[PATCH] 8217561 : X86: Add floating-point Math.min/max intrinsics, approval request
jatin.bhateja at intel.com
Sun Feb 17 09:02:37 UTC 2019
I have added a benchmark over the revision shared by Bernard
Following is the link to updated patch.
> -----Original Message-----
> From: Vladimir Kozlov [mailto:vladimir.kozlov at oracle.com]
> Sent: Saturday, February 16, 2019 7:19 AM
> To: B. Blaser <bsrbnd at gmail.com>; Nils Eliasson <nils.eliasson at oracle.com>
> Cc: Viswanathan, Sandhya <sandhya.viswanathan at intel.com>; hotspot-
> compiler-dev at openjdk.java.net; Bhateja, Jatin <jatin.bhateja at intel.com>;
> eric.caspole <eric.caspole at oracle.com>
> Subject: Re: [PATCH] 8217561 : X86: Add floating-point Math.min/max
> intrinsics, approval request
> Eric C. asked to add JMH microbanchmark to show performance of new
> code. You may be missed his e-mail:
> We need to see benefits from these changes and verify them before pushing
> On 2/15/19 12:35 PM, B. Blaser wrote:
> > Thanks for your approval Nils and for the other reviews from Vladimir
> > and Sandhya.
> > Are we waiting for some more benchmarks or can I already push Jatin's
> > contribution like this (webrev.04)?
> > Bernard
> > On Fri, 15 Feb 2019 at 14:33, Nils Eliasson <nils.eliasson at oracle.com>
> >> Looks good!
> >> What benchmark did you use? Is it in the microbenchmark suite?
> >> //N
> >> On 2019-02-15 12:49, Viswanathan, Sandhya wrote:
> >>> Hi Vladimir/Nils, Could one of you please review and approve the patch?
> You have looked at the previous versions of this patch so hoping it will be a
> quick review for you.
> >>> The updated patch from Bernard looks ok to me. I am not an official
> reviewer though.
> >>> https://bugs.openjdk.java.net/browse/JDK-8217561
> >>> http://cr.openjdk.java.net/~bsrbnd/jdk8217561/webrev.04/
> >>> Best Regards,
> >>> Sandhya
More information about the hotspot-compiler-dev