gpu Module Reference

Simple GPU - Fortran GPU Computing Library. More...

Data Types
interface	gpu_allocate
	Allocate GPU/CPU memory for arrays. More...
interface	gpu_allocate_c
	Allocate GPU/CPU memory (C binding) More...
type	gpu_blas
	Handle for BLAS operations. More...
interface	gpu_blas_create_c
interface	gpu_blas_destroy_c
interface	gpu_copy
	Copy data between GPU arrays. More...
interface	gpu_copy_c
interface	gpu_ddot_c
interface	gpu_deallocate
	Free GPU/CPU memory. More...
interface	gpu_deallocate_c
	Free GPU/CPU memory (C binding) More...
interface	gpu_dgeam_c
interface	gpu_dgemm_c
interface	gpu_dgemv_c
type	gpu_double1
	1-dimensional array of double precision values More...
type	gpu_double2
	2-dimensional array of double precision values More...
type	gpu_double3
	3-dimensional array of double precision values More...
type	gpu_double4
	4-dimensional array of double precision values More...
type	gpu_double5
	5-dimensional array of double precision values More...
type	gpu_double6
	6-dimensional array of double precision values More...
interface	gpu_download
	Download data from device (GPU) to host (CPU) More...
interface	gpu_download_c
interface	gpu_get_memory
	Query GPU memory usage (C binding) More...
interface	gpu_ndevices
	Get number of GPU devices (C binding) More...
type	gpu_real1
	1-dimensional array of single precision values More...
type	gpu_real2
	2-dimensional array of single precision values More...
type	gpu_real3
	3-dimensional array of single precision values More...
type	gpu_real4
	4-dimensional array of single precision values More...
type	gpu_real5
	5-dimensional array of single precision values More...
type	gpu_real6
	6-dimensional array of single precision values More...
interface	gpu_sdot_c
interface	gpu_set_device
	Set active GPU device (C binding) More...
interface	gpu_set_stream_c
interface	gpu_sgeam_c
interface	gpu_sgemm_c
interface	gpu_sgemv_c
type	gpu_stream
	Handle for CUDA streams. More...
interface	gpu_stream_create_c
interface	gpu_stream_destroy_c
interface	gpu_stream_synchronize
interface	gpu_synchronize
interface	gpu_upload
	Upload data from host (CPU) to device (GPU) More...
interface	gpu_upload_c
	Upload data to GPU (C binding) More...

Functions/Subroutines
subroutine	gpu_allocate_double1 (ptr, s)
	Allocate 1D double precision array (32-bit dimensions)
subroutine	gpu_allocate_double2 (ptr, s1, s2)
subroutine	gpu_allocate_double3 (ptr, s1, s2, s3)
subroutine	gpu_allocate_double4 (ptr, s1, s2, s3, s4)
subroutine	gpu_allocate_double5 (ptr, s1, s2, s3, s4, s5)
subroutine	gpu_allocate_double6 (ptr, s1, s2, s3, s4, s5, s6)
subroutine	gpu_allocate_double1_64 (ptr, s)
subroutine	gpu_allocate_double2_64 (ptr, s1, s2)
subroutine	gpu_allocate_double3_64 (ptr, s1, s2, s3)
subroutine	gpu_allocate_double4_64 (ptr, s1, s2, s3, s4)
subroutine	gpu_allocate_double5_64 (ptr, s1, s2, s3, s4, s5)
subroutine	gpu_allocate_double6_64 (ptr, s1, s2, s3, s4, s5, s6)
subroutine	gpu_allocate_real1 (ptr, s)
subroutine	gpu_allocate_real2 (ptr, s1, s2)
subroutine	gpu_allocate_real3 (ptr, s1, s2, s3)
subroutine	gpu_allocate_real4 (ptr, s1, s2, s3, s4)
subroutine	gpu_allocate_real5 (ptr, s1, s2, s3, s4, s5)
subroutine	gpu_allocate_real6 (ptr, s1, s2, s3, s4, s5, s6)
subroutine	gpu_allocate_real1_64 (ptr, s)
subroutine	gpu_allocate_real2_64 (ptr, s1, s2)
subroutine	gpu_allocate_real3_64 (ptr, s1, s2, s3)
subroutine	gpu_allocate_real4_64 (ptr, s1, s2, s3, s4)
subroutine	gpu_allocate_real5_64 (ptr, s1, s2, s3, s4, s5)
subroutine	gpu_allocate_real6_64 (ptr, s1, s2, s3, s4, s5, s6)
subroutine	gpu_deallocate_double1 (ptr)
subroutine	gpu_deallocate_double2 (ptr)
subroutine	gpu_deallocate_double3 (ptr)
subroutine	gpu_deallocate_double4 (ptr)
subroutine	gpu_deallocate_double5 (ptr)
subroutine	gpu_deallocate_double6 (ptr)
subroutine	gpu_deallocate_real1 (ptr)
subroutine	gpu_deallocate_real2 (ptr)
subroutine	gpu_deallocate_real3 (ptr)
subroutine	gpu_deallocate_real4 (ptr)
subroutine	gpu_deallocate_real5 (ptr)
subroutine	gpu_deallocate_real6 (ptr)
subroutine	gpu_upload_double0 (cpu_ptr, gpu_ptr, n)
subroutine	gpu_upload_double1 (cpu_ptr, gpu_ptr)
subroutine	gpu_upload_double2 (cpu_ptr, gpu_ptr)
subroutine	gpu_upload_double3 (cpu_ptr, gpu_ptr)
subroutine	gpu_upload_double4 (cpu_ptr, gpu_ptr)
subroutine	gpu_upload_double5 (cpu_ptr, gpu_ptr)
subroutine	gpu_upload_double6 (cpu_ptr, gpu_ptr)
subroutine	gpu_upload_real0 (cpu_ptr, gpu_ptr, n)
subroutine	gpu_upload_real1 (cpu_ptr, gpu_ptr)
subroutine	gpu_upload_real2 (cpu_ptr, gpu_ptr)
subroutine	gpu_upload_real3 (cpu_ptr, gpu_ptr)
subroutine	gpu_upload_real4 (cpu_ptr, gpu_ptr)
subroutine	gpu_upload_real5 (cpu_ptr, gpu_ptr)
subroutine	gpu_upload_real6 (cpu_ptr, gpu_ptr)
subroutine	gpu_download_double0 (gpu_ptr, cpu_ptr, n)
subroutine	gpu_download_double1 (gpu_ptr, cpu_ptr)
subroutine	gpu_download_double2 (gpu_ptr, cpu_ptr)
subroutine	gpu_download_double3 (gpu_ptr, cpu_ptr)
subroutine	gpu_download_double4 (gpu_ptr, cpu_ptr)
subroutine	gpu_download_double5 (gpu_ptr, cpu_ptr)
subroutine	gpu_download_double6 (gpu_ptr, cpu_ptr)
subroutine	gpu_download_real0 (gpu_ptr, cpu_ptr, n)
subroutine	gpu_download_real1 (gpu_ptr, cpu_ptr)
subroutine	gpu_download_real2 (gpu_ptr, cpu_ptr)
subroutine	gpu_download_real3 (gpu_ptr, cpu_ptr)
subroutine	gpu_download_real4 (gpu_ptr, cpu_ptr)
subroutine	gpu_download_real5 (gpu_ptr, cpu_ptr)
subroutine	gpu_download_real6 (gpu_ptr, cpu_ptr)
subroutine	gpu_copy_double0 (gpu_ptr_src, gpu_ptr_dest, n)
subroutine	gpu_copy_double1 (gpu_ptr_src, gpu_ptr_dest)
subroutine	gpu_copy_double2 (gpu_ptr_src, gpu_ptr_dest)
subroutine	gpu_copy_double3 (gpu_ptr_src, gpu_ptr_dest)
subroutine	gpu_copy_double4 (gpu_ptr_src, gpu_ptr_dest)
subroutine	gpu_copy_double5 (gpu_ptr_src, gpu_ptr_dest)
subroutine	gpu_copy_double6 (gpu_ptr_src, gpu_ptr_dest)
subroutine	gpu_copy_real0 (gpu_ptr_src, gpu_ptr_dest, n)
subroutine	gpu_copy_real1 (gpu_ptr_src, gpu_ptr_dest)
subroutine	gpu_copy_real2 (gpu_ptr_src, gpu_ptr_dest)
subroutine	gpu_copy_real3 (gpu_ptr_src, gpu_ptr_dest)
subroutine	gpu_copy_real4 (gpu_ptr_src, gpu_ptr_dest)
subroutine	gpu_copy_real5 (gpu_ptr_src, gpu_ptr_dest)
subroutine	gpu_copy_real6 (gpu_ptr_src, gpu_ptr_dest)
subroutine	gpu_stream_create (stream)
subroutine	gpu_stream_destroy (stream)
subroutine	gpu_set_stream (handle, stream)
subroutine	gpu_blas_create (handle)
	Create a BLAS handle.
subroutine	gpu_blas_destroy (handle)
	Destroy a BLAS handle.
subroutine	gpu_ddot (handle, n, dx, incx, dy, incy, res)
	Double precision dot product (32-bit dimensions)
subroutine	gpu_ddot_64 (handle, n, dx, incx, dy, incy, res)
	Double precision dot product (64-bit dimensions)
subroutine	gpu_sdot (handle, n, dx, incx, dy, incy, res)
subroutine	gpu_sdot_64 (handle, n, dx, incx, dy, incy, res)
subroutine	gpu_dgeam (handle, transa, transb, m, n, alpha, a, lda, beta, b, ldb, c, ldc)
subroutine	gpu_dgeam_64 (handle, transa, transb, m, n, alpha, a, lda, beta, b, ldb, c, ldc)
subroutine	gpu_sgeam (handle, transa, transb, m, n, alpha, a, lda, beta, b, ldb, c, ldc)
subroutine	gpu_sgeam_64 (handle, transa, transb, m, n, alpha, a, lda, beta, b, ldb, c, ldc)
subroutine	gpu_dgemv (handle, transa, m, n, alpha, a, lda, x, incx, beta, y, incy)
subroutine	gpu_dgemv_64 (handle, transa, m, n, alpha, a, lda, x, incx, beta, y, incy)
subroutine	gpu_sgemv (handle, transa, m, n, alpha, a, lda, x, incx, beta, y, incy)
subroutine	gpu_sgemv_64 (handle, transa, m, n, alpha, a, lda, x, incx, beta, y, incy)
subroutine	gpu_dgemm (handle, transa, transb, m, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
subroutine	gpu_dgemm_64 (handle, transa, transb, m, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
subroutine	gpu_sgemm (handle, transa, transb, m, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
subroutine	gpu_sgemm_64 (handle, transa, transb, m, n, k, alpha, a, lda, b, ldb, beta, c, ldc)

Detailed Description

Simple GPU - Fortran GPU Computing Library.

This module provides a Fortran interface for GPU computing with transparent support for both CPU (using standard BLAS) and NVIDIA GPU (using cuBLAS).

The library provides:

• GPU memory management (allocate, free, upload, download, copy)
• BLAS operations (Level 1, 2, and 3) for single and double precision
• Support for multidimensional arrays (1D through 6D)
• CUDA stream management for asynchronous operations
• Transparent API that works with both CPU and GPU backends

Author

Anthony Scemama

Date

2026

Function/Subroutine Documentation

gpu_allocate_double1()

subroutine gpu::gpu_allocate_double1	(	type(gpu_double1), intent(inout)	ptr,
		integer, intent(in)	s
	)

Allocate 1D double precision array (32-bit dimensions)

Definition at line 530 of file simple_gpu.F90.

gpu_allocate_double1_64()

subroutine gpu::gpu_allocate_double1_64	(	type(gpu_double1), intent(inout)	ptr,
		integer*8, intent(in)	s
	)

Definition at line 624 of file simple_gpu.F90.

gpu_allocate_double2()

subroutine gpu::gpu_allocate_double2	(	type(gpu_double2), intent(inout)	ptr,
		integer, intent(in)	s1,
		integer, intent(in)	s2
	)

Definition at line 543 of file simple_gpu.F90.

gpu_allocate_double2_64()

subroutine gpu::gpu_allocate_double2_64	(	type(gpu_double2), intent(inout)	ptr,
		integer*8, intent(in)	s1,
		integer*8, intent(in)	s2
	)

Definition at line 633 of file simple_gpu.F90.

gpu_allocate_double3()

subroutine gpu::gpu_allocate_double3	(	type(gpu_double3), intent(inout)	ptr,
		integer, intent(in)	s1,
		integer, intent(in)	s2,
		integer, intent(in)	s3
	)

Definition at line 557 of file simple_gpu.F90.

gpu_allocate_double3_64()

subroutine gpu::gpu_allocate_double3_64	(	type(gpu_double3), intent(inout)	ptr,
		integer*8, intent(in)	s1,
		integer*8, intent(in)	s2,
		integer*8, intent(in)	s3
	)

Definition at line 642 of file simple_gpu.F90.

gpu_allocate_double4()

subroutine gpu::gpu_allocate_double4	(	type(gpu_double4), intent(inout)	ptr,
		integer, intent(in)	s1,
		integer, intent(in)	s2,
		integer, intent(in)	s3,
		integer, intent(in)	s4
	)

Definition at line 572 of file simple_gpu.F90.

gpu_allocate_double4_64()

subroutine gpu::gpu_allocate_double4_64	(	type(gpu_double4), intent(inout)	ptr,
		integer*8, intent(in)	s1,
		integer*8, intent(in)	s2,
		integer*8, intent(in)	s3,
		integer*8, intent(in)	s4
	)

Definition at line 651 of file simple_gpu.F90.

gpu_allocate_double5()

subroutine gpu::gpu_allocate_double5	(	type(gpu_double5), intent(inout)	ptr,
		integer, intent(in)	s1,
		integer, intent(in)	s2,
		integer, intent(in)	s3,
		integer, intent(in)	s4,
		integer, intent(in)	s5
	)

Definition at line 588 of file simple_gpu.F90.

gpu_allocate_double5_64()

subroutine gpu::gpu_allocate_double5_64	(	type(gpu_double5), intent(inout)	ptr,
		integer*8, intent(in)	s1,
		integer*8, intent(in)	s2,
		integer*8, intent(in)	s3,
		integer*8, intent(in)	s4,
		integer*8, intent(in)	s5
	)

Definition at line 660 of file simple_gpu.F90.

gpu_allocate_double6()

subroutine gpu::gpu_allocate_double6	(	type(gpu_double6), intent(inout)	ptr,
		integer, intent(in)	s1,
		integer, intent(in)	s2,
		integer, intent(in)	s3,
		integer, intent(in)	s4,
		integer, intent(in)	s5,
		integer, intent(in)	s6
	)

Definition at line 605 of file simple_gpu.F90.

gpu_allocate_double6_64()

subroutine gpu::gpu_allocate_double6_64	(	type(gpu_double6), intent(inout)	ptr,
		integer*8, intent(in)	s1,
		integer*8, intent(in)	s2,
		integer*8, intent(in)	s3,
		integer*8, intent(in)	s4,
		integer*8, intent(in)	s5,
		integer*8, intent(in)	s6
	)

Definition at line 669 of file simple_gpu.F90.

gpu_allocate_real1()

subroutine gpu::gpu_allocate_real1	(	type(gpu_real1), intent(inout)	ptr,
		integer, intent(in)	s
	)

Definition at line 678 of file simple_gpu.F90.

gpu_allocate_real1_64()

subroutine gpu::gpu_allocate_real1_64	(	type(gpu_real1), intent(inout)	ptr,
		integer*8, intent(in)	s
	)

Definition at line 772 of file simple_gpu.F90.

gpu_allocate_real2()

subroutine gpu::gpu_allocate_real2	(	type(gpu_real2), intent(inout)	ptr,
		integer, intent(in)	s1,
		integer, intent(in)	s2
	)

Definition at line 691 of file simple_gpu.F90.

gpu_allocate_real2_64()

subroutine gpu::gpu_allocate_real2_64	(	type(gpu_real2), intent(inout)	ptr,
		integer*8, intent(in)	s1,
		integer*8, intent(in)	s2
	)

Definition at line 781 of file simple_gpu.F90.

gpu_allocate_real3()

subroutine gpu::gpu_allocate_real3	(	type(gpu_real3), intent(inout)	ptr,
		integer, intent(in)	s1,
		integer, intent(in)	s2,
		integer, intent(in)	s3
	)

Definition at line 705 of file simple_gpu.F90.

gpu_allocate_real3_64()

subroutine gpu::gpu_allocate_real3_64	(	type(gpu_real3), intent(inout)	ptr,
		integer*8, intent(in)	s1,
		integer*8, intent(in)	s2,
		integer*8, intent(in)	s3
	)

Definition at line 790 of file simple_gpu.F90.

gpu_allocate_real4()

subroutine gpu::gpu_allocate_real4	(	type(gpu_real4), intent(inout)	ptr,
		integer, intent(in)	s1,
		integer, intent(in)	s2,
		integer, intent(in)	s3,
		integer, intent(in)	s4
	)

Definition at line 720 of file simple_gpu.F90.

gpu_allocate_real4_64()

subroutine gpu::gpu_allocate_real4_64	(	type(gpu_real4), intent(inout)	ptr,
		integer*8, intent(in)	s1,
		integer*8, intent(in)	s2,
		integer*8, intent(in)	s3,
		integer*8, intent(in)	s4
	)

Definition at line 799 of file simple_gpu.F90.

gpu_allocate_real5()

subroutine gpu::gpu_allocate_real5	(	type(gpu_real5), intent(inout)	ptr,
		integer, intent(in)	s1,
		integer, intent(in)	s2,
		integer, intent(in)	s3,
		integer, intent(in)	s4,
		integer, intent(in)	s5
	)

Definition at line 736 of file simple_gpu.F90.

gpu_allocate_real5_64()

subroutine gpu::gpu_allocate_real5_64	(	type(gpu_real5), intent(inout)	ptr,
		integer*8, intent(in)	s1,
		integer*8, intent(in)	s2,
		integer*8, intent(in)	s3,
		integer*8, intent(in)	s4,
		integer*8, intent(in)	s5
	)

Definition at line 808 of file simple_gpu.F90.

gpu_allocate_real6()

subroutine gpu::gpu_allocate_real6	(	type(gpu_real6), intent(inout)	ptr,
		integer, intent(in)	s1,
		integer, intent(in)	s2,
		integer, intent(in)	s3,
		integer, intent(in)	s4,
		integer, intent(in)	s5,
		integer, intent(in)	s6
	)

Definition at line 753 of file simple_gpu.F90.

gpu_allocate_real6_64()

subroutine gpu::gpu_allocate_real6_64	(	type(gpu_real6), intent(inout)	ptr,
		integer*8, intent(in)	s1,
		integer*8, intent(in)	s2,
		integer*8, intent(in)	s3,
		integer*8, intent(in)	s4,
		integer*8, intent(in)	s5,
		integer*8, intent(in)	s6
	)

Definition at line 817 of file simple_gpu.F90.

gpu_blas_create()

subroutine gpu::gpu_blas_create

(

type(gpu_blas)

handle

)

Create a BLAS handle.

Initializes a BLAS handle for performing BLAS operations. Must be called before any BLAS functions. The handle should be destroyed with gpu_blas_destroy when no longer needed.

Parameters

[out]

handle

BLAS handle to create

Definition at line 1258 of file simple_gpu.F90.

gpu_blas_destroy()

subroutine gpu::gpu_blas_destroy

(

type(gpu_blas)

handle

)

Destroy a BLAS handle.

Frees resources associated with a BLAS handle. The handle should not be used after calling this.

Parameters

[in,out]

handle

BLAS handle to destroy

Definition at line 1269 of file simple_gpu.F90.

gpu_copy_double0()

subroutine gpu::gpu_copy_double0	(	double precision, intent(in), target	gpu_ptr_src,
		double precision, intent(in), target	gpu_ptr_dest,
		integer, intent(in)	n
	)

Definition at line 1126 of file simple_gpu.F90.

gpu_copy_double1()

subroutine gpu::gpu_copy_double1	(	type(gpu_double1), intent(in)	gpu_ptr_src,
		type(gpu_double1), intent(in)	gpu_ptr_dest
	)

Definition at line 1134 of file simple_gpu.F90.

gpu_copy_double2()

subroutine gpu::gpu_copy_double2	(	type(gpu_double2), intent(in)	gpu_ptr_src,
		type(gpu_double2), intent(in)	gpu_ptr_dest
	)

Definition at line 1141 of file simple_gpu.F90.

gpu_copy_double3()

subroutine gpu::gpu_copy_double3	(	type(gpu_double3), intent(in)	gpu_ptr_src,
		type(gpu_double3), intent(in)	gpu_ptr_dest
	)

Definition at line 1148 of file simple_gpu.F90.

gpu_copy_double4()

subroutine gpu::gpu_copy_double4	(	type(gpu_double4), intent(in)	gpu_ptr_src,
		type(gpu_double4), intent(in)	gpu_ptr_dest
	)

Definition at line 1155 of file simple_gpu.F90.

gpu_copy_double5()

subroutine gpu::gpu_copy_double5	(	type(gpu_double5), intent(in)	gpu_ptr_src,
		type(gpu_double5), intent(in)	gpu_ptr_dest
	)

Definition at line 1162 of file simple_gpu.F90.

gpu_copy_double6()

subroutine gpu::gpu_copy_double6	(	type(gpu_double6), intent(in)	gpu_ptr_src,
		type(gpu_double6), intent(in)	gpu_ptr_dest
	)

Definition at line 1169 of file simple_gpu.F90.

gpu_copy_real0()

subroutine gpu::gpu_copy_real0	(	real, intent(in), target	gpu_ptr_src,
		real, intent(in), target	gpu_ptr_dest,
		integer, intent(in)	n
	)

Definition at line 1176 of file simple_gpu.F90.

gpu_copy_real1()

subroutine gpu::gpu_copy_real1	(	type(gpu_real1), intent(in)	gpu_ptr_src,
		type(gpu_real1), intent(in)	gpu_ptr_dest
	)

Definition at line 1184 of file simple_gpu.F90.

gpu_copy_real2()

subroutine gpu::gpu_copy_real2	(	type(gpu_real2), intent(in)	gpu_ptr_src,
		type(gpu_real2), intent(in)	gpu_ptr_dest
	)

Definition at line 1191 of file simple_gpu.F90.

gpu_copy_real3()

subroutine gpu::gpu_copy_real3	(	type(gpu_real3), intent(in)	gpu_ptr_src,
		type(gpu_real3), intent(in)	gpu_ptr_dest
	)

Definition at line 1198 of file simple_gpu.F90.

gpu_copy_real4()

subroutine gpu::gpu_copy_real4	(	type(gpu_real4), intent(in)	gpu_ptr_src,
		type(gpu_real4), intent(in)	gpu_ptr_dest
	)

Definition at line 1205 of file simple_gpu.F90.

gpu_copy_real5()

subroutine gpu::gpu_copy_real5	(	type(gpu_real5), intent(in)	gpu_ptr_src,
		type(gpu_real5), intent(in)	gpu_ptr_dest
	)

Definition at line 1212 of file simple_gpu.F90.

gpu_copy_real6()

subroutine gpu::gpu_copy_real6	(	type(gpu_real6), intent(in)	gpu_ptr_src,
		type(gpu_real6), intent(in)	gpu_ptr_dest
	)

Definition at line 1219 of file simple_gpu.F90.

gpu_ddot()

subroutine gpu::gpu_ddot	(	type(gpu_blas), intent(in)	handle,
		integer*4	n,
		double precision, target	dx,
		integer*4	incx,
		double precision, target	dy,
		integer*4	incy,
		double precision, intent(out)	res
	)

Double precision dot product (32-bit dimensions)

Computes the dot product of two vectors: result = x^T * y

Parameters

[in]	handle	BLAS handle
[in]	n	Number of elements
[in]	dx	First element of vector x
[in]	incx	Stride for vector x
[in]	dy	First element of vector y
[in]	incy	Stride for vector y
[out]	res	Resulting dot product

Note

Pass the first element of the array: xf(1), not x

Definition at line 1293 of file simple_gpu.F90.

gpu_ddot_64()

subroutine gpu::gpu_ddot_64	(	type(gpu_blas), intent(in)	handle,
		integer*8	n,
		double precision, target	dx,
		integer*8	incx,
		double precision, target	dy,
		integer*8	incy,
		double precision, intent(out)	res
	)

Double precision dot product (64-bit dimensions)

Computes the dot product of two vectors: result = x^T * y This variant accepts 64-bit integers for dimensions.

Parameters

[in]	handle	BLAS handle
[in]	n	Number of elements (64-bit)
[in]	dx	First element of vector x
[in]	incx	Stride for vector x (64-bit)
[in]	dy	First element of vector y
[in]	incy	Stride for vector y (64-bit)
[out]	res	Resulting dot product

Definition at line 1314 of file simple_gpu.F90.

gpu_deallocate_double1()

subroutine gpu::gpu_deallocate_double1

(

type(gpu_double1), intent(inout)

ptr

)

Definition at line 829 of file simple_gpu.F90.

gpu_deallocate_double2()

subroutine gpu::gpu_deallocate_double2

(

type(gpu_double2), intent(inout)

ptr

)

Definition at line 836 of file simple_gpu.F90.

gpu_deallocate_double3()

subroutine gpu::gpu_deallocate_double3

(

type(gpu_double3), intent(inout)

ptr

)

Definition at line 843 of file simple_gpu.F90.

gpu_deallocate_double4()

subroutine gpu::gpu_deallocate_double4

(

type(gpu_double4), intent(inout)

ptr

)

Definition at line 850 of file simple_gpu.F90.

gpu_deallocate_double5()

subroutine gpu::gpu_deallocate_double5

(

type(gpu_double5), intent(inout)

ptr

)

Definition at line 857 of file simple_gpu.F90.

gpu_deallocate_double6()

subroutine gpu::gpu_deallocate_double6

(

type(gpu_double6), intent(inout)

ptr

)

Definition at line 864 of file simple_gpu.F90.

gpu_deallocate_real1()

subroutine gpu::gpu_deallocate_real1

(

type(gpu_real1), intent(inout)

ptr

)

Definition at line 872 of file simple_gpu.F90.

gpu_deallocate_real2()

subroutine gpu::gpu_deallocate_real2

(

type(gpu_real2), intent(inout)

ptr

)

Definition at line 879 of file simple_gpu.F90.

gpu_deallocate_real3()

subroutine gpu::gpu_deallocate_real3

(

type(gpu_real3), intent(inout)

ptr

)

Definition at line 886 of file simple_gpu.F90.

gpu_deallocate_real4()

subroutine gpu::gpu_deallocate_real4

(

type(gpu_real4), intent(inout)

ptr

)

Definition at line 893 of file simple_gpu.F90.

gpu_deallocate_real5()

subroutine gpu::gpu_deallocate_real5

(

type(gpu_real5), intent(inout)

ptr

)

Definition at line 900 of file simple_gpu.F90.

gpu_deallocate_real6()

subroutine gpu::gpu_deallocate_real6

(

type(gpu_real6), intent(inout)

ptr

)

Definition at line 907 of file simple_gpu.F90.

gpu_dgeam()

subroutine gpu::gpu_dgeam	(	type(gpu_blas), intent(in)	handle,
		character, intent(in)	transa,
		character, intent(in)	transb,
		integer*4, intent(in)	m,
		integer*4, intent(in)	n,
		double precision, intent(in)	alpha,
		double precision	a,
		integer*4, intent(in)	lda,
		double precision, intent(in)	beta,
		double precision	b,
		integer*4, intent(in)	ldb,
		double precision	c,
		integer*4, intent(in)	ldc
	)

Definition at line 1344 of file simple_gpu.F90.

gpu_dgeam_64()

subroutine gpu::gpu_dgeam_64	(	type(gpu_blas), intent(in)	handle,
		character, intent(in)	transa,
		character, intent(in)	transb,
		integer*8, intent(in)	m,
		integer*8, intent(in)	n,
		double precision, intent(in)	alpha,
		double precision	a,
		integer*8, intent(in)	lda,
		double precision, intent(in)	beta,
		double precision	b,
		integer*8, intent(in)	ldb,
		double precision	c,
		integer*8, intent(in)	ldc
	)

Definition at line 1356 of file simple_gpu.F90.

gpu_dgemm()

subroutine gpu::gpu_dgemm	(	type(gpu_blas), intent(in)	handle,
		character, intent(in)	transa,
		character, intent(in)	transb,
		integer*4, intent(in)	m,
		integer*4, intent(in)	n,
		integer*4, intent(in)	k,
		double precision, intent(in)	alpha,
		double precision	a,
		integer*4, intent(in)	lda,
		double precision	b,
		integer*4, intent(in)	ldb,
		double precision, intent(in)	beta,
		double precision	c,
		integer*4, intent(in)	ldc
	)

Definition at line 1448 of file simple_gpu.F90.

gpu_dgemm_64()

subroutine gpu::gpu_dgemm_64	(	type(gpu_blas), intent(in)	handle,
		character, intent(in)	transa,
		character, intent(in)	transb,
		integer*8, intent(in)	m,
		integer*8, intent(in)	n,
		integer*8, intent(in)	k,
		double precision, intent(in)	alpha,
		double precision	a,
		integer*8, intent(in)	lda,
		double precision	b,
		integer*8, intent(in)	ldb,
		double precision, intent(in)	beta,
		double precision	c,
		integer*8, intent(in)	ldc
	)

Definition at line 1460 of file simple_gpu.F90.

gpu_dgemv()

subroutine gpu::gpu_dgemv	(	type(gpu_blas), intent(in)	handle,
		character, intent(in)	transa,
		integer*4, intent(in)	m,
		integer*4, intent(in)	n,
		double precision, intent(in)	alpha,
		double precision	a,
		integer*4, intent(in)	lda,
		double precision	x,
		integer*4, intent(in)	incx,
		double precision, intent(in)	beta,
		double precision	y,
		integer*4, intent(in)	incy
	)

Definition at line 1395 of file simple_gpu.F90.

gpu_dgemv_64()

subroutine gpu::gpu_dgemv_64	(	type(gpu_blas), intent(in)	handle,
		character, intent(in)	transa,
		integer*8, intent(in)	m,
		integer*8, intent(in)	n,
		double precision, intent(in)	alpha,
		double precision	a,
		integer*8, intent(in)	lda,
		double precision	x,
		integer*8, intent(in)	incx,
		double precision, intent(in)	beta,
		double precision	y,
		integer*8, intent(in)	incy
	)

Definition at line 1407 of file simple_gpu.F90.

gpu_download_double0()

subroutine gpu::gpu_download_double0	(	double precision, intent(in), target	gpu_ptr,
		double precision, intent(in), target	cpu_ptr,
		integer, intent(in)	n
	)

Definition at line 1023 of file simple_gpu.F90.

gpu_download_double1()

subroutine gpu::gpu_download_double1	(	type(gpu_double1), intent(in)	gpu_ptr,
		double precision, dimension(:), intent(in), target	cpu_ptr
	)

Definition at line 1031 of file simple_gpu.F90.

gpu_download_double2()

subroutine gpu::gpu_download_double2	(	type(gpu_double2), intent(in)	gpu_ptr,
		double precision, dimension(:,:), intent(in), target	cpu_ptr
	)

Definition at line 1038 of file simple_gpu.F90.

gpu_download_double3()

subroutine gpu::gpu_download_double3	(	type(gpu_double3), intent(in)	gpu_ptr,
		double precision, dimension(:,:,:), intent(in), target	cpu_ptr
	)

Definition at line 1045 of file simple_gpu.F90.

gpu_download_double4()

subroutine gpu::gpu_download_double4	(	type(gpu_double4), intent(in)	gpu_ptr,
		double precision, dimension(:,:,:,:), intent(in), target	cpu_ptr
	)

Definition at line 1052 of file simple_gpu.F90.

gpu_download_double5()

subroutine gpu::gpu_download_double5	(	type(gpu_double5), intent(in)	gpu_ptr,
		double precision, dimension(:,:,:,:,:), intent(in), target	cpu_ptr
	)

Definition at line 1059 of file simple_gpu.F90.

gpu_download_double6()

subroutine gpu::gpu_download_double6	(	type(gpu_double6), intent(in)	gpu_ptr,
		double precision, dimension(:,:,:,:,:,:), intent(in), target	cpu_ptr
	)

Definition at line 1066 of file simple_gpu.F90.

gpu_download_real0()

subroutine gpu::gpu_download_real0	(	real, intent(in), target	gpu_ptr,
		real, intent(in), target	cpu_ptr,
		integer, intent(in)	n
	)

Definition at line 1073 of file simple_gpu.F90.

gpu_download_real1()

subroutine gpu::gpu_download_real1	(	type(gpu_real1), intent(in)	gpu_ptr,
		real, dimension(:), intent(in), target	cpu_ptr
	)

Definition at line 1081 of file simple_gpu.F90.

gpu_download_real2()

subroutine gpu::gpu_download_real2	(	type(gpu_real2), intent(in)	gpu_ptr,
		real, dimension(:,:), intent(in), target	cpu_ptr
	)

Definition at line 1088 of file simple_gpu.F90.

gpu_download_real3()

subroutine gpu::gpu_download_real3	(	type(gpu_real3), intent(in)	gpu_ptr,
		real, dimension(:,:,:), intent(in), target	cpu_ptr
	)

Definition at line 1095 of file simple_gpu.F90.

gpu_download_real4()

subroutine gpu::gpu_download_real4	(	type(gpu_real4), intent(in)	gpu_ptr,
		real, dimension(:,:,:,:), intent(in), target	cpu_ptr
	)

Definition at line 1102 of file simple_gpu.F90.

gpu_download_real5()

subroutine gpu::gpu_download_real5	(	type(gpu_real5), intent(in)	gpu_ptr,
		real, dimension(:,:,:,:,:), intent(in), target	cpu_ptr
	)

Definition at line 1109 of file simple_gpu.F90.

gpu_download_real6()

subroutine gpu::gpu_download_real6	(	type(gpu_real6), intent(in)	gpu_ptr,
		real, dimension(:,:,:,:,:,:), intent(in), target	cpu_ptr
	)

Definition at line 1116 of file simple_gpu.F90.

gpu_sdot()

subroutine gpu::gpu_sdot	(	type(gpu_blas), intent(in)	handle,
		integer*4	n,
		real, target	dx,
		integer*4	incx,
		real, target	dy,
		integer*4	incy,
		real, intent(out)	res
	)

Definition at line 1323 of file simple_gpu.F90.

gpu_sdot_64()

subroutine gpu::gpu_sdot_64	(	type(gpu_blas), intent(in)	handle,
		integer*8	n,
		real, target	dx,
		integer*8	incx,
		real, target	dy,
		integer*8	incy,
		real, intent(out)	res
	)

Definition at line 1332 of file simple_gpu.F90.

gpu_set_stream()

subroutine gpu::gpu_set_stream	(	type(gpu_blas)	handle,
		type(gpu_stream)	stream
	)

Definition at line 1240 of file simple_gpu.F90.

gpu_sgeam()

subroutine gpu::gpu_sgeam	(	type(gpu_blas), intent(in)	handle,
		character, intent(in)	transa,
		character, intent(in)	transb,
		integer*4, intent(in)	m,
		integer*4, intent(in)	n,
		real, intent(in)	alpha,
		real	a,
		integer*4, intent(in)	lda,
		real, intent(in)	beta,
		real	b,
		integer*4, intent(in)	ldb,
		real	c,
		integer*4, intent(in)	ldc
	)

Definition at line 1368 of file simple_gpu.F90.

gpu_sgeam_64()

subroutine gpu::gpu_sgeam_64	(	type(gpu_blas), intent(in)	handle,
		character, intent(in)	transa,
		character, intent(in)	transb,
		integer*8, intent(in)	m,
		integer*8, intent(in)	n,
		real, intent(in)	alpha,
		real	a,
		integer*8, intent(in)	lda,
		real, intent(in)	beta,
		real	b,
		integer*8, intent(in)	ldb,
		real	c,
		integer*8, intent(in)	ldc
	)

Definition at line 1380 of file simple_gpu.F90.

gpu_sgemm()

subroutine gpu::gpu_sgemm	(	type(gpu_blas), intent(in)	handle,
		character, intent(in)	transa,
		character, intent(in)	transb,
		integer*4, intent(in)	m,
		integer*4, intent(in)	n,
		integer*4, intent(in)	k,
		real, intent(in)	alpha,
		real	a,
		integer*4, intent(in)	lda,
		real	b,
		integer*4, intent(in)	ldb,
		real, intent(in)	beta,
		real	c,
		integer*4, intent(in)	ldc
	)

Definition at line 1471 of file simple_gpu.F90.

gpu_sgemm_64()

subroutine gpu::gpu_sgemm_64	(	type(gpu_blas), intent(in)	handle,
		character, intent(in)	transa,
		character, intent(in)	transb,
		integer*8, intent(in)	m,
		integer*8, intent(in)	n,
		integer*8, intent(in)	k,
		real, intent(in)	alpha,
		real	a,
		integer*8, intent(in)	lda,
		real	b,
		integer*8, intent(in)	ldb,
		real, intent(in)	beta,
		real	c,
		integer*8, intent(in)	ldc
	)

Definition at line 1483 of file simple_gpu.F90.

gpu_sgemv()

subroutine gpu::gpu_sgemv	(	type(gpu_blas), intent(in)	handle,
		character, intent(in)	transa,
		integer*4, intent(in)	m,
		integer*4, intent(in)	n,
		real, intent(in)	alpha,
		real	a,
		integer*4, intent(in)	lda,
		real	x,
		integer*4, intent(in)	incx,
		real, intent(in)	beta,
		real	y,
		integer*4, intent(in)	incy
	)

Definition at line 1420 of file simple_gpu.F90.

gpu_sgemv_64()

subroutine gpu::gpu_sgemv_64	(	type(gpu_blas), intent(in)	handle,
		character, intent(in)	transa,
		integer*8, intent(in)	m,
		integer*8, intent(in)	n,
		real, intent(in)	alpha,
		real	a,
		integer*8, intent(in)	lda,
		real	x,
		integer*8, intent(in)	incx,
		real, intent(in)	beta,
		real	y,
		integer*8, intent(in)	incy
	)

Definition at line 1432 of file simple_gpu.F90.

gpu_stream_create()

subroutine gpu::gpu_stream_create

(

type(gpu_stream)

stream

)

Definition at line 1230 of file simple_gpu.F90.

gpu_stream_destroy()

subroutine gpu::gpu_stream_destroy

(

type(gpu_stream)

stream

)

Definition at line 1235 of file simple_gpu.F90.

gpu_upload_double0()

subroutine gpu::gpu_upload_double0	(	double precision, intent(in), target	cpu_ptr,
		double precision, intent(in), target	gpu_ptr,
		integer, intent(in)	n
	)

Definition at line 918 of file simple_gpu.F90.

gpu_upload_double1()

subroutine gpu::gpu_upload_double1	(	double precision, dimension(*), intent(in), target	cpu_ptr,
		type(gpu_double1), intent(in)	gpu_ptr
	)

Definition at line 926 of file simple_gpu.F90.

gpu_upload_double2()

subroutine gpu::gpu_upload_double2	(	double precision, dimension(:,:), intent(in), target	cpu_ptr,
		type(gpu_double2), intent(in)	gpu_ptr
	)

Definition at line 933 of file simple_gpu.F90.

gpu_upload_double3()

subroutine gpu::gpu_upload_double3	(	double precision, dimension(:,:,:), intent(in), target	cpu_ptr,
		type(gpu_double3), intent(in)	gpu_ptr
	)

Definition at line 940 of file simple_gpu.F90.

gpu_upload_double4()

subroutine gpu::gpu_upload_double4	(	double precision, dimension(:,:,:,:), intent(in), target	cpu_ptr,
		type(gpu_double4), intent(in)	gpu_ptr
	)

Definition at line 947 of file simple_gpu.F90.

gpu_upload_double5()

subroutine gpu::gpu_upload_double5	(	double precision, dimension(:,:,:,:,:), intent(in), target	cpu_ptr,
		type(gpu_double5), intent(in)	gpu_ptr
	)

Definition at line 954 of file simple_gpu.F90.

gpu_upload_double6()

subroutine gpu::gpu_upload_double6	(	double precision, dimension(:,:,:,:,:,:), intent(in), target	cpu_ptr,
		type(gpu_double6), intent(in)	gpu_ptr
	)

Definition at line 961 of file simple_gpu.F90.

gpu_upload_real0()

subroutine gpu::gpu_upload_real0	(	real, intent(in), target	cpu_ptr,
		real, intent(in), target	gpu_ptr,
		integer, intent(in)	n
	)

Definition at line 969 of file simple_gpu.F90.

gpu_upload_real1()

subroutine gpu::gpu_upload_real1	(	real, dimension(*), intent(in), target	cpu_ptr,
		type(gpu_real1), intent(in)	gpu_ptr
	)

Definition at line 977 of file simple_gpu.F90.

gpu_upload_real2()

subroutine gpu::gpu_upload_real2	(	real, dimension(:,:), intent(in), target	cpu_ptr,
		type(gpu_real2), intent(in)	gpu_ptr
	)

Definition at line 984 of file simple_gpu.F90.

gpu_upload_real3()

subroutine gpu::gpu_upload_real3	(	real, dimension(:,:,:), intent(in), target	cpu_ptr,
		type(gpu_real3), intent(in)	gpu_ptr
	)

Definition at line 991 of file simple_gpu.F90.

gpu_upload_real4()

subroutine gpu::gpu_upload_real4	(	real, dimension(:,:,:,:), intent(in), target	cpu_ptr,
		type(gpu_real4), intent(in)	gpu_ptr
	)

Definition at line 998 of file simple_gpu.F90.

gpu_upload_real5()

subroutine gpu::gpu_upload_real5	(	real, dimension(:,:,:,:,:), intent(in), target	cpu_ptr,
		type(gpu_real5), intent(in)	gpu_ptr
	)

Definition at line 1005 of file simple_gpu.F90.

gpu_upload_real6()

subroutine gpu::gpu_upload_real6	(	real, dimension(:,:,:,:,:,:), intent(in), target	cpu_ptr,
		type(gpu_real6), intent(in)	gpu_ptr
	)

Definition at line 1012 of file simple_gpu.F90.