V11 of CUDA MB mod - attempt to restart freezed apps

[Geek@Play]

I still believe that it's a GPU problem.  That very same GPU has been freezing all weekend while the original computer and the GPU I swapped into it have been running without problems.  In short the freezing problem stayed with the a specific GPU even when swapped into a different computer.

Hope to hear from EVGA tomorrow but have received email confirming the RMA was requested by me.

[Geek@Play]

EVGA approved the RMA and I will be returning the GPU to them today.

[Vyper]

Ok cool. Actually it would be wonderful if it is that issue you have been having from the beginning so you can get rid of it.

Please keep us updated in this case, because if it works for you then a lot of manufacturers could face some hefty RMA from us s@h people

Kind regards Vyper

[Vyper]

Well i reported earlier that i did something that solved the freezing app issue..

After S@H got back up and started distributing WU's again two days ago i haven't had one lousy single cuda.exe hung whatsoever...
Frikking amazing but also somewhat disturbing..

So regarding the lockup version of this executable.. I can't test anything if i can't reproduce the problem that i'm having..

I hope that others out there doesn't need to test this version because actually who want's to deal with bugs that you need a workaround for anyway..

If my issues change and the cuda.exe starts hanging again i will definitely beta try this app and report further but as i said, my computer needs to start locking up again first...

Kind regards Vyper

[Raistmer]

Fine, I'm happy that problem disappears w/o such watchdog version using (it will slow down processing, very small slowdown though but ...).
I use it more than week on my dual GPU host and no more hangs I noticed. Maybe it helps, maybe simple no hangups - I didn't look in every stderr to findout there were restarts or not.
No invalid results from that host - it's enough for me for now.
If someone encounter some bug - he can report it. I'm busy with AP and  newly recived ATI cards now.

[Zeus Fab3r]

Fine, I'm happy that problem disappears w/o such watchdog version using (it will slow down processing, very small slowdown though but ...).
I use it more than week on my dual GPU host and no more hangs I noticed. Maybe it helps, maybe simple no hangups - I didn't look in every stderr to findout there were restarts or not.
No invalid results from that host - it's enough for me for now.
If someone encounter some bug - he can report it. I'm busy with AP and  newly recived ATI cards now.

Finally a solution that actually works 
3 days, 24/7 w/o single hangup and counting. Great job !
Still have a question. Can I run this cross_watch mod alongside opt AP?

Thanks again Raistmer.

[Raistmer]

Still have a question. Can I run this cross_watch mod alongside opt AP?

As long as at least 2 GPU apps running too. BOINC can't schedule correctly other apps with team MB pack. But you could try to use BOINC 6.6.20 (recommended one now) and this cross-watch build w/o CPU teampack part. Not usre what result it will give, but maybe it will work OK.

[chelski]

I'm using MB_6.08_mod_CUDA_V11_def_func_FFTW_ESTIMATE_update in a V9 structure (e.g. without a number_of_cpus) in order to run AP on CPUs and MB on GPUS and still get the occasional freeze at 0%.  Is there a workaround for this structure - e.g. cross watch not between CPU-MB app and GPU-MB app but between AP and GPU-MB app to solve the freezing issue?

Please correct me if I'm wrong about how this thing works... things have really moved on rather fast and it is not that clear anymore which app / build to use for which setup.  Thanks

[Borgholio]

Currently running the non-team V9 pack on Boinc 6.6.20 and it's working fine aside from the occasional CUDA app freeze.  I have a task set under Windows to run benchmarks every half hour which frees up most stuck tasks but from time to time a task will freeze that requires a restart of BOINC.  If I were to download V11, would I need to upgrade from V9 to V10 first, or could I simply extract the CUDA app from V11 and modify my app-info to look at the new file instead?

[Raistmer]

I'm using MB_6.08_mod_CUDA_V11_def_func_FFTW_ESTIMATE_update in a V9 structure (e.g. without a number_of_cpus) in order to run AP on CPUs and MB on GPUS and still get the occasional freeze at 0%.  Is there a workaround for this structure - e.g. cross watch not between CPU-MB app and GPU-MB app but between AP and GPU-MB app to solve the freezing issue?

Please correct me if I'm wrong about how this thing works... things have really moved on rather fast and it is not that clear anymore which app / build to use for which setup.  Thanks

No, there is no such mod. But it's possible.

[Raistmer]

Currently running the non-team V9 pack on Boinc 6.6.20 and it's working fine aside from the occasional CUDA app freeze.  I have a task set under Windows to run benchmarks every half hour which frees up most stuck tasks but from time to time a task will freeze that requires a restart of BOINC.  If I were to download V11, would I need to upgrade from V9 to V10 first, or could I simply extract the CUDA app from V11 and modify my app-info to look at the new file instead?

You need correct name of executable file listed in app_info. /moreover, modded Vx CPU files will call CUDA MB executables only with fixed names (different for different versions). So filename should match.

[Borgholio]

So would it be easier then to simply upgrade to v11 completely?

[Raistmer]

So would it be easier then to simply upgrade to v11 completely?

surerecommended method now is to upgrade to BOINC 6.6.20 and to use its scheduling abilities instead of "teamed" pack.

[Borgholio]

surerecommended method now is to upgrade to BOINC 6.6.20 and to use its scheduling abilities instead of "teamed" pack.

Already running 6.6.20.    Just wondering about the cuda frozen task kill.

[HSchmirPo]

surerecommended method now is to upgrade to BOINC 6.6.20 and to use its scheduling abilities instead of "teamed" pack.

Already running 6.6.20.    Just wondering about the cuda frozen task kill.

The same here: updated to 6.6.20 and frozen task.
Should we combine 6.6.20 with v11?