Skip to content

Latest commit

 

History

History
94 lines (80 loc) · 1.72 KB

README.md

File metadata and controls

94 lines (80 loc) · 1.72 KB

CUBLAS implementation progress

The following sections list the CUBLAS functions shown on the CUBLAS documentation page:

http://docs.nvidia.com/cuda/cublas/index.html

Level 1 (13 functions)

CUBLAS functions:

  • amax
  • amin
  • asum
  • axpy
  • copy
  • dot, dotc, dotu
  • nrm2
  • rot (not implemented in julia blas.jl)
  • rotg (not implemented in julia blas.jl)
  • rotm (not implemented in julia blas.jl)
  • rotmg (not implemented in julia blas.jl)
  • scal
  • swap (not implemented in julia blas.jl)

Level 2

Key:

  • ge: general
  • gb: general banded
  • sy: symmetric
  • sb: symmetric banded
  • sp: symmetric packed
  • tr: triangular
  • tb: triangular banded
  • tp: triangular packed
  • he: hermitian
  • hb: hermitian banded
  • hp: hermitian packed

CUBLAS functions:

  • gbmv (in julia/blas.jl)
  • gemv (in julia/blas.jl)
  • ger (in julia/blas.jl)
  • sbmv (in julia/blas.jl)
  • spmv
  • spr
  • spr2
  • symv (in julia/blas.jl)
  • syr (in julia/blas.jl)
  • syr2
  • tbmv
  • tbsv
  • tpmv
  • tpsv
  • trmv (in julia/blas.jl)
  • trsv (in julia/blas.jl)
  • hemv (in julia/blas.jl)
  • hbmv
  • hpmv
  • her (in julia/blas.jl)
  • her2
  • hpr
  • hpr2

Level 3

CUBLAS functions:

  • gemm (in julia/blas.jl)
  • gemmBatched
  • symm (in julia/blas.jl)
  • syrk (in julia/blas.jl)
  • syr2k (in julia/blas.jl)
  • syrkx
  • trmm (in julia/blas.jl)
  • trsm (in julia/blas.jl)
  • trsmBatched
  • hemm
  • herk (in julia/blas.jl)
  • her2k (in julia/blas.jl)
  • herkx

BLAS-like extensions

  • geam
  • dgmm
  • getrfBatched
  • getriBatched
  • geqrfBatched
  • gelsBatched
  • tpttr
  • trttp