Fix DLPack stream logic. #150217
Open
Fix DLPack stream logic. #150217
+97
−17
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/150217
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 57e26fd with merge base 191693a ( UNSTABLE - The following job is marked as unstable, possibly due to flakiness on trunk:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
This was referenced Mar 28, 2025
Divigroup-RAP
pushed a commit
to Divigroup-RAP/PYTORCH
that referenced
this pull request
Apr 22, 2025
This PR fixes the logic for dealing with CUDA and ROCm streams whenever we are trying to create a DLPack capsule from a tensor. In summary, this PR: - Uses the legacy default stream if `tensor.__dlpack__(stream=None)` is called for a CUDA tensor. - Errors if `tensor.__dlpack__(stream=2)` is called for a CUDA tensor: PyTorch doesn't support the per-thread default stream. - Errors if `tensor.__dlpack__(stream=stream)`, where `stream` is 1 or 2, is called for a CUDA tensor using ROCm. For more details, see [the documentation][1]. [1]: https://data-apis.org/array-api/latest/API_specification/generated/array_api.array.__dlpack__.html ghstack-source-id: cc0e31c Pull Request resolved: pytorch/pytorch#150217
msaroufim
approved these changes
May 14, 2025
albanD
approved these changes
May 14, 2025
torch/_tensor.py
Outdated
| @@ -1717,23 +1717,37 @@ def __dlpack__(self, stream=None, max_version=None): | |||
| # Stream pointers in CUDA/ROCm are uniquely numbered and can | |||
| # be retrieved from their integer value. | |||
| raise TypeError("stream must be ``int`` or ``none``") | |||
| elif stream is not None and stream != -1: | |||
| elif stream != -1: | |||
| if self.device.type == "cuda": | |||
| # NB: This logic handles the special case values for default | |||
There was a problem hiding this comment.
No update to dlpack.py ? :D
There was a problem hiding this comment.
No need. If stream is None, we still need to synchronize, assuming the legacy default stream.
torch/_tensor.py
Outdated
| if is_cuda and stream == 2: | ||
| raise BufferError("per-thread default stream is not supported.") | ||
|
|
||
| assert is_cuda or (is_rocm and stream not in (1, 2)), ( |
There was a problem hiding this comment.
Shouldn't this be a BufferError like above instead of AssertionError?
There was a problem hiding this comment.
I don't think so. The reason being that this assertion checks something the standard explicitly states as "unsupported" or "disallowed", i.e. something the consumer should know about. Moreover, the standard also says that:
Other errors are raised when export fails for other reasons (e.g., incorrect arguments passed or out of memory).
torch/_tensor.py
Outdated
| # Only synchronize on different streams | ||
| sync_stream = torch.cuda.current_stream() | ||
| if stream != sync_stream: | ||
| current_stream = torch.cuda.current_stream() |
There was a problem hiding this comment.
Do we care if self.device.index != torch.cuda.current_device() ?
There was a problem hiding this comment.
Good point. I think we should. I will add a check for that.
This PR fixes the logic for dealing with CUDA and ROCm streams whenever we are trying to create a DLPack capsule from a tensor. In summary, this PR: - Uses the legacy default stream if `tensor.__dlpack__(stream=None)` is called for a CUDA tensor. - Errors if `tensor.__dlpack__(stream=2)` is called for a CUDA tensor: PyTorch doesn't support the per-thread default stream. - Errors if `tensor.__dlpack__(stream=stream)`, where `stream` is 1 or 2, is called for a CUDA tensor using ROCm. For more details, see [the documentation][1]. [1]: https://data-apis.org/array-api/latest/API_specification/generated/array_api.array.__dlpack__.html [ghstack-poisoned]
|
Starting merge as part of PR stack under #150691 |
This PR fixes the logic for dealing with CUDA and ROCm streams whenever we are trying to create a DLPack capsule from a tensor. In summary, this PR: - Uses the legacy default stream if `tensor.__dlpack__(stream=None)` is called for a CUDA tensor. - Errors if `tensor.__dlpack__(stream=2)` is called for a CUDA tensor: PyTorch doesn't support the per-thread default stream. - Errors if `tensor.__dlpack__(stream=stream)`, where `stream` is 1 or 2, is called for a CUDA tensor using ROCm. For more details, see [the documentation][1]. [1]: https://data-apis.org/array-api/latest/API_specification/generated/array_api.array.__dlpack__.html [ghstack-poisoned]
superiwan
pushed a commit
to superiwan/pytorch
that referenced
this pull request
Jul 14, 2025
This PR fixes the logic for dealing with CUDA and ROCm streams whenever we are trying to create a DLPack capsule from a tensor. In summary, this PR: - Uses the legacy default stream if `tensor.__dlpack__(stream=None)` is called for a CUDA tensor. - Errors if `tensor.__dlpack__(stream=2)` is called for a CUDA tensor: PyTorch doesn't support the per-thread default stream. - Errors if `tensor.__dlpack__(stream=stream)`, where `stream` is 1 or 2, is called for a CUDA tensor using ROCm. For more details, see [the documentation][1]. [1]: https://data-apis.org/array-api/latest/API_specification/generated/array_api.array.__dlpack__.html ghstack-source-id: 443b5cb Pull Request resolved: pytorch/pytorch#150217
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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):
BufferErrorfor DLPack buffer-related errors. #150691This PR fixes the logic for dealing with CUDA and ROCm streams whenever
we are trying to create a DLPack capsule from a tensor.
In summary, this PR:
tensor.__dlpack__(stream=None)iscalled for a CUDA tensor.
tensor.__dlpack__(stream=2)is called for a CUDA tensor:PyTorch doesn't support the per-thread default stream.
tensor.__dlpack__(stream=stream), wherestreamis 1 or2, is called for a CUDA tensor using ROCm.
For more details, see the documentation.