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

TEST/GTEST: Added cuda gpu switching testing. #10388

Draft
wants to merge 1 commit into
base: master
Choose a base branch
from

Conversation

rakhmets
Copy link
Collaborator

What?

Added CUDA GPU switching testing.

The test should fail on a system with more than 1 GPU. Including this error in the output:

cuda_copy_md.c:649  UCX  ERROR cuMemGetAddressRange(0x7f8cd9e00000) error: named symbol not found

Test passes with this patch #9645.

@rakhmets rakhmets force-pushed the topic/gtest-switch-gpu branch from d31bb61 to 1adcb5e Compare December 17, 2024 18:09
class test_ucp_mmap_mgpu : public ucs::test {
};

UCS_TEST_F(test_ucp_mmap_mgpu, switch_gpu) {
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

do we need to add tests for transfer cuda_copy or cuda_ipc, mem type cuda or vmm/mallocasync:

  • buf1 on device1, buf2 on device2, copy happening under progress with context of unrelated device3

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Agree, we will need more test cases.
The test in this PR is the simplest one. And it fails. Once it is fixed, we will add more scenarios.

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

Successfully merging this pull request may close these issues.

2 participants