Repository navigation
Add workaround for ACPP allocation tracker destruction order - #343
Merged
Merged
Conversation
|
Check-perf-impact results: (ae6918621b46271c2f10d6eb978fe95d) ❓ No new benchmark data submitted. ❓ |
Coverage Report for CI Build 38424719415Coverage remained the same at 95.052%Details
Uncovered ChangesNo uncovered changes found. Coverage RegressionsNo coverage regressions found. Coverage Stats💛 - Coveralls |
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
AdaptiveCpp stores its USM allocation tracker in a function-local static that is initialized by the first allocation, i.e., after the runtime has registered its
atexitshutdown handler. Consequently, the tracker is destroyed before the runtime frees its allocations, causing eachsycl::free()call during shutdown to operate on a destroyed object. This results in heap corruption, manifesting as a hang on exit in release builds and a segmentation fault in debug builds.By allocating memory once in the
sycl_backendconstructor, the tracker is initialized before the shutdown handler is registered and therefore outlives the runtime.