<AWT Dev> Reviewer needed - awt/FontClass/CreateFont/DeleteFont test correction

[Dr Andrew John Hughes]

On 11:30 Thu 13 Jan     , Coleen Phillimore wrote:
> 
> >> After a couple rounds of testing, this change should be in hs20 which
> >> will be in jdk6u25.  I don't know how it gets into openjdk6.  I can
> >> barely follow our own release model.
> > 'our'?
> >
> Oracle's.

Ah, I thought that's what you meant but was confused because Oracle also
do the OpenJDK6 releases as far as I'm aware.

> >> I don't think you want to port it
> >> individually though.
> >>
> > OpenJDK6 is currently tracking the hs19 stable branch which has
> > 6938627 and hence this regression (including the broken AWT test).
> > Can you explain why it would not make sense to fix it?  There doesn't
> > seem to be a stable branch for hs20 on http://hg.openjdk.java.net yet
> > so it won't be imported into OpenJDK6 any time soon.
> >
> We're getting close to the stable hs20 branch.  I don't know how or when 
> it's imported to OpenJDK6 though so I shouldn't have tried to answer the 
> question.
> 

I usually do the work to import new versions of HotSpot into OpenJDK6 and Joe
approves it, as with all OpenJDK6 patches.  I've never really understood why
we need to have both a stable branch AND an OpenJDK6 HotSpot tree, given they
end up being essentially the same, but that's the way it is.

We're about to make a new release of OpenJDK6 (b21) with HotSpot 19.  We've
obviously don't want to import an entire new HotSpot at this juncture, but will
probably do so for b22 once the HotSpot tree becomes available.

>From what you say, my understanding is that 6938627 causes a regression from previous
HotSpot behaviour and your patch, 7009828, fixes that regression.  As OpenJDK6 has
6938627, would it not make sense to backport 7009828 as well to fix the regression?
If not, what am I missing here?  The patch doesn't seem hugely intrusive, and it reverts
back to using the previous expected behaviour of /tmp over java.io.tmpdir.

> Coleen
> >> Coleen
> >>>> Coleen
> >>>>
> >>>> On 1/12/2011 7:20 PM, Joe Darcy wrote:
> >>>>> Joe Darcy wrote:
> >>>>>> On 11/11/2010 4:01 AM, Pavel Tisnovsky wrote:
> >>>>>>> Phil Race wrote:
> >>>>>>>> Pavel,
> >>>>>>>>
> >>>>>>>> The font implementation and tests are 2D. Not AWT. Don't get thrown by
> >>>>>>>> package names.
> >>>>>>> Ah sorry, you are right, of course.
> >>>>>>>> I was a bit curious about the real reason for this failure.
> >>>>>>>> I wrote this test a couple of years ago and its never failed before.
> >>>>>>>> I don't see it failing on any 6-open build.
> >>>>>>>> I see it failing only in JDK 7 from b90. That seems to be due to
> >>>>>>>> HS18b03
> >>>>>>>> fix
> >>>>>>>> 6938627: Make temporary directory use property java.io.tmpdir when
> >>>>>>>> specified
> >>>>>>> It's failing on my side in OpenJDK 7 and in IcedTea6, because IT6 uses
> >>>>>>> some backports from OpenJDK 7 (from the log files I can only infer that
> >>>>>>> this bug first appeared in IcedTea6 when the hs19 were set as default
> >>>>>>> HotSpot on 2010-10-19). Is it still possible to push the test
> >>>>>>> correction
> >>>>>>> to OpenJDK 6 too, or only to OpenJDK 7?
> >>>>>>>
> >>>>>>> Also I'll need a bug id please.
> >>>>>>>
> >>>>>> Sorry for the belated reply, please use
> >>>>>>
> >>>>>> 7003327 Correct test awt/FontClass/CreateFont/DeleteFont
> >>>>>>
> >>>>>> Approved for OpenJDK 6.
> >>>>>>
> >>>>>> -Joe
> >>>>> Hello.
> >>>>>
> >>>>> If this can be pushed in the next day or two, I'm open to having this
> >>>>> bug fix in b21.
> >>>>>
> >>>>> Thanks,
> >>>>>
> >>>>> -Joe
> 

-- 
Andrew :)

Free Java Software Engineer
Red Hat, Inc. (http://www.redhat.com)

Support Free Java!
Contribute to GNU Classpath and IcedTea
http://www.gnu.org/software/classpath
http://icedtea.classpath.org
PGP Key: 94EFD9D8 (http://subkeys.pgp.net)
Fingerprint = F8EF F1EA 401E 2E60 15FA  7927 142C 2591 94EF D9D8