Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

p2pBandwidthLatencyTest has an oom error #276

Open
dusir opened this issue Jun 19, 2024 · 2 comments
Open

p2pBandwidthLatencyTest has an oom error #276

dusir opened this issue Jun 19, 2024 · 2 comments

Comments

@dusir
Copy link

dusir commented Jun 19, 2024

Hi!

We have a Gefroce RTX 3090 GPU oom error:

The env info:
My GPU is Geforce RTX 3090, 8 GPU device,the GPU driver is 535.171.04, cuda is 12.2.2_535.104.05 ,a linux os(CentOS Linux release 7.6.1810 (Core) ),the kernel version is 3.10.0-1160.31.1.el7.x86_64

the kernel cmdline:
BOOT_IMAGE=/boot/vmlinuz-3.10.0-1160.31.1.el7.x86_64 root=UUID=4b499d76-769a-40a0-93dc-4a31a59add28 ro crashkernel=auto console=ttyS0,115200 console=tty0 panic=5 net.ifnames=0 biosdevname=0 intel_idle.max_cstate=1 intel_pstate=disable nvidia.NVreg_NvLinkDisable=1

the problem is as follows:
image

p2pBandwidthLatencyTest test occurs an out of memory error,but the GPU memory usage is low, mem used very low .

and the 470 driver has no this issue,is this an driver bug? Please help me.

@icenotice
Copy link

I also encountered this problem. This problem can easily affect application startup. I hope NVIDIA will support and solve it as soon as possible.

@zpaixuanxuan
Copy link

I also encountered the problem of oom error. Is there any solution in the community?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants