File buffer_blas.h

Defines

cb_c cb_row
cb_fortran cb_column

Enums

enum cb_order

Values:

cb_row
cb_column
enum cb_side

Values:

cb_left
cb_right
enum cb_transpose

Values:

cb_no_trans
cb_trans
cb_conj_trans
enum cb_uplo

Values:

cb_upper
cb_lower

Functions

int gpublas_setup(gpucontext * ctx)
void gpublas_teardown(gpucontext * ctx)
const char* gpublas_error(gpucontext * ctx)
int gpublas_hdot(size_t N, gpudata * X, size_t offX, size_t incX, gpudata * Y, size_t offY, size_t incY, gpudata * Z, size_t offZ)
int gpublas_sdot(size_t N, gpudata * X, size_t offX, size_t incX, gpudata * Y, size_t offY, size_t incY, gpudata * Z, size_t offZ)
int gpublas_ddot(size_t N, gpudata * X, size_t offX, size_t incX, gpudata * Y, size_t offY, size_t incY, gpudata * Z, size_t offZ)
int gpublas_hgemv(cb_order order, cb_transpose transA, size_t M, size_t N, float alpha, gpudata * A, size_t offA, size_t lda, gpudata * X, size_t offX, int incX, float beta, gpudata * Y, size_t offY, int incY)
int gpublas_sgemv(cb_order order, cb_transpose transA, size_t M, size_t N, float alpha, gpudata * A, size_t offA, size_t lda, gpudata * X, size_t offX, int incX, float beta, gpudata * Y, size_t offY, int incY)
int gpublas_dgemv(cb_order order, cb_transpose transA, size_t M, size_t N, double alpha, gpudata * A, size_t offA, size_t lda, gpudata * X, size_t offX, int incX, double beta, gpudata * Y, size_t offY, int incY)
int gpublas_hgemm(cb_order order, cb_transpose transA, cb_transpose transB, size_t M, size_t N, size_t K, float alpha, gpudata * A, size_t offA, size_t lda, gpudata * B, size_t offB, size_t ldb, float beta, gpudata * C, size_t offC, size_t ldc)
int gpublas_sgemm(cb_order order, cb_transpose transA, cb_transpose transB, size_t M, size_t N, size_t K, float alpha, gpudata * A, size_t offA, size_t lda, gpudata * B, size_t offB, size_t ldb, float beta, gpudata * C, size_t offC, size_t ldc)
int gpublas_dgemm(cb_order order, cb_transpose transA, cb_transpose transB, size_t M, size_t N, size_t K, double alpha, gpudata * A, size_t offA, size_t lda, gpudata * B, size_t offB, size_t ldb, double beta, gpudata * C, size_t offC, size_t ldc)
int gpublas_hger(cb_order order, size_t M, size_t N, float alpha, gpudata * X, size_t offX, int incX, gpudata * Y, size_t offY, int incY, gpudata * A, size_t offA, size_t lda)
int gpublas_sger(cb_order order, size_t M, size_t N, float alpha, gpudata * X, size_t offX, int incX, gpudata * Y, size_t offY, int incY, gpudata * A, size_t offA, size_t lda)
int gpublas_dger(cb_order order, size_t M, size_t N, double alpha, gpudata * X, size_t offX, int incX, gpudata * Y, size_t offY, int incY, gpudata * A, size_t offA, size_t lda)
int gpublas_hgemmBatch(cb_order order, cb_transpose transA, cb_transpose transB, size_t M, size_t N, size_t K, float alpha, gpudata ** A, size_t * offA, size_t lda, gpudata ** B, size_t * offB, size_t ldb, float beta, gpudata ** C, size_t * offC, size_t ldc, size_t batchCount, int flags)
int gpublas_hgemm3D(cb_order order, cb_transpose transA, cb_transpose transB, size_t M, size_t N, size_t K, float alpha, gpudata * A, size_t offA, size_t lda, ssize_t strideA, gpudata * B, size_t offB, size_t ldb, ssize_t strideB, float beta, gpudata * C, size_t offC, size_t ldc, ssize_t strideC, size_t batchCount, int flags)
int gpublas_sgemm3D(cb_order order, cb_transpose transA, cb_transpose transB, size_t M, size_t N, size_t K, float alpha, gpudata * A, size_t offA, size_t lda, ssize_t strideA, gpudata * B, size_t offB, size_t ldb, ssize_t strideB, float beta, gpudata * C, size_t offC, size_t ldc, ssize_t strideC, size_t batchCount, int flags)
int gpublas_dgemm3D(cb_order order, cb_transpose transA, cb_transpose transB, size_t M, size_t N, size_t K, double alpha, gpudata * A, size_t offA, size_t lda, ssize_t strideA, gpudata * B, size_t offB, size_t ldb, ssize_t strideB, double beta, gpudata * C, size_t offC, size_t ldc, ssize_t strideC, size_t batchCount, int flags)
int gpublas_sgemmBatch(cb_order order, cb_transpose transA, cb_transpose transB, size_t M, size_t N, size_t K, float alpha, gpudata ** A, size_t * offA, size_t lda, gpudata ** B, size_t * offB, size_t ldb, float beta, gpudata ** C, size_t * offC, size_t ldc, size_t batchCount, int flags)
int gpublas_dgemmBatch(cb_order order, cb_transpose transA, cb_transpose transB, size_t M, size_t N, size_t K, double alpha, gpudata ** A, size_t * offA, size_t lda, gpudata ** B, size_t * offB, size_t ldb, double beta, gpudata ** C, size_t * offC, size_t ldc, size_t batchCount, int flags)
int gpublas_hgemvBatch(cb_order order, cb_transpose transA, size_t M, size_t N, float alpha, gpudata ** A, size_t * offA, size_t lda, gpudata ** x, size_t * offX, size_t incX, float beta, gpudata ** y, size_t * offY, size_t incY, size_t batchCount, int flags)
int gpublas_sgemvBatch(cb_order order, cb_transpose transA, size_t M, size_t N, float alpha, gpudata ** A, size_t * offA, size_t lda, gpudata ** x, size_t * offX, size_t incX, float beta, gpudata ** y, size_t * offY, size_t incY, size_t batchCount, int flags)
int gpublas_dgemvBatch(cb_order order, cb_transpose transA, size_t M, size_t N, double alpha, gpudata ** A, size_t * offA, size_t lda, gpudata ** x, size_t * offX, size_t incX, double beta, gpudata ** y, size_t * offY, size_t incY, size_t batchCount, int flags)
int gpublas_hgerBatch(cb_order order, size_t M, size_t N, float alpha, gpudata ** x, size_t * offX, size_t incX, gpudata ** y, size_t * offY, size_t incY, gpudata ** A, size_t * offA, size_t lda, size_t batchCount, int flags)
int gpublas_sgerBatch(cb_order order, size_t M, size_t N, float alpha, gpudata ** x, size_t * offX, size_t incX, gpudata ** y, size_t * offY, size_t incY, gpudata ** A, size_t * offA, size_t lda, size_t batchCount, int flags)
int gpublas_dgerBatch(cb_order order, size_t M, size_t N, double alpha, gpudata ** x, size_t * offX, size_t incX, gpudata ** y, size_t * offY, size_t incY, gpudata ** A, size_t * offA, size_t lda, size_t batchCount, int flags)