I see what you mean, but it's not really the point of the the BigInteger test, which is testing for an Arithmetic exception: The problem test is the one for 8021204 which constructs a String of "0"'s 2*31 long! which if you use 2bytes for each "0" is very likely to run out of memory! The problem is a "bug" in the testcase in that it doesn't quite ensure the environment is capable of running the test, by ensuring 1byte per char helps ensure that. Hotspot is that by default. I also suspect OOME is not certain either necessarily.
Andrew Leonard Java Runtimes Development IBM Hursley IBM United Kingdom Ltd internet email: [email protected] From: Brian Burkhalter <[email protected]> To: Andrew Leonard <[email protected]> Cc: Java Core Libs <[email protected]> Date: 17/03/2020 15:48 Subject: [EXTERNAL] Re: Sponsor Request: 8241097: java/math/BigInteger/largeMemory/SymmetricRangeTests.java requires -XX:+CompactStrings To clarify, could the compact strings setting be detected and maybe add another @run tag and expect an OOME? Not sure this is critical. Brian On Mar 17, 2020, at 8:06 AM, Brian Burkhalter <[email protected] > wrote: You’re welcome. I see. Let’s leave this hang out a bit for further comment. Also, is a test possible, or a modification of an existing test? Unless stated otherwise above: IBM United Kingdom Limited - Registered in England and Wales with number 741598. Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU
