RFR: 8320304: Refactor and simplify monitor deflation functions [v3]

David Holmes dholmes at openjdk.org
Tue Nov 21 00:31:09 UTC 2023


On Mon, 20 Nov 2023 23:35:59 GMT, Stefan Karlsson <stefank at openjdk.org> wrote:

>> The motivation for this is that I can hide the logging code inside this object instead of passing the LogStream and timer as parameters to the various functions. This simplifies the API, IMHO.
>
> This is a patch that removes the "safepointer" class:
> https://github.com/openjdk/jdk/commit/2025d7a8e8d48d2d9af548d95b4a88a57058427c
> 
> And this is what the entire patch would look like:
> https://github.com/openjdk/jdk/compare/master...stefank:jdk:8320304_cleanup_of_monitor_deflation_no_safepointer
> 
> Personally, I think this makes the code messier. Code that previously didn't have to care about logging now has to do so. It is also more obscure what function calls will perform the safepoint checking.

I appreciate you took the time to trial this - that was not necessary unless others have similar comments.  Personally I find the patch without safepointer less disruptive from the safepointing pov.

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

PR Review Comment: https://git.openjdk.org/jdk/pull/16706#discussion_r1399893265


More information about the hotspot-runtime-dev mailing list