RFR 8027575: b113 causing a lot of memory allocation and regression for wls_webapp_atomics

[Valerie Peng]

Xuelei and Tony,

Please find the updated webrev at
http://cr.openjdk.java.net/~valeriep/8027575/webrev.01/

I enhanced existing regression test and modified the fix slightly 
regarding to when to copy out the input for the same-src/dest-buffer 
case, i.e. do the encryption/decryption operation in place if the output 
offset is sufficiently apart from the input range, i.e. [(inOfs - 2 
blocks)....(inOfs + inLen)]

Thanks,
Valerie

On 6/20/2014 3:32 PM, Valerie Peng wrote:
>
> Hmm, maybe. I will add one.
> Thanks for the review.
> Valerie
>
> On 6/18/2014 10:02 PM, Anthony Scarpino wrote:
>> The code change looks fine.  Do you think this is worth a regression 
>> test?  Since the in-place operation wasn't caught it sound like there 
>> might be no in-place tests
>>
>> Tony
>>
>> On 06/18/2014 02:53 PM, Valerie Peng wrote:
>>>
>>> Tony or Vinnie, either of you have time to review?
>>> Thanks,
>>> Valerie
>>>
>>> On 5/15/2014 4:08 PM, Valerie (Yu-Ching) Peng wrote:
>>>>
>>>> Can someone help reviewing this SunJCE provider fix? Essentially,
>>>> CipherCore class is updated to minimize memory allocation and data
>>>> copying. The fix has been verified against the wls_webapp_atomics
>>>> benchmark.
>>>>
>>>> Webrev: http://cr.openjdk.java.net/~valeriep/8027575/webrev.00/
>>>>
>>>> Thanks,
>>>> Valerie
>>