[PATCH] 8217561 : X86: Add floating-point Math.min/max intrinsics, approval request

[B. Blaser]

Pushed:
http://hg.openjdk.java.net/jdk/jdk/rev/ff399127078a

Search tree printing commented here:
http://hg.openjdk.java.net/jdk/jdk/rev/ff399127078a#l8.221

Thanks to Jatin Bhateja for his contribution!
Bernard

On Thu, 7 Mar 2019 at 12:42, B. Blaser <bsrbnd at gmail.com> wrote:
>
> Thanks for your approval, Vladimir, I'll push it.
>
> I had timeouts on jdk/submit with the initial search tree but reducing
> the loop number made the test pass [1].
> It seems you have a stack overflow only when printing the tree at the
> end of the test on your SPARC system which isn't concerned by this
> fix.
> I'll comment out the recursive printing of the tree when pushing, the
> insertion being iterative.
> However, if we still have timeout/overflow reports on some systems,
> I'll comment out the search tree example as I added it only to try a
> realistic use case.
>
> Bernard
>
> [1] http://hg.openjdk.java.net/jdk/submit/rev/d164e0b595e6#l2.7
>
>
> On Thu, 7 Mar 2019 at 07:11, Vladimir Kozlov <vladimir.kozlov at oracle.com> wrote:
> >
> > I run tier1-3 testing with these changes.
> > All passed except compiler/intrinsics/math/TestFpMinMaxIntrinsics.java failed on SPARC:
> >
> > java.lang.reflect.InvocationTargetException
> >         at java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
> >         at java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62)
> > ...
> > Caused by: java.lang.StackOverflowError
> >         at compiler.intrinsics.math.TestFpMinMaxIntrinsics$Node.toString(TestFpMinMaxIntrinsics.java:262)
> >         at java.base/java.lang.invoke.StringConcatFactory$Stringifiers$ObjectStringifier.valueOf(StringConcatFactory.java:1702)
> >         at compiler.intrinsics.math.TestFpMinMaxIntrinsics$Node.toString(TestFpMinMaxIntrinsics.java:264)
> >         at java.base/java.lang.invoke.StringConcatFactory$Stringifiers$ObjectStringifier.valueOf(StringConcatFactory.java:1702)
> >         at compiler.intrinsics.math.TestFpMinMaxIntrinsics$Node.toString(TestFpMinMaxIntrinsics.java:264)
> >
> > Looks like deep recursion.
> >
> > It failed for last new @run commands with -XX:CompileCommand=dontinline,TestFpMinMaxIntrinsics.min*
> > TestFpMinMaxIntrinsics sortedSearchTree 1
> >
> > Vladimir
> >
> > On 3/6/19 4:55 PM, Vladimir Kozlov wrote:
> > > Okay. Lets push this version. Do you need sponsor to push?
> > >
> > > Thanks,
> > > Vladimir
> > >
> > > On 3/6/19 2:25 PM, B. Blaser wrote:
> > >> Here it is:
> > >>
> > >> http://cr.openjdk.java.net/~bsrbnd/jdk8217561/webrev.06/
> > >>
> > >> Any feedback is welcome (jdk/submit report is good),
> > >> Bernard