Questions and Answers :
Windows :
Problem with Cuda WU
Message board moderation
Author | Message |
---|---|
Nighthunter Send message Joined: 21 Jan 00 Posts: 2 Credit: 16,442,084 RAC: 0 |
All Cuda WU are ivalid or inconclusive, link: https://setiathome.berkeley.edu/results.php?hostid=7507044 Anyone can tell me "wheere is problem" ? Typicall Stderr is like this one: Stderr output <core_client_version>7.4.36</core_client_version> <![CDATA[ <stderr_txt> setiathome_CUDA: Found 1 CUDA device(s): Device 1: GeForce GTX 560 Ti, 1024 MiB, regsPerBlock 32768 computeCap 2.1, multiProcs 8 pciBusID = 1, pciSlotID = 0 clockRate = 1800 MHz In cudaAcc_initializeDevice(): Boinc passed DevPref 1 setiathome_CUDA: CUDA Device 1 specified, checking... Device 1: GeForce GTX 560 Ti is okay SETI@home using CUDA accelerated device GeForce GTX 560 Ti pulsefind: blocks per SM 4 (Fermi or newer default) pulsefind: periods per launch 100 (default) Priority of process set to BELOW_NORMAL (default) successfully Priority of worker thread set successfully setiathome enhanced x41zc, Cuda 4.20 Detected setiathome_enhanced_v7 task. Autocorrelations enabled, size 128k elements. Work Unit Info: ............... WU true angle range is : 2.713047 re-using dev_GaussFitResults array for dev_AutoCorrIn, 4194304 bytes re-using dev_GaussFitResults+524288x8 array for dev_AutoCorrOut, 4194304 bytes Thread call stack limit is: 1k cudaAcc_free() called... cudaAcc_free() running... cudaAcc_free() PulseFind freed... cudaAcc_free() Gaussfit freed... cudaAcc_free() AutoCorrelation freed... cudaAcc_free() DONE. Cuda sync'd & freed. Preemptively acknowledging a safe Exit. -> SETI@Home Informational message -9 result_overflow NOTE: The number of results detected equals the storage space allocated. Flopcounter: 157553587196.818360 Spike count: 0 Autocorr count: 30 Pulse count: 0 Triplet count: 0 Gaussian count: 0 Worker preemptively acknowledging an overflow exit.-> called boinc_finish Exit Status: 0 boinc_exit(): requesting safe worker shutdown -> boinc_exit(): received safe worker shutdown acknowledge -> Cuda threadsafe ExitProcess() initiated, rval 0 </stderr_txt> ]]> |
rob smith Send message Joined: 7 Mar 03 Posts: 22189 Credit: 416,307,556 RAC: 380 |
First thing that strikes me is that all have very short run times - a quick look at my rig with a GTX760 shows that the run times are substantially longer. Next, the number of "autocors" is 30, and so calculation will have stopped at that point. This is pointing to a problem with the GPU or the driver... A few questions: - did you notice when this string of invalid results started? - have you recently changed anything? - I see you are running the latest WHQL driver, did this driver come from Nvidia/Gforce or from Microsoft? Others may be along with further comments, suggestions or questions... Bob Smith Member of Seti PIPPS (Pluto is a Planet Protest Society) Somewhere in the (un)known Universe? |
BilBg Send message Joined: 27 May 07 Posts: 3720 Credit: 9,385,827 RAC: 0 |
 GeForce GTX 560 Ti is notorious for producing false overflows (e.g. "Autocorr count: 30") caused by too optimistic factory Overclock + too low GPU core voltage You need to lower MHz (of "shaders" / "CUDA cores") and/or up the GPU voltage For this you may use MSI Afterburner, EVGA Precision or similar tool: http://setiathome.berkeley.edu/forum_thread.php?id=65894&postid=1165854#1165854 jason_gee (the programmer of the CUDA apps): "The current school of thought is that some of these factory overclocked 560ti's need some extra GPU core voltage for stability running Cuda, the theory being that the manufacturers may have spent the effort getting them 'game stable' but overlooked Cuda/PhysX stability" http://setiathome.berkeley.edu/forum_thread.php?id=66124&postid=1170945#1170945 Links to voltage/MHz table posted by user "f_n_t": http://setiathome.berkeley.edu/forum_thread.php?id=63429&postid=1118632#1118632 Your: clockRate = 1800 MHz NVIDIA say: Processor Clock (MHz) 1645 or 1464 http://setiathome.berkeley.edu/forum_thread.php?id=69380&postid=1294814#1294814   - ALF - "Find out what you don't do well ..... then don't do it!" :)  |
Nighthunter Send message Joined: 21 Jan 00 Posts: 2 Credit: 16,442,084 RAC: 0 |
Thx a lot for answer. Best regards. |
©2024 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.