Hi Ivan, Thanks, Roger There is a pending enhancement to ProcessHandle to deal with uniqueness of pids. I can look at adding a note to process.getPid also. (The pid will be the pid when the process was alive.) Roger On 6/3/2015 6:58 AM, Ivan Gerasimov wrote:

Hi Roger!

Seems Okay to me.

Do you think it would make sense to make the doc for Process.getPid() clearer that it guarantees to return the same pid even after calling waitFor()? This's true for current implementation, but from the OS point of view, pid does not necessarily uniquely identifies the process after it has terminated.

Sincerely yours, Ivan

On 02.06.2015 21:33, Roger Riggs wrote:

...

Please review test cleanup to make process id testing more reliable. It now spawns javaChild to report the child pid instead of parsing the output of OS specific commands.

Webrev: http://cr.openjdk.java.net/~rriggs/webrev-builder-8067808/

Issue: http://cr.openjdk.java.net/~rriggs/webrev-builder-8067808/

Thanks, Roger

p.s. I'm looking for a review also for: 8081567 : java/lang/ProcessHandle/InfoTest.java failed Cannot run program "whoami"