RFR: 8303276: Secondary assertion failure in AdapterHandlerLibrary::contains during crash reporting

[David Holmes]

On Tue, 18 Apr 2023 13:15:21 GMT, Thomas Stuefe <stuefe at openjdk.org> wrote:

> we trade a guaranteed secondary error against a very unlikely one

I would say we trade a very obvious secondary error for a potentially very obscure one. If you don't hold the lock, nor are at a safepoint, then the access is inherently unsafe. You may not crash you may just get bad or misleading data which could then send you off on a wild goose chase.

Time will tell ...

-------------

PR Comment: https://git.openjdk.org/jdk/pull/13500#issuecomment-1514146892