RFR: 8371420: Still sporadic failures of gc/TestAlwaysPreTouchBehavior.java#<gcname> on Linux after JDK-8359104

[Matthias Baesken]

After [JDK-8359104](https://bugs.openjdk.org/browse/JDK-8359104) where getting RSS was improved, we still see sporadic failures of the test gc/TestAlwaysPreTouchBehavior.java#<gcname> , for different gcs.
E.g. gc/TestAlwaysPreTouchBehavior.java#Epsilon (with -Xlog:os=debug added for more output) 


----------System.err:(13/1278)----------
java.lang.RuntimeException: RSS of this process(24465408b) should be bigger than or equal to heap size(268435456b) (available memory: 16295006208). On Linux Kernel < 4.14 RSS can be inaccurate: expected 24465408 > 268435456


Looks like the RSS values we get are somewhat delayed/inaccurate .

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

Commit messages:
 - JDK-8371420

Changes: https://git.openjdk.org/jdk/pull/28735/files
  Webrev: https://webrevs.openjdk.org/?repo=jdk&pr=28735&range=00
  Issue: https://bugs.openjdk.org/browse/JDK-8371420
  Stats: 28 lines in 1 file changed: 8 ins; 0 del; 20 mod
  Patch: https://git.openjdk.org/jdk/pull/28735.diff
  Fetch: git fetch https://git.openjdk.org/jdk.git pull/28735/head:pull/28735

PR: https://git.openjdk.org/jdk/pull/28735