fix(docker): set TMS_CUDA_MAJOR for torch_memory_saver source build - #1774
Merged
Conversation
Contributor
|
Caution The consumer version of Gemini Code Assist on GitHub has been sunset. All code review activity has officially ceased. |
Zhichenzzz
force-pushed
the
zhichen/fix-docker-tms-cuda-major
branch
from
July 23, 2026 19:05
4170030 to
4e1d79e
Compare
Unpinning torch_memory_saver (#1773) crossed the multi-cuda-wheel change, whose setup.py now aborts a CUDA source build unless TMS_CUDA_MAJOR is set, breaking the Docker Build. Derive it from the image's torch so cu12 and cu13 variants both work.
Zhichenzzz
force-pushed
the
zhichen/fix-docker-tms-cuda-major
branch
from
July 23, 2026 19:06
4e1d79e to
5658359
Compare
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Problem
#1773 unpinned
torch_memory_saver(@d64a639→ master), which crossed the multi-cuda-wheel change. Itssetup.pynow aborts a CUDA source build unlessTMS_CUDA_MAJORis set:so the Docker Build & Push on main fails at the
pip install git+...torch_memory_saver.gitstep (failing run).Fix
Set
TMS_CUDA_MAJORfor that build, derived from the image's own torch (torch.version.cuda), so the cu12 and cu13 variants each build the matching wheel with no hardcoding.Notes
TMS_CUDA_MAJOR=<n> pip install git+...torch_memory_saver.git) builds cleanly against current TMS master on an 8×H200 CUDA-13 box.Dockerfile.rocmleft unpinned as chore(docker): upgrade torch_memory_saver to 6d5bce48 #1773 set it —TMS_CUDA_MAJORis a CUDA-only requirement; the ROCm build detects hip and does not need it.