This repository was archived by the owner on Jan 23, 2023. It is now read-only.
Ensure that ZeroMemoryInGCHeap writes in pointer-sized increments #12689
Conversation
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
by adding the volatile keyword (which disables memset optimization). Fixes #12207
|
@swgillespie @Petermarcu @brianrob @cmckinsey @Maoni0 - This should fix #12207 in master. Assuming it's green, we will want to port it to release/2.0.0 ASAP to get the the blocking bug cleared for 2.0. |
brianrob
approved these changes
Jul 8, 2017
|
@dotnet-bot test windows_nt release longgc |
swgillespie
approved these changes
Jul 8, 2017
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
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Outdated suggestions cannot be applied.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.
Enabling PGO optimizations exposed a latent but in the
ZeroMemoryInGCHeapfunction: it assumed that writing a constant (0) to a pointer-sized location in memory was inherently atomic. This is not, in fact, the case: the compiler can freely turn the write into a non-atomic sequence of smaller writes. This leads to a race condition with the GC running in another thread, as discovered by the test failure tracked by #12207.The fix is essentially the same as for #4341, and corroborated by advice from compiler devs: mark the buffer being cleared as volatile. This prevents the compiler from performing read/write optimizations against the buffer, and thus, that the store of a pointer-sized integer constant (given that the buffer access is aligned) remains a single, atomic store.