RFR 10 8183019: Diagnostics for onExitTest

[Langer, Christoph]

Hi Roger,

looks good. Reviewed from my side.

Best regards
Christoph

From: Roger Riggs [mailto:Roger.Riggs at Oracle.com]
Sent: Freitag, 30. Juni 2017 17:03
To: Langer, Christoph <christoph.langer at sap.com>; Core-Libs-Dev <core-libs-dev at openjdk.java.net>
Subject: Re: RFR 10 8183019: Diagnostics for onExitTest

Hi Christoph,

Thanks for the review.

webrev updated in place:
   http://cr.openjdk.java.net/~rriggs/webrev-onexit-8183019/
On 6/30/2017 2:24 AM, Langer, Christoph wrote:

Hi Roger,



the change looks good overall.



One thing that I probably don't fully understand:



Line 265: while (blines.isEmpty() && A_alive == A.isAlive()) {



Isn't this a bit dangerous for the case that A isn't alive any more, also wasn't at the time A_alive was initialized and blines is empty for some reason? Then you will end up with an endless loop. But maybe this is a bit theoretical...
Yes, the test can be simplified,  A_alive would be true (unless it died prematurely) so
the inner loop can just test for A.isAlive().

(jtreg timeouts would take care of the infinite loop but might leave a more confusing failure to diagnose).

Thanks, Roger



Best regards

Christoph