[11u] RFR 8234466: Class loading deadlock involving X509Factory#commitEvent()

Aleksey Shipilev shade at redhat.com
Thu Feb 6 10:03:15 UTC 2020


Thanks Paul, I added the approval request tag for this and the follow-up (8237508).

-Aleksey

On 2/5/20 7:14 PM, Hohensee, Paul wrote:
> Lgtm.
> 
> Paul
> 
> On 2/4/20, 5:12 AM, "jdk-updates-dev on behalf of Aleksey Shipilev" <jdk-updates-dev-bounces at openjdk.java.net on behalf of shade at redhat.com> wrote:
> 
>     Bug:
>       https://bugs.openjdk.java.net/browse/JDK-8234466
>       https://hg.openjdk.java.net/jdk/jdk/rev/6d386d360955
>     
>     The patch does not apply cleanly to 11u, because of two things:
>     
>       a) The move jdk.internal.misc -> jdk.internal.access happens in JDK-8211122, and that one is
>     already rejected from 11u backports. So, I had to rewrite the uses back to jdk.internal.misc.*.
>     
>       b) Test fails to compile due to the ambiguity in JarUtils: createJarFile(Path,Path,Path...) and
>     method createJarFile(Path,Path,String...). I believe it is caused by referencing one of the
>     non-updated JarUtils [2]. Instead of choosing the another testlibary (which probably has the impact
>     on the signing code), I opted to call the method that was available:
>      - JarUtils.createJarFile(Path.of(TEST_CLASS_PATH, jar.getFileName().toString()), xdir);
>      + JarUtils.createJarFile(Path.of(TEST_CLASS_PATH, jar.getFileName().toString()), xdir, Paths.get("."));
>     
>     11u webrev:
>       https://cr.openjdk.java.net/~shade/8234466/webrev.11u.01/
>     
>     Testing: new test (still passes), tier1, tier2, tier3
>     
>     -- 
>     Thanks,
>     -Aleksey
>     
>     [1] https://bugs.openjdk.java.net/browse/JDK-8211122
>     [2] https://mail.openjdk.java.net/pipermail/jdk-updates-dev/2020-February/002445.html



More information about the jdk-updates-dev mailing list