Heads-up: JDK-8230085 and MacOS Catalina

[Langer, Christoph]

Hi,

as per the latest discussion in the bug (https://bugs.openjdk.java.net/browse/JDK-8230085), we have agreed to bring this patch down to OpenJDK 11.0.5 and 8u232. So I've pushed it to jdk11u now (http://hg.openjdk.java.net/jdk-updates/jdk11u/rev/ee7128cf507a).

With this submit, the public part of OpenJDK 11.0.5 is done and all publicly known fixes that are contained in Oracle's 11.0.5 release have been backported, as per https://bugs.openjdk.java.net/issues/?filter=36515, which is empty now. �� Thanks to all people that were involved in this effort!

Best regards
Christoph

> -----Original Message-----
> From: Langer, Christoph
> Sent: Freitag, 20. September 2019 17:08
> To: Lindenmaier, Goetz <goetz.lindenmaier at sap.com>; Simon Tooke
> <stooke at redhat.com>; Andrew John Hughes <gnu.andrew at redhat.com>;
> jdk-updates-dev at openjdk.java.net; Aleksey Shipilev <shade at redhat.com>;
> Andrew Haley <aph at redhat.com>
> Subject: RE: Heads-up: JDK-8230085 and MacOS Catalina
> 
> Hi,
> 
> generally, I prefer to align with Oracle.
> 
> So, I'll hold back the push until next Tuesday (24th of September) which is
> the last open day for 11u before 11.0.5 CPU work starts. If I don't get any
> input that changes things until then, I'll push it to jdk11u-dev (11.0.6) after
> Tuesday.
> 
> Best regards
> Christoph
> 
> > -----Original Message-----
> > From: Lindenmaier, Goetz <goetz.lindenmaier at sap.com>
> > Sent: Donnerstag, 19. September 2019 14:23
> > To: Langer, Christoph <christoph.langer at sap.com>; Simon Tooke
> > <stooke at redhat.com>; Andrew John Hughes
> <gnu.andrew at redhat.com>;
> > jdk-updates-dev at openjdk.java.net; Aleksey Shipilev
> <shade at redhat.com>;
> > Andrew Haley <aph at redhat.com>
> > Subject: RE: Heads-up: JDK-8230085 and MacOS Catalina
> >
> > Hi,
> >
> > the change touches only mac files, except for removal of a 'private'
> > from a method.  So it should not break any other platform.
> > Catalina is to be released in October:
> > https://www.apple.com/macos/catalina/
> > If we deliver the fix only in January, people have to
> > deal with the issue until then.  Thus I would not object against
> > pushing this to 11.0.5.
> >
> > The only thing I'm not aware of is the importance of this fix.
> > Does isReadOnly() being true imply that you can not write to
> > any file any more? This would make 11.0.5 unusable on Catalina.
> > Or is this just an interface rarely used?
> >
> > Best regards,
> >   Goetz
> >
> >
> >
> >
> >
> > > -----Original Message-----
> > > From: jdk-updates-dev <jdk-updates-dev-bounces at openjdk.java.net>
> On
> > > Behalf Of Langer, Christoph
> > > Sent: Donnerstag, 19. September 2019 11:19
> > > To: Simon Tooke <stooke at redhat.com>; Andrew John Hughes
> > > <gnu.andrew at redhat.com>; jdk-updates-dev at openjdk.java.net;
> Aleksey
> > > Shipilev <shade at redhat.com>; Andrew Haley <aph at redhat.com>
> > > Subject: [CAUTION] RE: Heads-up: JDK-8230085 and MacOS Catalina
> > >
> > > Hi,
> > >
> > > I can see that Oracle has backported the change to their January update
> > > branches just last night. I'll ask if they have any plans to ship this already
> > with
> > > the October patches.
> > >
> > > Independent of that, I think we can decide for the OpenJDK releases to
> > have it
> > > already in October. What do people think? (Looping in aph...)
> > >
> > > Best regards
> > > Christoph
> > >
> > > > -----Original Message-----
> > > > From: jdk-updates-dev <jdk-updates-dev-bounces at openjdk.java.net>
> > On
> > > > Behalf Of Simon Tooke
> > > > Sent: Mittwoch, 18. September 2019 19:11
> > > > To: Andrew John Hughes <gnu.andrew at redhat.com>; jdk-updates-
> > > > dev at openjdk.java.net; Aleksey Shipilev <shade at redhat.com>
> > > > Subject: Re: Heads-up: JDK-8230085 and MacOS Catalina
> > > >
> > > >
> > > > On 9/17/2019 4:54 PM, Andrew John Hughes wrote:
> > > > >
> > > > > On 17/09/2019 20:13, Simon Tooke wrote:
> > > > >> On 9/16/2019 10:08 AM, Aleksey Shipilev wrote:
> > > > >>> On 9/16/19 4:04 PM, Langer, Christoph wrote:
> > > > >>>> does tck-red mean, there are issues with current TCK? In other
> > words,
> > > > it can't just simply be
> > > > >>>> backported before TCK has been adapted?
> > > > >>>
> https://wiki.openjdk.java.net/display/HotSpot/JBS+Label+Dictionary
> > > > says: "tck-red-(Rel): Used to
> > > > >>> identify TCK conformance stoppers (e.g. failure of a valid TCK test
> > that
> > > > exists in a shipped TCK)."
> > > > >>>
> > > > >>> From this and issue description it sounds like it is legit TCK failure on
> > > > Catalina.
> > > > >> I've now tested the fix on jdk8u and 11u on a Catalina beta, with
> > success.
> > > > >>
> > > > >> I've applied jdk8u-fix-request and jdk11u-fix-request labels, but if
> > > > >> they are critical (not my call) then the labels should be changed
> > > > >> accordingly.
> > > > >>
> > > > >> -Simon
> > > > >>
> > > > > If you want to request critical, you need to use the *-critical-request
> > > > > labels and make the case for it on the bug. In the event the issue is
> > > > > rejected for critical, you can always then apply for *-fix-request
> instead.
> > > > >
> > > > > Hope that helps,
> > > >
> > > > Sure, I'll do that.  I noticed that Christoph has already decided it's
> > > > not critical for 11u, though, so I won't be surprised if it gets rejected.
> > > >
> > > > Thanks,
> > > >
> > > > -simon
> > > >