CUDALucas (a.k.a. MaclucasFFTW/CUDA 2.3/CUFFTW) - Page 74

Brain · 2012-02-24, 15:17

Quote:

Originally Posted by msft

Ver 1.58
LaurV's Functional specification.

1.58 Win64, all untested:

CUDA 4.0 / SM 2.0
CUDA 4.1 / SM 2.0
CUDA 4.1 / SM 2.1

Interesting: CUDA 4.1 / SM 2.0 & SM 2.1 have same file size. I haven't made a diff. It may be that SM 2.1 is useless (I run a 560 Ti with CC2.1). Or I made a mistake. ;-)

bcp19 · 2012-02-24, 15:50

Quote:

Originally Posted by LaurV

You How can you be offended by a guy from the internet, which you don't know, and he doesn't know you for real? People tell a lot of sheesh...

<glances around the forum> Ummmmm, no comment.

flashjh · 2012-02-25, 04:23

Quote:

Originally Posted by LaurV

The first two DC's completed with CudaLucas1.49, drv4.0, cc2.0:

M( 26026433 )C, 0x457f73d49f90b822, n = 1572864, CUDALucas v1.49
M( 26176441 )C, 0x19283a19b247ba__, n = 1572864, CUDALucas v1.49

First is a match. Second is not (therefore I masked it). Both results coming from a gtx580 standard clock (no overclock this time). I am not going to repeat the second test as long as it is not confirmed as bad by a p95 run. After someone will clear the exponent, and if my test proved bad, I will repeat it to see if it come from the program.

edit: another small observation, the -c switch does not work for the screen, like in the older versions (I did not check if it really works for the checkpoint files too, but the screen effect is the first one observable) used to work. For -c30000, v1.49 still outputs to the screen every 10k iterations.

I finally got a match... I went back by going back to 1.49. Anyone gotten a match with a newer version?

Code:

 
M( 26026433 )C, 0x457f73d49f90b822, n = 1572864, CUDALucas v1.49

So now I'm going to try newer versions. Anyone who gets a match on another exponent, please post your results so I can use that to test also.

BTW - Here are the other failures I've had for M26026433:

Code:

 
M( 26026433 )C, 0x190df3dc67d21885, n = 1572864, CUDALucas v1.50 
M( 26026433 )C, 0xee1b55e2b3e0c8b5, n = 1572864, CUDALucas v1.50

EDIT: I'm running 1.49 and 1.58 (4.1 | 2.0) side-by-side on the same GPU to see how long the residues match... hopefully the whole time.

flashjh · 2012-02-25, 06:57

Quote:

Originally Posted by flashjh

EDIT: I'm running 1.49 and 1.58 (4.1 | 2.0) side-by-side on the same GPU to see how long the residues match... hopefully the whole time.

A ways in and they still both match... 1.58 is ahead now though, so it must be a little faster.

Brain · 2012-02-25, 10:51

One way to compile CUDALucas for Win64:
0. Have Win7 64 bit
1. Install Nvidia GPU Toolkit (e.g. version 4.1)
2. Install Nvidia GPU SDK (e.g. version 4.1)
3. Install Make for Windows
4. Install MS Visual Studio 2010 Professional Trial Edition (needed for 64bit, trial will not run out as only command line usage)
5. Set Path for nvcc, make and cl.exe (from VS/bin)
6. Edit given makefile for Win64: Adapt CUDA and SM parameter (e.g. 4.1/2.0). Rename it to makefile.
7. Enter "make" in console being in the CUDALucas/src directory.
8. Delete *.obj files.
9. Find the exe and be happy.

This should be it.

The day will come I won't be there to compile it. So a backup person/compiler will be needed. Any volunteers?

flashjh · 2012-02-25, 14:34

Quote:

Originally Posted by Brain

One way to compile CUDALucas for Win64:
0. Have Win7 64 bit
1. Install Nvidia GPU Toolkit (e.g. version 4.1)
2. Install Nvidia GPU SDK (e.g. version 4.1)
3. Install Make for Windows
4. Install MS Visual Studio 2010 Professional Trial Edition (needed for 64bit, trial will not run out as only command line usage)
5. Set Path for nvcc, make and cl.exe (from VS/bin)
6. Edit given makefile for Win64: Adapt CUDA and SM parameter (e.g. 4.1/2.0). Rename it to makefile.
7. Enter "make" in console being in the CUDALucas/src directory.
8. Delete *.obj files.
9. Find the exe and be happy.

This should be it.

The day will come I won't be there to compile it. So a backup person/compiler will be needed. Any volunteers?

I'll try to get it setup...

flashjh · 2012-02-26, 01:08

Here are my attempts at CUDA 1.58 4.1 sm_2.0 & sm_2.1

I can't make 4.0 yet because it's not installed. I tested briefly against 1.49 and got residue matches. Let me know if you test it.

Jerry

EDIT: I included my makefile. If you are trying to compile and have questions, let me know - I had quite a time getting this working so I can probably help!

flashjh · 2012-02-26, 04:23

Quote:

Originally Posted by Brain

One way to compile CUDALucas for Win64:
0. Have Win7 64 bit
1. Install Nvidia GPU Toolkit (e.g. version 4.1)
2. Install Nvidia GPU SDK (e.g. version 4.1)
3. Install Make for Windows
4. Install MS Visual Studio 2010 Professional Trial Edition (needed for 64bit, trial will not run out as only command line usage)
5. Set Path for nvcc, make and cl.exe (from VS/bin)
6. Edit given makefile for Win64: Adapt CUDA and SM parameter (e.g. 4.1/2.0). Rename it to makefile.
7. Enter "make" in console being in the CUDALucas/src directory.
8. Delete *.obj files.
9. Find the exe and be happy.

This should be it.

The day will come I won't be there to compile it. So a backup person/compiler will be needed. Any volunteers?

Brain,

When you compile, do you get any warnings?

I get the following:

Code:

CUDALucas.cu(524) : warning C4244: 'argument' : conversion from 'float' to 'size_t', possible loss of data
CUDALucas.cu(845) : warning C4018: '<' : signed/unsigned mismatch
CUDALucas.cu(1359) : warning C4018: '<' : signed/unsigned mismatch
CUDALucas.cu(1560) : warning C4018: '<' : signed/unsigned mismatch

So far the testing is going ok, but a lot more testing needs to be done.

I noticed an error during resume with 1.58. Is anyone else seeing problems with >=1.50 when resuming?

BTW - How did you do step 5 (from VS/bin)? Thanks.

msft · 2012-02-26, 06:51

Hi ,flashjh
I believe you can fix it.

Code:

  ip = (int *) malloc (((2 + sqrt ((float) n / 2)) * sizeof (int)));
-> ip = (int *) malloc (((size_t) (2 + sqrt ((float) n / 2)) * sizeof (int)));

  for (i = (STRIDE * STRIDE); i < (N + STRIDE * STRIDE);
->  for (i = (int) (STRIDE * STRIDE); i < (int) (N + STRIDE * STRIDE);

flashjh · 2012-02-26, 14:15

Quote:

Originally Posted by msft

Hi ,flashjh
I believe you can fix it.

Code:

  ip = (int *) malloc (((2 + sqrt ((float) n / 2)) * sizeof (int)));
-> ip = (int *) malloc (((size_t) (2 + sqrt ((float) n / 2)) * sizeof (int)));

  for (i = (STRIDE * STRIDE); i < (N + STRIDE * STRIDE);
->  for (i = (int) (STRIDE * STRIDE); i < (int) (N + STRIDE * STRIDE);

I'll change it later, thanks.

Brain · 2012-02-26, 16:17

Quote:

Originally Posted by flashjh

Brain,

When you compile, do you get any warnings?

I get the following:

Code:

CUDALucas.cu(524) : warning C4244: 'argument' : conversion from 'float' to 'size_t', possible loss of data
CUDALucas.cu(845) : warning C4018: '<' : signed/unsigned mismatch
CUDALucas.cu(1359) : warning C4018: '<' : signed/unsigned mismatch
CUDALucas.cu(1560) : warning C4018: '<' : signed/unsigned mismatch

I get the following:

Code:

------------------------
Compile output for 1.58
------------------------
"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v4.1/bin/nvcc" -c CUDALucas.cu -o CUDALucas.cuda4.1.sm_21.WIN64.obj -m64 --ptxas-options=-v "-ccbin=C:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\/bin" -DWIN64  -Xcompiler /EHsc,/W3,/nologo,/Ox,/Oy,/GL -arch=sm_21 -DMERS_PACKAGE -DBIT_SIEVE -DTESTING_SMALL_EXPONENTS -DSIEVE_SIZE_IN_BYTES=32 -DNUM_SMALL_PRIMES=32768 -DDO_NOT_USE_LONG_DOUBLE  "-IC:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v4.1/include" "-IC:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v4.1/include/cudart" "-IQ:\NVIDIA GPU Computing SDK 4.1/C/common/inc"  -D__x86_64__ -O3
tmpxft_00000ca8_00000000-14_CUDALucas.ii
CUDALucas.cu(524) : warning C4244: 'argument' : conversion from 'float' to 'size_t', possible loss of data
CUDALucas.cu(845) : warning C4018: '<' : signed/unsigned mismatch
CUDALucas.cu(1359) : warning C4018: '<' : signed/unsigned mismatch
CUDALucas.cu(1560) : warning C4018: '<' : signed/unsigned mismatch
cl /Ox /Oy /GL /W4 /fp:fast /nologo /c /Tp timeval.c /Fotimeval.WIN64.obj
timeval.c
link /nologo /LTCG  CUDALucas.cuda4.1.sm_21.WIN64.obj timeval.WIN64.obj "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v4.1/lib/x64/cudart.lib" "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v4.1/lib/x64/cufft.lib" /out:CUDALucas.cuda4.1.sm_21.WIN64.exe
Generating code
Finished generating code

Quote:

Originally Posted by flashjh

BTW - How did you do step 5 (from VS/bin)? Thanks.

Ehm, hard to remember. Nvidia tools automically set path, make was done manually, cl.exe probably also manually, not sure. Path is now:

Code:

------------------------
Path settings
------------------------
C:\Program Files (x86)\Microsoft Visual Studio  10.0\VC\BIN\amd64;C:\Windows\Microsoft.NET\Framework64\v4.0.30319;C:\Windows\Microsoft.NET\Framework64\v3.5;C:\Program  Files (x86)\Microsoft Visual Studio 10.0\VC\VCPackages;C:\Program Files  (x86)\Microsoft Visual Studio 10.0\Common7\IDE;C:\Program Files  (x86)\Microsoft Visual Studio 10.0\Common7\Tools;C:\Program Files  (x86)\HTML Help Workshop;C:\Program Files (x86)\Microsoft  SDKs\Windows\v7.0A\bin\NETFX 4.0 Tools\x64;C:\Program Files  (x86)\Microsoft SDKs\Windows\v7.0A\bin\x64;C:\Program Files  (x86)\Microsoft SDKs\Windows\v7.0A\bin;C:\Program Files (x86)\NVIDIA  Corporation\PhysX\Common;C:\Program Files\NVIDIA GPU Computing  Toolkit\CUDA\v4.1\\bin;C:\Program Files\NVIDIA GPU Computing  Toolkit\CUDA\v4.1\libnvvp\;C:\Program Files\Perl\site\bin;C:\Program  Files\Perl\bin;C:\Program Files\NVIDIA GPU Computing  Toolkit\CUDA\v4.0\bin\;C:\Program Files\Common Files\Microsoft  Shared\Windows Live;C:\Program Files (x86)\Common Files\Microsoft  Shared\Windows  Live;C:\Windows\system32;C:\Windows;C:\Windows\System32\Wbem;C:\Windows\System32\WindowsPowerShell\v1.0\;C:\Program  Files (x86)\Windows Live\Shared;C:\Program Files  (x86)\QuickTime\QTSystem\;C:\ProgramData\NVIDIA Corporation\NVIDIA GPU  Computing SDK 4.0\C\common\bin;C:\Program Files (x86)\Microsoft SQL  Server\100\Tools\Binn\;C:\Program Files\Microsoft SQL  Server\100\Tools\Binn\;C:\Program Files\Microsoft SQL  Server\100\DTS\Binn\;;C:\ProgramData\NVIDIA Corporation\NVIDIA GPU  Computing SDK 4.1\C\common\bin;C:\Program  Files\TortoiseSVN\bin;C:\Program Files (x86)\NVIDIA  Corporation\PhysX\Common;C:\Program Files\NVIDIA GPU Computing  Toolkit\CUDA\v4.1\\bin;C:\Program Files\NVIDIA GPU Computing  Toolkit\CUDA\v4.1\libnvvp\;C:\Program Files\Perl\site\bin;C:\Program  Files\Perl\bin;C:\Program Files\NVIDIA GPU Computing  Toolkit\CUDA\v4.0\bin\;C:\Program Files\Common Files\Microsoft  Shared\Windows Live;C:\Program Files (x86)\Common Files\Microsoft  Shared\Windows  Live;C:\Windows\system32;C:\Windows;C:\Windows\System32\Wbem;C:\Windows\System32\WindowsPowerShell\v1.0\;C:\Program  Files (x86)\Windows Live\Shared;C:\Program Files  (x86)\QuickTime\QTSystem\;C:\ProgramData\NVIDIA Corporation\NVIDIA GPU  Computing SDK 4.0\C\common\bin;C:\Program Files (x86)\Microsoft SQL  Server\100\Tools\Binn\;C:\Program Files\Microsoft SQL  Server\100\Tools\Binn\;C:\Program Files\Microsoft SQL  Server\100\DTS\Binn\;;C:\ProgramData\NVIDIA Corporation\NVIDIA GPU  Computing SDK 4.1\C\common\bin;C:\Program  Files\TortoiseSVN\bin;C:\Program Files (x86)\GnuWin32\bin;C:\Program  Files (x86)\Microsoft Visual Studio 10.0\VC\bin

Hard to read, I know. Me lazy.

2012-02-25, 10:51	#808
Brain Dec 2009 Peine, Germany 331 Posts	Backup compiler One way to compile CUDALucas for Win64: 0. Have Win7 64 bit 1. Install Nvidia GPU Toolkit (e.g. version 4.1) 2. Install Nvidia GPU SDK (e.g. version 4.1) 3. Install Make for Windows 4. Install MS Visual Studio 2010 Professional Trial Edition (needed for 64bit, trial will not run out as only command line usage) 5. Set Path for nvcc, make and cl.exe (from VS/bin) 6. Edit given makefile for Win64: Adapt CUDA and SM parameter (e.g. 4.1/2.0). Rename it to makefile. 7. Enter "make" in console being in the CUDALucas/src directory. 8. Delete *.obj files. 9. Find the exe and be happy. This should be it. The day will come I won't be there to compile it. So a backup person/compiler will be needed. Any volunteers?

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Don't DC/LL them with CudaLucas	LaurV	Data	131	2017-05-02 18:41
CUDALucas / cuFFT Performance on CUDA 7 / 7.5 / 8	Brain	GPU Computing	13	2016-02-19 15:53
CUDALucas: which binary to use?	Karl M Johnson	GPU Computing	15	2015-10-13 04:44
settings for cudaLucas	fairsky	GPU Computing	11	2013-11-03 02:08
Trying to run CUDALucas on Windows 8 CP	Rodrigo	GPU Computing	12	2012-03-07 23:20