Upstream Sync - up to: 27f501dbe011f4371bff938fe7e09311ab3002fa#56
Merged
LucasWilkinson merged 82 commits intomainfrom Mar 20, 2025
Merged
Upstream Sync - up to: 27f501dbe011f4371bff938fe7e09311ab3002fa#56LucasWilkinson merged 82 commits intomainfrom
LucasWilkinson merged 82 commits intomainfrom
Conversation
…ing (Dao-AILab#1446) * Always update both submodules to include them in sdist Always update both submodules, irrespectively of whether a CUDA or a ROCM build is being done, to ensure that the necessary files from both are present in sdist. Otherwise, attempt to perform a ROCM build from sdist fails because of missing `composable_kernel` srouces. * Include `*.py` files from composable_kernel in sdist Include the `*.py` files from `csrc` in sdist, to ensure that the `generate.py` script is present. * Replace the `os.system()` calls in `setup.py` with `subprocess.run()` * Add error checking to `subprocess.run()` calls in `setup.py` Add error checking to ensure that `setup.py` fails immediately if one of the commands fail. Otherwise, the failures result only in messages to stderr that could be missed, and could lead to more confusing errors later in the build process. * Call git in `setup.py` only when working in a git repository Call git commands in `setup.py` only when the `.git` directory is present, indicating that we are working in a git checkout. Otherwise, just assert that the needed files are there. With this, building from a source distribution no longer attempts to call git in an incorrect directory.
…1534) Generate a list of autotune configs based on device warp size to avoid triton error if maximum threads per block is exceeded.
…-sync-3-attempt-2 Signed-off-by: Lucas Wilkinson <lwilkinson@neuralmagic.com>
Signed-off-by: Lucas Wilkinson <lwilkinson@neuralmagic.com>
tlrmchlsmth
approved these changes
Mar 20, 2025
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
10 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.
Sync up to: Dao-AILab@27f501d
Add associated torch library wrapper for
get_scheduler_metadataand add AOT scheduling to the tests