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.

Sign up for GitHub

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

Jump to bottom

CUDA: fix LoRAs #3130

Merged
merged 1 commit into from
Sep 12, 2023
Merged

CUDA: fix LoRAs #3130

merged 1 commit into from
Sep 12, 2023

Conversation

JohannesGaessler
Copy link
Collaborator

As pointed out by #3110 (comment) , the CUDA code for LoRAs was broken by #3110 . This PR fixes this.

slaren
slaren reviewed Sep 11, 2023
View reviewed changes
ggml-cuda.cu Show resolved Hide resolved
@JohannesGaessler
Copy link
Collaborator Author

Previously ggml_cpy_tensor_2d was not called for LoRAs. As it turns out the logic for src0_on_device was incorrect so the tensor was being copied unnecessarily. I'm keeping the extended logic for ggml_cpy_tensor_2d.

@JohannesGaessler JohannesGaessler merged commit 4f7cd6b into ggerganov:master Sep 12, 2023
28 checks passed
pkrmf pushed a commit to morlockstudios-com/llama.cpp that referenced this pull request Sep 26, 2023
@JohannesGaessler @pkrmf
CUDA: fix LoRAs (ggerganov#3130)
b958c90
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@slaren slaren slaren approved these changes

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
2 participants
@JohannesGaessler @slaren