On 6/28/2016 3:51 PM, Sergey Bylokhov wrote:
On 28.06.16 12:49, Semyon Sadetsky wrote:
It seems the root cause is in the Util.generateOOME() method from the
regtesthelpers library. It should take into account the heap maximum
limit to generate OOME in a reasonable time.
The root cause is that the tests tries to allocate more than 250 GB of
memory, I do not see the reason why it should do it, since 128m is
enough for them.
Is it the case for a 250 GB RAM server only?
I think Util.generateOOME() should be improved in any case, to prevent
future tests to be slowdown/failed if running on large RAM servers.
This utility method is used in other tests which are not covered by the
fix but may fail due to the same reason.
Which one?
bug6795356.java
TwentyThousandTest.java
WindowsLeak.java
To find them all you can search for Util.generateOOME() usages
--Semyon
On 6/27/2016 6:25 PM, Sergey Bylokhov wrote:
Hello.
Please review the fix for jdk9.
Some of our tests fails with timeout when they tries to generate OOM.
This occur on the systems which have huge number of memory.
In the fix I added "-xm" option to minimize the available amount of
memory.
Bug: https://bugs.openjdk.java.net/browse/JDK-8159899
Webrev can be found at:
http://cr.openjdk.java.net/~serb/8159899/webrev.00
