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

initial support blackwell #747

Open
wants to merge 5 commits into
base: main
Choose a base branch
from
Open

initial support blackwell #747

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
0c2a721
initial support blackwell
johnnynunez Jan 21, 2025
5523aac
initial support blackwell
johnnynunez Jan 21, 2025
011036f
Update installation.rst
johnnynunez Jan 23, 2025
8c308d5
Update release_wheel.yml
johnnynunez Jan 23, 2025
aac6545
Merge branch 'main' into main
johnnynunez Jan 25, 2025
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
6 changes: 3 additions & 3 deletions .github/workflows/release_wheel.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -18,16 +18,16 @@ on:
# required: true

env:
TORCH_CUDA_ARCH_LIST: "7.5 8.0 8.9 9.0+PTX"
TORCH_CUDA_ARCH_LIST: "7.5 8.0 8.9 9.0 10.0 12.0+PTX"

jobs:
build:
strategy:
fail-fast: false
matrix:
python: ["3.8", "3.9", "3.10", "3.11", "3.12"]
cuda: ["11.8", "12.1", "12.4"]
torch: ["2.3", "2.4", "2.5"]
cuda: ["11.8", "12.1", "12.4", "12.6", "12.8"]
torch: ["2.3", "2.4", "2.5", "2.6"]
exclude: # for cuda 12.4, we only support torch 2.4+
- cuda: "12.4"
torch: "2.2"
Expand Down
2 changes: 1 addition & 1 deletion 3rdparty/cutlass
Open in desktop
Submodule cutlass updated 2034 files
2 changes: 1 addition & 1 deletion cmake/config.cmake
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -35,7 +35,7 @@ set(FLASHINFER_GEN_MASK_MODES 0 1 2)
# "native" is a special value for CMAKE_CUDA_ARCHITECTURES which means use the
# architectures of the host's GPU. it's new in CMake 3.24, if you are using an
# older of CMake or you want to use a different value, you can set its value
# here. Supported CUDA architectures include 80;86;89;90
# here. Supported CUDA architctures include 80;86;89;90;100;120
# NOTE(Zihao): using "native" might be slow because whenever compile a cuda file
# with `-arch=native`, nvcc will spawn a `__nvcc_device_query` process to get
# the architecture of the host's GPU, which could stall the compilation process.
Expand Down
4 changes: 2 additions & 2 deletions docs/installation.rst
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -155,7 +155,7 @@ You can follow the steps below to install FlashInfer from source code:
.. code-block:: bash

cd flashinfer
TORCH_CUDA_ARCH_LIST="7.5 8.0 8.9 9.0a" FLASHINFER_ENABLE_AOT=1 pip install --no-build-isolation --verbose --editable .
TORCH_CUDA_ARCH_LIST="7.5 8.0 8.9 9.0a 10.0 12.0 12.0a" FLASHINFER_ENABLE_AOT=1 pip install --no-build-isolation --verbose --editable .

5. Create FlashInfer distributions (optional):

Expand All @@ -182,7 +182,7 @@ You can follow the steps below to install FlashInfer from source code:
.. code-block:: bash

cd flashinfer
TORCH_CUDA_ARCH_LIST="7.5 8.0 8.9 9.0a" FLASHINFER_ENABLE_AOT=1 python -m build --no-isolation --wheel
TORCH_CUDA_ARCH_LIST="7.5 8.0 8.9 9.0a 10.0 12.0 12.0a" FLASHINFER_ENABLE_AOT=1 python -m build --no-isolation --wheel
ls -la dist/

C++ API
Expand Down