[prev in list] [next in list] [prev in thread] [next in thread] 

List:       openjdk-serviceability-dev
Subject:    Re: RFR: 8262271: SA: Add new stress test that tests getting the stack trace of an active thread
From:       Chris Plummer <cjplummer () openjdk ! java ! net>
Date:       2021-02-25 7:03:40
Message-ID: _K-vQe2ii1xQ6eUI1tlrMr1Jzln35Enmb_wO_XIFc4w=.9e23582a-2fcf-4044-bd97-d6c653b200d3 () github ! com
[Download RAW message or body]

On Thu, 25 Feb 2021 03:54:35 GMT, Chris Plummer <cjplummer@openjdk.org> wrote:

> This a new test that runs jstack on a process that is busy doing things. It runs \
> jstack 4 times as the process is starting up.

> Sorry but I don't see how this ensures it runs jstack as the jshell is
> starting up, the startup could be complete before even one jstack gets
> issued.

My observation is that at least the main thread has a different stack trace on each \
iteration. Also, each iteration takes progressively longer as jshell starts up. \
Although I don't fully understand the reason why, I think it's related to \
[JDK-8262098](https://bugs.openjdk.java.net/browse/JDK-8262098). Possibly the larger \
the program gets, the more SA needs to page in, even though you would think jstack \
would not need to touch that many pages. In any case, this is an indication that the \
program is changing its state.

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

PR: https://git.openjdk.java.net/jdk/pull/2720


[prev in list] [next in list] [prev in thread] [next in thread]