diff options
| author | Cedric Nugteren <web@cedricnugteren.nl> | 2015-09-18 10:23:16 +0200 |
|---|---|---|
| committer | Cedric Nugteren <web@cedricnugteren.nl> | 2015-09-18 10:23:16 +0200 |
| commit | 42db8ea968d9d2972446aa4fd73515a3d7aa093e (patch) | |
| tree | d74289f024341071ec11324a9827e2e33428f895 /src | |
| parent | 1c242100266e46cf97227c5506a1c32e51afbd00 (diff) | |
| parent | 4796c9bcbd84a9e8be1e2864ba47e0d6bf3e6632 (diff) | |
Merge pull request #26 from CNugteren/routine_definitions
Generated API interface and implementations
Diffstat (limited to 'src')
| -rw-r--r-- | src/clblast.cc | 1155 | ||||
| -rw-r--r-- | src/clblast_c.cc | 1058 |
2 files changed, 1959 insertions, 254 deletions
diff --git a/src/clblast.cc b/src/clblast.cc index 3303085e..a0dd8c70 100644 --- a/src/clblast.cc +++ b/src/clblast.cc @@ -42,11 +42,12 @@ #include "internal/routines/level3/xtrmm.h" namespace clblast { + // ================================================================================================= // BLAS level-1 (vector-vector) routines // ================================================================================================= -// SWAP +// Swap two vectors: SSWAP/DSWAP/CSWAP/ZSWAP template <typename T> StatusCode Swap(const size_t n, cl_mem x_buffer, const size_t x_offset, const size_t x_inc, @@ -64,21 +65,21 @@ StatusCode Swap(const size_t n, template StatusCode Swap<float>(const size_t, cl_mem, const size_t, const size_t, cl_mem, const size_t, const size_t, - cl_command_queue* queue, cl_event* event); + cl_command_queue*, cl_event*); template StatusCode Swap<double>(const size_t, cl_mem, const size_t, const size_t, cl_mem, const size_t, const size_t, - cl_command_queue* queue, cl_event* event); + cl_command_queue*, cl_event*); template StatusCode Swap<float2>(const size_t, cl_mem, const size_t, const size_t, cl_mem, const size_t, const size_t, - cl_command_queue* queue, cl_event* event); + cl_command_queue*, cl_event*); template StatusCode Swap<double2>(const size_t, cl_mem, const size_t, const size_t, cl_mem, const size_t, const size_t, - cl_command_queue* queue, cl_event* event); + cl_command_queue*, cl_event*); -// SCAL +// Vector scaling: SSCAL/DSCAL/CSCAL/ZSCAL template <typename T> StatusCode Scal(const size_t n, const T alpha, @@ -96,21 +97,21 @@ StatusCode Scal(const size_t n, template StatusCode Scal<float>(const size_t, const float, cl_mem, const size_t, const size_t, - cl_command_queue* queue, cl_event* event); + cl_command_queue*, cl_event*); template StatusCode Scal<double>(const size_t, const double, cl_mem, const size_t, const size_t, - cl_command_queue* queue, cl_event* event); + cl_command_queue*, cl_event*); template StatusCode Scal<float2>(const size_t, const float2, cl_mem, const size_t, const size_t, - cl_command_queue* queue, cl_event* event); + cl_command_queue*, cl_event*); template StatusCode Scal<double2>(const size_t, const double2, cl_mem, const size_t, const size_t, - cl_command_queue* queue, cl_event* event); + cl_command_queue*, cl_event*); -// COPY +// Vector copy: SCOPY/DCOPY/CCOPY/ZCOPY template <typename T> StatusCode Copy(const size_t n, const cl_mem x_buffer, const size_t x_offset, const size_t x_inc, @@ -128,21 +129,21 @@ StatusCode Copy(const size_t n, template StatusCode Copy<float>(const size_t, const cl_mem, const size_t, const size_t, cl_mem, const size_t, const size_t, - cl_command_queue* queue, cl_event* event); + cl_command_queue*, cl_event*); template StatusCode Copy<double>(const size_t, const cl_mem, const size_t, const size_t, cl_mem, const size_t, const size_t, - cl_command_queue* queue, cl_event* event); + cl_command_queue*, cl_event*); template StatusCode Copy<float2>(const size_t, const cl_mem, const size_t, const size_t, cl_mem, const size_t, const size_t, - cl_command_queue* queue, cl_event* event); + cl_command_queue*, cl_event*); template StatusCode Copy<double2>(const size_t, const cl_mem, const size_t, const size_t, cl_mem, const size_t, const size_t, - cl_command_queue* queue, cl_event* event); + cl_command_queue*, cl_event*); -// AXPY +// Vector-times-constant plus vector: SAXPY/DAXPY/CAXPY/ZAXPY template <typename T> StatusCode Axpy(const size_t n, const T alpha, @@ -163,24 +164,24 @@ template StatusCode Axpy<float>(const size_t, const float, const cl_mem, const size_t, const size_t, cl_mem, const size_t, const size_t, - cl_command_queue* queue, cl_event* event); + cl_command_queue*, cl_event*); template StatusCode Axpy<double>(const size_t, const double, const cl_mem, const size_t, const size_t, cl_mem, const size_t, const size_t, - cl_command_queue* queue, cl_event* event); + cl_command_queue*, cl_event*); template StatusCode Axpy<float2>(const size_t, const float2, const cl_mem, const size_t, const size_t, cl_mem, const size_t, const size_t, - cl_command_queue* queue, cl_event* event); + cl_command_queue*, cl_event*); template StatusCode Axpy<double2>(const size_t, const double2, const cl_mem, const size_t, const size_t, cl_mem, const size_t, const size_t, - cl_command_queue* queue, cl_event* event); + cl_command_queue*, cl_event*); -// DOT +// Dot product of two vectors: SDOT/DDOT template <typename T> StatusCode Dot(const size_t n, cl_mem dot_buffer, const size_t dot_offset, @@ -201,14 +202,14 @@ template StatusCode Dot<float>(const size_t, cl_mem, const size_t, const cl_mem, const size_t, const size_t, const cl_mem, const size_t, const size_t, - cl_command_queue* queue, cl_event* event); + cl_command_queue*, cl_event*); template StatusCode Dot<double>(const size_t, cl_mem, const size_t, const cl_mem, const size_t, const size_t, const cl_mem, const size_t, const size_t, - cl_command_queue* queue, cl_event* event); + cl_command_queue*, cl_event*); -// DOTU +// Dot product of two complex vectors: CDOTU/ZDOTU template <typename T> StatusCode Dotu(const size_t n, cl_mem dot_buffer, const size_t dot_offset, @@ -229,14 +230,14 @@ template StatusCode Dotu<float2>(const size_t, cl_mem, const size_t, const cl_mem, const size_t, const size_t, const cl_mem, const size_t, const size_t, - cl_command_queue* queue, cl_event* event); + cl_command_queue*, cl_event*); template StatusCode Dotu<double2>(const size_t, cl_mem, const size_t, const cl_mem, const size_t, const size_t, const cl_mem, const size_t, const size_t, - cl_command_queue* queue, cl_event* event); + cl_command_queue*, cl_event*); -// DOTC +// Dot product of two complex vectors, one conjugated: CDOTC/ZDOTC template <typename T> StatusCode Dotc(const size_t n, cl_mem dot_buffer, const size_t dot_offset, @@ -257,446 +258,1109 @@ template StatusCode Dotc<float2>(const size_t, cl_mem, const size_t, const cl_mem, const size_t, const size_t, const cl_mem, const size_t, const size_t, - cl_command_queue* queue, cl_event* event); + cl_command_queue*, cl_event*); template StatusCode Dotc<double2>(const size_t, cl_mem, const size_t, const cl_mem, const size_t, const size_t, const cl_mem, const size_t, const size_t, - cl_command_queue* queue, cl_event* event); + cl_command_queue*, cl_event*); // ================================================================================================= // BLAS level-2 (matrix-vector) routines // ================================================================================================= -// GEMV +// General matrix-vector multiplication: SGEMV/DGEMV/CGEMV/ZGEMV template <typename T> StatusCode Gemv(const Layout layout, const Transpose a_transpose, - const size_t m, const size_t n, const T alpha, + const size_t m, const size_t n, + const T alpha, const cl_mem a_buffer, const size_t a_offset, const size_t a_ld, - const cl_mem x_buffer, const size_t x_offset, const size_t x_inc, const T beta, + const cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + const T beta, cl_mem y_buffer, const size_t y_offset, const size_t y_inc, cl_command_queue* queue, cl_event* event) { - auto queue_cpp = Queue(*queue); auto event_cpp = Event(*event); auto routine = Xgemv<T>(queue_cpp, event_cpp); - - // Compiles the routine's device kernels auto status = routine.SetUp(); if (status != StatusCode::kSuccess) { return status; } - - // Runs the routine - return routine.DoGemv(layout, a_transpose, m, n, alpha, + return routine.DoGemv(layout, a_transpose, + m, n, + alpha, Buffer<T>(a_buffer), a_offset, a_ld, - Buffer<T>(x_buffer), x_offset, x_inc, beta, + Buffer<T>(x_buffer), x_offset, x_inc, + beta, Buffer<T>(y_buffer), y_offset, y_inc); } template StatusCode Gemv<float>(const Layout, const Transpose, - const size_t, const size_t, const float, + const size_t, const size_t, + const float, const cl_mem, const size_t, const size_t, - const cl_mem, const size_t, const size_t, const float, + const cl_mem, const size_t, const size_t, + const float, cl_mem, const size_t, const size_t, cl_command_queue*, cl_event*); template StatusCode Gemv<double>(const Layout, const Transpose, - const size_t, const size_t, const double, + const size_t, const size_t, + const double, + const cl_mem, const size_t, const size_t, const cl_mem, const size_t, const size_t, - const cl_mem, const size_t, const size_t, const double, + const double, cl_mem, const size_t, const size_t, cl_command_queue*, cl_event*); template StatusCode Gemv<float2>(const Layout, const Transpose, - const size_t, const size_t, const float2, + const size_t, const size_t, + const float2, const cl_mem, const size_t, const size_t, - const cl_mem, const size_t, const size_t, const float2, + const cl_mem, const size_t, const size_t, + const float2, cl_mem, const size_t, const size_t, cl_command_queue*, cl_event*); template StatusCode Gemv<double2>(const Layout, const Transpose, - const size_t, const size_t, const double2, + const size_t, const size_t, + const double2, + const cl_mem, const size_t, const size_t, const cl_mem, const size_t, const size_t, - const cl_mem, const size_t, const size_t, const double2, + const double2, cl_mem, const size_t, const size_t, cl_command_queue*, cl_event*); -// ================================================================================================= +// General banded matrix-vector multiplication: SGBMV/DGBMV/CGBMV/ZGBMV +template <typename T> +StatusCode Gbmv(const Layout, const Transpose, + const size_t, const size_t, const size_t, const size_t, + const T, + const cl_mem, const size_t, const size_t, + const cl_mem, const size_t, const size_t, + const T, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*) { + return StatusCode::kNotImplemented; +} +template StatusCode Gbmv<float>(const Layout, const Transpose, + const size_t, const size_t, const size_t, const size_t, + const float, + const cl_mem, const size_t, const size_t, + const cl_mem, const size_t, const size_t, + const float, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*); +template StatusCode Gbmv<double>(const Layout, const Transpose, + const size_t, const size_t, const size_t, const size_t, + const double, + const cl_mem, const size_t, const size_t, + const cl_mem, const size_t, const size_t, + const double, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*); +template StatusCode Gbmv<float2>(const Layout, const Transpose, + const size_t, const size_t, const size_t, const size_t, + const float2, + const cl_mem, const size_t, const size_t, + const cl_mem, const size_t, const size_t, + const float2, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*); +template StatusCode Gbmv<double2>(const Layout, const Transpose, + const size_t, const size_t, const size_t, const size_t, + const double2, + const cl_mem, const size_t, const size_t, + const cl_mem, const size_t, const size_t, + const double2, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*); -// HEMV +// Hermitian matrix-vector multiplication: CHEMV/ZHEMV template <typename T> StatusCode Hemv(const Layout layout, const Triangle triangle, - const size_t n, const T alpha, + const size_t n, + const T alpha, const cl_mem a_buffer, const size_t a_offset, const size_t a_ld, - const cl_mem x_buffer, const size_t x_offset, const size_t x_inc, const T beta, + const cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + const T beta, cl_mem y_buffer, const size_t y_offset, const size_t y_inc, cl_command_queue* queue, cl_event* event) { - auto queue_cpp = Queue(*queue); auto event_cpp = Event(*event); auto routine = Xhemv<T>(queue_cpp, event_cpp); - - // Compiles the routine's device kernels auto status = routine.SetUp(); if (status != StatusCode::kSuccess) { return status; } - - // Runs the routine - return routine.DoHemv(layout, triangle, n, alpha, + return routine.DoHemv(layout, triangle, + n, + alpha, Buffer<T>(a_buffer), a_offset, a_ld, - Buffer<T>(x_buffer), x_offset, x_inc, beta, + Buffer<T>(x_buffer), x_offset, x_inc, + beta, Buffer<T>(y_buffer), y_offset, y_inc); } template StatusCode Hemv<float2>(const Layout, const Triangle, - const size_t, const float2, + const size_t, + const float2, + const cl_mem, const size_t, const size_t, const cl_mem, const size_t, const size_t, - const cl_mem, const size_t, const size_t, const float2, + const float2, cl_mem, const size_t, const size_t, cl_command_queue*, cl_event*); template StatusCode Hemv<double2>(const Layout, const Triangle, - const size_t, const double2, + const size_t, + const double2, const cl_mem, const size_t, const size_t, - const cl_mem, const size_t, const size_t, const double2, + const cl_mem, const size_t, const size_t, + const double2, cl_mem, const size_t, const size_t, cl_command_queue*, cl_event*); -// ================================================================================================= +// Hermitian banded matrix-vector multiplication: CHBMV/ZHBMV +template <typename T> +StatusCode Hbmv(const Layout, const Triangle, + const size_t, const size_t, + const T, + const cl_mem, const size_t, const size_t, + const cl_mem, const size_t, const size_t, + const T, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*) { + return StatusCode::kNotImplemented; +} +template StatusCode Hbmv<float2>(const Layout, const Triangle, + const size_t, const size_t, + const float2, + const cl_mem, const size_t, const size_t, + const cl_mem, const size_t, const size_t, + const float2, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*); +template StatusCode Hbmv<double2>(const Layout, const Triangle, + const size_t, const size_t, + const double2, + const cl_mem, const size_t, const size_t, + const cl_mem, const size_t, const size_t, + const double2, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*); -// SYMV +// Hermitian packed matrix-vector multiplication: CHPMV/ZHPMV +template <typename T> +StatusCode Hpmv(const Layout, const Triangle, + const size_t, + const T, + const cl_mem, const size_t, + const cl_mem, const size_t, const size_t, + const T, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*) { + return StatusCode::kNotImplemented; +} +template StatusCode Hpmv<float2>(const Layout, const Triangle, + const size_t, + const float2, + const cl_mem, const size_t, + const cl_mem, const size_t, const size_t, + const float2, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*); +template StatusCode Hpmv<double2>(const Layout, const Triangle, + const size_t, + const double2, + const cl_mem, const size_t, + const cl_mem, const size_t, const size_t, + const double2, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*); + +// Symmetric matrix-vector multiplication: SSYMV/DSYMV template <typename T> StatusCode Symv(const Layout layout, const Triangle triangle, - const size_t n, const T alpha, + const size_t n, + const T alpha, const cl_mem a_buffer, const size_t a_offset, const size_t a_ld, - const cl_mem x_buffer, const size_t x_offset, const size_t x_inc, const T beta, + const cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + const T beta, cl_mem y_buffer, const size_t y_offset, const size_t y_inc, cl_command_queue* queue, cl_event* event) { - auto queue_cpp = Queue(*queue); auto event_cpp = Event(*event); auto routine = Xsymv<T>(queue_cpp, event_cpp); - - // Compiles the routine's device kernels auto status = routine.SetUp(); if (status != StatusCode::kSuccess) { return status; } - - // Runs the routine - return routine.DoSymv(layout, triangle, n, alpha, + return routine.DoSymv(layout, triangle, + n, + alpha, Buffer<T>(a_buffer), a_offset, a_ld, - Buffer<T>(x_buffer), x_offset, x_inc, beta, + Buffer<T>(x_buffer), x_offset, x_inc, + beta, Buffer<T>(y_buffer), y_offset, y_inc); } template StatusCode Symv<float>(const Layout, const Triangle, - const size_t, const float, + const size_t, + const float, + const cl_mem, const size_t, const size_t, const cl_mem, const size_t, const size_t, - const cl_mem, const size_t, const size_t, const float, + const float, cl_mem, const size_t, const size_t, cl_command_queue*, cl_event*); template StatusCode Symv<double>(const Layout, const Triangle, - const size_t, const double, + const size_t, + const double, + const cl_mem, const size_t, const size_t, + const cl_mem, const size_t, const size_t, + const double, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*); + +// Symmetric banded matrix-vector multiplication: SSBMV/DSBMV +template <typename T> +StatusCode Sbmv(const Layout, const Triangle, + const size_t, const size_t, + const T, + const cl_mem, const size_t, const size_t, + const cl_mem, const size_t, const size_t, + const T, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*) { + return StatusCode::kNotImplemented; +} +template StatusCode Sbmv<float>(const Layout, const Triangle, + const size_t, const size_t, + const float, + const cl_mem, const size_t, const size_t, + const cl_mem, const size_t, const size_t, + const float, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*); +template StatusCode Sbmv<double>(const Layout, const Triangle, + const size_t, const size_t, + const double, + const cl_mem, const size_t, const size_t, + const cl_mem, const size_t, const size_t, + const double, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*); + +// Symmetric packed matrix-vector multiplication: SSPMV/DSPMV +template <typename T> +StatusCode Spmv(const Layout, const Triangle, + const size_t, + const T, + const cl_mem, const size_t, + const cl_mem, const size_t, const size_t, + const T, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*) { + return StatusCode::kNotImplemented; +} +template StatusCode Spmv<float>(const Layout, const Triangle, + const size_t, + const float, + const cl_mem, const size_t, + const cl_mem, const size_t, const size_t, + const float, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*); +template StatusCode Spmv<double>(const Layout, const Triangle, + const size_t, + const double, + const cl_mem, const size_t, + const cl_mem, const size_t, const size_t, + const double, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*); + +// Triangular matrix-vector multiplication: STRMV/DTRMV/CTRMV/ZTRMV +template <typename T> +StatusCode Trmv(const Layout, const Triangle, const Transpose, const Diagonal, + const size_t, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*) { + return StatusCode::kNotImplemented; +} +template StatusCode Trmv<float>(const Layout, const Triangle, const Transpose, const Diagonal, + const size_t, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*); +template StatusCode Trmv<double>(const Layout, const Triangle, const Transpose, const Diagonal, + const size_t, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*); +template StatusCode Trmv<float2>(const Layout, const Triangle, const Transpose, const Diagonal, + const size_t, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*); +template StatusCode Trmv<double2>(const Layout, const Triangle, const Transpose, const Diagonal, + const size_t, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*); + +// Triangular banded matrix-vector multiplication: STBMV/DTBMV/CTBMV/ZTBMV +template <typename T> +StatusCode Tbmv(const Layout, const Triangle, const Transpose, const Diagonal, + const size_t, const size_t, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*) { + return StatusCode::kNotImplemented; +} +template StatusCode Tbmv<float>(const Layout, const Triangle, const Transpose, const Diagonal, + const size_t, const size_t, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*); +template StatusCode Tbmv<double>(const Layout, const Triangle, const Transpose, const Diagonal, + const size_t, const size_t, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*); +template StatusCode Tbmv<float2>(const Layout, const Triangle, const Transpose, const Diagonal, + const size_t, const size_t, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*); +template StatusCode Tbmv<double2>(const Layout, const Triangle, const Transpose, const Diagonal, + const size_t, const size_t, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*); + +// Triangular packed matrix-vector multiplication: STPMV/DTPMV/CTPMV/ZTPMV +template <typename T> +StatusCode Tpmv(const Layout, const Triangle, const Transpose, const Diagonal, + const size_t, + const cl_mem, const size_t, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*) { + return StatusCode::kNotImplemented; +} +template StatusCode Tpmv<float>(const Layout, const Triangle, const Transpose, const Diagonal, + const size_t, + const cl_mem, const size_t, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*); +template StatusCode Tpmv<double>(const Layout, const Triangle, const Transpose, const Diagonal, + const size_t, + const cl_mem, const size_t, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*); +template StatusCode Tpmv<float2>(const Layout, const Triangle, const Transpose, const Diagonal, + const size_t, + const cl_mem, const size_t, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*); +template StatusCode Tpmv<double2>(const Layout, const Triangle, const Transpose, const Diagonal, + const size_t, + const cl_mem, const size_t, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*); + +// Solves a triangular system of equations: STRSV/DTRSV/CTRSV/ZTRSV +template <typename T> +StatusCode Trsv(const Layout, const Triangle, const Transpose, const Diagonal, + const size_t, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*) { + return StatusCode::kNotImplemented; +} +template StatusCode Trsv<float>(const Layout, const Triangle, const Transpose, const Diagonal, + const size_t, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*); +template StatusCode Trsv<double>(const Layout, const Triangle, const Transpose, const Diagonal, + const size_t, const cl_mem, const size_t, const size_t, - const cl_mem, const size_t, const size_t, const double, cl_mem, const size_t, const size_t, cl_command_queue*, cl_event*); +template StatusCode Trsv<float2>(const Layout, const Triangle, const Transpose, const Diagonal, + const size_t, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*); +template StatusCode Trsv<double2>(const Layout, const Triangle, const Transpose, const Diagonal, + const size_t, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*); + +// Solves a banded triangular system of equations: STBSV/DTBSV/CTBSV/ZTBSV +template <typename T> +StatusCode Tbsv(const Layout, const Triangle, const Transpose, const Diagonal, + const size_t, const size_t, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*) { + return StatusCode::kNotImplemented; +} +template StatusCode Tbsv<float>(const Layout, const Triangle, const Transpose, const Diagonal, + const size_t, const size_t, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*); +template StatusCode Tbsv<double>(const Layout, const Triangle, const Transpose, const Diagonal, + const size_t, const size_t, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*); +template StatusCode Tbsv<float2>(const Layout, const Triangle, const Transpose, const Diagonal, + const size_t, const size_t, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*); +template StatusCode Tbsv<double2>(const Layout, const Triangle, const Transpose, const Diagonal, + const size_t, const size_t, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*); + +// Solves a packed triangular system of equations: STPSV/DTPSV/CTPSV/ZTPSV +template <typename T> +StatusCode Tpsv(const Layout, const Triangle, const Transpose, const Diagonal, + const size_t, + const cl_mem, const size_t, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*) { + return StatusCode::kNotImplemented; +} +template StatusCode Tpsv<float>(const Layout, const Triangle, const Transpose, const Diagonal, + const size_t, + const cl_mem, const size_t, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*); +template StatusCode Tpsv<double>(const Layout, const Triangle, const Transpose, const Diagonal, + const size_t, + const cl_mem, const size_t, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*); +template StatusCode Tpsv<float2>(const Layout, const Triangle, const Transpose, const Diagonal, + const size_t, + const cl_mem, const size_t, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*); +template StatusCode Tpsv<double2>(const Layout, const Triangle, const Transpose, const Diagonal, + const size_t, + const cl_mem, const size_t, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*); + +// General rank-1 matrix update: SGER/DGER +template <typename T> +StatusCode Ger(const Layout, + const size_t, const size_t, + const T, + const cl_mem, const size_t, const size_t, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*) { + return StatusCode::kNotImplemented; +} +template StatusCode Ger<float>(const Layout, + const size_t, const size_t, + const float, + const cl_mem, const size_t, const size_t, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*); +template StatusCode Ger<double>(const Layout, + const size_t, const size_t, + const double, + const cl_mem, const size_t, const size_t, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*); + +// General rank-1 complex matrix update: CGERU/ZGERU +template <typename T> +StatusCode Geru(const Layout, + const size_t, const size_t, + const T, + const cl_mem, const size_t, const size_t, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*) { + return StatusCode::kNotImplemented; +} +template StatusCode Geru<float2>(const Layout, + const size_t, const size_t, + const float2, + const cl_mem, const size_t, const size_t, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*); +template StatusCode Geru<double2>(const Layout, + const size_t, const size_t, + const double2, + const cl_mem, const size_t, const size_t, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*); + +// General rank-1 complex conjugated matrix update: CGERC/ZGERC +template <typename T> +StatusCode Gerc(const Layout, + const size_t, const size_t, + const T, + const cl_mem, const size_t, const size_t, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*) { + return StatusCode::kNotImplemented; +} +template StatusCode Gerc<float2>(const Layout, + const size_t, const size_t, + const float2, + const cl_mem, const size_t, const size_t, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*); +template StatusCode Gerc<double2>(const Layout, + const size_t, const size_t, + const double2, + const cl_mem, const size_t, const size_t, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*); + +// Hermitian rank-1 matrix update: CHER/ZHER +template <typename T> +StatusCode Her(const Layout, const Triangle, + const size_t, + const T, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*) { + return StatusCode::kNotImplemented; +} +template StatusCode Her<float>(const Layout, const Triangle, + const size_t, + const float, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*); +template StatusCode Her<double>(const Layout, const Triangle, + const size_t, + const double, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*); + +// Hermitian packed rank-1 matrix update: CHPR/ZHPR +template <typename T> +StatusCode Hpr(const Layout, const Triangle, + const size_t, + const T, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, + cl_command_queue*, cl_event*) { + return StatusCode::kNotImplemented; +} +template StatusCode Hpr<float>(const Layout, const Triangle, + const size_t, + const float, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, + cl_command_queue*, cl_event*); +template StatusCode Hpr<double>(const Layout, const Triangle, + const size_t, + const double, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, + cl_command_queue*, cl_event*); + +// Hermitian rank-2 matrix update: CHER2/ZHER2 +template <typename T> +StatusCode Her2(const Layout, const Triangle, + const size_t, + const T, + const cl_mem, const size_t, const size_t, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*) { + return StatusCode::kNotImplemented; +} +template StatusCode Her2<float2>(const Layout, const Triangle, + const size_t, + const float2, + const cl_mem, const size_t, const size_t, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*); +template StatusCode Her2<double2>(const Layout, const Triangle, + const size_t, + const double2, + const cl_mem, const size_t, const size_t, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*); + +// Hermitian packed rank-2 matrix update: CHPR2/ZHPR2 +template <typename T> +StatusCode Hpr2(const Layout, const Triangle, + const size_t, + const T, + const cl_mem, const size_t, const size_t, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, + cl_command_queue*, cl_event*) { + return StatusCode::kNotImplemented; +} +template StatusCode Hpr2<float2>(const Layout, const Triangle, + const size_t, + const float2, + const cl_mem, const size_t, const size_t, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, + cl_command_queue*, cl_event*); +template StatusCode Hpr2<double2>(const Layout, const Triangle, + const size_t, + const double2, + const cl_mem, const size_t, const size_t, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, + cl_command_queue*, cl_event*); + +// Symmetric rank-1 matrix update: SSYR/DSYR +template <typename T> +StatusCode Syr(const Layout, const Triangle, + const size_t, + const T, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*) { + return StatusCode::kNotImplemented; +} +template StatusCode Syr<float>(const Layout, const Triangle, + const size_t, + const float, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*); +template StatusCode Syr<double>(const Layout, const Triangle, + const size_t, + const double, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*); + +// Symmetric packed rank-1 matrix update: SSPR/DSPR +template <typename T> +StatusCode Spr(const Layout, const Triangle, + const size_t, + const T, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, + cl_command_queue*, cl_event*) { + return StatusCode::kNotImplemented; +} +template StatusCode Spr<float>(const Layout, const Triangle, + const size_t, + const float, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, + cl_command_queue*, cl_event*); +template StatusCode Spr<double>(const Layout, const Triangle, + const size_t, + const double, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, + cl_command_queue*, cl_event*); + +// Symmetric rank-2 matrix update: SSYR2/DSYR2 +template <typename T> +StatusCode Syr2(const Layout, const Triangle, + const size_t, + const T, + const cl_mem, const size_t, const size_t, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*) { + return StatusCode::kNotImplemented; +} +template StatusCode Syr2<float>(const Layout, const Triangle, + const size_t, + const float, + const cl_mem, const size_t, const size_t, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*); +template StatusCode Syr2<double>(const Layout, const Triangle, + const size_t, + const double, + const cl_mem, const size_t, const size_t, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*); + +// Symmetric packed rank-2 matrix update: SSPR2/DSPR2 +template <typename T> +StatusCode Spr2(const Layout, const Triangle, + const size_t, + const T, + const cl_mem, const size_t, const size_t, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, + cl_command_queue*, cl_event*) { + return StatusCode::kNotImplemented; +} +template StatusCode Spr2<float>(const Layout, const Triangle, + const size_t, + const float, + const cl_mem, const size_t, const size_t, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, + cl_command_queue*, cl_event*); +template StatusCode Spr2<double>(const Layout, const Triangle, + const size_t, + const double, + const cl_mem, const size_t, const size_t, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, + cl_command_queue*, cl_event*); // ================================================================================================= // BLAS level-3 (matrix-matrix) routines // ================================================================================================= -// GEMM +// General matrix-matrix multiplication: SGEMM/DGEMM/CGEMM/ZGEMM template <typename T> StatusCode Gemm(const Layout layout, const Transpose a_transpose, const Transpose b_transpose, - const size_t m, const size_t n, const size_t k, const T alpha, + const size_t m, const size_t n, const size_t k, + const T alpha, const cl_mem a_buffer, const size_t a_offset, const size_t a_ld, - const cl_mem b_buffer, const size_t b_offset, const size_t b_ld, const T beta, + const cl_mem b_buffer, const size_t b_offset, const size_t b_ld, + const T beta, cl_mem c_buffer, const size_t c_offset, const size_t c_ld, cl_command_queue* queue, cl_event* event) { auto queue_cpp = Queue(*queue); auto event_cpp = Event(*event); auto routine = Xgemm<T>(queue_cpp, event_cpp); - - // Compiles the routine's device kernels auto status = routine.SetUp(); if (status != StatusCode::kSuccess) { return status; } - - // Runs the routine - return routine.DoGemm(layout, a_transpose, b_transpose, m, n, k, alpha, + return routine.DoGemm(layout, a_transpose, b_transpose, + m, n, k, + alpha, Buffer<T>(a_buffer), a_offset, a_ld, - Buffer<T>(b_buffer), b_offset, b_ld, beta, + Buffer<T>(b_buffer), b_offset, b_ld, + beta, Buffer<T>(c_buffer), c_offset, c_ld); } template StatusCode Gemm<float>(const Layout, const Transpose, const Transpose, - const size_t, const size_t, const size_t, const float, + const size_t, const size_t, const size_t, + const float, + const cl_mem, const size_t, const size_t, const cl_mem, const size_t, const size_t, - const cl_mem, const size_t, const size_t, const float, + const float, cl_mem, const size_t, const size_t, cl_command_queue*, cl_event*); template StatusCode Gemm<double>(const Layout, const Transpose, const Transpose, - const size_t, const size_t, const size_t, const double, + const size_t, const size_t, const size_t, + const double, const cl_mem, const size_t, const size_t, - const cl_mem, const size_t, const size_t, const double, + const cl_mem, const size_t, const size_t, + const double, cl_mem, const size_t, const size_t, cl_command_queue*, cl_event*); template StatusCode Gemm<float2>(const Layout, const Transpose, const Transpose, - const size_t, const size_t, const size_t, const float2, + const size_t, const size_t, const size_t, + const float2, + const cl_mem, const size_t, const size_t, const cl_mem, const size_t, const size_t, - const cl_mem, const size_t, const size_t, const float2, + const float2, cl_mem, const size_t, const size_t, cl_command_queue*, cl_event*); template StatusCode Gemm<double2>(const Layout, const Transpose, const Transpose, - const size_t, const size_t, const size_t, const double2, + const size_t, const size_t, const size_t, + const double2, const cl_mem, const size_t, const size_t, - const cl_mem, const size_t, const size_t, const double2, + const cl_mem, const size_t, const size_t, + const double2, cl_mem, const size_t, const size_t, cl_command_queue*, cl_event*); -// ================================================================================================= - -// SYMM +// Symmetric matrix-matrix multiplication: SSYMM/DSYMM/CSYMM/ZSYMM template <typename T> StatusCode Symm(const Layout layout, const Side side, const Triangle triangle, - const size_t m, const size_t n, const T alpha, + const size_t m, const size_t n, + const T alpha, const cl_mem a_buffer, const size_t a_offset, const size_t a_ld, - const cl_mem b_buffer, const size_t b_offset, const size_t b_ld, const T beta, + const cl_mem b_buffer, const size_t b_offset, const size_t b_ld, + const T beta, cl_mem c_buffer, const size_t c_offset, const size_t c_ld, cl_command_queue* queue, cl_event* event) { auto queue_cpp = Queue(*queue); auto event_cpp = Event(*event); auto routine = Xsymm<T>(queue_cpp, event_cpp); - - // Compiles the routine's device kernels auto status = routine.SetUp(); if (status != StatusCode::kSuccess) { return status; } - - // Runs the routine - return routine.DoSymm(layout, side, triangle, m, n, alpha, + return routine.DoSymm(layout, side, triangle, + m, n, + alpha, Buffer<T>(a_buffer), a_offset, a_ld, - Buffer<T>(b_buffer), b_offset, b_ld, beta, + Buffer<T>(b_buffer), b_offset, b_ld, + beta, Buffer<T>(c_buffer), c_offset, c_ld); } template StatusCode Symm<float>(const Layout, const Side, const Triangle, - const size_t, const size_t, const float, + const size_t, const size_t, + const float, + const cl_mem, const size_t, const size_t, const cl_mem, const size_t, const size_t, - const cl_mem, const size_t, const size_t, const float, + const float, cl_mem, const size_t, const size_t, cl_command_queue*, cl_event*); template StatusCode Symm<double>(const Layout, const Side, const Triangle, - const size_t, const size_t, const double, + const size_t, const size_t, + const double, const cl_mem, const size_t, const size_t, - const cl_mem, const size_t, const size_t, const double, + const cl_mem, const size_t, const size_t, + const double, cl_mem, const size_t, const size_t, cl_command_queue*, cl_event*); template StatusCode Symm<float2>(const Layout, const Side, const Triangle, - const size_t, const size_t, const float2, + const size_t, const size_t, + const float2, + const cl_mem, const size_t, const size_t, const cl_mem, const size_t, const size_t, - const cl_mem, const size_t, const size_t, const float2, + const float2, cl_mem, const size_t, const size_t, cl_command_queue*, cl_event*); template StatusCode Symm<double2>(const Layout, const Side, const Triangle, - const size_t, const size_t, const double2, + const size_t, const size_t, + const double2, const cl_mem, const size_t, const size_t, - const cl_mem, const size_t, const size_t, const double2, + const cl_mem, const size_t, const size_t, + const double2, cl_mem, const size_t, const size_t, cl_command_queue*, cl_event*); -// ================================================================================================= - -// HEMM +// Hermitian matrix-matrix multiplication: CHEMM/ZHEMM template <typename T> StatusCode Hemm(const Layout layout, const Side side, const Triangle triangle, - const size_t m, const size_t n, const T alpha, + const size_t m, const size_t n, + const T alpha, const cl_mem a_buffer, const size_t a_offset, const size_t a_ld, - const cl_mem b_buffer, const size_t b_offset, const size_t b_ld, const T beta, + const cl_mem b_buffer, const size_t b_offset, const size_t b_ld, + const T beta, cl_mem c_buffer, const size_t c_offset, const size_t c_ld, cl_command_queue* queue, cl_event* event) { auto queue_cpp = Queue(*queue); auto event_cpp = Event(*event); auto routine = Xhemm<T>(queue_cpp, event_cpp); - - // Compiles the routine's device kernels auto status = routine.SetUp(); if (status != StatusCode::kSuccess) { return status; } - - // Runs the routine - return routine.DoHemm(layout, side, triangle, m, n, alpha, + return routine.DoHemm(layout, side, triangle, + m, n, + alpha, Buffer<T>(a_buffer), a_offset, a_ld, - Buffer<T>(b_buffer), b_offset, b_ld, beta, + Buffer<T>(b_buffer), b_offset, b_ld, + beta, Buffer<T>(c_buffer), c_offset, c_ld); } template StatusCode Hemm<float2>(const Layout, const Side, const Triangle, - const size_t, const size_t, const float2, + const size_t, const size_t, + const float2, + const cl_mem, const size_t, const size_t, const cl_mem, const size_t, const size_t, - const cl_mem, const size_t, const size_t, const float2, + const float2, cl_mem, const size_t, const size_t, cl_command_queue*, cl_event*); template StatusCode Hemm<double2>(const Layout, const Side, const Triangle, - const size_t, const size_t, const double2, + const size_t, const size_t, + const double2, const cl_mem, const size_t, const size_t, - const cl_mem, const size_t, const size_t, const double2, + const cl_mem, const size_t, const size_t, + const double2, cl_mem, const size_t, const size_t, cl_command_queue*, cl_event*); -// ================================================================================================= - -// SYRK +// Rank-K update of a symmetric matrix: SSYRK/DSYRK/CSYRK/ZSYRK template <typename T> StatusCode Syrk(const Layout layout, const Triangle triangle, const Transpose a_transpose, - const size_t n, const size_t k, const T alpha, - const cl_mem a_buffer, const size_t a_offset, const size_t a_ld, const T beta, + const size_t n, const size_t k, + const T alpha, + const cl_mem a_buffer, const size_t a_offset, const size_t a_ld, + const T beta, cl_mem c_buffer, const size_t c_offset, const size_t c_ld, cl_command_queue* queue, cl_event* event) { auto queue_cpp = Queue(*queue); auto event_cpp = Event(*event); auto routine = Xsyrk<T>(queue_cpp, event_cpp); - - // Compiles the routine's device kernels auto status = routine.SetUp(); if (status != StatusCode::kSuccess) { return status; } - - // Runs the routine - return routine.DoSyrk(layout, triangle, a_transpose, n, k, alpha, - Buffer<T>(a_buffer), a_offset, a_ld, beta, + return routine.DoSyrk(layout, triangle, a_transpose, + n, k, + alpha, + Buffer<T>(a_buffer), a_offset, a_ld, + beta, Buffer<T>(c_buffer), c_offset, c_ld); } template StatusCode Syrk<float>(const Layout, const Triangle, const Transpose, - const size_t, const size_t, const float, - const cl_mem, const size_t, const size_t, const float, + const size_t, const size_t, + const float, + const cl_mem, const size_t, const size_t, + const float, cl_mem, const size_t, const size_t, cl_command_queue*, cl_event*); template StatusCode Syrk<double>(const Layout, const Triangle, const Transpose, - const size_t, const size_t, const double, - const cl_mem, const size_t, const size_t, const double, + const size_t, const size_t, + const double, + const cl_mem, const size_t, const size_t, + const double, cl_mem, const size_t, const size_t, cl_command_queue*, cl_event*); template StatusCode Syrk<float2>(const Layout, const Triangle, const Transpose, - const size_t, const size_t, const float2, - const cl_mem, const size_t, const size_t, const float2, + const size_t, const size_t, + const float2, + const cl_mem, const size_t, const size_t, + const float2, cl_mem, const size_t, const size_t, cl_command_queue*, cl_event*); template StatusCode Syrk<double2>(const Layout, const Triangle, const Transpose, - const size_t, const size_t, const double2, - const cl_mem, const size_t, const size_t, const double2, + const size_t, const size_t, + const double2, + const cl_mem, const size_t, const size_t, + const double2, cl_mem, const size_t, const size_t, cl_command_queue*, cl_event*); -// ================================================================================================= - -// HERK +// Rank-K update of a hermitian matrix: CHERK/ZHERK template <typename T> StatusCode Herk(const Layout layout, const Triangle triangle, const Transpose a_transpose, - const size_t n, const size_t k, const T alpha, - const cl_mem a_buffer, const size_t a_offset, const size_t a_ld, const T beta, + const size_t n, const size_t k, + const T alpha, + const cl_mem a_buffer, const size_t a_offset, const size_t a_ld, + const T beta, cl_mem c_buffer, const size_t c_offset, const size_t c_ld, cl_command_queue* queue, cl_event* event) { auto queue_cpp = Queue(*queue); auto event_cpp = Event(*event); auto routine = Xherk<std::complex<T>,T>(queue_cpp, event_cpp); - - // Compiles the routine's device kernels auto status = routine.SetUp(); if (status != StatusCode::kSuccess) { return status; } - - // Runs the routine - return routine.DoHerk(layout, triangle, a_transpose, n, k, alpha, - Buffer<std::complex<T>>(a_buffer), a_offset, a_ld, beta, + return routine.DoHerk(layout, triangle, a_transpose, + n, k, + alpha, + Buffer<std::complex<T>>(a_buffer), a_offset, a_ld, + beta, Buffer<std::complex<T>>(c_buffer), c_offset, c_ld); } template StatusCode Herk<float>(const Layout, const Triangle, const Transpose, - const size_t, const size_t, const float, - const cl_mem, const size_t, const size_t, const float, + const size_t, const size_t, + const float, + const cl_mem, const size_t, const size_t, + const float, cl_mem, const size_t, const size_t, cl_command_queue*, cl_event*); template StatusCode Herk<double>(const Layout, const Triangle, const Transpose, - const size_t, const size_t, const double, - const cl_mem, const size_t, const size_t, const double, + const size_t, const size_t, + const double, + const cl_mem, const size_t, const size_t, + const double, cl_mem, const size_t, const size_t, cl_command_queue*, cl_event*); -// ================================================================================================= - -// SYR2K +// Rank-2K update of a symmetric matrix: SSYR2K/DSYR2K/CSYR2K/ZSYR2K template <typename T> StatusCode Syr2k(const Layout layout, const Triangle triangle, const Transpose ab_transpose, - const size_t n, const size_t k, const T alpha, + const size_t n, const size_t k, + const T alpha, const cl_mem a_buffer, const size_t a_offset, const size_t a_ld, - const cl_mem b_buffer, const size_t b_offset, const size_t b_ld, const T beta, + const cl_mem b_buffer, const size_t b_offset, const size_t b_ld, + const T beta, cl_mem c_buffer, const size_t c_offset, const size_t c_ld, cl_command_queue* queue, cl_event* event) { auto queue_cpp = Queue(*queue); auto event_cpp = Event(*event); auto routine = Xsyr2k<T>(queue_cpp, event_cpp); - - // Compiles the routine's device kernels auto status = routine.SetUp(); if (status != StatusCode::kSuccess) { return status; } - - // Runs the routine - return routine.DoSyr2k(layout, triangle, ab_transpose, n, k, alpha, + return routine.DoSyr2k(layout, triangle, ab_transpose, + n, k, + alpha, Buffer<T>(a_buffer), a_offset, a_ld, - Buffer<T>(b_buffer), b_offset, b_ld, beta, + Buffer<T>(b_buffer), b_offset, b_ld, + beta, Buffer<T>(c_buffer), c_offset, c_ld); } template StatusCode Syr2k<float>(const Layout, const Triangle, const Transpose, - const size_t, const size_t, const float, + const size_t, const size_t, + const float, + const cl_mem, const size_t, const size_t, const cl_mem, const size_t, const size_t, - const cl_mem, const size_t, const size_t, const float, + const float, cl_mem, const size_t, const size_t, cl_command_queue*, cl_event*); template StatusCode Syr2k<double>(const Layout, const Triangle, const Transpose, - const size_t, const size_t, const double, + const size_t, const size_t, + const double, const cl_mem, const size_t, const size_t, - const cl_mem, const size_t, const size_t, const double, + const cl_mem, const size_t, const size_t, + const double, cl_mem, const size_t, const size_t, cl_command_queue*, cl_event*); template StatusCode Syr2k<float2>(const Layout, const Triangle, const Transpose, - const size_t, const size_t, const float2, + const size_t, const size_t, + const float2, + const cl_mem, const size_t, const size_t, const cl_mem, const size_t, const size_t, - const cl_mem, const size_t, const size_t, const float2, + const float2, cl_mem, const size_t, const size_t, cl_command_queue*, cl_event*); template StatusCode Syr2k<double2>(const Layout, const Triangle, const Transpose, - const size_t, const size_t, const double2, + const size_t, const size_t, + const double2, const cl_mem, const size_t, const size_t, - const cl_mem, const size_t, const size_t, const double2, + const cl_mem, const size_t, const size_t, + const double2, cl_mem, const size_t, const size_t, cl_command_queue*, cl_event*); -// ================================================================================================= - -// SYR2K +// Rank-2K update of a hermitian matrix: CHER2K/ZHER2K template <typename T, typename U> StatusCode Her2k(const Layout layout, const Triangle triangle, const Transpose ab_transpose, - const size_t n, const size_t k, const T alpha, + const size_t n, const size_t k, + const T alpha, const cl_mem a_buffer, const size_t a_offset, const size_t a_ld, - const cl_mem b_buffer, const size_t b_offset, const size_t b_ld, const U beta, + const cl_mem b_buffer, const size_t b_offset, const size_t b_ld, + const U beta, cl_mem c_buffer, const size_t c_offset, const size_t c_ld, cl_command_queue* queue, cl_event* event) { auto queue_cpp = Queue(*queue); auto event_cpp = Event(*event); auto routine = Xher2k<T,U>(queue_cpp, event_cpp); - - // Compiles the routine's device kernels auto status = routine.SetUp(); if (status != StatusCode::kSuccess) { return status; } - - // Runs the routine - return routine.DoHer2k(layout, triangle, ab_transpose, n, k, alpha, + return routine.DoHer2k(layout, triangle, ab_transpose, + n, k, + alpha, Buffer<T>(a_buffer), a_offset, a_ld, - Buffer<T>(b_buffer), b_offset, b_ld, beta, + Buffer<T>(b_buffer), b_offset, b_ld, + beta, Buffer<T>(c_buffer), c_offset, c_ld); } template StatusCode Her2k<float2,float>(const Layout, const Triangle, const Transpose, - const size_t, const size_t, const float2, + const size_t, const size_t, + const float2, + const cl_mem, const size_t, const size_t, const cl_mem, const size_t, const size_t, - const cl_mem, const size_t, const size_t, const float, + const float, cl_mem, const size_t, const size_t, cl_command_queue*, cl_event*); template StatusCode Her2k<double2,double>(const Layout, const Triangle, const Transpose, - const size_t, const size_t, const double2, + const size_t, const size_t, + const double2, + const cl_mem, const size_t, const size_t, const cl_mem, const size_t, const size_t, - const cl_mem, const size_t, const size_t, const double, + const double, cl_mem, const size_t, const size_t, cl_command_queue*, cl_event*); -// ================================================================================================= - -// TRMM +// Triangular matrix-matrix multiplication: STRMM/DTRMM/CTRMM/ZTRMM template <typename T> -StatusCode Trmm(const Layout layout, const Side side, const Triangle triangle, - const Transpose a_transpose, const Diagonal diagonal, +StatusCode Trmm(const Layout layout, const Side side, const Triangle triangle, const Transpose a_transpose, const Diagonal diagonal, const size_t m, const size_t n, const T alpha, const cl_mem a_buffer, const size_t a_offset, const size_t a_ld, @@ -705,90 +1369,73 @@ StatusCode Trmm(const Layout layout, const Side side, const Triangle triangle, auto queue_cpp = Queue(*queue); auto event_cpp = Event(*event); auto routine = Xtrmm<T>(queue_cpp, event_cpp); - - // Compiles the routine's device kernels auto status = routine.SetUp(); if (status != StatusCode::kSuccess) { return status; } - - // Runs the routine - return routine.DoTrmm(layout, side, triangle, a_transpose, diagonal, m, n, alpha, + return routine.DoTrmm(layout, side, triangle, a_transpose, diagonal, + m, n, + alpha, Buffer<T>(a_buffer), a_offset, a_ld, Buffer<T>(b_buffer), b_offset, b_ld); } -template StatusCode Trmm<float>(const Layout, const Side, const Triangle, - const Transpose, const Diagonal, - const size_t, const size_t, const float, +template StatusCode Trmm<float>(const Layout, const Side, const Triangle, const Transpose, const Diagonal, + const size_t, const size_t, + const float, const cl_mem, const size_t, const size_t, cl_mem, const size_t, const size_t, cl_command_queue*, cl_event*); -template StatusCode Trmm<double>(const Layout, const Side, const Triangle, - const Transpose, const Diagonal, - const size_t, const size_t, const double, +template StatusCode Trmm<double>(const Layout, const Side, const Triangle, const Transpose, const Diagonal, + const size_t, const size_t, + const double, const cl_mem, const size_t, const size_t, cl_mem, const size_t, const size_t, cl_command_queue*, cl_event*); -template StatusCode Trmm<float2>(const Layout, const Side, const Triangle, - const Transpose, const Diagonal, - const size_t, const size_t, const float2, +template StatusCode Trmm<float2>(const Layout, const Side, const Triangle, const Transpose, const Diagonal, + const size_t, const size_t, + const float2, const cl_mem, const size_t, const size_t, cl_mem, const size_t, const size_t, cl_command_queue*, cl_event*); -template StatusCode Trmm<double2>(const Layout, const Side, const Triangle, - const Transpose, const Diagonal, - const size_t, const size_t, const double2, +template StatusCode Trmm<double2>(const Layout, const Side, const Triangle, const Transpose, const Diagonal, + const size_t, const size_t, + const double2, const cl_mem, const size_t, const size_t, cl_mem, const size_t, const size_t, cl_command_queue*, cl_event*); -// ================================================================================================= - -// TRSM -/* +// Solves a triangular system of equations: STRSM/DTRSM/CTRSM/ZTRSM template <typename T> -StatusCode Trsm(const Layout layout, const Side side, const Triangle triangle, - const Transpose a_transpose, const Diagonal diagonal, - const size_t m, const size_t n, - const T alpha, - const cl_mem a_buffer, const size_t a_offset, const size_t a_ld, - cl_mem b_buffer, const size_t b_offset, const size_t b_ld, - cl_command_queue* queue, cl_event* event) { - auto queue_cpp = Queue(*queue); - auto event_cpp = Event(*event); - auto routine = Xtrsm<T>(queue_cpp, event_cpp); - - // Compiles the routine's device kernels - auto status = routine.SetUp(); - if (status != StatusCode::kSuccess) { return status; } - - // Runs the routine - return routine.DoTrsm(layout, side, triangle, a_transpose, diagonal, m, n, alpha, - Buffer<T>(a_buffer), a_offset, a_ld, - Buffer<T>(b_buffer), b_offset, b_ld); +StatusCode Trsm(const Layout, const Side, const Triangle, const Transpose, const Diagonal, + const size_t, const size_t, + const T, + const cl_mem, const size_t, const size_t, + cl_mem, const size_t, const size_t, + cl_command_queue*, cl_event*) { + return StatusCode::kNotImplemented; } -template StatusCode Trsm<float>(const Layout, const Side, const Triangle, - const Transpose, const Diagonal, - const size_t, const size_t, const float, +template StatusCode Trsm<float>(const Layout, const Side, const Triangle, const Transpose, const Diagonal, + const size_t, const size_t, + const float, const cl_mem, const size_t, const size_t, cl_mem, const size_t, const size_t, cl_command_queue*, cl_event*); -template StatusCode Trsm<double>(const Layout, const Side, const Triangle, - const Transpose, const Diagonal, - const size_t, const size_t, const double, +template StatusCode Trsm<double>(const Layout, const Side, const Triangle, const Transpose, const Diagonal, + const size_t, const size_t, + const double, const cl_mem, const size_t, const size_t, cl_mem, const size_t, const size_t, cl_command_queue*, cl_event*); -template StatusCode Trsm<float2>(const Layout, const Side, const Triangle, - const Transpose, const Diagonal, - const size_t, const size_t, const float2, +template StatusCode Trsm<float2>(const Layout, const Side, const Triangle, const Transpose, const Diagonal, + const size_t, const size_t, + const float2, const cl_mem, const size_t, const size_t, cl_mem, const size_t, const size_t, cl_command_queue*, cl_event*); -template StatusCode Trsm<double2>(const Layout, const Side, const Triangle, - const Transpose, const Diagonal, - const size_t, const size_t, const double2, +template StatusCode Trsm<double2>(const Layout, const Side, const Triangle, const Transpose, const Diagonal, + const size_t, const size_t, + const double2, const cl_mem, const size_t, const size_t, cl_mem, const size_t, const size_t, cl_command_queue*, cl_event*); -*/ + // ================================================================================================= } // namespace clblast diff --git a/src/clblast_c.cc b/src/clblast_c.cc index eccf517f..fcec0951 100644 --- a/src/clblast_c.cc +++ b/src/clblast_c.cc @@ -363,6 +363,84 @@ StatusCode CLBlastZgemv(const Layout layout, const Transpose a_transpose, return static_cast<StatusCode>(status); } +// GBMV +StatusCode CLBlastSgbmv(const Layout layout, const Transpose a_transpose, + const size_t m, const size_t n, const size_t kl, const size_t ku, + const float alpha, + const cl_mem a_buffer, const size_t a_offset, const size_t a_ld, + const cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + const float beta, + cl_mem y_buffer, const size_t y_offset, const size_t y_inc, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Gbmv(static_cast<clblast::Layout>(layout), + static_cast<clblast::Transpose>(a_transpose), + m, n, kl, ku, + alpha, + a_buffer, a_offset, a_ld, + x_buffer, x_offset, x_inc, + beta, + y_buffer, y_offset, y_inc, + queue, event); + return static_cast<StatusCode>(status); +} +StatusCode CLBlastDgbmv(const Layout layout, const Transpose a_transpose, + const size_t m, const size_t n, const size_t kl, const size_t ku, + const double alpha, + const cl_mem a_buffer, const size_t a_offset, const size_t a_ld, + const cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + const double beta, + cl_mem y_buffer, const size_t y_offset, const size_t y_inc, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Gbmv(static_cast<clblast::Layout>(layout), + static_cast<clblast::Transpose>(a_transpose), + m, n, kl, ku, + alpha, + a_buffer, a_offset, a_ld, + x_buffer, x_offset, x_inc, + beta, + y_buffer, y_offset, y_inc, + queue, event); + return static_cast<StatusCode>(status); +} +StatusCode CLBlastCgbmv(const Layout layout, const Transpose a_transpose, + const size_t m, const size_t n, const size_t kl, const size_t ku, + const cl_float2 alpha, + const cl_mem a_buffer, const size_t a_offset, const size_t a_ld, + const cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + const cl_float2 beta, + cl_mem y_buffer, const size_t y_offset, const size_t y_inc, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Gbmv(static_cast<clblast::Layout>(layout), + static_cast<clblast::Transpose>(a_transpose), + m, n, kl, ku, + float2{alpha.s[0], alpha.s[1]}, + a_buffer, a_offset, a_ld, + x_buffer, x_offset, x_inc, + float2{beta.s[0], beta.s[1]}, + y_buffer, y_offset, y_inc, + queue, event); + return static_cast<StatusCode>(status); +} +StatusCode CLBlastZgbmv(const Layout layout, const Transpose a_transpose, + const size_t m, const size_t n, const size_t kl, const size_t ku, + const cl_double2 alpha, + const cl_mem a_buffer, const size_t a_offset, const size_t a_ld, + const cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + const cl_double2 beta, + cl_mem y_buffer, const size_t y_offset, const size_t y_inc, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Gbmv(static_cast<clblast::Layout>(layout), + static_cast<clblast::Transpose>(a_transpose), + m, n, kl, ku, + double2{alpha.s[0], alpha.s[1]}, + a_buffer, a_offset, a_ld, + x_buffer, x_offset, x_inc, + double2{beta.s[0], beta.s[1]}, + y_buffer, y_offset, y_inc, + queue, event); + return static_cast<StatusCode>(status); +} + // HEMV StatusCode CLBlastChemv(const Layout layout, const Triangle triangle, const size_t n, @@ -403,6 +481,86 @@ StatusCode CLBlastZhemv(const Layout layout, const Triangle triangle, return static_cast<StatusCode>(status); } +// HBMV +StatusCode CLBlastChbmv(const Layout layout, const Triangle triangle, + const size_t n, const size_t k, + const cl_float2 alpha, + const cl_mem a_buffer, const size_t a_offset, const size_t a_ld, + const cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + const cl_float2 beta, + cl_mem y_buffer, const size_t y_offset, const size_t y_inc, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Hbmv(static_cast<clblast::Layout>(layout), + static_cast<clblast::Triangle>(triangle), + n, k, + float2{alpha.s[0], alpha.s[1]}, + a_buffer, a_offset, a_ld, + x_buffer, x_offset, x_inc, + float2{beta.s[0], beta.s[1]}, + y_buffer, y_offset, y_inc, + queue, event); + return static_cast<StatusCode>(status); +} +StatusCode CLBlastZhbmv(const Layout layout, const Triangle triangle, + const size_t n, const size_t k, + const cl_double2 alpha, + const cl_mem a_buffer, const size_t a_offset, const size_t a_ld, + const cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + const cl_double2 beta, + cl_mem y_buffer, const size_t y_offset, const size_t y_inc, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Hbmv(static_cast<clblast::Layout>(layout), + static_cast<clblast::Triangle>(triangle), + n, k, + double2{alpha.s[0], alpha.s[1]}, + a_buffer, a_offset, a_ld, + x_buffer, x_offset, x_inc, + double2{beta.s[0], beta.s[1]}, + y_buffer, y_offset, y_inc, + queue, event); + return static_cast<StatusCode>(status); +} + +// HPMV +StatusCode CLBlastChpmv(const Layout layout, const Triangle triangle, + const size_t n, + const cl_float2 alpha, + const cl_mem ap_buffer, const size_t ap_offset, + const cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + const cl_float2 beta, + cl_mem y_buffer, const size_t y_offset, const size_t y_inc, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Hpmv(static_cast<clblast::Layout>(layout), + static_cast<clblast::Triangle>(triangle), + n, + float2{alpha.s[0], alpha.s[1]}, + ap_buffer, ap_offset, + x_buffer, x_offset, x_inc, + float2{beta.s[0], beta.s[1]}, + y_buffer, y_offset, y_inc, + queue, event); + return static_cast<StatusCode>(status); +} +StatusCode CLBlastZhpmv(const Layout layout, const Triangle triangle, + const size_t n, + const cl_double2 alpha, + const cl_mem ap_buffer, const size_t ap_offset, + const cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + const cl_double2 beta, + cl_mem y_buffer, const size_t y_offset, const size_t y_inc, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Hpmv(static_cast<clblast::Layout>(layout), + static_cast<clblast::Triangle>(triangle), + n, + double2{alpha.s[0], alpha.s[1]}, + ap_buffer, ap_offset, + x_buffer, x_offset, x_inc, + double2{beta.s[0], beta.s[1]}, + y_buffer, y_offset, y_inc, + queue, event); + return static_cast<StatusCode>(status); +} + // SYMV StatusCode CLBlastSsymv(const Layout layout, const Triangle triangle, const size_t n, @@ -443,6 +601,832 @@ StatusCode CLBlastDsymv(const Layout layout, const Triangle triangle, return static_cast<StatusCode>(status); } +// SBMV +StatusCode CLBlastSsbmv(const Layout layout, const Triangle triangle, + const size_t n, const size_t k, + const float alpha, + const cl_mem a_buffer, const size_t a_offset, const size_t a_ld, + const cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + const float beta, + cl_mem y_buffer, const size_t y_offset, const size_t y_inc, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Sbmv(static_cast<clblast::Layout>(layout), + static_cast<clblast::Triangle>(triangle), + n, k, + alpha, + a_buffer, a_offset, a_ld, + x_buffer, x_offset, x_inc, + beta, + y_buffer, y_offset, y_inc, + queue, event); + return static_cast<StatusCode>(status); +} +StatusCode CLBlastDsbmv(const Layout layout, const Triangle triangle, + const size_t n, const size_t k, + const double alpha, + const cl_mem a_buffer, const size_t a_offset, const size_t a_ld, + const cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + const double beta, + cl_mem y_buffer, const size_t y_offset, const size_t y_inc, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Sbmv(static_cast<clblast::Layout>(layout), + static_cast<clblast::Triangle>(triangle), + n, k, + alpha, + a_buffer, a_offset, a_ld, + x_buffer, x_offset, x_inc, + beta, + y_buffer, y_offset, y_inc, + queue, event); + return static_cast<StatusCode>(status); +} + +// SPMV +StatusCode CLBlastSspmv(const Layout layout, const Triangle triangle, + const size_t n, + const float alpha, + const cl_mem ap_buffer, const size_t ap_offset, + const cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + const float beta, + cl_mem y_buffer, const size_t y_offset, const size_t y_inc, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Spmv(static_cast<clblast::Layout>(layout), + static_cast<clblast::Triangle>(triangle), + n, + alpha, + ap_buffer, ap_offset, + x_buffer, x_offset, x_inc, + beta, + y_buffer, y_offset, y_inc, + queue, event); + return static_cast<StatusCode>(status); +} +StatusCode CLBlastDspmv(const Layout layout, const Triangle triangle, + const size_t n, + const double alpha, + const cl_mem ap_buffer, const size_t ap_offset, + const cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + const double beta, + cl_mem y_buffer, const size_t y_offset, const size_t y_inc, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Spmv(static_cast<clblast::Layout>(layout), + static_cast<clblast::Triangle>(triangle), + n, + alpha, + ap_buffer, ap_offset, + x_buffer, x_offset, x_inc, + beta, + y_buffer, y_offset, y_inc, + queue, event); + return static_cast<StatusCode>(status); +} + +// TRMV +StatusCode CLBlastStrmv(const Layout layout, const Triangle triangle, const Transpose a_transpose, const Diagonal diagonal, + const size_t n, + const cl_mem a_buffer, const size_t a_offset, const size_t a_ld, + cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Trmv<float>(static_cast<clblast::Layout>(layout), + static_cast<clblast::Triangle>(triangle), + static_cast<clblast::Transpose>(a_transpose), + static_cast<clblast::Diagonal>(diagonal), + n, + a_buffer, a_offset, a_ld, + x_buffer, x_offset, x_inc, + queue, event); + return static_cast<StatusCode>(status); +} +StatusCode CLBlastDtrmv(const Layout layout, const Triangle triangle, const Transpose a_transpose, const Diagonal diagonal, + const size_t n, + const cl_mem a_buffer, const size_t a_offset, const size_t a_ld, + cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Trmv<double>(static_cast<clblast::Layout>(layout), + static_cast<clblast::Triangle>(triangle), + static_cast<clblast::Transpose>(a_transpose), + static_cast<clblast::Diagonal>(diagonal), + n, + a_buffer, a_offset, a_ld, + x_buffer, x_offset, x_inc, + queue, event); + return static_cast<StatusCode>(status); +} +StatusCode CLBlastCtrmv(const Layout layout, const Triangle triangle, const Transpose a_transpose, const Diagonal diagonal, + const size_t n, + const cl_mem a_buffer, const size_t a_offset, const size_t a_ld, + cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Trmv<float2>(static_cast<clblast::Layout>(layout), + static_cast<clblast::Triangle>(triangle), + static_cast<clblast::Transpose>(a_transpose), + static_cast<clblast::Diagonal>(diagonal), + n, + a_buffer, a_offset, a_ld, + x_buffer, x_offset, x_inc, + queue, event); + return static_cast<StatusCode>(status); +} +StatusCode CLBlastZtrmv(const Layout layout, const Triangle triangle, const Transpose a_transpose, const Diagonal diagonal, + const size_t n, + const cl_mem a_buffer, const size_t a_offset, const size_t a_ld, + cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Trmv<double2>(static_cast<clblast::Layout>(layout), + static_cast<clblast::Triangle>(triangle), + static_cast<clblast::Transpose>(a_transpose), + static_cast<clblast::Diagonal>(diagonal), + n, + a_buffer, a_offset, a_ld, + x_buffer, x_offset, x_inc, + queue, event); + return static_cast<StatusCode>(status); +} + +// TBMV +StatusCode CLBlastStbmv(const Layout layout, const Triangle triangle, const Transpose a_transpose, const Diagonal diagonal, + const size_t n, const size_t k, + const cl_mem a_buffer, const size_t a_offset, const size_t a_ld, + cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Tbmv<float>(static_cast<clblast::Layout>(layout), + static_cast<clblast::Triangle>(triangle), + static_cast<clblast::Transpose>(a_transpose), + static_cast<clblast::Diagonal>(diagonal), + n, k, + a_buffer, a_offset, a_ld, + x_buffer, x_offset, x_inc, + queue, event); + return static_cast<StatusCode>(status); +} +StatusCode CLBlastDtbmv(const Layout layout, const Triangle triangle, const Transpose a_transpose, const Diagonal diagonal, + const size_t n, const size_t k, + const cl_mem a_buffer, const size_t a_offset, const size_t a_ld, + cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Tbmv<double>(static_cast<clblast::Layout>(layout), + static_cast<clblast::Triangle>(triangle), + static_cast<clblast::Transpose>(a_transpose), + static_cast<clblast::Diagonal>(diagonal), + n, k, + a_buffer, a_offset, a_ld, + x_buffer, x_offset, x_inc, + queue, event); + return static_cast<StatusCode>(status); +} +StatusCode CLBlastCtbmv(const Layout layout, const Triangle triangle, const Transpose a_transpose, const Diagonal diagonal, + const size_t n, const size_t k, + const cl_mem a_buffer, const size_t a_offset, const size_t a_ld, + cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Tbmv<float2>(static_cast<clblast::Layout>(layout), + static_cast<clblast::Triangle>(triangle), + static_cast<clblast::Transpose>(a_transpose), + static_cast<clblast::Diagonal>(diagonal), + n, k, + a_buffer, a_offset, a_ld, + x_buffer, x_offset, x_inc, + queue, event); + return static_cast<StatusCode>(status); +} +StatusCode CLBlastZtbmv(const Layout layout, const Triangle triangle, const Transpose a_transpose, const Diagonal diagonal, + const size_t n, const size_t k, + const cl_mem a_buffer, const size_t a_offset, const size_t a_ld, + cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Tbmv<double2>(static_cast<clblast::Layout>(layout), + static_cast<clblast::Triangle>(triangle), + static_cast<clblast::Transpose>(a_transpose), + static_cast<clblast::Diagonal>(diagonal), + n, k, + a_buffer, a_offset, a_ld, + x_buffer, x_offset, x_inc, + queue, event); + return static_cast<StatusCode>(status); +} + +// TPMV +StatusCode CLBlastStpmv(const Layout layout, const Triangle triangle, const Transpose a_transpose, const Diagonal diagonal, + const size_t n, + const cl_mem ap_buffer, const size_t ap_offset, + cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Tpmv<float>(static_cast<clblast::Layout>(layout), + static_cast<clblast::Triangle>(triangle), + static_cast<clblast::Transpose>(a_transpose), + static_cast<clblast::Diagonal>(diagonal), + n, + ap_buffer, ap_offset, + x_buffer, x_offset, x_inc, + queue, event); + return static_cast<StatusCode>(status); +} +StatusCode CLBlastDtpmv(const Layout layout, const Triangle triangle, const Transpose a_transpose, const Diagonal diagonal, + const size_t n, + const cl_mem ap_buffer, const size_t ap_offset, + cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Tpmv<double>(static_cast<clblast::Layout>(layout), + static_cast<clblast::Triangle>(triangle), + static_cast<clblast::Transpose>(a_transpose), + static_cast<clblast::Diagonal>(diagonal), + n, + ap_buffer, ap_offset, + x_buffer, x_offset, x_inc, + queue, event); + return static_cast<StatusCode>(status); +} +StatusCode CLBlastCtpmv(const Layout layout, const Triangle triangle, const Transpose a_transpose, const Diagonal diagonal, + const size_t n, + const cl_mem ap_buffer, const size_t ap_offset, + cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Tpmv<float2>(static_cast<clblast::Layout>(layout), + static_cast<clblast::Triangle>(triangle), + static_cast<clblast::Transpose>(a_transpose), + static_cast<clblast::Diagonal>(diagonal), + n, + ap_buffer, ap_offset, + x_buffer, x_offset, x_inc, + queue, event); + return static_cast<StatusCode>(status); +} +StatusCode CLBlastZtpmv(const Layout layout, const Triangle triangle, const Transpose a_transpose, const Diagonal diagonal, + const size_t n, + const cl_mem ap_buffer, const size_t ap_offset, + cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Tpmv<double2>(static_cast<clblast::Layout>(layout), + static_cast<clblast::Triangle>(triangle), + static_cast<clblast::Transpose>(a_transpose), + static_cast<clblast::Diagonal>(diagonal), + n, + ap_buffer, ap_offset, + x_buffer, x_offset, x_inc, + queue, event); + return static_cast<StatusCode>(status); +} + +// TRSV +StatusCode CLBlastStrsv(const Layout layout, const Triangle triangle, const Transpose a_transpose, const Diagonal diagonal, + const size_t n, + const cl_mem a_buffer, const size_t a_offset, const size_t a_ld, + cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Trsv<float>(static_cast<clblast::Layout>(layout), + static_cast<clblast::Triangle>(triangle), + static_cast<clblast::Transpose>(a_transpose), + static_cast<clblast::Diagonal>(diagonal), + n, + a_buffer, a_offset, a_ld, + x_buffer, x_offset, x_inc, + queue, event); + return static_cast<StatusCode>(status); +} +StatusCode CLBlastDtrsv(const Layout layout, const Triangle triangle, const Transpose a_transpose, const Diagonal diagonal, + const size_t n, + const cl_mem a_buffer, const size_t a_offset, const size_t a_ld, + cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Trsv<double>(static_cast<clblast::Layout>(layout), + static_cast<clblast::Triangle>(triangle), + static_cast<clblast::Transpose>(a_transpose), + static_cast<clblast::Diagonal>(diagonal), + n, + a_buffer, a_offset, a_ld, + x_buffer, x_offset, x_inc, + queue, event); + return static_cast<StatusCode>(status); +} +StatusCode CLBlastCtrsv(const Layout layout, const Triangle triangle, const Transpose a_transpose, const Diagonal diagonal, + const size_t n, + const cl_mem a_buffer, const size_t a_offset, const size_t a_ld, + cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Trsv<float2>(static_cast<clblast::Layout>(layout), + static_cast<clblast::Triangle>(triangle), + static_cast<clblast::Transpose>(a_transpose), + static_cast<clblast::Diagonal>(diagonal), + n, + a_buffer, a_offset, a_ld, + x_buffer, x_offset, x_inc, + queue, event); + return static_cast<StatusCode>(status); +} +StatusCode CLBlastZtrsv(const Layout layout, const Triangle triangle, const Transpose a_transpose, const Diagonal diagonal, + const size_t n, + const cl_mem a_buffer, const size_t a_offset, const size_t a_ld, + cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Trsv<double2>(static_cast<clblast::Layout>(layout), + static_cast<clblast::Triangle>(triangle), + static_cast<clblast::Transpose>(a_transpose), + static_cast<clblast::Diagonal>(diagonal), + n, + a_buffer, a_offset, a_ld, + x_buffer, x_offset, x_inc, + queue, event); + return static_cast<StatusCode>(status); +} + +// TBSV +StatusCode CLBlastStbsv(const Layout layout, const Triangle triangle, const Transpose a_transpose, const Diagonal diagonal, + const size_t n, const size_t k, + const cl_mem a_buffer, const size_t a_offset, const size_t a_ld, + cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Tbsv<float>(static_cast<clblast::Layout>(layout), + static_cast<clblast::Triangle>(triangle), + static_cast<clblast::Transpose>(a_transpose), + static_cast<clblast::Diagonal>(diagonal), + n, k, + a_buffer, a_offset, a_ld, + x_buffer, x_offset, x_inc, + queue, event); + return static_cast<StatusCode>(status); +} +StatusCode CLBlastDtbsv(const Layout layout, const Triangle triangle, const Transpose a_transpose, const Diagonal diagonal, + const size_t n, const size_t k, + const cl_mem a_buffer, const size_t a_offset, const size_t a_ld, + cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Tbsv<double>(static_cast<clblast::Layout>(layout), + static_cast<clblast::Triangle>(triangle), + static_cast<clblast::Transpose>(a_transpose), + static_cast<clblast::Diagonal>(diagonal), + n, k, + a_buffer, a_offset, a_ld, + x_buffer, x_offset, x_inc, + queue, event); + return static_cast<StatusCode>(status); +} +StatusCode CLBlastCtbsv(const Layout layout, const Triangle triangle, const Transpose a_transpose, const Diagonal diagonal, + const size_t n, const size_t k, + const cl_mem a_buffer, const size_t a_offset, const size_t a_ld, + cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Tbsv<float2>(static_cast<clblast::Layout>(layout), + static_cast<clblast::Triangle>(triangle), + static_cast<clblast::Transpose>(a_transpose), + static_cast<clblast::Diagonal>(diagonal), + n, k, + a_buffer, a_offset, a_ld, + x_buffer, x_offset, x_inc, + queue, event); + return static_cast<StatusCode>(status); +} +StatusCode CLBlastZtbsv(const Layout layout, const Triangle triangle, const Transpose a_transpose, const Diagonal diagonal, + const size_t n, const size_t k, + const cl_mem a_buffer, const size_t a_offset, const size_t a_ld, + cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Tbsv<double2>(static_cast<clblast::Layout>(layout), + static_cast<clblast::Triangle>(triangle), + static_cast<clblast::Transpose>(a_transpose), + static_cast<clblast::Diagonal>(diagonal), + n, k, + a_buffer, a_offset, a_ld, + x_buffer, x_offset, x_inc, + queue, event); + return static_cast<StatusCode>(status); +} + +// TPSV +StatusCode CLBlastStpsv(const Layout layout, const Triangle triangle, const Transpose a_transpose, const Diagonal diagonal, + const size_t n, + const cl_mem ap_buffer, const size_t ap_offset, + cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Tpsv<float>(static_cast<clblast::Layout>(layout), + static_cast<clblast::Triangle>(triangle), + static_cast<clblast::Transpose>(a_transpose), + static_cast<clblast::Diagonal>(diagonal), + n, + ap_buffer, ap_offset, + x_buffer, x_offset, x_inc, + queue, event); + return static_cast<StatusCode>(status); +} +StatusCode CLBlastDtpsv(const Layout layout, const Triangle triangle, const Transpose a_transpose, const Diagonal diagonal, + const size_t n, + const cl_mem ap_buffer, const size_t ap_offset, + cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Tpsv<double>(static_cast<clblast::Layout>(layout), + static_cast<clblast::Triangle>(triangle), + static_cast<clblast::Transpose>(a_transpose), + static_cast<clblast::Diagonal>(diagonal), + n, + ap_buffer, ap_offset, + x_buffer, x_offset, x_inc, + queue, event); + return static_cast<StatusCode>(status); +} +StatusCode CLBlastCtpsv(const Layout layout, const Triangle triangle, const Transpose a_transpose, const Diagonal diagonal, + const size_t n, + const cl_mem ap_buffer, const size_t ap_offset, + cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Tpsv<float2>(static_cast<clblast::Layout>(layout), + static_cast<clblast::Triangle>(triangle), + static_cast<clblast::Transpose>(a_transpose), + static_cast<clblast::Diagonal>(diagonal), + n, + ap_buffer, ap_offset, + x_buffer, x_offset, x_inc, + queue, event); + return static_cast<StatusCode>(status); +} +StatusCode CLBlastZtpsv(const Layout layout, const Triangle triangle, const Transpose a_transpose, const Diagonal diagonal, + const size_t n, + const cl_mem ap_buffer, const size_t ap_offset, + cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Tpsv<double2>(static_cast<clblast::Layout>(layout), + static_cast<clblast::Triangle>(triangle), + static_cast<clblast::Transpose>(a_transpose), + static_cast<clblast::Diagonal>(diagonal), + n, + ap_buffer, ap_offset, + x_buffer, x_offset, x_inc, + queue, event); + return static_cast<StatusCode>(status); +} + +// GER +StatusCode CLBlastSger(const Layout layout, + const size_t m, const size_t n, + const float alpha, + const cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + const cl_mem y_buffer, const size_t y_offset, const size_t y_inc, + cl_mem a_buffer, const size_t a_offset, const size_t a_ld, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Ger(static_cast<clblast::Layout>(layout), + m, n, + alpha, + x_buffer, x_offset, x_inc, + y_buffer, y_offset, y_inc, + a_buffer, a_offset, a_ld, + queue, event); + return static_cast<StatusCode>(status); +} +StatusCode CLBlastDger(const Layout layout, + const size_t m, const size_t n, + const double alpha, + const cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + const cl_mem y_buffer, const size_t y_offset, const size_t y_inc, + cl_mem a_buffer, const size_t a_offset, const size_t a_ld, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Ger(static_cast<clblast::Layout>(layout), + m, n, + alpha, + x_buffer, x_offset, x_inc, + y_buffer, y_offset, y_inc, + a_buffer, a_offset, a_ld, + queue, event); + return static_cast<StatusCode>(status); +} + +// GERU +StatusCode CLBlastCgeru(const Layout layout, + const size_t m, const size_t n, + const cl_float2 alpha, + const cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + const cl_mem y_buffer, const size_t y_offset, const size_t y_inc, + cl_mem a_buffer, const size_t a_offset, const size_t a_ld, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Geru(static_cast<clblast::Layout>(layout), + m, n, + float2{alpha.s[0], alpha.s[1]}, + x_buffer, x_offset, x_inc, + y_buffer, y_offset, y_inc, + a_buffer, a_offset, a_ld, + queue, event); + return static_cast<StatusCode>(status); +} +StatusCode CLBlastZgeru(const Layout layout, + const size_t m, const size_t n, + const cl_double2 alpha, + const cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + const cl_mem y_buffer, const size_t y_offset, const size_t y_inc, + cl_mem a_buffer, const size_t a_offset, const size_t a_ld, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Geru(static_cast<clblast::Layout>(layout), + m, n, + double2{alpha.s[0], alpha.s[1]}, + x_buffer, x_offset, x_inc, + y_buffer, y_offset, y_inc, + a_buffer, a_offset, a_ld, + queue, event); + return static_cast<StatusCode>(status); +} + +// GERC +StatusCode CLBlastCgerc(const Layout layout, + const size_t m, const size_t n, + const cl_float2 alpha, + const cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + const cl_mem y_buffer, const size_t y_offset, const size_t y_inc, + cl_mem a_buffer, const size_t a_offset, const size_t a_ld, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Gerc(static_cast<clblast::Layout>(layout), + m, n, + float2{alpha.s[0], alpha.s[1]}, + x_buffer, x_offset, x_inc, + y_buffer, y_offset, y_inc, + a_buffer, a_offset, a_ld, + queue, event); + return static_cast<StatusCode>(status); +} +StatusCode CLBlastZgerc(const Layout layout, + const size_t m, const size_t n, + const cl_double2 alpha, + const cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + const cl_mem y_buffer, const size_t y_offset, const size_t y_inc, + cl_mem a_buffer, const size_t a_offset, const size_t a_ld, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Gerc(static_cast<clblast::Layout>(layout), + m, n, + double2{alpha.s[0], alpha.s[1]}, + x_buffer, x_offset, x_inc, + y_buffer, y_offset, y_inc, + a_buffer, a_offset, a_ld, + queue, event); + return static_cast<StatusCode>(status); +} + +// HER +StatusCode CLBlastCher(const Layout layout, const Triangle triangle, + const size_t n, + const float alpha, + const cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + cl_mem a_buffer, const size_t a_offset, const size_t a_ld, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Her(static_cast<clblast::Layout>(layout), + static_cast<clblast::Triangle>(triangle), + n, + alpha, + x_buffer, x_offset, x_inc, + a_buffer, a_offset, a_ld, + queue, event); + return static_cast<StatusCode>(status); +} +StatusCode CLBlastZher(const Layout layout, const Triangle triangle, + const size_t n, + const double alpha, + const cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + cl_mem a_buffer, const size_t a_offset, const size_t a_ld, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Her(static_cast<clblast::Layout>(layout), + static_cast<clblast::Triangle>(triangle), + n, + alpha, + x_buffer, x_offset, x_inc, + a_buffer, a_offset, a_ld, + queue, event); + return static_cast<StatusCode>(status); +} + +// HPR +StatusCode CLBlastChpr(const Layout layout, const Triangle triangle, + const size_t n, + const float alpha, + const cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + cl_mem ap_buffer, const size_t ap_offset, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Hpr(static_cast<clblast::Layout>(layout), + static_cast<clblast::Triangle>(triangle), + n, + alpha, + x_buffer, x_offset, x_inc, + ap_buffer, ap_offset, + queue, event); + return static_cast<StatusCode>(status); +} +StatusCode CLBlastZhpr(const Layout layout, const Triangle triangle, + const size_t n, + const double alpha, + const cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + cl_mem ap_buffer, const size_t ap_offset, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Hpr(static_cast<clblast::Layout>(layout), + static_cast<clblast::Triangle>(triangle), + n, + alpha, + x_buffer, x_offset, x_inc, + ap_buffer, ap_offset, + queue, event); + return static_cast<StatusCode>(status); +} + +// HER2 +StatusCode CLBlastCher2(const Layout layout, const Triangle triangle, + const size_t n, + const cl_float2 alpha, + const cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + const cl_mem y_buffer, const size_t y_offset, const size_t y_inc, + cl_mem a_buffer, const size_t a_offset, const size_t a_ld, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Her2(static_cast<clblast::Layout>(layout), + static_cast<clblast::Triangle>(triangle), + n, + float2{alpha.s[0], alpha.s[1]}, + x_buffer, x_offset, x_inc, + y_buffer, y_offset, y_inc, + a_buffer, a_offset, a_ld, + queue, event); + return static_cast<StatusCode>(status); +} +StatusCode CLBlastZher2(const Layout layout, const Triangle triangle, + const size_t n, + const cl_double2 alpha, + const cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + const cl_mem y_buffer, const size_t y_offset, const size_t y_inc, + cl_mem a_buffer, const size_t a_offset, const size_t a_ld, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Her2(static_cast<clblast::Layout>(layout), + static_cast<clblast::Triangle>(triangle), + n, + double2{alpha.s[0], alpha.s[1]}, + x_buffer, x_offset, x_inc, + y_buffer, y_offset, y_inc, + a_buffer, a_offset, a_ld, + queue, event); + return static_cast<StatusCode>(status); +} + +// HPR2 +StatusCode CLBlastChpr2(const Layout layout, const Triangle triangle, + const size_t n, + const cl_float2 alpha, + const cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + const cl_mem y_buffer, const size_t y_offset, const size_t y_inc, + cl_mem ap_buffer, const size_t ap_offset, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Hpr2(static_cast<clblast::Layout>(layout), + static_cast<clblast::Triangle>(triangle), + n, + float2{alpha.s[0], alpha.s[1]}, + x_buffer, x_offset, x_inc, + y_buffer, y_offset, y_inc, + ap_buffer, ap_offset, + queue, event); + return static_cast<StatusCode>(status); +} +StatusCode CLBlastZhpr2(const Layout layout, const Triangle triangle, + const size_t n, + const cl_double2 alpha, + const cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + const cl_mem y_buffer, const size_t y_offset, const size_t y_inc, + cl_mem ap_buffer, const size_t ap_offset, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Hpr2(static_cast<clblast::Layout>(layout), + static_cast<clblast::Triangle>(triangle), + n, + double2{alpha.s[0], alpha.s[1]}, + x_buffer, x_offset, x_inc, + y_buffer, y_offset, y_inc, + ap_buffer, ap_offset, + queue, event); + return static_cast<StatusCode>(status); +} + +// SYR +StatusCode CLBlastSsyr(const Layout layout, const Triangle triangle, + const size_t n, + const float alpha, + const cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + cl_mem a_buffer, const size_t a_offset, const size_t a_ld, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Syr(static_cast<clblast::Layout>(layout), + static_cast<clblast::Triangle>(triangle), + n, + alpha, + x_buffer, x_offset, x_inc, + a_buffer, a_offset, a_ld, + queue, event); + return static_cast<StatusCode>(status); +} +StatusCode CLBlastDsyr(const Layout layout, const Triangle triangle, + const size_t n, + const double alpha, + const cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + cl_mem a_buffer, const size_t a_offset, const size_t a_ld, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Syr(static_cast<clblast::Layout>(layout), + static_cast<clblast::Triangle>(triangle), + n, + alpha, + x_buffer, x_offset, x_inc, + a_buffer, a_offset, a_ld, + queue, event); + return static_cast<StatusCode>(status); +} + +// SPR +StatusCode CLBlastSspr(const Layout layout, const Triangle triangle, + const size_t n, + const float alpha, + const cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + cl_mem ap_buffer, const size_t ap_offset, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Spr(static_cast<clblast::Layout>(layout), + static_cast<clblast::Triangle>(triangle), + n, + alpha, + x_buffer, x_offset, x_inc, + ap_buffer, ap_offset, + queue, event); + return static_cast<StatusCode>(status); +} +StatusCode CLBlastDspr(const Layout layout, const Triangle triangle, + const size_t n, + const double alpha, + const cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + cl_mem ap_buffer, const size_t ap_offset, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Spr(static_cast<clblast::Layout>(layout), + static_cast<clblast::Triangle>(triangle), + n, + alpha, + x_buffer, x_offset, x_inc, + ap_buffer, ap_offset, + queue, event); + return static_cast<StatusCode>(status); +} + +// SYR2 +StatusCode CLBlastSsyr2(const Layout layout, const Triangle triangle, + const size_t n, + const float alpha, + const cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + const cl_mem y_buffer, const size_t y_offset, const size_t y_inc, + cl_mem a_buffer, const size_t a_offset, const size_t a_ld, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Syr2(static_cast<clblast::Layout>(layout), + static_cast<clblast::Triangle>(triangle), + n, + alpha, + x_buffer, x_offset, x_inc, + y_buffer, y_offset, y_inc, + a_buffer, a_offset, a_ld, + queue, event); + return static_cast<StatusCode>(status); +} +StatusCode CLBlastDsyr2(const Layout layout, const Triangle triangle, + const size_t n, + const double alpha, + const cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + const cl_mem y_buffer, const size_t y_offset, const size_t y_inc, + cl_mem a_buffer, const size_t a_offset, const size_t a_ld, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Syr2(static_cast<clblast::Layout>(layout), + static_cast<clblast::Triangle>(triangle), + n, + alpha, + x_buffer, x_offset, x_inc, + y_buffer, y_offset, y_inc, + a_buffer, a_offset, a_ld, + queue, event); + return static_cast<StatusCode>(status); +} + +// SPR2 +StatusCode CLBlastSspr2(const Layout layout, const Triangle triangle, + const size_t n, + const float alpha, + const cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + const cl_mem y_buffer, const size_t y_offset, const size_t y_inc, + cl_mem ap_buffer, const size_t ap_offset, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Spr2(static_cast<clblast::Layout>(layout), + static_cast<clblast::Triangle>(triangle), + n, + alpha, + x_buffer, x_offset, x_inc, + y_buffer, y_offset, y_inc, + ap_buffer, ap_offset, + queue, event); + return static_cast<StatusCode>(status); +} +StatusCode CLBlastDspr2(const Layout layout, const Triangle triangle, + const size_t n, + const double alpha, + const cl_mem x_buffer, const size_t x_offset, const size_t x_inc, + const cl_mem y_buffer, const size_t y_offset, const size_t y_inc, + cl_mem ap_buffer, const size_t ap_offset, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Spr2(static_cast<clblast::Layout>(layout), + static_cast<clblast::Triangle>(triangle), + n, + alpha, + x_buffer, x_offset, x_inc, + y_buffer, y_offset, y_inc, + ap_buffer, ap_offset, + queue, event); + return static_cast<StatusCode>(status); +} + // ================================================================================================= // BLAS level-3 (matrix-matrix) routines // ================================================================================================= @@ -963,4 +1947,78 @@ StatusCode CLBlastZtrmm(const Layout layout, const Side side, const Triangle tri return static_cast<StatusCode>(status); } +// TRSM +StatusCode CLBlastStrsm(const Layout layout, const Side side, const Triangle triangle, const Transpose a_transpose, const Diagonal diagonal, + const size_t m, const size_t n, + const float alpha, + const cl_mem a_buffer, const size_t a_offset, const size_t a_ld, + cl_mem b_buffer, const size_t b_offset, const size_t b_ld, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Trsm(static_cast<clblast::Layout>(layout), + static_cast<clblast::Side>(side), + static_cast<clblast::Triangle>(triangle), + static_cast<clblast::Transpose>(a_transpose), + static_cast<clblast::Diagonal>(diagonal), + m, n, + alpha, + a_buffer, a_offset, a_ld, + b_buffer, b_offset, b_ld, + queue, event); + return static_cast<StatusCode>(status); +} +StatusCode CLBlastDtrsm(const Layout layout, const Side side, const Triangle triangle, const Transpose a_transpose, const Diagonal diagonal, + const size_t m, const size_t n, + const double alpha, + const cl_mem a_buffer, const size_t a_offset, const size_t a_ld, + cl_mem b_buffer, const size_t b_offset, const size_t b_ld, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Trsm(static_cast<clblast::Layout>(layout), + static_cast<clblast::Side>(side), + static_cast<clblast::Triangle>(triangle), + static_cast<clblast::Transpose>(a_transpose), + static_cast<clblast::Diagonal>(diagonal), + m, n, + alpha, + a_buffer, a_offset, a_ld, + b_buffer, b_offset, b_ld, + queue, event); + return static_cast<StatusCode>(status); +} +StatusCode CLBlastCtrsm(const Layout layout, const Side side, const Triangle triangle, const Transpose a_transpose, const Diagonal diagonal, + const size_t m, const size_t n, + const cl_float2 alpha, + const cl_mem a_buffer, const size_t a_offset, const size_t a_ld, + cl_mem b_buffer, const size_t b_offset, const size_t b_ld, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Trsm(static_cast<clblast::Layout>(layout), + static_cast<clblast::Side>(side), + static_cast<clblast::Triangle>(triangle), + static_cast<clblast::Transpose>(a_transpose), + static_cast<clblast::Diagonal>(diagonal), + m, n, + float2{alpha.s[0], alpha.s[1]}, + a_buffer, a_offset, a_ld, + b_buffer, b_offset, b_ld, + queue, event); + return static_cast<StatusCode>(status); +} +StatusCode CLBlastZtrsm(const Layout layout, const Side side, const Triangle triangle, const Transpose a_transpose, const Diagonal diagonal, + const size_t m, const size_t n, + const cl_double2 alpha, + const cl_mem a_buffer, const size_t a_offset, const size_t a_ld, + cl_mem b_buffer, const size_t b_offset, const size_t b_ld, + cl_command_queue* queue, cl_event* event) { + auto status = clblast::Trsm(static_cast<clblast::Layout>(layout), + static_cast<clblast::Side>(side), + static_cast<clblast::Triangle>(triangle), + static_cast<clblast::Transpose>(a_transpose), + static_cast<clblast::Diagonal>(diagonal), + m, n, + double2{alpha.s[0], alpha.s[1]}, + a_buffer, a_offset, a_ld, + b_buffer, b_offset, b_ld, + queue, event); + return static_cast<StatusCode>(status); +} + // ================================================================================================= |