mfakto: an OpenCL program for Mersenne prefactoring - Page 12

KingKurly · 2011-09-14, 01:40

I was able to test the Linux build as a success:

Code:

Selftest statistics
  number of tests           3332
  successfull tests         3332

selftest PASSED!


real    43m31.235s
user    9m22.360s
sys    31m10.151s

I did find it a bit strange that test cases 1552 through 1557 had no output to the terminal, but I will transition to using 0.08 now and I will begin submitting results to PrimeNet as they become available.

Do we need to redo "no factor" work that was done under 0.07 or can those be submitted? I have not downloaded the source to check a diff to see if it is reasonable thing to do.

Bdot · 2011-09-14, 08:53

Quote:

Originally Posted by KingKurly

I was able to test the Linux build as a success:

Code:

selftest PASSED!

Great to see!

Quote:

Originally Posted by KingKurly

I did find it a bit strange that test cases 1552 through 1557 had no output to the terminal, but I will transition to using 0.08 now and I will begin submitting results to PrimeNet as they become available.

The reason is, that test cases 1552 to 1557 are for factors of more than 91 bits. I had removed the 95-bit kernel because it was so terribly slow that you would not want to use it anyway. Therefore, mfakto currently only supports TF up to 91 bits.

Quote:

Originally Posted by KingKurly

Do we need to redo "no factor" work that was done under 0.07 or can those be submitted? I have not downloaded the source to check a diff to see if it is reasonable thing to do.

In version 0.07, the single-vectored MUL24 kernel did not work with Catalyst 11.8. In your self-compiled version you removed that kernel from the selftest, but not from the program. If you never changed the mfakto.ini-Parameter VectorSize (i.e. if you left it at 4), then that faulty kernel has not been used and you can submit the previous results without re-running them.

KingKurly · 2011-09-14, 14:50

Quote:

Originally Posted by Bdot

In version 0.07, the single-vectored MUL24 kernel did not work with Catalyst 11.8. In your self-compiled version you removed that kernel from the selftest, but not from the program. If you never changed the mfakto.ini-Parameter VectorSize (i.e. if you left it at 4), then that faulty kernel has not been used and you can submit the previous results without re-running them.

I have confirmed that the VectorSize never changed from 4. I am submitting the results, "closing the book" on 0.07, and moving to 0.08. Thank you so much!

Bdot · 2011-09-14, 16:03

Quote:

Originally Posted by KingKurly

moving to 0.08. Thank you so much!

Did you already check if mfakto_cl_barrett79 or mfakto_cl_71 are faster on your GPU? I'm really interested to see the two compared on different GPUs. On my HD5770, barrett is about 10% faster ...

Razor_FX_II · 2011-09-14, 16:28

Quote:

Originally Posted by Bdot

Did you already check if mfakto_cl_barrett79 or mfakto_cl_71 are faster on your GPU? I'm really interested to see the two compared on different GPUs. On my HD5770, barrett is about 10% faster ...

Using mfakto-0.08 on my HD4870's and HD4890's mfakto_cl_barrett79 is about 10% faster.
mfakto_cl_barrett79 avg rate: 55M/s
mfakto_cl_71 avg rate: 50M/s

KingKurly · 2011-09-14, 17:53

Quote:

Originally Posted by Bdot

Did you already check if mfakto_cl_barrett79 or mfakto_cl_71 are faster on your GPU? I'm really interested to see the two compared on different GPUs. On my HD5770, barrett is about 10% faster ...

I am finding similar results. My HD5450 seems to do about 8.6M/s on the mfakto_cl_71 and about 9.1M/s on the mfakto_cl_barrett79, doing TF on M41774351 from 68 to 69.

James Heinrich · 2011-09-14, 23:13

Since it appeared to be missing, I've created a stub article on MersenneWiki for mfakto:
http://www.mersennewiki.org/index.php/Mfakto

But since I don't actually use mfakto, perhaps someone else could fill in and fix all the details in the article.

DigiK-oz · 2011-09-20, 15:33

In the GPUGRID forum :

there's a bug in the latest sdk that makes a full use of a cpu-core whenever an opencl app is running.
They promised a fix, but still not here in 11.8
maybe in 11.9??

Maybe mfakto suffers from this as well? One of the threads using 100% of one cpu happens to be in the ATI libs....

Bdot · 2011-09-20, 18:23

Quote:

Originally Posted by DigiK-oz

In the GPUGRID forum :

there's a bug in the latest sdk that makes a full use of a cpu-core whenever an opencl app is running.
They promised a fix, but still not here in 11.8
maybe in 11.9??

Maybe mfakto suffers from this as well? One of the threads using 100% of one cpu happens to be in the ATI libs....

They seem to have implemented some kind of busy-wait (futex-based) whenever something needs to be synchronized with the GPU. As this is usually the CPU just waiting for the GPU to complete something, that is a total waste of CPU resources.

However, mfakto is not hit that badly as mfakto passes the prepared factor candidates to the GPU but does not wait for the results immediately. Instead, the next block of factor candidates is prepared on the CPU. Only when the CPU is faster preparing the stuff than the GPU can process it, then mfakto will synchronize with the GPU. And of course at the end of a class.

So yes, mfakto will also consume a full CPU core, but it will do something useful most of that time.

Samoflan · 2011-09-20, 18:33

Quote:

Originally Posted by Razor_FX_II

Using mfakto-0.08 on my HD4870's and HD4890's mfakto_cl_barrett79 is about 10% faster.
mfakto_cl_barrett79 avg rate: 55M/s
mfakto_cl_71 avg rate: 50M/s

I get similar results on my HD4890

mfakto_cl_barrett79 avg rate: 51.9M/s
mfakto_cl_71 avg rate: 48.7M/s

GPU load is 91-95%
CPU load will almost cap out 2 cores on my Phenom II x4 955

Bdot · 2011-09-26, 19:14

I´m sorry to report: yesterday I found a bug, mfakto up to 0.08 does not find the factor for k=3 for M6599953.

The reason is an invalid "optimization" that I made over the mfaktc-code. Mfaktc does not have this problem. I have fixed the bug and added a test case for it to the selftests.

The mfakto kernel "mfakto_cl_71" (all vector sizes) sometimes calculated a bad modulus when the factor candidate was <2⁴⁸. Smaller FCs (~2²⁴) had a higher chance for the error to occur, FCs >2⁴⁸ were always calculated correctly. The problem does not depend on the exponent size.

I´m sorry for possibly having wasted effort and resources, but I hope it´s not too many tests that need to be repeated as it´s only about small FCs. I will provide a fixed version within the next few days.

2011-09-14, 01:40	#122
KingKurly Sep 2010 Annapolis, MD, USA BD₁₆ Posts	I was able to test the Linux build as a success: Code: Selftest statistics number of tests 3332 successfull tests 3332 selftest PASSED! real 43m31.235s user 9m22.360s sys 31m10.151s I did find it a bit strange that test cases 1552 through 1557 had no output to the terminal, but I will transition to using 0.08 now and I will begin submitting results to PrimeNet as they become available. Do we need to redo "no factor" work that was done under 0.07 or can those be submitted? I have not downloaded the source to check a diff to see if it is reasonable thing to do.

2011-09-26, 19:14	#132
Bdot Nov 2010 Germany 3·199 Posts	Bug warning I´m sorry to report: yesterday I found a bug, mfakto up to 0.08 does not find the factor for k=3 for M6599953. The reason is an invalid "optimization" that I made over the mfaktc-code. Mfaktc does not have this problem. I have fixed the bug and added a test case for it to the selftests. The mfakto kernel "mfakto_cl_71" (all vector sizes) sometimes calculated a bad modulus when the factor candidate was <2⁴⁸. Smaller FCs (~2²⁴) had a higher chance for the error to occur, FCs >2⁴⁸ were always calculated correctly. The problem does not depend on the exponent size. I´m sorry for possibly having wasted effort and resources, but I hope it´s not too many tests that need to be repeated as it´s only about small FCs. I will provide a fixed version within the next few days.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
gpuOwL: an OpenCL program for Mersenne primality testing	preda	GpuOwl	2760	2022-05-15 00:00
mfaktc: a CUDA program for Mersenne prefactoring	TheJudger	GPU Computing	3541	2022-04-21 22:37
LL with OpenCL	msft	GPU Computing	433	2019-06-23 21:11
OpenCL for FPGAs	TObject	GPU Computing	2	2013-10-12 21:09
Program to TF Mersenne numbers with more than 1 sextillion digits?	Stargate38	Factoring	24	2011-11-03 00:34

2011-09-14, 23:13	#128
James Heinrich "James Heinrich" May 2004 ex-Northern Ontario 3²×7×59 Posts	Since it appeared to be missing, I've created a stub article on MersenneWiki for mfakto: http://www.mersennewiki.org/index.php/Mfakto But since I don't actually use mfakto, perhaps someone else could fill in and fix all the details in the article.

2011-09-20, 15:33	#129
DigiK-oz Jul 2008 30₈ Posts	In the GPUGRID forum : there's a bug in the latest sdk that makes a full use of a cpu-core whenever an opencl app is running. They promised a fix, but still not here in 11.8 maybe in 11.9?? Maybe mfakto suffers from this as well? One of the threads using 100% of one cpu happens to be in the ATI libs....