[Gmp-commit] /var/hg/gmp: Retune (mainly for DIV_1_VS_MUL_1_PERCENT).
mercurial at gmplib.org
mercurial at gmplib.org
Sun Jan 1 15:10:17 UTC 2017
details: /var/hg/gmp/rev/faa6086f89b6
changeset: 17196:faa6086f89b6
user: Torbjorn Granlund <tg at gmplib.org>
date: Sun Jan 01 16:10:15 2017 +0100
description:
Retune (mainly for DIV_1_VS_MUL_1_PERCENT).
diffstat:
mpn/arm/v6/gmp-mparam.h | 2 +
mpn/arm/v7a/cora15/gmp-mparam.h | 2 +
mpn/arm/v7a/cora7/gmp-mparam.h | 2 +
mpn/arm/v7a/cora8/gmp-mparam.h | 2 +
mpn/arm/v7a/cora9/gmp-mparam.h | 2 +
mpn/arm64/gmp-mparam.h | 2 +
mpn/arm64/xgene1/gmp-mparam.h | 2 +
mpn/powerpc32/p4/gmp-mparam.h | 2 +
mpn/powerpc64/mode32/p4/gmp-mparam.h | 195 ++++++++++++++++++----------------
mpn/powerpc64/mode64/p4/gmp-mparam.h | 2 +
mpn/x86/atom/gmp-mparam.h | 2 +
mpn/x86/bobcat/gmp-mparam.h | 2 +
mpn/x86/coreinhm/gmp-mparam.h | 2 +
mpn/x86/coreisbr/gmp-mparam.h | 2 +
mpn/x86/k10/gmp-mparam.h | 2 +
mpn/x86/pentium4/sse2/gmp-mparam.h | 2 +
mpn/x86_64/atom/gmp-mparam.h | 2 +
mpn/x86_64/bobcat/gmp-mparam.h | 2 +
mpn/x86_64/coreihwl/gmp-mparam.h | 2 +
mpn/x86_64/coreinhm/gmp-mparam.h | 2 +
mpn/x86_64/k10/gmp-mparam.h | 2 +
mpn/x86_64/pentium4/gmp-mparam.h | 2 +
22 files changed, 144 insertions(+), 93 deletions(-)
diffs (truncated from 507 to 300 lines):
diff -r 9a5d42a0d435 -r faa6086f89b6 mpn/arm/v6/gmp-mparam.h
--- a/mpn/arm/v6/gmp-mparam.h Sat Dec 31 22:09:11 2016 +0100
+++ b/mpn/arm/v6/gmp-mparam.h Sun Jan 01 16:10:15 2017 +0100
@@ -53,6 +53,8 @@
#define DIVEXACT_1_THRESHOLD 0 /* always (native) */
#define BMOD_1_TO_MOD_1_THRESHOLD 39
+#define DIV_1_VS_MUL_1_PERCENT 178
+
#define MUL_TOOM22_THRESHOLD 44
#define MUL_TOOM33_THRESHOLD 138
#define MUL_TOOM44_THRESHOLD 517
diff -r 9a5d42a0d435 -r faa6086f89b6 mpn/arm/v7a/cora15/gmp-mparam.h
--- a/mpn/arm/v7a/cora15/gmp-mparam.h Sat Dec 31 22:09:11 2016 +0100
+++ b/mpn/arm/v7a/cora15/gmp-mparam.h Sun Jan 01 16:10:15 2017 +0100
@@ -51,6 +51,8 @@
#define DIVEXACT_1_THRESHOLD 0 /* always (native) */
#define BMOD_1_TO_MOD_1_THRESHOLD 15
+#define DIV_1_VS_MUL_1_PERCENT 220
+
#define MUL_TOOM22_THRESHOLD 28
#define MUL_TOOM33_THRESHOLD 116
#define MUL_TOOM44_THRESHOLD 169
diff -r 9a5d42a0d435 -r faa6086f89b6 mpn/arm/v7a/cora7/gmp-mparam.h
--- a/mpn/arm/v7a/cora7/gmp-mparam.h Sat Dec 31 22:09:11 2016 +0100
+++ b/mpn/arm/v7a/cora7/gmp-mparam.h Sun Jan 01 16:10:15 2017 +0100
@@ -51,6 +51,8 @@
#define DIVEXACT_1_THRESHOLD 0 /* always (native) */
#define BMOD_1_TO_MOD_1_THRESHOLD 47
+#define DIV_1_VS_MUL_1_PERCENT 163
+
#define MUL_TOOM22_THRESHOLD 39
#define MUL_TOOM33_THRESHOLD 129
#define MUL_TOOM44_THRESHOLD 195
diff -r 9a5d42a0d435 -r faa6086f89b6 mpn/arm/v7a/cora8/gmp-mparam.h
--- a/mpn/arm/v7a/cora8/gmp-mparam.h Sat Dec 31 22:09:11 2016 +0100
+++ b/mpn/arm/v7a/cora8/gmp-mparam.h Sun Jan 01 16:10:15 2017 +0100
@@ -51,6 +51,8 @@
#define DIVEXACT_1_THRESHOLD 0 /* always (native) */
#define BMOD_1_TO_MOD_1_THRESHOLD 28
+#define DIV_1_VS_MUL_1_PERCENT 192
+
#define MUL_TOOM22_THRESHOLD 39
#define MUL_TOOM33_THRESHOLD 129
#define MUL_TOOM44_THRESHOLD 220
diff -r 9a5d42a0d435 -r faa6086f89b6 mpn/arm/v7a/cora9/gmp-mparam.h
--- a/mpn/arm/v7a/cora9/gmp-mparam.h Sat Dec 31 22:09:11 2016 +0100
+++ b/mpn/arm/v7a/cora9/gmp-mparam.h Sun Jan 01 16:10:15 2017 +0100
@@ -51,6 +51,8 @@
#define DIVEXACT_1_THRESHOLD 0 /* always (native) */
#define BMOD_1_TO_MOD_1_THRESHOLD 20
+#define DIV_1_VS_MUL_1_PERCENT 190
+
#define MUL_TOOM22_THRESHOLD 45
#define MUL_TOOM33_THRESHOLD 129
#define MUL_TOOM44_THRESHOLD 387
diff -r 9a5d42a0d435 -r faa6086f89b6 mpn/arm64/gmp-mparam.h
--- a/mpn/arm64/gmp-mparam.h Sat Dec 31 22:09:11 2016 +0100
+++ b/mpn/arm64/gmp-mparam.h Sun Jan 01 16:10:15 2017 +0100
@@ -53,6 +53,8 @@
#define DIVEXACT_1_THRESHOLD 0 /* always */
#define BMOD_1_TO_MOD_1_THRESHOLD 39
+#define DIV_1_VS_MUL_1_PERCENT 137
+
#define MUL_TOOM22_THRESHOLD 11
#define MUL_TOOM33_THRESHOLD 78
#define MUL_TOOM44_THRESHOLD 118
diff -r 9a5d42a0d435 -r faa6086f89b6 mpn/arm64/xgene1/gmp-mparam.h
--- a/mpn/arm64/xgene1/gmp-mparam.h Sat Dec 31 22:09:11 2016 +0100
+++ b/mpn/arm64/xgene1/gmp-mparam.h Sun Jan 01 16:10:15 2017 +0100
@@ -53,6 +53,8 @@
#define DIVEXACT_1_THRESHOLD 0 /* always */
#define BMOD_1_TO_MOD_1_THRESHOLD 29
+#define DIV_1_VS_MUL_1_PERCENT 313
+
#define MUL_TOOM22_THRESHOLD 16
#define MUL_TOOM33_THRESHOLD 73
#define MUL_TOOM44_THRESHOLD 160
diff -r 9a5d42a0d435 -r faa6086f89b6 mpn/powerpc32/p4/gmp-mparam.h
--- a/mpn/powerpc32/p4/gmp-mparam.h Sat Dec 31 22:09:11 2016 +0100
+++ b/mpn/powerpc32/p4/gmp-mparam.h Sun Jan 01 16:10:15 2017 +0100
@@ -51,6 +51,8 @@
#define DIVEXACT_1_THRESHOLD 0 /* always */
#define BMOD_1_TO_MOD_1_THRESHOLD 45
+#define DIV_1_VS_MUL_1_PERCENT 225
+
#define MUL_TOOM22_THRESHOLD 20
#define MUL_TOOM33_THRESHOLD 73
#define MUL_TOOM44_THRESHOLD 130
diff -r 9a5d42a0d435 -r faa6086f89b6 mpn/powerpc64/mode32/p4/gmp-mparam.h
--- a/mpn/powerpc64/mode32/p4/gmp-mparam.h Sat Dec 31 22:09:11 2016 +0100
+++ b/mpn/powerpc64/mode32/p4/gmp-mparam.h Sun Jan 01 16:10:15 2017 +0100
@@ -1,6 +1,6 @@
/* PowerPC-64 gmp-mparam.h -- Compiler/machine parameter header file.
-Copyright 2008, 2009, 2011 Free Software Foundation, Inc.
+Copyright 2008, 2009, 2011, 2017 Free Software Foundation, Inc.
This file is part of the GNU MP Library.
@@ -29,6 +29,8 @@
see https://www.gnu.org/licenses/. */
/* 1800 MHz PPC970 */
+/* FFT tuning limit = 0.5 M */
+/* Generated by tuneup.c, 2017-01-01, gcc 4.0 */
#define DIVREM_1_NORM_THRESHOLD 0 /* always */
#define DIVREM_1_UNNORM_THRESHOLD 0 /* always */
@@ -39,135 +41,142 @@
#define MOD_1U_TO_MOD_1_1_THRESHOLD 5
#define MOD_1_1_TO_MOD_1_2_THRESHOLD 6
#define MOD_1_2_TO_MOD_1_4_THRESHOLD 46
-#define PREINV_MOD_1_TO_MOD_1_THRESHOLD 14
+#define PREINV_MOD_1_TO_MOD_1_THRESHOLD 15
#define USE_PREINV_DIVREM_1 1
-#define DIV_QR_2_PI2_THRESHOLD 12
+#define DIV_QR_1N_PI1_METHOD 1
+#define DIV_QR_1_NORM_THRESHOLD 2
+#define DIV_QR_1_UNNORM_THRESHOLD 2
+#define DIV_QR_2_PI2_THRESHOLD 15
#define DIVEXACT_1_THRESHOLD 0 /* always */
-#define BMOD_1_TO_MOD_1_THRESHOLD 90
+#define BMOD_1_TO_MOD_1_THRESHOLD 88
-#define MUL_TOOM22_THRESHOLD 16
-#define MUL_TOOM33_THRESHOLD 57
-#define MUL_TOOM44_THRESHOLD 94
-#define MUL_TOOM6H_THRESHOLD 125
+#define DIV_1_VS_MUL_1_PERCENT 269
+
+#define MUL_TOOM22_THRESHOLD 18
+#define MUL_TOOM33_THRESHOLD 60
+#define MUL_TOOM44_THRESHOLD 88
+#define MUL_TOOM6H_THRESHOLD 124
#define MUL_TOOM8H_THRESHOLD 187
-#define MUL_TOOM32_TO_TOOM43_THRESHOLD 65
-#define MUL_TOOM32_TO_TOOM53_THRESHOLD 99
+#define MUL_TOOM32_TO_TOOM43_THRESHOLD 61
+#define MUL_TOOM32_TO_TOOM53_THRESHOLD 91
#define MUL_TOOM42_TO_TOOM53_THRESHOLD 61
-#define MUL_TOOM42_TO_TOOM63_THRESHOLD 56
-#define MUL_TOOM43_TO_TOOM54_THRESHOLD 70
+#define MUL_TOOM42_TO_TOOM63_THRESHOLD 60
+#define MUL_TOOM43_TO_TOOM54_THRESHOLD 74
#define SQR_BASECASE_THRESHOLD 4
-#define SQR_TOOM2_THRESHOLD 30
-#define SQR_TOOM3_THRESHOLD 98
-#define SQR_TOOM4_THRESHOLD 136
-#define SQR_TOOM6_THRESHOLD 180
+#define SQR_TOOM2_THRESHOLD 28
+#define SQR_TOOM3_THRESHOLD 90
+#define SQR_TOOM4_THRESHOLD 143
+#define SQR_TOOM6_THRESHOLD 181
#define SQR_TOOM8_THRESHOLD 272
#define MULMID_TOOM42_THRESHOLD 34
-#define MULMOD_BNM1_THRESHOLD 12
-#define SQRMOD_BNM1_THRESHOLD 13
+#define MULMOD_BNM1_THRESHOLD 10
+#define SQRMOD_BNM1_THRESHOLD 15
-#define MUL_FFT_MODF_THRESHOLD 244 /* k = 5 */
+#define MUL_FFT_MODF_THRESHOLD 252 /* k = 5 */
#define MUL_FFT_TABLE3 \
- { { 244, 5}, { 13, 6}, { 7, 5}, { 15, 6}, \
+ { { 252, 5}, { 11, 6}, { 6, 5}, { 13, 6}, \
+ { 7, 5}, { 15, 6}, { 13, 5}, { 27, 6}, \
{ 15, 7}, { 8, 6}, { 17, 7}, { 9, 6}, \
- { 19, 7}, { 13, 8}, { 7, 7}, { 17, 8}, \
- { 9, 7}, { 20, 8}, { 11, 7}, { 23, 8}, \
- { 13, 7}, { 29, 8}, { 19, 9}, { 11, 8}, \
- { 27,10}, { 7, 9}, { 15, 8}, { 33, 9}, \
+ { 19, 7}, { 17, 8}, { 9, 7}, { 20, 8}, \
+ { 11, 7}, { 23, 8}, { 13, 9}, { 7, 8}, \
+ { 21, 9}, { 11, 8}, { 27,10}, { 7, 9}, \
+ { 15, 8}, { 33, 9}, { 19, 8}, { 39, 9}, \
+ { 23, 8}, { 47, 9}, { 27,10}, { 15, 9}, \
+ { 39,10}, { 23, 9}, { 47,11}, { 15,10}, \
+ { 31, 9}, { 67,10}, { 39, 9}, { 83,10}, \
+ { 47, 9}, { 95, 8}, { 191,10}, { 55,11}, \
+ { 31,10}, { 63, 9}, { 127, 8}, { 255,10}, \
+ { 71, 9}, { 143, 8}, { 287,10}, { 79, 9}, \
+ { 159, 8}, { 319,11}, { 47,10}, { 95, 9}, \
+ { 191, 8}, { 383,10}, { 103,12}, { 31,11}, \
+ { 63,10}, { 127, 9}, { 255, 8}, { 511,10}, \
+ { 143, 9}, { 287,11}, { 79,10}, { 159, 9}, \
+ { 319, 8}, { 639,10}, { 175, 9}, { 351, 8}, \
+ { 703,11}, { 95,10}, { 191, 9}, { 383, 8}, \
+ { 767,10}, { 207, 9}, { 415,10}, { 223, 9}, \
+ { 447,12}, { 63,11}, { 127,10}, { 255, 9}, \
+ { 511,11}, { 143,10}, { 287, 9}, { 575,11}, \
+ { 159,10}, { 319, 9}, { 639,11}, { 175,10}, \
+ { 351, 9}, { 703,12}, { 95,11}, { 191,10}, \
+ { 383, 9}, { 767,11}, { 207,10}, { 415,11}, \
+ { 223,10}, { 447,13}, { 8192,14}, { 16384,15}, \
+ { 32768,16} }
+#define MUL_FFT_TABLE3_SIZE 105
+#define MUL_FFT_THRESHOLD 5248
+
+#define SQR_FFT_MODF_THRESHOLD 236 /* k = 5 */
+#define SQR_FFT_TABLE3 \
+ { { 236, 5}, { 13, 6}, { 15, 7}, { 8, 6}, \
+ { 17, 7}, { 9, 6}, { 19, 7}, { 17, 8}, \
+ { 9, 7}, { 20, 8}, { 11, 7}, { 24, 8}, \
+ { 13, 9}, { 7, 8}, { 19, 9}, { 11, 8}, \
+ { 25,10}, { 7, 9}, { 15, 8}, { 33, 9}, \
{ 19, 8}, { 39, 9}, { 23, 8}, { 47, 9}, \
{ 27,10}, { 15, 9}, { 39,10}, { 23, 9}, \
{ 47,11}, { 15,10}, { 31, 9}, { 67,10}, \
- { 39, 9}, { 83,10}, { 47, 9}, { 95, 8}, \
- { 191, 9}, { 99,10}, { 55,11}, { 31,10}, \
- { 63, 9}, { 127, 8}, { 255,10}, { 71, 9}, \
- { 143, 8}, { 287,10}, { 79, 9}, { 159, 8}, \
- { 319,11}, { 47,10}, { 95, 9}, { 191, 8}, \
- { 383,10}, { 103,12}, { 31,11}, { 63,10}, \
- { 127, 9}, { 255, 8}, { 511,10}, { 143, 9}, \
- { 287,11}, { 79,10}, { 159, 9}, { 319, 8}, \
- { 639,10}, { 175, 9}, { 351, 8}, { 703,11}, \
- { 95,10}, { 191, 9}, { 383, 8}, { 767,10}, \
- { 207, 9}, { 415,10}, { 223, 9}, { 447,12}, \
- { 63,11}, { 127,10}, { 255, 9}, { 511,11}, \
- { 143,10}, { 287, 9}, { 575, 8}, { 1151,11}, \
- { 159,10}, { 319, 9}, { 639,11}, { 175,10}, \
- { 351, 9}, { 703,12}, { 95,11}, { 191,10}, \
- { 383, 9}, { 767,11}, { 207,10}, { 415, 9}, \
- { 831,11}, { 223,10}, { 447,13}, { 8192,14}, \
- { 16384,15}, { 32768,16} }
-#define MUL_FFT_TABLE3_SIZE 106
-#define MUL_FFT_THRESHOLD 2688
-
-#define SQR_FFT_MODF_THRESHOLD 212 /* k = 5 */
-#define SQR_FFT_TABLE3 \
- { { 212, 5}, { 13, 6}, { 15, 7}, { 8, 6}, \
- { 17, 7}, { 9, 6}, { 19, 7}, { 13, 8}, \
- { 7, 7}, { 17, 8}, { 9, 7}, { 20, 8}, \
- { 11, 7}, { 23, 8}, { 13, 7}, { 27, 9}, \
- { 7, 8}, { 21, 9}, { 11, 8}, { 25,10}, \
- { 7, 9}, { 15, 8}, { 33, 9}, { 19, 8}, \
- { 39, 9}, { 23, 8}, { 47, 9}, { 27,10}, \
- { 15, 9}, { 39,10}, { 23, 9}, { 47,11}, \
- { 15,10}, { 31, 9}, { 63, 8}, { 127, 9}, \
- { 67,10}, { 39, 9}, { 79, 8}, { 159,10}, \
- { 47, 9}, { 95, 8}, { 191,11}, { 31,10}, \
- { 63, 9}, { 127, 8}, { 255,10}, { 71, 9}, \
- { 143, 8}, { 287,10}, { 79, 9}, { 159, 8}, \
- { 319,11}, { 47, 9}, { 191, 8}, { 383,12}, \
+ { 39, 9}, { 79, 8}, { 159,10}, { 47, 9}, \
+ { 95, 8}, { 191,11}, { 31,10}, { 63, 9}, \
+ { 127, 8}, { 255,10}, { 71, 9}, { 143, 8}, \
+ { 287,10}, { 79, 9}, { 159, 8}, { 319,11}, \
+ { 47,10}, { 95, 9}, { 191, 8}, { 383,12}, \
{ 31,11}, { 63,10}, { 127, 9}, { 255, 8}, \
{ 511,10}, { 143, 9}, { 287, 8}, { 575,11}, \
{ 79,10}, { 159, 9}, { 319, 8}, { 639,10}, \
- { 175, 9}, { 351, 8}, { 703,10}, { 191, 9}, \
- { 383, 8}, { 767,10}, { 207, 9}, { 415,11}, \
- { 111,10}, { 223,12}, { 63,11}, { 127,10}, \
+ { 175, 9}, { 351, 8}, { 703,11}, { 95,10}, \
+ { 191, 9}, { 383, 8}, { 767,10}, { 207, 9}, \
+ { 415,10}, { 223,12}, { 63,11}, { 127,10}, \
{ 255, 9}, { 511,11}, { 143,10}, { 287, 9}, \
{ 575,11}, { 159,10}, { 319, 9}, { 639,11}, \
- { 175,10}, { 351, 9}, { 703, 8}, { 1407,11}, \
- { 191,10}, { 383, 9}, { 767,11}, { 207,10}, \
- { 415,11}, { 223,10}, { 447,13}, { 8192,14}, \
- { 16384,15}, { 32768,16} }
-#define SQR_FFT_TABLE3_SIZE 102
-#define SQR_FFT_THRESHOLD 1984
+ { 175,10}, { 351, 9}, { 703,11}, { 191,10}, \
+ { 383, 9}, { 767,11}, { 207,10}, { 415,11}, \
+ { 223,10}, { 447,13}, { 8192,14}, { 16384,15}, \
+ { 32768,16} }
+#define SQR_FFT_TABLE3_SIZE 97
+#define SQR_FFT_THRESHOLD 3200
#define MULLO_BASECASE_THRESHOLD 0 /* always */
-#define MULLO_DC_THRESHOLD 55
-#define MULLO_MUL_N_THRESHOLD 5240
+#define MULLO_DC_THRESHOLD 56
+#define MULLO_MUL_N_THRESHOLD 8648
+#define SQRLO_BASECASE_THRESHOLD 2
+#define SQRLO_DC_THRESHOLD 106
+#define SQRLO_SQR_THRESHOLD 6293
-#define DC_DIV_QR_THRESHOLD 27
-#define DC_DIVAPPR_Q_THRESHOLD 108
+#define DC_DIV_QR_THRESHOLD 28
+#define DC_DIVAPPR_Q_THRESHOLD 102
#define DC_BDIV_QR_THRESHOLD 51
More information about the gmp-commit
mailing list