
+#define SQR_FFT_THRESHOLD                 2624
+
+#define MULLO_BASECASE_THRESHOLD             0  /* always */
+#define MULLO_DC_THRESHOLD                  31
+#define MULLO_MUL_N_THRESHOLD             6440
+#define SQRLO_BASECASE_THRESHOLD             9
+#define SQRLO_DC_THRESHOLD                 129
+#define SQRLO_SQR_THRESHOLD               5103
+
+#define DC_DIV_QR_THRESHOLD                 19
+#define DC_DIVAPPR_Q_THRESHOLD             123
+#define DC_BDIV_QR_THRESHOLD                79
+#define DC_BDIV_Q_THRESHOLD                154
+
+#define INV_MULMOD_BNM1_THRESHOLD           42
+#define INV_NEWTON_THRESHOLD               107
+#define INV_APPR_THRESHOLD                 107
+
+#define BINV_NEWTON_THRESHOLD              312
+#define REDC_1_TO_REDC_N_THRESHOLD          77
+
+#define MU_DIV_QR_THRESHOLD               1142
+#define MU_DIVAPPR_Q_THRESHOLD            1258
+#define MUPI_DIV_QR_THRESHOLD               30
+#define MU_BDIV_QR_THRESHOLD              1120
+#define MU_BDIV_Q_THRESHOLD               1394
+
+#define POWM_SEC_TABLE  6,19,203,579,2245
+
+#define GET_STR_DC_THRESHOLD                10
+#define GET_STR_PRECOMPUTE_THRESHOLD        18
+#define SET_STR_DC_THRESHOLD               115
+#define SET_STR_PRECOMPUTE_THRESHOLD      1941
+
+#define FAC_DSC_THRESHOLD                  182
+#define FAC_ODD_THRESHOLD                   44
+
+#define MATRIX22_STRASSEN_THRESHOLD         15
+#define HGCD2_DIV1_METHOD                    1  /* 13.04% faster than 3 */
+#define HGCD_THRESHOLD                      65
+#define HGCD_APPR_THRESHOLD                 50
+#define HGCD_REDUCE_THRESHOLD             2121
+#define GCD_DC_THRESHOLD                   293
+#define GCDEXT_DC_THRESHOLD                186
+#define JACOBI_BASE_METHOD                   1  /* 12.79% faster than 3 */


More information about the gmp-commit mailing list