Conversation
ContinuousBatchingState.__init__ only creates a torch.cuda.Stream when device.type == "cuda". On Ascend NPU or Intel XPU, compute_stream is set to None, losing async computation benefits. Add elif branches for torch.npu.Stream and torch.Stream (used by XPU).
Contributor
CI recapDashboard: View test results in Grafana
|
Member
|
cc @IlyasMoutawwakil for Ascend! |
Comment on lines
+120
to
+124
| self.compute_stream = (torch.cuda.Stream(device=self.device) if device.type == "cuda" else None) | ||
| if hasattr(torch, "npu") and device.type == "npu": | ||
| self.compute_stream = torch.npu.Stream(device=self.device) | ||
| elif hasattr(torch, "xpu") and device.type == "xpu": | ||
| self.compute_stream = torch.Stream(device=self.device) |
Member
There was a problem hiding this comment.
can't we use torch.Stream always when not cpu ?
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
3 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
What & why
ContinuousBatchingState.__init__only createstorch.cuda.Stream(device=self.device)whendevice.type == "cuda". On Ascend NPU or Intel XPU,compute_streamis set toNone, losing the benefit of async tensor transfers during continuous batching.Fix
Add additional branches for Ascend NPU (
torch.npu.Stream) and XPU (torch.Stream) when the corresponding device type is detected.