RFR: JDK-8136991: [REDO] Additional number of processed references printed with -XX:+PrintReferenceGC after JDK-8047125

[Bengt Rutisson]

Hi Thomas,

>
>
> -------- Forwarded Message --------
> Subject: 	Re: RFR: JDK-8136991: [REDO] Additional number of processed 
> references printed with -XX:+PrintReferenceGC after JDK-8047125
> Date: 	Wed, 23 Sep 2015 11:31:23 +0200
> From: 	Thomas Schatzl <thomas.schatzl at oracle.com>
> To: 	Bengt Rutisson <bengt.rutisson at oracle.com>
> CC: 	hotspot-gc-dev at openjdk.java.net <hotspot-gc-dev at openjdk.java.net>
>
>
>
> Hi Bengt,
>
> On Wed, 2015-09-23 at 10:33 +0200, Bengt Rutisson wrote:
> > Forgot to say that there is a diff compared to the previous version here:
> >http://cr.openjdk.java.net/~brutisso/8136991/webrev.02-03-diff/
>
>    the regexp in the test is not correct for product mode, as the number
> of references is not printed for JNI weak references there.
>
> Line
>
>    48       "#[0-9]+: \\[JNI Weak Reference, [0-9]+ refs, [0-9]+\\.[0-9]+
> secs\\]");
>
>
> should be
>
>    48       "#[0-9]+: \\[JNI Weak Reference, [[0-9]+ refs, ]?[0-9]+\
> \.[0-9]+ secs\\]");
>
> I think to optionally match the number of refs for JNI weak references.
> Otherwise one could make the regexp dependent on the build type it is
> run (using Platform.isDebug()).


Good catch! I went for the optional count.

Updated webrev:
http://cr.openjdk.java.net/~brutisso/8136991/webrev.04/

Only the test has been changed since the last webrev.

Thanks,
Bengt

>
> Thanks,
>    Thomas
>
>
>
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mail.openjdk.org/pipermail/hotspot-gc-dev/attachments/20150923/b359caba/attachment.htm>