-
Notifications
You must be signed in to change notification settings - Fork 372
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
multi-gpu at runtime error #988
Comments
Hi @ecilay , Thanks for reporting the issue. What's the error message that you got? |
|
Thanks, @ecilay ! Hmm, doesn't have any clue. If it's possible, could you share a small repro that would help us investigate? Thanks! |
@chenyang78 I think you can repro by using any two AIT model (or maybe they could be the same model), load them on different GPUs, and do inference, see if it works? If it does, would appreciate sharing your inference scripts, thanks. |
So say if I have two AIT convered models,
model0
oncuda0
andmodel1
oncuda1
.Even if I used
cudaSetDevice
to load the models properly on each cuda device, at run time, after running inference on model0 on cuda0, model1 fails to run. Once I move two models into same devices, problem resolved.Is this expected? Any possible short term fix? I did the experiment on A10g with 4 GPUs.
The text was updated successfully, but these errors were encountered: