Posts by zombie67 [MM]
log in
1) Message boards : Number crunching : Trial Factoring (Message 6679)
Posted 5 Aug 2020 by zombie67 [MM]
Does it work yet, running one task per GPU, with multiple GPUs on Windows? Back in April, I think it was still being worked on. But I haven't heard any update since then.


I have sent a PM to the dev. At the moment no.


Got it. I will stay tuned for more info.
2) Message boards : Number crunching : Trial Factoring (Message 6677)
Posted 5 Aug 2020 by zombie67 [MM]
Does it work yet, running one task per GPU, with multiple GPUs on Windows? Back in April, I think it was still being worked on. But I haven't heard any update since then.
3) Message boards : Number crunching : AVX-512? (Message 6561)
Posted 6 Jun 2020 by zombie67 [MM]
I am not so sure. Here is the result of a machine with AVX-512s:

Using zero-padded FMA3 FFT length 560K, Pass1=448, Pass2=1280, 5 threads, a = 3


FMA3 is not AVX-512.
4) Message boards : Number crunching : AVX-512? (Message 6559)
Posted 6 Jun 2020 by zombie67 [MM]
Do the CPU apps use AVX-512 if available?
5) Message boards : Number crunching : Trial Factoring (Message 6270)
Posted 19 Apr 2020 by zombie67 [MM]
Unsent is now zero.
6) Message boards : Number crunching : Sierpinski / Riesel Base - short (Message 6237)
Posted 16 Apr 2020 by zombie67 [MM]
More work is available now for this app.


Thanks!
7) Message boards : Number crunching : Sierpinski / Riesel Base - short (Message 6202)
Posted 15 Apr 2020 by zombie67 [MM]
Any idea when there will be more of these? I am just a hair short of the next badge level. :)
8) Message boards : Number crunching : Trial Factoring (Message 6194)
Posted 14 Apr 2020 by zombie67 [MM]
Got it. Ah well, I was going to add more GPUs to help speed up the current run. Bot most of my machines are multi-GPU. So I can add only one GPU each machine.
9) Message boards : Number crunching : Trial Factoring (Message 6190)
Posted 14 Apr 2020 by zombie67 [MM]
Question: Does it work yet, running one task per GPU, with multiple GPUs? I looked at the FAQ, and I see conflicting messages. Specifically, I am asking about NVIDIA on linux, and NVIDIA on Windows.

Running mfakto (opencl)
- Multi-GPU not supported

Running mfaktc (CUDA)
- Multi-GPU supported

app_config.xml
Multi-GPU (not working correctly)
10) Message boards : Number crunching : Trial Factoring (Message 6189)
Posted 14 Apr 2020 by zombie67 [MM]
I will add GPUs shortly.
11) Message boards : Number crunching : Trial Factoring tests (Message 5992)
Posted 3 Apr 2020 by zombie67 [MM]
Can you try standalone with:

./mfakto -d 00 -st
./mfakto -d 01 -st
./mfakto -d 10 -st
./mfakto -d 11 -st

mfakto is terrible to recognize the GPU

I have also below a possible compiling mfaktc app for mac if you are able to do this.



FYI, mfakto did not exist. It was still inside a zip package. I had to manually unzip it, cd into the unzipped directory, and then run the commands. Here are the results.

Notes:
- I currently have the cc_config.xml set to ignore the internal AMD Radeon Pro 580 (GPU1), and run only on the external AMD Radeon VII (GPU0).
- It looks like the first and third commands show the 580, and the second and fourth commands show the VII.
- BOINC startup info:

5 4/2/2020 12:37:57 PM OpenCL: AMD/ATI GPU 0: AMD Radeon VII Compute Engine (driver version 1.2 (Jan 13 2020 20:12:46), device version OpenCL 1.2, 16368MB, 16368MB available, 8645 GFLOPS peak)
6 4/2/2020 12:37:57 PM OpenCL: AMD/ATI GPU 1 (ignored by config): AMD Radeon Pro 580 Compute Engine (driver version 1.2 (Jan 13 2020 20:12:40), device version OpenCL 1.2, 8192MB, 8192MB available, 3456 GFLOPS peak)



kyoko:srbase.my-firewall.org_sr5 zombie67$ ./mfakto -d 00 -st
-bash: ./mfakto: No such file or directory
kyoko:srbase.my-firewall.org_sr5 zombie67$ ls
app_config.xml job_srbase_m64_00022.xml mfakto-mac64-v1.zip wrapper_26012_v2_x86_64-apple-darwin
job_GPU72_m64_00001.xml mfakto-mac64-v1 srbase_v8_m64.zip
kyoko:srbase.my-firewall.org_sr5 zombie67$ cd mfakto-mac64-v1
kyoko:mfakto-mac64-v1 zombie67$ ls
COPYING README.txt barrett15.cl common.cl gpusieve.cl mfakto.ini montgomery.cl tf_debug.h
Changelog-mfakto.txt barrett.cl barrett24.cl datatypes.h mfakto mfakto_Kernels.cl mul24.cl
kyoko:mfakto-mac64-v1 zombie67$ ./mfakto -d 00 -st
mfakto 0.15pre6 (64bit build)


Runtime options
Inifile mfakto.ini
Verbosity 1
SieveOnGPU yes
MoreClasses yes
GPUSievePrimes 82486
GPUSieveProcessSize 24Ki bits
GPUSieveSize 96Mi bits
FlushInterval 8
WorkFile worktodo.txt
ResultsFile results.txt
Checkpoints enabled
CheckpointDelay 300s
Stages enabled
StopAfterFactor class
PrintMode compact
V5UserID none
ComputerID none
TimeStampInResults no
VectorSize 2
GPUType AUTO
SmallExp no
UseBinfile mfakto_Kernels.elf
Compiletime options

Select device - Get device info:
WARNING: Unknown GPU name, assuming GCN. Please post the device name "AMD Radeon Pro 580 Compute Engine (AMD)" to http://www.mersenneforum.org/showthread.php?t=15646 to have it added to mfakto. Set GPUType in mfakto.ini to select a GPU type yourself to avoid this warning.

OpenCL device info
name AMD Radeon Pro 580 Compute Engine (AMD)
device (driver) version OpenCL 1.2 (1.2 (Jan 13 2020 20:12:40))
maximum threads per block 256
maximum threads per grid 16777216
number of multiprocessors 36 (2304 compute elements)
clock rate 1200MHz

Automatic parameters
threads per grid 0
optimizing kernels for GCN

Compiling kernels.
Warning: Dumping only the first of 2 binary formats - if loading the binary file mfakto_Kernels.elf fails, delete it and specify the -d <n> option for mfakto.
Error -46 (Invalid kernel name): Creating Kernel cl_barrett15_74_gs from program. (clCreateKernel)
ERROR: load_kernels(0) failed
kyoko:mfakto-mac64-v1 zombie67$ ./mfakto -d 01 -st
mfakto 0.15pre6 (64bit build)


Runtime options
Inifile mfakto.ini
Verbosity 1
SieveOnGPU yes
MoreClasses yes
GPUSievePrimes 82486
GPUSieveProcessSize 24Ki bits
GPUSieveSize 96Mi bits
FlushInterval 8
WorkFile worktodo.txt
ResultsFile results.txt
Checkpoints enabled
CheckpointDelay 300s
Stages enabled
StopAfterFactor class
PrintMode compact
V5UserID none
ComputerID none
TimeStampInResults no
VectorSize 2
GPUType AUTO
SmallExp no
UseBinfile mfakto_Kernels.elf
Compiletime options

Select device - Get device info:
WARNING: Unknown GPU name, assuming GCN. Please post the device name "AMD Radeon VII Compute Engine (AMD)" to http://www.mersenneforum.org/showthread.php?t=15646 to have it added to mfakto. Set GPUType in mfakto.ini to select a GPU type yourself to avoid this warning.

OpenCL device info
name AMD Radeon VII Compute Engine (AMD)
device (driver) version OpenCL 1.2 (1.2 (Jan 13 2020 20:12:46))
maximum threads per block 256
maximum threads per grid 16777216
number of multiprocessors 60 (3840 compute elements)
clock rate 1801MHz

Automatic parameters
threads per grid 0
optimizing kernels for GCN

Loading binary kernel file mfakto_Kernels.elf
Compiling kernels.
Error -46 (Invalid kernel name): Creating Kernel cl_barrett15_74_gs from program. (clCreateKernel)
ERROR: load_kernels(0) failed
kyoko:mfakto-mac64-v1 zombie67$ ./mfakto -d 10 -st
mfakto 0.15pre6 (64bit build)


Runtime options
Inifile mfakto.ini
Verbosity 1
SieveOnGPU yes
MoreClasses yes
GPUSievePrimes 82486
GPUSieveProcessSize 24Ki bits
GPUSieveSize 96Mi bits
FlushInterval 8
WorkFile worktodo.txt
ResultsFile results.txt
Checkpoints enabled
CheckpointDelay 300s
Stages enabled
StopAfterFactor class
PrintMode compact
V5UserID none
ComputerID none
TimeStampInResults no
VectorSize 2
GPUType AUTO
SmallExp no
UseBinfile mfakto_Kernels.elf
Compiletime options

Select device - Get device info:
WARNING: Unknown GPU name, assuming GCN. Please post the device name "AMD Radeon Pro 580 Compute Engine (AMD)" to http://www.mersenneforum.org/showthread.php?t=15646 to have it added to mfakto. Set GPUType in mfakto.ini to select a GPU type yourself to avoid this warning.

OpenCL device info
name AMD Radeon Pro 580 Compute Engine (AMD)
device (driver) version OpenCL 1.2 (1.2 (Jan 13 2020 20:12:40))
maximum threads per block 256
maximum threads per grid 16777216
number of multiprocessors 36 (2304 compute elements)
clock rate 1200MHz

Automatic parameters
threads per grid 0
optimizing kernels for GCN

Loading binary kernel file mfakto_Kernels.elf
Compiling kernels.
Error -46 (Invalid kernel name): Creating Kernel cl_barrett15_74_gs from program. (clCreateKernel)
ERROR: load_kernels(0) failed
kyoko:mfakto-mac64-v1 zombie67$ ./mfakto -d 11 -st
mfakto 0.15pre6 (64bit build)


Runtime options
Inifile mfakto.ini
Verbosity 1
SieveOnGPU yes
MoreClasses yes
GPUSievePrimes 82486
GPUSieveProcessSize 24Ki bits
GPUSieveSize 96Mi bits
FlushInterval 8
WorkFile worktodo.txt
ResultsFile results.txt
Checkpoints enabled
CheckpointDelay 300s
Stages enabled
StopAfterFactor class
PrintMode compact
V5UserID none
ComputerID none
TimeStampInResults no
VectorSize 2
GPUType AUTO
SmallExp no
UseBinfile mfakto_Kernels.elf
Compiletime options

Select device - Get device info:
WARNING: Unknown GPU name, assuming GCN. Please post the device name "AMD Radeon VII Compute Engine (AMD)" to http://www.mersenneforum.org/showthread.php?t=15646 to have it added to mfakto. Set GPUType in mfakto.ini to select a GPU type yourself to avoid this warning.

OpenCL device info
name AMD Radeon VII Compute Engine (AMD)
device (driver) version OpenCL 1.2 (1.2 (Jan 13 2020 20:12:46))
maximum threads per block 256
maximum threads per grid 16777216
number of multiprocessors 60 (3840 compute elements)
clock rate 1801MHz

Automatic parameters
threads per grid 0
optimizing kernels for GCN

Loading binary kernel file mfakto_Kernels.elf
Compiling kernels.
Error -46 (Invalid kernel name): Creating Kernel cl_barrett15_74_gs from program. (clCreateKernel)
ERROR: load_kernels(0) failed
kyoko:mfakto-mac64-v1 zombie67$
12) Message boards : Number crunching : Trial Factoring tests (Message 5988)
Posted 2 Apr 2020 by zombie67 [MM]
Running a radeon VII on a Mac fails. I assume it is because the GPU is not known, but not sure.

http://srbase.my-firewall.org/sr5/result.php?resultid=387845348

Stderr output

<core_client_version>7.14.4</core_client_version>
<![CDATA[
<message>
process exited with code 195 (0xc3, -61)</message>
<stderr_txt>
12:38:38 (2042): wrapper (7.2.26012): starting
12:38:38 (2042): wrapper: running ./mfakto ( --device 0)
ERROR: unknown option '--device'
12:38:39 (2042): ./mfakto exited; CPU time 0.012094
12:38:39 (2042): app exit status: 0x100
12:38:39 (2042): called boinc_finish

</stderr_txt>
]]>


Startup GPU identification:

Thu Apr 2 12:37:57 2020 | | OpenCL: AMD/ATI GPU 0: AMD Radeon VII Compute Engine (driver version 1.2 (Jan 13 2020 20:12:46), device version OpenCL 1.2, 16368MB, 16368MB available, 8645 GFLOPS peak)
13) Message boards : Number crunching : Trial Factoring tests (Message 5959)
Posted 29 Mar 2020 by zombie67 [MM]
If the app_config is not working try this here:

<exclude_gpu> <url>http://srbase.my-firewall.org/sr5/</url> <device_num>1</device_num> <type>NVIDIA|ATI</type> <app>GPU72</app> </exclude_gpu>


change the type what you have in your PC and put this code in the option section of cc_config


This is working for me. One GPU is running gpu72, the other GPU is running a task from another project.
14) Message boards : Number crunching : Trial Factoring tests (Message 5945)
Posted 29 Mar 2020 by zombie67 [MM]
<app_config>
<app_version>
<app_name>GPU72</app_name>
<plan_class>cuda100</plan_class> # 1
<cmdline>-d c</cmdline> # 2 - try multi-GPU, 2 tasks * 2 GPU
</app_version>
</app_config>


I am having trouble understanding what you mean with #2. I am already running two tasks, as I have two GPUs. But both tasks are running on just one of the GPUs*. What are you suggesting "c" should be, to get one task per GPU? It is currently set to "1", as it says in the FAQ.

Note: BOINCmgr shows one task running on GPU0 and one task on GPU1, even though both tasks are actually running on just one of the GPUs.
15) Message boards : Number crunching : Trial Factoring tests (Message 5935)
Posted 29 Mar 2020 by zombie67 [MM]
With regard to the app_config.xml file, to run one pask per GPU:

1) I am getting the following error in the event log, when BOINC reads the app_config.xml file:

Sun 29 Mar 2020 07:30:41 AM PDT | SRBase | Entry in app_config.xml for app 'GPU72', plan class '' doesn't match any app versions


2) Even with the error message, it seems to be working. I am running one task on each GPU, where it used to run both tasks on just one of the GPUs.

If you want host details, it is this one.

Edit: It was working for the first two tasks. But once they completed, and started new tasks, it went back to running both tasks on a single GPU.
16) Message boards : Number crunching : Trial Factoring (Message 5878)
Posted 28 Mar 2020 by zombie67 [MM]
linux v2 and Mac doesnt have an entry yet

I just tried running it on a Mac. It failed.

Here is the task:

http://srbase.my-firewall.org/sr5/result.php?resultid=387750937

Stderr output

<core_client_version>7.14.2</core_client_version>
<![CDATA[
<message>
process exited with code 195 (0xc3, -61)</message>
<stderr_txt>
20:02:04 (52956): wrapper (7.2.26012): starting
20:02:04 (52956): wrapper: running ./mfakto ( --device 0)
ERROR: unknown option '--device'
20:02:05 (52956): ./mfakto exited; CPU time 0.013171
20:02:05 (52956): app exit status: 0x100
20:02:05 (52956): called boinc_finish

</stderr_txt>
]]>


Here are the system and GPU details if you need them:

1 3/23/2020 3:49:43 PM Starting BOINC client version 7.14.2 for x86_64-apple-darwin
3 3/23/2020 3:49:43 PM Libraries: libcurl/7.58.0 OpenSSL/1.1.0g zlib/1.2.11 c-ares/1.13.0
5 3/23/2020 3:49:43 PM OpenCL: AMD/ATI GPU 0: AMD Radeon Pro 580 Compute Engine (driver version 1.2 (Jan 13 2020 20:12:40), device version OpenCL 1.2, 8192MB, 8192MB available, 3456 GFLOPS peak)
6 3/23/2020 3:49:43 PM OpenCL CPU: Intel(R) Core(TM) i7-7700K CPU @ 4.20GHz (OpenCL driver vendor: Apple, driver version 1.1, device version OpenCL 1.2)
8 3/23/2020 3:49:43 PM Processor: 8 GenuineIntel Intel(R) Core(TM) i7-7700K CPU @ 4.20GHz [x86 Family 6 Model 158 Stepping 9]
9 3/23/2020 3:49:43 PM Processor features: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clfsh ds acpi mmx fxsr sse sse2 ss htt tm pbe pni pclmulqdq dtes64 mon dscpl vmx est tm2 ssse3 fma cx16 tpr pdcm sse4_1 sse4_2 x2apic movbe popcnt aes pcid xsave o
10 3/23/2020 3:49:43 PM OS: Mac OS X 10.15.3 (Darwin 19.3.0)


Edit: I tried a second Mac, with a different GPU. Same error. Here is the line for the GPU form the event log:

5 3/27/2020 8:12:25 PM OpenCL: AMD/ATI GPU 0: AMD Radeon R9 M295X Compute Engine (driver version 1.2 (Jan 13 2020 20:12:40), device version OpenCL 1.2, 4096MB, 4096MB available, 2176 GFLOPS peak)
17) Message boards : Number crunching : Trial Factoring (Message 5875)
Posted 27 Mar 2020 by zombie67 [MM]
I thought the linux apps were updated so that installation of the toolkit was no longer required. I could swear I read that somewhere. Or am I going crazy?
18) Message boards : News : R430 Megaprime found (Message 2586)
Posted 23 May 2016 by zombie67 [MM]
:D
19) Message boards : Number crunching : short deadlines cause panic mode (Message 1731)
Posted 1 Aug 2015 by zombie67 [MM]
Agreed. Longer deadlines would solve the problem for everyone. Panic mode causes the BOINC client to do abnormal things.
20) Message boards : Number crunching : short deadlines cause panic mode (Message 1729)
Posted 1 Aug 2015 by zombie67 [MM]
That doesn't solve the problem.

But I did find a work-around using the app_config.xml:

<app_config>
<project_max_concurrent>N</project_max_concurrent>
</app_config>

It constrains the number of CPU threads available to the project. Setting this to be number of threads minus the number of threads needed for the GPUs.


Next 20

Main page · Your account · Message boards


Copyright © 2014-2020 BOINC Confederation / rebirther