I checked in the following; tasks with missing GPUs will be
marked as aborted (and reported the server)
rather than just being discarded.
The complexity of the "limbo state" idea outweighs the benefit.
-- David
- client: better behavior if a GPU goes away:
1) if an APP_VERSION is missing a coprocessor,
don't delete it and its files.
(If the coprocessor returns, we won't need to re-download)
2) if a RESULT uses an app version that is missing a coprocessor,
abort it (rather than deleting it).
The client will report the result on the next scheduler RPC,
and the server will make a new instance.
Raistmer wrote:
> Why BOINC should trash whole GPU cache if GPU was not found at current BOINC
> start ?
> Why tasks just can't be put in waiting or suspended (to not participate in
> scheduling decisions) state until GPU will be available again ?
>
> _______________________________________________
> 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.
