[aarch64-port-dev ] [aarch64-port-dev] [10] RFR: 8184049 : Matching rule for ubfiz

[Felix Yang]

Hi,

    I tried to modify the test case changing testI2() into:

    public static int testI2() {
        return (ia[0] & 0xf) << 30;
    }

    Then I got different execution results on aarch64 and x86:

aarch64:
java -XX:-TieredCompilation Test
2758214541841904631

x86:
java -XX:-TieredCompilation Test
2758195853365405696

    So I feel we are doing something wrong here. Please double check.
    Also I don't think we are handling the "format" of the matching rule
correctly: the shift information is ignored.

Thanks,
Felix

On 25 July 2017 at 20:38, Daniel Stewart <daniel.stewart at linaro.org> wrote:

> Good to push?
>
> On Mon, Jul 24, 2017 at 10:55 AM, Andrew Dinn <adinn at redhat.com> wrote:
>
> > On 24/07/17 14:37, Andrew Haley wrote:
> > > On 24/07/17 14:33, Daniel Stewart wrote:
> > >> Please see the webrev [1] for adding some matching rules to ubfiz [2].
> > >>
> > >> [1]: http://cr.openjdk.java.net/~njian/8184049/webrev.00/
> > >> [2]: https://bugs.openjdk.java.net/browse/JDK-8184049
> > >
> > > Looks very cool.  Send me a test case which exercises these patterns
> and
> > > I'll approve the patch.
> > Hmm, yes. Nice!
> >
> > regards,
> >
> >
> > Andrew Dinn
> > -----------
> > Senior Principal Software Engineer
> > Red Hat UK Ltd
> > Registered in England and Wales under Company Registration No. 03798903
> > Directors: Michael Cunningham, Michael ("Mike") O'Neill, Eric Shander
> >
>
>
>
> --
> Daniel Stewart
>