[8u] Request for approval for JDK-8169201: Montgomery multiply intrinsic should use correct name

[Rob McKenna]

Approved

    -Rob

On 13/04/18 17:20, Aleksey Shipilev wrote:
> (realized this was never pushed)
> 
> Ping?
> 
> -Aleksey
> 
> On 03/21/2018 04:54 PM, Aleksey Shipilev wrote:
> > It is a really trivial patch:
> >   http://hg.openjdk.java.net/jdk9/jdk9/hotspot/rev/110b1774a359
> > 
> > diff -r 96e9e479ff8a src/share/vm/opto/library_call.cpp
> > --- a/src/share/vm/opto/library_call.cpp	Thu Aug 24 01:27:02 2017 -0400
> > +++ b/src/share/vm/opto/library_call.cpp	Wed Mar 21 16:49:39 2018 +0100
> > @@ -6026,7 +6026,7 @@
> >    }
> > 
> >    assert(UseMontgomeryMultiplyIntrinsic, "not implemented on this platform");
> > -  const char* stubName = "montgomery_square";
> > +  const char* stubName = "montgomery_multiply";
> > 
> >    assert(callee()->signature()->size() == 7, "montgomeryMultiply has 7 parameters");
> > 
> > 
> > It applies with a bit of fuzz to jdk8-dev.
> > 
> > JDK 9 bug:
> >   https://bugs.openjdk.java.net/browse/JDK-8169201
> > 
> > JDK 9 RFR:
> >   http://mail.openjdk.java.net/pipermail/hotspot-compiler-dev/2016-November/024899.html
> > 
> > We need this to avoid weird bugs in code that relies on intrinsic names:
> >   http://mail.openjdk.java.net/pipermail/aarch64-port-dev/2018-March/005424.html
> > 
> > Thanks,
> > -Aleksey
> > 
> 
>