Regression bug in PKCS12 key wrapping

[Anders Rundgren]

Apparently there was a bug in BC.  The GitHub issue has been updated.
Case dismissed :)

thanx,
Anders

On 2021-11-25 4:10, Wei-Jun Wang wrote:
> 
>> On Nov 24, 2021, at 11:14 AM, Anders Rundgren <anders.rundgren.net at gmail.com> wrote:
>>
>> Hi List,
>>
>> Although this bug is for BC, I believe the problem is rather in JDK 17:
>> https://github.com/bcgit/bc-java/issues/823#issuecomment-977919380
> 
> Can you point out which part in JDK 17 the problem is?
> 
> Please note that in JDK 17 the default encryption algorithm for PKCS12 is upgraded to PBEWithHmacSHA256AndAES_256. If BC's decryption is not an exact reverse of SUN's encryption, then definitely the output will be garbage and you will not see the correct padding bytes.
> 
> Thanks,
> Max
> 
>>
>> Thanx,
>> Anders
>