RFR: 8326718: Test java/util/Formatter/Padding.java does not timeout on large inputs before JDK-8299677

Alan Bateman alanb at openjdk.org
Tue Feb 27 19:15:52 UTC 2024


On Tue, 27 Feb 2024 17:24:03 GMT, Chad Rakoczy <duke at openjdk.org> wrote:

> [JDK-8299677](https://bugs.openjdk.java.net/browse/JDK-8299677) fixes a bug with Formatter.format taking a long time when there is a lot of padding. This test runs Formatter.format with very large padding. Test fails before [JDK-8299677](https://bugs.openjdk.java.net/browse/JDK-8299677) and passes after.
> 
> Timeout for the test was set to 10 seconds. Test passes locally with as low as 1 (after [JDK-8299677](https://bugs.openjdk.java.net/browse/JDK-8299677)) and fails as high as 120 (before [JDK-8299677](https://bugs.openjdk.java.net/browse/JDK-8299677)) so it should be consistent.

test/jdk/java/util/Formatter/Padding.java line 29:

> 27:  * @summary Tests to excercise padding on int and double values,
> 28:  *      with various flag combinations.
> 29:  * @run junit/timeout=10 Padding

/timeout=10 looks problematic, I don't think you want that.

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

PR Review Comment: https://git.openjdk.org/jdk/pull/18033#discussion_r1504811214


More information about the core-libs-dev mailing list