On Mar 4, 2:24 pm, Bill Hart <goodwillh...@googlemail.com> wrote:
> Here are the figures for the 2.66 GHz Xeon Core 2 (Dunnington 6 core
> 16 MB cache) sage.math:
>
> wbh...@sage:~/mpir-core2/tune$ ./speed -c -s 1-40 mpn_add_n
> overhead 6.00 cycles, precision 1000000 units of 3.75e-10 secs, CPU
> freq 2666.76 MHz
> mpn_add_n
> 1 17.00
> 2 20.00
> 3 23.01
> 4 31.03
> 5 25.71
> 6 29.01
> 7 33.33
> 8 58.02
> 9 43.37
> 10 42.55
> 11 44.57
> 12 46.01
> 13 71.01
> 14 49.01
> 15 52.26
> 16 55.51
> 17 63.68
> 18 61.24
> 19 63.07
> 20 66.17
> 21 63.02
> 22 68.22
> 23 74.11
> 24 75.92
> 25 82.82
> 26 82.19
> 27 82.44
> 28 87.63
> 29 84.15
> 30 91.53
> 31 94.85
> 32 96.80
> 33 120.81
> 34 100.40
> 35 102.03
> 36 107.21
> 37 106.99
> 38 109.34
> 39 113.72
> 40 117.94
> wbh...@sage:~/mpir-core2/tune$ ./speed -c -s 1-40 mpn_addmul_1.333
> overhead 6.00 cycles, precision 1000000 units of 3.75e-10 secs, CPU
> freq 2666.76 MHz
> mpn_addmul_1.333
> 1 11.85
> 2 17.83
> 3 24.99
> 4 31.22
> 5 36.40
> 6 43.06
> 7 49.31
> 8 54.65
> 9 61.02
> 10 89.02
> 11 73.39
> 12 85.03
> 13 85.48
> 14 75.02
> 15 83.35
> 16 86.34
> 17 92.01
> 18 96.67
> 19 100.75
> 20 105.01
> 21 110.68
> 22 114.09
> 23 119.45
> 24 120.05
> 25 126.03
> 26 134.08
> 27 137.49
> 28 138.85
> 29 145.88
> 30 151.45
> 31 156.56
> 32 160.89
> 33 165.30
> 34 166.04
> 35 176.96
> 36 180.00
> 37 184.05
> 38 185.16
> 39 194.50
> 40 199.32
> wbh...@sage:~/mpir-core2/tune$ ./speed -c -s 1-40 mpn_mul_1.333
> overhead 6.00 cycles, precision 1000000 units of 3.75e-10 secs, CPU
> freq 2666.76 MHz
> mpn_mul_1.333
> 1 9.00
> 2 14.00
> 3 15.41
> 4 20.45
> 5 24.00
> 6 28.35
> 7 31.47
> 8 35.40
> 9 39.41
> 10 43.34
> 11 47.90
> 12 51.58
> 13 57.01
> 14 60.42
> 15 63.82
> 16 67.71
> 17 71.15
> 18 76.45
> 19 80.44
> 20 84.15
> 21 86.77
> 22 92.89
> 23 95.89
> 24 100.01
> 25 102.57
> 26 108.36
> 27 112.20
> 28 116.82
> 29 119.07
> 30 124.28
> 31 128.07
> 32 131.94
> 33 135.51
> 34 140.80
> 35 143.22
> 36 148.33
> 37 151.70
> 38 156.04
> 39 160.32
> 40 164.29
> wbh...@sage:~/mpir-core2/tune$ ./speed -c -s 1-40 mpn_lshift.23
> overhead 6.00 cycles, precision 1000000 units of 3.75e-10 secs, CPU
> freq 2666.76 MHz
> mpn_lshift.23
> 1 8.00
> 2 10.00
> 3 17.15
> 4 19.00
> 5 20.40
> 6 24.01
> 7 26.01
> 8 26.51
> 9 28.01
> 10 32.26
> 11 34.51
> 12 35.01
> 13 37.13
> 14 40.51
> 15 43.51
> 16 43.81
> 17 45.01
> 18 48.01
> 19 52.89
> 20 52.01
> 21 54.01
> 22 56.51
> 23 58.51
> 24 60.01
> 25 61.81
> 26 64.81
> 27 66.02
> 28 67.52
> 29 70.22
> 30 72.02
> 31 75.02
> 32 75.02
> 33 81.77
> 34 81.02
> 35 83.02
> 36 84.02
> 37 88.52
> 38 88.52
> 39 92.02
> 40 91.69
> wbh...@sage:~/mpir-core2/tune$ ./speed -c -s 1-40 mpn_rshift.23
> overhead 6.00 cycles, precision 1000000 units of 3.75e-10 secs, CPU
> freq 2666.76 MHz
> mpn_rshift.23
> 1 7.00
> 2 9.00
> 3 19.00
> 4 19.00
> 5 21.00
> 6 20.00
> 7 24.01
> 8 25.01
> 9 27.55
> 10 28.01
> 11 31.01
> 12 32.19
> 13 35.01
> 14 35.91
> 15 39.12
> 16 39.51
> 17 43.51
> 18 44.01
> 19 48.01
> 20 48.01
> 21 51.01
> 22 51.61
> 23 55.51
> 24 56.26
> 25 59.01
> 26 59.83
> 27 66.01
> 28 64.01
> 29 68.01
> 30 67.52
> 31 70.73
> 32 74.26
> 33 75.62
> 34 76.52
> 35 78.77
> 36 81.77
> 37 84.02
> 38 84.32
> 39 86.64
> 40 88.02
> wbh...@sage:~/mpir-core2/tune$ ./speed -c -s 1-40 mpn_lshift1
> overhead 6.00 cycles, precision 1000000 units of 3.75e-10 secs, CPU
> freq 2666.76 MHz
> mpn_lshift1
> 1 8.00
> 2 10.00
> 3 17.54
> 4 19.00
> 5 20.98
> 6 24.01
> 7 26.85
> 8 26.30
> 9 28.01
> 10 32.96
> 11 37.21
> 12 35.01
> 13 36.01
> 14 40.01
> 15 43.51
> 16 43.01
> 17 46.30
> 18 48.01
> 19 55.21
> 20 51.01
> 21 52.51
> 22 56.95
> 23 62.41
> 24 58.51
> 25 60.01
> 26 65.26
> 27 72.01
> 28 67.21
> 29 69.02
> 30 72.02
> 31 81.02
> 32 75.02
> 33 99.02
> 34 81.45
> 35 85.52
> 36 83.42
> 37 85.52
> 38 90.02
> 39 94.52
> 40 91.52
> wbh...@sage:~/mpir-core2/tune$ ./speed -c -s 1-40 mpn_rshift1
> overhead 6.00 cycles, precision 1000000 units of 3.75e-10 secs, CPU
> freq 2666.76 MHz
> mpn_rshift1
> 1 8.00
> 2 10.00
> 3 20.00
> 4 20.00
> 5 21.11
> 6 21.00
> 7 24.01
> 8 25.01
> 9 27.01
> 10 28.13
> 11 31.29
> 12 33.01
> 13 35.55
> 14 35.34
> 15 38.34
> 16 40.51
> 17 45.01
> 18 44.15
> 19 47.01
> 20 48.39
> 21 51.01
> 22 52.21
> 23 54.61
> 24 56.01
> 25 60.01
> 26 60.44
> 27 63.01
> 28 64.21
> 29 68.16
> 30 67.51
> 31 72.02
> 32 71.02
> 33 75.02
> 34 75.02
> 35 78.02
> 36 79.93
> 37 84.02
> 38 84.02
> 39 86.47
> 40 87.02
> wbh...@sage:~/mpir-core2/tune$ ./speed -c -s 1-40 mpn_mul_basecase
> overhead 6.00 cycles, precision 1000000 units of 3.75e-10 secs, CPU
> freq 2666.76 MHz
> mpn_mul_basecase
> 1 9.00
> 2 20.97
> 3 56.24
> 4 89.28
> 5 131.04
> 6 190.59
> 7 256.40
> 8 329.49
> 9 414.08
> 10 520.59
> 11 619.74
> 12 723.87
> 13 853.75
> 14 1003.21
> 15 1147.46
> 16 1305.83
> 17 1468.22
> 18 1669.51
> 19 1834.72
> 20 2043.40
> 21 2241.04
> 22 2485.90
> 23 2714.61
> 24 2943.96
> 25 3154.04
> 26 3487.28
> 27 3727.14
> 28 4019.71
> 29 4292.92
> 30 4637.73
> 31 4915.73
> 32 5239.38
> 33 5561.10
> 34 5960.99
> 35 6287.39
> 36 6634.71
> 37 7022.21
> 38 7441.70
> 39 7892.99
> 40 8658.11
> wbh...@sage:~/mpir-core2/tune$ ./speed -c -s 1-40 mpn_sqr_basecase
> overhead 6.00 cycles, precision 1000000 units of 3.75e-10 secs, CPU
> freq 2666.76 MHz
> mpn_sqr_basecase
> 1 8.00
> 2 18.01
> 3 45.01
> 4 86.02
> 5 116.69
> 6 161.37
> 7 194.06
> 8 240.56
> 9 297.58
> 10 362.08
> 11 411.06
> 12 475.73
> 13 540.43
> 14 617.80
> 15 696.23
> 16 785.11
> 17 878.98
> 18 973.20
> 19 1083.13
> 20 1159.81
> 21 1288.83
> 22 1413.27
> 23 1587.34
> 24 1693.37
> 25 1819.39
> 26 1965.45
> 27 2128.45
> 28 2277.46
> 29 2427.53
> 30 2606.62
> 31 2805.60
> 32 2976.68
> 33 3210.61
> 34 3365.72
> 35 3583.71
> 36 3752.86
> 37 3998.91
> 38 4183.31
> 39 4451.05
> 40 4677.09
>
> Bill.
>
> 2009/3/4 Jeff Gilchrist...
>
> read more »
Thanks for the new figures Bill. Jeff is looking to get some more as
well.
It takes me some time to produce these graphs so I hope you will
forgive me I don't republish immediately.
If I get a lot more figures I am going to have to do this a bit
differently as my current spreadsheet is getting very big!
Brian
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"mpir-devel" group.
To post to this group, send email to mpir-devel@googlegroups.com
To unsubscribe from this group, send email to
mpir-devel+unsubscr...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/mpir-devel?hl=en
-~----------~----~----~----~------~----~------~--~---
