RFR: 8323807: Async UL: Add a stalling mode to async UL [v10]
Johan Sjölen
jsjolen at openjdk.org
Thu Jan 23 12:56:50 UTC 2025
On Thu, 23 Jan 2025 10:32:53 GMT, Axel Boldt-Christmas <aboldtch at openjdk.org> wrote:
>> Hi Axel,
>>
>> I've already switched to having the consumer thread log anything it itself logs directly without going through the buffer so this won't be an issue. See `enqueue_if_locked` (I'll be removing that and only have `enqueue` soon, however, so check there if you can't find it).
>
> Not sure I understand what you are trying to say. Maybe it will be clearer when you have pushed code.
>
> But I am not talking about the consumer. I am talking about any thread which logs and calls into here, then fails to write to the buffer, and calls `os::malloc` which in turn may log, which calls back into this code. But maybe you are saying there are already guardrails around logging under the ProducerLock.
I forgot that the `os::malloc` is done in a producer thread, not the consumer thread.
-------------
PR Review Comment: https://git.openjdk.org/jdk/pull/22770#discussion_r1926933021
More information about the hotspot-runtime-dev
mailing list