AOCL-LAPACK 5.0

Highlights of AOCL-LAPACK 5.0 release

• Improved performance of the following APIs through AVX2 and AVX512 SIMD instructions:
  • Double Precision SVD (DGESVD)
  • LU Factorization/Solver routines for general matrices (DGETRF, ZGETRF, DGETRS, and DGESV)
  • Matrix inverse routine DGETRI for small sizes
  • Least Square solver DGELS for small sizes
  • Double Precision Auxiliary routine and DLARFG
• Improved performance of the following APIs using local AOCL-BLAS optimized kernels:
  • LU Factorization/Solver routines for band storage matrices (DGBTRF and DGBTRS)
• Option to set specific ISA code path at runtime through the AOCL_ENABLE_INSTRUCTIONS environment variable
• Sphinx-based AOCL-LAPACK API documentation
• pkgconfig support on Linux with CMake builds
• LAPACK API modifications:
  • Updated AOCL-LAPACK APIs return type to match with corresponding netlib subroutine prototypes
  • Removed xerbla and lsame definition from AOCL-LAPACK. Applications must invoke lsame from the BLAS library
• Test suite framework enhancements:
  • Improved accuracy tests including testing with different input generation mechanisms
  • Addition of extreme values, negative, and corner test cases
  • Addition of cases to test numerical stability
  • Support for LAPACKE interface test

AOCL-LAPACK 4.2

AOCL-LAPACK 4.2

Highlights of improvements on AMD “Zen” core based processors.

• Improved performance of the following APIs:
  • Double Precision SVD (DGESVD)
  • Factorization routines DGETRF and ZGETRF
  • Solver routines DGETRS and DGESV
• Option to link with AOCL-BLAS during build to enable invoking AOCL-BLAS internal APIs
• Applications using AOCL-LAPACK need to additionally link with AOCL-Utils library
• CMake improvements:
  • Ease of running tests using Ctest
  • ISA specific flags configurable during build
  • Code Coverage build option
• OpenMP parallelism enabled in the following APIs to match LAPACK 3.11 implementation:
  • {c,z}hetrd_hb2st, {s,d}sytrd_sb2st, iparam2stage
• Minor bug fixes and warnings addressed
• Test suite framework improvements and more test cases

AOCL-LAPACK 4.1

AOCL-LAPACK 4.1

Highlights of improvements on AMD “Zen” core based processors.

• Upgraded to Netlib LAPACK 3.11.0 specification that includes 14 new APIs, bug fixes, and improvements to the existing APIs
• Improved performance of the following APIs:
  - Double Precision SVD (DGESVD) for small matrices
  - Factorization routines (DGEQP3, SGETRF, ZGETRF, and DGETRF)
  - Double Precision solver API DGESV and DGESVD routines
• Framework changes to support the selection of optimized routines based on target CPU instruction set
• CMake build system for Linux and Windows
• Build system cleanup and minor bug fixes

AOCL-libFLAME 4.0

AOCL-libFLAME Version 4.0

Highlights of improvements on AMD EPYCTM processor family CPUs

• Upgrade to LAPACK 3.10.1 specification that includes several bug fixes from Netlib LAPACK
• Improved performance of the following APIs:
  • Eigen Value routine (ZGGEV)
  • SVD routines (DGESDD, CGESDD, and ZGESDD)
• Logging feature supports timing for real double precision libFLAME APIs
• AOCL-Progress feature that provides progress update on API computations running for a long time is extended for more APIs: {S/C/Z}GETRF, {S/D}POTRF,{S/D}GEQRF, {S/C/D/Z}GBTRF

AOCL-libFLAME 3.2

AOCL-libFLAME Version 3.2

Highlights of improvements on AMD EPYC^TM processor family CPUs

• Improved performance of the following for AMD “Zen” architecture
  • Eigen Value routines (DSYEVD and DSTEQR)
  • SVD routines (DGESVD)
• Feature AOCL_FLA_PROGRESS that provides progress update on API computations running for a long time; this support is available for double precision LU Factorization
• Improvements in libFLAME build system with new config flag for enabling optimization specific to AMD CPUs
• Improved test coverage

AMD Optimized libFLAME Version 3.1

AMD Optimized libFLAME Version 3.1

Highlights of improvements on AMD EPYC^TM processor family CPUs

• New APIs to compute partial LDLT factorization of a symmetric matrix using packed storage: ?spffrt2 and spffrtx
• New APIs to perform complete or incomplete LU factorization without pivoting of a general matrix: ?getrfnp and ?getrfnpi
• Test suite now supports LAPACK API tests for LU, Cholesky and QR operations
• Several bug fixes including handling denormal numbers in SVD functions
• New API to get version number of the library, FLA_Get_AOCL_Version()
• Library function tracing and input logging support added

AMD Optimized libFLAME Version 3.0.1

AMD Optimized libFLAME Version 3.0.1

Highlights of improvements on AMD EPYCTM processor family CPUs

• Improved performance of LU, QR and Cholesky Factorization
• Improved performance of routine that computes partial LDL^T factorization of a symmetric matrix using packed storage: spffrt2 and spffrtx
• Improved performance of routine that computes complete/incomplete LU factorization without pivoting of a general matrix: getrfnp and getrfnpi
• Library function tracing and input logging support added

AMD Optimized libFLAME Version 3.0

AMD Optimized libFLAME Version 3.0

Highlights of improvements on AMD EPYC^TM processor family CPUs

• New APIs to compute partial LDL^T factorization of a symmetric matrix using packed storage: ?spffrt2 and ?spffrtx
• New APIs to perform complete or incomplete LU factorization without pivoting of a general matrix: ?getrfnp and ?getrfnpi
• Test suite now supports LAPACK API tests for LU, Cholesky and QR operations
• Several bug fixes including handling denormal numbers in SVD functions
• New API to get version number of the library, FLA_Get_AOCL_Version()
• Library function tracing and input logging support added

AMD Optimized libFLAME Version 2.2

AMD Optimized libFLAME Version 2.2

Highlights of improvements on AMD EPYC^TM processor family CPUs

• libFLAME is now compatible with LAPACK 3.9.0 specification.
• More coverage of Netlib tests suite compatibility

AMD Optimized libFLAME Version 2.1

AMD optimized libFLAME version 2.1

Highlights of improvements on AMD EPYC^TM processor family CPUs

• Support for C++ Template APIs for all LAPACK functions
• Includes LAPACKE source in the libFLAME source directory