[upstream #2795] Histc raises error with integer input when deterministic algorithm is enabled#8
Draft
Stonepia wants to merge 1 commit into
Draft
[upstream #2795] Histc raises error with integer input when deterministic algorithm is enabled#8Stonepia wants to merge 1 commit into
Stonepia wants to merge 1 commit into
Conversation
[upstream pytorch#2795] Histc raises error with integer input when deterministic algorithm is enabled
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
[upstream pytorch#2795] Histc raises error with integer input when deterministic algorithm is enabled
Fixes https://github.com/ZhaoqiongZ/torch-xpu-ops-exp/issues/193
Root Cause: The XPU histc implementation in
SummaryOps.cppunconditionally callsglobalContext().alertNotDeterministic("_histc_xpu")for all input dtypes, whereas the CUDA implementation correctly guards the alert withif (at::isFloatingType(self.scalar_type())). Integer atomicAdd is exact and order-independent, making integer histc deterministic. Additionally,torch/_meta_registrations.pyhas a Python-level determinism guard for CUDA+float but no equivalent guard for XPU, so even fixing C++ alone may leave a gap.Diff stat: