Bug fixes #717
Merged
Bug fixes #717
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
Summary: This diff fixes two bugs that I found when creating a custom op and comparing results to PyTorch python implementations (next diff). 1) There is a segfault that occurred when n % 8 != 0 because the ukernel was storing out of bounds. There was an existing test for this case, but it passed because the output shape in the test was mistakenly too big and so no out of bound memory was written to in the test (it had shape m x k instead of shape m x n). This diff fixes the out-of-bound writes and the existing test. 2) The find_min_and_max function was incorrect. This corrects the function and adds tests for the reduction functions (find_min_and_max and compute_sum). (The find_min_and_max function is only used for dynamic quantization; there are existing tests for the quantization, but they passed because the existing find_min_and_max happened to return correct results in the tested case.) Reviewed By: digantdesai Differential Revision: D60773448
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/ao/717
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit d29a33e with merge base 1909171 ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
facebook-github-bot
added
the
CLA Signed
|
This pull request was exported from Phabricator. Differential Revision: D60773448 |
digantdesai
approved these changes
Aug 20, 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Outdated suggestions cannot be applied.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.
Summary:
This diff fixes two bugs that I found when creating a custom op and comparing results to PyTorch python implementations (next diff).
There is a segfault that occurred when n % 8 != 0 because the ukernel was storing out of bounds. There was an existing test for this case, but it passed because the output shape in the test was mistakenly too big and so no out of bound memory was written to in the test (it had shape m x k instead of shape m x n). This diff fixes the out-of-bound writes and the existing test.
The find_min_and_max function was incorrect. This corrects the function and adds tests for the reduction functions (find_min_and_max and compute_sum). (The find_min_and_max function is only used for dynamic quantization; there are existing tests for the quantization, but they passed because the existing find_min_and_max happened to return correct results in the tested case.)
Reviewed By: digantdesai
Differential Revision: D60773448