

I’ve downgraded to 430.40 then 430.26 and 418.87 (which was part of cuda 10.1.243 package).įor 418 I even downgraded to kernel-4.18.86_64 to properly install cuda 10.1 with its driver. Did you try downgrading the driver to either v430 or v418 to check if this is a regression? Since it’s ‘N’=No=Off this error message is really odd. Thread 481 "resolve" received signal SIGABRT, Aborted.Ġx00007fffc76a457f in raise () from /lib64/libc.so.6 Supports MultiDevice Co-op Kernel Launch: Noĭevice PCI Domain ID / Bus ID / location ID: 0 / 3 / 0ĭeviceQuery, CUDA Driver = CUDART, CUDA Driver Version = 10.1, CUDA Runtime Version = 10.0, NumDevs = 1 Support host page-locked memory mapping: Yesĭevice supports Unified Addressing (UVA): Yes Max dimension size of a grid size (x,y,z): (2147483647, 65535, 65535)Ĭoncurrent copy and kernel execution: Yes with 2 copy engine(s)

Max dimension size of a thread block (x,y,z): (1024, 1024, 64) Maximum number of threads per block: 1024 Maximum number of threads per multiprocessor: 2048 Total number of registers available per block: 65536 Total amount of shared memory per block: 49152 bytes

Total amount of constant memory: 65536 bytes Maximum Layered 1D Texture Size, (num) layers 1D=(16384), 2048 layers (24) Multiprocessors, (128) CUDA Cores/MP: 3072 CUDA Cores

Total amount of global memory: 12212 MBytes (12805668864 bytes) |-+-+-+ĬUDA Device Query (Runtime API) version (CUDART static linking)ĬUDA Driver Version / Runtime Version 10.1 / 10.0ĬUDA Capability Major/Minor version number: 5.2 lib/modprobe.d/nf:options nouveau modeset=0 $ grep modeset /etc/modprobe.d/* /lib/modprobe.d/* /etc/modprobe.d/nf:options nouveau modeset=0 I saw some other people having similar error and crash’es so I was hoping that someone here might have some ideas ] *ERROR* Failed to lookup gem object for fence attach: 0x00000004 When that happens, I am getting drm errors under dmesg: nvidia-uvm: Loaded the UVM driver, major device number 234. I have and issue with one app ‘DaVinci Resolve 16’ that crashes on load.
