Join PrimeGrid
Returning Participants
Community
Leader Boards
Results
Other
drummers-lowrise
|
Message boards :
Generalized Fermat Prime Search :
Genefer 3.2.8 testing
| Author |
Message |
|---|
|
|
|
Hi all,
A new release of Genefer is ready for testing, this includes two small bug-fixes/changes to the OpenCL transform and to the 32-bit x87 transform. Due to the limited changes, we will only do a reduced test compared with 3.2.6 (the last major release), concentrating only on the x87 and OpenCL transforms, both 32 and 64 bit binaries. We will also test that the new binaries work correctly with both PRPNet and
To participate in testing, please visit the Google sheet below and reserve tests - I won't always be available during the next few weeks so please keep the testing matrix up-to-date as best you can:
https://docs.google.com/spreadsheets/d/1kS5EWl74GZ2Hb5JdfRWu0EL-Rla2yqhZvgBVrzwxgpY/edit?usp=sharing
Binaries are available from SVN:
Mac: https://www.assembla.com/code/genefer/subversion/nodes/829/trunk/bin/mac
Windows: https://www.assembla.com/code/genefer/subversion/nodes/829/trunk/bin/windows
Linux: https://www.assembla.com/code/genefer/subversion/nodes/829/trunk/bin/linux
Any questions or problems, please post in this thread.
Thanks in advance for the help!
- Iain
____________
Twitter: IainBethune
Proud member of team "Aggie The Pew". Go Aggie!
3073428256125*2^1290000-1 is Prime! | |
|
 Artist Volunteer tester
Send message
Joined: 29 Sep 08
Posts: 88
ID: 29825
Credit: 413,046,355
RAC: 75,825
                         
|
|
Linux results for 157476^65536+1
$ ./genefer_linux64 -x x87 -q '157476^65536+1'
Command line: ./genefer_linux64 -x x87 -q 157476^65536+1
genefer 3.2.8-dev (Linux/CPU/64-bit)
Supported transform implementations: fma3 avx-intel sse4 sse2 default x87
Copyright 2001-2015, Yves Gallot
Copyright 2009, Mark Rodenkirch, David Underbakke
Copyright 2010-2012, Shoichiro Yamada, Ken Brazier
Copyright 2011-2015, Iain Bethune, Michael Goetz, Ronald Schneider
Genefer is free source code, under the MIT license.
Priority change succeeded.
Testing 157476^65536+1...
Using x87 (80-bit) transform
Resuming 157476^65536+1 from a checkpoint (1087894 iterations left)
Estimated time remaining for 157476^65536+1 is 0:53:04
157476^65536+1 is a probable composite. (RES=9f64b3f0d545615c) (340605 digits) (err = 0.0000) (time = 0:57:06) 23:51:27
$ ../genefer_linux32 -x x87 -q '157476^65536+1'
genefer 3.2.8-dev (Linux/CPU/32-bit)
Supported transform implementations: fma3 avx-intel sse4 sse2 default x87
Copyright 2001-2015, Yves Gallot
Copyright 2009, Mark Rodenkirch, David Underbakke
Copyright 2010-2012, Shoichiro Yamada, Ken Brazier
Copyright 2011-2015, Iain Bethune, Michael Goetz, Ronald Schneider
Genefer is free source code, under the MIT license.
Command line: ../genefer_linux32 -x x87 -q 157476^65536+1
Priority change succeeded.
Testing 157476^65536+1...
Using x87 (80-bit) transform
Starting initialization...
Initialization complete (0.080 seconds).
Estimated time remaining for 157476^65536+1 is 0:59:01
157476^65536+1 is a probable composite. (RES=9f64b3f0d545615c) (340605 digits) (err = 0.0000) (time = 0:59:31) 00:06:55
____________
144052 *5^2018290+1 is Prime! | |
|
|
|
|
Going to go to my linux box and try to get a GPU working in it, then run the tests, if I'm taking too long, feel free to replace my name(although I don't think it should take terribly long. | |
|
|
|
|
Thanks Andrew, is it possible for you to run the 32-bit OpenCL version too?
- Iain
Going to go to my linux box and try to get a GPU working in it, then run the tests, if I'm taking too long, feel free to replace my name(although I don't think it should take terribly long.
____________
Twitter: IainBethune
Proud member of team "Aggie The Pew". Go Aggie!
3073428256125*2^1290000-1 is Prime! | |
|
|
|
Thanks Andrew, is it possible for you to run the 32-bit OpenCL version too?
- Iain
Going to go to my linux box and try to get a GPU working in it, then run the tests, if I'm taking too long, feel free to replace my name(although I don't think it should take terribly long.
If it's still available I'll run it right now | |
|
|
|
Thanks Andrew, is it possible for you to run the 32-bit OpenCL version too?
- Iain
Going to go to my linux box and try to get a GPU working in it, then run the tests, if I'm taking too long, feel free to replace my name(although I don't think it should take terribly long.
If it's still available I'll run it right now
Not sure if it is because I am using a 64 bit distro, but geneferocl_linux32 finds no openCL devices, under any test.
Not sure if I should mark it as failed, but for me it was failed. Maybe someone with a 32 bit system could check it again to make sure the error persists.
An error (7) occured.
(just to try to keep from nitpicking too much, but occurred should have two r's) | |
|
|
|
Thanks Andrew, is it possible for you to run the 32-bit OpenCL version too?
- Iain
Going to go to my linux box and try to get a GPU working in it, then run the tests, if I'm taking too long, feel free to replace my name(although I don't think it should take terribly long.
If it's still available I'll run it right now
Not sure if it is because I am using a 64 bit distro, but geneferocl_linux32 finds no openCL devices, under any test.
Not sure if I should mark it as failed, but for me it was failed. Maybe someone with a 32 bit system could check it again to make sure the error persists.
An error (7) occured.
(just to try to keep from nitpicking too much, but occurred should have two r's)
I can confirm the error on 32-bit Mint, but I'm thinking it's driver related over program related. I'm working on getting it going now but I have to leave for work soon so don't expect quick results.
____________
Eating more cheese on Thursdays. | |
|
|
|
Not sure if I should mark it as failed, but for me it was failed. Maybe someone with a 32 bit system could check it again to make sure the error persists.
I can confirm the error on 32-bit Mint, but I'm thinking it's driver related over program related. I'm working on getting it going now but I have to leave for work soon so don't expect quick results.
Thanks for the reports - I have asked Ron who built the binary to test if it works on his machines, but if you have success please let me know - nothing has changed in the code or the build config/libraries that would affect device discovery, so most likely it is a driver issue - I'm not 100% sure what exactly is needed to run a 32-bit OpenCL binary...
____________
Twitter: IainBethune
Proud member of team "Aggie The Pew". Go Aggie!
3073428256125*2^1290000-1 is Prime! | |
|
rroonnaallddVolunteer developer
Volunteer tester
Send message
Joined: 3 Jul 09
Posts: 1213
ID: 42893
Credit: 34,634,263
RAC: 0
                
|
|
Okay. I made some tests and here are the results:
boinc@vmware2k-3:~/Cuda/Iain_Bethune-GENEFER/trunk/bin/linux$ ldd geneferocl*
geneferocl_linux32:
linux-gate.so.1 => (0xf77bd000)
libOpenCL.so.1 => /usr/lib/i386-linux-gnu/libOpenCL.so.1 (0xf779a000)
libstdc++.so.6 => /usr/lib32/libstdc++.so.6 (0xf76b2000)
libpthread.so.0 => /lib/i386-linux-gnu/libpthread.so.0 (0xf7695000)
libm.so.6 => /lib/i386-linux-gnu/libm.so.6 (0xf764f000)
libgcc_s.so.1 => /lib/i386-linux-gnu/libgcc_s.so.1 (0xf7632000)
libc.so.6 => /lib/i386-linux-gnu/libc.so.6 (0xf7484000)
libdl.so.2 => /lib/i386-linux-gnu/libdl.so.2 (0xf747f000)
/lib/ld-linux.so.2 (0xf77be000)
geneferocl_linux64:
linux-vdso.so.1 => (0x00007ffdb2558000)
libOpenCL.so.1 => /usr/lib/x86_64-linux-gnu/libOpenCL.so.1 (0x00007f4838528000)
libstdc++.so.6 => /usr/lib/x86_64-linux-gnu/libstdc++.so.6 (0x00007f4838224000)
libpthread.so.0 => /lib/x86_64-linux-gnu/libpthread.so.0 (0x00007f4838006000)
libm.so.6 => /lib/x86_64-linux-gnu/libm.so.6 (0x00007f4837d00000)
libgcc_s.so.1 => /lib/x86_64-linux-gnu/libgcc_s.so.1 (0x00007f4837aea000)
libc.so.6 => /lib/x86_64-linux-gnu/libc.so.6 (0x00007f4837725000)
libdl.so.2 => /lib/x86_64-linux-gnu/libdl.so.2 (0x00007f4837521000)
/lib64/ld-linux-x86-64.so.2 (0x00007f4838731000)
################################################
boinc@vmware2k-3:~/Cuda/Iain_Bethune-GENEFER/trunk/bin/linux$ ./geneferocl_linux32
geneferocl 3.2.8-dev (Linux/OpenCL/32-bit)
Copyright 2001-2015, Yves Gallot
Copyright 2009, Mark Rodenkirch, David Underbakke
Copyright 2010-2012, Shoichiro Yamada, Ken Brazier
Copyright 2011-2015, Iain Bethune, Michael Goetz, Ronald Schneider
Genefer is free source code, under the MIT license.
Command line: ./geneferocl_linux32
Priority change succeeded.
Usage: geneferocl [options] Options may be specified in any order
Options: -b run benchmarks
-b3 run a series of benchmarks to predict actual run times
-t run tests of known prime GFNs
-r run residue test on known composite GFNs
-l compute approximate usable upper limit of b at
each N
-q "b^N+1" test quick expression
-d N or --device N set device number=N(default 0)
-nvidia run opencl only on NVIDIA GPU
-amd or -ati run opencl only on AMD/ATI GPU
<filename> test GFNs in <filename>, one GFN per line,
in the format b N
-v or -V print the startup banner and immediately exit
-boinc operate as a BOINC client app
No options were specified, using interactive mode:
1. bench
2. test
3. test residue
4. find limits
5. normal
1
Running on platform 'NVIDIA CUDA', device 'GeForce GTS 450', version 'OpenCL 1.1 CUDA' and driver '331.113'.
Priority change failed (needs superuser privileges).
Generalized Fermat Number Bench
Running benchmarks for transform implementation "OCL"
2199064^8192+1 Time: 126 us/mul. Err: 0.2031 51956 digits
1798620^16384+1 Time: 175 us/mul. Err: 0.1875 102481 digits
1471094^32768+1 Time: 181 us/mul. Err: 0.1875 202102 digits
1203210^65536+1 Time: 374 us/mul. Err: 0.1719 398482 digits
984108^131072+1 Time: 760 us/mul. Err: 0.1797 785521 digits
804904^262144+1 Time: 1.57 ms/mul. Err: 0.1719 1548156 digits
658332^524288+1 Time: 3.03 ms/mul. Err: 0.1875 3050541 digits
538452^1048576+1 Time: 6.41 ms/mul. Err: 0.1680 6009544 digits
440400^2097152+1 Time: 13.3 ms/mul. Err: 0.1719 11836006 digits
360204^4194304+1 Time: 28.2 ms/mul. Err: 0.1582 23305854 digits
294612^8388608+1 Time: 61.9 ms/mul. Err: 0.1641 45879398 digits
Genefer Mark = 15.
Priority change succeeded.
################################################
boinc@vmware2k-3:~/Cuda/Iain_Bethune-GENEFER/trunk/bin/linux$ ./geneferocl_linux32 -q '157476^65536+1'
geneferocl 3.2.8-dev (Linux/OpenCL/32-bit)
Copyright 2001-2015, Yves Gallot
Copyright 2009, Mark Rodenkirch, David Underbakke
Copyright 2010-2012, Shoichiro Yamada, Ken Brazier
Copyright 2011-2015, Iain Bethune, Michael Goetz, Ronald Schneider
Genefer is free source code, under the MIT license.
Command line: ./geneferocl_linux32 -q 157476^65536+1
Priority change succeeded.
Testing 157476^65536+1...
Using OCL transform
Running on platform 'NVIDIA CUDA', device 'GeForce GTS 450', version 'OpenCL 1.1 CUDA' and driver '331.113'.
Starting initialization...
Initialization complete (0.149 seconds).
Estimated time remaining for 157476^65536+1 is 0:07:02
157476^65536+1 is a probable composite. (RES=9f64b3f0d545615c) (340605 digits) (err = 0.0037) (time = 0:07:04) 23:17:16
____________
Best wishes. Knowledge is power. by jjwhalen
 | |
|
|
|
|
One common difficulty with all builds I have used are tha they require glibc 2.14 on linux. Scientific/Centos 6/some other REHL distros only support 2.12 as of now, what would you guys recommend when it comes to this? | |
|
rroonnaallddVolunteer developer
Volunteer tester
Send message
Joined: 3 Jul 09
Posts: 1213
ID: 42893
Credit: 34,634,263
RAC: 0
|
|
I get no errors after updating the nvidia driver to 346:
boinc@vmware2k-3:~/Cuda/Iain_Bethune-GENEFER/trunk/bin/linux/neu$ ./geneferocl_linux32 -q '157476^65536+1'
geneferocl 3.2.8-dev (Linux/OpenCL/32-bit)
Copyright 2001-2015, Yves Gallot
Copyright 2009, Mark Rodenkirch, David Underbakke
Copyright 2010-2012, Shoichiro Yamada, Ken Brazier
Copyright 2011-2015, Iain Bethune, Michael Goetz, Ronald Schneider
Genefer is free source code, under the MIT license.
Command line: ./geneferocl_linux32 -q 157476^65536+1
Priority change succeeded.
Testing 157476^65536+1...
Using OCL transform
Running on platform 'NVIDIA CUDA', device 'GeForce GTS 450', version 'OpenCL 1.1 CUDA' and driver '346.72'.
Starting initialization...
Initialization complete (0.148 seconds).
Estimated time remaining for 157476^65536+1 is 0:06:45
157476^65536+1 is a probable composite. (RES=9f64b3f0d545615c) (340605 digits) (err = 0.0037) (time = 0:06:46) 17:17:40
boinc@vmware2k-3:~/Cuda/Iain_Bethune-GENEFER/trunk/bin/linux/neu$ ./geneferocl_linux64 -q '157476^65536+1'
geneferocl 3.2.8-dev (Linux/OpenCL/64-bit)
Copyright 2001-2015, Yves Gallot
Copyright 2009, Mark Rodenkirch, David Underbakke
Copyright 2010-2012, Shoichiro Yamada, Ken Brazier
Copyright 2011-2015, Iain Bethune, Michael Goetz, Ronald Schneider
Genefer is free source code, under the MIT license.
Command line: ./geneferocl_linux64 -q 157476^65536+1
Priority change succeeded.
Testing 157476^65536+1...
Using OCL transform
Running on platform 'NVIDIA CUDA', device 'GeForce GTS 450', version 'OpenCL 1.1 CUDA' and driver '346.72'.
Starting initialization...
Initialization complete (0.080 seconds).
Estimated time remaining for 157476^65536+1 is 0:06:41
157476^65536+1 is a probable composite. (RES=9f64b3f0d545615c) (340605 digits) (err = 0.0038) (time = 0:06:44) 17:31:12
#################################################
boinc@vmware2k-3:~/Cuda/Iain_Bethune-GENEFER/trunk/bin/linux/neu$ ldd geneferocl*
geneferocl_linux32:
linux-gate.so.1 => (0xf7786000)
libOpenCL.so.1 => /usr/lib/i386-linux-gnu/libOpenCL.so.1 (0xf7762000)
libstdc++.so.6 => /usr/lib32/libstdc++.so.6 (0xf767a000)
libpthread.so.0 => /lib/i386-linux-gnu/libpthread.so.0 (0xf765d000)
libm.so.6 => /lib/i386-linux-gnu/libm.so.6 (0xf7617000)
libgcc_s.so.1 => /lib/i386-linux-gnu/libgcc_s.so.1 (0xf75fa000)
libc.so.6 => /lib/i386-linux-gnu/libc.so.6 (0xf744c000)
libdl.so.2 => /lib/i386-linux-gnu/libdl.so.2 (0xf7447000)
/lib/ld-linux.so.2 (0xf7787000)
geneferocl_linux64:
linux-vdso.so.1 => (0x00007fff2e5f4000)
libOpenCL.so.1 => /usr/lib/x86_64-linux-gnu/libOpenCL.so.1 (0x00007ff1c331e000)
libstdc++.so.6 => /usr/lib/x86_64-linux-gnu/libstdc++.so.6 (0x00007ff1c301a000)
libpthread.so.0 => /lib/x86_64-linux-gnu/libpthread.so.0 (0x00007ff1c2dfc000)
libm.so.6 => /lib/x86_64-linux-gnu/libm.so.6 (0x00007ff1c2af6000)
libgcc_s.so.1 => /lib/x86_64-linux-gnu/libgcc_s.so.1 (0x00007ff1c28e0000)
libc.so.6 => /lib/x86_64-linux-gnu/libc.so.6 (0x00007ff1c251b000)
libdl.so.2 => /lib/x86_64-linux-gnu/libdl.so.2 (0x00007ff1c2317000)
/lib64/ld-linux-x86-64.so.2 (0x00007ff1c3523000)
____________
Best wishes. Knowledge is power. by jjwhalen
| |
|
rroonnaallddVolunteer developer
Volunteer tester
Send message
Joined: 3 Jul 09
Posts: 1213
ID: 42893
Credit: 34,634,263
RAC: 0
|
One common difficulty with all builds I have used are tha they require glibc 2.14 on linux. Scientific/Centos 6/some other REHL distros only support 2.12 as of now, what would you guys recommend when it comes to this?
I have no idea how to solve this. Recompiling the apps on your side is no solution because then you have to use the app_info.xml-construct. The last time i got this problem, i simply upgraded my entire linux installation and according to Glibc 2.14 needed is the best solution the upgrade to CentOS7.
This is all what i can say for the moment.
____________
Best wishes. Knowledge is power. by jjwhalen
| |
|
|
|
|
Installed a different distro (Ubuntu 15.04) and got the OpenCL part of the drivers working, so the 32-bit standalone testing completed just fine. Probably could have gotten them running right in Mint, but waiting 30-45 minutes for apt to update during each attempt wasn't cool.
BOINC testing will take longer...the no GPU support default is proving hard to change. Maybe I can get it working this weekend, and if so, I'll install 64 bit and do that one, too. I originally tried to fool BOINC into thinking the app is CPU and letting the program itself handle it, but it just led to 10 minute long error 7 (no OpenCL device) timeouts.
____________
Eating more cheese on Thursdays. | |
|
|
|
|
Thanks for all the testing so far. There are now 3 tests outstanding:
x87 on 64-bit Linux (Started by Andrew Hughes)
OpenCL BOINC on 32-bit Linux (started by Grebuloner)
OpenCL BOINC on 64-bit Linux (not started)
Please update the google sheet if these are in fact now complete. Andrew, are you able to do the last one, since you have 64-bit OpenCL set up already for the standalone tests? I can help with the app_info.xml if needed.
- Iain
____________
Twitter: IainBethune
Proud member of team "Aggie The Pew". Go Aggie!
3073428256125*2^1290000-1 is Prime! | |
|
288larssonVolunteer tester
Send message
Joined: 17 Apr 10
Posts: 136
ID: 58815
Credit: 5,966,817,125
RAC: 3,271,154
                              
|
|
Hi OpenCL BOINC on 64-bit Linux ( done) | |
|
|
|
Hi OpenCL BOINC on 64-bit Linux ( done)
Thanks!
____________
Twitter: IainBethune
Proud member of team "Aggie The Pew". Go Aggie!
3073428256125*2^1290000-1 is Prime! | |
|
|
|
Thanks for all the testing so far. There are now 3 tests outstanding:
x87 on 64-bit Linux (Started by Andrew Hughes)
OpenCL BOINC on 32-bit Linux (started by Grebuloner)
OpenCL BOINC on 64-bit Linux (not started)
Please update the google sheet if these are in fact now complete. Andrew, are you able to do the last one, since you have 64-bit OpenCL set up already for the standalone tests? I can help with the app_info.xml if needed.
- Iain
Sorry about being late here, there was a fire in the building(someone downstairs had a meth lab -_-) and thanks larsson for finishing it! | |
|
|
|
Thanks for all the testing so far. There are now 3 tests outstanding:
x87 on 64-bit Linux (Started by Andrew Hughes)
OpenCL BOINC on 32-bit Linux (started by Grebuloner)
OpenCL BOINC on 64-bit Linux (not started)
Please update the google sheet if these are in fact now complete. Andrew, are you able to do the last one, since you have 64-bit OpenCL set up already for the standalone tests? I can help with the app_info.xml if needed.
- Iain
Sorry about being late here, there was a fire in the building(someone downstairs had a meth lab -_-) and thanks larsson for finishing it!
Best excuse I've heard in a while! Did you get the long x87 test completed?
- Iain
____________
Twitter: IainBethune
Proud member of team "Aggie The Pew". Go Aggie!
3073428256125*2^1290000-1 is Prime! | |
|
|
|
Thanks for all the testing so far. There are now 3 tests outstanding:
x87 on 64-bit Linux (Started by Andrew Hughes)
OpenCL BOINC on 32-bit Linux (started by Grebuloner)
OpenCL BOINC on 64-bit Linux (not started)
Please update the google sheet if these are in fact now complete. Andrew, are you able to do the last one, since you have 64-bit OpenCL set up already for the standalone tests? I can help with the app_info.xml if needed.
- Iain
Sorry about being late here, there was a fire in the building(someone downstairs had a meth lab -_-) and thanks larsson for finishing it!
Best excuse I've heard in a while! Did you get the long x87 test completed?
- Iain
Yes, for some reason it took 187 hours for me to complete(maybe AVX support on my end?)
I go to a rather interesting Uni I suppose, one of four fires I believe in the housing areas :) | |
|
|
|
|
Thanks! Only OpenCL BOINC on 32-bit Linux left now, anyone able to take that one?
- Iain
____________
Twitter: IainBethune
Proud member of team "Aggie The Pew". Go Aggie!
3073428256125*2^1290000-1 is Prime! | |
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 14036
ID: 53948
Credit: 475,998,733
RAC: 222,579
                     
|
|
Genefer 3.2.8 is now live for both Short and World Record tasks.
____________
My lucky number is 75898524288+1 | |
|
|
|
Genefer 3.2.8 is now live for both Short and World Record tasks.
Thanks Mike! I will add them to the prpnet packages soon too.
____________
Twitter: IainBethune
Proud member of team "Aggie The Pew". Go Aggie!
3073428256125*2^1290000-1 is Prime! | |
|
Message boards :
Generalized Fermat Prime Search :
Genefer 3.2.8 testing |
