LL with OpenCL

[kracker]

Originally Posted by kracker

HD 7770

Iteration 10000 M( 22256453 )C, 0x3d9450d492b7e880, n = 1179648, clLucas v1.00 err = 0.2656 (2:39 real, 15.9270 ms/iter, ETA 98:23:35)
Iteration 10000 M( 24732709 )C, 0x81a12a304a754572, n = 1310720, clLucas v1.00 err = 0.2813 (3:04 real, 18.4028 ms/iter, ETA 126:21:56)
Iteration 10000 M( 29412433 )C, 0x27d7d112a73aa203, n = 1572864, clLucas v1.00 err = 0.25 (4:02 real, 24.1517 ms/iter, ETA 197:14:18)
Iteration 10000 M( 30620113 )C, 0x212dca3cec0acde2, n = 1638400, clLucas v1.00 err = 0.25 (5:11 real, 31.0753 ms/iter, ETA 264:13:34)
Iteration 10000 M( 32993419 )C, 0xcf86a69b844e35c0, n = 1769472, clLucas v1.00 err = 0.2813 (7:18 real, 43.7117 ms/iter, ETA 400:26:52)
Iteration 10000 M( 36418493 )C, 0x2f1388379572d5b4, n = 1966080, clLucas v1.00 err = 0.25 (5:30 real, 33.0295 ms/iter, ETA 333:57:52)
Iteration 10000 M( 38955173 )C, 0x8a45e3bbd4e4fc9b, n = 2097152, clLucas v1.00 err = 0.25 (2:03 real, 12.2586 ms/iter, ETA 132:35:49)
Iteration 10000 M( 43792559 )C, 0x7048d84bbfb0f810, n = 2359296, clLucas v1.00 err = 0.2813 (5:54 real, 35.3544 ms/iter, ETA 429:56:54)
Iteration 10000 M( 48375209 )C, 0xf957e240d591a99e, n = 2621440, clLucas v1.00 err = 0.2188 (6:33 real, 39.2538 ms/iter, ETA 527:18:36)
Iteration 10000 M( 57899201 )C, 0xa2ac01bbc76d92ee, n = 3145728, clLucas v1.00 err = 0.25 (9:00 real, 53.9709 ms/iter, ETA 867:43:57)
Iteration 10000 M( 60622229 )C, 0xd81c849f11fd1054, n = 3276800, clLucas v1.00 err = 0.2813 (11:06 real, 66.5953 ms/iter, ETA 1121:12:22)
Iteration 10000 M( 65066623 )C, 0xde7aeb8cc7a2a826, n = 3538944, clLucas v1.00 err = 0.2539 (16:00 real, 96.0663 ms/iter, ETA 1735:51:51)
Iteration 10000 M( 67662869 )C, 0xf854d1dee3fbb5d7, n = 3932160, clLucas v1.00 err = 0.05078 (11:59 real, 71.8933 ms/iter, ETA 1350:59:44)
Iteration 10000 M( 76722161 )C, 0x4b6ba0a6078e4bbb, n = 4194304, clLucas v1.00 err = 0.25 (4:14 real, 24.9663 ms/iter, ETA 540:43:00)

This is from msft's code, but the windows exe I have given is not (yet).

[TeknoHog]

Latest data on my 7970 after updating the fglrx driver from 13.6 to 13.8 beta2, which fixed the speed issue:

Iteration 31850000 M( 31874257 )C, 0x0d7f34784f593253, n = 2097152, clLucas v1.0
0 err = 0.003784 (0:38 real, 3.7953 ms/iter, ETA 1:15)
Iteration 31860000 M( 31874257 )C, 0xf108c23e501041fa, n = 2097152, clLucas v1.0
0 err = 0.003784 (0:38 real, 3.7934 ms/iter, ETA 0:37)
Iteration 31870000 M( 31874257 )C, 0xa928d1f275659189, n = 2097152, clLucas v1.0
0 err = 0.003784 (0:38 real, 3.7974 ms/iter, ETA 0:00)
M( 31874257 )C, 0xc12e777157d7662d, n = 2097152, clLucas v1.00

So now the question is, how do we submit this back to mersenne.org, as it doesn't accept the clLucas line at the moment? Is there an equivalent format for CUDALucas that we could use, for example? My getwork/submission script is working otherwise...

Edit: submitted as CUDALucas :-/

[Bdot]

Hi,

I could not spend a lot of time, but a few observations:
(HD7850@1050MHz)

• -aggressive does not change iteration times a lot (if at all), it just consumes one CPU core.
• when continuing from a checkpoint, the first output line does not consider the real number of iterations, but instead calculates based on the reporting interval.
• the default 256 threads is too much - I'm much faster with 128, and yet a bit faster with 64.
• an -fftbench option would be good (or did I just not find it?)

[H:\software\clLucas]clLucas_x64.exe 3666666
...
Iteration 70000 M( 3666666 )C, 0x481c56c64dbd1f96, n = 196608, clLucas v1.00 err = 0.1191 (0:09 real, 0.8653 ms/iter, ETA 51:46)
Iteration 80000 M( 3666666 )C, 0xb2d63153bd472e1b, n = 196608, clLucas v1.00 err = 0.1191 (0:08 real, 0.8693 ms/iter, ETA 51:52)

[H:\software\clLucas]clLucas_x64.exe 1398296
...
Iteration 460000 M( 1398296 )C, 0x85705dae01da5965, n = 73728, clLucas v1.00 err = 0.1172 (0:04 real, 0.4200 ms/iter, ETA 6:30)
Iteration 470000 M( 1398296 )C, 0x1043ab81f5b11820, n = 73728, clLucas v1.00 err = 0.1172 (0:04 real, 0.4103 ms/iter, ETA 6:17)

[H:\software\clLucas]clLucas_x64.exe 36666666
...
Iteration 20000 M( 36666666 )C, 0x8c022e364d0eac22, n = 2097152, clLucas  v1.00 err = 0.08203 (1:28 real, 8.8128 ms/iter, ETA 89:41:41)
Iteration 30000 M( 36666666 )C, 0x581cb1c8d6065b84, n = 2097152, clLucas  v1.00 err = 0.08203 (1:28 real, 8.7944 ms/iter, ETA 89:28:57)

[H:\software\clLucas]clLucas_x64.exe -c 1000 57885161
...
Iteration 10000 M( 57885161 )C, 0x76c27556683cd84d, n = 3276800, clLucas  v1.00 err = 0.09766 (1:04 real, 64.3920 ms/iter, ETA 1035:11:28)
Iteration 11000 M( 57885161 )C, 0x837d0b600890ab5b, n = 3276800, clLucas  v1.00 err = 0.1309 (1:05 real, 64.9171 ms/iter, ETA 1043:36:54)

[H:\software\clLucas]clLucas_x64.exe -c 1000 -aggressive  57885161
...
Iteration 13000 M( 57885161 )C, 0xa2cef1503943d3bf, n = 3276800, clLucas  v1.00 err = 0.1094 (1:04 real, 64.0899 ms/iter, ETA 1030:16:49)
Iteration 14000 M( 57885161 )C, 0x947b2df0e7be2dea, n = 3276800, clLucas  v1.00 err = 0.1094 (1:05 real, 64.5999 ms/iter, ETA 1038:27:41)

[H:\software\clLucas]clLucas_x64.exe -c 1000 -aggressive -threads 128  57885161
...
Iteration 16000 M( 57885161 )C, 0xdcfa32ac6d8c107a, n = 3276800, clLucas  v1.00 err = 0.1016 (0:50 real, 49.6628 ms/iter, ETA 798:18:58)
Iteration 17000 M( 57885161 )C, 0x9451050ea89a2770, n = 3276800, clLucas  v1.00 err = 0.1035 (0:49 real, 49.6535 ms/iter, ETA 798:09:09)

[H:\software\clLucas]clLucas_x64.exe -c 1000 -threads 128  57885161
...
Iteration 22000 M( 57885161 )C, 0x0258edf2015b2203, n = 3276800, clLucas  v1.00 err = 0.1094 (0:50 real, 50.1777 ms/iter, ETA 806:30:32)
Iteration 23000 M( 57885161 )C, 0x0a0a2a85bde18654, n = 3276800, clLucas  v1.00 err = 0.1094 (0:50 real, 50.2124 ms/iter, ETA 807:03:10)

[H:\software\clLucas]clLucas_x64.exe -c 1000 -threads 64 -aggressive  57885161
...
Iteration 26000 M( 57885161 )C, 0x69cac186e70a4535, n = 3276800, clLucas  v1.00 err = 0.1094 (0:49 real, 48.6486 ms/iter, ETA 781:52:36)
Iteration 27000 M( 57885161 )C, 0x926467beb9cf48a7, n = 3276800, clLucas  v1.00 err = 0.1094 (0:49 real, 48.6174 ms/iter, ETA 781:21:43)

[H:\software\clLucas]clLucas_x64.exe -c 1000 -threads 64  57885161
...
Iteration 29000 M( 57885161 )C, 0x0d405d12a12d19f8, n = 3276800, clLucas  v1.00 err = 0.1094 (0:49 real, 49.0277 ms/iter, ETA 787:55:45)
Iteration 30000 M( 57885161 )C, 0xce0d85ab0065a232, n = 3276800, clLucas  v1.00 err = 0.1094 (0:49 real, 49.0770 ms/iter, ETA 788:42:29)

[H:\software\clLucas]clLucas_x64.exe -c 10000 -threads 64  1398269
Platform :Advanced Micro Devices, Inc.
Device 0 : Pitcairn


start M1398269 fft length = 73728
Iteration 10000 M( 1398269 )C, 0xa4a6d2f0e34629db, n = 73728, clLucas v1.00 err = 0.09766 (0:04 real, 0.4366 ms/iter, ETA 10:02)
Iteration 20000 M( 1398269 )C, 0x4483211b6f1ce250, n = 73728, clLucas v1.00 err = 0.09766 (0:05 real, 0.4315 ms/iter, ETA 9:51)
...
Iteration 1380000 M( 1398269 )C, 0x6650b95f00196d18, n = 73728, clLucas v1.00 err = 0.1123 (0:04 real, 0.4209 ms/iter, ETA 0:04)
Iteration 1390000 M( 1398269 )C, 0x554ae339bfea8fae, n = 73728, clLucas v1.00 err = 0.1123 (0:05 real, 0.4275 ms/iter, ETA 0:00)
M( 1398269 )P, n = 73728, clLucas v1.00

[H:\software\clLucas]clLucas_x64.exe -c 10000 -threads 64 -aggressive  -f 2097152 31874257
...
Iteration 20000 M( 31874257 )C, 0xe022bb72ce6cdb82, n = 2097152, clLucas  v1.00 err = 0.003357 (1:24 real, 8.3428 ms/iter, ETA 73:48:37)
Iteration 30000 M( 31874257 )C, 0xbdfe7180597003a5, n = 2097152, clLucas  v1.00 err = 0.003357 (1:23 real, 8.3476 ms/iter, ETA 73:49:48)

[kracker]

Also, I noticed I can not pass -threads and -f at the same time, it gives me "over specifications Grid = 65536"

[Robish]

Originally Posted by Robish

7870 any good to you?

Thanks Kracker

Tried 3666666 and got the following seems very FAST :-)

C:\Users\ati2\Desktop\clLucas_x64_test-2-1.0>clLucas_x64.exe 3666666
Platform :Advanced Micro Devices, Inc.
Device 0 : Pitcairn


start M3666666 fft length = 196608
Iteration 10000 M( 3666666 )C, 0xe86cc2b3a93ada07, n = 196608, clLucas v1.00 err
= 0.1191 (0:08 real, 0.8151 ms/iter, ETA 49:35)
Iteration 20000 M( 3666666 )C, 0x16c15dc858ec9064, n = 196608, clLucas v1.00 err
= 0.1191 (0:08 real, 0.8115 ms/iter, ETA 49:13)
Iteration 30000 M( 3666666 )C, 0xc6e1c0a2601d7f33, n = 196608, clLucas v1.00 err
= 0.1191 (0:09 real, 0.8118 ms/iter, ETA 49:06)
Iteration 40000 M( 3666666 )C, 0xb1b040472d81720a, n = 196608, clLucas v1.00 err
= 0.1191 (0:08 real, 0.8012 ms/iter, ETA 48:20)
Iteration 50000 M( 3666666 )C, 0xc6d33607654b3465, n = 196608, clLucas v1.00 err
= 0.1191 (0:08 real, 0.8087 ms/iter, ETA 48:39)
Iteration 60000 M( 3666666 )C, 0x7dd1b41fded4e79c, n = 196608, clLucas v1.00 err
= 0.1191 (0:08 real, 0.8070 ms/iter, ETA 48:25)
Iteration 70000 M( 3666666 )C, 0x481c56c64dbd1f96, n = 196608, clLucas v1.00 err
= 0.1191 (0:08 real, 0.8091 ms/iter, ETA 48:24)
Iteration 80000 M( 3666666 )C, 0xb2d63153bd472e1b, n = 196608, clLucas v1.00 err
= 0.1191 (0:08 real, 0.8095 ms/iter, ETA 48:17)
Iteration 90000 M( 3666666 )C, 0x4b22ed5b307efdca, n = 196608, clLucas v1.00 err
= 0.1191 (0:08 real, 0.8058 ms/iter, ETA 47:56)
Iteration 100000 M( 3666666 )C, 0xe0897573171fe6ae, n = 196608, clLucas v1.00 er
r = 0.1191 (0:08 real, 0.8076 ms/iter, ETA 47:55)
Iteration 110000 M( 3666666 )C, 0x1485e0f69f23750f, n = 196608, clLucas v1.00 er
r = 0.1191 (0:08 real, 0.8092 ms/iter, ETA 47:52)
Iteration 120000 M( 3666666 )C, 0x6c1b03db058b2ce1, n = 196608, clLucas v1.00 er
r = 0.1191 (0:08 real, 0.8098 ms/iter, ETA 47:46)
Iteration 130000 M( 3666666 )C, 0x2aa04a97b707fca3, n = 196608, clLucas v1.00 er
r = 0.1191 (0:08 real, 0.8108 ms/iter, ETA 47:42)
Iteration 140000 M( 3666666 )C, 0x5839fce56f4630ff, n = 196608, clLucas v1.00 er
r = 0.1191 (0:08 real, 0.8071 ms/iter, ETA 47:20)
Iteration 150000 M( 3666666 )C, 0x0fd38750d4621e4a, n = 196608, clLucas v1.00 er
r = 0.1191 (0:09 real, 0.8044 ms/iter, ETA 47:03)
Iteration 160000 M( 3666666 )C, 0x7a8c543b06082160, n = 196608, clLucas v1.00 er
r = 0.1191 (0:08 real, 0.8059 ms/iter, ETA 47:00)
Iteration 170000 M( 3666666 )C, 0x9c37e31d1f60c262, n = 196608, clLucas v1.00 er
r = 0.1191 (0:08 real, 0.8128 ms/iter, ETA 47:16)
Iteration 180000 M( 3666666 )C, 0x4a4a5a934fd4ea0c, n = 196608, clLucas v1.00 er
r = 0.1191 (0:08 real, 0.8082 ms/iter, ETA 46:52)
Iteration 190000 M( 3666666 )C, 0x260c6c3206fd92b1, n = 196608, clLucas v1.00 er
r = 0.1191 (0:08 real, 0.8052 ms/iter, ETA 46:34)
Iteration 200000 M( 3666666 )C, 0xb85f9849ab11bf89, n = 196608, clLucas v1.00 er
r = 0.1191 (0:08 real, 0.8087 ms/iter, ETA 46:38)
Iteration 210000 M( 3666666 )C, 0x4c1f2d4754b54d25, n = 196608, clLucas v1.00 er
r = 0.1191 (0:08 real, 0.8087 ms/iter, ETA 46:30)
Iteration 220000 M( 3666666 )C, 0xfa826e206a66edbf, n = 196608, clLucas v1.00 er
r = 0.1191 (0:08 real, 0.8118 ms/iter, ETA 46:32)
Iteration 230000 M( 3666666 )C, 0x649d5b98822df542, n = 196608, clLucas v1.00 er
r = 0.1191 (0:08 real, 0.8122 ms/iter, ETA 46:25)
Iteration 240000 M( 3666666 )C, 0xdb7a99437ae25bcc, n = 196608, clLucas v1.00 er
r = 0.1191 (0:08 real, 0.8051 ms/iter, ETA 45:53)
Iteration 250000 M( 3666666 )C, 0x2147b1a2e3d18cce, n = 196608, clLucas v1.00 er
r = 0.1191 (0:08 real, 0.8049 ms/iter, ETA 45:44)
Iteration 260000 M( 3666666 )C, 0x899266bd7f9907a8, n = 196608, clLucas v1.00 er
r = 0.1191 (0:09 real, 0.8079 ms/iter, ETA 45:46)
Iteration 270000 M( 3666666 )C, 0x026d37bfe64c3536, n = 196608, clLucas v1.00 er
r = 0.1191 (0:08 real, 0.8080 ms/iter, ETA 45:39)
Iteration 280000 M( 3666666 )C, 0x4e863fc47d380f42, n = 196608, clLucas v1.00 er
r = 0.1191 (0:08 real, 0.8128 ms/iter, ETA 45:47)
Iteration 290000 M( 3666666 )C, 0xe93f5eaff11d6056, n = 196608, clLucas v1.00 er
r = 0.1191 (0:08 real, 0.8124 ms/iter, ETA 45:37)
Iteration 300000 M( 3666666 )C, 0x3910168db3215a61, n = 196608, clLucas v1.00 er
r = 0.1191 (0:08 real, 0.8110 ms/iter, ETA 45:24)
Iteration 310000 M( 3666666 )C, 0x9e7d2b141aacfe11, n = 196608, clLucas v1.00 er
r = 0.1191 (0:08 real, 0.8322 ms/iter, ETA 46:27)
Iteration 320000 M( 3666666 )C, 0x866715d108a8c45c, n = 196608, clLucas v1.00 er
r = 0.1191 (0:08 real, 0.8114 ms/iter, ETA 45:10)

[TeknoHog]

Originally Posted by kracker

Also, I noticed I can not pass -threads and -f at the same time, it gives me "over specifications Grid = 65536"

This happened to me with -f 4194304 -threads 64, but the same FFT with -threads 128 seems OK.

[kracker]

Originally Posted by Robish

Thanks Kracker

Tried 3666666 and got the following seems very FAST :-)

C:\Users\ati2\Desktop\clLucas_x64_test-2-1.0>clLucas_x64.exe 3666666
Platform :Advanced Micro Devices, Inc.
Device 0 : Pitcairn


start M3666666 fft length = 196608
Iteration 10000 M( 3666666 )C, 0xe86cc2b3a93ada07, n = 196608, clLucas v1.00 err
= 0.1191 (0:08 real, 0.8151 ms/iter, ETA 49:35)

Actually, what msft wants is 36666666 not 3666666


Thanks

[Robish]

Originally Posted by kracker

Actually, what msft wants is 36666666 not 3666666


Thanks

oops ill get on it now , right after i find my glasses :-)

this however doesnt look so good ;-(


start M62868347 fft length = 3145728
err = 0.498047, increasing n from 3145728

start M62868347 fft length = 3276800
err = 0.362305, increasing n from 3276800

start M62868347 fft length = 3538944
^C caught. Writing checkpoint.

C:\Users\ati2\Desktop\clLucas_x64_test-2-1.0>clLucas_x64.exe 62868347
Platform :Advanced Micro Devices, Inc.
Device 0 : Pitcairn


continuing work from a partial result M62868347 fft length = 3538944 iteration =
8184
Iteration 10000 M( 62868347 )C, 0x2fead152a6afa7d8, n = 3538944, clLucas v1.00 e
rr = 0.1406 (1:38 real, 9.7877 ms/iter, ETA 170:52:39)
Iteration 20000 M( 62868347 )C, 0x06a9133da73deab9, n = 3538944, clLucas v1.00 e
rr = 0.1719 (8:58 real, 53.7943 ms/iter, ETA 939:00:31)
Iteration 30000 M( 62868347 )C, 0x130b4bbd5e6fd089, n = 3538944, clLucas v1.00 e
rr = 0.1719 (8:58 real, 53.7580 ms/iter, ETA 938:13:33)
Iteration 40000 M( 62868347 )C, 0x71bf6180dbb3ab34, n = 3538944, clLucas v1.00 e
rr = 0.1719 (8:58 real, 53.7607 ms/iter, ETA 938:07:26)
Iteration 50000 M( 62868347 )C, 0x16af1ab076be35d8, n = 3538944, clLucas v1.00 e
rr = 0.1719 (8:57 real, 53.7578 ms/iter, ETA 937:55:28)
Iteration 60000 M( 62868347 )C, 0x539c50f655ed11c2, n = 3538944, clLucas v1.00 e
rr = 0.1719 (8:58 real, 53.7378 ms/iter, ETA 937:25:35)
Iteration 70000 M( 62868347 )C, 0x772c2f498e3396bc, n = 3538944, clLucas v1.00 e
rr = 0.1719 (8:57 real, 53.7215 ms/iter, ETA 936:59:32)
Iteration 80000 M( 62868347 )C, 0xc92b41585ec2e013, n = 3538944, clLucas v1.00 e
rr = 0.1719 (8:57 real, 53.7171 ms/iter, ETA 936:45:59)
Iteration 90000 M( 62868347 )C, 0xc347a291759ed429, n = 3538944, clLucas v1.00 e
rr = 0.1758 (8:58 real, 53.7148 ms/iter, ETA 936:34:39)
Iteration 100000 M( 62868347 )C, 0xb2cc1c96baa5d5bb, n = 3538944, clLucas v1.00
err = 0.1758 (8:57 real, 53.7248 ms/iter, ETA 936:36:10)
Iteration 110000 M( 62868347 )C, 0x748fff43c0fc8da3, n = 3538944, clLucas v1.00
err = 0.1758 (8:57 real, 53.7437 ms/iter, ETA 936:47:00)
Iteration 120000 M( 62868347 )C, 0xb9f285166c1839aa, n = 3538944, clLucas v1.00
err = 0.1758 (8:58 real, 53.7087 ms/iter, ETA 936:01:26)
Iteration 130000 M( 62868347 )C, 0x86c68b8b33f4167c, n = 3538944, clLucas v1.00
err = 0.1758 (8:57 real, 53.7184 ms/iter, ETA 936:02:35)
Iteration 140000 M( 62868347 )C, 0xc9288f53d58e8ea3, n = 3538944, clLucas v1.00
err = 0.1758 (8:57 real, 53.7157 ms/iter, ETA 935:50:48)
Iteration 150000 M( 62868347 )C, 0xce673f8a45d7ddf5, n = 3538944, clLucas v1.00
err = 0.1758 (8:58 real, 53.7388 ms/iter, ETA 936:06:02)
Iteration 160000 M( 62868347 )C, 0xdcc765228dd55241, n = 3538944, clLucas v1.00
err = 0.1758 (8:57 real, 53.7086 ms/iter, ETA 935:25:26)
Iteration 170000 M( 62868347 )C, 0x1727f1a054a5f28a, n = 3538944, clLucas v1.00
err = 0.1758 (8:57 real, 53.7227 ms/iter, ETA 935:31:17)
Iteration 180000 M( 62868347 )C, 0xc8ff7d9c2748d4e9, n = 3538944, clLucas v1.00
err = 0.1758 (8:58 real, 53.7287 ms/iter, ETA 935:28:31)
Iteration 190000 M( 62868347 )C, 0xcd63ba84c503b2f9, n = 3538944, clLucas v1.00
err = 0.1758 (8:57 real, 53.7156 ms/iter, ETA 935:05:54)
Iteration 200000 M( 62868347 )C, 0x49631c026979a4e8, n = 3538944, clLucas v1.00
err = 0.1758 (8:57 real, 53.7132 ms/iter, ETA 934:54:31)
Iteration 210000 M( 62868347 )C, 0x7f9273f72f5e7f26, n = 3538944, clLucas v1.00
err = 0.1758 (8:57 real, 53.7303 ms/iter, ETA 935:03:25)
Iteration 220000 M( 62868347 )C, 0x45774e159ec4ccce, n = 3538944, clLucas v1.00
err = 0.1758 (8:57 real, 53.7227 ms/iter, ETA 934:46:30)
Iteration 230000 M( 62868347 )C, 0x1081c1ed5af77851, n = 3538944, clLucas v1.00
err = 0.1758 (8:57 real, 53.7038 ms/iter, ETA 934:17:46)
Iteration 240000 M( 62868347 )C, 0x5c2e90e0eb1c0010, n = 3538944, clLucas v1.00
err = 0.1758 (8:57 real, 53.7073 ms/iter, ETA 934:12:28)
Iteration 250000 M( 62868347 )C, 0xfeb5c371a9a3b534, n = 3538944, clLucas v1.00
err = 0.1758 (8:58 real, 53.7348 ms/iter, ETA 934:32:14)
Iteration 260000 M( 62868347 )C, 0x5d18321fe2ba3c1d, n = 3538944, clLucas v1.00
err = 0.1758 (8:57 real, 53.7017 ms/iter, ETA 933:48:45)

[kracker]

Try clLucas_x64.exe 36666666
.

[Robish]

Originally Posted by kracker

Try clLucas_x64.exe 36666666
.

C:\Users\ati2\Desktop\clLucas_x64_test-2-1.0>clLucas_x64.exe 36666666
Platform :Advanced Micro Devices, Inc.
Device 0 : Pitcairn


start M36666666 fft length = 1966080
err = 0.359375, increasing n from 1966080

start M36666666 fft length = 2097152
Iteration 10000 M( 36666666 )C, 0xded2eec2ad4c020b, n = 2097152, clLucas v1.00 e
rr = 0.08203 (1:25 real, 8.5243 ms/iter, ETA 86:46:55)
Iteration 20000 M( 36666666 )C, 0x8c022e364d0eac22, n = 2097152, clLucas v1.00 e
rr = 0.08203 (1:26 real, 8.5993 ms/iter, ETA 87:31:19)
Iteration 30000 M( 36666666 )C, 0x581cb1c8d6065b84, n = 2097152, clLucas v1.00 e
rr = 0.08203 (1:28 real, 8.7684 ms/iter, ETA 89:13:05)
Iteration 40000 M( 36666666 )C, 0x21b58443efd8f52f, n = 2097152, clLucas v1.00 e
rr = 0.08203 (1:28 real, 8.7668 ms/iter, ETA 89:10:41)
Iteration 50000 M( 36666666 )C, 0xa0dbfc5013b70951, n = 2097152, clLucas v1.00 e
rr = 0.08203 (1:28 real, 8.7679 ms/iter, ETA 89:09:51)

[sanaris]

Hey, I've got link errors at link stage when using MSVC2012 on Win8 x64 environment. AMD FFT was compiled separately and fed through include/lib. Project files were from matrix multiplication.

1>------ Build started: Project: clLucas, Configuration: Release x64 ------
1>  Kernels.cpp
1>Kernels.obj : error LNK2005: "unsigned __int64 plan" (?plan@@3_KA) already defined in clLucas.obj
1>Kernels.obj : error LNK2005: "class cl::Buffer g_inv2" (?g_inv2@@3VBuffer@cl@@A) already defined in clLucas.obj
1>Kernels.obj : error LNK2005: "class cl::Buffer g_ttmp3" (?g_ttmp3@@3VBuffer@cl@@A) already defined in clLucas.obj
1>Kernels.obj : error LNK2005: "class std::vector<class cl::Device,class std::allocator<class cl::Device> > m_devices" (?m_devices@@3V?$vector@VDevice@cl@@V?$allocator@VDevice@cl@@@std@@@std@@A) already defined in clLucas.obj
1>Kernels.obj : error LNK2005: "class cl::Buffer g_ttmp" (?g_ttmp@@3VBuffer@cl@@A) already defined in clLucas.obj
1>Kernels.obj : error LNK2005: "class cl::Buffer g_ttp" (?g_ttp@@3VBuffer@cl@@A) already defined in clLucas.obj
1>Kernels.obj : error LNK2005: "class cl::Buffer g_inv" (?g_inv@@3VBuffer@cl@@A) already defined in clLucas.obj
1>Kernels.obj : error LNK2005: "class cl::Context context" (?context@@3VContext@cl@@A) already defined in clLucas.obj
1>Kernels.obj : error LNK2005: "class cl::CommandQueue m_cmdQueue" (?m_cmdQueue@@3VCommandQueue@cl@@A) already defined in clLucas.obj
1>Kernels.obj : error LNK2005: "class cl::Buffer g_ttmpp" (?g_ttmpp@@3VBuffer@cl@@A) already defined in clLucas.obj
1>Kernels.obj : error LNK2005: "class cl::Context m_context" (?m_context@@3VContext@cl@@A) already defined in clLucas.obj
1>Kernels.obj : error LNK2005: "class cl::CommandQueue commandQueue" (?commandQueue@@3VCommandQueue@cl@@A) already defined in clLucas.obj
1>Kernels.obj : error LNK2005: "class cl::Program m_program" (?m_program@@3VProgram@cl@@A) already defined in clLucas.obj
1>Kernels.obj : error LNK2005: "class cl::Buffer g_inv3" (?g_inv3@@3VBuffer@cl@@A) already defined in clLucas.obj
1>Kernels.obj : error LNK2005: "class Lucas clLucas" (?clLucas@@3VLucas@@A) already defined in clLucas.obj
1>Kernels.obj : error LNK2005: "class cl::Buffer g_ttmp2" (?g_ttmp2@@3VBuffer@cl@@A) already defined in clLucas.obj
1>Kernels.obj : error LNK2005: "class cl::Buffer g_err" (?g_err@@3VBuffer@cl@@A) already defined in clLucas.obj
1>Kernels.obj : error LNK2005: "class cl::Buffer g_carry" (?g_carry@@3VBuffer@cl@@A) already defined in clLucas.obj
1>Kernels.obj : error LNK2005: "class cl::Buffer g_ttp3" (?g_ttp3@@3VBuffer@cl@@A) already defined in clLucas.obj
1>Kernels.obj : error LNK2005: "class cl::Buffer g_x" (?g_x@@3VBuffer@cl@@A) already defined in clLucas.obj
1>Kernels.obj : error LNK2005: "class cl::Buffer g_ttp2" (?g_ttp2@@3VBuffer@cl@@A) already defined in clLucas.obj
1>Kernels.obj : error LNK2005: "public: int __cdecl Lucas::fft_close(void)" (?fft_close@Lucas@@QEAAHXZ) already defined in clLucas.obj
1>Kernels.obj : error LNK2005: "public: int __cdecl Lucas::fft_setup(int)" (?fft_setup@Lucas@@QEAAHH@Z) already defined in clLucas.obj
1>Kernels.obj : error LNK2005: "class std::basic_string<char,struct std::char_traits<char>,class std::allocator<char> > __cdecl prettyPrintclFFTStatus(int const &)" (?prettyPrintclFFTStatus@@YA?AV?$basic_string@DU?$char_traits@D@std@@V?$allocator@D@2@@std@@AEBH@Z) already defined in clLucas.obj
1>Kernels.obj : error LNK2005: "public: int __cdecl Lucas::set_deviceId(int)" (?set_deviceId@Lucas@@QEAAHH@Z) already defined in clLucas.obj
1>Kernels.obj : error LNK2005: "public: int __cdecl Lucas::normalize2_runCLKernels(void)" (?normalize2_runCLKernels@Lucas@@QEAAHXZ) already defined in clLucas.obj
1>Kernels.obj : error LNK2005: "public: int __cdecl Lucas::normalize_runCLKernels(void)" (?normalize_runCLKernels@Lucas@@QEAAHXZ) already defined in clLucas.obj
1>Kernels.obj : error LNK2005: "public: virtual int __cdecl Lucas::setup(void)" (?setup@Lucas@@UEAAHXZ) already defined in clLucas.obj
1>Kernels.obj : error LNK2005: "public: int __cdecl Lucas::mul_runCLKernels(void)" (?mul_runCLKernels@Lucas@@QEAAHXZ) already defined in clLucas.obj
1>Kernels.obj : error LNK2005: "public: int __cdecl Lucas::setupCL(void)" (?setupCL@Lucas@@QEAAHXZ) already defined in clLucas.obj
1>timeval.obj : error LNK2005: "int __cdecl gettimeofday(struct timeval *,struct timezone *)" (?gettimeofday@@YAHPEAUtimeval@@PEAUtimezone@@@Z) already defined in clLucas.obj
1>C:\Users\yury\Documents\AMD APP\\samples\opencl\bin\release\x86_64\clLucas.exe : fatal error LNK1169: one or more multiply defined symbols found
========== Build: 0 succeeded, 1 failed, 0 up-to-date, 0 skipped ==========