Skip to content

Fix cudaErrIllegalAddress by using correct device pointer #6862

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
Merged
mohanchen merged 1 commit into from
Jan 15, 2026
Merged

Fix cudaErrIllegalAddress by using correct device pointer #6862

mohanchen merged 1 commit into from
Jan 15, 2026

Conversation

@Cstandardlib
Copy link
Collaborator

@Cstandardlib Cstandardlib commented Jan 15, 2026 •
edited
Loading

Linked Issue

Fix #6850 #6563

What's changed?

  • The error was caused by passing a host pointer to a device function matrix_mul_vector_op.
  • CPU vector eigenvalue_iter->data() on the host is replaced with the device copy e_temp_hd in the call.

@Cstandardlib Cstandardlib added Bugs Bugs that only solvable with sufficient knowledge of DFT GPU & DCU & HPC GPU and DCU and HPC related any issues Diago Issues related to diagonalizaiton methods labels Jan 15, 2026
@mohanchen mohanchen merged commit 880cd2f into deepmodeling:develop Jan 15, 2026
14 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@mohanchen mohanchen mohanchen approved these changes

Assignees

No one assigned

Labels

Bugs Bugs that only solvable with sufficient knowledge of DFT Diago Issues related to diagonalizaiton methods GPU & DCU & HPC GPU and DCU and HPC related any issues

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

Davidson subspace diagonalization method crashes at basically any CUDA GPU

2 participants

@Cstandardlib @mohanchen