Skip to content
This repository was archived by the owner on Aug 15, 2025. It is now read-only.

Update CUDA_UPGRADE_GUIDE.MD #2060

Closed
wants to merge 1 commit into from
Closed

Update CUDA_UPGRADE_GUIDE.MD #2060

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
15 changes: 6 additions & 9 deletions CUDA_UPGRADE_GUIDE.MD
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -9,26 +9,23 @@ Here is the supported matrix for CUDA and CUDNN (versions can be looked up in ht

| CUDA | CUDNN | additional details |
| --- | --- | --- |
| 11.8 | 8.7.0.84 | Legacy CUDA Release |
| 12.1 | 8.9.2.26 | Stable CUDA Release |
| 12.4 | 8.9.7.29 | Latest CUDA Nightly |
| 11.8 | 9.1.0.70 | Legacy CUDA Release |
| 12.6 (.3) | 9.5.1.17 | Stable CUDA Release |
| 12.8 (.1) | 9.7.1.26 | Latest CUDA Nightly |
Copy link
Contributor

Choose a reason for hiding this comment

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

We are currently building with 12.8.0 (not .1) if I'm not mistaken: https://github.com/pytorch/pytorch/blob/562328501e167206dc7d4b16895b5ae538520e06/.github/scripts/generate_binary_build_matrix.py#L75-L88 matches: https://docs.nvidia.com/cuda/archive/12.8.0/cuda-toolkit-release-notes/index.html#id4

Also, it seems torch==2.7.0+cu128 uses cuDNN==9.7.1.26, but the nightly builds use 9.8? https://github.com/pytorch/pytorch/blob/562328501e167206dc7d4b16895b5ae538520e06/.github/scripts/generate_binary_build_matrix.py#L78
We might need to clarify versioned releases vs. nightly builds.

Copy link
Contributor Author

Choose a reason for hiding this comment

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

Thanks for checking, indeed we are using 12.8.0 - https://github.com/pytorch/pytorch/blob/main/.ci/docker/common/install_cuda.sh#L224.

Wondering why for CUDNN versions it is still 9.7.1.26 for 2.7.0, the update to 9.8 is merged on 3/13 and should be in 2.7.0 branch cut? (pytorch/pytorch#148963)



### B. Check the package availability

Package availability to validate before starting upgrade process :

1) CUDA and CUDNN is available for Linux and Windows:
https://developer.download.nvidia.com/compute/cuda/11.5.0/local_installers/cuda_11.5.0_495.29.05_linux.run
https://developer.download.nvidia.com/compute/redist/cudnn/v8.3.2/local_installers/11.5/
https://developer.nvidia.com/cuda-downloads (select arch type for download)

2) CUDA is available on conda via nvidia channel : https://anaconda.org/nvidia/cuda/files

3) CUDA is available on Docker hub images : https://hub.docker.com/r/nvidia/cuda
2) CUDA is available on Docker hub images : https://hub.docker.com/r/nvidia/cuda
Following example is for cuda 12.4: https://gitlab.com/nvidia/container-images/cuda/-/tree/master/dist/12.4.0/ubuntu2204/devel?ref_type=heads
(Make sure to use version without CUDNN, it should be installed separately by install script)

4) Validate new driver availability: https://docs.nvidia.com/cuda/cuda-toolkit-release-notes/index.html. Check following table: Table 3. CUDA Toolkit and Corresponding Driver Versions
3) Validate new driver availability: https://docs.nvidia.com/cuda/cuda-toolkit-release-notes/index.html. Check following table: Table 3. CUDA Toolkit and Corresponding Driver Versions


## 1. Maintain Progress and Updates
Expand Down