RFR(XS): 8198510: Limit GC threads for small Java heap

[Stefan Johansson]

Hi Martin,

The functional tests looks good. We have also started a set of 
performance benchmarks that will run over the weekend to make sure our 
baseline won't be to affected by this (it shouldn't, but we don't want 
false regressions showing up).

It would be really nice if you could update the patch with a new flag 
text, something like what Thomas suggested and then update the CSR 
Specification section. Once that is done I can review it:
https://bugs.openjdk.java.net/browse/JDK-8198547

Thanks,
Stefan

On 2018-02-22 11:45, Doerr, Martin wrote:
> Hi Thomas,
>
> thank you very much for taking care of this issue.
> I'll wait for the results.
>
> Best regards,
> Martin
>
>
> -----Original Message-----
> From: Thomas Schatzl [mailto:thomas.schatzl at oracle.com]
> Sent: Donnerstag, 22. Februar 2018 11:32
> To: Doerr, Martin <martin.doerr at sap.com>; hotspot-gc-dev at openjdk.java.net
> Subject: Re: RFR(XS): 8198510: Limit GC threads for small Java heap
>
> Hi,
>
> On Wed, 2018-02-21 at 16:17 +0000, Doerr, Martin wrote:
>> Hi Thomas,
>>
>> thanks for discussing this topic with your colleagues.
>>
>> I have a webrev which simply switches on UseDynamicNumberOfGCThreads
>> by default:
>> http://cr.openjdk.java.net/~mdoerr/8198510_gc_threads/webrev.01/
>>
>> I'll test it in our nightly tests. If I get approval from 2 reviewers
>> and no objections, I'll push it to the submission forest and later to
>> jdk/hs.
>    - the description of the flag mentions Parallel GC explicitly - this
> functionality is available in all (parallel) collectors.
>
> Something like:
>
> "Dynamically choose the number of threads up to a maximum of
> ParallelGCThreads parallel collectors will use for garbage collection
> work"
>
>    - I just started a run in our test system with that change. Let's see
>   if and what fails due to that.
>
>    - unfortunately this is a public interface change, so we need to wait
> for the CSR to be approved I filed for you, JDK-8198547.
>
> I filled in all parts except for the specification - it would be nice
> to be able to put the inline diff there (since it's so small) and link
> the final webrev and a comment containing a reference to this the mail
> thread.
>
> I will add (an internal) link to test results when it finishes
> hopefully without any new issues.
>
> When all this is done, we need at least one Reviewer (me and maybe
> somebody else) to review the CSR and add themselves to the "Reviewed-
> by" field (Use the "Edit" button on the top left) so that we can fast-
> track it, ie. set it to "Finalized" and then wait for comment from the
> CSR approval board :)
>
> Thanks,
>    Thomas
>
>