Hi, I just bought two machines AMD-Vi capable with a NVIDIA TESLA. Has anyone tried to assign any NVIDIA to a virtual host and use CUDA on it? I assigned the NVIDIA tesla and installed the nvidia driver, as follows: root@tesla2:~# kvm -name sat -cpu host -smp 16 -daemonize -vnc :$1 \ -localtime -m 60g \ -net nic,macaddr=00:16:3e:aa:aa:05,model=virtio,vlan=112\ -net tap,vlan=112,ifname=sat112\ -drive file=/raw/sat2.raw,if=virtio,boot=on\ -device pci-assign,host=03:00.0 on my virtual host I get: root@sat:~# lspci |grep -i nvidia 00:04.0 3D controller: nVidia Corporation GT200 [Tesla C1060] (rev a1) and loaded the nvidia driver root@sat:~# dmesg ... [ 4.004588] NVRM: loading NVIDIA UNIX x86_64 Kernel Module 256.35 Wed Jun 16 18:42:44 PDT 2010 ... But when i run any app that uses CUDA a get the following on dmesg [ 136.157604] NVRM: RmInitAdapter failed! (0x23:0x2f:639) [ 136.157982] NVRM: rm_init_adapter(0) failed [ 136.161767] NVRM: RmInitAdapter failed! (0x23:0x2f:639) [ 136.162179] NVRM: rm_init_adapter(0) failed [ 136.165351] NVRM: RmInitAdapter failed! (0x23:0x2f:639) [ 136.165695] NVRM: rm_init_adapter(0) failed Did anyone got this working? Or someone can guarantee that it is impossible to do that? Thanks in advance, -- Bruno Ribas - ribas@xxxxxxxxxxxx http://www.inf.ufpr.br/ribas C3SL: http://www.c3sl.ufpr.br -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html