[aarch64-port-dev ] Fwd: RFR(M): 8181616: FMA Vectorization on x86

[Yang Zhang]

Hi Andrew

I'd like to take a look at this task. If there is anything we can do,
I will open a bugs.openjdk.java.net bug.

Regards
Yang

On 7 June 2017 at 03:52, Andrew Haley <aph at redhat.com> wrote:
> If anybody fancies taking a look at this one, please let me know.  I
> think it'd be quite fun.  Mind you, we only have 128-bit vectors so it
> might be that there's nothing we can do until we have SVE to take
> advantage of this.
>
> Andrew.
>
>
> ---------- Forwarded message ----------
> From: "Deshpande, Vivek R" <vivek.r.deshpande at intel.com>
> To: "hotspot-compiler-dev at openjdk.java.net" <hotspot-compiler-dev at openjdk.java.net>, Vladimir Kozlov <vladimir.kozlov at oracle.com>
> Cc: "Viswanathan, Sandhya" <sandhya.viswanathan at intel.com>
> Bcc:
> Date: Mon, 5 Jun 2017 21:42:47 +0000
> Subject: RFR(M): 8181616: FMA Vectorization on x86
> Hi All
>
> The enhancement generates Vector FMA instructions using SuperWord Analysis for X86.
> We observed upto 6x gain on Skylake and KNL using 512 bit Vector FMA instructions using FMA API operating on independent arrays.
> Webrev:
> http://cr.openjdk.java.net/~vdeshpande/8181616/webrev.00/
> I have also updated the JBS entry.
> https://bugs.openjdk.java.net/browse/JDK-8181616
> Would you please review and sponsor it.
>
> Regards,
> Vivek
>
>