Hello, I’m a brand new student working with setting up Unet segmentations in university. I’m using Pytorch with two nvidia rtx 6000 ada and running windows 11 with WSL2 Ubuntu 24.04 LTS as required by my university. CUDA 12.8 and nvcc-12.8.
I try to use DataParallel tutorial code from pytorch and get the below errors.
Any and all assistance would be appreciated. Thank you!
3G72CW3:134448:134448 [0] NCCL INFO cudaDriverVersion 12080
3G72CW3:134448:134448 [0] NCCL INFO Bootstrap : Using eth0:172.19.58.132<0>
3G72CW3:134448:134448 [0] NCCL INFO NET/Plugin: No plugin found (libnccl-net.so)
3G72CW3:134448:134448 [0] NCCL INFO NET/Plugin: Plugin load returned 2 : libnccl-net.so: cannot open shared object file: No such file or directory : when loading libnccl-net.so
3G72CW3:134448:134448 [0] NCCL INFO NET/Plugin: Using internal network plugin.
NCCL version 2.21.5+cuda12.4
3G72CW3:134448:134484 [0] NCCL INFO NET/IB : No device found.
3G72CW3:134448:134484 [0] NCCL INFO NET/Socket : Using [0]eth0:172.19.58.132<0>
3G72CW3:134448:134484 [0] NCCL INFO Using non-device net plugin version 0
3G72CW3:134448:134484 [0] NCCL INFO Using network Socket
3G72CW3:134448:134485 [1] NCCL INFO Using non-device net plugin version 0
3G72CW3:134448:134485 [1] NCCL INFO Using network Socket
3G72CW3:134448:134484 [0] NCCL INFO ncclCommInitRank comm 0x1620c530 rank 0 nranks 2 cudaDev 0 nvmlDev 0 busId 1000 commId 0x100f7bc3fbfee128 - Init START
3G72CW3:134448:134485 [1] NCCL INFO ncclCommInitRank comm 0x16211e10 rank 1 nranks 2 cudaDev 1 nvmlDev 1 busId c1000 commId 0x100f7bc3fbfee128 - Init START
3G72CW3:134448:134485 [1] NCCL INFO Could not enable P2P between dev 1(=c1000) and dev 0(=1000)
3G72CW3:134448:134485 [1] NCCL INFO Could not enable P2P between dev 0(=1000) and dev 1(=c1000)
3G72CW3:134448:134485 [1] NCCL INFO Could not enable P2P between dev 1(=c1000) and dev 0(=1000)
3G72CW3:134448:134485 [1] NCCL INFO Could not enable P2P between dev 0(=1000) and dev 1(=c1000)
3G72CW3:134448:134484 [0] NCCL INFO Could not enable P2P between dev 1(=c1000) and dev 0(=1000)
3G72CW3:134448:134484 [0] NCCL INFO Could not enable P2P between dev 0(=1000) and dev 1(=c1000)
3G72CW3:134448:134484 [0] NCCL INFO Could not enable P2P between dev 1(=c1000) and dev 0(=1000)
3G72CW3:134448:134484 [0] NCCL INFO Could not enable P2P between dev 0(=1000) and dev 1(=c1000)
3G72CW3:134448:134484 [0] NCCL INFO comm 0x1620c530 rank 0 nRanks 2 nNodes 1 localRanks 2 localRank 0 MNNVL 0
3G72CW3:134448:134484 [0] NCCL INFO Channel 00/02 : 0 1
…
3G72CW3:134448:134484 [0] NCCL INFO group.cc:64 → 1 [Async thread]
3G72CW3:134448:134448 [0] NCCL INFO group.cc:418 → 1
3G72CW3:134448:134448 [0] NCCL INFO group.cc:95 → 1
3G72CW3:134448:134448 [0] NCCL INFO init.cc:1892 → 1