- Added changelog
- Added hipblas-bench with support for:
- gemv, trsm, gemm
- Added rocSOLVER as a cpack dependency
- Added hipblasSetAtomicsMode and hipblasGetAtomicsMode
- No longer look for CUDA backend unless --cuda build flag is passed
- Make device memory reallocate on demand
- Added --static build flag to allow for creating a static library
- Added --rocblas-path command line option to choose path to pre-built rocBLAS
- Added sgetriBatched, dgetriBatched, cgetriBatched, and zgetriBatched
- Added TrsmEx, TrsmBatchedEx, and TrsmStridedBatchedEx
- Added hipblasSetVectorAsync and hipblasGetVectorAsync
- Added hipblasSetMatrixAsync and hipblasGetMatrixAsync
- Added Fortran support for getrf, getrs, geqrf and all variants thereof
- Added the following functions. All added functions include batched and strided-batched support with rocBLAS backend:
- stbsv, dtbsv, ctbsv, ztbsv
- ssymm, dsymm, csymm, zsymm
- cgeam, zgeam
- chemm, zhemm
- strtri, dtrtri, ctrtri, ztrtri
- sdgmm, ddgmm, cdgmm, zdgmm
- Added GemmBatchedEx and GemmStridedBatchedEx
- Added Fortran support for BLAS functions
- Added the following functions. All added functions include batched and strided-batched support with rocBLAS backend:
- sgbmv, dgbmv, cgbmv, zgbmv
- chemv, zhemv
- stbmv, dtbmv, ctbmv, ztbmv
- strmv, trmv, ctrmv, ztrmv
- chbmv, zhbmv
- cher, zher
- cher2, zher2
- chpmv, zhpmv
- chpr, zhpr
- chpr2, zhpr2
- ssbmv, dsbmv
- sspmv, dspmv
- ssymv, dsymv, csymv, zsymv
- stpmv, dtpmv, ctpmv, ztpmv
- cgeru, cgerc, zgeru, zgerc
- sspr, dspr, cspr, zspr
- sspr2, dspr2
- csyr, zsyr
- ssyr2, dsyr2, csyr2, zsyr2
- stpsv, dtpsv, ctpsv, ztpsv
- ctrsv, ztrsv
- cherk, zherk
- cherkx, zherkx
- cher2k, zher2k
- ssyrk, dsyrk, csyrk, zsyrk
- ssyr2k, dsyr2k, csyr2k, zsyr2k
- ssyrkx, dsyrkx, csyrkx, zsyrkx
- ctrmm, ztrmm
- ctrsm, ztrsm