8184800: Streamline RTM flag validity testing with generic flag testing support.
lutz.schmidt at sap.com
Thu Jul 20 09:29:25 UTC 2017
the changes look good to me. Thanks for your efforts.
How about a copyright update in
No new webrev required for that, of course.
On 20.07.2017, 08:56, "hotspot-compiler-dev on behalf of Lindenmaier, Goetz" <hotspot-compiler-dev-bounces at openjdk.java.net on behalf of goetz.lindenmaier at sap.com> wrote:
> -----Original Message-----
> From: Vladimir Kozlov [mailto:vladimir.kozlov at oracle.com]
> Sent: Mittwoch, 19. Juli 2017 17:13
> To: Lindenmaier, Goetz <goetz.lindenmaier at sap.com>; hotspot-compiler-
> dev at openjdk.java.net
> Subject: Re: 8184800: Streamline RTM flag validity testing with generic flag
> testing support.
> Thank you, Goetz
> Looks good to me.
> On 7/19/17 3:37 AM, Lindenmaier, Goetz wrote:
> > Hi,
> > please review this change. I please need a sponsor.
> > http://cr.openjdk.java.net/~goetz/wr17/8184800-rtmFlags/webrev.01/
> > RTM was implemented before specifying ranges for flags was established
> > in openJdk. It has custom checks implemented per platform in
> > vm_version_<cpu>.cpp as well as tests for these checks in
> > 8180612 added ranges for a row of RTM flags. Tests for RTMAbortRatio
> > and RTMTotalCountIncrRate now fail, because the range check failure
> > message is different.
> > This change removes the custom checks, implements a generic method
> > RTMTotalCountIncrRateConstraintFunc() and updates the tests. Some
> > tests are now superfluous as the ranges will be tested by
> > TestOptionsWithRanges.
> > I tested on x86/ppc machines with and without RTM support, as well
> > as on the other platforms we test automated.
> > Best regards,
> > Goetz.
More information about the hotspot-compiler-dev