There is a simple regedit fix if you continue to have the stuck WU issue but updating the driver is a must. 12.11beta4
Printable View
out of need and curiosity: is it possible to run two different cpu projects and one gpu at all? trying the gfam app_info and its not working. maybe because i also have hcc cpu in my cache?
What GPU are you using ... I have app_info files for either nvidia or amd that do HCC GPU, HCC CPU, and GFAM CPU
a hd7950.
Here it is.
Side note ... GFAM WUs come from WCG in both 32 and 64 bit versions and BOINC will randomly test the different versions to make sure you *mostly* get the version that runs faster. I made an assumption that you want to run 64bit so I told the 32bit version to use the 64bit executables. This works perfectly fine (I am running this myself on an nvidia rig)
Code:<app_info>
<app>
<name>hcc1</name>
<user_friendly_name>XtremeSystems Crunches Cancer</user_friendly_name>
</app>
<app>
<name>gfam</name>
<user_friendly_name>XtremeSystems Fights Malaria</user_friendly_name>
</app>
<file_info>
<name>wcg_hcc1_img_6.56_windows_intelx86__ati_hcc1</name>
<executable/>
</file_info>
<file_info>
<name>hcckernel.cl.6.56</name>
<executable/>
</file_info>
<file_info>
<name>wcg_hcc1_img_6.56_windows_intelx86</name>
<executable/>
</file_info>
<file_info>
<name>wcgrid_gfam_vina_6.12_windows_x86_64</name>
<executable/>
</file>
<file_info>
<name>wcgrid_gfam_vina_prod_x86_64.exe.6.12</name>
<executable/>
</file>
<app_version>
<app_name>hcc1</app_name>
<version_num>656</version_num>
<platform>windows_intelx86</platform>
<plan_class>ati_hcc1</plan_class>
<avg_ncpus>1.000000</avg_ncpus>
<max_ncpus>0.750000</max_ncpus>
<coproc>
<type>ATI</type>
<count>0.125000</count>
</coproc>
<file_ref>
<file_name>wcg_hcc1_img_6.56_windows_intelx86__ati_hcc1</file_name>
<main_program/>
</file_ref>
<file_ref>
<file_name>hcckernel.cl.6.56</file_name>
<open_name>hcckernel.cl</open_name>
</file_ref>
</app_version>
<app_version>
<app_name>hcc1</app_name>
<version_num>656</version_num>
<platform>windows_intelx86</platform>
<avg_ncpus>1.000000</avg_ncpus>
<max_ncpus>1.000000</max_ncpus>
<flops>3231932486.733310</flops>
<api_version>6.13.0</api_version>
<file_ref>
<file_name>wcg_hcc1_img_6.56_windows_intelx86</file_name>
<main_program/>
</file_ref>
</app_version>
<app_version>
<app_name>gfam</app_name>
<version_num>612</version_num>
<platform>windows_x86_64</platform>
<avg_ncpus>1.000000</avg_ncpus>
<max_ncpus>1.000000</max_ncpus>
<flops>3868351853.053741</flops>
<api_version>7.1.0</api_version>
<file_ref>
<file_name>wcgrid_gfam_vina_6.12_windows_x86_64</file_name>
<main_program/>
</file_ref>
<file_ref>
<file_name>wcgrid_gfam_vina_prod_x86_64.exe.6.12</file_name>
<open_name>AutoDockVina64.exe</open_name>
</file_ref>
</app_version>
<app_version>
<app_name>gfam</app_name>
<version_num>612</version_num>
<platform>windows_x86</platform>
<avg_ncpus>1.000000</avg_ncpus>
<max_ncpus>1.000000</max_ncpus>
<flops>3868351853.053741</flops>
<api_version>7.1.0</api_version>
<file_ref>
<file_name>wcgrid_gfam_vina_6.12_windows_x86_64</file_name>
<main_program/>
</file_ref>
<file_ref>
<file_name>wcgrid_gfam_vina_prod_x86_64.exe.6.12</file_name>
<open_name>AutoDockVina64.exe</open_name>
</file_ref>
</app_version>
</app_info>
another side note ... I've make diretories outside of my BOINC folders to store the executable files so I can just pop them in and switch up my app_info for the projects I want to run. This saves me from having run without app_info while I wait for the executables to download for all of the projects I want. Another trick is that if you want to run on someone else's account and you have their weak key you make an account file and drop it in the main BOINC ProgramData folder
C:\ProgramData\BOINC
account_www.worldcommunitygrid.org.xml
Open BOINC Manager and "Remove" WCG.Code:<account>
<master_url>http://www.worldcommunitygrid.org/</master_url>
<authenticator>XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX</authenticator>
<project_name>World Community Grid</project_name>
</account>
Close BOINC completely.
Create a folder in the BOINC projects folder and name it www.worldcommunitygrid.org
drop in the app_info and the executables, restart BOINC ... almost no downtime :-)
wcg tells me theres a parse error check syntax in the xml file.
checking now...
I just copied it into my app_info and it works perfectly fine on my machine. Please post the messages from the log when you first restart BOINC so we can take a look at what's happening on your machine.
perhaps you were talking about the account file in my last post ... sorry, I just wrapped it better a moment ago :am:
no i wasnt. i copied the app info in the editor saved it to programdata/boinc/wcg and started boinc again (the message was written in red and was displayed within seconds of program start and before the not using a proxy message),though i am currently crunching on oc account. so i need to follow that as well?
I forgot about HPF2. One machine grabbed a WU and stopped getting GPU work. It's the same story as GFAM, it apparently has an x86 & x86_64 executable. Once I added in a section for "windows_x86_64" with the same files as above, it worked fine.
Relevant info:
Just stick it somewhere between <app_info> and </app_info>Code:<app>
<name>hpf2</name>
<user_friendly_name>Human Proteome Folding - Phase 2</user_friendly_name>
</app>
<file_info>
<name>wcg_hpf2_rosetta_6.40_windows_intelx86</name>
<executable/>
</file_info>
<file_info>
<name>wcg_hpf2_6.40.tga</name>
<executable/>
</file_info>
<file_info>
<name>wcg_hpf2_rosetta_graphics_6.40_windows_intelx86</name>
<executable/>
</file_info>
<file_info>
<name>hpf2_6.40_win_paths.txt</name>
<executable/>
</file_info>
<app_version>
<app_name>hpf2</app_name>
<version_num>640</version_num>
<platform>windows_intelx86</platform>
<avg_ncpus>1.000000</avg_ncpus>
<max_ncpus>1.000000</max_ncpus>
<flops>3570131549.238372</flops>
<api_version>6.3.3</api_version>
<file_ref>
<file_name>wcg_hpf2_rosetta_6.40_windows_intelx86</file_name>
<main_program/>
</file_ref>
<file_ref>
<file_name>wcg_hpf2_6.40.tga</file_name>
<open_name>wc_grid_hpf2.tga</open_name>
</file_ref>
<file_ref>
<file_name>wcg_hpf2_rosetta_graphics_6.40_windows_intelx86</file_name>
<open_name>graphics_app</open_name>
</file_ref>
<file_ref>
<file_name>hpf2_6.40_win_paths.txt</file_name>
<open_name>paths.txt</open_name>
</file_ref>
</app_version>
<app_version>
<app_name>hpf2</app_name>
<version_num>640</version_num>
<platform>windows_x86_64</platform>
<avg_ncpus>1.000000</avg_ncpus>
<max_ncpus>1.000000</max_ncpus>
<flops>3570131549.238372</flops>
<api_version>6.3.3</api_version>
<file_ref>
<file_name>wcg_hpf2_rosetta_6.40_windows_intelx86</file_name>
<main_program/>
</file_ref>
<file_ref>
<file_name>wcg_hpf2_6.40.tga</file_name>
<open_name>wc_grid_hpf2.tga</open_name>
</file_ref>
<file_ref>
<file_name>wcg_hpf2_rosetta_graphics_6.40_windows_intelx86</file_name>
<open_name>graphics_app</open_name>
</file_ref>
<file_ref>
<file_name>hpf2_6.40_win_paths.txt</file_name>
<open_name>paths.txt</open_name>
</file_ref>
</app_version>
-starting boinc 6.10.58
- found app info using anonymous platform
- parse error in app info xml check syntax in xml file
-no application found for task: discarding windows x86_64 612
then a lot of no app discarding....
-general prefs for home
-reading prefs override file...//preferences
-not using proxy
-suspending computation, initial delay
"check syntax in xml file"
no typos in the xml file?
I used that exact app_info ... in fact I copied it out of the post and tested it after haschioz reported an issue the first time.
haschioz ... can you post back the app info inside [code] tags so we can see if anything got lost in translation?
heres the info i pulled out of the file (with the editor)
<app_info>
<app>
<name>hcc1</name>
<user_friendly_name>XtremeSystems Crunches Cancer</user_friendly_name>
</app>
<app>
<name>gfam</name>
<user_friendly_name>XtremeSystems Fights Malaria</user_friendly_name>
</app>
<file_info>
<name>wcg_hcc1_img_6.56_windows_intelx86__ati_hcc1 </name>
<executable/>
</file_info>
<file_info>
<name>hcckernel.cl.6.56</name>
<executable/>
</file_info>
<file_info>
<name>wcg_hcc1_img_6.56_windows_intelx86</name>
<executable/>
</file_info>
<file_info>
<name>wcgrid_gfam_vina_6.12_windows_x86_64</name>
<executable/>
</file>
<file_info>
<name>wcgrid_gfam_vina_prod_x86_64.exe.6.12</name>
<executable/>
</file>
<app_version>
<app_name>hcc1</app_name>
<version_num>656</version_num>
<platform>windows_intelx86</platform>
<plan_class>ati_hcc1</plan_class>
<avg_ncpus>1.000000</avg_ncpus>
<max_ncpus>0.750000</max_ncpus>
<coproc>
<type>ATI</type>
<count>0.125000</count>
</coproc>
<file_ref>
<file_name>wcg_hcc1_img_6.56_windows_intelx86__ati _hcc1</file_name>
<main_program/>
</file_ref>
<file_ref>
<file_name>hcckernel.cl.6.56</file_name>
<open_name>hcckernel.cl</open_name>
</file_ref>
</app_version>
<app_version>
<app_name>hcc1</app_name>
<version_num>656</version_num>
<platform>windows_intelx86</platform>
<avg_ncpus>1.000000</avg_ncpus>
<max_ncpus>1.000000</max_ncpus>
<flops>3231932486.733310</flops>
<api_version>6.13.0</api_version>
<file_ref>
<file_name>wcg_hcc1_img_6.56_windows_intelx86</file_name>
<main_program/>
</file_ref>
</app_version>
<app_version>
<app_name>gfam</app_name>
<version_num>612</version_num>
<platform>windows_x86_64</platform>
<avg_ncpus>1.000000</avg_ncpus>
<max_ncpus>1.000000</max_ncpus>
<flops>3868351853.053741</flops>
<api_version>7.1.0</api_version>
<file_ref>
<file_name>wcgrid_gfam_vina_6.12_windows_x86_64</file_name>
<main_program/>
</file_ref>
<file_ref>
<file_name>wcgrid_gfam_vina_prod_x86_64.exe.6.12 </file_name>
<open_name>AutoDockVina64.exe</open_name>
</file_ref>
</app_version>
<app_version>
<app_name>gfam</app_name>
<version_num>612</version_num>
<platform>windows_x86</platform>
<avg_ncpus>1.000000</avg_ncpus>
<max_ncpus>1.000000</max_ncpus>
<flops>3868351853.053741</flops>
<api_version>7.1.0</api_version>
<file_ref>
<file_name>wcgrid_gfam_vina_6.12_windows_x86_64</file_name>
<main_program/>
</file_ref>
<file_ref>
<file_name>wcgrid_gfam_vina_prod_x86_64.exe.6.12 </file_name>
<open_name>AutoDockVina64.exe</open_name>
</file_ref>
</app_version>
</app_info>
IMHO your boinc version 6.10.58 is a big part of your problem. During the beta testing the 6.x.xx versions had by far the most errors when I checked my resend logs.
Update to 7.0.36 The 7 versions are much better optimized to run GPU tasks. The latest listed on the site is 7.0.38 but I haven't tried that one yet and I know 7.0.36 works well.
Uninstall the old then install the new.
New app_info for NVIDIA ... HCC_CPU, HCC_GPU, GFAM_CPU, Configured to run 2 concurrent HCC_GPU
Code:<app_info>
<app>
<name>hcc1</name>
<user_friendly_name>XtremeSystems Crunches Cancer</user_friendly_name>
</app>
<app>
<name>gfam</name>
<user_friendly_name>XtremeSystems Fights Malaria</user_friendly_name>
</app>
<file_info>
<name>wcg_hcc1_img_6.56_windows_intelx86__nvidia_hcc1</name>
<executable/>
</file_info>
<file_info>
<name>hcckernel.cl.6.56</name>
<executable/>
</file_info>
<file_info>
<name>wcg_hcc1_img_6.56_windows_intelx86</name>
<executable/>
</file_info>
<file_info>
<name>wcgrid_gfam_vina_6.12_windows_x86_64</name>
<executable/>
</file>
<file_info>
<name>wcgrid_gfam_vina_prod_x86_64.exe.6.12</name>
<executable/>
</file>
<app_version>
<app_name>hcc1</app_name>
<version_num>656</version_num>
<platform>windows_intelx86</platform>
<plan_class>nvidia_hcc1</plan_class>
<avg_ncpus>1.000000</avg_ncpus>
<max_ncpus>0.750000</max_ncpus>
<coproc>
<type>CUDA</type>
<count>0.500000</count>
</coproc>
<file_ref>
<file_name>wcg_hcc1_img_6.56_windows_intelx86__nvidia_hcc1</file_name>
<main_program/>
</file_ref>
<file_ref>
<file_name>hcckernel.cl.6.56</file_name>
<open_name>hcckernel.cl</open_name>
</file_ref>
</app_version>
<app_version>
<app_name>hcc1</app_name>
<version_num>656</version_num>
<platform>windows_intelx86</platform>
<avg_ncpus>1.000000</avg_ncpus>
<max_ncpus>1.000000</max_ncpus>
<flops>3231932486.733310</flops>
<api_version>6.13.0</api_version>
<file_ref>
<file_name>wcg_hcc1_img_6.56_windows_intelx86</file_name>
<main_program/>
</file_ref>
</app_version>
<app_version>
<app_name>gfam</app_name>
<version_num>612</version_num>
<platform>windows_x86_64</platform>
<avg_ncpus>1.000000</avg_ncpus>
<max_ncpus>1.000000</max_ncpus>
<flops>3868351853.053741</flops>
<api_version>7.1.0</api_version>
<file_ref>
<file_name>wcgrid_gfam_vina_6.12_windows_x86_64</file_name>
<main_program/>
</file_ref>
<file_ref>
<file_name>wcgrid_gfam_vina_prod_x86_64.exe.6.12</file_name>
<open_name>AutoDockVina64.exe</open_name>
</file_ref>
</app_version>
<app_version>
<app_name>gfam</app_name>
<version_num>612</version_num>
<platform>windows_intelx86</platform>
<avg_ncpus>1.000000</avg_ncpus>
<max_ncpus>1.000000</max_ncpus>
<flops>3347548492.458962</flops>
<api_version>7.1.0</api_version>
<file_ref>
<file_name>wcgrid_gfam_vina_6.12_windows_x86_64</file_name>
<main_program/>
</file_ref>
<file_ref>
<file_name>wcgrid_gfam_vina_prod_x86_64.exe.6.12</file_name>
<open_name>AutoDockVina64.exe</open_name>
</file_ref>
</app_version>
</app_info>
The only thing I can see in the XML which should not matter anyway is that at the end of a couple of the file name tags there is a space between the actual file name and the close tag ... for example...
<name>wcg_hcc1_img_6.56_windows_intelx86__ati_hc c1 </name>
should be
<name>wcg_hcc1_img_6.56_windows_intelx86__ati_hcc1 </name>
there are a couple more but once you know what to look for they are pretty easy to spot.
alright i will update.
Edit: since a couple ofhours it wont dl any gpu work at all. the cache for cpu is bigger than what i set it at (boinctasks says 15d) for a 1d buffer on a 2600k
Ok, I give up. I can't find it.
There's a way to "cheat" the number of cores, I think Snow Crash posted it somewhere.
What I'm trying to do is run lucky 13 gpu threads on a 12 cpu thread machine. I set .076 in the app_info but, it's still only running 12 threads.
Should I even bother trying? (990x with HT on, and a 7970)
Edit: @Sparky: I've never experienced a work unit hang. The 12.11 beta driver came out before I got my card, so I started with that driver. Had an ATI card before the new one, running probably 11.8 or something driver. I just uninstalled the old, and then installed the new driver and everything has been working fine.
<avg_ncpus>0.8</avg_ncpus> <this number here = 1/number of threads you wish to run multiplied by the number of cores available so for 12 cores this would be 1/13*12=0.923
<max_ncpus>1.0</max_ncpus>
<coproc>
<type>ATI</type>
<count>.1</count> <assuming one card this number is 1/number of threads you wish to run so for 13=0.0769
I am currently trialling 0.8 and 0.1 for 10 threads on one card on a 7950 and things work well on an 8 thread rig
Sure you should give it a go as the 7970 is more capable
Added this to first post ;)
It is the avg_cpu and max_cpu (close enough to find) that needs to change to be
(your CPU count) / (number of GPU tasks)
Thanks Old Chap and Snow Crash. I'm running 13 gpu threads now. Fingers crossed.
ok, Dave the "app_info" challenged here.:D
I'm reading that people are running TWO GPU work units on 5870's and it's working well.
So tell me, what exactly do I need to run two WU and my cpu WU as this is on the beast and I'm not willing to shut off all the potential of the cpu's( 32 threads)
All you will lose is 1 thread for each additional gpu instance you have running on YOUR rig :D
The main question is what are you running besides GPU? HCC?