For posterity, here are the results on the Atom N450 running in 32 bit mode:
#############################################################################
# With assembly file
#############################################################################
./speed -s 1-100 -t 1 mpn_sqr_basecase >> atom.tune
overhead 0.000000005 secs, precision 1000000 units of 6.00e-10 secs, CPU
freq 1667.00 MHz
mpn_sqr_basecase
1 0.000000013
2 0.000000029
3 0.000000061
4 0.000000117
5 0.000000173
6 0.000000248
7 0.000000320
8 0.000000400
9 0.000000485
10 0.000000578
11 0.000000677
12 0.000000782
13 0.000000906
14 0.000001023
15 0.000001171
16 0.000001302
17 0.000001441
18 0.000001594
19 0.000001747
20 0.000001906
21 0.000002071
22 0.000002242
23 0.000002421
24 0.000002605
25 0.000002797
26 0.000002995
27 0.000003199
28 0.000003411
29 0.000003629
30 0.000003854
31 0.000004087
32 0.000004326
33 0.000004570
34 0.000004821
35 0.000005079
36 0.000005343
37 0.000005613
38 0.000005891
39 0.000006174
40 0.000006465
41 0.000006762
42 0.000007065
43 0.000007376
44 0.000007692
45 0.000008017
46 0.000008347
47 0.000008686
48 0.000009030
49 0.000009380
50 0.000009737
zsh: segmentation fault (core dumped) ./speed -s 1-100 -t 1
mpn_sqr_basecase
make tune
./tuneup
Parameters for ./mpn/x86/k7/gmp-mparam.h
Using: CPU cycle counter, supplemented by microsecond getrusage()
speed_precision 1000000, speed_unittime 1.00e-09 secs, CPU freq 1000.00 MHz
DEFAULT_MAX_SIZE 1000, fft_max_size 50000
/* Generated by tuneup.c, 2014-02-17, gcc 4.7 */
#define MUL_KARATSUBA_THRESHOLD 20
#define MUL_TOOM3_THRESHOLD 114
#define MUL_TOOM4_THRESHOLD 184
#define MUL_TOOM8H_THRESHOLD 286
#define SQR_BASECASE_THRESHOLD 0 /* always (native) */
#define SQR_KARATSUBA_THRESHOLD 36
#define SQR_TOOM3_THRESHOLD 117
#define SQR_TOOM4_THRESHOLD 298
#define SQR_TOOM8_THRESHOLD 324
#define POWM_THRESHOLD 128
#define HGCD_THRESHOLD 179
#define GCD_DC_THRESHOLD 442
#define GCDEXT_DC_THRESHOLD 309
#define JACOBI_BASE_METHOD 1
#define USE_PREINV_DIVREM_1 1 /* native */
#define USE_PREINV_MOD_1 1 /* native */
#define DIVREM_2_THRESHOLD 0 /* always */
#define DIVEXACT_1_THRESHOLD 0 /* always (native) */
#define MODEXACT_1_ODD_THRESHOLD 0 /* always (native) */
#define MOD_1_1_THRESHOLD 83
#define MOD_1_2_THRESHOLD 195
#define MOD_1_3_THRESHOLD 502
#define DIVREM_HENSEL_QR_1_THRESHOLD 74
#define RSH_DIVREM_HENSEL_QR_1_THRESHOLD 987
#define DIVREM_EUCLID_HENSEL_THRESHOLD 58
#define ROOTREM_THRESHOLD 6
#define GET_STR_DC_THRESHOLD 13
#define GET_STR_PRECOMPUTE_THRESHOLD 25
#define SET_STR_DC_THRESHOLD 208
#define SET_STR_PRECOMPUTE_THRESHOLD 272
#define MUL_FFT_FULL_THRESHOLD 3648
#define SQR_FFT_FULL_THRESHOLD 3008
#define MULLOW_BASECASE_THRESHOLD 4
#define MULLOW_DC_THRESHOLD 39
#define MULLOW_MUL_THRESHOLD 2966
#define MULHIGH_BASECASE_THRESHOLD 8
#define MULHIGH_DC_THRESHOLD 39
#define MULHIGH_MUL_THRESHOLD 2908
#define MULMOD_2EXPM1_THRESHOLD 20
#define FAC_UI_THRESHOLD 2849
#define DC_DIV_QR_THRESHOLD 68
#define DC_DIVAPPR_Q_N_THRESHOLD 22
#define INV_DIV_QR_THRESHOLD 2130
#define INV_DIVAPPR_Q_N_THRESHOLD 22
#define DC_DIV_Q_THRESHOLD 225
#define INV_DIV_Q_THRESHOLD 2747
#define DC_DIVAPPR_Q_THRESHOLD 217
#define INV_DIVAPPR_Q_THRESHOLD 7910
#define DC_BDIV_QR_THRESHOLD 68
#define DC_BDIV_Q_THRESHOLD 128
/* fft_tuning -- autogenerated by tune-fft */
#define FFT_TAB \
{ { 3, 3 }, { 3, 2 }, { 2, 1 }, { 1, 0 }, { 1, 0 } }
#define MULMOD_TAB \
{ 4, 3, 3, 3, 3, 2, 2, 2, 2, 2, 2, 2, 2, 1, 1 }
#define FFT_N_NUM 15
#define FFT_MULMOD_2EXPP1_CUTOFF 128
/* Tuneup completed successfully, took 794 seconds */
#############################################################################
# Without assembly file mpn/x86/k7/sqr_basecase.asm
#############################################################################
overhead 0.000000008 secs, precision 1000000 units of 1.00e-09 secs, CPU
freq 1000.00 MHz
mpn_sqr_basecase
1 0.000000022
2 0.000000054
3 0.000000136
4 0.000000211
5 0.000000301
6 0.000000407
7 0.000000529
8 0.000000667
9 0.000000850
10 0.000001019
11 0.000001222
12 0.000001438
13 0.000001670
14 0.000002148
15 0.000002227
16 0.000002524
17 0.000002843
18 0.000003155
19 0.000003483
20 0.000003827
21 0.000004187
22 0.000004563
23 0.000004955
24 0.000005363
25 0.000005787
26 0.000006228
27 0.000006684
28 0.000007156
29 0.000007644
30 0.000008148
31 0.000008668
32 0.000009205
33 0.000009757
34 0.000010325
35 0.000010909
36 0.000011509
37 0.000012125
38 0.000012757
39 0.000013405
40 0.000014069
41 0.000014750
42 0.000015446
43 0.000016158
44 0.000016886
45 0.000017631
46 0.000018391
47 0.000019167
48 0.000019959
49 0.000020767
50 0.000021592
51 0.000022432
52 0.000023289
53 0.000024160
54 0.000025048
55 0.000025953
56 0.000026873
57 0.000027809
58 0.000028762
59 0.000029729
60 0.000030713
61 0.000031715
62 0.000032730
63 0.000033763
64 0.000034810
65 0.000035876
66 0.000036955
67 0.000038051
68 0.000039163
69 0.000040292
70 0.000041436
71 0.000042596
72 0.000043772
73 0.000044966
74 0.000046173
75 0.000047397
76 0.000048639
77 0.000049894
78 0.000051166
79 0.000052456
80 0.000053760
81 0.000055080
82 0.000056415
83 0.000057768
84 0.000059137
85 0.000060522
86 0.000061922
87 0.000063338
88 0.000064771
89 0.000066218
90 0.000067682
91 0.000069163
92 0.000070659
93 0.000072172
94 0.000073703
95 0.000075246
96 0.000076805
97 0.000078381
98 0.000079974
99 0.000081581
100 0.000083205
--
You received this message because you are subscribed to the Google Groups
"mpir-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to mpir-devel+unsubscr...@googlegroups.com.
To post to this group, send email to mpir-devel@googlegroups.com.
Visit this group at http://groups.google.com/group/mpir-devel.
For more options, visit https://groups.google.com/groups/opt_out.
