Advanced search

Message boards : Number crunching : SWAN : FATAL : Cuda driver error 3 in file 'swanlibnv2.cpp' in line 446.

Author Message
pvh
Send message
Joined: 17 Mar 10
Posts: 23
Credit: 1,173,824,416
RAC: 0
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 36133 - Posted: 7 Apr 2014 | 10:22:02 UTC

Since 5 April I am getting lots of these errors:

<core_client_version>7.2.33</core_client_version>
<![CDATA[
<message>
process exited with code 199 (0xc7, -57)
</message>
<stderr_txt>
SWAN : FATAL : Cuda driver error 3 in file 'swanlibnv2.cpp' in line 446.
# SWAN swan_assert -57

</stderr_txt>
]]>


The WUs immediately crash after starting up. Some of my WUs make it through to the end and validate OK, but most crash like this. This happens both to short and long WUs on both my rigs. Before April 5 both rigs ran tasks just fine... Any clues what is going on here? I am running on 64-bit Linux (openSUSE 13.1). My driver version is 331.20.

Richard Haselgrove
Send message
Joined: 11 Jul 09
Posts: 1620
Credit: 9,183,509,048
RAC: 19,199,223
Level
Tyr
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 36134 - Posted: 7 Apr 2014 | 10:37:02 UTC - in response to Message 36133.

Looks like all your errors are with the cuda60 version of the app, including the v8.21 which MJH deployed yesterday. We'd best let him know.

[It's a project problem, which they should be able to fix in a day or two. Not your fault. I think.]

Profile Stoneageman
Avatar
Send message
Joined: 25 May 09
Posts: 224
Credit: 34,057,374,498
RAC: 35
Level
Trp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 36137 - Posted: 7 Apr 2014 | 11:23:20 UTC

With those cards your best option is to upgrade to 331.49 which should see little impact on performance.

pvh
Send message
Joined: 17 Mar 10
Posts: 23
Credit: 1,173,824,416
RAC: 0
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 36167 - Posted: 8 Apr 2014 | 18:31:17 UTC - in response to Message 36137.

With those cards your best option is to upgrade to 331.49 which should see little impact on performance.


So far it looks like this solved the problem, thanks!

Profile microchip
Avatar
Send message
Joined: 4 Sep 11
Posts: 110
Credit: 326,102,587
RAC: 0
Level
Asp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 36185 - Posted: 9 Apr 2014 | 16:26:25 UTC
Last modified: 9 Apr 2014 | 16:27:26 UTC

Same here on Linux with latest beta NV driver on Short. Most tasks error out immediately and every once in a while a task completes. All tasks are from SANTI
____________

Team Belgium

Michal Kinďura
Send message
Joined: 22 Jun 11
Posts: 2
Credit: 136,044,421
RAC: 0
Level
Cys
Scientific publications
watwatwatwatwatwatwat
Message 36187 - Posted: 9 Apr 2014 | 18:58:22 UTC

Hi,
I have nV GTS450 and I have a problem with the cuda60 WUs.

http://www.gpugrid.net/results.php?userid=77759&offset=0&show_names=0&state=5&appid=

cuda55 works correctly.

Profile microchip
Avatar
Send message
Joined: 4 Sep 11
Posts: 110
Credit: 326,102,587
RAC: 0
Level
Asp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 36574 - Posted: 23 Apr 2014 | 7:49:19 UTC
Last modified: 23 Apr 2014 | 7:49:40 UTC

Bump?

Still having issues with the CUDA6 tasks. most error out on my GTX 560. They only seem to run reliably on my low-end GT 440. CUDA5 tasks have no problem on both my cards
____________

Team Belgium

Profile skgiven
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3968
Credit: 1,995,359,260
RAC: 0
Level
His
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 36576 - Posted: 23 Apr 2014 | 8:00:00 UTC - in response to Message 36574.

Maybe it's been fixed:

686x-SANTI_MAR419cap310-60-84-RND9954_0 6455855 21 Apr 2014 | 1:23:59 UTC 21 Apr 2014 | 7:15:39 UTC Completed and validated 20,788.57 2,199.78 18,300.00 Short runs (2-3 hours on fastest card) v8.21 (cuda60)

1_4_9-NATHAN_CMYB_run1-18-40-RND4568_0 6450430 20 Apr 2014 | 0:37:10 UTC 20 Apr 2014 | 5:43:28 UTC Completed and validated 17,994.03 1,632.46 16,650.00 Short runs (2-3 hours on fastest card) v8.21 (cuda60)
____________
FAQ's

HOW TO:
- Opt out of Beta Tests
- Ask for Help

Post to thread

Message boards : Number crunching : SWAN : FATAL : Cuda driver error 3 in file 'swanlibnv2.cpp' in line 446.

//