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

Don't do a kernel search by mangled string every time #100

Open
eyalroz opened this issue Dec 17, 2024 · 0 comments
Open

Don't do a kernel search by mangled string every time #100

eyalroz opened this issue Dec 17, 2024 · 0 comments
Assignees
@eyalroz
Labels
task

Comments

@eyalroz
Copy link
Owner

eyalroz commented Dec 17, 2024

We shouldn't have to search for the kernel in the module every run; let's just find it once and keep it in the CUDA-specific member of the execution context, explicitly. This would probably let us avoid having to keep the mangled name there.
@eyalroz eyalroz self-assigned this Dec 17, 2024
@eyalroz eyalroz added the task label Dec 17, 2024
eyalroz pushed a commit that referenced this issue Dec 19, 2024
Fixes #99 + #100: CUDA changes related to static shared memory and ke…
1ac081c 
…rnel wrappers

* Not search for the mangled kernel by name with every run
* Execution context now explicitly holds a `cuda::kernel_t`
* Execution context no longer holds the mangled kernel name
* Now printing the kernel's amount of static shared memory used, after the build
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@eyalroz eyalroz

Labels
task
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
1 participant
@eyalroz