Hi Mani,

I think Martijn's suggestion works better : - " it will be good to be able to encourage various OpenJDK
members to contribute / port tests into it and measure that. Small steps"

Rgds,Rory

On 19/01/2015 22:56, Mani Sarkar wrote:
Hi all,

On the below I was going to suggest, is it possible to have JCov metrics from other non-published tests and sources run on various environments and then combine them with the public version (from our servers).

John Oliver did a patch for JaCoCO long ago, that way we can see pathways from other tests which only improve overall coverage metrics. Is there something available with the JCov tool itself ?

Is this an idea that is feasible ?

Cheers,
Mani

--
@theNeomatrix369 <http://twitter.com/theNeomatrix369>*| **Blog <http://neomatrix369.wordpress.com/>**| *LJC Associate & LJC Advocate (@adoptopenjdk & @adoptajsr programs) *Meet-a-Project - *MutabilityDetector <https://github.com/MutabilityDetector>*| **Bitbucket <https://bitbucket.org/neomatrix369>* * | **Github <https://github.com/neomatrix369>** | **LinkedIn <http://uk.linkedin.com/pub/mani-sarkar/71/a77/39b>*
*Come to Devoxx UK 2015:* http://www.devoxx.co.uk/

*/Don't chase success, rather aim for "Excellence", and success will come chasing after you!/*


    Message: 4
    Date: Mon, 19 Jan 2015 17:28:11 +0000
    From: Martijn Verburg <martijnverb...@gmail.com
    <mailto:martijnverb...@gmail.com>>
    To: "Rory O'Donnell" <rory.odonn...@oracle.com
    <mailto:rory.odonn...@oracle.com>>
    Cc: "quality-discuss@openjdk.java.net
    <mailto:quality-discuss@openjdk.java.net>"
            <quality-discuss@openjdk.java.net
    <mailto:quality-discuss@openjdk.java.net>>
    Subject: Re: We have some Code Coverage results from JCov/JTreg!
    Message-ID:
<cap7yuas4rkwjfybhci_r5q0qr08mpr-sz_7hnxrln6n39cp...@mail.gmail.com <mailto:cap7yuas4rkwjfybhci_r5q0qr08mpr-sz_7hnxrln6n39cp...@mail.gmail.com>>
    Content-Type: text/plain; charset="utf-8"

    Hi Rory,

    Understood - it will be good to be able to encourage various OpenJDK
    members to contribute / port tests into it and measure that. Small
    steps
    :-).

    On 19 January 2015 at 16:35, Rory O'Donnell
    <rory.odonn...@oracle.com <mailto:rory.odonn...@oracle.com>>
    wrote:

    >
    > On 19/01/2015 15:36, Martijn Verburg wrote:
    >
    > Hi Rory,
    >
    >  Thanks - we're looking into the existing Cloudbees Jenkins
    instance to
    > do this.
    >
    > ok
    >
    >  A second question quick question - do the numbers we're
    publishing look
    > right compared to your internal ones?
    >
    >   Appreciate non OpenJDK tests run by Oracle means that they
    cover more.
    >
    >  Comparing our internal numbers with yours would be like
    comparing apples
    > with pears
    > I'm afraid.
    >
    > Rgds,Rory
    >
    >   Cheers,
    > Martijn
    >
    > On 19 January 2015 at 15:18, Rory O'Donnell
    <rory.odonn...@oracle.com <mailto:rory.odonn...@oracle.com>>
    > wrote:
    >
    >>
    >> On 19/01/2015 10:55, Martijn Verburg wrote:
    >>
    >>> Hi all,
    >>>
    >>> John Oliver and Mani Sarkar spent some time on the most recent
    Adopt
    >>> OpenJDK hackday and managed to get what looks like to be
    meaningful code
    >>> coverage numbers for OpenJDK using the jcov/jtreg tools:
    >>>
    >>> Results for jdk9: http://sticky.uwcs.co.uk/jcov/
    >>>
    >>> ========Code Tools Dev========
    >>>
    >>> The configuration John used was as follows (is this the
    correct usage
    >>> pattern?):
    >>>
    >>> Build jdk images
    >>> install jtreg with the jcov
    >>>
    >>> export the normal vars:
    >>>
    >>> ```
    >>> export SOURCE_CODE=/home/joliver/workspace/jdk9/
    >>> export JTREG_INSTALL=/home/joliver/workspace/jtreg
    >>> export JT_HOME=$JTREG_INSTALL
    >>> export JTREG_HOME=$JTREG_INSTALL
    >>> export
    >>>
    
PRODUCT_HOME=$SOURCE_CODE/build/linux-x86_64-normal-server-release/images/jdk
    >>>
    >>> export JPRT_JTREG_HOME=${JT_HOME}
    >>> export JPRT_JAVA_HOME=${PRODUCT_HOME}
    >>> export JTREG_TIMEOUT_FACTOR=5
    >>> export CONCURRENCY=8
    >>> ```
    >>>
    >>> cd into jdk/test
    >>>
    >>> edit the Makefile and add the following:
    >>>
    >>> ```
    >>> jdkroot=/home/joliver/workspace/jdk9/
    >>>
    >>> JTREG_TEST_OPTIONS +=
    >>>
    
-jcov/classes:$(jdkroot)/build/linux-x86_64-normal-server-release/jdk/modules/java.base
    >>> JTREG_TEST_OPTIONS +=
    >>> -jcov/source:$(jdkroot)/jdk/src/java.base/share/classes
    >>> JTREG_TEST_OPTIONS += -jcov/include:*
    >>> ```
    >>>
    >>> just before the line: # Make sure jtreg exists
    >>>
    >>> then just run "make test" inside the root
    >>>
    >>> =======Quality Discuss=======
    >>>
    >>  Hi Martijn,
    >>
    >> Posting the results on our wiki won't work, so I can provide a
    link to the
    >> results. Let me know when  you have decided on the link locations.
    >>
    >> Rgds,Rory
    >>
    >>>
    >>> Is this something that could be hosted by the quality group
    for the
    >>> major OpenJDK code lines (7u, 8u and jdk9)?
    >>>
    >>> If not then the Adoption Group can host it on one of their
    external
    >>> servers temporarily and we could link to that from the
    wiki(s)/project
    >>> page(s).
    >>>
    >>> Cheers,
    >>> Martijn
    >>>
    >>
    >>   --
    >> Rgds,Rory O'Donnell
    >> Quality Engineering Manager
    >> Oracle EMEA , Dublin, Ireland
    >>
    >>
    >
    > --
    > Rgds,Rory O'Donnell
    > Quality Engineering Manager
    > Oracle EMEA , Dublin, Ireland
    >
    >
    -------------- next part --------------
    An HTML attachment was scrubbed...
    URL:
    
<http://mail.openjdk.java.net/pipermail/quality-discuss/attachments/20150119/2c08545b/attachment.html>

    End of quality-discuss Digest, Vol 39, Issue 4
    **********************************************


--
Rgds,Rory O'Donnell
Quality Engineering Manager
Oracle EMEA , Dublin, Ireland

Reply via email to