Support torch.xpu.mem_get_info API #141230
Closed
Conversation
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
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/141230
Note: Links to docs will display an error until the docs builds have been completed. ✅ You can merge normally! (1 Unrelated Failure)As of commit e7cc11e with merge base d633cf1 ( FLAKY - The following job failed but was likely due to flakiness present on trunk:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
guangyey
changed the title
guangyey
added
ciflow/xpu
Collaborator
Author
|
@pytorchbot rebase |
Collaborator
|
@pytorchbot started a rebase job onto refs/remotes/origin/viable/strict. Check the current status here |
Collaborator
Author
|
We have filed an issue to track this unrelated failure, refer to #141861 |
ezyang
reviewed
Dec 4, 2024
torch/xpu/memory.py
Outdated
| if :attr:`device` is ``None`` (default). | ||
| Returns: | ||
| int: the memory avialble on the device in units of bytes. |
albanD
approved these changes
Dec 4, 2024
torch/csrc/xpu/Module.cpp
Outdated
| Py_RETURN_NONE; | ||
| } | ||
|
|
||
| PyObject* THXPModule_getMemoryInfo(PyObject* self, PyObject* arg) { |
Collaborator
There was a problem hiding this comment.
nit: use pybind11 for this kind of functions. It makes the code simpler to maintain.
Collaborator
Author
There was a problem hiding this comment.
Thanks, @albanD, pybind11 makes my life easier. We will file another PR to bind these methods left in this file via pybind11.
Collaborator
Author
|
@pytorchbot merge |
Collaborator
Merge startedYour change will be merged once all checks pass (ETA 0-4 Hours). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
dvrogozh
added a commit
to dvrogozh/accelerate
that referenced
this pull request
Dec 6, 2024
torch.xpu.mem_get_info API is available starting from PyTorch 2.6 (and in nightly 2.6.0.dev20241206+xpu or later). To work properly this method requires PyTorch built with the SYCL runtime which supports API to query device memory stats. If not available, exception will be raised. Requires: pytorch/pytorch#141230 Fixes: huggingface#2929 Fixes: huggingface/transformers#31922 Signed-off-by: Dmitry Rogozhkin <[email protected]>
dvrogozh
added a commit
to dvrogozh/accelerate
that referenced
this pull request
Dec 6, 2024
torch.xpu.mem_get_info API is available starting from PyTorch 2.6 (and in nightly 2.6.0.dev20241206+xpu or later). To work properly this method requires PyTorch built with the SYCL runtime which supports API to query device memory stats. If not available, exception will be raised. Requires: pytorch/pytorch#141230 Fixes: huggingface#2929 Fixes: huggingface/transformers#31922 Signed-off-by: Dmitry Rogozhkin <[email protected]>
This was referenced Dec 6, 2024
AmdSampsa
pushed a commit
to AmdSampsa/pytorch
that referenced
this pull request
Dec 9, 2024
# Motivate Fix pytorch#130599 This PR intends to add a new API, `torch.xpu.mem_get_info,` which is widely used in popular model workloads. For example, [here](https://github.com/huggingface/accelerate/blob/403c0714d1dd019a481022afce5df75a9963ecd9/src/accelerate/utils/modeling.py#L721) we need to get current GPU memory usage to split or load the model. Pull Request resolved: pytorch#141230 Approved by: https://github.com/EikanWang, https://github.com/albanD
dvrogozh
added a commit
to dvrogozh/accelerate
that referenced
this pull request
Dec 9, 2024
torch.xpu.mem_get_info API is available starting from PyTorch 2.6 (and in nightly 2.6.0.dev20241206+xpu or later). To work properly this method requires PyTorch built with the SYCL runtime which supports API to query device memory stats. If not available, exception will be raised. Requires: pytorch/pytorch#141230 Fixes: huggingface#2929 Fixes: huggingface/transformers#31922 Signed-off-by: Dmitry Rogozhkin <[email protected]>
SunMarc
pushed a commit
to huggingface/accelerate
that referenced
this pull request
Dec 24, 2024
torch.xpu.mem_get_info API is available starting from PyTorch 2.6 (and in nightly 2.6.0.dev20241206+xpu or later). To work properly this method requires PyTorch built with the SYCL runtime which supports API to query device memory stats. If not available, exception will be raised. Requires: pytorch/pytorch#141230 Fixes: #2929 Fixes: huggingface/transformers#31922 Signed-off-by: Dmitry Rogozhkin <[email protected]>
73 tasks
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
7 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.
Stack from ghstack (oldest at bottom):
Motivate
Fix #130599
This PR intends to add a new API,
torch.xpu.mem_get_info,which is widely used in popular model workloads.For example, here we need to get current GPU memory usage to split or load the model.
cc @gujinghui @EikanWang @fengyuan14