Loading... (original) (raw)
- Type:
Enhancement
- Resolution: Fixed
- Priority:
P3
- Fix Version/s: 9
- Affects Version/s: 9
The optimization is achieved through intrinsics for BigInteger squareToLen and mulAdd methods.
We see ~50% improvement for SPECjvm2008 crypto.rsa with this patch.
backported by
JDK-8135484 Use Intel x64 CPU instructions for RSA acceleration
- Resolved
JDK-8150024 Use Intel x64 CPU instructions for RSA acceleration
- Resolved
JDK-8152448 Use Intel x64 CPU instructions for RSA acceleration
- Resolved
JDK-8153423 Use Intel x64 CPU instructions for RSA acceleration
- Resolved
JDK-8153442 Use Intel x64 CPU instructions for RSA acceleration
- Resolved
JDK-8155359 Use Intel x64 CPU instructions for RSA acceleration
- Resolved
JDK-8156260 Use Intel x64 CPU instructions for RSA acceleration
- Resolved
JDK-8161779 Use Intel x64 CPU instructions for RSA acceleration
- Resolved
JDK-8162057 Use Intel x64 CPU instructions for RSA acceleration
- Resolved
relates to
JDK-8069539 RSA Intrinsics
- Resolved
JDK-8151522 Disable 8130150 and 8081778 intrinsics by default
- Resolved
JDK-8133781 TLShandshake causes high cpu usage in BigInteger.montReduce
- Closed
JDK-8154945 Enable 8130150 and 8081778 intrinsics by default
- Resolved
JDK-8149554 high CPU utilization seen in execution of class "java.math.BigInteger"
- Closed
(4 backported by, 5 relates to)