RFR: 8323791: Parallel: Using non-atomic for live bytes update during Full GC [v4]

Albert Mingkun Yang ayang at openjdk.org
Fri Feb 9 12:14:06 UTC 2024


On Tue, 30 Jan 2024 14:35:09 GMT, Thomas Schatzl <tschatzl at openjdk.org> wrote:

> However I have seen slight regressions (or at least no difference) with e.g. gcbasher 128m (with

With large-heap, obj-iteration (even in parallel) can be slow, which can overshadow the gain from non-atomic updates. Therefore, I have tried the cache-solution as you suggested: https://github.com/openjdk/jdk/pull/17788 It has better improvement and avoids obj-iteration.

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

PR Comment: https://git.openjdk.org/jdk/pull/17442#issuecomment-1935819839


More information about the hotspot-gc-dev mailing list