On 13 January 2011 00:22, Satish Balay <balay at mcs.anl.gov> wrote: > There is deviceQuery example that shows the version. Perhaps we have > to write a small test that replicates this example code. > > balay at bb30:~/NVIDIA_GPU_Computing_SDK/C/bin/linux/release>./deviceQueryDrv > |grep Capability > ?CUDA Capability Major/Minor version number: ? ?1.3 > ?CUDA Capability Major/Minor version number: ? ?1.3 > > Potential issues: > - the machine has 2 nvidia cards with different version > - the card is locked by a different user - running cuda code [at configure > time] > - the compile node doesn't have the nvidia cards [like on a cluster..] > > satish >
What about starting by letting users pass --with-cuda-arch=sm_20 to configure? -- Lisandro Dalcin --------------- CIMEC (INTEC/CONICET-UNL) Predio CONICET-Santa Fe Colectora RN 168 Km 472, Paraje El Pozo Tel: +54-342-4511594 (ext 1011) Tel/Fax: +54-342-4511169
