Fixing failing regression tests caused by the JEP 400: UTF-8 by Default. `JcmdOutputEncodingTest` test case uses `file.encoding=UTF-8` in `C` locale. The output from the agent library is in `UTF-8` so it succeeded before the JEP has been implemented, as System.out used `UTF-8` converter. After the JEP, it started failing because System.out is using `US-ASCII` which generates series of '?', ending up with assertion failure in the test.
The proposed fix is to pass `sun.stdout.encoding=UTF-8` as well as `file.encoding` so that tool process' System.out is in `UTF-8` as well. ------------- Commit messages: - Merge branch 'master' into JDK-8273187 - Reverted command encoding, then modified the test case to use sun.stdout.encoding - Used `transferTo()` - Merge branch 'master' into JDK-8273187 - Nuked UTF-8 conversion in tools - Merge branch 'master' into JDK-8273187 - 8273187: jtools tests fail with missing markerName check Changes: https://git.openjdk.java.net/jdk/pull/5539/files Webrev: https://webrevs.openjdk.java.net/?repo=jdk&pr=5539&range=00 Issue: https://bugs.openjdk.java.net/browse/JDK-8273187 Stats: 10 lines in 3 files changed: 3 ins; 4 del; 3 mod Patch: https://git.openjdk.java.net/jdk/pull/5539.diff Fetch: git fetch https://git.openjdk.java.net/jdk pull/5539/head:pull/5539 PR: https://git.openjdk.java.net/jdk/pull/5539
