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

[OMPD] The LLVM OMPD ompd_device_initialize() function has a few problems #63

Open
jdelsign opened this issue Feb 27, 2019 · 0 comments

Comments

@jdelsign
Copy link

The LLVM OMPD ompd_device_initialize() has a few problems for NVIDIA devices (based on a brief review I did of the sources):

  • It does not check the kind argument to make sure that it is OMPD_DEVICE_KIND_CUDA.
  • It does not check that the sizeof_id is 8.
  • It assumes that (*id) is 64-bit aligned, though that's probably OK if the size is 8.
  • It saves the id argument in ompd_address_space_handle_t::id, which is the pointer to the 64-bit context, not context value itself. But, I'm not sure that field is used anywhere.
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

1 participant