RFR(XS): 7193946: Move warnings associated with UseMemSetInBOT flag

John Cuthbertson john.cuthbertson at oracle.com
Fri Sep 14 16:31:27 UTC 2012

Hi Bengt,

Thanks for the review. I too went back and forth about where to put the 
check. I'm not sure there is a clear division but I'll see which 
location looks a bitter 'fit'.


On 09/13/12 23:39, Bengt Rutisson wrote:
> Hi John,
> Thanks for fixing this!
> It looks good to me. I'm not really sure about how the work in 
> arguments.cpp is supposed to be divided. You added the new check to 
> Arguments::parse() but I guess it would be possible to put the code 
> into Snippet Arguments::check_vm_args_consistency() instead. It looks 
> to me like it might fit better there, but I am fine with leaving it in 
> Arguments::parse() as well. Just glad to get rid of the duplicated code.
> Thanks,
> Bengt
> On 2012-09-13 18:58, John Cuthbertson wrote:
>> Hi Everyone,
>> Can I have a couple of volunteers review the changes for this CR? 
>> They are fairly small. The webrev can be found at: 
>> http://cr.openjdk.java.net/~johnc/7193946/webrev.0/
>> Summary:
>> In the review comments for the fix for 7192128, Bengt suggested to 
>> move the individual warnings from concurrentMarkSweepGeneration.cpp 
>> and g1Collectedheap.cpp and place a single warning in a common piece 
>> of code. These changes address that review comment. The suggested 
>> location (vm_version_sparc.cpp) was unsuitable as the routine which 
>> would be the natural choice is called twice and the warning would be 
>> issued twice. A better place is when we check the other GC flags for 
>> consistency in arguments.cpp.
>> Testing:
>> * command line testing
>> * GC basher and GCOld on sun4v with and without UseMemSetInBOT set
>> Thanks,
>> JohnC

More information about the hotspot-gc-dev mailing list