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

[metal] Add kernel side memory allocator #1175

Merged
merged 3 commits into from
Jun 11, 2020
Merged

[metal] Add kernel side memory allocator #1175

merged 3 commits into from
Jun 11, 2020

Conversation

k-ye
Copy link
Member

@k-ye k-ye commented Jun 7, 2020

This PR has no functionality changes. It just reimplements Metal's ListManager via the dynamic memory allocator on the GPU. All the existing tests have passed, and taichi_sparse.py ran just fine.

The concrete design is listed in #1174 itself.

Related issue = #1174

[Click here for the format server]

@k-ye k-ye requested a review from yuanming-hu June 7, 2020 11:46
Copy link
Member

@yuanming-hu yuanming-hu left a comment

Choose a reason for hiding this comment

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

Thanks! Due to time constraints, I could only take a very rough pass which clearly isn't enough to identify potential problems. Feel free to merge - I don't think I will have enough time recently to do a very careful review on this... Sorry.

@archibate archibate added the LGTM label Jun 10, 2020
metal::memory_order_relaxed);
}

[[maybe_unused]] PtrOffset mtl_listmgr_ensure_chunk(thread ListManager *l,
Copy link
Member Author

Choose a reason for hiding this comment

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

Thanks! Honestly I think this mtl_listmgr_ensure_chunk is all that deserves some attention. It implements the memory allocation..

@k-ye k-ye merged commit 482e82d into taichi-dev:master Jun 11, 2020
@k-ye k-ye deleted the mem branch June 11, 2020 11:47
@FantasyVR FantasyVR mentioned this pull request Jun 14, 2020
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.

4 participants