[8u-dev, ppc] RFR for (almost clean) backport of 8158232
gromero at linux.vnet.ibm.com
Mon May 13 19:59:43 UTC 2019
Thanks for the backport and for the webrev.
I understand that offset adjustments in general, and particularly for this
backport, are not considered a change that needs to be reviewed again.
That said, and although I'm not a Reviewer, I tested it against SPECjvm and
microbenchmarks for byte, int, and long and reviewed the change for jdk8u-dev.
It looks good.
Please, provide a "Fix Request" comment to the original bug explaining that
the backport is low risk and affects PPC64-only, accordingly to  and .
Then please add the label "jdk8u-fix-request" to it.
Once the approval to push is granted I'll sponsor the change.
On 05/10/2019 03:55 AM, Kazunori Ogata wrote:
> Sorry, I forgot to put the links to the bug report and the original
> changeset Also forgot to mention that this changeset is needed to
> backport AES intrinsics support  on ppc64 big-endian.
> Bug report:
> Original change set
>  https://bugs.openjdk.java.net/browse/JDK-8188868
> "hotspot-compiler-dev" <hotspot-compiler-dev-bounces at openjdk.java.net>
> wrote on 2019/05/10 15:30:05:
>> From: "Kazunori Ogata" <OGATAK at jp.ibm.com>
>> To: hotspot-compiler-dev at openjdk.java.net, jdk8u-dev at openjdk.java.net
>> Date: 2019/05/10 15:31
>> Subject: [8u-dev, ppc] RFR for (almost clean) backport of 8158232
>> Sent by: "hotspot-compiler-dev"
> <hotspot-compiler-dev-bounces at openjdk.java.net>
>> May I get review for backport of 8158232: PPC64: improve byte, int and
>> long array copy stubs by using VSX instructions?
>> This changeset looks no conflict with the latest jdk8u-dev code, but the
>> patch command failed to apply it. It seems the patch command lost the
>> code regions to apply patches because stubGenerator_ppc.cpp has sets of
>> similar (but slightly different) functions.
>> I created new webrev mainly to update line numbers in the patch file. I
>> verified I can build fastdebug and release builds and there was no
>> degradation in "make test" results.
More information about the hotspot-compiler-dev