RFR: 8331008: Implement JEP 478: Key Derivation Function API (Preview) [v14]

[Kevin Driver]

On Fri, 23 Aug 2024 21:29:29 GMT, Kevin Driver <kdriver at openjdk.org> wrote:

>> test/jdk/com/sun/crypto/provider/KDF/TestHKDF.java line 187:
>> 
>>> 185:             actualPRK, testData.info,
>>> 186:             testData.outLen);
>>> 187:         actualOKM = kdfExpand.deriveKey("RAW", kdfParameterSpecExpand);
>> 
>> We use "RAW" for key format, but in this file, it's used as key algorithm. Maybe use a standard key algorithm name?
>
> These are "RAW" since the lengths are somewhat arbitrary in the KATs. It's perhaps a bit misleading to create standard keys with strange output lengths. Would you prefer "Generic" over "RAW"?

Addressed in https://github.com/openjdk/jdk/pull/20301/commits/9f050b6a1a4a83d8623e206323071c2c77c90bb2. Please indicate if resolved.

>> test/jdk/javax/crypto/KDF/Threading.java line 63:
>> 
>>> 61:     }
>>> 62: 
>>> 63:     @Test(threadPoolSize = 50, invocationCount = 100, timeOut = 30)
>> 
>> When running on the group server, I observe frequent test failures due to the short `timeOut` value. Perhaps we should increase the value?
>
> I'm happy to increase the value, but this will vary by machine capability. I think I tuned it to a shorter `timeOut` recently, but I can increase the value again.

I double-checked, and the value was actually lengthened recently. Do you have a value in mind?

-------------

PR Review Comment: https://git.openjdk.org/jdk/pull/20301#discussion_r1729544766
PR Review Comment: https://git.openjdk.org/jdk/pull/20301#discussion_r1729545419