• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Core running in safe mode?!

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

Lancelot

Member
Joined
Feb 12, 2001
Location
the Netherlands
I've noticed this before; after a system crash (last night we had a power-outage due to a storm) the F@H client runs extremely slow. This morning I restarted the fileserver in my sig and when I got home noticed it took approx. 1hr and 5mins to complete a step! (project3040) WTF!? After a shutdown and normal Boot of the system it is now taking about 28mins per step!? Is the client running in some kind of safe mode after an improper core shutdown or what?

First it does this:
[06:14:00] Preparing to commence simulation
[06:14:00] - Ensuring status. Please wait.
[06:14:17] - Looking at optimizations...
[06:14:17] - Working with standard loops on this execution.
[06:14:17] - Previous termination of core was improper.
[06:14:17] - Files status OK
[06:14:17] - Expanded 290961 -> 1461493 (decompressed 502.2 percent)
[06:14:17]
[06:14:17] Project: 3040 (Run 16, Clone 108, Gen 16)
[06:14:17]
[06:14:17] Entering M.D.
[06:14:38] (Starting from checkpoint)
[06:14:38] Protein: p3040_supervillin-03
[06:14:38]
[06:14:38] Writing local files
[06:14:38] Completed 1000000 out of 5000000 steps (20)
[07:21:15] Writing local files
[07:21:15] Completed 1050000 out of 5000000 steps (21)
[08:27:21] Writing local files
[08:27:21] Completed 1100000 out of 5000000 steps (22)
[09:33:33] Writing local files
[09:33:33] Completed 1150000 out of 5000000 steps (23)
[10:39:42] Writing local files
[10:39:42] Completed 1200000 out of 5000000 steps (24)
[11:45:44] Writing local files
[11:45:44] Completed 1250000 out of 5000000 steps (25)
[12:51:47] Writing local files
[12:51:47] Completed 1300000 out of 5000000 steps (26)
[13:57:46] Writing local files
[13:57:46] Completed 1350000 out of 5000000 steps (27)
[15:03:48] Writing local files
[15:03:48] Completed 1400000 out of 5000000 steps (28)

Folding@Home Client Shutdown.

And after a reboot it does this;

[16:58:16] Preparing to commence simulation
[16:58:16] - Looking at optimizations...
[16:58:16] - Files status OK
[16:58:16] - Expanded 290961 -> 1461493 (decompressed 502.2 percent)
[16:58:17]
[16:58:17] Project: 3040 (Run 16, Clone 108, Gen 16)
[16:58:17]
[16:58:17] Assembly optimizations on if available.
[16:58:17] Entering M.D.
[16:58:38] (Starting from checkpoint)
[16:58:38] Protein: p3040_supervillin-03
[16:58:38]
[16:58:38] Writing local files
[16:58:38] Completed 1428128 out of 5000000 steps (29)
[16:58:38] [16:58:16] Preparing to commence simulation
[16:58:16] - Looking at optimizations...
[16:58:16] - Files status OK
[16:58:16] - Expanded 290961 -> 1461493 (decompressed 502.2 percent)
[16:58:17]
[16:58:17] Project: 3040 (Run 16, Clone 108, Gen 16)
[16:58:17]
[16:58:17] Assembly optimizations on if available.
[16:58:17] Entering M.D.
[16:58:38] (Starting from checkpoint)
[16:58:38] Protein: p3040_supervillin-03
[16:58:38]
[16:58:38] Writing local files
[16:58:38] Completed 1428128 out of 5000000 steps (29)
[16:58:38] Extra SSE boost OK.
[17:10:18] Writing local files
[17:10:18] Completed 1450000 out of 5000000 steps (29)
[17:37:03] Writing local files
[17:37:03] Completed 1500000 out of 5000000 steps (30)
[18:03:44] Writing local files
[18:03:44] Completed 1550000 out of 5000000 steps (31)
[18:30:25] Writing local files
[18:30:25] Completed 1600000 out of 5000000 steps (32)
[17:10:18] Writing local files
[17:10:18] Completed 1450000 out of 5000000 steps (29)
[17:37:03] Writing local files
[17:37:03] Completed 1500000 out of 5000000 steps (30)
[18:03:44] Writing local files
[18:03:44] Completed 1550000 out of 5000000 steps (31)
[18:30:25] Writing local files
[18:30:25] Completed 1600000 out of 5000000 steps (32)

See the time difference per step?! I notice the line "Extra SSE boost OK." is missing after the first start, could that be it?!
 
Last edited:
Yes, it is:
06:14:17] - Working with standard loops on this execution.
[06:14:17] - Previous termination of core was improper.

That's why you want "forceasm" in your initialization string for the FAH client. :burn: :burn:

Standard loops = no SSE.
 
That's the first time I've ever heard about that forceasm thingy. How and where do I put that? My rig could've done twice the work today, grrrr!
 
Lancelot said:
That's the first time I've ever heard about that forceasm thingy. How and where do I put that? My rig could've done twice the work today, grrrr!

it is a command line switch -forceasm. If running from the desktop/shortcut add it to the end of the target string (separated by a blank) Running as a service you need to add it to the parms in the registry.

I always put it on every new instance i setup along with the -verbosity 9 switch.
 
Last edited:
ihrsetrdr said:
Hey 7im, what's the status over on FC.org?

Yes, the Folding-Community Support forum is down temporarily. Please be patient while the ISP hosting the site works to correct the problem.

You can watch the Stanford News page for updates. http://folding.stanford.edu/news.html

Or just post here in your own forum. I'll be dropping by a little more frequently until the support site is back up.

Thanks.
 
ihrsetrdr said:
Thanks 7im.
Hey, speaking of the -forceasm argument, didn't someone mention that this flag is irrelevant on a Linux box?

IIRC, that's half right. Don't need -forceasm on a Linux box running the SMP client. -forceasm is hard coded to on in the SMP beta. ;)
 
7im said:
IIRC, that's half right. Don't need -forceasm on a Linux box running the SMP client. -forceasm is hard coded to on in the SMP beta. ;)

Thanks for the clarification, 7im. It's better to be half right than uh, howze that go...oh yea, having a glass that's half empty!

:D
 
Back