Integrated: JDK-8319314: NMT detail report slow or hangs for large number of mappings
Thomas Stuefe
stuefe at openjdk.org
Sat Nov 11 06:58:08 UTC 2023
On Thu, 2 Nov 2023 16:42:58 GMT, Thomas Stuefe <stuefe at openjdk.org> wrote:
> If a lot of mappings have been registered with NMT, NMT detail report may get so slow as to hang seemingly.
>
> 100 k mappings will take > 1 minute
> 1 million mappings never returns (I canceled the test after 90 minutes)
>
> We burn all the time doing a sorted insert into the target array of the MemBaseline object. That sorted insert is pointless since the source array is already sorted by region base address.
>
> Omitting the pointless sort drops the runtime for 1 million mappings from > 90 minutes to 48 seconds.
>
> Tests: Tested locally. GHAs (Windows error unrelated).
This pull request has now been integrated.
Changeset: 9938b3f6
Author: Thomas Stuefe <stuefe at openjdk.org>
URL: https://git.openjdk.org/jdk/commit/9938b3f62babfc35ee682bd979a6bf08ac7cd348
Stats: 18 lines in 1 file changed: 8 ins; 5 del; 5 mod
8319314: NMT detail report slow or hangs for large number of mappings
Reviewed-by: jsjolen, gziemski
-------------
PR: https://git.openjdk.org/jdk/pull/16480
More information about the hotspot-runtime-dev
mailing list