Ivan,
I'll sponsor the change, but I want to push it after this week's
RT_Baseline snapshot. We snapshot on Tuesday (2014.09.09 at 1900 PT)
and if the nightly test results look reasonable Wednesday morning,
then I'll push it on Wednesday.
Does this sound acceptable to you?
Dan
On 9/7/14 10:03 PM, Ivan Gerasimov wrote:
Thanks Daniel and David!
I'll need a sponsor to help me push the change.
Sincerely yours,
Ivan
On 08.09.2014 6:03, David Holmes wrote:
Hi Ivan,
On 7/09/2014 4:07 PM, Ivan Gerasimov wrote:
Hello!
This is a proposal to address issue with wrong exit codes from a Java
processes on Windows.
In order to avoid a race, calls to _endthread(), exit and _exit() are
explicitly synchronized.
We allow simultaneous calls to _endthread() by multiple threads.
However, at the time exit() or _exit() is called, no calls to
_endthread() are allowed.
Some instrumentation added with JDK-8055338 remain in the code to help
diagnose the failures if they still occur.
BUGURL: https://bugs.openjdk.java.net/browse/JDK-8057744
WEBREV: http://cr.openjdk.java.net/~igerasim/8057744/0/webrev/
As per preliminary discussions this all looks okay.
One small concern I have is that the exit() may be arbitrarily
delayed if the mutexes are not assigned fairly, and the application
(more likely a test) is creating shortlived threads concurrently with
the exit attempt. But I guess we will just have to deal with that if
it arises.
Thanks,
David
Sincerely yours,
Ivan
