RFR(XS): JDK-8010941: MinJumpTableSize is set to 18, investigate if that's still optimal

Christian Thalinger christian.thalinger at oracle.com
Tue Sep 10 11:44:33 PDT 2013


Are Zero and Shark builds broken after this change?

-- Chris

On Sep 10, 2013, at 1:49 AM, Niclas Adlertz <niclas.adlertz at oracle.com> wrote:

> On 2013-09-09 11:33, Aleksey Shipilev wrote:
>> On 09/09/2013 01:16 PM, Niclas Adlertz wrote:
>>> On 2013-09-06 14:02, Aleksey Shipilev wrote:
>>>> On 09/06/2013 03:54 PM, Niclas Adlertz wrote:
>>>>>> It is customary for us (perf guys) to create multiple
>>>>>> @GMB methods with different problems sizes, and run them in single JMH
>>>>>> session.
>>>>> That would have been smarter, yes. I'll do that next time. Thanks.
>>>>> And thank you for helping me with JMH.
>>>> Sure, any time.
>>>> 
>>>>>> Yes, "10" seems the turn-point for X64; for SPARC, I'd set it to "5".
>>>>> Ok, I'll have one value for each platform instead.
>>>>> 
>>>>> WEBREV: http://cr.openjdk.java.net/~adlertz/JDK-8010941/webrev02/
>>>> Good!
>>>> 
>>>> I worry about the 32-bit x86 though. It seems the register pressure with
>>>> jump tables is higher? If so, the turn-point for x86 can be even larger.
>>>> Can you double-check the 32-bit x86 VM?
>>>> 
>>>> -Aleksey.
>>>> 
>>> Hi Aleksey.
>>> 
>>> I don't think the register pressure is higher using a jump table since
>>> it's just a jump to an address with an offset.
>>> In fact, tests seem to show the opposite; with a lookup table
>>> (implemented as a binary search) fewer registers seem to have a negative
>>> impact.
>>> When we have 6 cases, it's actually faster to use a jump table on 32-bit
>>> x86 instead of a lookup table (compared to 10 on x64).
>> 
>> Thumbs up. Let's go with your change then.
>> 
>> -Aleksey.
>> 
> 
> Thank you Aleksey.
> 



More information about the hotspot-compiler-dev mailing list