Hi,
Anyone can help push Andrew Haley's fix ASAP since current master cannot build?
http://cr.openjdk.java.net/~aph/8211207/
--
Thanks,
Pengfei
> -----Original Message-----
>
> Hi,
>
> On 09/27/2018 11:14 AM, Dmitry Samersoff wrote:
>
> > My $0.2
> >
> > It might be better to refactor
> > switch(tag) in c1_LIRGenerator_aarch64.cpp to usual case/break form,
> > rather than have multiple returns.
> >
> > i.e.
> >
> > switch (tag) {
> > 583 case floatTag: // fall trough
> > 584 case doubleTag:
> > do_ArithmeticOp_FPU(x);
> > break;
> > 585 case longTag:
> > do_ArithmeticOp_Long(x);
> > break;
> > 586 case intTag:
> > do_ArithmeticOp_Int(x);
> > break;
> > 587 default:
> > ShouldNotReachHere();
> > 588 }
> >
> > return;
>
> It would, but we follow x86 as much as we can, which makes porting changes
> from there much easier. This recently helped with the "Better Byte
> Behaviour" and "Better Interface Invocations" fixes, where the
> aarch64 patch required relatively little work beyond translating the
> x86 instructions into their obvious equivalents.
>
> So, we prefer consistency over style where this makes maintenance easier.
>
> --
> Andrew Haley
> Java Platform Lead Engineer
> Red Hat UK Ltd. <https://www.redhat.com>
> EAC8 43EB D3EF DB98 CC77 2FAD A5CD 6035 332F A671
