Spurious test failure codegen-llvm/issue-107681-unwrap_unchecked.rs #158500

Activity

1. JonathanBrouwer

   added

   A-LLVMArea: Code generation parts specific to LLVM. Both correctness bugs and optimization-related issues.Area: Code generation parts specific to LLVM. Both correctness bugs and optimization-related issues.

   C-bugCategory: This is a bug.Category: This is a bug.

   I-flaky-testIssue: A test is flaky/unreliable/spuriously failsIssue: A test is flaky/unreliable/spuriously fails

   on Jun 27, 2026
2. rustbot

   added

   needs-triageThis issue may need triage. Remove when done. See docs forge.rust-lang.org/release/issue-triagingThis issue may need triage. Remove when done. See docs forge.rust-lang.org/release/issue-triaging

   on Jun 27, 2026
3. workingjubilee commented on Jun 27, 2026

   

   workingjubilee

   on Jun 27, 2026

   Last edited by workingjubilee

   Member

   More actions

   To summarize our current findings from that discussion thread:

   #158067 was the PR that introduced the sporadic failures

   cc @TimNN @dianqk

   The issue is probably due to randomize-layout in the codegen-llvm suite.

   The test was made more imprecise in a way that is probably unnecessary, while being made more precise in ways that are brittle in the face of randomize-layout.

   The line

       // CHECK: [[INNER:%.*]] = load ptr, ptr %x
   

   is not resilient to the load sometimes being from %x and sometimes being from a pointer assigned to %0 that is itself GEP'd from %x

   Revert posted at #158502
4. saethlin

   added

   T-compilerRelevant to the compiler team, which will review and decide on the PR/issue.Relevant to the compiler team, which will review and decide on the PR/issue.

   and removed

   needs-triageThis issue may need triage. Remove when done. See docs forge.rust-lang.org/release/issue-triagingThis issue may need triage. Remove when done. See docs forge.rust-lang.org/release/issue-triaging

   on Jun 27, 2026
5. rust-bors

   added a commit that references this issue on Jun 27, 2026

   Auto merge of #158502 - saethlin:revert-flaky-codegen-test-fix, r=Jon…

   ...

   47ab177
6. JonathanBrouwer

   added a commit that references this issue on Jun 27, 2026

   Rollup merge of rust-lang#158502 - saethlin:revert-flaky-codegen-test…

   ...

   39b11c0
7. rust-bors

   added a commit that references this issue on Jun 27, 2026

   Auto merge of #158502 - saethlin:revert-flaky-codegen-test-fix, r=Jon…

   ...

   8b95a26
8. rust-bors

   closed this as completedin 8b95a26on Jun 28, 2026
9. TimNN commented on Jun 28, 2026

   

   TimNN

   on Jun 28, 2026

   Contributor

   More actions

   Thank you for the revert and sorry for the trouble, I wasn't aware of the Layout randomization part.

   ❤️React with ❤️2Reacted by Jonathan Brouwer and Jubilee
10. workingjubilee commented on Jun 28, 2026

    

    workingjubilee

    on Jun 28, 2026

    Member

    More actions

    Honestly, it should not have been trouble, so it's quite fine. We probably need to try to figure out how to make the layout randomization not engage for codegen-llvm tests, but the Zulip thread also describes why that is not a quick 1-minute fix.