Build failing on macOS in hotspot module

Jaikiran Pai jai.forums2013 at gmail.com
Fri Jul 19 12:21:09 UTC 2019


Thank you Alan.

Sorry, my last update was early this morning and I should have fetched
latest before sending out this mail. Fetched the latest now.

-Jaikiran

On 19/07/19 5:45 PM, Alan Bateman wrote:
> This seems to have been fixed a few hours ago:
>    http://hg.openjdk.java.net/jdk/jdk/rev/4d421888ad63
>
> On 19/07/2019 13:12, Jaikiran Pai wrote:
>> I just updated my local jdk workspace to the latest default branch
>> and did:
>>
>> bash configure
>>
>> make clean
>>
>> make images
>>
>> The build keeps failing with:
>>
>> ERROR: Build failed for target 'images' in configuration
>> 'macosx-x86_64-server-release' (exit code 2)
>> Stopping sjavac server
>>
>> === Output from failing command(s) repeated here ===
>> * For target hotspot_variant-server_libjvm_objs_shenandoahSupport.o:
>> openjdk/jdk/src/hotspot/share/gc/shenandoah/c2/shenandoahSupport.cpp:3019:33:
>>
>> error: operator '?:' has lower precedence than '+'; '+' will be
>> evaluated first [-Werror,-Wparentheses]
>>    return Node::hash() + _native ? 1 : 0;
>>           ~~~~~~~~~~~~~~~~~~~~~~ ^
>> openjdk/jdk/src/hotspot/share/gc/shenandoah/c2/shenandoahSupport.cpp:3019:33:
>>
>> note: place parentheses around the '+' expression to silence this
>> warning
>>    return Node::hash() + _native ? 1 : 0;
>>                                  ^
>>           (                     )
>> openjdk/jdk/src/hotspot/share/gc/shenandoah/c2/shenandoahSupport.cpp:3019:33:
>>
>> note: place parentheses around the '?:' expression to evaluate it first
>>    return Node::hash() + _native ? 1 : 0;
>>                                  ^
>>                          (              )
>> 1 error generated.
>>
>> * All command lines available in
>> openjdk/jdk/build/macosx-x86_64-server-release/make-support/failure-logs.
>>
>> === End of repeated output ===
>>
>>
>> I'm on macOS 10.14.1. The last working state was a few days back, I
>> think last week, when my local builds were working fine. Is this an
>> expected error, requiring some tool upgrade? Or does this need a fix in
>> the cpp code?
>>
>> -Jaikiran
>


More information about the build-dev mailing list