RFR: 8168850: Mark module entries that have been specified by --patch-module
Ioi Lam
ioi.lam at oracle.com
Mon Dec 5 05:34:28 UTC 2016
On 12/1/16 5:44 PM, David Holmes wrote:
> Hi Rachel,
>
> On 2/12/2016 7:18 AM, Rachel Protacio wrote:
>> Hi,
>>
>> Please review this fix, which adds a boolean to ModuleEntry to specify
>> whether the module has been patched using the command line
>> --patch-module. This is needed for CDS to know whether to load a class
>> from the shared archive. Their tests for those changes will encompass my
>> change here. I verified that it worked properly through the logging
>> messages, which can be triggered with
>> -Xlog:modules+patch=trace
>> Passes JPRT.
>>
>> Bug: https://bugs.openjdk.java.net/browse/JDK-8168850
>> Open webrev: http://cr.openjdk.java.net/~rprotacio/8168850.00/
>
> Looks okay.
>
> My only query is with the change in the initialization order for
> create_javabase. Does JPRT testing sufficiently exercise this?
>
The change in the initialization seems OK to me.
setup_patch_mod_entries() doesn't seem to rely on the result of
create_javabase(), so it should be OK to move create_javabase() after
setup_patch_mod_entries().
The JPRT tests usually run without --patch-modules, but also uses
--patch-modules in some test cases:
$grep Patch linux_x64_3.8-fastdebug-c2-hotspot_fast_runtime.log
TEST: runtime/modules/PatchModule/PatchModuleCDS.java
TEST: runtime/modules/PatchModule/PatchModuleDupJavaBase.java
TEST: runtime/modules/PatchModule/PatchModule2Dirs.java
TEST: runtime/modules/PatchModule/PatchModuleClassList.java
TEST: runtime/modules/PatchModule/PatchModuleDupModule.java
TEST: runtime/modules/PatchModule/PatchModuleJavaBase.java
TEST: runtime/modules/PatchModule/PatchModuleTest.java
TEST: runtime/modules/PatchModule/PatchModuleTestJar.java
TEST: runtime/modules/PatchModule/PatchModuleTestJarDir.java
TEST: runtime/modules/PatchModule/PatchModuleTraceCL.java
TEST: runtime/modules/Visibility/PatchModuleVisibility.java
So I think the coverage is pretty good.
Thanks
- Ioi
> Thanks,
> David
>
>> Thank you!
>> Rachel
More information about the hotspot-runtime-dev
mailing list