On Mon, 19 Oct 2026 17:16:48 GMT, Mat Carter <[email protected]> wrote:
>> Mat Carter has updated the pull request with a new target base due to a >> merge or a rebase. The incremental webrev excludes the unrelated changes >> brought in by the merge/rebase. The pull request contains five additional >> commits since the last revision: >> >> - Add non-LSE variant regression tests >> - Need the derived class to also use the new accessor >> - Merge branch 'master' into macarte/nolse >> - Use getVolatile instead of getAcquire >> - With -UseLSE the Starvation test on Windows timeouts close to 100% >> whereas this only happens on Linux on larger machines with many cores. The >> issue is that the LDR following the CAS can execute before the STLXR >> breaking the Dekker protocol, replacing with LADR by using getAcquire the >> solves the issue as it won't be reordered before the STLXR. > > Thank you for the reviews and your patience! Post-integration question to @macarte: > In the xfer and tryMatchData we call an intrinsic (lets call it dekkerFence) > prior to reading the waiter. OK, but the fix only does `xfer`, and not `tryMatchData`, AFAICS? Is this affected the same? ------------- PR Comment: https://git.openjdk.org/jdk/pull/31465#issuecomment-6286621631
