Re: [Patch] Losing many samples when profiling multiple JVMs

On 07/10/2015 04:31 AM, Andrew Haley wrote:
> On 07/07/15 21:31, William Cohen wrote:
>> Sorry, it has taken me a while to get back to this.
>>
>> I am including Andrew Haley on this email because he works on the JVM
>> and has used oprofile to monitor performance of JVM generated code.
>> He might be able to give some feedback this issue and whether the
>> proposed fix is sufficient.
> 
> It's certainly true that the JVM can split and remap memory areas, so
> you have to be able to cope with that.
> 
> Andrew.
> 

Hi Andrew,

Thanks for the confirmation that the JVM might do that.  I have reworked the patch based on Rei Odaira's comments.  Andrew, this should address the problem you encountered in mid-May with samples being lost due to no permanent mapping on arm64.  Could you give the scratch built Fedora 22 rpms a try:

x86, x86-64, armv7 arm http://koji.fedoraproject.org/koji/taskinfo?taskID=10331313

64-bit arm64:  http://arm.koji.fedoraproject.org/koji/taskinfo?taskID=3063105

Assuming this patch is good, I would like to roll this patch into the upstream oprofile and make a release candiate.

--Will



------------------------------------------------------------------------------
Don't Limit Your Business. Reach for the Cloud.
GigeNET's Cloud Solutions provide you with the tools and support that
you need to offload your IT needs and focus on growing your business.
Configured For All Businesses. Start Your Cloud Today.
https://www.gigenetcloud.com/