RFR(XS) : 8038240 : new WB API to get nmethod
Morris Meyer
morris.meyer at oracle.com
Wed Mar 26 22:30:09 UTC 2014
Igor,
I wasn't sure about the distinction. Your original comments are fine.
Thanks for tracking into this nit.
--morris
On 3/25/14, 3:25 PM, Igor Ignatyev wrote:
> Morris,
>
> is it about copyright years?
> if so, according to Mark's email[1] we have to use a comma as a
> separator:
>> The Oracle formats are, respectively:
>>
>> Copyright (c) $X, Oracle and/or its affiliates. All rights reserved.
>>
>> and
>>
>> Copyright (c) $X, $Y, Oracle and/or its affiliates. All rights
>> reserved.
>>
>> The multi-year copyright is different ("$X, $Y" instead of "$X-$Y"), but
>> the meaning is the same.
> [1] http://mail.openjdk.java.net/pipermail/jdk7-dev/2010-May/001321.html
>
> Igor
>
> On 03/25/2014 05:56 PM, Morris Meyer wrote:
>> Igor,
>>
>> I think the comments in WhiteBox.java, whitebox.cpp and whitebox.hpp
>> need to read 2012-2014 to be inclusive of changes from 2013.
>>
>> Other than that this looks fine.
>>
>> --morris
>>
>> On 3/25/14, 5:06 AM, Igor Ignatyev wrote:
>>> Vladimir/Morris,
>>>
>>> Thanks for review, I've added a nmethod structure w/ code's array and
>>> compilation level. Since to use NMethod class directly in WhiteBox
>>> class, class should be added to bootclasspath, I decided to return the
>>> structure as an object array instead.
>>>
>>> I've replaced getMethodCompilationLevel in CompilerWhiteBoxTest and
>>> planning to remove this method completely in a separate CR.
>>>
>>> Also I've removed CLEAR_PENDING_EXCEPTION from CHECK_JNI_EXCEPTION
>>> macros
>>>
>>> http://cr.openjdk.java.net/~iignatyev/8038240/webrev.01/
>>>
>>> Igor
>>>
>>> On 03/25/2014 01:11 AM, Vladimir Kozlov wrote:
>>>> Igor,
>>>>
>>>> Can we pass a structure instead of code's array? We can add an
>>>> additional information about nmethod into such structure later if
>>>> needed. For now it would have just one pointer to jbyteArray array.
>>>>
>>>> Thanks,
>>>> Vladimir
>>> On 03/25/2014 02:23 AM, Morris Meyer wrote:
>>>> Looks good. (Not a reviewer).
>>>>
>>>> --morris
>>>>
>>>>
>>>> --mm
>>>>
>>>>
>>>> On 3/24/14 10:24 AM, Igor Ignatyev wrote:
>>>>> Hi all,
>>>>>
>>>>> Please review the patch which introduces a new WhiteBox method --
>>>>> byte[]
>>>>> getNMethod(). We need this method to write reliable tests for
>>>>> JDK-8031321 : 'Support Intel bit manipulation instructions'.
>>>>>
>>>>> webrev: http://cr.openjdk.java.net/~iignatyev/8038240/webrev.00/
>>>>> jbs: https://bugs.openjdk.java.net/browse/JDK-8038240
>>>>> testing: jprt
>>
More information about the hotspot-compiler-dev
mailing list