JMC-6122: Method profiling rule should show the hottest path

Marcus Hirt marcus.hirt at oracle.com
Fri Nov 30 01:28:06 UTC 2018


Hi Josh,

 

If Mario is busy, I can help sponsor this.

 

Kind regards,

Marcus

 

From: Joshua Matsuoka <jmatsuok at redhat.com>
Date: Wednesday, 28 November 2018 at 16:59
To: Marcus Hirt <marcus.hirt at oracle.com>, Mario Torre <neugens at redhat.com>
Cc: <jmc-dev at openjdk.java.net>
Subject: Re: JMC-6122: Method profiling rule should show the hottest path

 

Hi Marcus, 

 

Here's an exported patch with the requested change. Mario, could you sponsor it?

 

Cheers,

 

- Josh

 

On Tue, Nov 27, 2018 at 3:11 PM Marcus Hirt <marcus.hirt at oracle.com> wrote:

Hi Josh,

Looks good! Perhaps a new paragraph before the “The most common stack trace was”?

Kind regards,
Marcus


From: Joshua Matsuoka <jmatsuok at redhat.com>
Date: Tuesday, 27 November 2018 at 20:32
To: Marcus Hirt <marcus.hirt at oracle.com>
Cc: <jmc-dev at openjdk.java.net>
Subject: Re: JMC-6122: Method profiling rule should show the hottest path

Hi Marcus,

Thanks for the review! Here's an updated webrev

http://cr.openjdk.java.net/~jmatsuoka/JMC-6122/webrev.02/

Cheers,

- Josh

On Mon, Nov 26, 2018 at 5:31 PM Marcus Hirt <mailto:marcus.hirt at oracle.com> wrote:
Hi Josh,

1. Map.replace was introduced in JDK 8, and core is still compiled with JDK 7.

2. Just a nit - could the assignment of the fields for the
   MethodProfilingWindowResult please be in the order of the arguments?

Kind regards,
Marcus

On 2018-11-26, 22:26, "jmc-dev on behalf of Joshua Matsuoka" <mailto:jmc-dev-bounces at openjdk.java.net on behalf of mailto:jmatsuok at redhat.com> wrote:

    Forgot the patch: http://cr.openjdk.java.net/~jmatsuoka/JMC-6122/webrev.01/

    [1]
    https://bugs.openjdk.java.net/projects/JMC/issues/JMC-6122?filter=allopenissues

    Cheers,

    - Josh

    On Mon, Nov 26, 2018 at 4:24 PM Joshua Matsuoka <mailto:jmatsuok at redhat.com> wrote:

    > Hi,
    >
    > The following patch addresses JMC-6122 by changing the grouping aggregator
    > to instead group on stack trace so we can retrieve the hottest path
    > alongside the hottest method and display it in the rule results.
    >
    >
    > As for the comment on the bug [1], I think that belongs more in a separate
    > bug addressed by another patch.
    >
    > Thoughts?
    >
    > [1]
    >







More information about the jmc-dev mailing list