Re: RFR(s): 8212828: (process) Change the Process launch mechanism default on Linux to be posix_spawn

On Mon, Feb 11, 2019 at 9:18 PM Martin Buchholz wrote: > Looks good to me. > > Thank you, I just pushed. > It's true that these tests depending on external tools are very brittle. > In particular, strace is in the middle of a flag re-org > >-e trace=%process >-e trace=process

Re: RFR 8218228 : The constructor StringBuffer(CharSequence) violates spec for negatively sized argument

Hi Roger! On 2/11/19 7:30 AM, Roger Riggs wrote: Hi Ivan, I called it out because the CSR does not mention that the behavior of some of the cases (-1..-16) is changing and some of the emails asserted there was no change in behavior. I'm fine with one changeset as long as both changes are

Re: JDK-6982173: Small problem causing thousands of wasted CPU hours

The current implementation seems very counter-intuitive with anything that doesn't have the same comparison semantics, for instance a TreeSet/Map with a Comparator, Identity-based Set/Map etc. The output of the following snippet would probably surprise most: /* --- snip --- */ import

Java SSLSocketChannel/SSLSelector?

Hi java-core community, I have been directed to this channel to discuss matters related to Java performant ssl communications. Maybe this is a topic that has been discussed in the past, in which case I would appreciate if someone pointed me to that particular topic. Back in 2002 I personally

Re: RFR(s): 8212828: (process) Change the Process launch mechanism default on Linux to be posix_spawn

Looks good to me. It's true that these tests depending on external tools are very brittle. In particular, strace is in the middle of a flag re-org -e trace=%process -e trace=process (deprecated) Nevertheless, we have such tests - are they worth the maintenance burden? My own

Re: RFR(s): 8212828: (process) Change the Process launch mechanism default on Linux to be posix_spawn

Thank you Roger! On Mon, Feb 11, 2019 at 7:55 PM Roger Riggs wrote: > Great! Looks fine. > > Thanks, Roger > > On 02/11/2019 01:50 PM, Thomas Stüfe wrote: > > Hi Roger, Martin, > > hopefully final version: > > >

Re: RFR(s): 8212828: (process) Change the Process launch mechanism default on Linux to be posix_spawn

Great!  Looks fine. Thanks, Roger On 02/11/2019 01:50 PM, Thomas Stüfe wrote: Hi Roger, Martin, hopefully final version: http://cr.openjdk.java.net/~stuefe/webrevs/8213192--(process)-change-the-process-launch-mechanism-default-on-linux-to-be-posix_spawn/webrev.03/webrev/

Re: RFR(s): 8212828: (process) Change the Process launch mechanism default on Linux to be posix_spawn

Hi Roger, Martin, hopefully final version: http://cr.openjdk.java.net/~stuefe/webrevs/8213192--(process)-change-the-process-launch-mechanism-default-on-linux-to-be-posix_spawn/webrev.03/webrev/ I removed the test and the changes in the test library made for the test. Test is just too brittle

Re: RFR 8218228 : The constructor StringBuffer(CharSequence) violates spec for negatively sized argument

Hi Ivan, I called it out because the CSR does not mention that the behavior of some of the cases (-1..-16) is changing and some of the emails asserted there was no change in behavior. I'm fine with one changeset as long as both changes are explicit. The bug and the CSR should both be updated.

Re: RFR(s): 8212828: (process) Change the Process launch mechanism default on Linux to be posix_spawn

Hi Thomas, I'd be fine with leaving it out.  It only provides confirmation of the library call, nothing algorithmic or complex and there is no bug to verify. So yes, drop it and save the test time and maintenance. Thanks, Roger On 02/09/2019 10:24 AM, Thomas Stüfe wrote: Hi Roger, I start

Re: RFR: jsr166 integration 2019-02

> On 9 Feb 2019, at 19:25, Martin Buchholz wrote: > > Thanks, Chris. > > On Sat, Feb 9, 2019 at 6:34 AM Chris Hegarty > wrote: >> 8215359: InnocuousForkJoinWorkerThread.setContextClassLoader needlessly >> throws >>