RFR (XS): 8021301: better event messages

Rickard Bäckman rickard.backman at oracle.com
Fri Aug 2 09:07:19 PDT 2013


Looks good.

/R

On Aug 2, 2013, at 4:54 PM, Albert Noll wrote:

> Hi,
> 
> do we need a second review for this?
> 
> Best,
> Albert
> 
> On 31.07.2013 15:43, Vladimir Kozlov wrote:
>> Good.
>> 
>> thanks,
>> Vladimir
>> 
>> On 7/31/13 2:23 PM, Albert Noll wrote:
>>> Hi,
>>> 
>>> thanks again Vladimir. Here is the new version that includes your
>>> proposed changes.
>>> 
>>> http://cr.openjdk.java.net/~anoll/8021301/webrev.02/
>>> <http://cr.openjdk.java.net/%7Eanoll/8021301/webrev.02/>
>>> 
>>> Best,
>>> Albert
>>> 
>>> On 30.07.2013 17:48, Vladimir Kozlov wrote:
>>>> Looks good. Could you modify both places to enclose the message into
>>>> <> and remove space before ):
>>>> 
>>>> current:
>>>> 
>>>> Exception <a 'java/lang/NoSuchMethodError'>: Method
>>>> sun.misc.Unsafe.prefetchRead(Ljava/lang/Object;J)V name or signature
>>>> does not match (0x00000006ab00cbf8 )
>>>> 
>>>> I think would be better:
>>>> 
>>>> Exception <a 'java/lang/NoSuchMethodError': Method
>>>> sun.misc.Unsafe.prefetchRead(Ljava/lang/Object;J)V name or signature
>>>> does not match> (0x00000006ab00cbf8)
>>>> 
>>>> "Exception <%s%s%s> (" INTPTR_FORMAT ")
>>>> 
>>>> Thanks,
>>>> Vladimir
>>>> 
>>>> On 7/30/13 5:16 PM, Albert Noll wrote:
>>>>> Hi Vladimir,
>>>>> 
>>>>> thanks for looking at the patch. I changed the output format. Do you
>>>>> think it is fine now?
>>>>> Here is the new webrev:
>>>>> http://cr.openjdk.java.net/~anoll/8021301/webrev.01/
>>>>> <http://cr.openjdk.java.net/%7Eanoll/8021301/webrev.01/>
>>>>> 
>>>>> Best,
>>>>> Albert
>>>>> 
>>>>> 
>>>>> On 30.07.2013 10:24, Vladimir Kozlov wrote:
>>>>>> Could you use format similar to one in print_cr()?
>>>>>> 
>>>>>> Vladimir
>>>>>> 
>>>>>> On 7/30/13 9:28 AM, Albert Noll wrote:
>>>>>>> Hi all,
>>>>>>> 
>>>>>>> please review this small patch.
>>>>>>> 
>>>>>>> jbs: https://jbs.oracle.com/bugs/browse/JDK-8021301
>>>>>>> webrev: http://cr.openjdk.java.net/~anoll/8021301/webrev.00/
>>>>>>> <http://cr.openjdk.java.net/%7Eanoll/8021301/webrev.00/>
>>>>>>> 
>>>>>>> Problem:
>>>>>>> Having event message printing hex-address is not very helpful:
>>>>>>> 
>>>>>>> Event: 57.947 Thread 0x00007f1f3e955800 Threw 0x00000008060c6e90 at
>>>>>>> /HUDSON/workspace/2-build-linux-amd64/jdk8/4338/hotspot/src/share/vm/prims/jvm.cpp:1203 
>>>>>>> 
>>>>>>> 
>>>>>>> 
>>>>>>> Especially when the address is some temporary symbol address you
>>>>>>> can't not use in a debugger, like for class loading:
>>>>>>> 
>>>>>>> EventMark m("loading class " INTPTR_FORMAT, (address)h_name);
>>>>>>> 
>>>>>>> Solution:
>>>>>>> Make output better readable.
>>>>>>> 
>>>>>>> Best,
>>>>>>> Albert
>>>>> 
>>> 
> 



More information about the hotspot-compiler-dev mailing list