Questionable new-build failure
Kelly O'Hair
kelly.ohair at oracle.com
Fri Jan 25 21:55:31 UTC 2013
Please file JBS Issues.
And keep in mind every "Solaris 10" system could be different.
It depends on which Solaris 10 Update it is, and also potentially if it has all the latest patches or not.
We try and use Solaris 10u6 for building, usually with all latest required and recommended patches.
The changes to Solaris 10, via Updates and even patches has become quite aggressive, at least much
more aggressive than in the past. Which creates quite a moving target for anyone using the utilities
or expecting the exact same behaviors. I realize that this may be a goodness to some people. ;^)
And it's not easy to determine what you have.
Solaris 11 has introduced a 'uname -v' which will tell you if it is 11.0 or 11.1 (Update 1), but Solaris 10
does not have that. Best you can do is look at the /etc/release file.
-kto
On Jan 24, 2013, at 3:00 PM, David Holmes wrote:
> On 25/01/2013 7:44 AM, David Chase wrote:
>> I'm trying new-build on a Solaris 10 box, and it goes okay till this comparison:
>>
>> COMPARING
>> /export/drchase/jdk8tl/build/solaris-x86_64-normal-server-fastdebug/jdk/gensrc_x11wrappers/sizer/sizes.64
>> and
>> /export/drchase/jdk8tl/jdk/src/solaris/classes/sun/awt/X11/generator/sizes.64-solaris-i386
>>
>> The two files are identical if sorted, but the comparison does not pre-sort, and it fails after spewing all the differences.
>>
>> Any suggestions? Is this a bug, or yet another pilot error?
>
> It's a bug. Changes to this were only pushed in the past couple of days. As you note the generated file needs to be sorted before doing the comparison.
>
> David
>
>> I'm poking through WrapperGenerator.java to see if I see anything obvious.
>>
>> David
>>
More information about the build-dev
mailing list