Skip to content

ci: upgrade apache-tvm-ffi==0.1.0b15 #1860

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

Merged
merged 1 commit into from
Oct 4, 2025
Merged

ci: upgrade apache-tvm-ffi==0.1.0b15 #1860

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 4 additions & 4 deletions ci/docker-tags.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,4 +1,4 @@
flashinfer/flashinfer-ci-cu126: 20251002-52089b5
flashinfer/flashinfer-ci-cu128: 20251002-52089b5
flashinfer/flashinfer-ci-cu129: 20251002-52089b5
flashinfer/flashinfer-ci-cu130: 20251002-52089b5
flashinfer/flashinfer-ci-cu126: 20251004-ef9dd60
flashinfer/flashinfer-ci-cu128: 20251004-ef9dd60
flashinfer/flashinfer-ci-cu129: 20251004-ef9dd60
flashinfer/flashinfer-ci-cu130: 20251004-ef9dd60
2 changes: 1 addition & 1 deletion flashinfer-cubin/pyproject.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,5 +1,5 @@
[build-system]
requires = ["setuptools>=61.0", "wheel", "requests", "filelock", "torch", "tqdm", "numpy", "apache-tvm-ffi==0.1.0b11"]
requires = ["setuptools>=61.0", "wheel", "requests", "filelock", "torch", "tqdm", "numpy", "apache-tvm-ffi==0.1.0b15"]
Copy link
Contributor

Choose a reason for hiding this comment

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

medium

The version for apache-tvm-ffi is hardcoded here. This version is also specified in flashinfer-jit-cache/pyproject.toml and the root setup.py. This duplication can make future dependency updates error-prone and lead to inconsistencies.

To improve maintainability, I recommend centralizing dependency versions to have a single source of truth. While this can be challenging for build-system.requires in a monorepo structure, it's worth exploring solutions to avoid this duplication.

build-backend = "build_backend"
backend-path = ["."]

Expand Down
2 changes: 1 addition & 1 deletion flashinfer-jit-cache/pyproject.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,5 +1,5 @@
[build-system]
requires = ["setuptools>=77", "packaging>=24", "wheel", "torch", "ninja", "requests", "numpy", "nvidia-ml-py", "nvidia-nvshmem-cu12", "apache-tvm-ffi==0.1.0b11"]
requires = ["setuptools>=77", "packaging>=24", "wheel", "torch", "ninja", "requests", "numpy", "nvidia-ml-py", "nvidia-nvshmem-cu12", "apache-tvm-ffi==0.1.0b15"]
Copy link
Contributor

Choose a reason for hiding this comment

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

medium

Similar to other configuration files in this PR, the version for apache-tvm-ffi is hardcoded here and duplicated in other files. Centralizing dependency versions would improve project maintainability and reduce the risk of inconsistencies.

build-backend = "build_backend"
backend-path = ["."]

Expand Down
2 changes: 1 addition & 1 deletion setup.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -51,7 +51,7 @@ def generate_build_meta() -> None:
"click",
"tqdm",
"tabulate",
"apache-tvm-ffi==0.1.0b11",
"apache-tvm-ffi==0.1.0b15",
Copy link
Contributor

Choose a reason for hiding this comment

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

medium

The apache-tvm-ffi version is hardcoded here, and also in flashinfer-cubin/pyproject.toml and flashinfer-jit-cache/pyproject.toml. To improve maintainability and avoid potential inconsistencies, consider defining this version in a single place and referencing it from all three locations.

"packaging>=24.2",
"nvidia-cudnn-frontend>=1.13.0",
"nvidia-cutlass-dsl>=4.2.1",
Expand Down