[11u] RFR 8224487: outputStream should not be copyable

[Thomas Stüfe]

Thank you Christoph. Patch has been tested in 11u-dev for the last ~10
nights.

Cheers, Thomas

On Mon, Jun 24, 2019 at 9:44 AM Langer, Christoph <christoph.langer at sap.com>
wrote:

> Hi Thomas,
>
> I think this is good. Looks like you resolved the rejects correctly.
>
> Will you run it through our test system? I don't see the patch in there
> currently...
>
> Thanks & Best regards
> Christoph
>
> > -----Original Message-----
> > From: hotspot-dev <hotspot-dev-bounces at openjdk.java.net> On Behalf Of
> > Thomas Stüfe
> > Sent: Donnerstag, 13. Juni 2019 18:29
> > To: jdk-updates-dev at openjdk.java.net
> > Cc: HotSpot Open Source Developers <hotspot-dev at openjdk.java.net>
> > Subject: [11u] RFR 8224487: outputStream should not be copyable
> >
> > Hi all,
> >
> > I would like to backport to 11u
> >
> > https://bugs.openjdk.java.net/browse/JDK-8224487.
> >
> > It is a precondition to backport three other fixes surrounding
> > stringStream:
> > -  https://bugs.openjdk.java.net/browse/JDK-8224193 (stringStream should
> > not use Resource Area)
> > -  https://bugs.openjdk.java.net/browse/JDK-8220394 (bufferedStream
> > does
> > not honor size limit)
> > - https://bugs.openjdk.java.net/browse/JDK-8225225 (stringStream
> internal
> > buffer should always be zero terminated)
> >
> > Original RFR discussion:
> > https://mail.openjdk.java.net/pipermail/hotspot-dev/2019-
> > May/038208.html
> > Original patch: http://hg.openjdk.java.net/jdk/jdk/rev/0927d8c7296f
> >
> > Full patch (with 11u corrections):
> > http://cr.openjdk.java.net/~stuefe/webrevs/backports/8224487-make-
> > streams-not-copyable.patch
> > Delta to original patch:
> > http://cr.openjdk.java.net/~stuefe/webrevs/backports/8224487-make-
> > streams-not-copyable-11u-changes.patch
> >
> > The patch disables copy and assignment on outputStream child classes,
> since
> > this has been a source of errors (unintended sharing of the stream
> backing
> > buffer between two instances of stringStream, for instance). It fixes all
> > resulting build errors - which mostly indicate real errors.
> >
> > Patch did not apply cleanly since 11u misses some work in the event log
> > Coleen did in 12, and a small change Lutz Schmidt did for the code heap
> > printer.
> >
> > Thanks for the review.
> >
> > Cheers, Thomas
>