tcp.custom_op support for torch.aten.arange with dynamic input #100
Conversation
This file contains 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
navahgar
reviewed
Oct 7, 2024
Comment on lines
+544
to
+550
| // convert constant int to fp if possible | ||
| int64_t constIntVal; | ||
| if (matchPattern(value, torch::Torch::m_TorchConstantInt(&constIntVal))) { | ||
| attrs.push_back(rewriter.getNamedAttr( | ||
| attrName, rewriter.getF64FloatAttr(static_cast<double>(constIntVal)))); | ||
| return true; | ||
| } |
There was a problem hiding this comment.
Why do we have to convert int to fp here? Why not convert them to IntAttr?
There was a problem hiding this comment.
aten.arange.start_step can take either int or fp for start and step arguments. However, trt-mlir only accepts F64Attr. Converting all constants to FPAttr saves us some time for type checking.
I can also add the conversion in tcp-to-trtmlir. LMKWYT
There was a problem hiding this comment.
Thanks for the clarification. Can you add a comment in the code to explain this?
navahgar
approved these changes
Oct 7, 2024
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Outdated suggestions cannot be applied.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.
tcp.custom_op support for
torch.aten.arangewith dynamic input. Static case will be handled through TOSA dialect.To test:
(in docker)
bazel test //...