Joined: 27 Jul 99
I have been seeing this happening in Windows XP Home and Professional on various of my systems utilizing AMD processors but never on Intel ones. If I suspend the one chewing up time and doing nothing, it will start another unit that usually will do the same thing. After about 3 or more suspensions one unit will finally take off. To get the suspended ones to run sometimes quitting BOINC and restarting it will get the unit to run and other times it requires a system reboot to get them to run. This is annoying and wasteful. Either some trash is not cleaned in the code from the previous run; a temp file is not closed/deleted or something, a variable not being reset, or some reason the code stops in the 'Choosing' routine. Seems rather sloppy coding maybe on the AMD processor side. This needs to be looked into as it has been happening for quite a few releases of BOINC now.
Joined: 9 Jun 99
Seeing how I don't have any problems, and never had on any of my systems, I'm inclined to believe it's something on your own systems that's causing this. If it were really sloppy coding everyone would have this problem. I don't see several threads on any of the forums of people saying they have this problem.
And if it were sloppy coding, it'd be in the science application, since BOINC is just a managing program. It doesn't do any of the active crunching (and thus hanging or not of a task).
With all that said, what else is running on all your affected systems that could cause this?
Ancient Astronaut Theorists suggest that in many ways, you can be considered an alien conspiracy!
Joined: 27 May 07
I have been seeing this happening in Windows XP Home and Professional on various of my systems utilizing AMD processors but never on Intel ones.
(Please, try to type in paragraphs, I stopped reading after the first sentence, it's hard to follow 'raw' text)
Yes, it was reported by some users that some AMD CPUs sometimes hang at "Optimal function choice":
(it is hard to test/debug/fix something that happens only for some/sometimes)
And of course this code is not in BOINC but in standard SETI app so it's not strange that it "has been happening for quite a few releases of BOINC now".
(in the links is given possible solution/workaround - using optimized applications)
SETI@home v7 app is in works so it may fix this problem.
- ALF - "Find out what you don't do well ..... then don't do it!" :)
©2016 University of California
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.