RFR: JDK-8267371: Concurrent gtests take too long

Wed May 19 09:30:40 UTC 2021

On Wed, 19 May 2021 09:04:51 GMT, Aleksey Shipilev <shade at openjdk.org> wrote:

> This looks good to me. I was kinda wondering the last time I looked at that test if 30 threads is excessive for the test as well.

I'm honestly not sure why these tests are run concurrently; it looks to me like there is no VM internal synchronization involved which we could test, this is basically just a bunch of parallel mmap()s. I may miss something here.

Had been introduced (as jtreg-started whitebox test) with https://bugs.openjdk.java.net/browse/JDK-8027237. The associated RFR thread does not discuss the run parameters.

@stefank Stefan, do you think we could reduce the number of Threads for these tests too?

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

PR: https://git.openjdk.java.net/jdk/pull/4110