[8u-dev] RFR for non-clean backport of 8154156: PPC64: improve array copy stubs by using vector instructions

[Kazunori Ogata]

Hi Gustavo,

Thank you for reviewing and pointing the issue.  I wasn't aware that it 
can cause build error.  I'll be cautious.

Also thank you to sponsor this backport.


Regards,
Ogata

"Gustavo Romero" <gromero at linux.vnet.ibm.com> wrote on 2019/04/22 
23:02:18:

> From: "Gustavo Romero" <gromero at linux.vnet.ibm.com>
> To: Kazunori Ogata/Japan/IBM at IBMJP, jdk8u-dev at openjdk.java.net
> Cc: "Doerr, Martin" <martin.doerr at sap.com>, "Andrew Hughes" 
<gnu.andrew at redhat.com>
> Date: 2019/04/22 23:02
> Subject: Re: [8u-dev] RFR for non-clean backport of 8154156: PPC64: 
> improve array copy stubs by using vector instructions
> 
> Hi Ogata,
> 
> On 03/28/2019 09:55 AM, Kazunori Ogata wrote:
> > May I get review for non-clean backport of 8154156: PPC64: improve 
array
> > copy stubs by using vector instructions?
> > 
> <snip>
> > 
> > Original patch: http://hg.openjdk.java.net/jdk/jdk/rev/c9d756fa846e
> > Weberv: 
http://cr.openjdk.java.net/~horii/jdk8u_aes_be/8154156/webrev.02/
> 
> Change looks good.
> 
> Only a nit: it's missing a small change from Volker to cast a pointor to 
in:
> 
> user:        simonis
> date:        Mon May 04 18:41:53 2015 +0200
> summary:     8079280: Fix format warning/error in vm_version_ppc.cpp
> 
> diff -r 3d55863a66ba -r 751a7698b62e 
hotspot/src/cpu/ppc/vm/vm_version_ppc.cpp
> --- a/hotspot/src/cpu/ppc/vm/vm_version_ppc.cpp Sat May 02 00:16:29 2015 
-0700
> +++ b/hotspot/src/cpu/ppc/vm/vm_version_ppc.cpp Mon May 04 18:41:53 2015 
+0200
> @@ -629,7 +629,7 @@
>     // Print the detection code.
>     if (PrintAssembly) {
>       ttyLocker ttyl;
> -    tty->print_cr("Decoding dscr configuration stub at " INTPTR_FORMAT 
" 
> before execution:", code);
> +    tty->print_cr("Decoding dscr configuration stub at " INTPTR_FORMAT 
" 
> before execution:", p2i(code));
>       Disassembler::decode((u_char*)code, (u_char*)code_end, tty);
>     }
> 
> Without that change it will fail to compile on some distros, like 
Ubuntu.
> 
> I don't need a new webrev tho. I'll fix it before pushing.
> 
> So, hearing no objections I'll push it tomorrow to 
jdk8u/jdk8u-dev/hotspot.
> 
> Thanks for downporting it.
> 
> @Martin, thanks a lot for reviewing it.
> 
> Best regards,
> Gustavo