RFR(M) 8213348: jdk.internal.vm.compiler.management service providers missing in module descriptor
dean.long at oracle.com
dean.long at oracle.com
Thu Nov 8 20:37:11 UTC 2018
On 11/8/18 11:49 AM, Mandy Chung wrote:
> This is strange. upgrade module path precedes the application module
> path in the search path. I would not expect --module-path would help.
>
Sorry, the error is if I remove both --upgrade-module-path and
--module-path.
Putting back either one makes it work again, so I can remove either one.
I don't remember why I added --module-path, but it was probably when I was
experimented with using --module-source-path instead of -sourcepath.
Should I go ahead and remove --module-path?
> Can you get the full javac command to see what are being compiled
> and all the paths set in these options?
>
It's very long, but I can summarize. -sourcepath contains source paths
under
src/jdk.internal.vm.compiler only. The files we are compiling come from
_gensrc_proc_files, which contains only files under
jdk.internal.vm.compiler.
dl
> Mandy
>
> On 11/8/18 11:28 AM, dean.long at oracle.com wrote:
>> I get this without --module-path:
>>
>> Fatal Error: Unable to find package java.lang in classpath or
>> bootclasspath
>>
>> I guess before, when it was using the unnamed module, that it was
>> picking up these
>> system classes from the jdk10 bootjdk, which doesn't seem ideal.
>>
>> It works without --upgrade-module-path, so I can remove that.
>>
>> dl
>>
>> On 11/7/18 8:12 PM, Mandy Chung wrote:
>>> Hi Dean,
>>>
>>> On 11/7/18 7:56 PM, dean.long at oracle.com wrote:
>>>> https://bugs.openjdk.java.net/browse/JDK-8213348
>>>> https://bugs.openjdk.java.net/browse/JDK-8211781
>>>> http://cr.openjdk.java.net/~dlong/8213348/webrev/
>>>>
>>> 108 --module-path $(JDK_OUTPUTDIR)/modules \
>>> 109 --upgrade-module-path $(JDK_OUTPUTDIR)/modules --system none \
>>>
>>> You added `--module-path` option to javac. What error did you
>>> get without it? It looks a bit suspicious. I wonder if
>>> the `--upgrade-module-path` is needed as --system none is set.
>>>
>>> We need Jon to advice on this javac command.
>>>
>>> Mandy
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.openjdk.java.net/pipermail/hotspot-compiler-dev/attachments/20181108/35e7d8cf/attachment-0001.html>
More information about the hotspot-compiler-dev
mailing list