The least capable device affected by particular driver version bug. Please re-read opening post. The project can to not send tasks for ALL NV devices under particular driver version to avoid wrong computations. This would mean loss in performance though. Cause capable devices will be banned along with incapable ones and BOINC provides no measures to discriminate between them.
So yes, workaround is possible. And no, it's not proper solution to issue. Consider this as design enhancement request with example why it is needed. Thu, 2 Oct 2014 15:59:47 +0000 от "McLeod, John" <[email protected]>: >Can the project send work that will run on the least capable device and have >it run on both? > >Sent from my Android phone using TouchDown (www.nitrodesk.com) > >-----Original Message----- >From: Raistmer the Sorcerer [[email protected]] >Received: Thursday, 02 Oct 2014, 11:47AM >To: [email protected] [[email protected]] >Subject: Re: [boinc_dev] BOINCscheduling & plan class possible issue > >This solution requires user intervention. The question is how to avoid this >server-side. User can be not aware that his host generates invalid result and >project can't prevent it now. > > >Thu, 02 Oct 2014 17:06:05 +0200 от David Anderson <[email protected]>: >>Yes. >>You can also use <exclude_gpu> to be more specific >>about which GPU to use for which project. >> >>On 02-Oct-2014 4:59 PM, McLeod, John wrote: >>> Would that then prevent BOINC from running tasks on some of the GPS for any >>> tasks? >>> >>> Sent from my Android phone using TouchDown ( www.nitrodesk.com ) >>> >>> -----Original Message----- >>> *From:* David Anderson [[email protected]] >>> *Received:* Thursday, 02 Oct 2014, 10:58AM >>> *To:* [email protected] [[email protected]] >>> *Subject:* Re: [boinc_dev] BOINCscheduling & plan class possible issue >>> >>> I assume these are cases where the user has set <use_all_gpus>. >>> If that flag is removed the problem should go away. >>> -- David >>> >>> On 02-Oct-2014 4:28 PM, Raistmer the Sorcerer wrote: >>>> It was discovered that OpenCL AstroPulse application in SETI can't >>>> properly work >>>> under 340.52 nVidia driver on pre-FERMI hardware (single pulses missing, >>>> issue >>>> reported and reproduced by NV already). To avoid wrong computations Eric >>>> created >>>> 2 plan classes that should avoid work sending on pre-FERMI GPUs running >>>> under >>>> that faulty driver. >>>> >>>> But it was discovered (look here: >>>> >>>> http://setiweb.ssl.berkeley.edu/beta/forum_thread.php?id=2182&postid=52632 >>>> ) >>>> that multi-GPU hosts containing both FERMI and pre-FERMI GPUs are able to >>>> recive >>>> tasks under another common plan class for FERMI devices and execute those >>>> recived >>>> tasks on pre-FERMI devices. That is, though server obey plan class and >>>> ignore >>>> pre-FERMI devices on 340.52 driver, BOINC client doesn't obey plan class >>>> and >>>> schedule to run tasks for FERMI devices on pre-FERMI ones (when both >>>> present in >>>> host). >>>> >>>> Could someone confirm this issue? And how one should act to exclude >>>> pre-FERMI >>>> devices on 340.52 driver in such case of mixed NV devices in host? >>>> >>>> >>>> >>> _______________________________________________ >>> boinc_dev mailing list >>> [email protected] >>> http://lists.ssl.berkeley.edu/mailman/listinfo/boinc_dev >>> To unsubscribe, visit the above URL and >>> (near bottom of page) enter your email address. >>_______________________________________________ >>boinc_dev mailing list >>[email protected] >> http://lists.ssl.berkeley.edu/mailman/listinfo/boinc_dev >>To unsubscribe, visit the above URL and >>(near bottom of page) enter your email address. > >_______________________________________________ >boinc_dev mailing list >[email protected] >http://lists.ssl.berkeley.edu/mailman/listinfo/boinc_dev >To unsubscribe, visit the above URL and >(near bottom of page) enter your email address. _______________________________________________ boinc_dev mailing list [email protected] http://lists.ssl.berkeley.edu/mailman/listinfo/boinc_dev To unsubscribe, visit the above URL and (near bottom of page) enter your email address.
