I've seen that, for the specific case case of CPU workunits that had to
be suspended due
to BOINC running out of free memory it's allowed to use, the suspended
workunit is not
saved in a state that allows resuming from its current point when it has
not yet written
any checkpoints at all. It therefore kept starting from the beginning,
over and over.
In my case, this appears to be due to BOINC under 64-bit Windows Vista
needing a tighter
limit on 32-bit memory space it can use than on the total 64-bit memory
it can use.
Date: Thu, 16 May 2013 11:16:59 -0700 From: Eric J Korpela
<[email protected]> To: Raistmer the Sorcerer
<[email protected]> Cc: "[email protected]"
<[email protected]> Subject: Re: [boinc_dev] The ways in that
BOINC tells app to quit I had "keeps running when computer is in use"
problem last night with an Einstein CUDA app, so it's clear that this
isn't restricted to SETI@home. There has to be a BOINC issue or some
common flaw with how exiting is handled. I'm wondering is there's some
way that writing a checkpoint (or reporting a checkpoint) is going
wrong. I assume that even GPU apps would suspend rather than exit if
there has been no checkpoint since it was started? Of course in this
case it's not suspending either, it's continuing to run. David, any
comments? On Sun, May 12, 2013 at 11:47 PM, Raistmer the Sorcerer
<[email protected]>wrote:
> Then looks like I should report bug in BOINC API - sometimes it doesn't
>and set only "suspend" flag w/o "quit" flag.
>Possible relevant info: user had changed idle interval to 10 min instead
>of default.
>
>
>???????????, 12 ??? 2013, 21:06 -07:00 ?? David Anderson <
>[email protected]>:
> >
> >
> >On 12-May-2013 4:44 AM, Raistmer the Sorcerer wrote:
> >>Can I get definitive answer, please. In case when user set BOINC not to
>use
> >>GPU while user active (USE GPU only when PC idle): should BOINC set
> >>boinc_status.quit_request flag when user becomes active by BOINC devs
> >>opinion or should not?
> >
> >Yes.
> >
> >-- David
> >
> >>
> >>
> >>???????, 11 ??? 2013, 8:49 +04:00 ?? Raistmer the Sorcerer
> >><[email protected] >:
> >>>If you look corresponding thread you will see that app does check.
>There is
> >>>boinc_status.suspended flag also, but is this designed behavior, not
>to set
> >>>boinc_status.quit_request for GPU app in all cases when its
>exit=suspend
> >>>required ?
> >>>
> >>>
> >>>
> >>>???????, 10 ??? 2013, 17:24 -07:00 ?? Eric J Korpela <
> >>> [email protected] >:
> >>>>I can think of one other option. If an OpenCL routine never exits,
>the
> >>>>application might not get to the point of checking the flags (unless
>you
> >>>>are checking the flags while waiting on OpenCL routines to finish). I
> >>>>haven't checked the driver revision the users in question have
>installed.
> >>>>Are they using suspect driver versions?
> >>>>
> >>>>
> >>>>On Fri, May 10, 2013 at 4:46 PM, Raistmer the Sorcerer <
> >>>> [email protected] > wrote:
> >>>>>Trying to solve non-suspending issue listed in this thread:
> >>>>>
>http://setiathome.berkeley.edu/forum_thread.php?id=71581&postid=1365551
> >>>>>
> >>>>>
> >I came to conclusion that either BOINC does something wrong in this
>situation or
> >app doesn't check all needed flags and not aware about exit request.
> >>>>>
> >>>>>I check these flags: boinc_status.quit_request
> >>>>>boinc_status.abort_request
> >>>>>
> >>>>>Maybe some another flag BOINC sets to inform app that suspend
>required
_______________________________________________
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.
