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

[TensorIR][M2a] Decompose-Reduction #9041

Merged
merged 16 commits into from
Oct 2, 2021
Merged

[TensorIR][M2a] Decompose-Reduction #9041

merged 16 commits into from
Oct 2, 2021

Conversation

spectrometerHBH
Copy link
Contributor

This PR is part of the TensorIR upstreaming effort (#7527), which adds the following schedule primitives:

decompose_reduction
Co-authored-by: Junru Shao junrushao1994@gmail.com
Co-authored-by: Ruihang Lai lairuihangdongdong@qq.com
Co-authored-by: Hongyi Jin 3231950289@qq.com
Co-authored-by: Wuwei Lin wuwei@apache.org
Co-authored-by: Siyuan Feng Hzfengsy@sjtu.edu.cn

@spectrometerHBH
Copy link
Contributor Author

@junrushao junrushao self-assigned this Sep 19, 2021
@junrushao
Copy link
Member

Sorry I have a deadline on Monday...Will review do after that

src/tir/schedule/primitive/reduction.cc Outdated Show resolved Hide resolved
src/tir/schedule/primitive/reduction.cc Outdated Show resolved Hide resolved
src/tir/schedule/primitive/reduction.cc Outdated Show resolved Hide resolved
Comment on lines +74 to +81
Stmt VisitStmt_(const SeqStmtNode* seq) final {
Array<Stmt> new_stmts;
new_stmts.reserve(seq->seq.size());
for (const Stmt& old_stmt : seq->seq) {
new_stmts.push_back(VisitStmt(old_stmt));
}
return SeqStmt::Flatten(new_stmts);
}
Copy link
Member

Choose a reason for hiding this comment

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

Why do we need special handling of SeqStmt btw?

@junrushao
Copy link
Member

@spectrometerHBH Sorry for the delay! I think I finally did a complete with the code review. Let me know if it makes sense to you

@spectrometerHBH
Copy link
Contributor Author

spectrometerHBH commented Sep 28, 2021

@spectrometerHBH Sorry for the delay! I think I finally did a complete with the code review. Let me know if it makes sense to you

@junrushao1994 Thanks!I might be slow to response in the next 2 days

@junrushao
Copy link
Member

Would be nice if we could get this merged this week or weekend :-)

Copy link
Member

@junrushao junrushao left a comment

Choose a reason for hiding this comment

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

@spectrometerHBH and I discussed for quite a few days and finally addressed all my concerns. Thanks for the great effort!

@junrushao junrushao merged commit 6b3fe95 into apache:main Oct 2, 2021
ylc pushed a commit to ylc/tvm that referenced this pull request Jan 7, 2022
This PR is part of the TensorIR upstreaming effort (apache#7527),
which adds the `decompose-reduction` scheduling primitive.

Co-authored-by: Junru Shao <junrushao1994@gmail.com>
Co-authored-by: Ruihang Lai <lairuihangdongdong@qq.com>
Co-authored-by: Hongyi Jin <3231950289@qq.com>
Co-authored-by: Wuwei Lin <wuwei@apache.org>
Co-authored-by: Siyuan Feng <Hzfengsy@sjtu.edu.cn>
ylc pushed a commit to ylc/tvm that referenced this pull request Jan 13, 2022
This PR is part of the TensorIR upstreaming effort (apache#7527),
which adds the `decompose-reduction` scheduling primitive.

Co-authored-by: Junru Shao <junrushao1994@gmail.com>
Co-authored-by: Ruihang Lai <lairuihangdongdong@qq.com>
Co-authored-by: Hongyi Jin <3231950289@qq.com>
Co-authored-by: Wuwei Lin <wuwei@apache.org>
Co-authored-by: Siyuan Feng <Hzfengsy@sjtu.edu.cn>
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.

5 participants