Re: [PR] [TOPI] Fix strided_slice_with_axes to handle negative axis values [tvm]
tlopex merged PR #18917: URL: https://github.com/apache/tvm/pull/18917 -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] - To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
Re: [PR] [TOPI] Fix strided_slice_with_axes to handle negative axis values [tvm]
mshr-h commented on PR #18917: URL: https://github.com/apache/tvm/pull/18917#issuecomment-4095726556 cc @tlopex @guan404ming -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] - To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
Re: [PR] [TOPI] Fix strided_slice_with_axes to handle negative axis values [tvm]
gemini-code-assist[bot] commented on code in PR #18917:
URL: https://github.com/apache/tvm/pull/18917#discussion_r2963809939
##
include/tvm/topi/transform.h:
##
@@ -904,28 +904,41 @@ inline Tensor strided_slice_with_axes(const Tensor& x,
const ffi::Array
std::string slice_mode = "end",
std::string name =
"T_strided_slice_with_axes",
std::string tag = kInjective) {
- const size_t src_tensor_dim = x->shape.size();
- TVM_FFI_ICHECK(axes.size() <= src_tensor_dim);
+ const int64_t src_tensor_dim = static_cast(x->shape.size());
+ TVM_FFI_ICHECK(static_cast(axes.size()) <= src_tensor_dim);
TVM_FFI_ICHECK(axes.size() == begin.size() && axes.size() == end.size() &&
axes.size() == strides.size());
+ // Normalize negative axes
+ ffi::Array normalized_axes;
+ for (size_t i = 0; i < axes.size(); ++i) {
+int64_t axis = axes[i].IntValue();
+if (axis < 0) {
+ axis += src_tensor_dim;
+}
+TVM_FFI_ICHECK(axis >= 0 && axis < src_tensor_dim)
+<< "Axis " << axes[i].IntValue() << " is out of bounds for tensor with
" << src_tensor_dim
+<< " dimensions";
+normalized_axes.push_back(Integer(axis));
+ }
Review Comment:
While normalizing axes, it's good practice to also check for duplicate axes.
Currently, if an axis is repeated, the slice parameters for that axis will be
silently overwritten, which can lead to unexpected behavior. Other operators
like `transpose` already perform this check. You can use a `std::unordered_set`
to efficiently track seen axes and ensure each axis is specified only once.
```suggestion
ffi::Array normalized_axes;
std::unordered_set seen_axes;
for (size_t i = 0; i < axes.size(); ++i) {
int64_t axis = axes[i].IntValue();
if (axis < 0) {
axis += src_tensor_dim;
}
TVM_FFI_ICHECK(axis >= 0 && axis < src_tensor_dim)
<< "Axis " << axes[i].IntValue() << " is out of bounds for tensor
with " << src_tensor_dim
<< " dimensions";
auto [it, inserted] = seen_axes.insert(axis);
TVM_FFI_ICHECK(inserted) << "Axis " << axis << " is specified more than
once.";
normalized_axes.push_back(Integer(axis));
}
```
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
-
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
Re: [PR] [TOPI] Fix strided_slice_with_axes to handle negative axis values [tvm]
gemini-code-assist[bot] commented on PR #18917: URL: https://github.com/apache/tvm/pull/18917#issuecomment-4095270468 ## Summary of Changes Hello, I'm Gemini Code Assist[^1]! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed! This pull request addresses a critical bug in the `strided_slice_with_axes` function where negative axis values were not properly handled, leading to runtime errors. The changes introduce a robust mechanism to normalize these negative indices to their positive equivalents, ensuring the slicing operation functions as expected and improving the stability of tensor manipulations. A dedicated test case has been added to validate this fix. ### Highlights * **Negative Axis Handling**: Implemented normalization for negative axis values in `strided_slice_with_axes` to convert them into positive, zero-based indices. This prevents `IndexError` during `LegalizeOps` when negative axes are used directly. * **API Integration**: The normalized axes are now consistently used when calling `StridedSliceCanonicalizeBegin`, `StridedSliceOutputShape`, and within the compute lambda for index calculation, ensuring correct behavior across the slicing logic. * **Test Coverage**: Added a new test case `test_strided_slice_negative_axes` to verify that `strided_slice_with_axes` correctly processes negative axis inputs after the `LegalizeOps` pass. 🧠**New Feature in Public Preview:** You can now enable **Memory** to help **Gemini Code Assist** learn from your team's feedback. This makes future code reviews more consistent and personalized to your project's style. **Click [here](https://codeassist.google/code-review/login) to enable Memory in your admin console.** Using Gemini Code Assist The full guide for Gemini Code Assist can be found on our [documentation page](https://developers.google.com/gemini-code-assist/docs/review-github-code), here are some quick tips. Invoking Gemini You can request assistance from Gemini at any point by creating a comment using either `/gemini ` or `@gemini-code-assist `. Below is a summary of the supported commands on the current page. Feature | Command | Description --- | --- | --- Code Review | `/gemini review` | Performs a code review for the current pull request in its current state. Pull Request Summary | `/gemini summary` | Provides a summary of the current pull request in its current state. Comment | @gemini-code-assist | Responds in comments when explicitly tagged, both in pull request comments and review comments. Help | `/gemini help` | Displays a list of available commands. Customization To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a `.gemini/` folder in the base of the repository. Detailed instructions can be found [here](https://developers.google.com/gemini-code-assist/docs/customize-gemini-behavior-github). Limitations & Feedback Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with :thumbsup: and :thumbsdown: on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for GitHub and other Google products, sign up [here](https://google.qualtrics.com/jfe/form/SV_2cyuGuTWsEw84yG). [^1]: Review the [Privacy Notices](https://policies.google.com/privacy), [Generative AI Prohibited Use Policy](https://policies.google.com/terms/generative-ai/use-policy), [Terms of Service](https://policies.google.com/terms), and learn how to configure Gemini Code Assist in GitHub [here](https://developers.google.com/gemini-code-assist/docs/customize-gemini-behavior-github). Gemini can make mistakes, so double check it and [use code with caution](https://support.google.com/legal/answer/13505487). -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] - To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
[PR] [TOPI] Fix strided_slice_with_axes to handle negative axis values [tvm]
mshr-h opened a new pull request, #18917: URL: https://github.com/apache/tvm/pull/18917 Negative axis values (e.g., `axes=[-1]`) in `strided_slice_with_axes` were used directly as array indices without normalization, causing an `IndexError` during `LegalizeOps`. This PR normalizes negative axes to positive equivalents before passing them to `StridedSliceCanonicalizeBegin`, `StridedSliceOutputShape`, and the compute lambda. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] - To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
