Re: RFR(XS): 8198585: add asserts to verify that ServiceUtil::visible_oop is not needed

2018-02-23 Thread Chris Plummer

Ok. I'll make that change.

thanks,

Chris

On 2/23/18 6:45 AM, Daniel D. Daugherty wrote:

Instead of "assert(false, ...", I recommend "fatal(...".
This will cause a failure in all build configs including 'release' bits.

Dan


On 2/22/18 8:16 PM, Chris Plummer wrote:

Hello,

Please review the following:

https://bugs.openjdk.java.net/browse/JDK-8198585
http://cr.openjdk.java.net/~cjplummer/8198585/webrev.00/webrev/

Before removing ServiceUtil::visible_oop(), I want to make sure it 
really isn't needed. Supposedly it should never return false, thus 
negating the need for its existence. This change adds asserts 
whenever false is returned. If it makes it all the way through 
promotion testing, then I'll delete the ServiceUtil::visible_oop() 
code and the references to it.


I tested by running all jdk and hotspot tier1-3 tests.

thanks,

Chris






Re: RFR(XS): 8198585: add asserts to verify that ServiceUtil::visible_oop is not needed

2018-02-23 Thread Daniel D. Daugherty

Instead of "assert(false, ...", I recommend "fatal(...".
This will cause a failure in all build configs including 'release' bits.

Dan


On 2/22/18 8:16 PM, Chris Plummer wrote:

Hello,

Please review the following:

https://bugs.openjdk.java.net/browse/JDK-8198585
http://cr.openjdk.java.net/~cjplummer/8198585/webrev.00/webrev/

Before removing ServiceUtil::visible_oop(), I want to make sure it 
really isn't needed. Supposedly it should never return false, thus 
negating the need for its existence. This change adds asserts whenever 
false is returned. If it makes it all the way through promotion 
testing, then I'll delete the ServiceUtil::visible_oop() code and the 
references to it.


I tested by running all jdk and hotspot tier1-3 tests.

thanks,

Chris




Re: RFR(XS): 8198585: add asserts to verify that ServiceUtil::visible_oop is not needed

2018-02-23 Thread serguei.spit...@oracle.com

Hi Chris,

+1

Thanks,
Serguei


On 2/23/18 01:58, Stefan Karlsson wrote:

Looks good.

StefanK

On 2018-02-23 02:16, Chris Plummer wrote:

Hello,

Please review the following:

https://bugs.openjdk.java.net/browse/JDK-8198585
http://cr.openjdk.java.net/~cjplummer/8198585/webrev.00/webrev/

Before removing ServiceUtil::visible_oop(), I want to make sure it 
really isn't needed. Supposedly it should never return false, thus 
negating the need for its existence. This change adds asserts 
whenever false is returned. If it makes it all the way through 
promotion testing, then I'll delete the ServiceUtil::visible_oop() 
code and the references to it.


I tested by running all jdk and hotspot tier1-3 tests.

thanks,

Chris







Re: RFR(XS): 8198585: add asserts to verify that ServiceUtil::visible_oop is not needed

2018-02-23 Thread Stefan Karlsson

Looks good.

StefanK

On 2018-02-23 02:16, Chris Plummer wrote:

Hello,

Please review the following:

https://bugs.openjdk.java.net/browse/JDK-8198585
http://cr.openjdk.java.net/~cjplummer/8198585/webrev.00/webrev/

Before removing ServiceUtil::visible_oop(), I want to make sure it 
really isn't needed. Supposedly it should never return false, thus 
negating the need for its existence. This change adds asserts whenever 
false is returned. If it makes it all the way through promotion 
testing, then I'll delete the ServiceUtil::visible_oop() code and the 
references to it.


I tested by running all jdk and hotspot tier1-3 tests.

thanks,

Chris