RFR (S): 8190703: TestSystemGCWith* infrequently times out on SPARC

Aleksey Shipilev shade at redhat.com
Mon Nov 6 19:46:52 UTC 2017


On 11/06/2017 12:27 PM, Thomas Schatzl wrote:
>   can I have reviews for this change the fixes intermittent timeouts in
> a stress test when run on slow machines and debug builds.
> 
> The fix is to simply let the test bail out early if time is up - this
> allows the test run through as intended in the many configurations
> where it does not take excessive amount of time.

Makes sense. Real timeouts would still surface.

> CR:
> https://bugs.openjdk.java.net/browse/JDK-8190703
> Webrev:
> http://cr.openjdk.java.net/~tschatzl/8190703/webrev

*) This change seems like a leftover:

  93 System.err.println("SystemGCTask with delay " + delayMS);

*) This change also does not look right -- should we instead increase delays in callers?

 101             ThreadUtils.sleep(delayMS * 10);


*) I think you can just say:

 for (int i = 0; (i < 4) && (System.currentTimeMillis() < endTime); i++)

 ...instead of rewriting the whole loop.

-Aleksey

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: OpenPGP digital signature
URL: <https://mail.openjdk.org/pipermail/hotspot-gc-dev/attachments/20171106/445f6dae/signature.asc>


More information about the hotspot-gc-dev mailing list