|
NOTE: I think we're not going to do this testing afterall. This post can be resurrected later for a new round of testing, if needed.
We're getting close to releasing the new Genefer into production. It's got lots of improvements, most notably having all the CPU transforms in a single program. You no longer need to pick and choose which app to run. The only choices now are CPU, Nvidia CUDA, Nvidia OpenCL, or ATI OpenCL.
You can help with the testing by running BOINC tasks with the new executable using the app_info anonymous platform mechanism.
Reservations:
I'm handling reservations the same way as we did with the FMA3 LLR app. There's a Google Drive spreadsheet where you can see what tests still need to be done. Anything that is grey is available. If it's yellow, blue, or green, someone is already doing that test.
We want to verify that a BOINC task can run through to completion in each of the categories in that matrix. We'll be testing Windows, Linux, and Mac. We're testing short and WR, although only short tasks on CPUs. We want to task AMD CPUs, pre Sandy Bridge Intel, Sandy Bridge or Ivy Bridge, and Haswell. We want to test ATI GPUs, Nvidia 5xx and earlier GPUs, and Nvidia 6xx+ and later (including TITAN) GPUs. Finally, on Both classes of Nvidia GPUs, we want to test both the CUDA and the OpenCL versions of Genefer.
In all that's 42 different tests that need to be done.
If you wish to reserve one of the tests, post in this thread and I'll update the spreadsheet. Let me know the workunit ID of the test you're running. Once I have the ID, I can track its progress.
Downloads:
Windows:
Windows CPU (64 bit only)
Windows Nvidia CUDA
Windows OpenCL (Nvidia or ATI)
Linux
Linux CPU (64 bit only)
Linux Nvidia CUDA
Linux OpenCL (Nvidia or ATI)
Mac
Mac CPU (64 bit only)
Mac Nvidia CUDA
Mac OpenCL (Nvidia or ATI)
How to use app_info:
RULE # 1: TURNING APP_INFO ON OR OFF, OR CHANGING IT, USUALLY ABORTS ALL TASKS THAT ARE ON YOUR COMPUTER. FINISH ALL YOUR TASKS BEFORE PLAYING WITH APP_INFO!!!!!
Rule #2: See Rule #1.
Please, make sure all of your PrimeGrid work is finished before turning app_info on or off.
The following directions are for Windows, but the same process is used on Linux and Mac, although the filenames and directories are slightly different.
To use app_info, you need to create a file called app_info.xml in your ...boinc\projects\www.primegrid.com directory. The contents contain information about each subproject you want to run. You can NOT turn on app_info for some sub-projects; if you use app_info, you must use it for everything you wish to run.
Here's what a sample file with just two projects in it:
<app_info>
<app>
<name>genefer</name>
<user_friendly_name>Genefer</user_friendly_name>
</app>
<app>
<name>genefer_wr</name>
<user_friendly_name>Genefer (World Record)</user_friendly_name>
</app>
<file_info>
<name>geneferCUDA-windows.exe</name>
<executable/>
</file_info>
<file_info>
<name>cudart32_32_16.dll</name>
</file_info>
<file_info>
<name>cufft32_32_16.dll</name>
</file_info>
<app_version>
<app_name>genefer</app_name>
<version_num>007</version_num>
<api_version>6.10.25</api_version>
<file_ref>
<file_name>geneferCUDA-windows.exe</file_name>
<main_program/>
</file_ref>
<file_ref>
<file_name>cudart32_32_16.dll</file_name>
<open_name>cudart32_32_16.dll</open_name>
<copy_file/>
</file_ref>
<file_ref>
<file_name>cufft32_32_16.dll</file_name>
<open_name>cufft32_32_16.dll</open_name>
<copy_file/>
</file_ref>
<platform>windows_intelx86</platform>
<plan_class>cudaGFN</plan_class>
<coproc>
<type>CUDA</type>
<count>1.000000</count>
</coproc>
</app_version>
<app_version>
<app_name>genefer_wr</app_name>
<version_num>007</version_num>
<api_version>6.10.25</api_version>
<file_ref>
<file_name>geneferCUDA-windows.exe</file_name>
<main_program/>
</file_ref>
<file_ref>
<file_name>cudart32_32_16.dll</file_name>
<open_name>cudart32_32_16.dll</open_name>
<copy_file/>
</file_ref>
<file_ref>
<file_name>cufft32_32_16.dll</file_name>
<open_name>cufft32_32_16.dll</open_name>
<copy_file/>
</file_ref>
<platform>windows_intelx86</platform>
<plan_class>cudaGFNWR</plan_class>
<coproc>
<type>CUDA</type>
<count>1.000000</count>
</coproc>
</app_version>
</app_info>
Most of this is copied from sched_reply_www.primegrid.com.xml, which is created by BOINC when you request work for one of the subprojects. You take what it gives you, and change it ever so slightly to make it use your own application, instead of the server.
App_info contains three different types of entries:
<app> defines the subproject -- in this case, genefer or genefer_wr. Copy what I have in the sample file above.
The second section, <file_info>, is the file definitions, where you tell boinc all the files you're using. Defined above is genefercuda and the two cuda libraries it requires. For a CPU or OpenCL app, you only have to define the genefer program as there aren't any libraries.
Finally, the <app_version> sections tie it all together and is what defines the program that you want to run tasks for that app. In the sample file above, I've defined cuda app_versions for both the genefer and genefer_wr apps. It defines both the executable file as well as the cuda libraries.
If you wanted to run a different cuda program, you would use the app_version section above, but change the executable name to the name of your program. If you need different libraries, you'd change that too.
If you wanted to run CPU programs, you'd use an app version section that looks something like this:
<app_version>
<app_name>genefer</app_name>
<version_num>007</version_num>
<api_version>6.10.25</api_version>
<file_ref>
<file_name>geneferAVX.exe</file_name>
<main_program/>
</file_ref>
<platform>windows_intelx86_64</platform>
</app_version>
We got rid of the libraries, the plan class, and some GPU specific stuff, but notice that the platform is different: it's now the 64 bit windows platform.
For an OpenCL version running on an Nvidia GPU, you would use something like this:
<app_version>
<app_name>genefer</app_name>
<version_num>007</version_num>
<api_version>6.10.25</api_version>
<file_ref>
<file_name>geneferocl-windows.exe</file_name>
<main_program/>
</file_ref>
<platform>windows_intelx86</platform>
<plan_class>OCLcudaGFN</plan_class>
<coproc>
<type>CUDA</type>
<count>1.000000</count>
</coproc>
</app_version>
For an ATI GPU running the OpenCL app, you would change <type>CUDA</type> to <type>ATI</type>.
The file names should match the files you downloaded. You can rename the files if you wish.
Finally, once the file is set up, you need to restart your boinc-client for it to take affect. This is NOT the same as restarting the boinc manager window. If you're not sure how to do this, the easiest thing is to restart your computer. (The method varies depending on your O.S., how you installed BOINC, and which version of BOINC you're running.)
After switching to app_info and restarting BOINC, or when modifying app_info and restarting BOINC, you should look at the beginning of your BOINC log to see if there were any errors detected in the app_info file.
____________
My lucky number is 75898524288+1 |
