Differences between OpenJDK 8u and IcedTea8

[Volker Simonis]

On Thu, Jan 9, 2020 at 6:57 PM Volker Simonis <volker.simonis at gmail.com> wrote:
>
> On Thu, Jan 9, 2020 at 5:58 PM Severin Gehwolf <sgehwolf at redhat.com> wrote:
> >
> > Hi Volker,
> >
> > On Thu, 2020-01-09 at 17:36 +0100, Volker Simonis wrote:
> > > Hi Severin,
> > >
> > > thanks for the quick reply.
> > >
> > > On Thu, Jan 9, 2020 at 1:32 PM Severin Gehwolf <sgehwolf at redhat.com> wrote:
> > > > Hi Volker,
> > > >
> > > > On Thu, 2020-01-09 at 12:11 +0100, Volker Simonis wrote:
> > > > > Hi,
> > > > >
> > > > > I've just realized that ojk8u-dev currently doesn't build correctly if
> > > > > configured with "--with-zlib=system".
> > > >
> > > > Could you be more specific? What's working (if any) and what's broken
> > > > exactly? Is it this?
> > > >
> > > > https://icedtea.classpath.org/bugzilla/show_bug.cgi?id=1413
> > > >
> > >
> > > Yes, it's exactly that one. I have included a link to the patch [2] in
> > > my previous mail:
> > >
> > > [2] http://icedtea.classpath.org//hg/icedtea8-forest/jdk?cmd=changeset;node=c5b4565befea
> > >
> > > > > This was surprising to me,
> > > > > because I know that most  (all?) Linux distributions link their ojdk8
> > > > > packages dynamically against the system provided libz (and I could
> > > > > verify this on my Ubuntu machine).
> > > > >
> > > > > So I took a peek at the IcedTea 8 repo [1] and found that this issue
> > > > > was fixed there since a long time ago [2]. I was also surprised to see
> > > > > that the IcedTea 8 jdk repository actually contains quite some fixes
> > > > > and downports which are not in OpenJDK 8u yet [3] (e.g. HiDPI
> > > > > support).
> > > > >
> > > > > Now that Red Hat is leading the OpenJDK 8 updates project, is there
> > > > > any specific reason (except the lack of time :)) why you still keep
> > > > > these changes in IcedTea instead of pushing them upstream to
> > > > > jdk8u-dev?
> > > >
> > > > Yes.
> > > >
> > > > We don't use the IcedTea 8 tree for the Red Hat OpenJDK builds.
> > > > Although, I know some other distros use it. We use [1] plus some
> > > > patches. No IcedTea per se involved. We want to stay as close to
> > > > upstream as possible. Feel free to have a look at the SRPM if you're
> > > > interested.
> > > >
> > >
> > > That's interesting, although it doesn't really simplify the situation
> > > (at least not for me :). So why are the IcedTea 8 repos then still
> > > kept up to date?
> >
> > It's open source. Anyone can pick it up/maintain and use it :) AFAIK,
> > Andrew Hughes is mostly working on it in his spare time.
> >
> > > > "--with-zlib-system" works for us[2]. I haven't investigated why it
> > > > doesn't work for you.
> > > >
> > >
> > > Well "--with-zlib=system", definitely doesn't work for the plain
> > > http://hg.openjdk.java.net/aarch64-port/jdk8u-shenandoah. It also
> > > doesn't work for the plain OpenJDK sources contained in the source rpm
> > > I've downloaded from [1]. I realized that the source rpm contains a
> > > set of additional patches but I couldn't easily apply them manually.
> > > Nevertheless I greped through them and they don't seem to contain  the
> > > mentioned fix for PR1423 [2].
> >
> > My quick reading of the bug suggested it wasn't reproducible
> > everywhere. Could this be an Ubuntu/SUSE specific thing?
> >
>
> May be. I'm on Ubuntu. "ld" on Ubuntu seems to have "--as-needed" as
> default. This means that when you a command line like "-lXXX YYY.o"
> you will get undefined symbols in YYY.o if these symbols are satisfied
> by libXXX.so". The opposite is "--no-as-needed" which still makes
> libXXX.so a dependency of the resulting output. Do you know if the
> linker/gcc use "--no-as-needed" as default?

Wanted to ask if you know whether the linker/gcc on RHEL/Fedora  use
"--no-as-needed" as default?

>
> > > So maybe the fix finally gets into the
> > > sources somehow magically in a way I don't understand, but the
> > > question remains, why such trivial fixes don't get up-streamed?
> >
> > Not sure about that. I think it's mostly a matter of somebody-will-
> > have-to-spend-time-to-get-it-upstream question like you mentioned
> > earlier.
> >
> > Thanks,
> > Severin
> >
> > > [1] https://kojipkgs.fedoraproject.org//packages/java-1.8.0-openjdk/1.8.0.242.b04/0.0.ea.fc32/src/java-1.8.0-openjdk-1.8.0.242.b04-0.0.ea.fc32.src.rpm
> > > [2] https://icedtea.classpath.org/bugzilla/show_bug.cgi?id=1413
> > >
> > > > Thanks,
> > > > Severin
> > > >
> > > > [1] http://hg.openjdk.java.net/aarch64-port/jdk8u-shenandoah/
> > > > [2] For example: https://src.fedoraproject.org/rpms/java-1.8.0-openjdk/blob/master/f/java-1.8.0-openjdk.spec#_1659
> > > >
> > > > > Thank you and best regards,
> > > > > Volker
> > > > >
> > > > > [1]  http://icedtea.classpath.org/hg/icedtea8-forest/jdk
> > > > > [2] http://icedtea.classpath.org//hg/icedtea8-forest/jdk?cmd=changeset;node=c5b4565befea
> > > > > [3] http://icedtea.classpath.org/hg/icedtea8-forest/jdk/search/?rev=PR1&revcount=100
> > > > >
> >