Preface

This document describes the NVIDIA Fortran interfaces to cuBLAS, cuFFT, cuRAND, and cuSPARSE, which are CUDA Libraries used in scientific and engineering applications built upon the CUDA computing architecture.

Intended Audience

This guide is intended for application programmers, scientists and engineers proficient in programming with the Fortran language. This guide assumes some familiarity with either CUDA Fortran or OpenACC.

Organization

The organization of this document is as follows:

Introduction
contains a general introduction to Fortran interfaces, OpenACC, CUDA Fortran, and CUDA Library functions
BLAS Runtime Library APIs
describes the Fortran interfaces to the various cuBLAS libraries
FFT Runtime Library APIs
describes the module types, definitions and Fortran interfaces to the cuFFT library
Random Number Runtime APIs
describes the Fortran interfaces to the host and device cuRAND libraries
Sparse Matrix Runtime APIs
describes the module types, definitions and Fortran interfaces to the cuSPARSE Library
Matrix Solver Runtime APIs
describes the module types, definitions and Fortran interfaces to the cuSOLVER Library
Tensor Primitives Runtime APIs
describes the module types, definitions and Fortran interfaces to the cuTENSOR Library
NVIDIA Collective Communications Library APIs
describes the module types, definitions and Fortran interfaces to the NCCL Library
NVSHMEM Communication Library APIs
describes the module types, definitions and Fortran interfaces to the NVSHMEM Library
Examples
provides sample code and an explanation of each of the simple examples.

Conventions

This guide uses the following conventions:

italic
is used for emphasis.
Constant Width
is used for filenames, directories, arguments, options, examples, and for language statements in the text, including assembly language statements.
Bold
is used for commands.
[ item1 ]
in general, square brackets indicate optional items. In this case item1 is optional. In the context of p/t-sets, square brackets are required to specify a p/t-set.
{ item2 | item 3 }
braces indicate that a selection is required. In this case, you must select either item2 or item3.
filename ...
ellipsis indicate a repetition. Zero or more of the preceding item may occur. In this example, multiple filenames are allowed.
FORTRAN
Fortran language statements are shown in the text of this guide using a reduced fixed point size.
C++ and C
C++ and C language statements are shown in the test of this guide using a reduced fixed point size.

Terminology

If there are terms in this guide with which you are unfamiliar, see the NVIDIA HPC glossary.

1. Introduction

This document provides a reference for calling CUDA Library functions from NVIDIA Fortran. It can be used from Fortran code using the OpenACC or OpenMP programming models, or from NVIDIA CUDA Fortran. Currently, the CUDA libraries which NVIDIA provides pre-built interface modules for, and which are documented here, are:

  • cuBLAS, an implementation of the BLAS.
  • cuFFT, a library of Fast Fourier Transform (FFT) routines.
  • cuRAND, a library for random number generation.
  • cuSPARSE, a library of linear algebra routines used with sparse matrices.
  • cuSOLVER, a library of equation solvers used with dense or other matrices.
  • cuTENSOR, a library for tensor primitive operations.
  • NCCL, a collective communications librarys.
  • NVSHMEM, a library implementation of OpenSHMEM on GPUs.

The OpenACC Application Program Interface is a collection of compiler directives and runtime routines that allows the programmer to specify loops and regions of code for offloading from a host CPU to an attached accelerator, such as a GPU. The OpenACC API was designed and is maintained by an industry consortium. See the OpenACC website for more information about the OpenACC API.

OpenMP is a specification for a set of compiler directives, an applications programming interface (API), and a set of environment variables that can be used to specify parallel execution from Fortran (and other languages). The OpenMP target offload capabilities are similar in many respects to OpenACC. The methods for passing device arrays to library functions from host code differ only in syntax compared to those used in OpenACC. For general information about using OpenMP and to obtain a copy of the OpenMP specification, refer to the OpenMP organization's website.

CUDA Fortran is a small set of extensions to Fortran that supports and is built upon the CUDA computing architecture. CUDA Fortran includes a Fortran 2003 compiler and tool chain for programming NVIDIA GPUs using Fortran, and is an analog to NVIDIA's CUDA C compiler. Compared to the NVIDIA Accelerator and OpenACC directives-based model and compilers, CUDA Fortran is a lower-level explicit programming model with substantial runtime library components that give expert programmers direct control of all aspects of GPGPU programming.

This document does not contain explanations or purposes of the library functions, nor does it contain details of the approach used in the CUDA implementation to target GPUs. For that information, please see the appropriate library document that comes with the NVIDIA CUDA Toolkit. This document does provide the Fortran module contents: derived types, enumerations, and interfaces, to make use of the libraries from Fortran rather than from C or C++.

Many of the examples used in this document are provided in the HPC compiler and tools distribution, along with Makefiles, and are stored in the yearly directory, such as 2020/examples/CUDA-Libraries.

1.1. Fortran Interfaces and Wrappers

Almost all of the function interfaces shown in this document make use of features from the Fortran 2003 iso_c_binding intrinsic module. This module provides a standard way for dealing with isues such as inter-language data types, capitalization, adding underscores to symbol names, or passing arguments by value.

Often, the iso_c_binding module enables Fortran programs containing properly written interfaces to call directly into the C library functions. In some cases, NVIDIA has written small wrappers around the C library function, to make the Fortran call site more "Fortran-like", hiding some issues exposed in the C interfaces like handle management, host vs. device pointer management, or character and complex data type issues.

In a small number of cases, the C Library may contain multiple entry points to handle different data types, perhaps an int in one function and a size_t in another, otherwise the functions are identical. In these cases, NVIDIA may provide just one generic Fortran interface, and will call the appropriate C function under the hood.

1.2. Using CUDA Libraries from OpenACC Host Code

All of the libraries covered in this document contain functions which are callable from OpenACC host code. Most functions take some arguments which are expected to be device pointers (the address of a variable in device global memory). There are several ways to do that in OpenACC.

If the call is lexically nested within an OpenACC data directive, the NVIDIA Fortran compiler, in the presence of an explicit interface such as those provided by the NVIDIA library modules, will default to passing the device pointer when required. 

subroutine hostcall(a, b, n)
use cublas
real a(n), b(n)
!$acc data copy(a, b)
call cublasSswap(n, a, 1, b, 1)
!$acc end data
return
end

A Fortran interface is made explicit when you use the module that contains it, as in the line use cublas in the example above. If you look ahead to the actual interface for cublasSswap, you will see that the arrays a and b are declared with the CUDA Fortran device attribute, so they take only device addresses as arguments.

It is more acceptable and general when using OpenACC to pass device pointers to subprograms by using the host_data clause as most implementations don't have a way to mark arguments as device pointers. The host_data construct with the use_device clause makes the device addresses available in host code for passing to the subprogram. 

use cufft
use openacc
. . .
!$acc data copyin(a), copyout(b,c)
ierr = cufftPlan2D(iplan1,m,n,CUFFT_C2C)
ierr = ierr + cufftSetStream(iplan1,acc_get_cuda_stream(acc_async_sync))
!$acc host_data use_device(a,b,c)
ierr = ierr + cufftExecC2C(iplan1,a,b,CUFFT_FORWARD)
ierr = ierr + cufftExecC2C(iplan1,b,c,CUFFT_INVERSE)
!$acc end host_data

! scale c
!$acc kernels
c = c / (m*n)
!$acc end kernels
!$acc end data

This code snippet also shows an example of sharing the stream that OpenACC and the cuFFT library use. Every library in this document has a function for setting the CUDA stream which the library runs on. Usually, when using OpenACC, you want the OpenACC kernels to run on the same stream as the library functions. In the case above, this guarantees that the kernel c = c / (m*n) does not start until the FFT operations complete. The function acc_get_cuda_stream and the definition for acc_async_sync are in the openacc module.

1.3. Using CUDA Libraries from OpenACC Device Code

Two libraries are currently available from within OpenACC compute regions. Certain functions in both the openacc_curand module and the nvshmem module are marked acc routine seq.

The cuRAND device library is all contained within CUDA header files. In device code, it is designed to return one or a small number of random numbers per thread. The thread's random generators run independently of each other, and it is usually advised for performance reasons to give each thread a different seed, rather than a different offset. 

program t
use openacc_curand
integer, parameter :: n = 500
real a(n,n,4)
type(curandStateXORWOW) :: h
integer(8) :: seed, seq, offset
a = 0.0
!$acc parallel num_gangs(n) vector_length(n) copy(a)
!$acc loop gang
do j = 1, n
!$acc loop vector private(h)
  do i = 1, n
    seed = 12345_8 + j*n*n + i*2
    seq = 0_8
    offset = 0_8
    call curand_init(seed, seq, offset, h)
!$acc loop seq
    do k = 1, 4
      a(i,j,k) = curand_uniform(h)
    end do
  end do
end do
!$acc end parallel
print *,maxval(a),minval(a),sum(a)/(n*n*4)
end

When using the openacc_curand module, since all the code is contained in CUDA header files, you do not need any additional libraries on the link line.

1.4. Using CUDA Libraries from CUDA Fortran Host Code

The predominant usage model for the library functions listed in this document is to call them from CUDA Host code. CUDA Fortran allows some special capabilities in that the compiler is able to recognize the device and managed attribute in resolving generic interfaces. Device actual arguments can only match the interface's device dummy arguments; managed actual arguments, by precedence, match managed dummy arguments first, then device dummies, then host. 

program testisamax  ! link with -cudalib=cublas -lblas
use cublas
real*4              x(1000)
real*4, device  :: xd(1000)
real*4, managed :: xm(1000)

call random_number(x)

! Call host BLAS
j = isamax(1000,x,1)

xd = x
! Call cuBLAS
k = isamax(1000,xd,1)
print *,j.eq.k

xm = x
! Also calls cuBLAS
k = isamax(1000,xm,1)
print *,j.eq.k
end

Using the cudafor module, the full set of CUDA functionality is available to programmers for managing CUDA events, streams, synchronization, and asynchronous behaviors. CUDA Fortran can be used in OpenMP programs, and the CUDA Libraries in this document are thread safe with respect to host CPU threads. Further examples are included in chapter Examples.

1.5. Using CUDA Libraries from CUDA Fortran Device Code

The cuRAND and NVSHMEM libraries have functions callable from CUDA Fortran device code, and their interfaces are accessed via the curand_device and nvshmem modules, respectively. The module interfaces are very similar to the modules used in OpenACC device code, but for CUDA Fortran, each subroutine and function is declared attributes([host,]device), and the subroutines and functions do not need to be marked as acc routine seq. 

module mrand
    use curand_device
    integer, parameter :: n = 500
    contains
	attributes(global) subroutine randsub(a)
	real, device :: a(n,n,4)
	type(curandStateXORWOW) :: h
	integer(8) :: seed, seq, offset
	j = blockIdx%x; i = threadIdx%x
	seed = 12345_8 + j*n*n + i*2
	seq = 0_8
	offset = 0_8
	call curand_init(seed, seq, offset, h)
	do k = 1, 4
	    a(i,j,k) = curand_uniform(h)
	end do
	end subroutine
end module

program t   ! nvfortran t.cuf
use mrand
use cudafor ! recognize maxval, minval, sum w/managed
real, managed :: a(n,n,4)
a = 0.0
call randsub<<<n,n>>>(a)
print *,maxval(a),minval(a),sum(a)/(n*n*4)
end program

1.6. Pointer Modes in cuBLAS and cuSPARSE

Because the NVIDIA Fortran compiler can distinguish between host and device arguments, the NVIDIA modules for interfacing to cuBLAS and cuSPARSE handle pointer modes differently than CUDA C, which requires setting the mode explicitly for scalar arguments. Examples of scalar arguments which can reside either on the host or device are the alpha and beta scale factors to the *gemm functions.

Typically, when using the normal "non-_v2" interfaces in the cuBLAS and cuSPARSE modules, the runtime wrappers will implicitly add the setting and restoring of the library pointer modes behind the scenes. This adds some negligible but non-zero overhead to the calls.

To avoid the implicit getting and setting of the pointer mode with every invocation of a library function do the following:

  • For the BLAS, use the cublas_v2 module, and the v2 entry points, such as cublasIsamax_v2. It is the programmer's responsibility to properly set the pointer mode when needed. Examples of scalar arguments which do require setting the pointer mode are the alpha and beta scale factors passed to the *gemm routines, and the scalar results returned from the v2 versions of the *amax(), *amin(), *asum(), *rotg(), *rotmg(), *nrm2(), and *dot() functions. In the v2 interfaces shown in the chapter 2, these scalar arguments will have the comment ! device or host variable. Examples of scalar arguments which do not require setting the pointer mode are increments, extents, and lengths such as incx, incy, n, lda, ldb, and ldc.

  • For the cuSPARSE library, each function listed in chapter 5 which contains scalar arguments with the comment ! device or host variable has a corresponding v2 interface, though it is not documented here. For instance, in addition to the interface named cusparseSaxpyi, there is another interface named cusparseSaxpyi_v2 with the exact same argument list which calls into the cuSPARSE library directly and will not implicitly get or set the library pointer mode.

The CUDA default pointer mode is that the scalar arguments reside on the host. The NVIDIA runtime does not change that setting.

1.7. Writing Your Own CUDA Interfaces

Despite the large number of interfaces included in the modules described in this document, users will have the need from time-to-time to write their own interfaces to new libraries or their own tuned CUDA, perhaps written in C/C++. There are some standard techniques to use, and some non-standard NVIDIA extensions which can make creating working interfaces easier. 

! cufftExecC2C
interface cufftExecC2C
    integer function cufftExecC2C( plan, idata, odata, direction ) &
        bind(C,name='cufftExecC2C')
        integer, value :: plan
        complex, device, dimension(*) :: idata, odata
        integer, value :: direction
    end function cufftExecC2C
end interface cufftExecC2C

This interface calls the C library function directly. You can deal with Fortran's capitalization issues by putting the properly capitalized C function in the bind(C) attribute. If the C function expects input arguments passed by value, you can add the value attribute to the dummy declaration as well. A nice feature of Fortran is that the interface can change, but the code at the call site may not have to. The compiler changes the details of the call to fit the interface.

Now suppose a user of this interface would like to call this function with REAL data (F77 code is notorious for mixing REAL and COMPLEX declarations). There are two ways to do this: 

! cufftExecC2C
interface cufftExecC2C
    integer function cufftExecC2C( plan, idata, odata, direction ) &
        bind(C,name='cufftExecC2C')
        integer, value :: plan
        complex, device, dimension(*) :: idata, odata
        integer, value :: direction
    end function cufftExecC2C
    integer function cufftExecR2R( plan, idata, odata, direction ) &
        bind(C,name='cufftExecC2C')
        integer, value :: plan
        real, device, dimension(*) :: idata, odata
        integer, value :: direction
    end function cufftExecR2R
end interface cufftExecC2C

Here the C name hasn't changed. The compiler will now accept actual arguments corresponding to idata and odata that are declared REAL. A generic interface is created named cufftExecC2C. If you have problems debugging your generic interface, as a debugging aid you can try calling the specific name, cufftExecR2R in this case, to help diagnose the problem.

A commonly used extension which is supported by NVIDIA is ignore_tkr. A programmer can use it in an interface to instruct the compiler to ignore any combination of the type, kind, and rank during the interface matching process. The previous example using ignore_tkr looks like this: 

! cufftExecC2C
interface cufftExecC2C
    integer function cufftExecC2C( plan, idata, odata, direction ) &
        bind(C,name='cufftExecC2C')
        integer, value :: plan
        !dir$ ignore_tkr(tr) idata, (tr) odata
        complex, device, dimension(*) :: idata, odata
        integer, value :: direction
    end function cufftExecC2C
end interface cufftExecC2C

Now the compiler will ignore both the type and rank (F77 could also be sloppy in its handling of array dimensions) of idata and odata when matching the call site to the interface. An unfortunate side-effect is that the interface will now allow integer, logical, and character data for idata and odata. It is up to the implementor to determine if that is acceptable.

A final aid, specific to NVIDIA, worth mentioning here is ignore_tkr (d), which ignores the device attribute of an actual argument during interface matching.

Of course, if you write a wrapper, a narrow strip of code between the Fortran call and your library function, you are not limited by the simple transormations that a compiler can do, such as those listed here. As mentioned earlier, many of the interfaces provided in the cuBLAS and cuSPARSE modules use wrappers.

A common request is a way for Fortran programmers to take advantage of the thrust library. Explaining thrust and C++ programming is outside of the scope of this document, but this simple example can show how to take advantage of the excellent sort capabilities in thrust: 

// Filename: csort.cu
// nvcc -c -arch sm_35 csort.cu
#include <thrust/device_vector.h>
#include <thrust/copy.h>
#include <thrust/sort.h>

extern "C" {

    //Sort for integer arrays
    void thrust_int_sort_wrapper( int *data, int N)
    {
    thrust::device_ptr <int> dev_ptr(data);
    thrust::sort(dev_ptr, dev_ptr+N);
    }

    //Sort for float arrays
    void thrust_float_sort_wrapper( float *data, int N)
    {
    thrust::device_ptr <float> dev_ptr(data);
    thrust::sort(dev_ptr, dev_ptr+N);
    }

    //Sort for double arrays
    void thrust_double_sort_wrapper( double *data, int N)
    {
    thrust::device_ptr <double> dev_ptr(data);
    thrust::sort(dev_ptr, dev_ptr+N);
    }
}

Set up interface to the sort subroutine in Fortran and calls are simple:

program t
interface sort
   subroutine sort_int(array, n) &
        bind(C,name='thrust_int_sort_wrapper')
        integer(4), device, dimension(*) :: array
        integer(4), value :: n
    end subroutine
end interface
integer(4), parameter :: n = 100
integer(4), device :: a_d(n)
integer(4) :: a_h(n)
!$cuf kernel do
do i = 1, n
    a_d(i) = 1 + mod(47*i,n)
end do
call sort(a_d, n)
a_h = a_d
nres  = count(a_h .eq. (/(i,i=1,n)/))
if (nres.eq.n) then
    print *,"test PASSED"
else
    print *,"test FAILED"
endif
end

1.8. NVIDIA Fortran Compiler Options

The NVIDIA Fortran compiler driver is called nvfortran. General information on the compiler options which can be passed to nvfortran can be obtained by typing nvfortran -help. To enable targeting NVIDIA GPUs using OpenACC, use nvfortran -acc=gpu. To enable targeting NVIDIA GPUs using CUDA Fortran, use nvfortran -cuda. CUDA Fortran is also supported by the NVIDIA Fortran compilers when the filename uses the .cuf extension. Uppercase file extensions, .F90 or .CUF, for example, may also be used, in which case the program is processed by the preprocessor before being compiled.

Other options which are pertinent to the examples in this document are:

  • -⁠cudalib[=cublas|cufft|cufftw|curand|cusolver|cusparse|cutensor|nvblas|nccl|nvshmem]: this option adds the appropriate versions of the CUDA-optimized libraries to the link line. It handles static and dynamic linking, and platform (Linux, Windows) differences unobtrusively.

  • -⁠gpu=cc70: this option compiles for compute capability 7.0. Certain library functionality may require minimum compute capability of 6.0, 7.0, or higher.

  • -⁠gpu=cudaX.Y: this option compiles and links with a particular CUDA Toolkit version. Certain library functionality may require a newer (or older, for deprecated functions) CUDA runtime version.

2. BLAS Runtime APIs

This section describes the Fortran interfaces to the CUDA BLAS libraries. There are currently three separate collections of function entry points which are commonly referred to as the cuBLAS:

  • The original CUDA implementation of the BLAS routines, referred to as the legacy API, which are callable from the host and expect and operate on device data.

  • The newer "v2" CUDA implementation of the BLAS routines, plus some extensions for batched operations. These are also callable from the host and operate on device data. In Fortran terms, these entry points have been changed from subroutines to functions which return status.

  • The cuBLAS XT library which can target multiple GPUs using only host-resident data.

NVIDIA currently ships with three Fortran modules which programmers can use to call into this cuBLAS functionality:

  • cublas, which provides interfaces to into the main cublas library. Both the legacy and v2 names are supported. In this module, the cublas names (such as cublasSaxpy) use the legacy calling conventions. Interfaces to a host BLAS library (for instance libblas.a in the NVIDIA distribution) are also included in the cublas module. These interfaces are exposed by adding the line 
    use cublas
    to your program unit.
  • cublas_v2, which is similar to the cublas module in most ways except the cublas names (such as cublasSaxpy) use the v2 calling conventions. For instance, instead of a subroutine, cublasSaxpy is a function which takes a handle as the first argument and returns an integer containing the status of the call. These interfaces are exposed by adding the line 
    use cublas_v2
    to your program unit.
  • cublasxt, which interfaces directly to the cublasXT API. These interfaces are exposed by adding the line 
    use cublasxt
    to your program unit.

The v2 routines are integer functions that return an error status code; they return a value of CUBLAS_STATUS_SUCCESS if the call was successful, or other cuBLAS status return value if there was an error.

Documented interfaces to the traditional BLAS names in the subsequent sections, which contain the comment ! device or host variable should not be confused with the pointer mode issue from section 1.6. The traditional BLAS names are overloaded generic names in the cublas module. For instance, in this interface 

 subroutine scopy(n, x, incx, y, incy)
   integer :: n
   real(4), device, dimension(*) :: x, y ! device or host variable
   integer :: incx, incy

The arrays x and y can either both be device arrays, in which case cublasScopy is called via the generic interface, or they can both be host arrays, in which case scopy from the host BLAS library is called. Using CUDA Fortran managed data as actual arguments to scopy poses an interesting case, and calling cublasScopy is chosen by default. If you wish to call the host library version of scopy with managed data, don't expose the generic scopy interface at the call site.

Unless a specific kind is provided, in the following interfaces the plain integer type implies integer(4) and the plain real type implies real(4).

2.1. CUBLAS Definitions and Helper Functions

This section contains definitions and data types used in the cuBLAS library and interfaces to the cuBLAS Helper Functions.

The cublas module contains the following derived type definitions:

TYPE cublasHandle
  TYPE(C_PTR)  :: handle
END TYPE

The cuBLAS module contains the following enumerations:

enum, bind(c)
    enumerator :: CUBLAS_STATUS_SUCCESS         =0
    enumerator :: CUBLAS_STATUS_NOT_INITIALIZED =1
    enumerator :: CUBLAS_STATUS_ALLOC_FAILED    =3
    enumerator :: CUBLAS_STATUS_INVALID_VALUE   =7
    enumerator :: CUBLAS_STATUS_ARCH_MISMATCH   =8
    enumerator :: CUBLAS_STATUS_MAPPING_ERROR   =11
    enumerator :: CUBLAS_STATUS_EXECUTION_FAILED=13
    enumerator :: CUBLAS_STATUS_INTERNAL_ERROR  =14
end enum
enum, bind(c)
    enumerator :: CUBLAS_FILL_MODE_LOWER=0
    enumerator :: CUBLAS_FILL_MODE_UPPER=1
end enum
enum, bind(c)
    enumerator :: CUBLAS_DIAG_NON_UNIT=0
    enumerator :: CUBLAS_DIAG_UNIT=1
end enum
enum, bind(c)
    enumerator :: CUBLAS_SIDE_LEFT =0
    enumerator :: CUBLAS_SIDE_RIGHT=1
end enum
enum, bind(c)
    enumerator :: CUBLAS_OP_N=0
    enumerator :: CUBLAS_OP_T=1
    enumerator :: CUBLAS_OP_C=2
end enum
enum, bind(c)
    enumerator :: CUBLAS_POINTER_MODE_HOST   = 0
    enumerator :: CUBLAS_POINTER_MODE_DEVICE = 1
end enum

2.1.1. cublasCreate

This function initializes the CUBLAS library and creates a handle to an opaque structure holding the CUBLAS library context. It allocates hardware resources on the host and device and must be called prior to making any other CUBLAS library calls. The CUBLAS library context is tied to the current CUDA device. To use the library on multiple devices, one CUBLAS handle needs to be created for each device. Furthermore, for a given device, multiple CUBLAS handles with different configuration can be created. Because cublasCreate allocates some internal resources and the release of those resources by calling cublasDestroy will implicitly call cublasDeviceSynchronize, it is recommended to minimize the number of cublasCreate/cublasDestroy occurences. For multi-threaded applications that use the same device from different threads, the recommended programming model is to create one CUBLAS handle per thread and use that CUBLAS handle for the entire life of the thread. 

 integer(4) function cublasCreate(handle)
   type(cublasHandle) :: handle

2.1.2. cublasDestroy

This function releases hardware resources used by the CUBLAS library. This function is usually the last call with a particular handle to the CUBLAS library. Because cublasCreate allocates some internal resources and the release of those resources by calling cublasDestroy will implicitly call cublasDeviceSynchronize, it is recommended to minimize the number of cublasCreate/cublasDestroy occurences. 

 integer(4) function cublasDestroy(handle)
   type(cublasHandle) :: handle

2.1.3. cublasGetVersion

This function returns the version number of the cuBLAS library. 

 integer(4) function cublasGetVersion(handle, version)
   type(cublasHandle) :: handle
   integer(4) :: version

2.1.4. cublasSetStream

This function sets the cuBLAS library stream, which will be used to execute all subsequent calls to the cuBLAS library functions. If the cuBLAS library stream is not set, all kernels use the default NULL stream. In particular, this routine can be used to change the stream between kernel launches and then to reset the cuBLAS library stream back to NULL. 

 integer(4) function cublasSetStream(handle, stream)
   type(cublasHandle) :: handle
   integer(kind=cuda_stream_kind()) :: stream

2.1.5. cublasGetStream

This function gets the cuBLAS library stream, which is being used to execute all calls to the cuBLAS library functions. If the cuBLAS library stream is not set, all kernels use the default NULL stream. 

 integer(4) function cublasGetStream(handle, stream)
   type(cublasHandle) :: handle
   integer(kind=cuda_stream_kind()) :: stream

2.1.6. cublasGetPointerMode

This function obtains the pointer mode used by the cuBLAS library. In the cublas module, the pointer mode is set and reset on a call-by-call basis depending on the whether the device attribute is set on scalar actual arguments. See section 1.6 for a discussion of pointer modes. 

 integer(4) function cublasGetPointerMode(handle, mode)
   type(cublasHandle) :: handle
   integer(4) :: mode

2.1.7. cublasSetPointerMode

This function sets the pointer mode used by the cuBLAS library. When using the cublas module, the pointer mode is set on a call-by-call basis depending on the whether the device attribute is set on scalar actual arguments. When using the cublas_v2 module with v2 interfaces, it is the programmer's responsibility to make calls to cublasSetPointerMode so scalar arguments are handled correctly by the library. See section 1.6 for a discussion of pointer modes. 

 integer(4) function cublasSetPointerMode(handle, mode)
   type(cublasHandle) :: handle
   integer(4) :: mode

2.1.8. cublasGetAtomicsMode

This function obtains the atomics mode used by the cuBLAS library. 

 integer(4) function cublasGetAtomicsMode(handle, mode)
   type(cublasHandle) :: handle
   integer(4) :: mode

2.1.9. cublasSetAtomicsMode

This function sets the atomics mode used by the cuBLAS library. Some routines in the cuBLAS library have alternate implementations that use atomics to accumulate results. These alternate implementations may run faster but may also generate results which are not identical from one run to the other. The default is to not allow atomics in cuBLAS functions. 

 integer(4) function cublasSetAtomicsMode(handle, mode)
   type(cublasHandle) :: handle
   integer(4) :: mode

2.1.10. cublasGetMathMode

This function obtains the math mode used by the cuBLAS library. 

 integer(4) function cublasGetMathMode(handle, mode)
   type(cublasHandle) :: handle
   integer(4) :: mode

2.1.11. cublasSetMathMode

This function sets the math mode used by the cuBLAS library. Some routines in the cuBLAS library allow you to choose the compute precision used to generate results. These alternate approaches may run faster but may also generate different, less accurate results. 

 integer(4) function cublasSetMathMode(handle, mode)
   type(cublasHandle) :: handle
   integer(4) :: mode

2.1.12. cublasGetHandle

This function gets the cuBLAS handle currently in use by a thread. The CUDA Fortran runtime keeps track of a CPU thread's current handle, if you are either using the legacy BLAS API, or do not wish to pass the handle through to low-level functions or subroutines manually. 

 type(cublashandle) function cublasGetHandle() 
 integer(4) function cublasGetHandle(handle)
   type(cublasHandle) :: handle

2.1.13. cublasSetVector

This function copies n elements from a vector x in host memory space to a vector y in GPU memory space. It is assumed that each element requires storage of elemSize bytes. In CUDA Fortran, the type of vector x and y is overloaded to take any data type, but the size of the data type must still be specified in bytes. This functionality can also be implemented using cudaMemcpy or array assignment statements. 

 integer(4) function cublassetvector(n, elemsize, x, incx, y, incy)
   integer :: n, elemsize, incx, incy
   integer*1, dimension(*) :: x
   integer*1, device, dimension(*) :: y

2.1.14. cublasGetVector

This function copies n elements from a vector x in GPU memory space to a vector y in host memory space. It is assumed that each element requires storage of elemSize bytes. In CUDA Fortran, the type of vector x and y is overloaded to take any data type, but the size of the data type must still be specified in bytes. This functionality can also be implemented using cudaMemcpy or array assignment statements. 

 integer(4) function cublasgetvector(n, elemsize, x, incx, y, incy)
   integer :: n, elemsize, incx, incy
   integer*1, device, dimension(*) :: x
   integer*1, dimension(*) :: y

2.1.15. cublasSetMatrix

This function copies a tile of rows x cols elements from a matrix A in host memory space to a matrix B in GPU memory space. It is assumed that each element requires storage of elemSize bytes. In CUDA Fortran, the type of Matrix A and B is overloaded to take any data type, but the size of the data type must still be specified in bytes. This functionality can also be implemented using cudaMemcpy, cudaMemcpy2D, or array assignment statements. 

 integer(4) function cublassetmatrix(rows, cols, elemsize, a, lda, b, ldb)
   integer :: rows, cols, elemsize, lda, ldb
   integer*1, dimension(lda, *) :: a
   integer*1, device, dimension(ldb, *) :: b

2.1.16. cublasGetMatrix

This function copies a tile of rows x cols elements from a matrix A in GPU memory space to a matrix B in host memory space. It is assumed that each element requires storage of elemSize bytes. In CUDA Fortran, the type of Matrix A and B is overloaded to take any data type, but the size of the data type must still be specified in bytes. This functionality can also be implemented using cudaMemcpy, cudaMemcpy2D, or array assignment statements. 

 integer(4) function cublasgetmatrix(rows, cols, elemsize, a, lda, b, ldb)
   integer :: rows, cols, elemsize, lda, ldb
   integer*1, device, dimension(lda, *) :: a
   integer*1, dimension(ldb, *) :: b

2.1.17. cublasSetVectorAsync

This function copies n elements from a vector x in host memory space to a vector y in GPU memory space, asynchronously, on the given CUDA stream. It is assumed that each element requires storage of elemSize bytes. In CUDA Fortran, the type of vector x and y is overloaded to take any data type, but the size of the data type must still be specified in bytes. This functionality can also be implemented using cudaMemcpyAsync. 

 integer(4) function cublassetvectorasync(n, elemsize, x, incx, y, incy, stream)
   integer :: n, elemsize, incx, incy
   integer*1, dimension(*) :: x
   integer*1, device, dimension(*) :: y
   integer(kind=cuda_stream_kind()) :: stream

2.1.18. cublasGetVectorAsync

This function copies n elements from a vector x in host memory space to a vector y in GPU memory space, asynchronously, on the given CUDA stream. It is assumed that each element requires storage of elemSize bytes. In CUDA Fortran, the type of vector x and y is overloaded to take any data type, but the size of the data type must still be specified in bytes. This functionality can also be implemented using cudaMemcpyAsync. 

 integer(4) function cublasgetvectorasync(n, elemsize, x, incx, y, incy, stream)
   integer :: n, elemsize, incx, incy
   integer*1, device, dimension(*) :: x
   integer*1, dimension(*) :: y
   integer(kind=cuda_stream_kind()) :: stream

2.1.19. cublasSetMatrixAsync

This function copies a tile of rows x cols elements from a matrix A in host memory space to a matrix B in GPU memory space, asynchronously using the specified stream. It is assumed that each element requires storage of elemSize bytes. In CUDA Fortran, the type of Matrix A and B is overloaded to take any data type, but the size of the data type must still be specified in bytes. This functionality can also be implemented using cudaMemcpyAsync or cudaMemcpy2DAsync. 

 integer(4) function cublassetmatrixasync(rows, cols, elemsize, a, lda, b, ldb, stream)
   integer :: rows, cols, elemsize, lda, ldb
   integer*1, dimension(lda, *) :: a
   integer*1, device, dimension(ldb, *) :: b
   integer(kind=cuda_stream_kind()) :: stream

2.1.20. cublasGetMatrixAsync

This function copies a tile of rows x cols elements from a matrix A in GPU memory space to a matrix B in host memory space, asynchronously, using the specified stream. It is assumed that each element requires storage of elemSize bytes. In CUDA Fortran, the type of Matrix A and B is overloaded to take any data type, but the size of the data type must still be specified in bytes. This functionality can also be implemented using cudaMemcpyAsync or cudaMemcpy2DAsync. 

 integer(4) function cublasgetmatrixasync(rows, cols, elemsize, a, lda, b, ldb, stream)
   integer :: rows, cols, elemsize, lda, ldb
   integer*1, device, dimension(lda, *) :: a
   integer*1, dimension(ldb, *) :: b
   integer(kind=cuda_stream_kind()) :: stream

2.2. Single Precision Functions and Subroutines

This section contains interfaces to the single precision BLAS and cuBLAS functions and subroutines.

2.2.1. isamax

ISAMAX finds the index of the element having the maximum absolute value. 

 integer(4) function isamax(n, x, incx)
   integer :: n
   real(4), device, dimension(*) :: x ! device or host variable
   integer :: incx 
 integer(4) function cublasIsamax(n, x, incx)
   integer :: n
   real(4), device, dimension(*) :: x
   integer :: incx 
 integer(4) function cublasIsamax_v2(h, n, x, incx, res)
   type(cublasHandle) :: h
   integer :: n
   real(4), device, dimension(*) :: x
   integer :: incx
   integer, device :: res ! device or host variable

2.2.2. isamin

ISAMIN finds the index of the element having the minimum absolute value. 

 integer(4) function isamin(n, x, incx)
   integer :: n
   real(4), device, dimension(*) :: x ! device or host variable
   integer :: incx 
 integer(4) function cublasIsamin(n, x, incx)
   integer :: n
   real(4), device, dimension(*) :: x
   integer :: incx 
 integer(4) function cublasIsamin_v2(h, n, x, incx, res)
   type(cublasHandle) :: h
   integer :: n
   real(4), device, dimension(*) :: x
   integer :: incx
   integer, device :: res ! device or host variable

2.2.3. sasum

SASUM takes the sum of the absolute values. 

 real(4) function sasum(n, x, incx)
   integer :: n
   real(4), device, dimension(*) :: x ! device or host variable
   integer :: incx 
 real(4) function cublasSasum(n, x, incx)
   integer :: n
   real(4), device, dimension(*) :: x
   integer :: incx 
 integer(4) function cublasSasum_v2(h, n, x, incx, res)
   type(cublasHandle) :: h
   integer :: n
   real(4), device, dimension(*) :: x
   integer :: incx
   real(4), device :: res ! device or host variable

2.2.4. saxpy

SAXPY constant times a vector plus a vector. 

 subroutine saxpy(n, a, x, incx, y, incy)
   integer :: n
   real(4), device :: a ! device or host variable
   real(4), device, dimension(*) :: x, y ! device or host variable
   integer :: incx, incy 
 subroutine cublasSaxpy(n, a, x, incx, y, incy)
   integer :: n
   real(4), device :: a ! device or host variable
   real(4), device, dimension(*) :: x, y
   integer :: incx, incy 
 integer(4) function cublasSaxpy_v2(h, n, a, x, incx, y, incy)
   type(cublasHandle) :: h
   integer :: n
   real(4), device :: a ! device or host variable
   real(4), device, dimension(*) :: x, y
   integer :: incx, incy

2.2.5. scopy

SCOPY copies a vector, x, to a vector, y. 

 subroutine scopy(n, x, incx, y, incy)
   integer :: n
   real(4), device, dimension(*) :: x, y ! device or host variable
   integer :: incx, incy 
 subroutine cublasScopy(n, x, incx, y, incy)
   integer :: n
   real(4), device, dimension(*) :: x, y
   integer :: incx, incy 
 integer(4) function cublasScopy_v2(h, n, x, incx, y, incy)
   type(cublasHandle) :: h
   integer :: n
   real(4), device, dimension(*) :: x, y
   integer :: incx, incy

2.2.6. sdot

SDOT forms the dot product of two vectors. 

 real(4) function sdot(n, x, incx, y, incy)
   integer :: n
   real(4), device, dimension(*) :: x, y ! device or host variable
   integer :: incx, incy 
 real(4) function cublasSdot(n, x, incx, y, incy)
   integer :: n
   real(4), device, dimension(*) :: x, y
   integer :: incx, incy 
 integer(4) function cublasSdot_v2(h, n, x, incx, y, incy, res)
   type(cublasHandle) :: h
   integer :: n
   real(4), device, dimension(*) :: x, y
   integer :: incx, incy
   real(4), device :: res ! device or host variable

2.2.7. snrm2

SNRM2 returns the euclidean norm of a vector via the function name, so that SNRM2 := sqrt( x'*x ). 

 real(4) function snrm2(n, x, incx)
   integer :: n
   real(4), device, dimension(*) :: x ! device or host variable
   integer :: incx 
 real(4) function cublasSnrm2(n, x, incx)
   integer :: n
   real(4), device, dimension(*) :: x
   integer :: incx 
 integer(4) function cublasSnrm2_v2(h, n, x, incx, res)
   type(cublasHandle) :: h
   integer :: n
   real(4), device, dimension(*) :: x
   integer :: incx
   real(4), device :: res ! device or host variable

2.2.8. srot

SROT applies a plane rotation. 

 subroutine srot(n, x, incx, y, incy, sc, ss)
   integer :: n
   real(4), device :: sc, ss ! device or host variable
   real(4), device, dimension(*) :: x, y ! device or host variable
   integer :: incx, incy 
 subroutine cublasSrot(n, x, incx, y, incy, sc, ss)
   integer :: n
   real(4), device :: sc, ss ! device or host variable
   real(4), device, dimension(*) :: x, y
   integer :: incx, incy 
 integer(4) function cublasSrot_v2(h, n, x, incx, y, incy, sc, ss)
   type(cublasHandle) :: h
   integer :: n
   real(4), device :: sc, ss ! device or host variable
   real(4), device, dimension(*) :: x, y
   integer :: incx, incy

2.2.9. srotg

SROTG constructs a Givens plane rotation. 

 subroutine srotg(sa, sb, sc, ss)
   real(4), device :: sa, sb, sc, ss ! device or host variable 
 subroutine cublasSrotg(sa, sb, sc, ss)
   real(4), device :: sa, sb, sc, ss ! device or host variable 
 integer(4) function cublasSrotg_v2(h, sa, sb, sc, ss)
   type(cublasHandle) :: h
   real(4), device :: sa, sb, sc, ss ! device or host variable

2.2.10. srotm

SROTM applies the modified Givens transformation, H, to the 2 by N matrix (SX**T) , where **T indicates transpose. The elements of SX are in (SX**T) SX(LX+I*INCX), I = 0 to N-1, where LX = 1 if INCX .GE. 0, ELSE LX = (-INCX)*N, and similarly for SY using LY and INCY. With SPARAM(1)=SFLAG, H has one of the following forms.. SFLAG=-1.E0 SFLAG=0.E0 SFLAG=1.E0 SFLAG=-2.E0 (SH11 SH12) (1.E0 SH12) (SH11 1.E0) (1.E0 0.E0) H=( ) ( ) ( ) ( ) (SH21 SH22), (SH21 1.E0), (-1.E0 SH22), (0.E0 1.E0). See SROTMG for a description of data storage in SPARAM. 

 subroutine srotm(n, x, incx, y, incy, param)
   integer :: n
   real(4), device, dimension(*) :: x, y ! device or host variable
   integer :: incx, incy 
 subroutine cublasSrotm(n, x, incx, y, incy, param)
   integer :: n
   real(4), device, dimension(*) :: x, y
   integer :: incx, incy
   real(4), device :: param(*) ! device or host variable 
 integer(4) function cublasSrotm_v2(h, n, x, incx, y, incy, param)
   type(cublasHandle) :: h
   integer :: n
   real(4), device :: param(*) ! device or host variable
   real(4), device, dimension(*) :: x, y
   integer :: incx, incy

2.2.11. srotmg

SROTMG constructs the modified Givens transformation matrix H which zeros the second component of the 2-vector (SQRT(SD1)*SX1,SQRT(SD2)*SY2)**T. With SPARAM(1)=SFLAG, H has one of the following forms.. SFLAG=-1.E0 SFLAG=0.E0 SFLAG=1.E0 SFLAG=-2.E0 (SH11 SH12) (1.E0 SH12) (SH11 1.E0) (1.E0 0.E0) H=( ) ( ) ( ) ( ) (SH21 SH22), (SH21 1.E0), (-1.E0 SH22), (0.E0 1.E0). Locations 2-4 of SPARAM contain SH11,SH21,SH12, and SH22 respectively. (Values of 1.E0, -1.E0, or 0.E0 implied by the value of SPARAM(1) are not stored in SPARAM.) 

 subroutine srotmg(d1, d2, x1, y1, param)
   real(4), device :: d1, d2, x1, y1, param(*) ! device or host variable 
 subroutine cublasSrotmg(d1, d2, x1, y1, param)
   real(4), device :: d1, d2, x1, y1, param(*) ! device or host variable 
 integer(4) function cublasSrotmg_v2(h, d1, d2, x1, y1, param)
   type(cublasHandle) :: h
   real(4), device :: d1, d2, x1, y1, param(*) ! device or host variable

2.2.12. sscal

SSCAL scales a vector by a constant. 

 subroutine sscal(n, a, x, incx)
   integer :: n
   real(4), device :: a ! device or host variable
   real(4), device, dimension(*) :: x ! device or host variable
   integer :: incx 
 subroutine cublasSscal(n, a, x, incx)
   integer :: n
   real(4), device :: a ! device or host variable
   real(4), device, dimension(*) :: x
   integer :: incx 
 integer(4) function cublasSscal_v2(h, n, a, x, incx)
   type(cublasHandle) :: h
   integer :: n
   real(4), device :: a ! device or host variable
   real(4), device, dimension(*) :: x
   integer :: incx

2.2.13. sswap

SSWAP interchanges two vectors. 

 subroutine sswap(n, x, incx, y, incy)
   integer :: n
   real(4), device, dimension(*) :: x, y ! device or host variable
   integer :: incx, incy 
 subroutine cublasSswap(n, x, incx, y, incy)
   integer :: n
   real(4), device, dimension(*) :: x, y
   integer :: incx, incy 
 integer(4) function cublasSswap_v2(h, n, x, incx, y, incy)
   type(cublasHandle) :: h
   integer :: n
   real(4), device, dimension(*) :: x, y
   integer :: incx, incy

2.2.14. sgbmv

SGBMV performs one of the matrix-vector operations y := alpha*A*x + beta*y, or y := alpha*A**T*x + beta*y, where alpha and beta are scalars, x and y are vectors and A is an m by n band matrix, with kl sub-diagonals and ku super-diagonals. 

 subroutine sgbmv(t, m, n, kl, ku, alpha, a, lda, x, incx, beta, y, incy)
   character*1 :: t
   integer :: m, n, kl, ku, lda, incx, incy
   real(4), device, dimension(lda, *) :: a ! device or host variable
   real(4), device, dimension(*) :: x, y ! device or host variable
   real(4), device :: alpha, beta ! device or host variable 
 subroutine cublasSgbmv(t, m, n, kl, ku, alpha, a, lda, x, incx, beta, y, incy)
   character*1 :: t
   integer :: m, n, kl, ku, lda, incx, incy
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(*) :: x, y
   real(4), device :: alpha, beta ! device or host variable 
 integer(4) function cublasSgbmv_v2(h, t, m, n, kl, ku, alpha, a, lda, x, incx, beta, y, incy)
   type(cublasHandle) :: h
   integer :: t
   integer :: m, n, kl, ku, lda, incx, incy
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(*) :: x, y
   real(4), device :: alpha, beta ! device or host variable

2.2.15. sgemv

SGEMV performs one of the matrix-vector operations y := alpha*A*x + beta*y, or y := alpha*A**T*x + beta*y, where alpha and beta are scalars, x and y are vectors and A is an m by n matrix. 

 subroutine sgemv(t, m, n, alpha, a, lda, x, incx, beta, y, incy)
   character*1 :: t
   integer :: m, n, lda, incx, incy
   real(4), device, dimension(lda, *) :: a ! device or host variable
   real(4), device, dimension(*) :: x, y ! device or host variable
   real(4), device :: alpha, beta ! device or host variable 
 subroutine cublasSgemv(t, m, n, alpha, a, lda, x, incx, beta, y, incy)
   character*1 :: t
   integer :: m, n, lda, incx, incy
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(*) :: x, y
   real(4), device :: alpha, beta ! device or host variable 
 integer(4) function cublasSgemv_v2(h, t, m, n, alpha, a, lda, x, incx, beta, y, incy)
   type(cublasHandle) :: h
   integer :: t
   integer :: m, n, lda, incx, incy
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(*) :: x, y
   real(4), device :: alpha, beta ! device or host variable

2.2.16. sger

SGER performs the rank 1 operation A := alpha*x*y**T + A, where alpha is a scalar, x is an m element vector, y is an n element vector and A is an m by n matrix. 

 subroutine sger(m, n, alpha, x, incx, y, incy, a, lda)
   integer :: m, n, lda, incx, incy
   real(4), device, dimension(lda, *) :: a ! device or host variable
   real(4), device, dimension(*) :: x, y ! device or host variable
   real(4), device :: alpha ! device or host variable 
 subroutine cublasSger(m, n, alpha, x, incx, y, incy, a, lda)
   integer :: m, n, lda, incx, incy
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(*) :: x, y
   real(4), device :: alpha ! device or host variable 
 integer(4) function cublasSger_v2(h, m, n, alpha, x, incx, y, incy, a, lda)
   type(cublasHandle) :: h
   integer :: m, n, lda, incx, incy
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(*) :: x, y
   real(4), device :: alpha ! device or host variable

2.2.17. ssbmv

SSBMV performs the matrix-vector operation y := alpha*A*x + beta*y, where alpha and beta are scalars, x and y are n element vectors and A is an n by n symmetric band matrix, with k super-diagonals. 

 subroutine ssbmv(t, n, k, alpha, a, lda, x, incx, beta, y, incy)
   character*1 :: t
   integer :: k, n, lda, incx, incy
   real(4), device, dimension(lda, *) :: a ! device or host variable
   real(4), device, dimension(*) :: x, y ! device or host variable
   real(4), device :: alpha, beta ! device or host variable 
 subroutine cublasSsbmv(t, n, k, alpha, a, lda, x, incx, beta, y, incy)
   character*1 :: t
   integer :: k, n, lda, incx, incy
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(*) :: x, y
   real(4), device :: alpha, beta ! device or host variable 
 integer(4) function cublasSsbmv_v2(h, t, n, k, alpha, a, lda, x, incx, beta, y, incy)
   type(cublasHandle) :: h
   integer :: t
   integer :: k, n, lda, incx, incy
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(*) :: x, y
   real(4), device :: alpha, beta ! device or host variable

2.2.18. sspmv

SSPMV performs the matrix-vector operation y := alpha*A*x + beta*y, where alpha and beta are scalars, x and y are n element vectors and A is an n by n symmetric matrix, supplied in packed form. 

 subroutine sspmv(t, n, alpha, a, x, incx, beta, y, incy)
   character*1 :: t
   integer :: n, incx, incy
   real(4), device, dimension(*) :: a, x, y ! device or host variable
   real(4), device :: alpha, beta ! device or host variable 
 subroutine cublasSspmv(t, n, alpha, a, x, incx, beta, y, incy)
   character*1 :: t
   integer :: n, incx, incy
   real(4), device, dimension(*) :: a, x, y
   real(4), device :: alpha, beta ! device or host variable 
 integer(4) function cublasSspmv_v2(h, t, n, alpha, a, x, incx, beta, y, incy)
   type(cublasHandle) :: h
   integer :: t
   integer :: n, incx, incy
   real(4), device, dimension(*) :: a, x, y
   real(4), device :: alpha, beta ! device or host variable

2.2.19. sspr

SSPR performs the symmetric rank 1 operation A := alpha*x*x**T + A, where alpha is a real scalar, x is an n element vector and A is an n by n symmetric matrix, supplied in packed form. 

 subroutine sspr(t, n, alpha, x, incx, a)
   character*1 :: t
   integer :: n, incx
   real(4), device, dimension(*) :: a, x ! device or host variable
   real(4), device :: alpha ! device or host variable 
 subroutine cublasSspr(t, n, alpha, x, incx, a)
   character*1 :: t
   integer :: n, incx
   real(4), device, dimension(*) :: a, x
   real(4), device :: alpha ! device or host variable 
 integer(4) function cublasSspr_v2(h, t, n, alpha, x, incx, a)
   type(cublasHandle) :: h
   integer :: t
   integer :: n, incx
   real(4), device, dimension(*) :: a, x
   real(4), device :: alpha ! device or host variable

2.2.20. sspr2

SSPR2 performs the symmetric rank 2 operation A := alpha*x*y**T + alpha*y*x**T + A, where alpha is a scalar, x and y are n element vectors and A is an n by n symmetric matrix, supplied in packed form. 

 subroutine sspr2(t, n, alpha, x, incx, y, incy, a)
   character*1 :: t
   integer :: n, incx, incy
   real(4), device, dimension(*) :: a, x, y ! device or host variable
   real(4), device :: alpha ! device or host variable 
 subroutine cublasSspr2(t, n, alpha, x, incx, y, incy, a)
   character*1 :: t
   integer :: n, incx, incy
   real(4), device, dimension(*) :: a, x, y
   real(4), device :: alpha ! device or host variable 
 integer(4) function cublasSspr2_v2(h, t, n, alpha, x, incx, y, incy, a)
   type(cublasHandle) :: h
   integer :: t
   integer :: n, incx, incy
   real(4), device, dimension(*) :: a, x, y
   real(4), device :: alpha ! device or host variable

2.2.21. ssymv

SSYMV performs the matrix-vector operation y := alpha*A*x + beta*y, where alpha and beta are scalars, x and y are n element vectors and A is an n by n symmetric matrix. 

 subroutine ssymv(uplo, n, alpha, a, lda, x, incx, beta, y, incy)
   character*1 :: uplo
   integer :: n, lda, incx, incy
   real(4), device, dimension(lda, *) :: a ! device or host variable
   real(4), device, dimension(*) :: x, y ! device or host variable
   real(4), device :: alpha, beta ! device or host variable 
 subroutine cublasSsymv(uplo, n, alpha, a, lda, x, incx, beta, y, incy)
   character*1 :: uplo
   integer :: n, lda, incx, incy
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(*) :: x, y
   real(4), device :: alpha, beta ! device or host variable 
 integer(4) function cublasSsymv_v2(h, uplo, n, alpha, a, lda, x, incx, beta, y, incy)
   type(cublasHandle) :: h
   integer :: uplo
   integer :: n, lda, incx, incy
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(*) :: x, y
   real(4), device :: alpha, beta ! device or host variable

2.2.22. ssyr

SSYR performs the symmetric rank 1 operation A := alpha*x*x**T + A, where alpha is a real scalar, x is an n element vector and A is an n by n symmetric matrix. 

 subroutine ssyr(t, n, alpha, x, incx, a, lda)
   character*1 :: t
   integer :: n, incx, lda
   real(4), device, dimension(lda, *) :: a ! device or host variable
   real(4), device, dimension(*) :: x ! device or host variable
   real(4), device :: alpha ! device or host variable 
 subroutine cublasSsyr(t, n, alpha, x, incx, a, lda)
   character*1 :: t
   integer :: n, incx, lda
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(*) :: x
   real(4), device :: alpha ! device or host variable 
 integer(4) function cublasSsyr_v2(h, t, n, alpha, x, incx, a, lda)
   type(cublasHandle) :: h
   integer :: t
   integer :: n, incx, lda
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(*) :: x
   real(4), device :: alpha ! device or host variable

2.2.23. ssyr2

SSYR2 performs the symmetric rank 2 operation A := alpha*x*y**T + alpha*y*x**T + A, where alpha is a scalar, x and y are n element vectors and A is an n by n symmetric matrix. 

 subroutine ssyr2(t, n, alpha, x, incx, y, incy, a, lda)
   character*1 :: t
   integer :: n, incx, incy, lda
   real(4), device, dimension(lda, *) :: a ! device or host variable
   real(4), device, dimension(*) :: x, y ! device or host variable
   real(4), device :: alpha ! device or host variable 
 subroutine cublasSsyr2(t, n, alpha, x, incx, y, incy, a, lda)
   character*1 :: t
   integer :: n, incx, incy, lda
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(*) :: x, y
   real(4), device :: alpha ! device or host variable 
 integer(4) function cublasSsyr2_v2(h, t, n, alpha, x, incx, y, incy, a, lda)
   type(cublasHandle) :: h
   integer :: t
   integer :: n, incx, incy, lda
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(*) :: x, y
   real(4), device :: alpha ! device or host variable

2.2.24. stbmv

STBMV performs one of the matrix-vector operations x := A*x, or x := A**T*x, where x is an n element vector and A is an n by n unit, or non-unit, upper or lower triangular band matrix, with ( k + 1 ) diagonals. 

 subroutine stbmv(u, t, d, n, k, a, lda, x, incx)
   character*1 :: u, t, d
   integer :: n, k, incx, lda
   real(4), device, dimension(lda, *) :: a ! device or host variable
   real(4), device, dimension(*) :: x ! device or host variable 
 subroutine cublasStbmv(u, t, d, n, k, a, lda, x, incx)
   character*1 :: u, t, d
   integer :: n, k, incx, lda
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(*) :: x 
 integer(4) function cublasStbmv_v2(h, u, t, d, n, k, a, lda, x, incx)
   type(cublasHandle) :: h
   integer :: u, t, d
   integer :: n, k, incx, lda
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(*) :: x

2.2.25. stbsv

STBSV solves one of the systems of equations A*x = b, or A**T*x = b, where b and x are n element vectors and A is an n by n unit, or non-unit, upper or lower triangular band matrix, with ( k + 1 ) diagonals. No test for singularity or near-singularity is included in this routine. Such tests must be performed before calling this routine. 

 subroutine stbsv(u, t, d, n, k, a, lda, x, incx)
   character*1 :: u, t, d
   integer :: n, k, incx, lda
   real(4), device, dimension(lda, *) :: a ! device or host variable
   real(4), device, dimension(*) :: x ! device or host variable 
 subroutine cublasStbsv(u, t, d, n, k, a, lda, x, incx)
   character*1 :: u, t, d
   integer :: n, k, incx, lda
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(*) :: x 
 integer(4) function cublasStbsv_v2(h, u, t, d, n, k, a, lda, x, incx)
   type(cublasHandle) :: h
   integer :: u, t, d
   integer :: n, k, incx, lda
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(*) :: x

2.2.26. stpmv

STPMV performs one of the matrix-vector operations x := A*x, or x := A**T*x, where x is an n element vector and A is an n by n unit, or non-unit, upper or lower triangular matrix, supplied in packed form. 

 subroutine stpmv(u, t, d, n, a, x, incx)
   character*1 :: u, t, d
   integer :: n, incx
   real(4), device, dimension(*) :: a, x ! device or host variable 
 subroutine cublasStpmv(u, t, d, n, a, x, incx)
   character*1 :: u, t, d
   integer :: n, incx
   real(4), device, dimension(*) :: a, x 
 integer(4) function cublasStpmv_v2(h, u, t, d, n, a, x, incx)
   type(cublasHandle) :: h
   integer :: u, t, d
   integer :: n, incx
   real(4), device, dimension(*) :: a, x

2.2.27. stpsv

STPSV solves one of the systems of equations A*x = b, or A**T*x = b, where b and x are n element vectors and A is an n by n unit, or non-unit, upper or lower triangular matrix, supplied in packed form. No test for singularity or near-singularity is included in this routine. Such tests must be performed before calling this routine. 

 subroutine stpsv(u, t, d, n, a, x, incx)
   character*1 :: u, t, d
   integer :: n, incx
   real(4), device, dimension(*) :: a, x ! device or host variable 
 subroutine cublasStpsv(u, t, d, n, a, x, incx)
   character*1 :: u, t, d
   integer :: n, incx
   real(4), device, dimension(*) :: a, x 
 integer(4) function cublasStpsv_v2(h, u, t, d, n, a, x, incx)
   type(cublasHandle) :: h
   integer :: u, t, d
   integer :: n, incx
   real(4), device, dimension(*) :: a, x

2.2.28. strmv

STRMV performs one of the matrix-vector operations x := A*x, or x := A**T*x, where x is an n element vector and A is an n by n unit, or non-unit, upper or lower triangular matrix. 

 subroutine strmv(u, t, d, n, a, lda, x, incx)
   character*1 :: u, t, d
   integer :: n, incx, lda
   real(4), device, dimension(lda, *) :: a ! device or host variable
   real(4), device, dimension(*) :: x ! device or host variable 
 subroutine cublasStrmv(u, t, d, n, a, lda, x, incx)
   character*1 :: u, t, d
   integer :: n, incx, lda
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(*) :: x 
 integer(4) function cublasStrmv_v2(h, u, t, d, n, a, lda, x, incx)
   type(cublasHandle) :: h
   integer :: u, t, d
   integer :: n, incx, lda
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(*) :: x

2.2.29. strsv

STRSV solves one of the systems of equations A*x = b, or A**T*x = b, where b and x are n element vectors and A is an n by n unit, or non-unit, upper or lower triangular matrix. No test for singularity or near-singularity is included in this routine. Such tests must be performed before calling this routine. 

 subroutine strsv(u, t, d, n, a, lda, x, incx)
   character*1 :: u, t, d
   integer :: n, incx, lda
   real(4), device, dimension(lda, *) :: a ! device or host variable
   real(4), device, dimension(*) :: x ! device or host variable 
 subroutine cublasStrsv(u, t, d, n, a, lda, x, incx)
   character*1 :: u, t, d
   integer :: n, incx, lda
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(*) :: x 
 integer(4) function cublasStrsv_v2(h, u, t, d, n, a, lda, x, incx)
   type(cublasHandle) :: h
   integer :: u, t, d
   integer :: n, incx, lda
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(*) :: x

2.2.30. sgemm

SGEMM performs one of the matrix-matrix operations C := alpha*op( A )*op( B ) + beta*C, where op( X ) is one of op( X ) = X or op( X ) = X**T, alpha and beta are scalars, and A, B and C are matrices, with op( A ) an m by k matrix, op( B ) a k by n matrix and C an m by n matrix. 

 subroutine sgemm(transa, transb, m, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   character*1 :: transa, transb
   integer :: m, n, k, lda, ldb, ldc
   real(4), device, dimension(lda, *) :: a ! device or host variable
   real(4), device, dimension(ldb, *) :: b ! device or host variable
   real(4), device, dimension(ldc, *) :: c ! device or host variable
   real(4), device :: alpha, beta ! device or host variable 
 subroutine cublasSgemm(transa, transb, m, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   character*1 :: transa, transb
   integer :: m, n, k, lda, ldb, ldc
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(ldb, *) :: b
   real(4), device, dimension(ldc, *) :: c
   real(4), device :: alpha, beta ! device or host variable 
 integer(4) function cublasSgemm_v2(h, transa, transb, m, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: transa, transb
   integer :: m, n, k, lda, ldb, ldc
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(ldb, *) :: b
   real(4), device, dimension(ldc, *) :: c
   real(4), device :: alpha, beta ! device or host variable

2.2.31. ssymm

SSYMM performs one of the matrix-matrix operations C := alpha*A*B + beta*C, or C := alpha*B*A + beta*C, where alpha and beta are scalars, A is a symmetric matrix and B and C are m by n matrices. 

 subroutine ssymm(side, uplo, m, n, alpha, a, lda, b, ldb, beta, c, ldc)
   character*1 :: side, uplo
   integer :: m, n, lda, ldb, ldc
   real(4), device, dimension(lda, *) :: a ! device or host variable
   real(4), device, dimension(ldb, *) :: b ! device or host variable
   real(4), device, dimension(ldc, *) :: c ! device or host variable
   real(4), device :: alpha, beta ! device or host variable 
 subroutine cublasSsymm(side, uplo, m, n, alpha, a, lda, b, ldb, beta, c, ldc)
   character*1 :: side, uplo
   integer :: m, n, lda, ldb, ldc
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(ldb, *) :: b
   real(4), device, dimension(ldc, *) :: c
   real(4), device :: alpha, beta ! device or host variable 
 integer(4) function cublasSsymm_v2(h, side, uplo, m, n, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: side, uplo
   integer :: m, n, lda, ldb, ldc
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(ldb, *) :: b
   real(4), device, dimension(ldc, *) :: c
   real(4), device :: alpha, beta ! device or host variable

2.2.32. ssyrk

SSYRK performs one of the symmetric rank k operations C := alpha*A*A**T + beta*C, or C := alpha*A**T*A + beta*C, where alpha and beta are scalars, C is an n by n symmetric matrix and A is an n by k matrix in the first case and a k by n matrix in the second case. 

 subroutine ssyrk(uplo, trans, n, k, alpha, a, lda, beta, c, ldc)
   character*1 :: uplo, trans
   integer :: n, k, lda, ldc
   real(4), device, dimension(lda, *) :: a ! device or host variable
   real(4), device, dimension(ldc, *) :: c ! device or host variable
   real(4), device :: alpha, beta ! device or host variable 
 subroutine cublasSsyrk(uplo, trans, n, k, alpha, a, lda, beta, c, ldc)
   character*1 :: uplo, trans
   integer :: n, k, lda, ldc
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(ldc, *) :: c
   real(4), device :: alpha, beta ! device or host variable 
 integer(4) function cublasSsyrk_v2(h, uplo, trans, n, k, alpha, a, lda, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: uplo, trans
   integer :: n, k, lda, ldc
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(ldc, *) :: c
   real(4), device :: alpha, beta ! device or host variable

2.2.33. ssyr2k

SSYR2K performs one of the symmetric rank 2k operations C := alpha*A*B**T + alpha*B*A**T + beta*C, or C := alpha*A**T*B + alpha*B**T*A + beta*C, where alpha and beta are scalars, C is an n by n symmetric matrix and A and B are n by k matrices in the first case and k by n matrices in the second case. 

 subroutine ssyr2k(uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   character*1 :: uplo, trans
   integer :: n, k, lda, ldb, ldc
   real(4), device, dimension(lda, *) :: a ! device or host variable
   real(4), device, dimension(ldb, *) :: b ! device or host variable
   real(4), device, dimension(ldc, *) :: c ! device or host variable
   real(4), device :: alpha, beta ! device or host variable 
 subroutine cublasSsyr2k(uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   character*1 :: uplo, trans
   integer :: n, k, lda, ldb, ldc
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(ldb, *) :: b
   real(4), device, dimension(ldc, *) :: c
   real(4), device :: alpha, beta ! device or host variable 
 integer(4) function cublasSsyr2k_v2(h, uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: uplo, trans
   integer :: n, k, lda, ldb, ldc
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(ldb, *) :: b
   real(4), device, dimension(ldc, *) :: c
   real(4), device :: alpha, beta ! device or host variable

2.2.34. ssyrkx

SSYRKX performs a variation of the symmetric rank k update C := alpha*A*B**T + beta*C, where alpha and beta are scalars, C is an n by n symmetric matrix stored in lower or upper mode, and A and B are n by k matrices. This routine can be used when B is in such a way that the result is guaranteed to be symmetric. See the CUBLAS documentation for more details. 

 subroutine ssyrkx(uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   character*1 :: uplo, trans
   integer :: n, k, lda, ldb, ldc
   real(4), device, dimension(lda, *) :: a ! device or host variable
   real(4), device, dimension(ldb, *) :: b ! device or host variable
   real(4), device, dimension(ldc, *) :: c ! device or host variable
   real(4), device :: alpha, beta ! device or host variable 
 subroutine cublasSsyrkx(uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   character*1 :: uplo, trans
   integer :: n, k, lda, ldb, ldc
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(ldb, *) :: b
   real(4), device, dimension(ldc, *) :: c
   real(4), device :: alpha, beta ! device or host variable 
 integer(4) function cublasSsyrkx_v2(h, uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: uplo, trans
   integer :: n, k, lda, ldb, ldc
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(ldb, *) :: b
   real(4), device, dimension(ldc, *) :: c
   real(4), device :: alpha, beta ! device or host variable

2.2.35. strmm

STRMM performs one of the matrix-matrix operations B := alpha*op( A )*B, or B := alpha*B*op( A ), where alpha is a scalar, B is an m by n matrix, A is a unit, or non-unit, upper or lower triangular matrix and op( A ) is one of op( A ) = A or op( A ) = A**T. 

 subroutine strmm(side, uplo, transa, diag, m, n, alpha, a, lda, b, ldb)
   character*1 :: side, uplo, transa, diag
   integer :: m, n, lda, ldb
   real(4), device, dimension(lda, *) :: a ! device or host variable
   real(4), device, dimension(ldb, *) :: b ! device or host variable
   real(4), device :: alpha ! device or host variable 
 subroutine cublasStrmm(side, uplo, transa, diag, m, n, alpha, a, lda, b, ldb)
   character*1 :: side, uplo, transa, diag
   integer :: m, n, lda, ldb
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(ldb, *) :: b
   real(4), device :: alpha ! device or host variable 
 integer(4) function cublasStrmm_v2(h, side, uplo, transa, diag, m, n, alpha, a, lda, b, ldb, c, ldc)
   type(cublasHandle) :: h
   integer :: side, uplo, transa, diag
   integer :: m, n, lda, ldb, ldc
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(ldb, *) :: b
   real(4), device, dimension(ldc, *) :: c
   real(4), device :: alpha ! device or host variable

2.2.36. strsm

STRSM solves one of the matrix equations op( A )*X = alpha*B, or X*op( A ) = alpha*B, where alpha is a scalar, X and B are m by n matrices, A is a unit, or non-unit, upper or lower triangular matrix and op( A ) is one of op( A ) = A or op( A ) = A**T. The matrix X is overwritten on B. 

 subroutine strsm(side, uplo, transa, diag, m, n, alpha, a, lda, b, ldb)
   character*1 :: side, uplo, transa, diag
   integer :: m, n, lda, ldb
   real(4), device, dimension(lda, *) :: a ! device or host variable
   real(4), device, dimension(ldb, *) :: b ! device or host variable
   real(4), device :: alpha ! device or host variable 
 subroutine cublasStrsm(side, uplo, transa, diag, m, n, alpha, a, lda, b, ldb)
   character*1 :: side, uplo, transa, diag
   integer :: m, n, lda, ldb
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(ldb, *) :: b
   real(4), device :: alpha ! device or host variable 
 integer(4) function cublasStrsm_v2(h, side, uplo, transa, diag, m, n, alpha, a, lda, b, ldb)
   type(cublasHandle) :: h
   integer :: side, uplo, transa, diag
   integer :: m, n, lda, ldb
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(ldb, *) :: b
   real(4), device :: alpha ! device or host variable

2.2.37. cublasSgetrfBatched

SGETRF computes an LU factorization of a general M-by-N matrix A using partial pivoting with row interchanges. The factorization has the form A = P * L * U where P is a permutation matrix, L is lower triangular with unit diagonal elements (lower trapezoidal if m > n), and U is upper triangular (upper trapezoidal if m < n). This is the right-looking Level 3 BLAS version of the algorithm. 

 integer(4) function cublasSgetrfBatched(h, n, Aarray, lda, ipvt, info, batchCount)
   type(cublasHandle) :: h
   integer :: n
   type(c_devptr), device :: Aarray(*)
   integer :: lda
   integer, device :: ipvt(*)
   integer, device :: info(*)
   integer :: batchCount

2.2.38. cublasSgetriBatched

SGETRI computes the inverse of a matrix using the LU factorization computed by SGETRF. This method inverts U and then computes inv(A) by solving the system inv(A)*L = inv(U) for inv(A). 

 integer(4) function cublasSgetriBatched(h, n, Aarray, lda, ipvt, Carray, ldc, info, batchCount)
   type(cublasHandle) :: h
   integer :: n
   type(c_devptr), device :: Aarray(*)
   integer :: lda
   integer, device :: ipvt(*)
   type(c_devptr), device :: Carray(*)
   integer :: ldc
   integer, device :: info(*)
   integer :: batchCount

2.2.39. cublasSgetrsBatched

SGETRS solves a system of linear equations A * X = B or A**T * X = B with a general N-by-N matrix A using the LU factorization computed by SGETRF. 

 integer(4) function cublasSgetrsBatched(h, trans, n, nrhs, Aarray, lda, ipvt, Barray, ldb, info, batchCount)
   type(cublasHandle) :: h
   integer :: trans ! integer or character(1) variable
   integer :: n, nrhs
   type(c_devptr), device :: Aarray(*)
   integer :: lda
   integer, device :: ipvt(*)
   type(c_devptr), device :: Barray(*)
   integer :: ldb
   integer :: info(*)
   integer :: batchCount

2.2.40. cublasSgemmBatched

SGEMM performs one of the matrix-matrix operations C := alpha*op( A )*op( B ) + beta*C, where op( X ) is one of op( X ) = X or op( X ) = X**T, alpha and beta are scalars, and A, B and C are matrices, with op( A ) an m by k matrix, op( B ) a k by n matrix and C an m by n matrix. 

 integer(4) function cublasSgemmBatched(h, transa, transb, m, n, k, alpha, Aarray, lda, Barray, ldb, beta, Carray, ldc, batchCount)
   type(cublasHandle) :: h
   integer :: transa ! integer or character(1) variable
   integer :: transb ! integer or character(1) variable
   integer :: m, n, k
   real(4), device :: alpha ! device or host variable
   type(c_devptr), device :: Aarray(*)
   integer :: lda
   type(c_devptr), device :: Barray(*)
   integer :: ldb
   real(4), device :: beta ! device or host variable
   type(c_devptr), device :: Carray(*)
   integer :: ldc
   integer :: batchCount 
 integer(4) function cublasSgemmBatched_v2(h, transa, transb, m, n, k, alpha,           Aarray, lda, Barray, ldb, beta, Carray, ldc, batchCount)
   type(cublasHandle) :: h
   integer :: transa
   integer :: transb
   integer :: m, n, k
   real(4), device :: alpha ! device or host variable
   type(c_devptr), device :: Aarray(*)
   integer :: lda
   type(c_devptr), device :: Barray(*)
   integer :: ldb
   real(4), device :: beta ! device or host variable
   type(c_devptr), device :: Carray(*)
   integer :: ldc
   integer :: batchCount

2.2.41. cublasStrsmBatched

STRSM solves one of the matrix equations op( A )*X = alpha*B, or X*op( A ) = alpha*B, where alpha is a scalar, X and B are m by n matrices, A is a unit, or non-unit, upper or lower triangular matrix and op( A ) is one of op( A ) = A or op( A ) = A**T. The matrix X is overwritten on B. 

 integer(4) function cublasStrsmBatched( h, side, uplo, trans, diag, m, n, alpha, A, lda, B, ldb, batchCount)
   type(cublasHandle) :: h
   integer :: side ! integer or character(1) variable
   integer :: uplo ! integer or character(1) variable
   integer :: trans ! integer or character(1) variable
   integer :: diag ! integer or character(1) variable
   integer :: m, n
   real(4), device :: alpha ! device or host variable
   type(c_devptr), device :: A(*)
   integer :: lda
   type(c_devptr), device :: B(*)
   integer :: ldb
   integer :: batchCount 
 integer(4) function cublasStrsmBatched_v2( h, side, uplo, trans, diag, m, n, alpha, A, lda, B, ldb, batchCount)
   type(cublasHandle) :: h
   integer :: side
   integer :: uplo
   integer :: trans
   integer :: diag
   integer :: m, n
   real(4), device :: alpha ! device or host variable
   type(c_devptr), device :: A(*)
   integer :: lda
   type(c_devptr), device :: B(*)
   integer :: ldb
   integer :: batchCount

2.2.42. cublasSmatinvBatched

cublasSmatinvBatched is a short cut of cublasSgetrfBatched plus cublasSgetriBatched. However it only works if n is less than 32. If not, the user has to go through cublasSgetrfBatched and cublasSgetriBatched. 

 integer(4) function cublasSmatinvBatched(h, n, Aarray, lda, Ainv, lda_inv, info, batchCount)
   type(cublasHandle) :: h
   integer :: n
   type(c_devptr), device :: Aarray(*)
   integer :: lda
   type(c_devptr), device :: Ainv(*)
   integer :: lda_inv
   integer, device :: info(*)
   integer :: batchCount

2.2.43. cublasSgeqrfBatched

SGEQRF computes a QR factorization of a real M-by-N matrix A: A = Q * R. 

 integer(4) function cublasSgeqrfBatched(h, m, n, Aarray, lda, Tau, info, batchCount)
   type(cublasHandle) :: h
   integer :: m, n
   type(c_devptr), device :: Aarray(*)
   integer :: lda
   type(c_devptr), device :: Tau(*)
   integer :: info(*)
   integer :: batchCount

2.2.44. cublasSgelsBatched

SGELS solves overdetermined or underdetermined real linear systems involving an M-by-N matrix A, or its transpose, using a QR or LQ factorization of A. It is assumed that A has full rank. The following options are provided: 1. If TRANS = 'N' and m >= n: find the least squares solution of an overdetermined system, i.e., solve the least squares problem minimize || B - A*X ||. 2. If TRANS = 'N' and m < n: find the minimum norm solution of an underdetermined system A * X = B. 3. If TRANS = 'T' and m >= n: find the minimum norm solution of an undetermined system A**T * X = B. 4. If TRANS = 'T' and m < n: find the least squares solution of an overdetermined system, i.e., solve the least squares problem minimize || B - A**T * X ||. Several right hand side vectors b and solution vectors x can be handled in a single call; they are stored as the columns of the M-by-NRHS right hand side matrix B and the N-by-NRHS solution matrix X. 

 integer(4) function cublasSgelsBatched(h, trans, m, n, nrhs, Aarray, lda, Carray, ldc, info, devinfo, batchCount)
   type(cublasHandle) :: h
   integer :: trans ! integer or character(1) variable
   integer :: m, n, nrhs
   type(c_devptr), device :: Aarray(*)
   integer :: lda
   type(c_devptr), device :: Carray(*)
   integer :: ldc
   integer :: info(*)
   integer, device :: devinfo(*)
   integer :: batchCount

2.2.45. cublasSgemmStridedBatched

SGEMM performs one of the matrix-matrix operations C := alpha*op( A )*op( B ) + beta*C, where op( X ) is one of op( X ) = X or op( X ) = X**T, alpha and beta are scalars, and A, B and C are matrices, with op( A ) an m by k matrix, op( B ) a k by n matrix and C an m by n matrix. 

 integer(4) function cublasSgemmStridedBatched(h, transa, transb, m, n, k, alpha, Aarray, lda, strideA, Barray, ldb, strideB, beta, Carray, ldc, strideC, batchCount)
   type(cublasHandle) :: h
   integer :: transa ! integer or character(1) variable
   integer :: transb ! integer or character(1) variable
   integer :: m, n, k
   real(4), device :: alpha ! device or host variable
   real(4), device :: Aarray(*)
   integer :: lda
   integer :: strideA
   real(4), device :: Barray(*)
   integer :: ldb
   integer :: strideB
   real(4), device :: beta ! device or host variable
   real(4), device :: Carray(*)
   integer :: ldc
   integer :: strideC
   integer :: batchCount 
 integer(4) function cublasSgemmStridedBatched_v2(h, transa, transb, m, n, k, alpha,           Aarray, lda, strideA, Barray, ldb, strideB, beta, Carray, ldc, strideC, batchCount)
   type(cublasHandle) :: h
   integer :: transa
   integer :: transb
   integer :: m, n, k
   real(4), device :: alpha ! device or host variable
   real(4), device :: Aarray(*)
   integer :: lda
   integer :: strideA
   real(4), device :: Barray(*)
   integer :: ldb
   integer :: strideB
   real(4), device :: beta ! device or host variable
   real(4), device :: Carray(*)
   integer :: ldc
   integer :: strideC
   integer :: batchCount

2.3. Double Precision Functions and Subroutines

This section contains interfaces to the double precision BLAS and cuBLAS functions and subroutines.

2.3.1. idamax

IDAMAX finds the the index of the element having the maximum absolute value. 

 integer(4) function idamax(n, x, incx)
   integer :: n
   real(8), device, dimension(*) :: x ! device or host variable
   integer :: incx 
 integer(4) function cublasIdamax(n, x, incx)
   integer :: n
   real(8), device, dimension(*) :: x
   integer :: incx 
 integer(4) function cublasIdamax_v2(h, n, x, incx, res)
   type(cublasHandle) :: h
   integer :: n
   real(8), device, dimension(*) :: x
   integer :: incx
   integer, device :: res ! device or host variable

2.3.2. idamin

IDAMIN finds the index of the element having the minimum absolute value. 

 integer(4) function idamin(n, x, incx)
   integer :: n
   real(8), device, dimension(*) :: x ! device or host variable
   integer :: incx 
 integer(4) function cublasIdamin(n, x, incx)
   integer :: n
   real(8), device, dimension(*) :: x
   integer :: incx 
 integer(4) function cublasIdamin_v2(h, n, x, incx, res)
   type(cublasHandle) :: h
   integer :: n
   real(8), device, dimension(*) :: x
   integer :: incx
   integer, device :: res ! device or host variable

2.3.3. dasum

DASUM takes the sum of the absolute values. 

 real(8) function dasum(n, x, incx)
   integer :: n
   real(8), device, dimension(*) :: x ! device or host variable
   integer :: incx 
 real(8) function cublasDasum(n, x, incx)
   integer :: n
   real(8), device, dimension(*) :: x
   integer :: incx 
 integer(4) function cublasDasum_v2(h, n, x, incx, res)
   type(cublasHandle) :: h
   integer :: n
   real(8), device, dimension(*) :: x
   integer :: incx
   real(8), device :: res ! device or host variable

2.3.4. daxpy

DAXPY constant times a vector plus a vector. 

 subroutine daxpy(n, a, x, incx, y, incy)
   integer :: n
   real(8), device :: a ! device or host variable
   real(8), device, dimension(*) :: x, y ! device or host variable
   integer :: incx, incy 
 subroutine cublasDaxpy(n, a, x, incx, y, incy)
   integer :: n
   real(8), device :: a ! device or host variable
   real(8), device, dimension(*) :: x, y
   integer :: incx, incy 
 integer(4) function cublasDaxpy_v2(h, n, a, x, incx, y, incy)
   type(cublasHandle) :: h
   integer :: n
   real(8), device :: a ! device or host variable
   real(8), device, dimension(*) :: x, y
   integer :: incx, incy

2.3.5. dcopy

DCOPY copies a vector, x, to a vector, y. 

 subroutine dcopy(n, x, incx, y, incy)
   integer :: n
   real(8), device, dimension(*) :: x, y ! device or host variable
   integer :: incx, incy 
 subroutine cublasDcopy(n, x, incx, y, incy)
   integer :: n
   real(8), device, dimension(*) :: x, y
   integer :: incx, incy 
 integer(4) function cublasDcopy_v2(h, n, x, incx, y, incy)
   type(cublasHandle) :: h
   integer :: n
   real(8), device, dimension(*) :: x, y
   integer :: incx, incy

2.3.6. ddot

DDOT forms the dot product of two vectors. 

 real(8) function ddot(n, x, incx, y, incy)
   integer :: n
   real(8), device, dimension(*) :: x, y ! device or host variable
   integer :: incx, incy 
 real(8) function cublasDdot(n, x, incx, y, incy)
   integer :: n
   real(8), device, dimension(*) :: x, y
   integer :: incx, incy 
 integer(4) function cublasDdot_v2(h, n, x, incx, y, incy, res)
   type(cublasHandle) :: h
   integer :: n
   real(8), device, dimension(*) :: x, y
   integer :: incx, incy
   real(8), device :: res ! device or host variable

2.3.7. dnrm2

DNRM2 returns the euclidean norm of a vector via the function name, so that DNRM2 := sqrt( x'*x ) 

 real(8) function dnrm2(n, x, incx)
   integer :: n
   real(8), device, dimension(*) :: x ! device or host variable
   integer :: incx 
 real(8) function cublasDnrm2(n, x, incx)
   integer :: n
   real(8), device, dimension(*) :: x
   integer :: incx 
 integer(4) function cublasDnrm2_v2(h, n, x, incx, res)
   type(cublasHandle) :: h
   integer :: n
   real(8), device, dimension(*) :: x
   integer :: incx
   real(8), device :: res ! device or host variable

2.3.8. drot

DROT applies a plane rotation. 

 subroutine drot(n, x, incx, y, incy, sc, ss)
   integer :: n
   real(8), device :: sc, ss ! device or host variable
   real(8), device, dimension(*) :: x, y ! device or host variable
   integer :: incx, incy 
 subroutine cublasDrot(n, x, incx, y, incy, sc, ss)
   integer :: n
   real(8), device :: sc, ss ! device or host variable
   real(8), device, dimension(*) :: x, y
   integer :: incx, incy 
 integer(4) function cublasDrot_v2(h, n, x, incx, y, incy, sc, ss)
   type(cublasHandle) :: h
   integer :: n
   real(8), device :: sc, ss ! device or host variable
   real(8), device, dimension(*) :: x, y
   integer :: incx, incy

2.3.9. drotg

DROTG constructs a Givens plane rotation. 

 subroutine drotg(sa, sb, sc, ss)
   real(8), device :: sa, sb, sc, ss ! device or host variable 
 subroutine cublasDrotg(sa, sb, sc, ss)
   real(8), device :: sa, sb, sc, ss ! device or host variable 
 integer(4) function cublasDrotg_v2(h, sa, sb, sc, ss)
   type(cublasHandle) :: h
   real(8), device :: sa, sb, sc, ss ! device or host variable

2.3.10. drotm

DROTM applies the modified Givens transformation, H, to the 2 by N matrix (DX**T) , where **T indicates transpose. The elements of DX are in (DX**T) DX(LX+I*INCX), I = 0 to N-1, where LX = 1 if INCX .GE. 0, ELSE LX = (-INCX)*N, and similarly for DY using LY and INCY. With DPARAM(1)=DFLAG, H has one of the following forms.. DFLAG=-1.D0 DFLAG=0.D0 DFLAG=1.D0 DFLAG=-2.D0 (DH11 DH12) (1.D0 DH12) (DH11 1.D0) (1.D0 0.D0) H=( ) ( ) ( ) ( ) (DH21 DH22), (DH21 1.D0), (-1.D0 DH22), (0.D0 1.D0). See DROTMG for a description of data storage in DPARAM. 

 subroutine drotm(n, x, incx, y, incy, param)
   integer :: n
   real(8), device, dimension(*) :: x, y ! device or host variable
   integer :: incx, incy 
 subroutine cublasDrotm(n, x, incx, y, incy, param)
   integer :: n
   real(8), device, dimension(*) :: x, y
   integer :: incx, incy
   real(8), device :: param(*) ! device or host variable 
 integer(4) function cublasDrotm_v2(h, n, x, incx, y, incy, param)
   type(cublasHandle) :: h
   integer :: n
   real(8), device :: param(*) ! device or host variable
   real(8), device, dimension(*) :: x, y
   integer :: incx, incy

2.3.11. drotmg

DROTMG constructs the modified Givens transformation matrix H which zeros the second component of the 2-vector (SQRT(DD1)*DX1,SQRT(DD2)*DY2)**T. With DPARAM(1)=DFLAG, H has one of the following forms.. DFLAG=-1.D0 DFLAG=0.D0 DFLAG=1.D0 DFLAG=-2.D0 (DH11 DH12) (1.D0 DH12) (DH11 1.D0) (1.D0 0.D0) H=( ) ( ) ( ) ( ) (DH21 DH22), (DH21 1.D0), (-1.D0 DH22), (0.D0 1.D0). Locations 2-4 of DPARAM contain DH11, DH21, DH12, and DH22 respectively. (Values of 1.D0, -1.D0, of 0.D0 implied by the value of DPARAM(1) are not stored in DPARAM.) 

 subroutine drotmg(d1, d2, x1, y1, param)
   real(8), device :: d1, d2, x1, y1, param(*) ! device or host variable 
 subroutine cublasDrotmg(d1, d2, x1, y1, param)
   real(8), device :: d1, d2, x1, y1, param(*) ! device or host variable 
 integer(4) function cublasDrotmg_v2(h, d1, d2, x1, y1, param)
   type(cublasHandle) :: h
   real(8), device :: d1, d2, x1, y1, param(*) ! device or host variable

2.3.12. dscal

DSCAL scales a vector by a constant. 

 subroutine dscal(n, a, x, incx)
   integer :: n
   real(8), device :: a ! device or host variable
   real(8), device, dimension(*) :: x ! device or host variable
   integer :: incx 
 subroutine cublasDscal(n, a, x, incx)
   integer :: n
   real(8), device :: a ! device or host variable
   real(8), device, dimension(*) :: x
   integer :: incx 
 integer(4) function cublasDscal_v2(h, n, a, x, incx)
   type(cublasHandle) :: h
   integer :: n
   real(8), device :: a ! device or host variable
   real(8), device, dimension(*) :: x
   integer :: incx

2.3.13. dswap

interchanges two vectors. 

 subroutine dswap(n, x, incx, y, incy)
   integer :: n
   real(8), device, dimension(*) :: x, y ! device or host variable
   integer :: incx, incy 
 subroutine cublasDswap(n, x, incx, y, incy)
   integer :: n
   real(8), device, dimension(*) :: x, y
   integer :: incx, incy 
 integer(4) function cublasDswap_v2(h, n, x, incx, y, incy)
   type(cublasHandle) :: h
   integer :: n
   real(8), device, dimension(*) :: x, y
   integer :: incx, incy

2.3.14. dgbmv

DGBMV performs one of the matrix-vector operations y := alpha*A*x + beta*y, or y := alpha*A**T*x + beta*y, where alpha and beta are scalars, x and y are vectors and A is an m by n band matrix, with kl sub-diagonals and ku super-diagonals. 

 subroutine dgbmv(t, m, n, kl, ku, alpha, a, lda, x, incx, beta, y, incy)
   character*1 :: t
   integer :: m, n, kl, ku, lda, incx, incy
   real(8), device, dimension(lda, *) :: a ! device or host variable
   real(8), device, dimension(*) :: x, y ! device or host variable
   real(8), device :: alpha, beta ! device or host variable 
 subroutine cublasDgbmv(t, m, n, kl, ku, alpha, a, lda, x, incx, beta, y, incy)
   character*1 :: t
   integer :: m, n, kl, ku, lda, incx, incy
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(*) :: x, y
   real(8), device :: alpha, beta ! device or host variable 
 integer(4) function cublasDgbmv_v2(h, t, m, n, kl, ku, alpha, a, lda, x, incx, beta, y, incy)
   type(cublasHandle) :: h
   integer :: t
   integer :: m, n, kl, ku, lda, incx, incy
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(*) :: x, y
   real(8), device :: alpha, beta ! device or host variable

2.3.15. dgemv

DGEMV performs one of the matrix-vector operations y := alpha*A*x + beta*y, or y := alpha*A**T*x + beta*y, where alpha and beta are scalars, x and y are vectors and A is an m by n matrix. 

 subroutine dgemv(t, m, n, alpha, a, lda, x, incx, beta, y, incy)
   character*1 :: t
   integer :: m, n, lda, incx, incy
   real(8), device, dimension(lda, *) :: a ! device or host variable
   real(8), device, dimension(*) :: x, y ! device or host variable
   real(8), device :: alpha, beta ! device or host variable 
 subroutine cublasDgemv(t, m, n, alpha, a, lda, x, incx, beta, y, incy)
   character*1 :: t
   integer :: m, n, lda, incx, incy
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(*) :: x, y
   real(8), device :: alpha, beta ! device or host variable 
 integer(4) function cublasDgemv_v2(h, t, m, n, alpha, a, lda, x, incx, beta, y, incy)
   type(cublasHandle) :: h
   integer :: t
   integer :: m, n, lda, incx, incy
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(*) :: x, y
   real(8), device :: alpha, beta ! device or host variable

2.3.16. dger

DGER performs the rank 1 operation A := alpha*x*y**T + A, where alpha is a scalar, x is an m element vector, y is an n element vector and A is an m by n matrix. 

 subroutine dger(m, n, alpha, x, incx, y, incy, a, lda)
   integer :: m, n, lda, incx, incy
   real(8), device, dimension(lda, *) :: a ! device or host variable
   real(8), device, dimension(*) :: x, y ! device or host variable
   real(8), device :: alpha ! device or host variable 
 subroutine cublasDger(m, n, alpha, x, incx, y, incy, a, lda)
   integer :: m, n, lda, incx, incy
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(*) :: x, y
   real(8), device :: alpha ! device or host variable 
 integer(4) function cublasDger_v2(h, m, n, alpha, x, incx, y, incy, a, lda)
   type(cublasHandle) :: h
   integer :: m, n, lda, incx, incy
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(*) :: x, y
   real(8), device :: alpha ! device or host variable

2.3.17. dsbmv

DSBMV performs the matrix-vector operation y := alpha*A*x + beta*y, where alpha and beta are scalars, x and y are n element vectors and A is an n by n symmetric band matrix, with k super-diagonals. 

 subroutine dsbmv(t, n, k, alpha, a, lda, x, incx, beta, y, incy)
   character*1 :: t
   integer :: k, n, lda, incx, incy
   real(8), device, dimension(lda, *) :: a ! device or host variable
   real(8), device, dimension(*) :: x, y ! device or host variable
   real(8), device :: alpha, beta ! device or host variable 
 subroutine cublasDsbmv(t, n, k, alpha, a, lda, x, incx, beta, y, incy)
   character*1 :: t
   integer :: k, n, lda, incx, incy
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(*) :: x, y
   real(8), device :: alpha, beta ! device or host variable 
 integer(4) function cublasDsbmv_v2(h, t, n, k, alpha, a, lda, x, incx, beta, y, incy)
   type(cublasHandle) :: h
   integer :: t
   integer :: k, n, lda, incx, incy
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(*) :: x, y
   real(8), device :: alpha, beta ! device or host variable

2.3.18. dspmv

DSPMV performs the matrix-vector operation y := alpha*A*x + beta*y, where alpha and beta are scalars, x and y are n element vectors and A is an n by n symmetric matrix, supplied in packed form. 

 subroutine dspmv(t, n, alpha, a, x, incx, beta, y, incy)
   character*1 :: t
   integer :: n, incx, incy
   real(8), device, dimension(*) :: a, x, y ! device or host variable
   real(8), device :: alpha, beta ! device or host variable 
 subroutine cublasDspmv(t, n, alpha, a, x, incx, beta, y, incy)
   character*1 :: t
   integer :: n, incx, incy
   real(8), device, dimension(*) :: a, x, y
   real(8), device :: alpha, beta ! device or host variable 
 integer(4) function cublasDspmv_v2(h, t, n, alpha, a, x, incx, beta, y, incy)
   type(cublasHandle) :: h
   integer :: t
   integer :: n, incx, incy
   real(8), device, dimension(*) :: a, x, y
   real(8), device :: alpha, beta ! device or host variable

2.3.19. dspr

DSPR performs the symmetric rank 1 operation A := alpha*x*x**T + A, where alpha is a real scalar, x is an n element vector and A is an n by n symmetric matrix, supplied in packed form. 

 subroutine dspr(t, n, alpha, x, incx, a)
   character*1 :: t
   integer :: n, incx
   real(8), device, dimension(*) :: a, x ! device or host variable
   real(8), device :: alpha ! device or host variable 
 subroutine cublasDspr(t, n, alpha, x, incx, a)
   character*1 :: t
   integer :: n, incx
   real(8), device, dimension(*) :: a, x
   real(8), device :: alpha ! device or host variable 
 integer(4) function cublasDspr_v2(h, t, n, alpha, x, incx, a)
   type(cublasHandle) :: h
   integer :: t
   integer :: n, incx
   real(8), device, dimension(*) :: a, x
   real(8), device :: alpha ! device or host variable

2.3.20. dspr2

DSPR2 performs the symmetric rank 2 operation A := alpha*x*y**T + alpha*y*x**T + A, where alpha is a scalar, x and y are n element vectors and A is an n by n symmetric matrix, supplied in packed form. 

 subroutine dspr2(t, n, alpha, x, incx, y, incy, a)
   character*1 :: t
   integer :: n, incx, incy
   real(8), device, dimension(*) :: a, x, y ! device or host variable
   real(8), device :: alpha ! device or host variable 
 subroutine cublasDspr2(t, n, alpha, x, incx, y, incy, a)
   character*1 :: t
   integer :: n, incx, incy
   real(8), device, dimension(*) :: a, x, y
   real(8), device :: alpha ! device or host variable 
 integer(4) function cublasDspr2_v2(h, t, n, alpha, x, incx, y, incy, a)
   type(cublasHandle) :: h
   integer :: t
   integer :: n, incx, incy
   real(8), device, dimension(*) :: a, x, y
   real(8), device :: alpha ! device or host variable

2.3.21. dsymv

DSYMV performs the matrix-vector operation y := alpha*A*x + beta*y, where alpha and beta are scalars, x and y are n element vectors and A is an n by n symmetric matrix. 

 subroutine dsymv(uplo, n, alpha, a, lda, x, incx, beta, y, incy)
   character*1 :: uplo
   integer :: n, lda, incx, incy
   real(8), device, dimension(lda, *) :: a ! device or host variable
   real(8), device, dimension(*) :: x, y ! device or host variable
   real(8), device :: alpha, beta ! device or host variable 
 subroutine cublasDsymv(uplo, n, alpha, a, lda, x, incx, beta, y, incy)
   character*1 :: uplo
   integer :: n, lda, incx, incy
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(*) :: x, y
   real(8), device :: alpha, beta ! device or host variable 
 integer(4) function cublasDsymv_v2(h, uplo, n, alpha, a, lda, x, incx, beta, y, incy)
   type(cublasHandle) :: h
   integer :: uplo
   integer :: n, lda, incx, incy
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(*) :: x, y
   real(8), device :: alpha, beta ! device or host variable

2.3.22. dsyr

DSYR performs the symmetric rank 1 operation A := alpha*x*x**T + A, where alpha is a real scalar, x is an n element vector and A is an n by n symmetric matrix. 

 subroutine dsyr(t, n, alpha, x, incx, a, lda)
   character*1 :: t
   integer :: n, incx, lda
   real(8), device, dimension(lda, *) :: a ! device or host variable
   real(8), device, dimension(*) :: x ! device or host variable
   real(8), device :: alpha ! device or host variable 
 subroutine cublasDsyr(t, n, alpha, x, incx, a, lda)
   character*1 :: t
   integer :: n, incx, lda
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(*) :: x
   real(8), device :: alpha ! device or host variable 
 integer(4) function cublasDsyr_v2(h, t, n, alpha, x, incx, a, lda)
   type(cublasHandle) :: h
   integer :: t
   integer :: n, incx, lda
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(*) :: x
   real(8), device :: alpha ! device or host variable

2.3.23. dsyr2

DSYR2 performs the symmetric rank 2 operation A := alpha*x*y**T + alpha*y*x**T + A, where alpha is a scalar, x and y are n element vectors and A is an n by n symmetric matrix. 

 subroutine dsyr2(t, n, alpha, x, incx, y, incy, a, lda)
   character*1 :: t
   integer :: n, incx, incy, lda
   real(8), device, dimension(lda, *) :: a ! device or host variable
   real(8), device, dimension(*) :: x, y ! device or host variable
   real(8), device :: alpha ! device or host variable 
 subroutine cublasDsyr2(t, n, alpha, x, incx, y, incy, a, lda)
   character*1 :: t
   integer :: n, incx, incy, lda
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(*) :: x, y
   real(8), device :: alpha ! device or host variable 
 integer(4) function cublasDsyr2_v2(h, t, n, alpha, x, incx, y, incy, a, lda)
   type(cublasHandle) :: h
   integer :: t
   integer :: n, incx, incy, lda
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(*) :: x, y
   real(8), device :: alpha ! device or host variable

2.3.24. dtbmv

DTBMV performs one of the matrix-vector operations x := A*x, or x := A**T*x, where x is an n element vector and A is an n by n unit, or non-unit, upper or lower triangular band matrix, with ( k + 1 ) diagonals. 

 subroutine dtbmv(u, t, d, n, k, a, lda, x, incx)
   character*1 :: u, t, d
   integer :: n, k, incx, lda
   real(8), device, dimension(lda, *) :: a ! device or host variable
   real(8), device, dimension(*) :: x ! device or host variable 
 subroutine cublasDtbmv(u, t, d, n, k, a, lda, x, incx)
   character*1 :: u, t, d
   integer :: n, k, incx, lda
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(*) :: x 
 integer(4) function cublasDtbmv_v2(h, u, t, d, n, k, a, lda, x, incx)
   type(cublasHandle) :: h
   integer :: u, t, d
   integer :: n, k, incx, lda
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(*) :: x

2.3.25. dtbsv

DTBSV solves one of the systems of equations A*x = b, or A**T*x = b, where b and x are n element vectors and A is an n by n unit, or non-unit, upper or lower triangular band matrix, with ( k + 1 ) diagonals. No test for singularity or near-singularity is included in this routine. Such tests must be performed before calling this routine. 

 subroutine dtbsv(u, t, d, n, k, a, lda, x, incx)
   character*1 :: u, t, d
   integer :: n, k, incx, lda
   real(8), device, dimension(lda, *) :: a ! device or host variable
   real(8), device, dimension(*) :: x ! device or host variable 
 subroutine cublasDtbsv(u, t, d, n, k, a, lda, x, incx)
   character*1 :: u, t, d
   integer :: n, k, incx, lda
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(*) :: x 
 integer(4) function cublasDtbsv_v2(h, u, t, d, n, k, a, lda, x, incx)
   type(cublasHandle) :: h
   integer :: u, t, d
   integer :: n, k, incx, lda
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(*) :: x

2.3.26. dtpmv

DTPMV performs one of the matrix-vector operations x := A*x, or x := A**T*x, where x is an n element vector and A is an n by n unit, or non-unit, upper or lower triangular matrix, supplied in packed form. 

 subroutine dtpmv(u, t, d, n, a, x, incx)
   character*1 :: u, t, d
   integer :: n, incx
   real(8), device, dimension(*) :: a, x ! device or host variable 
 subroutine cublasDtpmv(u, t, d, n, a, x, incx)
   character*1 :: u, t, d
   integer :: n, incx
   real(8), device, dimension(*) :: a, x 
 integer(4) function cublasDtpmv_v2(h, u, t, d, n, a, x, incx)
   type(cublasHandle) :: h
   integer :: u, t, d
   integer :: n, incx
   real(8), device, dimension(*) :: a, x

2.3.27. dtpsv

DTPSV solves one of the systems of equations A*x = b, or A**T*x = b, where b and x are n element vectors and A is an n by n unit, or non-unit, upper or lower triangular matrix, supplied in packed form. No test for singularity or near-singularity is included in this routine. Such tests must be performed before calling this routine. 

 subroutine dtpsv(u, t, d, n, a, x, incx)
   character*1 :: u, t, d
   integer :: n, incx
   real(8), device, dimension(*) :: a, x ! device or host variable 
 subroutine cublasDtpsv(u, t, d, n, a, x, incx)
   character*1 :: u, t, d
   integer :: n, incx
   real(8), device, dimension(*) :: a, x 
 integer(4) function cublasDtpsv_v2(h, u, t, d, n, a, x, incx)
   type(cublasHandle) :: h
   integer :: u, t, d
   integer :: n, incx
   real(8), device, dimension(*) :: a, x

2.3.28. dtrmv

DTRMV performs one of the matrix-vector operations x := A*x, or x := A**T*x, where x is an n element vector and A is an n by n unit, or non-unit, upper or lower triangular matrix. 

 subroutine dtrmv(u, t, d, n, a, lda, x, incx)
   character*1 :: u, t, d
   integer :: n, incx, lda
   real(8), device, dimension(lda, *) :: a ! device or host variable
   real(8), device, dimension(*) :: x ! device or host variable 
 subroutine cublasDtrmv(u, t, d, n, a, lda, x, incx)
   character*1 :: u, t, d
   integer :: n, incx, lda
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(*) :: x 
 integer(4) function cublasDtrmv_v2(h, u, t, d, n, a, lda, x, incx)
   type(cublasHandle) :: h
   integer :: u, t, d
   integer :: n, incx, lda
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(*) :: x

2.3.29. dtrsv

DTRSV solves one of the systems of equations A*x = b, or A**T*x = b, where b and x are n element vectors and A is an n by n unit, or non-unit, upper or lower triangular matrix. No test for singularity or near-singularity is included in this routine. Such tests must be performed before calling this routine. 

 subroutine dtrsv(u, t, d, n, a, lda, x, incx)
   character*1 :: u, t, d
   integer :: n, incx, lda
   real(8), device, dimension(lda, *) :: a ! device or host variable
   real(8), device, dimension(*) :: x ! device or host variable 
 subroutine cublasDtrsv(u, t, d, n, a, lda, x, incx)
   character*1 :: u, t, d
   integer :: n, incx, lda
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(*) :: x 
 integer(4) function cublasDtrsv_v2(h, u, t, d, n, a, lda, x, incx)
   type(cublasHandle) :: h
   integer :: u, t, d
   integer :: n, incx, lda
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(*) :: x

2.3.30. dgemm

DGEMM performs one of the matrix-matrix operations C := alpha*op( A )*op( B ) + beta*C, where op( X ) is one of op( X ) = X or op( X ) = X**T, alpha and beta are scalars, and A, B and C are matrices, with op( A ) an m by k matrix, op( B ) a k by n matrix and C an m by n matrix. 

 subroutine dgemm(transa, transb, m, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   character*1 :: transa, transb
   integer :: m, n, k, lda, ldb, ldc
   real(8), device, dimension(lda, *) :: a ! device or host variable
   real(8), device, dimension(ldb, *) :: b ! device or host variable
   real(8), device, dimension(ldc, *) :: c ! device or host variable
   real(8), device :: alpha, beta ! device or host variable 
 subroutine cublasDgemm(transa, transb, m, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   character*1 :: transa, transb
   integer :: m, n, k, lda, ldb, ldc
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(ldb, *) :: b
   real(8), device, dimension(ldc, *) :: c
   real(8), device :: alpha, beta ! device or host variable 
 integer(4) function cublasDgemm_v2(h, transa, transb, m, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: transa, transb
   integer :: m, n, k, lda, ldb, ldc
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(ldb, *) :: b
   real(8), device, dimension(ldc, *) :: c
   real(8), device :: alpha, beta ! device or host variable

2.3.31. dsymm

DSYMM performs one of the matrix-matrix operations C := alpha*A*B + beta*C, or C := alpha*B*A + beta*C, where alpha and beta are scalars, A is a symmetric matrix and B and C are m by n matrices. 

 subroutine dsymm(side, uplo, m, n, alpha, a, lda, b, ldb, beta, c, ldc)
   character*1 :: side, uplo
   integer :: m, n, lda, ldb, ldc
   real(8), device, dimension(lda, *) :: a ! device or host variable
   real(8), device, dimension(ldb, *) :: b ! device or host variable
   real(8), device, dimension(ldc, *) :: c ! device or host variable
   real(8), device :: alpha, beta ! device or host variable 
 subroutine cublasDsymm(side, uplo, m, n, alpha, a, lda, b, ldb, beta, c, ldc)
   character*1 :: side, uplo
   integer :: m, n, lda, ldb, ldc
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(ldb, *) :: b
   real(8), device, dimension(ldc, *) :: c
   real(8), device :: alpha, beta ! device or host variable 
 integer(4) function cublasDsymm_v2(h, side, uplo, m, n, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: side, uplo
   integer :: m, n, lda, ldb, ldc
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(ldb, *) :: b
   real(8), device, dimension(ldc, *) :: c
   real(8), device :: alpha, beta ! device or host variable

2.3.32. dsyrk

DSYRK performs one of the symmetric rank k operations C := alpha*A*A**T + beta*C, or C := alpha*A**T*A + beta*C, where alpha and beta are scalars, C is an n by n symmetric matrix and A is an n by k matrix in the first case and a k by n matrix in the second case. 

 subroutine dsyrk(uplo, trans, n, k, alpha, a, lda, beta, c, ldc)
   character*1 :: uplo, trans
   integer :: n, k, lda, ldc
   real(8), device, dimension(lda, *) :: a ! device or host variable
   real(8), device, dimension(ldc, *) :: c ! device or host variable
   real(8), device :: alpha, beta ! device or host variable 
 subroutine cublasDsyrk(uplo, trans, n, k, alpha, a, lda, beta, c, ldc)
   character*1 :: uplo, trans
   integer :: n, k, lda, ldc
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(ldc, *) :: c
   real(8), device :: alpha, beta ! device or host variable 
 integer(4) function cublasDsyrk_v2(h, uplo, trans, n, k, alpha, a, lda, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: uplo, trans
   integer :: n, k, lda, ldc
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(ldc, *) :: c
   real(8), device :: alpha, beta ! device or host variable

2.3.33. dsyr2k

DSYR2K performs one of the symmetric rank 2k operations C := alpha*A*B**T + alpha*B*A**T + beta*C, or C := alpha*A**T*B + alpha*B**T*A + beta*C, where alpha and beta are scalars, C is an n by n symmetric matrix and A and B are n by k matrices in the first case and k by n matrices in the second case. 

 subroutine dsyr2k(uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   character*1 :: uplo, trans
   integer :: n, k, lda, ldb, ldc
   real(8), device, dimension(lda, *) :: a ! device or host variable
   real(8), device, dimension(ldb, *) :: b ! device or host variable
   real(8), device, dimension(ldc, *) :: c ! device or host variable
   real(8), device :: alpha, beta ! device or host variable 
 subroutine cublasDsyr2k(uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   character*1 :: uplo, trans
   integer :: n, k, lda, ldb, ldc
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(ldb, *) :: b
   real(8), device, dimension(ldc, *) :: c
   real(8), device :: alpha, beta ! device or host variable 
 integer(4) function cublasDsyr2k_v2(h, uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: uplo, trans
   integer :: n, k, lda, ldb, ldc
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(ldb, *) :: b
   real(8), device, dimension(ldc, *) :: c
   real(8), device :: alpha, beta ! device or host variable

2.3.34. dsyrkx

DSYRKX performs a variation of the symmetric rank k update C := alpha*A*B**T + beta*C, where alpha and beta are scalars, C is an n by n symmetric matrix stored in lower or upper mode, and A and B are n by k matrices. This routine can be used when B is in such a way that the result is guaranteed to be symmetric. See the CUBLAS documentation for more details. 

 subroutine dsyrkx(uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   character*1 :: uplo, trans
   integer :: n, k, lda, ldb, ldc
   real(8), device, dimension(lda, *) :: a ! device or host variable
   real(8), device, dimension(ldb, *) :: b ! device or host variable
   real(8), device, dimension(ldc, *) :: c ! device or host variable
   real(8), device :: alpha, beta ! device or host variable 
 subroutine cublasDsyrkx(uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   character*1 :: uplo, trans
   integer :: n, k, lda, ldb, ldc
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(ldb, *) :: b
   real(8), device, dimension(ldc, *) :: c
   real(8), device :: alpha, beta ! device or host variable 
 integer(4) function cublasDsyrkx_v2(h, uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: uplo, trans
   integer :: n, k, lda, ldb, ldc
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(ldb, *) :: b
   real(8), device, dimension(ldc, *) :: c
   real(8), device :: alpha, beta ! device or host variable

2.3.35. dtrmm

DTRMM performs one of the matrix-matrix operations B := alpha*op( A )*B, or B := alpha*B*op( A ), where alpha is a scalar, B is an m by n matrix, A is a unit, or non-unit, upper or lower triangular matrix and op( A ) is one of op( A ) = A or op( A ) = A**T. 

 subroutine dtrmm(side, uplo, transa, diag, m, n, alpha, a, lda, b, ldb)
   character*1 :: side, uplo, transa, diag
   integer :: m, n, lda, ldb
   real(8), device, dimension(lda, *) :: a ! device or host variable
   real(8), device, dimension(ldb, *) :: b ! device or host variable
   real(8), device :: alpha ! device or host variable 
 subroutine cublasDtrmm(side, uplo, transa, diag, m, n, alpha, a, lda, b, ldb)
   character*1 :: side, uplo, transa, diag
   integer :: m, n, lda, ldb
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(ldb, *) :: b
   real(8), device :: alpha ! device or host variable 
 integer(4) function cublasDtrmm_v2(h, side, uplo, transa, diag, m, n, alpha, a, lda, b, ldb, c, ldc)
   type(cublasHandle) :: h
   integer :: side, uplo, transa, diag
   integer :: m, n, lda, ldb, ldc
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(ldb, *) :: b
   real(8), device, dimension(ldc, *) :: c
   real(8), device :: alpha ! device or host variable

2.3.36. dtrsm

DTRSM solves one of the matrix equations op( A )*X = alpha*B, or X*op( A ) = alpha*B, where alpha is a scalar, X and B are m by n matrices, A is a unit, or non-unit, upper or lower triangular matrix and op( A ) is one of op( A ) = A or op( A ) = A**T. The matrix X is overwritten on B. 

 subroutine dtrsm(side, uplo, transa, diag, m, n, alpha, a, lda, b, ldb)
   character*1 :: side, uplo, transa, diag
   integer :: m, n, lda, ldb
   real(8), device, dimension(lda, *) :: a ! device or host variable
   real(8), device, dimension(ldb, *) :: b ! device or host variable
   real(8), device :: alpha ! device or host variable 
 subroutine cublasDtrsm(side, uplo, transa, diag, m, n, alpha, a, lda, b, ldb)
   character*1 :: side, uplo, transa, diag
   integer :: m, n, lda, ldb
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(ldb, *) :: b
   real(8), device :: alpha ! device or host variable 
 integer(4) function cublasDtrsm_v2(h, side, uplo, transa, diag, m, n, alpha, a, lda, b, ldb)
   type(cublasHandle) :: h
   integer :: side, uplo, transa, diag
   integer :: m, n, lda, ldb
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(ldb, *) :: b
   real(8), device :: alpha ! device or host variable

2.3.37. cublasDgetrfBatched

DGETRF computes an LU factorization of a general M-by-N matrix A using partial pivoting with row interchanges. The factorization has the form A = P * L * U where P is a permutation matrix, L is lower triangular with unit diagonal elements (lower trapezoidal if m > n), and U is upper triangular (upper trapezoidal if m < n). This is the right-looking Level 3 BLAS version of the algorithm. 

 integer(4) function cublasDgetrfBatched(h, n, Aarray, lda, ipvt, info, batchCount)
   type(cublasHandle) :: h
   integer :: n
   type(c_devptr), device :: Aarray(*)
   integer :: lda
   integer, device :: ipvt(*)
   integer, device :: info(*)
   integer :: batchCount

2.3.38. cublasDgetriBatched

DGETRI computes the inverse of a matrix using the LU factorization computed by DGETRF. This method inverts U and then computes inv(A) by solving the system inv(A)*L = inv(U) for inv(A). 

 integer(4) function cublasDgetriBatched(h, n, Aarray, lda, ipvt, Carray, ldc, info, batchCount)
   type(cublasHandle) :: h
   integer :: n
   type(c_devptr), device :: Aarray(*)
   integer :: lda
   integer, device :: ipvt(*)
   type(c_devptr), device :: Carray(*)
   integer :: ldc
   integer, device :: info(*)
   integer :: batchCount

2.3.39. cublasDgetrsBatched

DGETRS solves a system of linear equations A * X = B or A**T * X = B with a general N-by-N matrix A using the LU factorization computed by DGETRF. 

 integer(4) function cublasDgetrsBatched(h, trans, n, nrhs, Aarray, lda, ipvt, Barray, ldb, info, batchCount)
   type(cublasHandle) :: h
   integer :: trans ! integer or character(1) variable
   integer :: n, nrhs
   type(c_devptr), device :: Aarray(*)
   integer :: lda
   integer, device :: ipvt(*)
   type(c_devptr), device :: Barray(*)
   integer :: ldb
   integer :: info(*)
   integer :: batchCount

2.3.40. cublasDgemmBatched

DGEMM performs one of the matrix-matrix operations C := alpha*op( A )*op( B ) + beta*C, where op( X ) is one of op( X ) = X or op( X ) = X**T, alpha and beta are scalars, and A, B and C are matrices, with op( A ) an m by k matrix, op( B ) a k by n matrix and C an m by n matrix. 

 integer(4) function cublasDgemmBatched(h, transa, transb, m, n, k, alpha, Aarray, lda, Barray, ldb, beta, Carray, ldc, batchCount)
   type(cublasHandle) :: h
   integer :: transa ! integer or character(1) variable
   integer :: transb ! integer or character(1) variable
   integer :: m, n, k
   real(8), device :: alpha ! device or host variable
   type(c_devptr), device :: Aarray(*)
   integer :: lda
   type(c_devptr), device :: Barray(*)
   integer :: ldb
   real(8), device :: beta ! device or host variable
   type(c_devptr), device :: Carray(*)
   integer :: ldc
   integer :: batchCount 
 integer(4) function cublasDgemmBatched_v2(h, transa, transb, m, n, k, alpha,           Aarray, lda, Barray, ldb, beta, Carray, ldc, batchCount)
   type(cublasHandle) :: h
   integer :: transa
   integer :: transb
   integer :: m, n, k
   real(8), device :: alpha ! device or host variable
   type(c_devptr), device :: Aarray(*)
   integer :: lda
   type(c_devptr), device :: Barray(*)
   integer :: ldb
   real(8), device :: beta ! device or host variable
   type(c_devptr), device :: Carray(*)
   integer :: ldc
   integer :: batchCount

2.3.41. cublasDtrsmBatched

DTRSM solves one of the matrix equations op( A )*X = alpha*B, or X*op( A ) = alpha*B, where alpha is a scalar, X and B are m by n matrices, A is a unit, or non-unit, upper or lower triangular matrix and op( A ) is one of op( A ) = A or op( A ) = A**T. The matrix X is overwritten on B. 

 integer(4) function cublasDtrsmBatched( h, side, uplo, trans, diag, m, n, alpha, A, lda, B, ldb, batchCount)
   type(cublasHandle) :: h
   integer :: side ! integer or character(1) variable
   integer :: uplo ! integer or character(1) variable
   integer :: trans ! integer or character(1) variable
   integer :: diag ! integer or character(1) variable
   integer :: m, n
   real(8), device :: alpha ! device or host variable
   type(c_devptr), device :: A(*)
   integer :: lda
   type(c_devptr), device :: B(*)
   integer :: ldb
   integer :: batchCount 
 integer(4) function cublasDtrsmBatched_v2( h, side, uplo, trans, diag, m, n, alpha, A, lda, B, ldb, batchCount)
   type(cublasHandle) :: h
   integer :: side
   integer :: uplo
   integer :: trans
   integer :: diag
   integer :: m, n
   real(8), device :: alpha ! device or host variable
   type(c_devptr), device :: A(*)
   integer :: lda
   type(c_devptr), device :: B(*)
   integer :: ldb
   integer :: batchCount

2.3.42. cublasDmatinvBatched

cublasDmatinvBatched is a short cut of cublasDgetrfBatched plus cublasDgetriBatched. However it only works if n is less than 32. If not, the user has to go through cublasDgetrfBatched and cublasDgetriBatched. 

 integer(4) function cublasDmatinvBatched(h, n, Aarray, lda, Ainv, lda_inv, info, batchCount)
   type(cublasHandle) :: h
   integer :: n
   type(c_devptr), device :: Aarray(*)
   integer :: lda
   type(c_devptr), device :: Ainv(*)
   integer :: lda_inv
   integer, device :: info(*)
   integer :: batchCount

2.3.43. cublasDgeqrfBatched

DGEQRF computes a QR factorization of a real M-by-N matrix A: A = Q * R. 

 integer(4) function cublasDgeqrfBatched(h, m, n, Aarray, lda, Tau, info, batchCount)
   type(cublasHandle) :: h
   integer :: m, n
   type(c_devptr), device :: Aarray(*)
   integer :: lda
   type(c_devptr), device :: Tau(*)
   integer :: info(*)
   integer :: batchCount

2.3.44. cublasDgelsBatched

DGELS solves overdetermined or underdetermined real linear systems involving an M-by-N matrix A, or its transpose, using a QR or LQ factorization of A. It is assumed that A has full rank. The following options are provided: 1. If TRANS = 'N' and m >= n: find the least squares solution of an overdetermined system, i.e., solve the least squares problem minimize || B - A*X ||. 2. If TRANS = 'N' and m < n: find the minimum norm solution of an underdetermined system A * X = B. 3. If TRANS = 'T' and m >= n: find the minimum norm solution of an undetermined system A**T * X = B. 4. If TRANS = 'T' and m < n: find the least squares solution of an overdetermined system, i.e., solve the least squares problem minimize || B - A**T * X ||. Several right hand side vectors b and solution vectors x can be handled in a single call; they are stored as the columns of the M-by-NRHS right hand side matrix B and the N-by-NRHS solution matrix X. 

 integer(4) function cublasDgelsBatched(h, trans, m, n, nrhs, Aarray, lda, Carray, ldc, info, devinfo, batchCount)
   type(cublasHandle) :: h
   integer :: trans ! integer or character(1) variable
   integer :: m, n, nrhs
   type(c_devptr), device :: Aarray(*)
   integer :: lda
   type(c_devptr), device :: Carray(*)
   integer :: ldc
   integer :: info(*)
   integer, device :: devinfo(*)
   integer :: batchCount

cublasDgemmStridedBatched

DGEMM performs one of the matrix-matrix operations C := alpha*op( A )*op( B ) + beta*C, where op( X ) is one of op( X ) = X or op( X ) = X**T, alpha and beta are scalars, and A, B and C are matrices, with op( A ) an m by k matrix, op( B ) a k by n matrix and C an m by n matrix. 

 integer(4) function cublasDgemmStridedBatched(h, transa, transb, m, n, k, alpha, Aarray, lda, strideA, Barray, ldb, strideB, beta, Carray, ldc, strideC, batchCount)
   type(cublasHandle) :: h
   integer :: transa ! integer or character(1) variable
   integer :: transb ! integer or character(1) variable
   integer :: m, n, k
   real(8), device :: alpha ! device or host variable
   real(8), device :: Aarray(*)
   integer :: lda
   integer :: strideA
   real(8), device :: Barray(*)
   integer :: ldb
   integer :: strideB
   real(8), device :: beta ! device or host variable
   real(8), device :: Carray(*)
   integer :: ldc
   integer :: strideC
   integer :: batchCount 
 integer(4) function cublasDgemmStridedBatched_v2(h, transa, transb, m, n, k, alpha,           Aarray, lda, strideA, Barray, ldb, strideB, beta, Carray, ldc, strideC, batchCount)
   type(cublasHandle) :: h
   integer :: transa
   integer :: transb
   integer :: m, n, k
   real(8), device :: alpha ! device or host variable
   real(8), device :: Aarray(*)
   integer :: lda
   integer :: strideA
   real(8), device :: Barray(*)
   integer :: ldb
   integer :: strideB
   real(8), device :: beta ! device or host variable
   real(8), device :: Carray(*)
   integer :: ldc
   integer :: strideC
   integer :: batchCount

2.4. Single Precision Complex Functions and Subroutines

This section contains interfaces to the single precision complex BLAS and cuBLAS functions and subroutines.

2.4.1. icamax

ICAMAX finds the index of the element having the maximum absolute value. 

 integer(4) function icamax(n, x, incx)
   integer :: n
   complex(4), device, dimension(*) :: x ! device or host variable
   integer :: incx 
 integer(4) function cublasIcamax(n, x, incx)
   integer :: n
   complex(4), device, dimension(*) :: x
   integer :: incx 
 integer(4) function cublasIcamax_v2(h, n, x, incx, res)
   type(cublasHandle) :: h
   integer :: n
   complex(4), device, dimension(*) :: x
   integer :: incx
   integer, device :: res ! device or host variable

2.4.2. icamin

ICAMIN finds the index of the element having the minimum absolute value. 

 integer(4) function icamin(n, x, incx)
   integer :: n
   complex(4), device, dimension(*) :: x ! device or host variable
   integer :: incx 
 integer(4) function cublasIcamin(n, x, incx)
   integer :: n
   complex(4), device, dimension(*) :: x
   integer :: incx 
 integer(4) function cublasIcamin_v2(h, n, x, incx, res)
   type(cublasHandle) :: h
   integer :: n
   complex(4), device, dimension(*) :: x
   integer :: incx
   integer, device :: res ! device or host variable

2.4.3. scasum

SCASUM takes the sum of the absolute values of a complex vector and returns a single precision result. 

 real(4) function scasum(n, x, incx)
   integer :: n
   complex(4), device, dimension(*) :: x ! device or host variable
   integer :: incx 
 real(4) function cublasScasum(n, x, incx)
   integer :: n
   complex(4), device, dimension(*) :: x
   integer :: incx 
 integer(4) function cublasScasum_v2(h, n, x, incx, res)
   type(cublasHandle) :: h
   integer :: n
   complex(4), device, dimension(*) :: x
   integer :: incx
   real(4), device :: res ! device or host variable

2.4.4. caxpy

CAXPY constant times a vector plus a vector. 

 subroutine caxpy(n, a, x, incx, y, incy)
   integer :: n
   complex(4), device :: a ! device or host variable
   complex(4), device, dimension(*) :: x, y ! device or host variable
   integer :: incx, incy 
 subroutine cublasCaxpy(n, a, x, incx, y, incy)
   integer :: n
   complex(4), device :: a ! device or host variable
   complex(4), device, dimension(*) :: x, y
   integer :: incx, incy 
 integer(4) function cublasCaxpy_v2(h, n, a, x, incx, y, incy)
   type(cublasHandle) :: h
   integer :: n
   complex(4), device :: a ! device or host variable
   complex(4), device, dimension(*) :: x, y
   integer :: incx, incy

2.4.5. ccopy

CCOPY copies a vector x to a vector y. 

 subroutine ccopy(n, x, incx, y, incy)
   integer :: n
   complex(4), device, dimension(*) :: x, y ! device or host variable
   integer :: incx, incy 
 subroutine cublasCcopy(n, x, incx, y, incy)
   integer :: n
   complex(4), device, dimension(*) :: x, y
   integer :: incx, incy 
 integer(4) function cublasCcopy_v2(h, n, x, incx, y, incy)
   type(cublasHandle) :: h
   integer :: n
   complex(4), device, dimension(*) :: x, y
   integer :: incx, incy

2.4.6. cdotc

forms the dot product of two vectors, conjugating the first vector. 

 complex(4) function cdotc(n, x, incx, y, incy)
   integer :: n
   complex(4), device, dimension(*) :: x, y ! device or host variable
   integer :: incx, incy 
 complex(4) function cublasCdotc(n, x, incx, y, incy)
   integer :: n
   complex(4), device, dimension(*) :: x, y
   integer :: incx, incy 
 integer(4) function cublasCdotc_v2(h, n, x, incx, y, incy, res)
   type(cublasHandle) :: h
   integer :: n
   complex(4), device, dimension(*) :: x, y
   integer :: incx, incy
   complex(4), device :: res ! device or host variable

2.4.7. cdotu

CDOTU forms the dot product of two vectors. 

 complex(4) function cdotu(n, x, incx, y, incy)
   integer :: n
   complex(4), device, dimension(*) :: x, y ! device or host variable
   integer :: incx, incy 
 complex(4) function cublasCdotu(n, x, incx, y, incy)
   integer :: n
   complex(4), device, dimension(*) :: x, y
   integer :: incx, incy 
 integer(4) function cublasCdotu_v2(h, n, x, incx, y, incy, res)
   type(cublasHandle) :: h
   integer :: n
   complex(4), device, dimension(*) :: x, y
   integer :: incx, incy
   complex(4), device :: res ! device or host variable

2.4.8. scnrm2

SCNRM2 returns the euclidean norm of a vector via the function name, so that SCNRM2 := sqrt( x**H*x ) 

 real(4) function scnrm2(n, x, incx)
   integer :: n
   complex(4), device, dimension(*) :: x ! device or host variable
   integer :: incx 
 real(4) function cublasScnrm2(n, x, incx)
   integer :: n
   complex(4), device, dimension(*) :: x
   integer :: incx 
 integer(4) function cublasScnrm2_v2(h, n, x, incx, res)
   type(cublasHandle) :: h
   integer :: n
   complex(4), device, dimension(*) :: x
   integer :: incx
   real(4), device :: res ! device or host variable

2.4.9. crot

CROT applies a plane rotation, where the cos (C) is real and the sin (S) is complex, and the vectors CX and CY are complex. 

 subroutine crot(n, x, incx, y, incy, sc, ss)
   integer :: n
   real(4), device :: sc ! device or host variable
   complex(4), device :: ss ! device or host variable
   complex(4), device, dimension(*) :: x, y ! device or host variable
   integer :: incx, incy 
 subroutine cublasCrot(n, x, incx, y, incy, sc, ss)
   integer :: n
   real(4), device :: sc ! device or host variable
   complex(4), device :: ss ! device or host variable
   complex(4), device, dimension(*) :: x, y
   integer :: incx, incy 
 integer(4) function cublasCrot_v2(h, n, x, incx, y, incy, sc, ss)
   type(cublasHandle) :: h
   integer :: n
   real(4), device :: sc ! device or host variable
   complex(4), device :: ss ! device or host variable
   complex(4), device, dimension(*) :: x, y
   integer :: incx, incy

2.4.10. csrot

CSROT applies a plane rotation, where the cos and sin (c and s) are real and the vectors cx and cy are complex. 

 subroutine csrot(n, x, incx, y, incy, sc, ss)
   integer :: n
   real(4), device :: sc, ss ! device or host variable
   complex(4), device, dimension(*) :: x, y ! device or host variable
   integer :: incx, incy 
 subroutine cublasCsrot(n, x, incx, y, incy, sc, ss)
   integer :: n
   real(4), device :: sc, ss ! device or host variable
   complex(4), device, dimension(*) :: x, y
   integer :: incx, incy 
 integer(4) function cublasCsrot_v2(h, n, x, incx, y, incy, sc, ss)
   type(cublasHandle) :: h
   integer :: n
   real(4), device :: sc, ss ! device or host variable
   complex(4), device, dimension(*) :: x, y
   integer :: incx, incy

2.4.11. crotg

CROTG determines a complex Givens rotation. 

 subroutine crotg(sa, sb, sc, ss)
   complex(4), device :: sa, sb, ss ! device or host variable
   real(4), device :: sc ! device or host variable 
 subroutine cublasCrotg(sa, sb, sc, ss)
   complex(4), device :: sa, sb, ss ! device or host variable
   real(4), device :: sc ! device or host variable 
 integer(4) function cublasCrotg_v2(h, sa, sb, sc, ss)
   type(cublasHandle) :: h
   complex(4), device :: sa, sb, ss ! device or host variable
   real(4), device :: sc ! device or host variable

2.4.12. cscal

CSCAL scales a vector by a constant. 

 subroutine cscal(n, a, x, incx)
   integer :: n
   complex(4), device :: a ! device or host variable
   complex(4), device, dimension(*) :: x ! device or host variable
   integer :: incx 
 subroutine cublasCscal(n, a, x, incx)
   integer :: n
   complex(4), device :: a ! device or host variable
   complex(4), device, dimension(*) :: x
   integer :: incx 
 integer(4) function cublasCscal_v2(h, n, a, x, incx)
   type(cublasHandle) :: h
   integer :: n
   complex(4), device :: a ! device or host variable
   complex(4), device, dimension(*) :: x
   integer :: incx

2.4.13. csscal

CSSCAL scales a complex vector by a real constant. 

 subroutine csscal(n, a, x, incx)
   integer :: n
   real(4), device :: a ! device or host variable
   complex(4), device, dimension(*) :: x ! device or host variable
   integer :: incx 
 subroutine cublasCsscal(n, a, x, incx)
   integer :: n
   real(4), device :: a ! device or host variable
   complex(4), device, dimension(*) :: x
   integer :: incx 
 integer(4) function cublasCsscal_v2(h, n, a, x, incx)
   type(cublasHandle) :: h
   integer :: n
   real(4), device :: a ! device or host variable
   complex(4), device, dimension(*) :: x
   integer :: incx

2.4.14. cswap

CSWAP interchanges two vectors. 

 subroutine cswap(n, x, incx, y, incy)
   integer :: n
   complex(4), device, dimension(*) :: x, y ! device or host variable
   integer :: incx, incy 
 subroutine cublasCswap(n, x, incx, y, incy)
   integer :: n
   complex(4), device, dimension(*) :: x, y
   integer :: incx, incy 
 integer(4) function cublasCswap_v2(h, n, x, incx, y, incy)
   type(cublasHandle) :: h
   integer :: n
   complex(4), device, dimension(*) :: x, y
   integer :: incx, incy

2.4.15. cgbmv

CGBMV performs one of the matrix-vector operations y := alpha*A*x + beta*y, or y := alpha*A**T*x + beta*y, or y := alpha*A**H*x + beta*y, where alpha and beta are scalars, x and y are vectors and A is an m by n band matrix, with kl sub-diagonals and ku super-diagonals. 

 subroutine cgbmv(t, m, n, kl, ku, alpha, a, lda, x, incx, beta, y, incy)
   character*1 :: t
   integer :: m, n, kl, ku, lda, incx, incy
   complex(4), device, dimension(lda, *) :: a ! device or host variable
   complex(4), device, dimension(*) :: x, y ! device or host variable
   complex(4), device :: alpha, beta ! device or host variable 
 subroutine cublasCgbmv(t, m, n, kl, ku, alpha, a, lda, x, incx, beta, y, incy)
   character*1 :: t
   integer :: m, n, kl, ku, lda, incx, incy
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(*) :: x, y
   complex(4), device :: alpha, beta ! device or host variable 
 integer(4) function cublasCgbmv_v2(h, t, m, n, kl, ku, alpha, a, lda, x, incx, beta, y, incy)
   type(cublasHandle) :: h
   integer :: t
   integer :: m, n, kl, ku, lda, incx, incy
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(*) :: x, y
   complex(4), device :: alpha, beta ! device or host variable

2.4.16. cgemv

CGEMV performs one of the matrix-vector operations y := alpha*A*x + beta*y, or y := alpha*A**T*x + beta*y, or y := alpha*A**H*x + beta*y, where alpha and beta are scalars, x and y are vectors and A is an m by n matrix. 

 subroutine cgemv(t, m, n, alpha, a, lda, x, incx, beta, y, incy)
   character*1 :: t
   integer :: m, n, lda, incx, incy
   complex(4), device, dimension(lda, *) :: a ! device or host variable
   complex(4), device, dimension(*) :: x, y ! device or host variable
   complex(4), device :: alpha, beta ! device or host variable 
 subroutine cublasCgemv(t, m, n, alpha, a, lda, x, incx, beta, y, incy)
   character*1 :: t
   integer :: m, n, lda, incx, incy
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(*) :: x, y
   complex(4), device :: alpha, beta ! device or host variable 
 integer(4) function cublasCgemv_v2(h, t, m, n, alpha, a, lda, x, incx, beta, y, incy)
   type(cublasHandle) :: h
   integer :: t
   integer :: m, n, lda, incx, incy
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(*) :: x, y
   complex(4), device :: alpha, beta ! device or host variable

2.4.17. cgerc

CGERC performs the rank 1 operation A := alpha*x*y**H + A, where alpha is a scalar, x is an m element vector, y is an n element vector and A is an m by n matrix. 

 subroutine cgerc(m, n, alpha, x, incx, y, incy, a, lda)
   integer :: m, n, lda, incx, incy
   complex(4), device, dimension(lda, *) :: a ! device or host variable
   complex(4), device, dimension(*) :: x, y ! device or host variable
   complex(4), device :: alpha ! device or host variable 
 subroutine cublasCgerc(m, n, alpha, x, incx, y, incy, a, lda)
   integer :: m, n, lda, incx, incy
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(*) :: x, y
   complex(4), device :: alpha ! device or host variable 
 integer(4) function cublasCgerc_v2(h, m, n, alpha, x, incx, y, incy, a, lda)
   type(cublasHandle) :: h
   integer :: m, n, lda, incx, incy
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(*) :: x, y
   complex(4), device :: alpha ! device or host variable

2.4.18. cgeru

CGERU performs the rank 1 operation A := alpha*x*y**T + A, where alpha is a scalar, x is an m element vector, y is an n element vector and A is an m by n matrix. 

 subroutine cgeru(m, n, alpha, x, incx, y, incy, a, lda)
   integer :: m, n, lda, incx, incy
   complex(4), device, dimension(lda, *) :: a ! device or host variable
   complex(4), device, dimension(*) :: x, y ! device or host variable
   complex(4), device :: alpha ! device or host variable 
 subroutine cublasCgeru(m, n, alpha, x, incx, y, incy, a, lda)
   integer :: m, n, lda, incx, incy
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(*) :: x, y
   complex(4), device :: alpha ! device or host variable 
 integer(4) function cublasCgeru_v2(h, m, n, alpha, x, incx, y, incy, a, lda)
   type(cublasHandle) :: h
   integer :: m, n, lda, incx, incy
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(*) :: x, y
   complex(4), device :: alpha ! device or host variable

2.4.19. csymv

CSYMV performs the matrix-vector operation y := alpha*A*x + beta*y, where alpha and beta are scalars, x and y are n element vectors and A is an n by n symmetric matrix. 

 subroutine csymv(uplo, n, alpha, a, lda, x, incx, beta, y, incy)
   character*1 :: uplo
   integer :: n, lda, incx, incy
   complex(4), device, dimension(lda, *) :: a ! device or host variable
   complex(4), device, dimension(*) :: x, y ! device or host variable
   complex(4), device :: alpha, beta ! device or host variable 
 subroutine cublasCsymv(uplo, n, alpha, a, lda, x, incx, beta, y, incy)
   character*1 :: uplo
   integer :: n, lda, incx, incy
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(*) :: x, y
   complex(4), device :: alpha, beta ! device or host variable 
 integer(4) function cublasCsymv_v2(h, uplo, n, alpha, a, lda, x, incx, beta, y, incy)
   type(cublasHandle) :: h
   integer :: uplo
   integer :: n, lda, incx, incy
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(*) :: x, y
   complex(4), device :: alpha, beta ! device or host variable

2.4.20. csyr

CSYR performs the symmetric rank 1 operation A := alpha*x*x**H + A, where alpha is a complex scalar, x is an n element vector and A is an n by n symmetric matrix. 

 subroutine csyr(t, n, alpha, x, incx, a, lda)
   character*1 :: t
   integer :: n, incx, lda
   complex(4), device, dimension(lda, *) :: a ! device or host variable
   complex(4), device, dimension(*) :: x ! device or host variable
   complex(4), device :: alpha ! device or host variable 
 subroutine cublasCsyr(t, n, alpha, x, incx, a, lda)
   character*1 :: t
   integer :: n, incx, lda
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(*) :: x
   complex(4), device :: alpha ! device or host variable 
 integer(4) function cublasCsyr_v2(h, t, n, alpha, x, incx, a, lda)
   type(cublasHandle) :: h
   integer :: t
   integer :: n, incx, lda
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(*) :: x
   complex(4), device :: alpha ! device or host variable

2.4.21. csyr2

CSYR2 performs the symmetric rank 2 operation A := alpha*x*y' + alpha*y*x' + A, where alpha is a complex scalar, x and y are n element vectors and A is an n by n SY matrix. 

 subroutine csyr2(t, n, alpha, x, incx, y, incy, a, lda)
   character*1 :: t
   integer :: n, incx, incy, lda
   complex(4), device, dimension(lda, *) :: a ! device or host variable
   complex(4), device, dimension(*) :: x, y ! device or host variable
   complex(4), device :: alpha ! device or host variable 
 subroutine cublasCsyr2(t, n, alpha, x, incx, y, incy, a, lda)
   character*1 :: t
   integer :: n, incx, incy, lda
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(*) :: x, y
   complex(4), device :: alpha ! device or host variable 
 integer(4) function cublasCsyr2_v2(h, t, n, alpha, x, incx, y, incy, a, lda)
   type(cublasHandle) :: h
   integer :: t
   integer :: n, incx, incy, lda
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(*) :: x, y
   complex(4), device :: alpha ! device or host variable

2.4.22. ctbmv

CTBMV performs one of the matrix-vector operations x := A*x, or x := A**T*x, or x := A**H*x, where x is an n element vector and A is an n by n unit, or non-unit, upper or lower triangular band matrix, with ( k + 1 ) diagonals. 

 subroutine ctbmv(u, t, d, n, k, a, lda, x, incx)
   character*1 :: u, t, d
   integer :: n, k, incx, lda
   complex(4), device, dimension(lda, *) :: a ! device or host variable
   complex(4), device, dimension(*) :: x ! device or host variable 
 subroutine cublasCtbmv(u, t, d, n, k, a, lda, x, incx)
   character*1 :: u, t, d
   integer :: n, k, incx, lda
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(*) :: x 
 integer(4) function cublasCtbmv_v2(h, u, t, d, n, k, a, lda, x, incx)
   type(cublasHandle) :: h
   integer :: u, t, d
   integer :: n, k, incx, lda
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(*) :: x

2.4.23. ctbsv

CTBSV solves one of the systems of equations A*x = b, or A**T*x = b, or A**H*x = b, where b and x are n element vectors and A is an n by n unit, or non-unit, upper or lower triangular band matrix, with ( k + 1 ) diagonals. No test for singularity or near-singularity is included in this routine. Such tests must be performed before calling this routine. 

 subroutine ctbsv(u, t, d, n, k, a, lda, x, incx)
   character*1 :: u, t, d
   integer :: n, k, incx, lda
   complex(4), device, dimension(lda, *) :: a ! device or host variable
   complex(4), device, dimension(*) :: x ! device or host variable 
 subroutine cublasCtbsv(u, t, d, n, k, a, lda, x, incx)
   character*1 :: u, t, d
   integer :: n, k, incx, lda
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(*) :: x 
 integer(4) function cublasCtbsv_v2(h, u, t, d, n, k, a, lda, x, incx)
   type(cublasHandle) :: h
   integer :: u, t, d
   integer :: n, k, incx, lda
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(*) :: x

2.4.24. ctpmv

CTPMV performs one of the matrix-vector operations x := A*x, or x := A**T*x, or x := A**H*x, where x is an n element vector and A is an n by n unit, or non-unit, upper or lower triangular matrix, supplied in packed form. 

 subroutine ctpmv(u, t, d, n, a, x, incx)
   character*1 :: u, t, d
   integer :: n, incx
   complex(4), device, dimension(*) :: a, x ! device or host variable 
 subroutine cublasCtpmv(u, t, d, n, a, x, incx)
   character*1 :: u, t, d
   integer :: n, incx
   complex(4), device, dimension(*) :: a, x 
 integer(4) function cublasCtpmv_v2(h, u, t, d, n, a, x, incx)
   type(cublasHandle) :: h
   integer :: u, t, d
   integer :: n, incx
   complex(4), device, dimension(*) :: a, x

2.4.25. ctpsv

CTPSV solves one of the systems of equations A*x = b, or A**T*x = b, or A**H*x = b, where b and x are n element vectors and A is an n by n unit, or non-unit, upper or lower triangular matrix, supplied in packed form. No test for singularity or near-singularity is included in this routine. Such tests must be performed before calling this routine. 

 subroutine ctpsv(u, t, d, n, a, x, incx)
   character*1 :: u, t, d
   integer :: n, incx
   complex(4), device, dimension(*) :: a, x ! device or host variable 
 subroutine cublasCtpsv(u, t, d, n, a, x, incx)
   character*1 :: u, t, d
   integer :: n, incx
   complex(4), device, dimension(*) :: a, x 
 integer(4) function cublasCtpsv_v2(h, u, t, d, n, a, x, incx)
   type(cublasHandle) :: h
   integer :: u, t, d
   integer :: n, incx
   complex(4), device, dimension(*) :: a, x

2.4.26. ctrmv

CTRMV performs one of the matrix-vector operations x := A*x, or x := A**T*x, or x := A**H*x, where x is an n element vector and A is an n by n unit, or non-unit, upper or lower triangular matrix. 

 subroutine ctrmv(u, t, d, n, a, lda, x, incx)
   character*1 :: u, t, d
   integer :: n, incx, lda
   complex(4), device, dimension(lda, *) :: a ! device or host variable
   complex(4), device, dimension(*) :: x ! device or host variable 
 subroutine cublasCtrmv(u, t, d, n, a, lda, x, incx)
   character*1 :: u, t, d
   integer :: n, incx, lda
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(*) :: x 
 integer(4) function cublasCtrmv_v2(h, u, t, d, n, a, lda, x, incx)
   type(cublasHandle) :: h
   integer :: u, t, d
   integer :: n, incx, lda
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(*) :: x

2.4.27. ctrsv

CTRSV solves one of the systems of equations A*x = b, or A**T*x = b, or A**H*x = b, where b and x are n element vectors and A is an n by n unit, or non-unit, upper or lower triangular matrix. No test for singularity or near-singularity is included in this routine. Such tests must be performed before calling this routine. 

 subroutine ctrsv(u, t, d, n, a, lda, x, incx)
   character*1 :: u, t, d
   integer :: n, incx, lda
   complex(4), device, dimension(lda, *) :: a ! device or host variable
   complex(4), device, dimension(*) :: x ! device or host variable 
 subroutine cublasCtrsv(u, t, d, n, a, lda, x, incx)
   character*1 :: u, t, d
   integer :: n, incx, lda
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(*) :: x 
 integer(4) function cublasCtrsv_v2(h, u, t, d, n, a, lda, x, incx)
   type(cublasHandle) :: h
   integer :: u, t, d
   integer :: n, incx, lda
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(*) :: x

2.4.28. chbmv

CHBMV performs the matrix-vector operation y := alpha*A*x + beta*y, where alpha and beta are scalars, x and y are n element vectors and A is an n by n hermitian band matrix, with k super-diagonals. 

 subroutine chbmv(uplo, n, k, alpha, a, lda, x, incx, beta, y, incy)
   character*1 :: uplo
   integer :: k, n, lda, incx, incy
   complex(4), device, dimension(lda, *) :: a ! device or host variable
   complex(4), device, dimension(*) :: x, y ! device or host variable
   complex(4), device :: alpha, beta ! device or host variable 
 subroutine cublasChbmv(uplo, n, k, alpha, a, lda, x, incx, beta, y, incy)
   character*1 :: uplo
   integer :: k, n, lda, incx, incy
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(*) :: x, y
   complex(4), device :: alpha, beta ! device or host variable 
 integer(4) function cublasChbmv_v2(h, uplo, n, k, alpha, a, lda, x, incx, beta, y, incy)
   type(cublasHandle) :: h
   integer :: uplo
   integer :: k, n, lda, incx, incy
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(*) :: x, y
   complex(4), device :: alpha, beta ! device or host variable

2.4.29. chemv

CHEMV performs the matrix-vector operation y := alpha*A*x + beta*y, where alpha and beta are scalars, x and y are n element vectors and A is an n by n hermitian matrix. 

 subroutine chemv(uplo, n, alpha, a, lda, x, incx, beta, y, incy)
   character*1 :: uplo
   integer :: n, lda, incx, incy
   complex(4), device, dimension(lda, *) :: a ! device or host variable
   complex(4), device, dimension(*) :: x, y ! device or host variable
   complex(4), device :: alpha, beta ! device or host variable 
 subroutine cublasChemv(uplo, n, alpha, a, lda, x, incx, beta, y, incy)
   character*1 :: uplo
   integer :: n, lda, incx, incy
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(*) :: x, y
   complex(4), device :: alpha, beta ! device or host variable 
 integer(4) function cublasChemv_v2(h, uplo, n, alpha, a, lda, x, incx, beta, y, incy)
   type(cublasHandle) :: h
   integer :: uplo
   integer :: n, lda, incx, incy
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(*) :: x, y
   complex(4), device :: alpha, beta ! device or host variable

2.4.30. chpmv

CHPMV performs the matrix-vector operation y := alpha*A*x + beta*y, where alpha and beta are scalars, x and y are n element vectors and A is an n by n hermitian matrix, supplied in packed form. 

 subroutine chpmv(uplo, n, alpha, a, x, incx, beta, y, incy)
   character*1 :: uplo
   integer :: n, incx, incy
   complex(4), device, dimension(*) :: a, x, y ! device or host variable
   complex(4), device :: alpha, beta ! device or host variable 
 subroutine cublasChpmv(uplo, n, alpha, a, x, incx, beta, y, incy)
   character*1 :: uplo
   integer :: n, incx, incy
   complex(4), device, dimension(*) :: a, x, y
   complex(4), device :: alpha, beta ! device or host variable 
 integer(4) function cublasChpmv_v2(h, uplo, n, alpha, a, x, incx, beta, y, incy)
   type(cublasHandle) :: h
   integer :: uplo
   integer :: n, incx, incy
   complex(4), device, dimension(*) :: a, x, y
   complex(4), device :: alpha, beta ! device or host variable

2.4.31. cher

CHER performs the hermitian rank 1 operation A := alpha*x*x**H + A, where alpha is a real scalar, x is an n element vector and A is an n by n hermitian matrix. 

 subroutine cher(t, n, alpha, x, incx, a, lda)
   character*1 :: t
   integer :: n, incx, lda
   complex(4), device, dimension(*) :: a, x ! device or host variable
   real(4), device :: alpha ! device or host variable 
 subroutine cublasCher(t, n, alpha, x, incx, a, lda)
   character*1 :: t
   integer :: n, incx, lda
   complex(4), device, dimension(*) :: a, x
   real(4), device :: alpha ! device or host variable 
 integer(4) function cublasCher_v2(h, t, n, alpha, x, incx, a, lda)
   type(cublasHandle) :: h
   integer :: t
   integer :: n, incx, lda
   complex(4), device, dimension(*) :: a, x
   real(4), device :: alpha ! device or host variable

2.4.32. cher2

CHER2 performs the hermitian rank 2 operation A := alpha*x*y**H + conjg( alpha )*y*x**H + A, where alpha is a scalar, x and y are n element vectors and A is an n by n hermitian matrix. 

 subroutine cher2(t, n, alpha, x, incx, y, incy, a, lda)
   character*1 :: t
   integer :: n, incx, incy, lda
   complex(4), device, dimension(*) :: a, x, y ! device or host variable
   complex(4), device :: alpha ! device or host variable 
 subroutine cublasCher2(t, n, alpha, x, incx, y, incy, a, lda)
   character*1 :: t
   integer :: n, incx, incy, lda
   complex(4), device, dimension(*) :: a, x, y
   complex(4), device :: alpha ! device or host variable 
 integer(4) function cublasCher2_v2(h, t, n, alpha, x, incx, y, incy, a, lda)
   type(cublasHandle) :: h
   integer :: t
   integer :: n, incx, incy, lda
   complex(4), device, dimension(*) :: a, x, y
   complex(4), device :: alpha ! device or host variable

2.4.33. chpr

CHPR performs the hermitian rank 1 operation A := alpha*x*x**H + A, where alpha is a real scalar, x is an n element vector and A is an n by n hermitian matrix, supplied in packed form. 

 subroutine chpr(t, n, alpha, x, incx, a)
   character*1 :: t
   integer :: n, incx
   complex(4), device, dimension(*) :: a, x ! device or host variable
   real(4), device :: alpha ! device or host variable 
 subroutine cublasChpr(t, n, alpha, x, incx, a)
   character*1 :: t
   integer :: n, incx
   complex(4), device, dimension(*) :: a, x
   real(4), device :: alpha ! device or host variable 
 integer(4) function cublasChpr_v2(h, t, n, alpha, x, incx, a)
   type(cublasHandle) :: h
   integer :: t
   integer :: n, incx
   complex(4), device, dimension(*) :: a, x
   real(4), device :: alpha ! device or host variable

2.4.34. chpr2

CHPR2 performs the hermitian rank 2 operation A := alpha*x*y**H + conjg( alpha )*y*x**H + A, where alpha is a scalar, x and y are n element vectors and A is an n by n hermitian matrix, supplied in packed form. 

 subroutine chpr2(t, n, alpha, x, incx, y, incy, a)
   character*1 :: t
   integer :: n, incx, incy
   complex(4), device, dimension(*) :: a, x, y ! device or host variable
   complex(4), device :: alpha ! device or host variable 
 subroutine cublasChpr2(t, n, alpha, x, incx, y, incy, a)
   character*1 :: t
   integer :: n, incx, incy
   complex(4), device, dimension(*) :: a, x, y
   complex(4), device :: alpha ! device or host variable 
 integer(4) function cublasChpr2_v2(h, t, n, alpha, x, incx, y, incy, a)
   type(cublasHandle) :: h
   integer :: t
   integer :: n, incx, incy
   complex(4), device, dimension(*) :: a, x, y
   complex(4), device :: alpha ! device or host variable

2.4.35. cgemm

CGEMM performs one of the matrix-matrix operations C := alpha*op( A )*op( B ) + beta*C, where op( X ) is one of op( X ) = X or op( X ) = X**T or op( X ) = X**H, alpha and beta are scalars, and A, B and C are matrices, with op( A ) an m by k matrix, op( B ) a k by n matrix and C an m by n matrix. 

 subroutine cgemm(transa, transb, m, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   character*1 :: transa, transb
   integer :: m, n, k, lda, ldb, ldc
   complex(4), device, dimension(lda, *) :: a ! device or host variable
   complex(4), device, dimension(ldb, *) :: b ! device or host variable
   complex(4), device, dimension(ldc, *) :: c ! device or host variable
   complex(4), device :: alpha, beta ! device or host variable 
 subroutine cublasCgemm(transa, transb, m, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   character*1 :: transa, transb
   integer :: m, n, k, lda, ldb, ldc
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(ldb, *) :: b
   complex(4), device, dimension(ldc, *) :: c
   complex(4), device :: alpha, beta ! device or host variable 
 integer(4) function cublasCgemm_v2(h, transa, transb, m, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: transa, transb
   integer :: m, n, k, lda, ldb, ldc
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(ldb, *) :: b
   complex(4), device, dimension(ldc, *) :: c
   complex(4), device :: alpha, beta ! device or host variable

2.4.36. csymm

CSYMM performs one of the matrix-matrix operations C := alpha*A*B + beta*C, or C := alpha*B*A + beta*C, where alpha and beta are scalars, A is a symmetric matrix and B and C are m by n matrices. 

 subroutine csymm(side, uplo, m, n, alpha, a, lda, b, ldb, beta, c, ldc)
   character*1 :: side, uplo
   integer :: m, n, lda, ldb, ldc
   complex(4), device, dimension(lda, *) :: a ! device or host variable
   complex(4), device, dimension(ldb, *) :: b ! device or host variable
   complex(4), device, dimension(ldc, *) :: c ! device or host variable
   complex(4), device :: alpha, beta ! device or host variable 
 subroutine cublasCsymm(side, uplo, m, n, alpha, a, lda, b, ldb, beta, c, ldc)
   character*1 :: side, uplo
   integer :: m, n, lda, ldb, ldc
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(ldb, *) :: b
   complex(4), device, dimension(ldc, *) :: c
   complex(4), device :: alpha, beta ! device or host variable 
 integer(4) function cublasCsymm_v2(h, side, uplo, m, n, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: side, uplo
   integer :: m, n, lda, ldb, ldc
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(ldb, *) :: b
   complex(4), device, dimension(ldc, *) :: c
   complex(4), device :: alpha, beta ! device or host variable

2.4.37. csyrk

CSYRK performs one of the symmetric rank k operations C := alpha*A*A**T + beta*C, or C := alpha*A**T*A + beta*C, where alpha and beta are scalars, C is an n by n symmetric matrix and A is an n by k matrix in the first case and a k by n matrix in the second case. 

 subroutine csyrk(uplo, trans, n, k, alpha, a, lda, beta, c, ldc)
   character*1 :: uplo, trans
   integer :: n, k, lda, ldc
   complex(4), device, dimension(lda, *) :: a ! device or host variable
   complex(4), device, dimension(ldc, *) :: c ! device or host variable
   complex(4), device :: alpha, beta ! device or host variable 
 subroutine cublasCsyrk(uplo, trans, n, k, alpha, a, lda, beta, c, ldc)
   character*1 :: uplo, trans
   integer :: n, k, lda, ldc
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(ldc, *) :: c
   complex(4), device :: alpha, beta ! device or host variable 
 integer(4) function cublasCsyrk_v2(h, uplo, trans, n, k, alpha, a, lda, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: uplo, trans
   integer :: n, k, lda, ldc
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(ldc, *) :: c
   complex(4), device :: alpha, beta ! device or host variable

2.4.38. csyr2k

CSYR2K performs one of the symmetric rank 2k operations C := alpha*A*B**T + alpha*B*A**T + beta*C, or C := alpha*A**T*B + alpha*B**T*A + beta*C, where alpha and beta are scalars, C is an n by n symmetric matrix and A and B are n by k matrices in the first case and k by n matrices in the second case. 

 subroutine csyr2k(uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   character*1 :: uplo, trans
   integer :: n, k, lda, ldb, ldc
   complex(4), device, dimension(lda, *) :: a ! device or host variable
   complex(4), device, dimension(ldb, *) :: b ! device or host variable
   complex(4), device, dimension(ldc, *) :: c ! device or host variable
   complex(4), device :: alpha, beta ! device or host variable 
 subroutine cublasCsyr2k(uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   character*1 :: uplo, trans
   integer :: n, k, lda, ldb, ldc
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(ldb, *) :: b
   complex(4), device, dimension(ldc, *) :: c
   complex(4), device :: alpha, beta ! device or host variable 
 integer(4) function cublasCsyr2k_v2(h, uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: uplo, trans
   integer :: n, k, lda, ldb, ldc
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(ldb, *) :: b
   complex(4), device, dimension(ldc, *) :: c
   complex(4), device :: alpha, beta ! device or host variable

2.4.39. csyrkx

CSYRKX performs a variation of the symmetric rank k update C := alpha*A*B**T + beta*C, where alpha and beta are scalars, C is an n by n symmetric matrix stored in lower or upper mode, and A and B are n by k matrices. This routine can be used when B is in such a way that the result is guaranteed to be symmetric. See the CUBLAS documentation for more details. 

 subroutine csyrkx(uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   character*1 :: uplo, trans
   integer :: n, k, lda, ldb, ldc
   complex(4), device, dimension(lda, *) :: a ! device or host variable
   complex(4), device, dimension(ldb, *) :: b ! device or host variable
   complex(4), device, dimension(ldc, *) :: c ! device or host variable
   complex(4), device :: alpha, beta ! device or host variable 
 subroutine cublasCsyrkx(uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   character*1 :: uplo, trans
   integer :: n, k, lda, ldb, ldc
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(ldb, *) :: b
   complex(4), device, dimension(ldc, *) :: c
   complex(4), device :: alpha, beta ! device or host variable 
 integer(4) function cublasCsyrkx_v2(h, uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: uplo, trans
   integer :: n, k, lda, ldb, ldc
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(ldb, *) :: b
   complex(4), device, dimension(ldc, *) :: c
   complex(4), device :: alpha, beta ! device or host variable

2.4.40. ctrmm

CTRMM performs one of the matrix-matrix operations B := alpha*op( A )*B, or B := alpha*B*op( A ) where alpha is a scalar, B is an m by n matrix, A is a unit, or non-unit, upper or lower triangular matrix and op( A ) is one of op( A ) = A or op( A ) = A**T or op( A ) = A**H. 

 subroutine ctrmm(side, uplo, transa, diag, m, n, alpha, a, lda, b, ldb)
   character*1 :: side, uplo, transa, diag
   integer :: m, n, lda, ldb
   complex(4), device, dimension(lda, *) :: a ! device or host variable
   complex(4), device, dimension(ldb, *) :: b ! device or host variable
   complex(4), device :: alpha ! device or host variable 
 subroutine cublasCtrmm(side, uplo, transa, diag, m, n, alpha, a, lda, b, ldb)
   character*1 :: side, uplo, transa, diag
   integer :: m, n, lda, ldb
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(ldb, *) :: b
   complex(4), device :: alpha ! device or host variable 
 integer(4) function cublasCtrmm_v2(h, side, uplo, transa, diag, m, n, alpha, a, lda, b, ldb, c, ldc)
   type(cublasHandle) :: h
   integer :: side, uplo, transa, diag
   integer :: m, n, lda, ldb, ldc
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(ldb, *) :: b
   complex(4), device, dimension(ldc, *) :: c
   complex(4), device :: alpha ! device or host variable

2.4.41. ctrsm

CTRSM solves one of the matrix equations op( A )*X = alpha*B, or X*op( A ) = alpha*B, where alpha is a scalar, X and B are m by n matrices, A is a unit, or non-unit, upper or lower triangular matrix and op( A ) is one of op( A ) = A or op( A ) = A**T or op( A ) = A**H. The matrix X is overwritten on B. 

 subroutine ctrsm(side, uplo, transa, diag, m, n, alpha, a, lda, b, ldb)
   character*1 :: side, uplo, transa, diag
   integer :: m, n, lda, ldb
   complex(4), device, dimension(lda, *) :: a ! device or host variable
   complex(4), device, dimension(ldb, *) :: b ! device or host variable
   complex(4), device :: alpha ! device or host variable 
 subroutine cublasCtrsm(side, uplo, transa, diag, m, n, alpha, a, lda, b, ldb)
   character*1 :: side, uplo, transa, diag
   integer :: m, n, lda, ldb
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(ldb, *) :: b
   complex(4), device :: alpha ! device or host variable 
 integer(4) function cublasCtrsm_v2(h, side, uplo, transa, diag, m, n, alpha, a, lda, b, ldb)
   type(cublasHandle) :: h
   integer :: side, uplo, transa, diag
   integer :: m, n, lda, ldb
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(ldb, *) :: b
   complex(4), device :: alpha ! device or host variable

2.4.42. chemm

CHEMM performs one of the matrix-matrix operations C := alpha*A*B + beta*C, or C := alpha*B*A + beta*C, where alpha and beta are scalars, A is an hermitian matrix and B and C are m by n matrices. 

 subroutine chemm(side, uplo, m, n, alpha, a, lda, b, ldb, beta, c, ldc)
   character*1 :: side, uplo
   integer :: m, n, lda, ldb, ldc
   complex(4), device, dimension(lda, *) :: a ! device or host variable
   complex(4), device, dimension(ldb, *) :: b ! device or host variable
   complex(4), device, dimension(ldc, *) :: c ! device or host variable
   complex(4), device :: alpha, beta ! device or host variable 
 subroutine cublasChemm(side, uplo, m, n, alpha, a, lda, b, ldb, beta, c, ldc)
   character*1 :: side, uplo
   integer :: m, n, lda, ldb, ldc
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(ldb, *) :: b
   complex(4), device, dimension(ldc, *) :: c
   complex(4), device :: alpha, beta ! device or host variable 
 integer(4) function cublasChemm_v2(h, side, uplo, m, n, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: side, uplo
   integer :: m, n, lda, ldb, ldc
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(ldb, *) :: b
   complex(4), device, dimension(ldc, *) :: c
   complex(4), device :: alpha, beta ! device or host variable

2.4.43. cherk

CHERK performs one of the hermitian rank k operations C := alpha*A*A**H + beta*C, or C := alpha*A**H*A + beta*C, where alpha and beta are real scalars, C is an n by n hermitian matrix and A is an n by k matrix in the first case and a k by n matrix in the second case. 

 subroutine cherk(uplo, trans, n, k, alpha, a, lda, beta, c, ldc)
   character*1 :: uplo, trans
   integer :: n, k, lda, ldc
   complex(4), device, dimension(lda, *) :: a ! device or host variable
   complex(4), device, dimension(ldc, *) :: c ! device or host variable
   real(4), device :: alpha, beta ! device or host variable 
 subroutine cublasCherk(uplo, trans, n, k, alpha, a, lda, beta, c, ldc)
   character*1 :: uplo, trans
   integer :: n, k, lda, ldc
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(ldc, *) :: c
   real(4), device :: alpha, beta ! device or host variable 
 integer(4) function cublasCherk_v2(h, uplo, trans, n, k, alpha, a, lda, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: uplo, trans
   integer :: n, k, lda, ldc
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(ldc, *) :: c
   real(4), device :: alpha, beta ! device or host variable

2.4.44. cher2k

CHER2K performs one of the hermitian rank 2k operations C := alpha*A*B**H + conjg( alpha )*B*A**H + beta*C, or C := alpha*A**H*B + conjg( alpha )*B**H*A + beta*C, where alpha and beta are scalars with beta real, C is an n by n hermitian matrix and A and B are n by k matrices in the first case and k by n matrices in the second case. 

 subroutine cher2k(uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   character*1 :: uplo, trans
   integer :: n, k, lda, ldb, ldc
   complex(4), device, dimension(lda, *) :: a ! device or host variable
   complex(4), device, dimension(ldb, *) :: b ! device or host variable
   complex(4), device, dimension(ldc, *) :: c ! device or host variable
   complex(4), device :: alpha ! device or host variable
   real(4), device :: beta ! device or host variable 
 subroutine cublasCher2k(uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   character*1 :: uplo, trans
   integer :: n, k, lda, ldb, ldc
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(ldb, *) :: b
   complex(4), device, dimension(ldc, *) :: c
   complex(4), device :: alpha ! device or host variable
   real(4), device :: beta ! device or host variable 
 integer(4) function cublasCher2k_v2(h, uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: uplo, trans
   integer :: n, k, lda, ldb, ldc
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(ldb, *) :: b
   complex(4), device, dimension(ldc, *) :: c
   complex(4), device :: alpha ! device or host variable
   real(4), device :: beta ! device or host variable

2.4.45. cherkx

CHERKX performs a variation of the hermitian rank k operations C := alpha*A*B**H + beta*C, where alpha and beta are real scalars, C is an n by n hermitian matrix stored in lower or upper mode, and A and B are n by k matrices. See the CUBLAS documentation for more details. 

 subroutine cherkx(uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   character*1 :: uplo, trans
   integer :: n, k, lda, ldb, ldc
   complex(4), device, dimension(lda, *) :: a ! device or host variable
   complex(4), device, dimension(ldb, *) :: b ! device or host variable
   complex(4), device, dimension(ldc, *) :: c ! device or host variable
   complex(4), device :: alpha ! device or host variable
   real(4), device :: beta ! device or host variable 
 subroutine cublasCherkx(uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   character*1 :: uplo, trans
   integer :: n, k, lda, ldb, ldc
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(ldb, *) :: b
   complex(4), device, dimension(ldc, *) :: c
   complex(4), device :: alpha ! device or host variable
   real(4), device :: beta ! device or host variable 
 integer(4) function cublasCherkx_v2(h, uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: uplo, trans
   integer :: n, k, lda, ldb, ldc
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(ldb, *) :: b
   complex(4), device, dimension(ldc, *) :: c
   complex(4), device :: alpha ! device or host variable
   real(4), device :: beta ! device or host variable

2.4.46. cublasCgetrfBatched

CGETRF computes an LU factorization of a general M-by-N matrix A using partial pivoting with row interchanges. The factorization has the form A = P * L * U where P is a permutation matrix, L is lower triangular with unit diagonal elements (lower trapezoidal if m > n), and U is upper triangular (upper trapezoidal if m < n). This is the right-looking Level 3 BLAS version of the algorithm. 

 integer(4) function cublasCgetrfBatched(h, n, Aarray, lda, ipvt, info, batchCount)
   type(cublasHandle) :: h
   integer :: n
   type(c_devptr), device :: Aarray(*)
   integer :: lda
   integer, device :: ipvt(*)
   integer, device :: info(*)
   integer :: batchCount

2.4.47. cublasCgetriBatched

CGETRI computes the inverse of a matrix using the LU factorization computed by CGETRF. This method inverts U and then computes inv(A) by solving the system inv(A)*L = inv(U) for inv(A). 

 integer(4) function cublasCgetriBatched(h, n, Aarray, lda, ipvt, Carray, ldc, info, batchCount)
   type(cublasHandle) :: h
   integer :: n
   type(c_devptr), device :: Aarray(*)
   integer :: lda
   integer, device :: ipvt(*)
   type(c_devptr), device :: Carray(*)
   integer :: ldc
   integer, device :: info(*)
   integer :: batchCount

2.4.48. cublasCgetrsBatched

CGETRS solves a system of linear equations A * X = B, A**T * X = B, or A**H * X = B with a general N-by-N matrix A using the LU factorization computed by CGETRF. 

 integer(4) function cublasCgetrsBatched(h, trans, n, nrhs, Aarray, lda, ipvt, Barray, ldb, info, batchCount)
   type(cublasHandle) :: h
   integer :: trans ! integer or character(1) variable
   integer :: n, nrhs
   type(c_devptr), device :: Aarray(*)
   integer :: lda
   integer, device :: ipvt(*)
   type(c_devptr), device :: Barray(*)
   integer :: ldb
   integer :: info(*)
   integer :: batchCount

2.4.49. cublasCgemmBatched

CGEMM performs one of the matrix-matrix operations C := alpha*op( A )*op( B ) + beta*C, where op( X ) is one of op( X ) = X or op( X ) = X**T or op( X ) = X**H, alpha and beta are scalars, and A, B and C are matrices, with op( A ) an m by k matrix, op( B ) a k by n matrix and C an m by n matrix. 

 integer(4) function cublasCgemmBatched(h, transa, transb, m, n, k, alpha, Aarray, lda, Barray, ldb, beta, Carray, ldc, batchCount)
   type(cublasHandle) :: h
   integer :: transa ! integer or character(1) variable
   integer :: transb ! integer or character(1) variable
   integer :: m, n, k
   complex(4), device :: alpha ! device or host variable
   type(c_devptr), device :: Aarray(*)
   integer :: lda
   type(c_devptr), device :: Barray(*)
   integer :: ldb
   complex(4), device :: beta ! device or host variable
   type(c_devptr), device :: Carray(*)
   integer :: ldc
   integer :: batchCount 
 integer(4) function cublasCgemmBatched_v2(h, transa, transb, m, n, k, alpha,           Aarray, lda, Barray, ldb, beta, Carray, ldc, batchCount)
   type(cublasHandle) :: h
   integer :: transa
   integer :: transb
   integer :: m, n, k
   complex(4), device :: alpha ! device or host variable
   type(c_devptr), device :: Aarray(*)
   integer :: lda
   type(c_devptr), device :: Barray(*)
   integer :: ldb
   complex(4), device :: beta ! device or host variable
   type(c_devptr), device :: Carray(*)
   integer :: ldc
   integer :: batchCount

2.4.50. cublasCtrsmBatched

CTRSM solves one of the matrix equations op( A )*X = alpha*B, or X*op( A ) = alpha*B, where alpha is a scalar, X and B are m by n matrices, A is a unit, or non-unit, upper or lower triangular matrix and op( A ) is one of op( A ) = A or op( A ) = A**T or op( A ) = A**H. The matrix X is overwritten on B. 

 integer(4) function cublasCtrsmBatched( h, side, uplo, trans, diag, m, n, alpha, A, lda, B, ldb, batchCount)
   type(cublasHandle) :: h
   integer :: side ! integer or character(1) variable
   integer :: uplo ! integer or character(1) variable
   integer :: trans ! integer or character(1) variable
   integer :: diag ! integer or character(1) variable
   integer :: m, n
   complex(4), device :: alpha ! device or host variable
   type(c_devptr), device :: A(*)
   integer :: lda
   type(c_devptr), device :: B(*)
   integer :: ldb
   integer :: batchCount 
 integer(4) function cublasCtrsmBatched_v2( h, side, uplo, trans, diag, m, n, alpha, A, lda, B, ldb, batchCount)
   type(cublasHandle) :: h
   integer :: side
   integer :: uplo
   integer :: trans
   integer :: diag
   integer :: m, n
   complex(4), device :: alpha ! device or host variable
   type(c_devptr), device :: A(*)
   integer :: lda
   type(c_devptr), device :: B(*)
   integer :: ldb
   integer :: batchCount

2.4.51. cublasCmatinvBatched

cublasCmatinvBatched is a short cut of cublasCgetrfBatched plus cublasCgetriBatched. However it only works if n is less than 32. If not, the user has to go through cublasCgetrfBatched and cublasCgetriBatched. 

 integer(4) function cublasCmatinvBatched(h, n, Aarray, lda, Ainv, lda_inv, info, batchCount)
   type(cublasHandle) :: h
   integer :: n
   type(c_devptr), device :: Aarray(*)
   integer :: lda
   type(c_devptr), device :: Ainv(*)
   integer :: lda_inv
   integer, device :: info(*)
   integer :: batchCount

2.4.52. cublasCgeqrfBatched

CGEQRF computes a QR factorization of a complex M-by-N matrix A: A = Q * R. 

 integer(4) function cublasCgeqrfBatched(h, m, n, Aarray, lda, Tau, info, batchCount)
   type(cublasHandle) :: h
   integer :: m, n
   type(c_devptr), device :: Aarray(*)
   integer :: lda
   type(c_devptr), device :: Tau(*)
   integer :: info(*)
   integer :: batchCount

2.4.53. cublasCgelsBatched

CGELS solves overdetermined or underdetermined complex linear systems involving an M-by-N matrix A, or its conjugate-transpose, using a QR or LQ factorization of A. It is assumed that A has full rank. The following options are provided: 1. If TRANS = 'N' and m >= n: find the least squares solution of an overdetermined system, i.e., solve the least squares problem minimize || B - A*X ||. 2. If TRANS = 'N' and m < n: find the minimum norm solution of an underdetermined system A * X = B. 3. If TRANS = 'C' and m >= n: find the minimum norm solution of an undetermined system A**H * X = B. 4. If TRANS = 'C' and m < n: find the least squares solution of an overdetermined system, i.e., solve the least squares problem minimize || B - A**H * X ||. Several right hand side vectors b and solution vectors x can be handled in a single call; they are stored as the columns of the M-by-NRHS right hand side matrix B and the N-by-NRHS solution matrix X. 

 integer(4) function cublasCgelsBatched(h, trans, m, n, nrhs, Aarray, lda, Carray, ldc, info, devinfo, batchCount)
   type(cublasHandle) :: h
   integer :: trans ! integer or character(1) variable
   integer :: m, n, nrhs
   type(c_devptr), device :: Aarray(*)
   integer :: lda
   type(c_devptr), device :: Carray(*)
   integer :: ldc
   integer :: info(*)
   integer, device :: devinfo(*)
   integer :: batchCount

2.4.54. cublasCgemmStridedBatched

CGEMM performs one of the matrix-matrix operations C := alpha*op( A )*op( B ) + beta*C, where op( X ) is one of op( X ) = X or op( X ) = X**T, alpha and beta are scalars, and A, B and C are matrices, with op( A ) an m by k matrix, op( B ) a k by n matrix and C an m by n matrix. 

 integer(4) function cublasCgemmStridedBatched(h, transa, transb, m, n, k, alpha, Aarray, lda, strideA, Barray, ldb, strideB, beta, Carray, ldc, strideC, batchCount)
   type(cublasHandle) :: h
   integer :: transa ! integer or character(1) variable
   integer :: transb ! integer or character(1) variable
   integer :: m, n, k
   complex(4), device :: alpha ! device or host variable
   complex(4), device :: Aarray(*)
   integer :: lda
   integer :: strideA
   complex(4), device :: Barray(*)
   integer :: ldb
   integer :: strideB
   complex(4), device :: beta ! device or host variable
   complex(4), device :: Carray(*)
   integer :: ldc
   integer :: strideC
   integer :: batchCount 
 integer(4) function cublasCgemmStridedBatched_v2(h, transa, transb, m, n, k, alpha,           Aarray, lda, strideA, Barray, ldb, strideB, beta, Carray, ldc, strideC, batchCount)
   type(cublasHandle) :: h
   integer :: transa
   integer :: transb
   integer :: m, n, k
   complex(4), device :: alpha ! device or host variable
   complex(4), device :: Aarray(*)
   integer :: lda
   integer :: strideA
   complex(4), device :: Barray(*)
   integer :: ldb
   integer :: strideB
   complex(4), device :: beta ! device or host variable
   complex(4), device :: Carray(*)
   integer :: ldc
   integer :: strideC
   integer :: batchCount

2.5. Double Precision Complex Functions and Subroutines

This section contains interfaces to the double precision complex BLAS and cuBLAS functions and subroutines.

2.5.1. izamax

IZAMAX finds the index of the element having the maximum absolute value. 

 integer(4) function izamax(n, x, incx)
   integer :: n
   complex(8), device, dimension(*) :: x ! device or host variable
   integer :: incx 
 integer(4) function cublasIzamax(n, x, incx)
   integer :: n
   complex(8), device, dimension(*) :: x
   integer :: incx 
 integer(4) function cublasIzamax_v2(h, n, x, incx, res)
   type(cublasHandle) :: h
   integer :: n
   complex(8), device, dimension(*) :: x
   integer :: incx
   integer, device :: res ! device or host variable

2.5.2. izamin

IZAMIN finds the index of the element having the minimum absolute value. 

 integer(4) function izamin(n, x, incx)
   integer :: n
   complex(8), device, dimension(*) :: x ! device or host variable
   integer :: incx 
 integer(4) function cublasIzamin(n, x, incx)
   integer :: n
   complex(8), device, dimension(*) :: x
   integer :: incx 
 integer(4) function cublasIzamin_v2(h, n, x, incx, res)
   type(cublasHandle) :: h
   integer :: n
   complex(8), device, dimension(*) :: x
   integer :: incx
   integer, device :: res ! device or host variable

2.5.3. dzasum

DZASUM takes the sum of the absolute values. 

 real(8) function dzasum(n, x, incx)
   integer :: n
   complex(8), device, dimension(*) :: x ! device or host variable
   integer :: incx 
 real(8) function cublasDzasum(n, x, incx)
   integer :: n
   complex(8), device, dimension(*) :: x
   integer :: incx 
 integer(4) function cublasDzasum_v2(h, n, x, incx, res)
   type(cublasHandle) :: h
   integer :: n
   complex(8), device, dimension(*) :: x
   integer :: incx
   real(8), device :: res ! device or host variable

2.5.4. zaxpy

ZAXPY constant times a vector plus a vector. 

 subroutine zaxpy(n, a, x, incx, y, incy)
   integer :: n
   complex(8), device :: a ! device or host variable
   complex(8), device, dimension(*) :: x, y ! device or host variable
   integer :: incx, incy 
 subroutine cublasZaxpy(n, a, x, incx, y, incy)
   integer :: n
   complex(8), device :: a ! device or host variable
   complex(8), device, dimension(*) :: x, y
   integer :: incx, incy 
 integer(4) function cublasZaxpy_v2(h, n, a, x, incx, y, incy)
   type(cublasHandle) :: h
   integer :: n
   complex(8), device :: a ! device or host variable
   complex(8), device, dimension(*) :: x, y
   integer :: incx, incy

2.5.5. zcopy

ZCOPY copies a vector, x, to a vector, y. 

 subroutine zcopy(n, x, incx, y, incy)
   integer :: n
   complex(8), device, dimension(*) :: x, y ! device or host variable
   integer :: incx, incy 
 subroutine cublasZcopy(n, x, incx, y, incy)
   integer :: n
   complex(8), device, dimension(*) :: x, y
   integer :: incx, incy 
 integer(4) function cublasZcopy_v2(h, n, x, incx, y, incy)
   type(cublasHandle) :: h
   integer :: n
   complex(8), device, dimension(*) :: x, y
   integer :: incx, incy

2.5.6. zdotc

ZDOTC forms the dot product of a vector. 

 complex(8) function zdotc(n, x, incx, y, incy)
   integer :: n
   complex(8), device, dimension(*) :: x, y ! device or host variable
   integer :: incx, incy 
 complex(8) function cublasZdotc(n, x, incx, y, incy)
   integer :: n
   complex(8), device, dimension(*) :: x, y
   integer :: incx, incy 
 integer(4) function cublasZdotc_v2(h, n, x, incx, y, incy, res)
   type(cublasHandle) :: h
   integer :: n
   complex(8), device, dimension(*) :: x, y
   integer :: incx, incy
   complex(8), device :: res ! device or host variable

2.5.7. zdotu

ZDOTU forms the dot product of two vectors. 

 complex(8) function zdotu(n, x, incx, y, incy)
   integer :: n
   complex(8), device, dimension(*) :: x, y ! device or host variable
   integer :: incx, incy 
 complex(8) function cublasZdotu(n, x, incx, y, incy)
   integer :: n
   complex(8), device, dimension(*) :: x, y
   integer :: incx, incy 
 integer(4) function cublasZdotu_v2(h, n, x, incx, y, incy, res)
   type(cublasHandle) :: h
   integer :: n
   complex(8), device, dimension(*) :: x, y
   integer :: incx, incy
   complex(8), device :: res ! device or host variable

2.5.8. dznrm2

DZNRM2 returns the euclidean norm of a vector via the function name, so that DZNRM2 := sqrt( x**H*x ) 

 real(8) function dznrm2(n, x, incx)
   integer :: n
   complex(8), device, dimension(*) :: x ! device or host variable
   integer :: incx 
 real(8) function cublasDznrm2(n, x, incx)
   integer :: n
   complex(8), device, dimension(*) :: x
   integer :: incx 
 integer(4) function cublasDznrm2_v2(h, n, x, incx, res)
   type(cublasHandle) :: h
   integer :: n
   complex(8), device, dimension(*) :: x
   integer :: incx
   real(8), device :: res ! device or host variable

2.5.9. zrot

ZROT applies a plane rotation, where the cos (C) is real and the sin (S) is complex, and the vectors CX and CY are complex. 

 subroutine zrot(n, x, incx, y, incy, sc, ss)
   integer :: n
   real(8), device :: sc ! device or host variable
   complex(8), device :: ss ! device or host variable
   complex(8), device, dimension(*) :: x, y ! device or host variable
   integer :: incx, incy 
 subroutine cublasZrot(n, x, incx, y, incy, sc, ss)
   integer :: n
   real(8), device :: sc ! device or host variable
   complex(8), device :: ss ! device or host variable
   complex(8), device, dimension(*) :: x, y
   integer :: incx, incy 
 integer(4) function cublasZrot_v2(h, n, x, incx, y, incy, sc, ss)
   type(cublasHandle) :: h
   integer :: n
   real(8), device :: sc ! device or host variable
   complex(8), device :: ss ! device or host variable
   complex(8), device, dimension(*) :: x, y
   integer :: incx, incy

2.5.10. zsrot

ZSROT applies a plane rotation, where the cos and sin (c and s) are real and the vectors cx and cy are complex. 

 subroutine zsrot(n, x, incx, y, incy, sc, ss)
   integer :: n
   real(8), device :: sc, ss ! device or host variable
   complex(8), device, dimension(*) :: x, y ! device or host variable
   integer :: incx, incy 
 subroutine cublasZsrot(n, x, incx, y, incy, sc, ss)
   integer :: n
   real(8), device :: sc, ss ! device or host variable
   complex(8), device, dimension(*) :: x, y
   integer :: incx, incy 
 integer(4) function cublasZsrot_v2(h, n, x, incx, y, incy, sc, ss)
   type(cublasHandle) :: h
   integer :: n
   real(8), device :: sc, ss ! device or host variable
   complex(8), device, dimension(*) :: x, y
   integer :: incx, incy

2.5.11. zrotg

ZROTG determines a double complex Givens rotation. 

 subroutine zrotg(sa, sb, sc, ss)
   complex(8), device :: sa, sb, ss ! device or host variable
   real(8), device :: sc ! device or host variable 
 subroutine cublasZrotg(sa, sb, sc, ss)
   complex(8), device :: sa, sb, ss ! device or host variable
   real(8), device :: sc ! device or host variable 
 integer(4) function cublasZrotg_v2(h, sa, sb, sc, ss)
   type(cublasHandle) :: h
   complex(8), device :: sa, sb, ss ! device or host variable
   real(8), device :: sc ! device or host variable

2.5.12. zscal

ZSCAL scales a vector by a constant. 

 subroutine zscal(n, a, x, incx)
   integer :: n
   complex(8), device :: a ! device or host variable
   complex(8), device, dimension(*) :: x ! device or host variable
   integer :: incx 
 subroutine cublasZscal(n, a, x, incx)
   integer :: n
   complex(8), device :: a ! device or host variable
   complex(8), device, dimension(*) :: x
   integer :: incx 
 integer(4) function cublasZscal_v2(h, n, a, x, incx)
   type(cublasHandle) :: h
   integer :: n
   complex(8), device :: a ! device or host variable
   complex(8), device, dimension(*) :: x
   integer :: incx

2.5.13. zdscal

ZDSCAL scales a vector by a constant. 

 subroutine zdscal(n, a, x, incx)
   integer :: n
   real(8), device :: a ! device or host variable
   complex(8), device, dimension(*) :: x ! device or host variable
   integer :: incx 
 subroutine cublasZdscal(n, a, x, incx)
   integer :: n
   real(8), device :: a ! device or host variable
   complex(8), device, dimension(*) :: x
   integer :: incx 
 integer(4) function cublasZdscal_v2(h, n, a, x, incx)
   type(cublasHandle) :: h
   integer :: n
   real(8), device :: a ! device or host variable
   complex(8), device, dimension(*) :: x
   integer :: incx

2.5.14. zswap

ZSWAP interchanges two vectors. 

 subroutine zswap(n, x, incx, y, incy)
   integer :: n
   complex(8), device, dimension(*) :: x, y ! device or host variable
   integer :: incx, incy 
 subroutine cublasZswap(n, x, incx, y, incy)
   integer :: n
   complex(8), device, dimension(*) :: x, y
   integer :: incx, incy 
 integer(4) function cublasZswap_v2(h, n, x, incx, y, incy)
   type(cublasHandle) :: h
   integer :: n
   complex(8), device, dimension(*) :: x, y
   integer :: incx, incy

2.5.15. zgbmv

ZGBMV performs one of the matrix-vector operations y := alpha*A*x + beta*y, or y := alpha*A**T*x + beta*y, or y := alpha*A**H*x + beta*y, where alpha and beta are scalars, x and y are vectors and A is an m by n band matrix, with kl sub-diagonals and ku super-diagonals. 

 subroutine zgbmv(t, m, n, kl, ku, alpha, a, lda, x, incx, beta, y, incy)
   character*1 :: t
   integer :: m, n, kl, ku, lda, incx, incy
   complex(8), device, dimension(lda, *) :: a ! device or host variable
   complex(8), device, dimension(*) :: x, y ! device or host variable
   complex(8), device :: alpha, beta ! device or host variable 
 subroutine cublasZgbmv(t, m, n, kl, ku, alpha, a, lda, x, incx, beta, y, incy)
   character*1 :: t
   integer :: m, n, kl, ku, lda, incx, incy
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(*) :: x, y
   complex(8), device :: alpha, beta ! device or host variable 
 integer(4) function cublasZgbmv_v2(h, t, m, n, kl, ku, alpha, a, lda, x, incx, beta, y, incy)
   type(cublasHandle) :: h
   integer :: t
   integer :: m, n, kl, ku, lda, incx, incy
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(*) :: x, y
   complex(8), device :: alpha, beta ! device or host variable

2.5.16. zgemv

ZGEMV performs one of the matrix-vector operations y := alpha*A*x + beta*y, or y := alpha*A**T*x + beta*y, or y := alpha*A**H*x + beta*y, where alpha and beta are scalars, x and y are vectors and A is an m by n matrix. 

 subroutine zgemv(t, m, n, alpha, a, lda, x, incx, beta, y, incy)
   character*1 :: t
   integer :: m, n, lda, incx, incy
   complex(8), device, dimension(lda, *) :: a ! device or host variable
   complex(8), device, dimension(*) :: x, y ! device or host variable
   complex(8), device :: alpha, beta ! device or host variable 
 subroutine cublasZgemv(t, m, n, alpha, a, lda, x, incx, beta, y, incy)
   character*1 :: t
   integer :: m, n, lda, incx, incy
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(*) :: x, y
   complex(8), device :: alpha, beta ! device or host variable 
 integer(4) function cublasZgemv_v2(h, t, m, n, alpha, a, lda, x, incx, beta, y, incy)
   type(cublasHandle) :: h
   integer :: t
   integer :: m, n, lda, incx, incy
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(*) :: x, y
   complex(8), device :: alpha, beta ! device or host variable

2.5.17. zgerc

ZGERC performs the rank 1 operation A := alpha*x*y**H + A, where alpha is a scalar, x is an m element vector, y is an n element vector and A is an m by n matrix. 

 subroutine zgerc(m, n, alpha, x, incx, y, incy, a, lda)
   integer :: m, n, lda, incx, incy
   complex(8), device, dimension(lda, *) :: a ! device or host variable
   complex(8), device, dimension(*) :: x, y ! device or host variable
   complex(8), device :: alpha ! device or host variable 
 subroutine cublasZgerc(m, n, alpha, x, incx, y, incy, a, lda)
   integer :: m, n, lda, incx, incy
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(*) :: x, y
   complex(8), device :: alpha ! device or host variable 
 integer(4) function cublasZgerc_v2(h, m, n, alpha, x, incx, y, incy, a, lda)
   type(cublasHandle) :: h
   integer :: m, n, lda, incx, incy
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(*) :: x, y
   complex(8), device :: alpha ! device or host variable

2.5.18. zgeru

ZGERU performs the rank 1 operation A := alpha*x*y**T + A, where alpha is a scalar, x is an m element vector, y is an n element vector and A is an m by n matrix. 

 subroutine zgeru(m, n, alpha, x, incx, y, incy, a, lda)
   integer :: m, n, lda, incx, incy
   complex(8), device, dimension(lda, *) :: a ! device or host variable
   complex(8), device, dimension(*) :: x, y ! device or host variable
   complex(8), device :: alpha ! device or host variable 
 subroutine cublasZgeru(m, n, alpha, x, incx, y, incy, a, lda)
   integer :: m, n, lda, incx, incy
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(*) :: x, y
   complex(8), device :: alpha ! device or host variable 
 integer(4) function cublasZgeru_v2(h, m, n, alpha, x, incx, y, incy, a, lda)
   type(cublasHandle) :: h
   integer :: m, n, lda, incx, incy
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(*) :: x, y
   complex(8), device :: alpha ! device or host variable

2.5.19. zsymv

ZSYMV performs the matrix-vector operation y := alpha*A*x + beta*y, where alpha and beta are scalars, x and y are n element vectors and A is an n by n symmetric matrix. 

 subroutine zsymv(uplo, n, alpha, a, lda, x, incx, beta, y, incy)
   character*1 :: uplo
   integer :: n, lda, incx, incy
   complex(8), device, dimension(lda, *) :: a ! device or host variable
   complex(8), device, dimension(*) :: x, y ! device or host variable
   complex(8), device :: alpha, beta ! device or host variable 
 subroutine cublasZsymv(uplo, n, alpha, a, lda, x, incx, beta, y, incy)
   character*1 :: uplo
   integer :: n, lda, incx, incy
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(*) :: x, y
   complex(8), device :: alpha, beta ! device or host variable 
 integer(4) function cublasZsymv_v2(h, uplo, n, alpha, a, lda, x, incx, beta, y, incy)
   type(cublasHandle) :: h
   integer :: uplo
   integer :: n, lda, incx, incy
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(*) :: x, y
   complex(8), device :: alpha, beta ! device or host variable

2.5.20. zsyr

ZSYR performs the symmetric rank 1 operation A := alpha*x*x**H + A, where alpha is a complex scalar, x is an n element vector and A is an n by n symmetric matrix. 

 subroutine zsyr(t, n, alpha, x, incx, a, lda)
   character*1 :: t
   integer :: n, incx, lda
   complex(8), device, dimension(lda, *) :: a ! device or host variable
   complex(8), device, dimension(*) :: x ! device or host variable
   complex(8), device :: alpha ! device or host variable 
 subroutine cublasZsyr(t, n, alpha, x, incx, a, lda)
   character*1 :: t
   integer :: n, incx, lda
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(*) :: x
   complex(8), device :: alpha ! device or host variable 
 integer(4) function cublasZsyr_v2(h, t, n, alpha, x, incx, a, lda)
   type(cublasHandle) :: h
   integer :: t
   integer :: n, incx, lda
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(*) :: x
   complex(8), device :: alpha ! device or host variable

2.5.21. zsyr2

ZSYR2 performs the symmetric rank 2 operation A := alpha*x*y' + alpha*y*x' + A, where alpha is a complex scalar, x and y are n element vectors and A is an n by n SY matrix. 

 subroutine zsyr2(t, n, alpha, x, incx, y, incy, a, lda)
   character*1 :: t
   integer :: n, incx, incy, lda
   complex(8), device, dimension(lda, *) :: a ! device or host variable
   complex(8), device, dimension(*) :: x, y ! device or host variable
   complex(8), device :: alpha ! device or host variable 
 subroutine cublasZsyr2(t, n, alpha, x, incx, y, incy, a, lda)
   character*1 :: t
   integer :: n, incx, incy, lda
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(*) :: x, y
   complex(8), device :: alpha ! device or host variable 
 integer(4) function cublasZsyr2_v2(h, t, n, alpha, x, incx, y, incy, a, lda)
   type(cublasHandle) :: h
   integer :: t
   integer :: n, incx, incy, lda
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(*) :: x, y
   complex(8), device :: alpha ! device or host variable

2.5.22. ztbmv

ZTBMV performs one of the matrix-vector operations x := A*x, or x := A**T*x, or x := A**H*x, where x is an n element vector and A is an n by n unit, or non-unit, upper or lower triangular band matrix, with ( k + 1 ) diagonals. 

 subroutine ztbmv(u, t, d, n, k, a, lda, x, incx)
   character*1 :: u, t, d
   integer :: n, k, incx, lda
   complex(8), device, dimension(lda, *) :: a ! device or host variable
   complex(8), device, dimension(*) :: x ! device or host variable 
 subroutine cublasZtbmv(u, t, d, n, k, a, lda, x, incx)
   character*1 :: u, t, d
   integer :: n, k, incx, lda
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(*) :: x 
 integer(4) function cublasZtbmv_v2(h, u, t, d, n, k, a, lda, x, incx)
   type(cublasHandle) :: h
   integer :: u, t, d
   integer :: n, k, incx, lda
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(*) :: x

2.5.23. ztbsv

ZTBSV solves one of the systems of equations A*x = b, or A**T*x = b, or A**H*x = b, where b and x are n element vectors and A is an n by n unit, or non-unit, upper or lower triangular band matrix, with ( k + 1 ) diagonals. No test for singularity or near-singularity is included in this routine. Such tests must be performed before calling this routine. 

 subroutine ztbsv(u, t, d, n, k, a, lda, x, incx)
   character*1 :: u, t, d
   integer :: n, k, incx, lda
   complex(8), device, dimension(lda, *) :: a ! device or host variable
   complex(8), device, dimension(*) :: x ! device or host variable 
 subroutine cublasZtbsv(u, t, d, n, k, a, lda, x, incx)
   character*1 :: u, t, d
   integer :: n, k, incx, lda
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(*) :: x 
 integer(4) function cublasZtbsv_v2(h, u, t, d, n, k, a, lda, x, incx)
   type(cublasHandle) :: h
   integer :: u, t, d
   integer :: n, k, incx, lda
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(*) :: x

2.5.24. ztpmv

ZTPMV performs one of the matrix-vector operations x := A*x, or x := A**T*x, or x := A**H*x, where x is an n element vector and A is an n by n unit, or non-unit, upper or lower triangular matrix, supplied in packed form. 

 subroutine ztpmv(u, t, d, n, a, x, incx)
   character*1 :: u, t, d
   integer :: n, incx
   complex(8), device, dimension(*) :: a, x ! device or host variable 
 subroutine cublasZtpmv(u, t, d, n, a, x, incx)
   character*1 :: u, t, d
   integer :: n, incx
   complex(8), device, dimension(*) :: a, x 
 integer(4) function cublasZtpmv_v2(h, u, t, d, n, a, x, incx)
   type(cublasHandle) :: h
   integer :: u, t, d
   integer :: n, incx
   complex(8), device, dimension(*) :: a, x

2.5.25. ztpsv

ZTPSV solves one of the systems of equations A*x = b, or A**T*x = b, or A**H*x = b, where b and x are n element vectors and A is an n by n unit, or non-unit, upper or lower triangular matrix, supplied in packed form. No test for singularity or near-singularity is included in this routine. Such tests must be performed before calling this routine. 

 subroutine ztpsv(u, t, d, n, a, x, incx)
   character*1 :: u, t, d
   integer :: n, incx
   complex(8), device, dimension(*) :: a, x ! device or host variable 
 subroutine cublasZtpsv(u, t, d, n, a, x, incx)
   character*1 :: u, t, d
   integer :: n, incx
   complex(8), device, dimension(*) :: a, x 
 integer(4) function cublasZtpsv_v2(h, u, t, d, n, a, x, incx)
   type(cublasHandle) :: h
   integer :: u, t, d
   integer :: n, incx
   complex(8), device, dimension(*) :: a, x

2.5.26. ztrmv

ZTRMV performs one of the matrix-vector operations x := A*x, or x := A**T*x, or x := A**H*x, where x is an n element vector and A is an n by n unit, or non-unit, upper or lower triangular matrix. 

 subroutine ztrmv(u, t, d, n, a, lda, x, incx)
   character*1 :: u, t, d
   integer :: n, incx, lda
   complex(8), device, dimension(lda, *) :: a ! device or host variable
   complex(8), device, dimension(*) :: x ! device or host variable 
 subroutine cublasZtrmv(u, t, d, n, a, lda, x, incx)
   character*1 :: u, t, d
   integer :: n, incx, lda
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(*) :: x 
 integer(4) function cublasZtrmv_v2(h, u, t, d, n, a, lda, x, incx)
   type(cublasHandle) :: h
   integer :: u, t, d
   integer :: n, incx, lda
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(*) :: x

2.5.27. ztrsv

ZTRSV solves one of the systems of equations A*x = b, or A**T*x = b, or A**H*x = b, where b and x are n element vectors and A is an n by n unit, or non-unit, upper or lower triangular matrix. No test for singularity or near-singularity is included in this routine. Such tests must be performed before calling this routine. 

 subroutine ztrsv(u, t, d, n, a, lda, x, incx)
   character*1 :: u, t, d
   integer :: n, incx, lda
   complex(8), device, dimension(lda, *) :: a ! device or host variable
   complex(8), device, dimension(*) :: x ! device or host variable 
 subroutine cublasZtrsv(u, t, d, n, a, lda, x, incx)
   character*1 :: u, t, d
   integer :: n, incx, lda
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(*) :: x 
 integer(4) function cublasZtrsv_v2(h, u, t, d, n, a, lda, x, incx)
   type(cublasHandle) :: h
   integer :: u, t, d
   integer :: n, incx, lda
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(*) :: x

2.5.28. zhbmv

ZHBMV performs the matrix-vector operation y := alpha*A*x + beta*y, where alpha and beta are scalars, x and y are n element vectors and A is an n by n hermitian band matrix, with k super-diagonals. 

 subroutine zhbmv(uplo, n, k, alpha, a, lda, x, incx, beta, y, incy)
   character*1 :: uplo
   integer :: k, n, lda, incx, incy
   complex(8), device, dimension(lda, *) :: a ! device or host variable
   complex(8), device, dimension(*) :: x, y ! device or host variable
   complex(8), device :: alpha, beta ! device or host variable 
 subroutine cublasZhbmv(uplo, n, k, alpha, a, lda, x, incx, beta, y, incy)
   character*1 :: uplo
   integer :: k, n, lda, incx, incy
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(*) :: x, y
   complex(8), device :: alpha, beta ! device or host variable 
 integer(4) function cublasZhbmv_v2(h, uplo, n, k, alpha, a, lda, x, incx, beta, y, incy)
   type(cublasHandle) :: h
   integer :: uplo
   integer :: k, n, lda, incx, incy
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(*) :: x, y
   complex(8), device :: alpha, beta ! device or host variable

2.5.29. zhemv

ZHEMV performs the matrix-vector operation y := alpha*A*x + beta*y, where alpha and beta are scalars, x and y are n element vectors and A is an n by n hermitian matrix. 

 subroutine zhemv(uplo, n, alpha, a, lda, x, incx, beta, y, incy)
   character*1 :: uplo
   integer :: n, lda, incx, incy
   complex(8), device, dimension(lda, *) :: a ! device or host variable
   complex(8), device, dimension(*) :: x, y ! device or host variable
   complex(8), device :: alpha, beta ! device or host variable 
 subroutine cublasZhemv(uplo, n, alpha, a, lda, x, incx, beta, y, incy)
   character*1 :: uplo
   integer :: n, lda, incx, incy
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(*) :: x, y
   complex(8), device :: alpha, beta ! device or host variable 
 integer(4) function cublasZhemv_v2(h, uplo, n, alpha, a, lda, x, incx, beta, y, incy)
   type(cublasHandle) :: h
   integer :: uplo
   integer :: n, lda, incx, incy
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(*) :: x, y
   complex(8), device :: alpha, beta ! device or host variable

2.5.30. zhpmv

ZHPMV performs the matrix-vector operation y := alpha*A*x + beta*y, where alpha and beta are scalars, x and y are n element vectors and A is an n by n hermitian matrix, supplied in packed form. 

 subroutine zhpmv(uplo, n, alpha, a, x, incx, beta, y, incy)
   character*1 :: uplo
   integer :: n, incx, incy
   complex(8), device, dimension(*) :: a, x, y ! device or host variable
   complex(8), device :: alpha, beta ! device or host variable 
 subroutine cublasZhpmv(uplo, n, alpha, a, x, incx, beta, y, incy)
   character*1 :: uplo
   integer :: n, incx, incy
   complex(8), device, dimension(*) :: a, x, y
   complex(8), device :: alpha, beta ! device or host variable 
 integer(4) function cublasZhpmv_v2(h, uplo, n, alpha, a, x, incx, beta, y, incy)
   type(cublasHandle) :: h
   integer :: uplo
   integer :: n, incx, incy
   complex(8), device, dimension(*) :: a, x, y
   complex(8), device :: alpha, beta ! device or host variable

2.5.31. zher

ZHER performs the hermitian rank 1 operation A := alpha*x*x**H + A, where alpha is a real scalar, x is an n element vector and A is an n by n hermitian matrix. 

 subroutine zher(t, n, alpha, x, incx, a, lda)
   character*1 :: t
   integer :: n, incx, lda
   complex(8), device, dimension(*) :: a, x ! device or host variable
   real(8), device :: alpha ! device or host variable 
 subroutine cublasZher(t, n, alpha, x, incx, a, lda)
   character*1 :: t
   integer :: n, incx, lda
   complex(8), device, dimension(*) :: a, x
   real(8), device :: alpha ! device or host variable 
 integer(4) function cublasZher_v2(h, t, n, alpha, x, incx, a, lda)
   type(cublasHandle) :: h
   integer :: t
   integer :: n, incx, lda
   complex(8), device, dimension(*) :: a, x
   real(8), device :: alpha ! device or host variable

2.5.32. zher2

ZHER2 performs the hermitian rank 2 operation A := alpha*x*y**H + conjg( alpha )*y*x**H + A, where alpha is a scalar, x and y are n element vectors and A is an n by n hermitian matrix. 

 subroutine zher2(t, n, alpha, x, incx, y, incy, a, lda)
   character*1 :: t
   integer :: n, incx, incy
   complex(8), device, dimension(*) :: a, x, y ! device or host variable
   complex(8), device :: alpha ! device or host variable 
 subroutine cublasZher2(t, n, alpha, x, incx, y, incy, a, lda)
   character*1 :: t
   integer :: n, incx, incy, lda
   complex(8), device, dimension(*) :: a, x, y
   complex(8), device :: alpha ! device or host variable 
 integer(4) function cublasZher2_v2(h, t, n, alpha, x, incx, y, incy, a, lda)
   type(cublasHandle) :: h
   integer :: t
   integer :: n, incx, incy, lda
   complex(8), device, dimension(*) :: a, x, y
   complex(8), device :: alpha ! device or host variable

2.5.33. zhpr

ZHPR performs the hermitian rank 1 operation A := alpha*x*x**H + A, where alpha is a real scalar, x is an n element vector and A is an n by n hermitian matrix, supplied in packed form. 

 subroutine zhpr(t, n, alpha, x, incx, a)
   character*1 :: t
   integer :: n, incx
   complex(8), device, dimension(*) :: a, x ! device or host variable
   real(8), device :: alpha ! device or host variable 
 subroutine cublasZhpr(t, n, alpha, x, incx, a)
   character*1 :: t
   integer :: n, incx
   complex(8), device, dimension(*) :: a, x
   real(8), device :: alpha ! device or host variable 
 integer(4) function cublasZhpr_v2(h, t, n, alpha, x, incx, a)
   type(cublasHandle) :: h
   integer :: t
   integer :: n, incx
   complex(8), device, dimension(*) :: a, x
   real(8), device :: alpha ! device or host variable

2.5.34. zhpr2

ZHPR2 performs the hermitian rank 2 operation A := alpha*x*y**H + conjg( alpha )*y*x**H + A, where alpha is a scalar, x and y are n element vectors and A is an n by n hermitian matrix, supplied in packed form. 

 subroutine zhpr2(t, n, alpha, x, incx, y, incy, a)
   character*1 :: t
   integer :: n, incx, incy
   complex(8), device, dimension(*) :: a, x, y ! device or host variable
   complex(8), device :: alpha ! device or host variable 
 subroutine cublasZhpr2(t, n, alpha, x, incx, y, incy, a)
   character*1 :: t
   integer :: n, incx, incy
   complex(8), device, dimension(*) :: a, x, y
   complex(8), device :: alpha ! device or host variable 
 integer(4) function cublasZhpr2_v2(h, t, n, alpha, x, incx, y, incy, a)
   type(cublasHandle) :: h
   integer :: t
   integer :: n, incx, incy
   complex(8), device, dimension(*) :: a, x, y
   complex(8), device :: alpha ! device or host variable

2.5.35. zgemm

ZGEMM performs one of the matrix-matrix operations C := alpha*op( A )*op( B ) + beta*C, where op( X ) is one of op( X ) = X or op( X ) = X**T or op( X ) = X**H, alpha and beta are scalars, and A, B and C are matrices, with op( A ) an m by k matrix, op( B ) a k by n matrix and C an m by n matrix. 

 subroutine zgemm(transa, transb, m, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   character*1 :: transa, transb
   integer :: m, n, k, lda, ldb, ldc
   complex(8), device, dimension(lda, *) :: a ! device or host variable
   complex(8), device, dimension(ldb, *) :: b ! device or host variable
   complex(8), device, dimension(ldc, *) :: c ! device or host variable
   complex(8), device :: alpha, beta ! device or host variable 
 subroutine cublasZgemm(transa, transb, m, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   character*1 :: transa, transb
   integer :: m, n, k, lda, ldb, ldc
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(ldb, *) :: b
   complex(8), device, dimension(ldc, *) :: c
   complex(8), device :: alpha, beta ! device or host variable 
 integer(4) function cublasZgemm_v2(h, transa, transb, m, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: transa, transb
   integer :: m, n, k, lda, ldb, ldc
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(ldb, *) :: b
   complex(8), device, dimension(ldc, *) :: c
   complex(8), device :: alpha, beta ! device or host variable

2.5.36. zsymm

ZSYMM performs one of the matrix-matrix operations C := alpha*A*B + beta*C, or C := alpha*B*A + beta*C, where alpha and beta are scalars, A is a symmetric matrix and B and C are m by n matrices. 

 subroutine zsymm(side, uplo, m, n, alpha, a, lda, b, ldb, beta, c, ldc)
   character*1 :: side, uplo
   integer :: m, n, lda, ldb, ldc
   complex(8), device, dimension(lda, *) :: a ! device or host variable
   complex(8), device, dimension(ldb, *) :: b ! device or host variable
   complex(8), device, dimension(ldc, *) :: c ! device or host variable
   complex(8), device :: alpha, beta ! device or host variable 
 subroutine cublasZsymm(side, uplo, m, n, alpha, a, lda, b, ldb, beta, c, ldc)
   character*1 :: side, uplo
   integer :: m, n, lda, ldb, ldc
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(ldb, *) :: b
   complex(8), device, dimension(ldc, *) :: c
   complex(8), device :: alpha, beta ! device or host variable 
 integer(4) function cublasZsymm_v2(h, side, uplo, m, n, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: side, uplo
   integer :: m, n, lda, ldb, ldc
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(ldb, *) :: b
   complex(8), device, dimension(ldc, *) :: c
   complex(8), device :: alpha, beta ! device or host variable

2.5.37. zsyrk

ZSYRK performs one of the symmetric rank k operations C := alpha*A*A**T + beta*C, or C := alpha*A**T*A + beta*C, where alpha and beta are scalars, C is an n by n symmetric matrix and A is an n by k matrix in the first case and a k by n matrix in the second case. 

 subroutine zsyrk(uplo, trans, n, k, alpha, a, lda, beta, c, ldc)
   character*1 :: uplo, trans
   integer :: n, k, lda, ldc
   complex(8), device, dimension(lda, *) :: a ! device or host variable
   complex(8), device, dimension(ldc, *) :: c ! device or host variable
   complex(8), device :: alpha, beta ! device or host variable 
 subroutine cublasZsyrk(uplo, trans, n, k, alpha, a, lda, beta, c, ldc)
   character*1 :: uplo, trans
   integer :: n, k, lda, ldc
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(ldc, *) :: c
   complex(8), device :: alpha, beta ! device or host variable 
 integer(4) function cublasZsyrk_v2(h, uplo, trans, n, k, alpha, a, lda, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: uplo, trans
   integer :: n, k, lda, ldc
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(ldc, *) :: c
   complex(8), device :: alpha, beta ! device or host variable

2.5.38. zsyr2k

ZSYR2K performs one of the symmetric rank 2k operations C := alpha*A*B**T + alpha*B*A**T + beta*C, or C := alpha*A**T*B + alpha*B**T*A + beta*C, where alpha and beta are scalars, C is an n by n symmetric matrix and A and B are n by k matrices in the first case and k by n matrices in the second case. 

 subroutine zsyr2k(uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   character*1 :: uplo, trans
   integer :: n, k, lda, ldb, ldc
   complex(8), device, dimension(lda, *) :: a ! device or host variable
   complex(8), device, dimension(ldb, *) :: b ! device or host variable
   complex(8), device, dimension(ldc, *) :: c ! device or host variable
   complex(8), device :: alpha, beta ! device or host variable 
 subroutine cublasZsyr2k(uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   character*1 :: uplo, trans
   integer :: n, k, lda, ldb, ldc
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(ldb, *) :: b
   complex(8), device, dimension(ldc, *) :: c
   complex(8), device :: alpha, beta ! device or host variable 
 integer(4) function cublasZsyr2k_v2(h, uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: uplo, trans
   integer :: n, k, lda, ldb, ldc
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(ldb, *) :: b
   complex(8), device, dimension(ldc, *) :: c
   complex(8), device :: alpha, beta ! device or host variable

2.5.39. zsyrkx

ZSYRKX performs a variation of the symmetric rank k update C := alpha*A*B**T + beta*C, where alpha and beta are scalars, C is an n by n symmetric matrix stored in lower or upper mode, and A and B are n by k matrices. This routine can be used when B is in such a way that the result is guaranteed to be symmetric. See the CUBLAS documentation for more details. 

 subroutine zsyrkx(uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   character*1 :: uplo, trans
   integer :: n, k, lda, ldb, ldc
   complex(8), device, dimension(lda, *) :: a ! device or host variable
   complex(8), device, dimension(ldb, *) :: b ! device or host variable
   complex(8), device, dimension(ldc, *) :: c ! device or host variable
   complex(8), device :: alpha, beta ! device or host variable 
 subroutine cublasZsyrkx(uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   character*1 :: uplo, trans
   integer :: n, k, lda, ldb, ldc
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(ldb, *) :: b
   complex(8), device, dimension(ldc, *) :: c
   complex(8), device :: alpha, beta ! device or host variable 
 integer(4) function cublasZsyrkx_v2(h, uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: uplo, trans
   integer :: n, k, lda, ldb, ldc
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(ldb, *) :: b
   complex(8), device, dimension(ldc, *) :: c
   complex(8), device :: alpha, beta ! device or host variable

2.5.40. ztrmm

ZTRMM performs one of the matrix-matrix operations B := alpha*op( A )*B, or B := alpha*B*op( A ) where alpha is a scalar, B is an m by n matrix, A is a unit, or non-unit, upper or lower triangular matrix and op( A ) is one of op( A ) = A or op( A ) = A**T or op( A ) = A**H. 

 subroutine ztrmm(side, uplo, transa, diag, m, n, alpha, a, lda, b, ldb)
   character*1 :: side, uplo, transa, diag
   integer :: m, n, lda, ldb
   complex(8), device, dimension(lda, *) :: a ! device or host variable
   complex(8), device, dimension(ldb, *) :: b ! device or host variable
   complex(8), device :: alpha ! device or host variable 
 subroutine cublasZtrmm(side, uplo, transa, diag, m, n, alpha, a, lda, b, ldb)
   character*1 :: side, uplo, transa, diag
   integer :: m, n, lda, ldb
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(ldb, *) :: b
   complex(8), device :: alpha ! device or host variable 
 integer(4) function cublasZtrmm_v2(h, side, uplo, transa, diag, m, n, alpha, a, lda, b, ldb, c, ldc)
   type(cublasHandle) :: h
   integer :: side, uplo, transa, diag
   integer :: m, n, lda, ldb, ldc
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(ldb, *) :: b
   complex(8), device, dimension(ldc, *) :: c
   complex(8), device :: alpha ! device or host variable

2.5.41. ztrsm

ZTRSM solves one of the matrix equations op( A )*X = alpha*B, or X*op( A ) = alpha*B, where alpha is a scalar, X and B are m by n matrices, A is a unit, or non-unit, upper or lower triangular matrix and op( A ) is one of op( A ) = A or op( A ) = A**T or op( A ) = A**H. The matrix X is overwritten on B. 

 subroutine ztrsm(side, uplo, transa, diag, m, n, alpha, a, lda, b, ldb)
   character*1 :: side, uplo, transa, diag
   integer :: m, n, lda, ldb
   complex(8), device, dimension(lda, *) :: a ! device or host variable
   complex(8), device, dimension(ldb, *) :: b ! device or host variable
   complex(8), device :: alpha ! device or host variable 
 subroutine cublasZtrsm(side, uplo, transa, diag, m, n, alpha, a, lda, b, ldb)
   character*1 :: side, uplo, transa, diag
   integer :: m, n, lda, ldb
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(ldb, *) :: b
   complex(8), device :: alpha ! device or host variable 
 integer(4) function cublasZtrsm_v2(h, side, uplo, transa, diag, m, n, alpha, a, lda, b, ldb)
   type(cublasHandle) :: h
   integer :: side, uplo, transa, diag
   integer :: m, n, lda, ldb
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(ldb, *) :: b
   complex(8), device :: alpha ! device or host variable

2.5.42. zhemm

ZHEMM performs one of the matrix-matrix operations C := alpha*A*B + beta*C, or C := alpha*B*A + beta*C, where alpha and beta are scalars, A is an hermitian matrix and B and C are m by n matrices. 

 subroutine zhemm(side, uplo, m, n, alpha, a, lda, b, ldb, beta, c, ldc)
   character*1 :: side, uplo
   integer :: m, n, lda, ldb, ldc
   complex(8), device, dimension(lda, *) :: a ! device or host variable
   complex(8), device, dimension(ldb, *) :: b ! device or host variable
   complex(8), device, dimension(ldc, *) :: c ! device or host variable
   complex(8), device :: alpha, beta ! device or host variable 
 subroutine cublasZhemm(side, uplo, m, n, alpha, a, lda, b, ldb, beta, c, ldc)
   character*1 :: side, uplo
   integer :: m, n, lda, ldb, ldc
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(ldb, *) :: b
   complex(8), device, dimension(ldc, *) :: c
   complex(8), device :: alpha, beta ! device or host variable 
 integer(4) function cublasZhemm_v2(h, side, uplo, m, n, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: side, uplo
   integer :: m, n, lda, ldb, ldc
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(ldb, *) :: b
   complex(8), device, dimension(ldc, *) :: c
   complex(8), device :: alpha, beta ! device or host variable

2.5.43. zherk

ZHERK performs one of the hermitian rank k operations C := alpha*A*A**H + beta*C, or C := alpha*A**H*A + beta*C, where alpha and beta are real scalars, C is an n by n hermitian matrix and A is an n by k matrix in the first case and a k by n matrix in the second case. 

 subroutine zherk(uplo, trans, n, k, alpha, a, lda, beta, c, ldc)
   character*1 :: uplo, trans
   integer :: n, k, lda, ldc
   complex(8), device, dimension(lda, *) :: a ! device or host variable
   complex(8), device, dimension(ldc, *) :: c ! device or host variable
   real(8), device :: alpha, beta ! device or host variable 
 subroutine cublasZherk(uplo, trans, n, k, alpha, a, lda, beta, c, ldc)
   character*1 :: uplo, trans
   integer :: n, k, lda, ldc
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(ldc, *) :: c
   real(8), device :: alpha, beta ! device or host variable 
 integer(4) function cublasZherk_v2(h, uplo, trans, n, k, alpha, a, lda, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: uplo, trans
   integer :: n, k, lda, ldc
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(ldc, *) :: c
   real(8), device :: alpha, beta ! device or host variable

2.5.44. zher2k

ZHER2K performs one of the hermitian rank 2k operations C := alpha*A*B**H + conjg( alpha )*B*A**H + beta*C, or C := alpha*A**H*B + conjg( alpha )*B**H*A + beta*C, where alpha and beta are scalars with beta real, C is an n by n hermitian matrix and A and B are n by k matrices in the first case and k by n matrices in the second case. 

 subroutine zher2k(uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   character*1 :: uplo, trans
   integer :: n, k, lda, ldb, ldc
   complex(8), device, dimension(lda, *) :: a ! device or host variable
   complex(8), device, dimension(ldb, *) :: b ! device or host variable
   complex(8), device, dimension(ldc, *) :: c ! device or host variable
   complex(8), device :: alpha ! device or host variable
   real(8), device :: beta ! device or host variable 
 subroutine cublasZher2k(uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   character*1 :: uplo, trans
   integer :: n, k, lda, ldb, ldc
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(ldb, *) :: b
   complex(8), device, dimension(ldc, *) :: c
   complex(8), device :: alpha ! device or host variable
   real(8), device :: beta ! device or host variable 
 integer(4) function cublasZher2k_v2(h, uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: uplo, trans
   integer :: n, k, lda, ldb, ldc
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(ldb, *) :: b
   complex(8), device, dimension(ldc, *) :: c
   complex(8), device :: alpha ! device or host variable
   real(8), device :: beta ! device or host variable

2.5.45. zherkx

ZHERKX performs a variation of the hermitian rank k operations C := alpha*A*B**H + beta*C, where alpha and beta are real scalars, C is an n by n hermitian matrix stored in lower or upper mode, and A and B are n by k matrices. See the CUBLAS documentation for more details. 

 subroutine zherkx(uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   character*1 :: uplo, trans
   integer :: n, k, lda, ldb, ldc
   complex(8), device, dimension(lda, *) :: a ! device or host variable
   complex(8), device, dimension(ldb, *) :: b ! device or host variable
   complex(8), device, dimension(ldc, *) :: c ! device or host variable
   complex(8), device :: alpha ! device or host variable
   real(8), device :: beta ! device or host variable 
 subroutine cublasZherkx(uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   character*1 :: uplo, trans
   integer :: n, k, lda, ldb, ldc
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(ldb, *) :: b
   complex(8), device, dimension(ldc, *) :: c
   complex(8), device :: alpha ! device or host variable
   real(8), device :: beta ! device or host variable 
 integer(4) function cublasZherkx_v2(h, uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: uplo, trans
   integer :: n, k, lda, ldb, ldc
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(ldb, *) :: b
   complex(8), device, dimension(ldc, *) :: c
   complex(8), device :: alpha ! device or host variable
   real(8), device :: beta ! device or host variable

2.5.46. cublasZgetrfBatched

ZGETRF computes an LU factorization of a general M-by-N matrix A using partial pivoting with row interchanges. The factorization has the form A = P * L * U where P is a permutation matrix, L is lower triangular with unit diagonal elements (lower trapezoidal if m > n), and U is upper triangular (upper trapezoidal if m < n). This is the right-looking Level 3 BLAS version of the algorithm. 

 integer(4) function cublasZgetrfBatched(h, n, Aarray, lda, ipvt, info, batchCount)
   type(cublasHandle) :: h
   integer :: n
   type(c_devptr), device :: Aarray(*)
   integer :: lda
   integer, device :: ipvt(*)
   integer, device :: info(*)
   integer :: batchCount

2.5.47. cublasZgetriBatched

ZGETRI computes the inverse of a matrix using the LU factorization computed by ZGETRF. This method inverts U and then computes inv(A) by solving the system inv(A)*L = inv(U) for inv(A). 

 integer(4) function cublasZgetriBatched(h, n, Aarray, lda, ipvt, Carray, ldc, info, batchCount)
   type(cublasHandle) :: h
   integer :: n
   type(c_devptr), device :: Aarray(*)
   integer :: lda
   integer, device :: ipvt(*)
   type(c_devptr), device :: Carray(*)
   integer :: ldc
   integer, device :: info(*)
   integer :: batchCount

2.5.48. cublasZgetrsBatched

ZGETRS solves a system of linear equations A * X = B, A**T * X = B, or A**H * X = B with a general N-by-N matrix A using the LU factorization computed by ZGETRF. 

 integer(4) function cublasZgetrsBatched(h, trans, n, nrhs, Aarray, lda, ipvt, Barray, ldb, info, batchCount)
   type(cublasHandle) :: h
   integer :: trans ! integer or character(1) variable
   integer :: n, nrhs
   type(c_devptr), device :: Aarray(*)
   integer :: lda
   integer, device :: ipvt(*)
   type(c_devptr), device :: Barray(*)
   integer :: ldb
   integer :: info(*)
   integer :: batchCount

2.5.49. cublasZgemmBatched

ZGEMM performs one of the matrix-matrix operations C := alpha*op( A )*op( B ) + beta*C, where op( X ) is one of op( X ) = X or op( X ) = X**T or op( X ) = X**H, alpha and beta are scalars, and A, B and C are matrices, with op( A ) an m by k matrix, op( B ) a k by n matrix and C an m by n matrix. 

 integer(4) function cublasZgemmBatched(h, transa, transb, m, n, k, alpha, Aarray, lda, Barray, ldb, beta, Carray, ldc, batchCount)
   type(cublasHandle) :: h
   integer :: transa ! integer or character(1) variable
   integer :: transb ! integer or character(1) variable
   integer :: m, n, k
   complex(8), device :: alpha ! device or host variable
   type(c_devptr), device :: Aarray(*)
   integer :: lda
   type(c_devptr), device :: Barray(*)
   integer :: ldb
   complex(8), device :: beta ! device or host variable
   type(c_devptr), device :: Carray(*)
   integer :: ldc
   integer :: batchCount 
 integer(4) function cublasZgemmBatched_v2(h, transa, transb, m, n, k, alpha,           Aarray, lda, Barray, ldb, beta, Carray, ldc, batchCount)
   type(cublasHandle) :: h
   integer :: transa
   integer :: transb
   integer :: m, n, k
   complex(8), device :: alpha ! device or host variable
   type(c_devptr), device :: Aarray(*)
   integer :: lda
   type(c_devptr), device :: Barray(*)
   integer :: ldb
   complex(8), device :: beta ! device or host variable
   type(c_devptr), device :: Carray(*)
   integer :: ldc
   integer :: batchCount

2.5.50. cublasZtrsmBatched

ZTRSM solves one of the matrix equations op( A )*X = alpha*B, or X*op( A ) = alpha*B, where alpha is a scalar, X and B are m by n matrices, A is a unit, or non-unit, upper or lower triangular matrix and op( A ) is one of op( A ) = A or op( A ) = A**T or op( A ) = A**H. The matrix X is overwritten on B. 

 integer(4) function cublasZtrsmBatched( h, side, uplo, trans, diag, m, n, alpha, A, lda, B, ldb, batchCount)
   type(cublasHandle) :: h
   integer :: side ! integer or character(1) variable
   integer :: uplo ! integer or character(1) variable
   integer :: trans ! integer or character(1) variable
   integer :: diag ! integer or character(1) variable
   integer :: m, n
   complex(8), device :: alpha ! device or host variable
   type(c_devptr), device :: A(*)
   integer :: lda
   type(c_devptr), device :: B(*)
   integer :: ldb
   integer :: batchCount 
 integer(4) function cublasZtrsmBatched_v2( h, side, uplo, trans, diag, m, n, alpha, A, lda, B, ldb, batchCount)
   type(cublasHandle) :: h
   integer :: side
   integer :: uplo
   integer :: trans
   integer :: diag
   integer :: m, n
   complex(8), device :: alpha ! device or host variable
   type(c_devptr), device :: A(*)
   integer :: lda
   type(c_devptr), device :: B(*)
   integer :: ldb
   integer :: batchCount

2.5.51. cublasZmatinvBatched

cublasZmatinvBatched is a short cut of cublasZgetrfBatched plus cublasZgetriBatched. However it only works if n is less than 32. If not, the user has to go through cublasZgetrfBatched and cublasZgetriBatched. 

 integer(4) function cublasZmatinvBatched(h, n, Aarray, lda, Ainv, lda_inv, info, batchCount)
   type(cublasHandle) :: h
   integer :: n
   type(c_devptr), device :: Aarray(*)
   integer :: lda
   type(c_devptr), device :: Ainv(*)
   integer :: lda_inv
   integer, device :: info(*)
   integer :: batchCount

2.5.52. cublasZgeqrfBatched

ZGEQRF computes a QR factorization of a complex M-by-N matrix A: A = Q * R. 

 integer(4) function cublasZgeqrfBatched(h, m, n, Aarray, lda, Tau, info, batchCount)
   type(cublasHandle) :: h
   integer :: m, n
   type(c_devptr), device :: Aarray(*)
   integer :: lda
   type(c_devptr), device :: Tau(*)
   integer :: info(*)
   integer :: batchCount

2.5.53. cublasZgelsBatched

ZGELS solves overdetermined or underdetermined complex linear systems involving an M-by-N matrix A, or its conjugate-transpose, using a QR or LQ factorization of A. It is assumed that A has full rank. The following options are provided: 1. If TRANS = 'N' and m >= n: find the least squares solution of an overdetermined system, i.e., solve the least squares problem minimize || B - A*X ||. 2. If TRANS = 'N' and m < n: find the minimum norm solution of an underdetermined system A * X = B. 3. If TRANS = 'C' and m >= n: find the minimum norm solution of an undetermined system A**H * X = B. 4. If TRANS = 'C' and m < n: find the least squares solution of an overdetermined system, i.e., solve the least squares problem minimize || B - A**H * X ||. Several right hand side vectors b and solution vectors x can be handled in a single call; they are stored as the columns of the M-by-NRHS right hand side matrix B and the N-by-NRHS solution matrix X. 

 integer(4) function cublasZgelsBatched(h, trans, m, n, nrhs, Aarray, lda, Carray, ldc, info, devinfo, batchCount)
   type(cublasHandle) :: h
   integer :: trans ! integer or character(1) variable
   integer :: m, n, nrhs
   type(c_devptr), device :: Aarray(*)
   integer :: lda
   type(c_devptr), device :: Carray(*)
   integer :: ldc
   integer :: info(*)
   integer, device :: devinfo(*)
   integer :: batchCount

2.5.54. cublasZgemmStridedBatched

ZGEMM performs one of the matrix-matrix operations C := alpha*op( A )*op( B ) + beta*C, where op( X ) is one of op( X ) = X or op( X ) = X**T, alpha and beta are scalars, and A, B and C are matrices, with op( A ) an m by k matrix, op( B ) a k by n matrix and C an m by n matrix. 

 integer(4) function cublasZgemmStridedBatched(h, transa, transb, m, n, k, alpha, Aarray, lda, strideA, Barray, ldb, strideB, beta, Carray, ldc, strideC, batchCount)
   type(cublasHandle) :: h
   integer :: transa ! integer or character(1) variable
   integer :: transb ! integer or character(1) variable
   integer :: m, n, k
   complex(8), device :: alpha ! device or host variable
   complex(8), device :: Aarray(*)
   integer :: lda
   integer :: strideA
   complex(8), device :: Barray(*)
   integer :: ldb
   integer :: strideB
   complex(8), device :: beta ! device or host variable
   complex(8), device :: Carray(*)
   integer :: ldc
   integer :: strideC
   integer :: batchCount 
 integer(4) function cublasZgemmStridedBatched_v2(h, transa, transb, m, n, k, alpha,           Aarray, lda, strideA, Barray, ldb, strideB, beta, Carray, ldc, strideC, batchCount)
   type(cublasHandle) :: h
   integer :: transa
   integer :: transb
   integer :: m, n, k
   complex(8), device :: alpha ! device or host variable
   complex(8), device :: Aarray(*)
   integer :: lda
   integer :: strideA
   complex(8), device :: Barray(*)
   integer :: ldb
   integer :: strideB
   complex(8), device :: beta ! device or host variable
   complex(8), device :: Carray(*)
   integer :: ldc
   integer :: strideC
   integer :: batchCount

2.6. CUBLAS V2 Module Functions

This section contains interfaces to the cuBLAS V2 Module Functions. Users can access this module by inserting the line use cublas_v2 into the program unit. One major difference in the cublas_v2 versus the cublas module is the cublas entry points, such as cublasIsamax are changed to take the handle as the first argument. The second difference in the cublas_v2 module is the v2 entry points, such as cublasIsamax_v2 do not implicitly handle the pointer modes for the user. It is up to the programmer to make calls to cublasSetPointerMode to tell the library if scalar arguments reside on the host or device. The actual interfaces to the v2 entry points do not change, and are not listed in this section.

2.6.1. Single Precision Functions and Subroutines

This section contains the V2 interfaces to the single precision BLAS and cuBLAS functions and subroutines.

2.6.1.1. isamax

If you use the cublas_v2 module, the interface for cublasIsamax is changed to the following: 

 integer(4) function cublasIsamax(h, n, x, incx, res)
   type(cublasHandle) :: h
   integer :: n
   real(4), device, dimension(*) :: x
   integer :: incx
   integer, device :: res ! device or host variable

2.6.1.2. isamin

If you use the cublas_v2 module, the interface for cublasIsamin is changed to the following: 

 integer(4) function cublasIsamin(h, n, x, incx, res)
   type(cublasHandle) :: h
   integer :: n
   real(4), device, dimension(*) :: x
   integer :: incx
   integer, device :: res ! device or host variable

2.6.1.3. sasum

If you use the cublas_v2 module, the interface for cublasSasum is changed to the following: 

 integer(4) function cublasSasum(h, n, x, incx, res)
   type(cublasHandle) :: h
   integer :: n
   real(4), device, dimension(*) :: x
   integer :: incx
   real(4), device :: res ! device or host variable

2.6.1.4. saxpy

If you use the cublas_v2 module, the interface for cublasSaxpy is changed to the following: 

 integer(4) function cublasSaxpy(h, n, a, x, incx, y, incy)
   type(cublasHandle) :: h
   integer :: n
   real(4), device :: a ! device or host variable
   real(4), device, dimension(*) :: x, y
   integer :: incx, incy

2.6.1.5. scopy

If you use the cublas_v2 module, the interface for cublasScopy is changed to the following: 

 integer(4) function cublasScopy(h, n, x, incx, y, incy)
   type(cublasHandle) :: h
   integer :: n
   real(4), device, dimension(*) :: x, y
   integer :: incx, incy

2.6.1.6. sdot

If you use the cublas_v2 module, the interface for cublasSdot is changed to the following: 

 integer(4) function cublasSdot(h, n, x, incx, y, incy, res)
   type(cublasHandle) :: h
   integer :: n
   real(4), device, dimension(*) :: x, y
   integer :: incx, incy
   real(4), device :: res ! device or host variable

2.6.1.7. snrm2

If you use the cublas_v2 module, the interface for cublasSnrm2 is changed to the following: 

 integer(4) function cublasSnrm2(h, n, x, incx, res)
   type(cublasHandle) :: h
   integer :: n
   real(4), device, dimension(*) :: x
   integer :: incx
   real(4), device :: res ! device or host variable

2.6.1.8. srot

If you use the cublas_v2 module, the interface for cublasSrot is changed to the following: 

 integer(4) function cublasSrot(h, n, x, incx, y, incy, sc, ss)
   type(cublasHandle) :: h
   integer :: n
   real(4), device :: sc, ss ! device or host variable
   real(4), device, dimension(*) :: x, y
   integer :: incx, incy

2.6.1.9. srotg

If you use the cublas_v2 module, the interface for cublasSrotg is changed to the following: 

 integer(4) function cublasSrotg(h, sa, sb, sc, ss)
   type(cublasHandle) :: h
   real(4), device :: sa, sb, sc, ss ! device or host variable

2.6.1.10. srotm

If you use the cublas_v2 module, the interface for cublasSrotm is changed to the following: 

 integer(4) function cublasSrotm(h, n, x, incx, y, incy, param)
   type(cublasHandle) :: h
   integer :: n
   real(4), device :: param(*) ! device or host variable
   real(4), device, dimension(*) :: x, y
   integer :: incx, incy

2.6.1.11. srotmg

If you use the cublas_v2 module, the interface for cublasSrotmg is changed to the following: 

 integer(4) function cublasSrotmg(h, d1, d2, x1, y1, param)
   type(cublasHandle) :: h
   real(4), device :: d1, d2, x1, y1, param(*) ! device or host variable

2.6.1.12. sscal

If you use the cublas_v2 module, the interface for cublasSscal is changed to the following: 

 integer(4) function cublasSscal(h, n, a, x, incx)
   type(cublasHandle) :: h
   integer :: n
   real(4), device :: a ! device or host variable
   real(4), device, dimension(*) :: x
   integer :: incx

2.6.1.13. sswap

If you use the cublas_v2 module, the interface for cublasSswap is changed to the following: 

 integer(4) function cublasSswap(h, n, x, incx, y, incy)
   type(cublasHandle) :: h
   integer :: n
   real(4), device, dimension(*) :: x, y
   integer :: incx, incy

2.6.1.14. sgbmv

If you use the cublas_v2 module, the interface for cublasSgbmv is changed to the following: 

 integer(4) function cublasSgbmv(h, t, m, n, kl, ku, alpha, a, lda, x, incx, beta, y, incy)
   type(cublasHandle) :: h
   integer :: t
   integer :: m, n, kl, ku, lda, incx, incy
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(*) :: x, y
   real(4), device :: alpha, beta ! device or host variable

2.6.1.15. sgemv

If you use the cublas_v2 module, the interface for cublasSgemv is changed to the following: 

 integer(4) function cublasSgemv(h, t, m, n, alpha, a, lda, x, incx, beta, y, incy)
   type(cublasHandle) :: h
   integer :: t
   integer :: m, n, lda, incx, incy
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(*) :: x, y
   real(4), device :: alpha, beta ! device or host variable

2.6.1.16. sger

If you use the cublas_v2 module, the interface for cublasSger is changed to the following: 

 integer(4) function cublasSger(h, m, n, alpha, x, incx, y, incy, a, lda)
   type(cublasHandle) :: h
   integer :: m, n, lda, incx, incy
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(*) :: x, y
   real(4), device :: alpha ! device or host variable

2.6.1.17. ssbmv

If you use the cublas_v2 module, the interface for cublasSsbmv is changed to the following: 

 integer(4) function cublasSsbmv(h, t, n, k, alpha, a, lda, x, incx, beta, y, incy)
   type(cublasHandle) :: h
   integer :: t
   integer :: k, n, lda, incx, incy
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(*) :: x, y
   real(4), device :: alpha, beta ! device or host variable

2.6.1.18. sspmv

If you use the cublas_v2 module, the interface for cublasSspmv is changed to the following: 

 integer(4) function cublasSspmv(h, t, n, alpha, a, x, incx, beta, y, incy)
   type(cublasHandle) :: h
   integer :: t
   integer :: n, incx, incy
   real(4), device, dimension(*) :: a, x, y
   real(4), device :: alpha, beta ! device or host variable

2.6.1.19. sspr

If you use the cublas_v2 module, the interface for cublasSspr is changed to the following: 

 integer(4) function cublasSspr(h, t, n, alpha, x, incx, a)
   type(cublasHandle) :: h
   integer :: t
   integer :: n, incx
   real(4), device, dimension(*) :: a, x
   real(4), device :: alpha ! device or host variable

2.6.1.20. sspr2

If you use the cublas_v2 module, the interface for cublasSspr2 is changed to the following: 

 integer(4) function cublasSspr2(h, t, n, alpha, x, incx, y, incy, a)
   type(cublasHandle) :: h
   integer :: t
   integer :: n, incx, incy
   real(4), device, dimension(*) :: a, x, y
   real(4), device :: alpha ! device or host variable

2.6.1.21. ssymv

If you use the cublas_v2 module, the interface for cublasSsymv is changed to the following: 

 integer(4) function cublasSsymv(h, uplo, n, alpha, a, lda, x, incx, beta, y, incy)
   type(cublasHandle) :: h
   integer :: uplo
   integer :: n, lda, incx, incy
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(*) :: x, y
   real(4), device :: alpha, beta ! device or host variable

2.6.1.22. ssyr

If you use the cublas_v2 module, the interface for cublasSsyr is changed to the following: 

 integer(4) function cublasSsyr(h, t, n, alpha, x, incx, a, lda)
   type(cublasHandle) :: h
   integer :: t
   integer :: n, incx, lda
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(*) :: x
   real(4), device :: alpha ! device or host variable

2.6.1.23. ssyr2

If you use the cublas_v2 module, the interface for cublasSsyr2 is changed to the following: 

 integer(4) function cublasSsyr2(h, t, n, alpha, x, incx, y, incy, a, lda)
   type(cublasHandle) :: h
   integer :: t
   integer :: n, incx, incy, lda
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(*) :: x, y
   real(4), device :: alpha ! device or host variable

2.6.1.24. stbmv

If you use the cublas_v2 module, the interface for cublasStbmv is changed to the following: 

 integer(4) function cublasStbmv(h, u, t, d, n, k, a, lda, x, incx)
   type(cublasHandle) :: h
   integer :: u, t, d
   integer :: n, k, incx, lda
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(*) :: x

2.6.1.25. stbsv

If you use the cublas_v2 module, the interface for cublasStbsv is changed to the following: 

 integer(4) function cublasStbsv(h, u, t, d, n, k, a, lda, x, incx)
   type(cublasHandle) :: h
   integer :: u, t, d
   integer :: n, k, incx, lda
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(*) :: x

2.6.1.26. stpmv

If you use the cublas_v2 module, the interface for cublasStpmv is changed to the following: 

 integer(4) function cublasStpmv(h, u, t, d, n, a, x, incx)
   type(cublasHandle) :: h
   integer :: u, t, d
   integer :: n, incx
   real(4), device, dimension(*) :: a, x

2.6.1.27. stpsv

If you use the cublas_v2 module, the interface for cublasStpsv is changed to the following: 

 integer(4) function cublasStpsv(h, u, t, d, n, a, x, incx)
   type(cublasHandle) :: h
   integer :: u, t, d
   integer :: n, incx
   real(4), device, dimension(*) :: a, x

2.6.1.28. strmv

If you use the cublas_v2 module, the interface for cublasStrmv is changed to the following: 

 integer(4) function cublasStrmv(h, u, t, d, n, a, lda, x, incx)
   type(cublasHandle) :: h
   integer :: u, t, d
   integer :: n, incx, lda
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(*) :: x

2.6.1.29. strsv

If you use the cublas_v2 module, the interface for cublasStrsv is changed to the following: 

 integer(4) function cublasStrsv(h, u, t, d, n, a, lda, x, incx)
   type(cublasHandle) :: h
   integer :: u, t, d
   integer :: n, incx, lda
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(*) :: x

2.6.1.30. sgemm

If you use the cublas_v2 module, the interface for cublasSgemm is changed to the following: 

 integer(4) function cublasSgemm(h, transa, transb, m, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: transa, transb
   integer :: m, n, k, lda, ldb, ldc
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(ldb, *) :: b
   real(4), device, dimension(ldc, *) :: c
   real(4), device :: alpha, beta ! device or host variable

2.6.1.31. ssymm

If you use the cublas_v2 module, the interface for cublasSsymm is changed to the following: 

 integer(4) function cublasSsymm(h, side, uplo, m, n, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: side, uplo
   integer :: m, n, lda, ldb, ldc
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(ldb, *) :: b
   real(4), device, dimension(ldc, *) :: c
   real(4), device :: alpha, beta ! device or host variable

2.6.1.32. ssyrk

If you use the cublas_v2 module, the interface for cublasSsyrk is changed to the following: 

 integer(4) function cublasSsyrk(h, uplo, trans, n, k, alpha, a, lda, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: uplo, trans
   integer :: n, k, lda, ldc
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(ldc, *) :: c
   real(4), device :: alpha, beta ! device or host variable

2.6.1.33. ssyr2k

If you use the cublas_v2 module, the interface for cublasSsyr2k is changed to the following: 

 integer(4) function cublasSsyr2k(h, uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: uplo, trans
   integer :: n, k, lda, ldb, ldc
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(ldb, *) :: b
   real(4), device, dimension(ldc, *) :: c
   real(4), device :: alpha, beta ! device or host variable

2.6.1.34. ssyrkx

If you use the cublas_v2 module, the interface for cublasSsyrkx is changed to the following: 

 integer(4) function cublasSsyrkx(h, uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: uplo, trans
   integer :: n, k, lda, ldb, ldc
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(ldb, *) :: b
   real(4), device, dimension(ldc, *) :: c
   real(4), device :: alpha, beta ! device or host variable

2.6.1.35. strmm

If you use the cublas_v2 module, the interface for cublasStrmm is changed to the following: 

 integer(4) function cublasStrmm(h, side, uplo, transa, diag, m, n, alpha, a, lda, b, ldb, c, ldc)
   type(cublasHandle) :: h
   integer :: side, uplo, transa, diag
   integer :: m, n, lda, ldb, ldc
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(ldb, *) :: b
   real(4), device, dimension(ldc, *) :: c
   real(4), device :: alpha ! device or host variable

2.6.1.36. strsm

If you use the cublas_v2 module, the interface for cublasStrsm is changed to the following: 

 integer(4) function cublasStrsm(h, side, uplo, transa, diag, m, n, alpha, a, lda, b, ldb)
   type(cublasHandle) :: h
   integer :: side, uplo, transa, diag
   integer :: m, n, lda, ldb
   real(4), device, dimension(lda, *) :: a
   real(4), device, dimension(ldb, *) :: b
   real(4), device :: alpha ! device or host variable

2.6.2. Double Precision Functions and Subroutines

This section contains the V2 interfaces to the double precision BLAS and cuBLAS functions and subroutines.

2.6.2.1. idamax

If you use the cublas_v2 module, the interface for cublasIdamax is changed to the following: 

 integer(4) function cublasIdamax(h, n, x, incx, res)
   type(cublasHandle) :: h
   integer :: n
   real(8), device, dimension(*) :: x
   integer :: incx
   integer, device :: res ! device or host variable

2.6.2.2. idamin

If you use the cublas_v2 module, the interface for cublasIdamin is changed to the following: 

 integer(4) function cublasIdamin(h, n, x, incx, res)
   type(cublasHandle) :: h
   integer :: n
   real(8), device, dimension(*) :: x
   integer :: incx
   integer, device :: res ! device or host variable

2.6.2.3. dasum

If you use the cublas_v2 module, the interface for cublasDasum is changed to the following: 

 integer(4) function cublasDasum(h, n, x, incx, res)
   type(cublasHandle) :: h
   integer :: n
   real(8), device, dimension(*) :: x
   integer :: incx
   real(8), device :: res ! device or host variable

2.6.2.4. daxpy

If you use the cublas_v2 module, the interface for cublasDaxpy is changed to the following: 

 integer(4) function cublasDaxpy(h, n, a, x, incx, y, incy)
   type(cublasHandle) :: h
   integer :: n
   real(8), device :: a ! device or host variable
   real(8), device, dimension(*) :: x, y
   integer :: incx, incy

2.6.2.5. dcopy

If you use the cublas_v2 module, the interface for cublasDcopy is changed to the following: 

 integer(4) function cublasDcopy(h, n, x, incx, y, incy)
   type(cublasHandle) :: h
   integer :: n
   real(8), device, dimension(*) :: x, y
   integer :: incx, incy

2.6.2.6. ddot

If you use the cublas_v2 module, the interface for cublasDdot is changed to the following: 

 integer(4) function cublasDdot(h, n, x, incx, y, incy, res)
   type(cublasHandle) :: h
   integer :: n
   real(8), device, dimension(*) :: x, y
   integer :: incx, incy
   real(8), device :: res ! device or host variable

2.6.2.7. dnrm2

If you use the cublas_v2 module, the interface for cublasDnrm2 is changed to the following: 

 integer(4) function cublasDnrm2(h, n, x, incx, res)
   type(cublasHandle) :: h
   integer :: n
   real(8), device, dimension(*) :: x
   integer :: incx
   real(8), device :: res ! device or host variable

2.6.2.8. drot

If you use the cublas_v2 module, the interface for cublasDrot is changed to the following: 

 integer(4) function cublasDrot(h, n, x, incx, y, incy, sc, ss)
   type(cublasHandle) :: h
   integer :: n
   real(8), device :: sc, ss ! device or host variable
   real(8), device, dimension(*) :: x, y
   integer :: incx, incy

2.6.2.9. drotg

If you use the cublas_v2 module, the interface for cublasDrotg is changed to the following: 

 integer(4) function cublasDrotg(h, sa, sb, sc, ss)
   type(cublasHandle) :: h
   real(8), device :: sa, sb, sc, ss ! device or host variable

2.6.2.10. drotm

If you use the cublas_v2 module, the interface for cublasDrotm is changed to the following: 

 integer(4) function cublasDrotm(h, n, x, incx, y, incy, param)
   type(cublasHandle) :: h
   integer :: n
   real(8), device :: param(*) ! device or host variable
   real(8), device, dimension(*) :: x, y
   integer :: incx, incy

2.6.2.11. drotmg

If you use the cublas_v2 module, the interface for cublasDrotmg is changed to the following: 

 integer(4) function cublasDrotmg(h, d1, d2, x1, y1, param)
   type(cublasHandle) :: h
   real(8), device :: d1, d2, x1, y1, param(*) ! device or host variable

2.6.2.12. dscal

If you use the cublas_v2 module, the interface for cublasDscal is changed to the following: 

 integer(4) function cublasDscal(h, n, a, x, incx)
   type(cublasHandle) :: h
   integer :: n
   real(8), device :: a ! device or host variable
   real(8), device, dimension(*) :: x
   integer :: incx

2.6.2.13. dswap

If you use the cublas_v2 module, the interface for cublasDswap is changed to the following: 

 integer(4) function cublasDswap(h, n, x, incx, y, incy)
   type(cublasHandle) :: h
   integer :: n
   real(8), device, dimension(*) :: x, y
   integer :: incx, incy

2.6.2.14. dgbmv

If you use the cublas_v2 module, the interface for cublasDgbmv is changed to the following: 

 integer(4) function cublasDgbmv(h, t, m, n, kl, ku, alpha, a, lda, x, incx, beta, y, incy)
   type(cublasHandle) :: h
   integer :: t
   integer :: m, n, kl, ku, lda, incx, incy
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(*) :: x, y
   real(8), device :: alpha, beta ! device or host variable

2.6.2.15. dgemv

If you use the cublas_v2 module, the interface for cublasDgemv is changed to the following: 

 integer(4) function cublasDgemv(h, t, m, n, alpha, a, lda, x, incx, beta, y, incy)
   type(cublasHandle) :: h
   integer :: t
   integer :: m, n, lda, incx, incy
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(*) :: x, y
   real(8), device :: alpha, beta ! device or host variable

2.6.2.16. dger

If you use the cublas_v2 module, the interface for cublasDger is changed to the following: 

 integer(4) function cublasDger(h, m, n, alpha, x, incx, y, incy, a, lda)
   type(cublasHandle) :: h
   integer :: m, n, lda, incx, incy
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(*) :: x, y
   real(8), device :: alpha ! device or host variable

2.6.2.17. dsbmv

If you use the cublas_v2 module, the interface for cublasDsbmv is changed to the following: 

 integer(4) function cublasDsbmv(h, t, n, k, alpha, a, lda, x, incx, beta, y, incy)
   type(cublasHandle) :: h
   integer :: t
   integer :: k, n, lda, incx, incy
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(*) :: x, y
   real(8), device :: alpha, beta ! device or host variable

2.6.2.18. dspmv

If you use the cublas_v2 module, the interface for cublasDspmv is changed to the following: 

 integer(4) function cublasDspmv(h, t, n, alpha, a, x, incx, beta, y, incy)
   type(cublasHandle) :: h
   integer :: t
   integer :: n, incx, incy
   real(8), device, dimension(*) :: a, x, y
   real(8), device :: alpha, beta ! device or host variable

2.6.2.19. dspr

If you use the cublas_v2 module, the interface for cublasDspr is changed to the following: 

 integer(4) function cublasDspr(h, t, n, alpha, x, incx, a)
   type(cublasHandle) :: h
   integer :: t
   integer :: n, incx
   real(8), device, dimension(*) :: a, x
   real(8), device :: alpha ! device or host variable

2.6.2.20. dspr2

If you use the cublas_v2 module, the interface for cublasDspr2 is changed to the following: 

 integer(4) function cublasDspr2(h, t, n, alpha, x, incx, y, incy, a)
   type(cublasHandle) :: h
   integer :: t
   integer :: n, incx, incy
   real(8), device, dimension(*) :: a, x, y
   real(8), device :: alpha ! device or host variable

2.6.2.21. dsymv

If you use the cublas_v2 module, the interface for cublasDsymv is changed to the following: 

 integer(4) function cublasDsymv(h, uplo, n, alpha, a, lda, x, incx, beta, y, incy)
   type(cublasHandle) :: h
   integer :: uplo
   integer :: n, lda, incx, incy
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(*) :: x, y
   real(8), device :: alpha, beta ! device or host variable

2.6.2.22. dsyr

If you use the cublas_v2 module, the interface for cublasDsyr is changed to the following: 

 integer(4) function cublasDsyr(h, t, n, alpha, x, incx, a, lda)
   type(cublasHandle) :: h
   integer :: t
   integer :: n, incx, lda
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(*) :: x
   real(8), device :: alpha ! device or host variable

2.6.2.23. dsyr2

If you use the cublas_v2 module, the interface for cublasDsyr2 is changed to the following: 

 integer(4) function cublasDsyr2(h, t, n, alpha, x, incx, y, incy, a, lda)
   type(cublasHandle) :: h
   integer :: t
   integer :: n, incx, incy, lda
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(*) :: x, y
   real(8), device :: alpha ! device or host variable

2.6.2.24. dtbmv

If you use the cublas_v2 module, the interface for cublasDtbmv is changed to the following: 

 integer(4) function cublasDtbmv(h, u, t, d, n, k, a, lda, x, incx)
   type(cublasHandle) :: h
   integer :: u, t, d
   integer :: n, k, incx, lda
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(*) :: x

2.6.2.25. dtbsv

If you use the cublas_v2 module, the interface for cublasDtbsv is changed to the following: 

 integer(4) function cublasDtbsv(h, u, t, d, n, k, a, lda, x, incx)
   type(cublasHandle) :: h
   integer :: u, t, d
   integer :: n, k, incx, lda
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(*) :: x

2.6.2.26. dtpmv

If you use the cublas_v2 module, the interface for cublasDtpmv is changed to the following: 

 integer(4) function cublasDtpmv(h, u, t, d, n, a, x, incx)
   type(cublasHandle) :: h
   integer :: u, t, d
   integer :: n, incx
   real(8), device, dimension(*) :: a, x

2.6.2.27. dtpsv

If you use the cublas_v2 module, the interface for cublasDtpsv is changed to the following: 

 integer(4) function cublasDtpsv(h, u, t, d, n, a, x, incx)
   type(cublasHandle) :: h
   integer :: u, t, d
   integer :: n, incx
   real(8), device, dimension(*) :: a, x

2.6.2.28. dtrmv

If you use the cublas_v2 module, the interface for cublasDtrmv is changed to the following: 

 integer(4) function cublasDtrmv(h, u, t, d, n, a, lda, x, incx)
   type(cublasHandle) :: h
   integer :: u, t, d
   integer :: n, incx, lda
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(*) :: x

2.6.2.29. dtrsv

If you use the cublas_v2 module, the interface for cublasDtrsv is changed to the following: 

 integer(4) function cublasDtrsv(h, u, t, d, n, a, lda, x, incx)
   type(cublasHandle) :: h
   integer :: u, t, d
   integer :: n, incx, lda
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(*) :: x

2.6.2.30. dgemm

If you use the cublas_v2 module, the interface for cublasDgemm is changed to the following: 

 integer(4) function cublasDgemm(h, transa, transb, m, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: transa, transb
   integer :: m, n, k, lda, ldb, ldc
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(ldb, *) :: b
   real(8), device, dimension(ldc, *) :: c
   real(8), device :: alpha, beta ! device or host variable

2.6.2.31. dsymm

If you use the cublas_v2 module, the interface for cublasDsymm is changed to the following: 

 integer(4) function cublasDsymm(h, side, uplo, m, n, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: side, uplo
   integer :: m, n, lda, ldb, ldc
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(ldb, *) :: b
   real(8), device, dimension(ldc, *) :: c
   real(8), device :: alpha, beta ! device or host variable

2.6.2.32. dsyrk

If you use the cublas_v2 module, the interface for cublasDsyrk is changed to the following: 

 integer(4) function cublasDsyrk(h, uplo, trans, n, k, alpha, a, lda, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: uplo, trans
   integer :: n, k, lda, ldc
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(ldc, *) :: c
   real(8), device :: alpha, beta ! device or host variable

2.6.2.33. dsyr2k

If you use the cublas_v2 module, the interface for cublasDsyr2k is changed to the following: 

 integer(4) function cublasDsyr2k(h, uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: uplo, trans
   integer :: n, k, lda, ldb, ldc
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(ldb, *) :: b
   real(8), device, dimension(ldc, *) :: c
   real(8), device :: alpha, beta ! device or host variable

2.6.2.34. dsyrkx

If you use the cublas_v2 module, the interface for cublasDsyrkx is changed to the following: 

 integer(4) function cublasDsyrkx(h, uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: uplo, trans
   integer :: n, k, lda, ldb, ldc
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(ldb, *) :: b
   real(8), device, dimension(ldc, *) :: c
   real(8), device :: alpha, beta ! device or host variable

2.6.2.35. dtrmm

If you use the cublas_v2 module, the interface for cublasDtrmm is changed to the following: 

 integer(4) function cublasDtrmm(h, side, uplo, transa, diag, m, n, alpha, a, lda, b, ldb, c, ldc)
   type(cublasHandle) :: h
   integer :: side, uplo, transa, diag
   integer :: m, n, lda, ldb, ldc
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(ldb, *) :: b
   real(8), device, dimension(ldc, *) :: c
   real(8), device :: alpha ! device or host variable

2.6.2.36. dtrsm

If you use the cublas_v2 module, the interface for cublasDtrsm is changed to the following: 

 integer(4) function cublasDtrsm(h, side, uplo, transa, diag, m, n, alpha, a, lda, b, ldb)
   type(cublasHandle) :: h
   integer :: side, uplo, transa, diag
   integer :: m, n, lda, ldb
   real(8), device, dimension(lda, *) :: a
   real(8), device, dimension(ldb, *) :: b
   real(8), device :: alpha ! device or host variable

2.6.3. Single Precision Complex Functions and Subroutines

This section contains the V2 interfaces to the single precision complex BLAS and cuBLAS functions and subroutines.

2.6.3.1. icamax

If you use the cublas_v2 module, the interface for cublasIcamax is changed to the following: 

 integer(4) function cublasIcamax(h, n, x, incx, res)
   type(cublasHandle) :: h
   integer :: n
   complex(4), device, dimension(*) :: x
   integer :: incx
   integer, device :: res ! device or host variable

2.6.3.2. icamin

If you use the cublas_v2 module, the interface for cublasIcamin is changed to the following: 

 integer(4) function cublasIcamin(h, n, x, incx, res)
   type(cublasHandle) :: h
   integer :: n
   complex(4), device, dimension(*) :: x
   integer :: incx
   integer, device :: res ! device or host variable

2.6.3.3. scasum

If you use the cublas_v2 module, the interface for cublasScasum is changed to the following: 

 integer(4) function cublasScasum(h, n, x, incx, res)
   type(cublasHandle) :: h
   integer :: n
   complex(4), device, dimension(*) :: x
   integer :: incx
   real(4), device :: res ! device or host variable

2.6.3.4. caxpy

If you use the cublas_v2 module, the interface for cublasCaxpy is changed to the following: 

 integer(4) function cublasCaxpy(h, n, a, x, incx, y, incy)
   type(cublasHandle) :: h
   integer :: n
   complex(4), device :: a ! device or host variable
   complex(4), device, dimension(*) :: x, y
   integer :: incx, incy

2.6.3.5. ccopy

If you use the cublas_v2 module, the interface for cublasCcopy is changed to the following: 

 integer(4) function cublasCcopy(h, n, x, incx, y, incy)
   type(cublasHandle) :: h
   integer :: n
   complex(4), device, dimension(*) :: x, y
   integer :: incx, incy

2.6.3.6. cdotc

If you use the cublas_v2 module, the interface for cublasCdotc is changed to the following: 

 integer(4) function cublasCdotc(h, n, x, incx, y, incy, res)
   type(cublasHandle) :: h
   integer :: n
   complex(4), device, dimension(*) :: x, y
   integer :: incx, incy
   complex(4), device :: res ! device or host variable

2.6.3.7. cdotu

If you use the cublas_v2 module, the interface for cublasCdotu is changed to the following: 

 integer(4) function cublasCdotu(h, n, x, incx, y, incy, res)
   type(cublasHandle) :: h
   integer :: n
   complex(4), device, dimension(*) :: x, y
   integer :: incx, incy
   complex(4), device :: res ! device or host variable

2.6.3.8. scnrm2

If you use the cublas_v2 module, the interface for cublasScnrm2 is changed to the following: 

 integer(4) function cublasScnrm2(h, n, x, incx, res)
   type(cublasHandle) :: h
   integer :: n
   complex(4), device, dimension(*) :: x
   integer :: incx
   real(4), device :: res ! device or host variable

2.6.3.9. crot

If you use the cublas_v2 module, the interface for cublasCrot is changed to the following: 

 integer(4) function cublasCrot(h, n, x, incx, y, incy, sc, ss)
   type(cublasHandle) :: h
   integer :: n
   real(4), device :: sc ! device or host variable
   complex(4), device :: ss ! device or host variable
   complex(4), device, dimension(*) :: x, y
   integer :: incx, incy

2.6.3.10. csrot

If you use the cublas_v2 module, the interface for cublasCsrot is changed to the following: 

 integer(4) function cublasCsrot(h, n, x, incx, y, incy, sc, ss)
   type(cublasHandle) :: h
   integer :: n
   real(4), device :: sc, ss ! device or host variable
   complex(4), device, dimension(*) :: x, y
   integer :: incx, incy

2.6.3.11. crotg

If you use the cublas_v2 module, the interface for cublasCrotg is changed to the following: 

 integer(4) function cublasCrotg(h, sa, sb, sc, ss)
   type(cublasHandle) :: h
   complex(4), device :: sa, sb, ss ! device or host variable
   real(4), device :: sc ! device or host variable

2.6.3.12. cscal

If you use the cublas_v2 module, the interface for cublasCscal is changed to the following: 

 integer(4) function cublasCscal(h, n, a, x, incx)
   type(cublasHandle) :: h
   integer :: n
   complex(4), device :: a ! device or host variable
   complex(4), device, dimension(*) :: x
   integer :: incx

2.6.3.13. csscal

If you use the cublas_v2 module, the interface for cublasCsscal is changed to the following: 

 integer(4) function cublasCsscal(h, n, a, x, incx)
   type(cublasHandle) :: h
   integer :: n
   real(4), device :: a ! device or host variable
   complex(4), device, dimension(*) :: x
   integer :: incx

2.6.3.14. cswap

If you use the cublas_v2 module, the interface for cublasCswap is changed to the following: 

 integer(4) function cublasCswap(h, n, x, incx, y, incy)
   type(cublasHandle) :: h
   integer :: n
   complex(4), device, dimension(*) :: x, y
   integer :: incx, incy

2.6.3.15. cgbmv

If you use the cublas_v2 module, the interface for cublasCgbmv is changed to the following: 

 integer(4) function cublasCgbmv(h, t, m, n, kl, ku, alpha, a, lda, x, incx, beta, y, incy)
   type(cublasHandle) :: h
   integer :: t
   integer :: m, n, kl, ku, lda, incx, incy
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(*) :: x, y
   complex(4), device :: alpha, beta ! device or host variable

2.6.3.16. cgemv

If you use the cublas_v2 module, the interface for cublasCgemv is changed to the following: 

 integer(4) function cublasCgemv(h, t, m, n, alpha, a, lda, x, incx, beta, y, incy)
   type(cublasHandle) :: h
   integer :: t
   integer :: m, n, lda, incx, incy
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(*) :: x, y
   complex(4), device :: alpha, beta ! device or host variable

2.6.3.17. cgerc

If you use the cublas_v2 module, the interface for cublasCgerc is changed to the following: 

 integer(4) function cublasCgerc(h, m, n, alpha, x, incx, y, incy, a, lda)
   type(cublasHandle) :: h
   integer :: m, n, lda, incx, incy
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(*) :: x, y
   complex(4), device :: alpha ! device or host variable

2.6.3.18. cgeru

If you use the cublas_v2 module, the interface for cublasCgeru is changed to the following: 

 integer(4) function cublasCgeru(h, m, n, alpha, x, incx, y, incy, a, lda)
   type(cublasHandle) :: h
   integer :: m, n, lda, incx, incy
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(*) :: x, y
   complex(4), device :: alpha ! device or host variable

2.6.3.19. csymv

If you use the cublas_v2 module, the interface for cublasCsymv is changed to the following: 

 integer(4) function cublasCsymv(h, uplo, n, alpha, a, lda, x, incx, beta, y, incy)
   type(cublasHandle) :: h
   integer :: uplo
   integer :: n, lda, incx, incy
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(*) :: x, y
   complex(4), device :: alpha, beta ! device or host variable

2.6.3.20. csyr

If you use the cublas_v2 module, the interface for cublasCsyr is changed to the following: 

 integer(4) function cublasCsyr(h, t, n, alpha, x, incx, a, lda)
   type(cublasHandle) :: h
   integer :: t
   integer :: n, incx, lda
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(*) :: x
   complex(4), device :: alpha ! device or host variable

2.6.3.21. csyr2

If you use the cublas_v2 module, the interface for cublasCsyr2 is changed to the following: 

 integer(4) function cublasCsyr2(h, t, n, alpha, x, incx, y, incy, a, lda)
   type(cublasHandle) :: h
   integer :: t
   integer :: n, incx, incy, lda
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(*) :: x, y
   complex(4), device :: alpha ! device or host variable

2.6.3.22. ctbmv

If you use the cublas_v2 module, the interface for cublasCtbmv is changed to the following: 

 integer(4) function cublasCtbmv(h, u, t, d, n, k, a, lda, x, incx)
   type(cublasHandle) :: h
   integer :: u, t, d
   integer :: n, k, incx, lda
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(*) :: x

2.6.3.23. ctbsv

If you use the cublas_v2 module, the interface for cublasCtbsv is changed to the following: 

 integer(4) function cublasCtbsv(h, u, t, d, n, k, a, lda, x, incx)
   type(cublasHandle) :: h
   integer :: u, t, d
   integer :: n, k, incx, lda
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(*) :: x

2.6.3.24. ctpmv

If you use the cublas_v2 module, the interface for cublasCtpmv is changed to the following: 

 integer(4) function cublasCtpmv(h, u, t, d, n, a, x, incx)
   type(cublasHandle) :: h
   integer :: u, t, d
   integer :: n, incx
   complex(4), device, dimension(*) :: a, x

2.6.3.25. ctpsv

If you use the cublas_v2 module, the interface for cublasCtpsv is changed to the following: 

 integer(4) function cublasCtpsv(h, u, t, d, n, a, x, incx)
   type(cublasHandle) :: h
   integer :: u, t, d
   integer :: n, incx
   complex(4), device, dimension(*) :: a, x

2.6.3.26. ctrmv

If you use the cublas_v2 module, the interface for cublasCtrmv is changed to the following: 

 integer(4) function cublasCtrmv(h, u, t, d, n, a, lda, x, incx)
   type(cublasHandle) :: h
   integer :: u, t, d
   integer :: n, incx, lda
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(*) :: x

2.6.3.27. ctrsv

If you use the cublas_v2 module, the interface for cublasCtrsv is changed to the following: 

 integer(4) function cublasCtrsv(h, u, t, d, n, a, lda, x, incx)
   type(cublasHandle) :: h
   integer :: u, t, d
   integer :: n, incx, lda
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(*) :: x

2.6.3.28. chbmv

If you use the cublas_v2 module, the interface for cublasChbmv is changed to the following: 

 integer(4) function cublasChbmv(h, uplo, n, k, alpha, a, lda, x, incx, beta, y, incy)
   type(cublasHandle) :: h
   integer :: uplo
   integer :: k, n, lda, incx, incy
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(*) :: x, y
   complex(4), device :: alpha, beta ! device or host variable

2.6.3.29. chemv

If you use the cublas_v2 module, the interface for cublasChemv is changed to the following: 

 integer(4) function cublasChemv(h, uplo, n, alpha, a, lda, x, incx, beta, y, incy)
   type(cublasHandle) :: h
   integer :: uplo
   integer :: n, lda, incx, incy
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(*) :: x, y
   complex(4), device :: alpha, beta ! device or host variable

2.6.3.30. chpmv

If you use the cublas_v2 module, the interface for cublasChpmv is changed to the following: 

 integer(4) function cublasChpmv(h, uplo, n, alpha, a, x, incx, beta, y, incy)
   type(cublasHandle) :: h
   integer :: uplo
   integer :: n, incx, incy
   complex(4), device, dimension(*) :: a, x, y
   complex(4), device :: alpha, beta ! device or host variable

2.6.3.31. cher

If you use the cublas_v2 module, the interface for cublasCher is changed to the following: 

 integer(4) function cublasCher(h, t, n, alpha, x, incx, a, lda)
   type(cublasHandle) :: h
   integer :: t
   integer :: n, incx, lda
   complex(4), device, dimension(*) :: a, x
   real(4), device :: alpha ! device or host variable

2.6.3.32. cher2

If you use the cublas_v2 module, the interface for cublasCher2 is changed to the following: 

 integer(4) function cublasCher2(h, t, n, alpha, x, incx, y, incy, a, lda)
   type(cublasHandle) :: h
   integer :: t
   integer :: n, incx, incy, lda
   complex(4), device, dimension(*) :: a, x, y
   complex(4), device :: alpha ! device or host variable

2.6.3.33. chpr

If you use the cublas_v2 module, the interface for cublasChpr is changed to the following: 

 integer(4) function cublasChpr(h, t, n, alpha, x, incx, a)
   type(cublasHandle) :: h
   integer :: t
   integer :: n, incx
   complex(4), device, dimension(*) :: a, x
   real(4), device :: alpha ! device or host variable

2.6.3.34. chpr2

If you use the cublas_v2 module, the interface for cublasChpr2 is changed to the following: 

 integer(4) function cublasChpr2(h, t, n, alpha, x, incx, y, incy, a)
   type(cublasHandle) :: h
   integer :: t
   integer :: n, incx, incy
   complex(4), device, dimension(*) :: a, x, y
   complex(4), device :: alpha ! device or host variable

2.6.3.35. cgemm

If you use the cublas_v2 module, the interface for cublasCgemm is changed to the following: 

 integer(4) function cublasCgemm(h, transa, transb, m, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: transa, transb
   integer :: m, n, k, lda, ldb, ldc
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(ldb, *) :: b
   complex(4), device, dimension(ldc, *) :: c
   complex(4), device :: alpha, beta ! device or host variable

2.6.3.36. csymm

If you use the cublas_v2 module, the interface for cublasCsymm is changed to the following: 

 integer(4) function cublasCsymm(h, side, uplo, m, n, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: side, uplo
   integer :: m, n, lda, ldb, ldc
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(ldb, *) :: b
   complex(4), device, dimension(ldc, *) :: c
   complex(4), device :: alpha, beta ! device or host variable

2.6.3.37. csyrk

If you use the cublas_v2 module, the interface for cublasCsyrk is changed to the following: 

 integer(4) function cublasCsyrk(h, uplo, trans, n, k, alpha, a, lda, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: uplo, trans
   integer :: n, k, lda, ldc
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(ldc, *) :: c
   complex(4), device :: alpha, beta ! device or host variable

2.6.3.38. csyr2k

If you use the cublas_v2 module, the interface for cublasCsyr2k is changed to the following: 

 integer(4) function cublasCsyr2k(h, uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: uplo, trans
   integer :: n, k, lda, ldb, ldc
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(ldb, *) :: b
   complex(4), device, dimension(ldc, *) :: c
   complex(4), device :: alpha, beta ! device or host variable

2.6.3.39. csyrkx

If you use the cublas_v2 module, the interface for cublasCsyrkx is changed to the following: 

 integer(4) function cublasCsyrkx(h, uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: uplo, trans
   integer :: n, k, lda, ldb, ldc
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(ldb, *) :: b
   complex(4), device, dimension(ldc, *) :: c
   complex(4), device :: alpha, beta ! device or host variable

2.6.3.40. ctrmm

If you use the cublas_v2 module, the interface for cublasCtrmm is changed to the following: 

 integer(4) function cublasCtrmm(h, side, uplo, transa, diag, m, n, alpha, a, lda, b, ldb, c, ldc)
   type(cublasHandle) :: h
   integer :: side, uplo, transa, diag
   integer :: m, n, lda, ldb, ldc
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(ldb, *) :: b
   complex(4), device, dimension(ldc, *) :: c
   complex(4), device :: alpha ! device or host variable

2.6.3.41. ctrsm

If you use the cublas_v2 module, the interface for cublasCtrsm is changed to the following: 

 integer(4) function cublasCtrsm(h, side, uplo, transa, diag, m, n, alpha, a, lda, b, ldb)
   type(cublasHandle) :: h
   integer :: side, uplo, transa, diag
   integer :: m, n, lda, ldb
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(ldb, *) :: b
   complex(4), device :: alpha ! device or host variable

2.6.3.42. chemm

If you use the cublas_v2 module, the interface for cublasChemm is changed to the following: 

 integer(4) function cublasChemm(h, side, uplo, m, n, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: side, uplo
   integer :: m, n, lda, ldb, ldc
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(ldb, *) :: b
   complex(4), device, dimension(ldc, *) :: c
   complex(4), device :: alpha, beta ! device or host variable

2.6.3.43. cherk

If you use the cublas_v2 module, the interface for cublasCherk is changed to the following: 

 integer(4) function cublasCherk(h, uplo, trans, n, k, alpha, a, lda, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: uplo, trans
   integer :: n, k, lda, ldc
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(ldc, *) :: c
   real(4), device :: alpha, beta ! device or host variable

2.6.3.44. cher2k

If you use the cublas_v2 module, the interface for cublasCher2k is changed to the following: 

 integer(4) function cublasCher2k(h, uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: uplo, trans
   integer :: n, k, lda, ldb, ldc
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(ldb, *) :: b
   complex(4), device, dimension(ldc, *) :: c
   complex(4), device :: alpha ! device or host variable
   real(4), device :: beta ! device or host variable

2.6.3.45. cherkx

If you use the cublas_v2 module, the interface for cublasCherkx is changed to the following: 

 integer(4) function cublasCherkx(h, uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: uplo, trans
   integer :: n, k, lda, ldb, ldc
   complex(4), device, dimension(lda, *) :: a
   complex(4), device, dimension(ldb, *) :: b
   complex(4), device, dimension(ldc, *) :: c
   complex(4), device :: alpha ! device or host variable
   real(4), device :: beta ! device or host variable

2.6.4. Double Precision Complex Functions and Subroutines

This section contains the V2 interfaces to the double precision complex BLAS and cuBLAS functions and subroutines.

2.6.4.1. izamax

If you use the cublas_v2 module, the interface for cublasIzamax is changed to the following: 

 integer(4) function cublasIzamax(h, n, x, incx, res)
   type(cublasHandle) :: h
   integer :: n
   complex(8), device, dimension(*) :: x
   integer :: incx
   integer, device :: res ! device or host variable

2.6.4.2. izamin

If you use the cublas_v2 module, the interface for cublasIzamin is changed to the following: 

 integer(4) function cublasIzamin(h, n, x, incx, res)
   type(cublasHandle) :: h
   integer :: n
   complex(8), device, dimension(*) :: x
   integer :: incx
   integer, device :: res ! device or host variable

2.6.4.3. dzasum

If you use the cublas_v2 module, the interface for cublasDzasum is changed to the following: 

 integer(4) function cublasDzasum(h, n, x, incx, res)
   type(cublasHandle) :: h
   integer :: n
   complex(8), device, dimension(*) :: x
   integer :: incx
   real(8), device :: res ! device or host variable

2.6.4.4. zaxpy

If you use the cublas_v2 module, the interface for cublasZaxpy is changed to the following: 

 integer(4) function cublasZaxpy(h, n, a, x, incx, y, incy)
   type(cublasHandle) :: h
   integer :: n
   complex(8), device :: a ! device or host variable
   complex(8), device, dimension(*) :: x, y
   integer :: incx, incy

2.6.4.5. zcopy

If you use the cublas_v2 module, the interface for cublasZcopy is changed to the following: 

 integer(4) function cublasZcopy(h, n, x, incx, y, incy)
   type(cublasHandle) :: h
   integer :: n
   complex(8), device, dimension(*) :: x, y
   integer :: incx, incy

2.6.4.6. zdotc

If you use the cublas_v2 module, the interface for cublasZdotc is changed to the following: 

 integer(4) function cublasZdotc(h, n, x, incx, y, incy, res)
   type(cublasHandle) :: h
   integer :: n
   complex(8), device, dimension(*) :: x, y
   integer :: incx, incy
   complex(8), device :: res ! device or host variable

2.6.4.7. zdotu

If you use the cublas_v2 module, the interface for cublasZdotu is changed to the following: 

 integer(4) function cublasZdotu(h, n, x, incx, y, incy, res)
   type(cublasHandle) :: h
   integer :: n
   complex(8), device, dimension(*) :: x, y
   integer :: incx, incy
   complex(8), device :: res ! device or host variable

2.6.4.8. dznrm2

If you use the cublas_v2 module, the interface for cublasDznrm2 is changed to the following: 

 integer(4) function cublasDznrm2(h, n, x, incx, res)
   type(cublasHandle) :: h
   integer :: n
   complex(8), device, dimension(*) :: x
   integer :: incx
   real(8), device :: res ! device or host variable

2.6.4.9. zrot

If you use the cublas_v2 module, the interface for cublasZrot is changed to the following: 

 integer(4) function cublasZrot(h, n, x, incx, y, incy, sc, ss)
   type(cublasHandle) :: h
   integer :: n
   real(8), device :: sc ! device or host variable
   complex(8), device :: ss ! device or host variable
   complex(8), device, dimension(*) :: x, y
   integer :: incx, incy

2.6.4.10. zsrot

If you use the cublas_v2 module, the interface for cublasZsrot is changed to the following: 

 integer(4) function cublasZsrot(h, n, x, incx, y, incy, sc, ss)
   type(cublasHandle) :: h
   integer :: n
   real(8), device :: sc, ss ! device or host variable
   complex(8), device, dimension(*) :: x, y
   integer :: incx, incy

2.6.4.11. zrotg

If you use the cublas_v2 module, the interface for cublasZrotg is changed to the following: 

 integer(4) function cublasZrotg(h, sa, sb, sc, ss)
   type(cublasHandle) :: h
   complex(8), device :: sa, sb, ss ! device or host variable
   real(8), device :: sc ! device or host variable

2.6.4.12. zscal

If you use the cublas_v2 module, the interface for cublasZscal is changed to the following: 

 integer(4) function cublasZscal(h, n, a, x, incx)
   type(cublasHandle) :: h
   integer :: n
   complex(8), device :: a ! device or host variable
   complex(8), device, dimension(*) :: x
   integer :: incx

2.6.4.13. zdscal

If you use the cublas_v2 module, the interface for cublasZdscal is changed to the following: 

 integer(4) function cublasZdscal(h, n, a, x, incx)
   type(cublasHandle) :: h
   integer :: n
   real(8), device :: a ! device or host variable
   complex(8), device, dimension(*) :: x
   integer :: incx

2.6.4.14. zswap

If you use the cublas_v2 module, the interface for cublasZswap is changed to the following: 

 integer(4) function cublasZswap(h, n, x, incx, y, incy)
   type(cublasHandle) :: h
   integer :: n
   complex(8), device, dimension(*) :: x, y
   integer :: incx, incy

2.6.4.15. zgbmv

If you use the cublas_v2 module, the interface for cublasZgbmv is changed to the following: 

 integer(4) function cublasZgbmv(h, t, m, n, kl, ku, alpha, a, lda, x, incx, beta, y, incy)
   type(cublasHandle) :: h
   integer :: t
   integer :: m, n, kl, ku, lda, incx, incy
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(*) :: x, y
   complex(8), device :: alpha, beta ! device or host variable

2.6.4.16. zgemv

If you use the cublas_v2 module, the interface for cublasZgemv is changed to the following: 

 integer(4) function cublasZgemv(h, t, m, n, alpha, a, lda, x, incx, beta, y, incy)
   type(cublasHandle) :: h
   integer :: t
   integer :: m, n, lda, incx, incy
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(*) :: x, y
   complex(8), device :: alpha, beta ! device or host variable

2.6.4.17. zgerc

If you use the cublas_v2 module, the interface for cublasZgerc is changed to the following: 

 integer(4) function cublasZgerc(h, m, n, alpha, x, incx, y, incy, a, lda)
   type(cublasHandle) :: h
   integer :: m, n, lda, incx, incy
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(*) :: x, y
   complex(8), device :: alpha ! device or host variable

2.6.4.18. zgeru

If you use the cublas_v2 module, the interface for cublasZgeru is changed to the following: 

 integer(4) function cublasZgeru(h, m, n, alpha, x, incx, y, incy, a, lda)
   type(cublasHandle) :: h
   integer :: m, n, lda, incx, incy
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(*) :: x, y
   complex(8), device :: alpha ! device or host variable

2.6.4.19. zsymv

If you use the cublas_v2 module, the interface for cublasZsymv is changed to the following: 

 integer(4) function cublasZsymv(h, uplo, n, alpha, a, lda, x, incx, beta, y, incy)
   type(cublasHandle) :: h
   integer :: uplo
   integer :: n, lda, incx, incy
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(*) :: x, y
   complex(8), device :: alpha, beta ! device or host variable

2.6.4.20. zsyr

If you use the cublas_v2 module, the interface for cublasZsyr is changed to the following: 

 integer(4) function cublasZsyr(h, t, n, alpha, x, incx, a, lda)
   type(cublasHandle) :: h
   integer :: t
   integer :: n, incx, lda
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(*) :: x
   complex(8), device :: alpha ! device or host variable

2.6.4.21. zsyr2

If you use the cublas_v2 module, the interface for cublasZsyr2 is changed to the following: 

 integer(4) function cublasZsyr2(h, t, n, alpha, x, incx, y, incy, a, lda)
   type(cublasHandle) :: h
   integer :: t
   integer :: n, incx, incy, lda
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(*) :: x, y
   complex(8), device :: alpha ! device or host variable

2.6.4.22. ztbmv

If you use the cublas_v2 module, the interface for cublasZtbmv is changed to the following: 

 integer(4) function cublasZtbmv(h, u, t, d, n, k, a, lda, x, incx)
   type(cublasHandle) :: h
   integer :: u, t, d
   integer :: n, k, incx, lda
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(*) :: x

2.6.4.23. ztbsv

If you use the cublas_v2 module, the interface for cublasZtbsv is changed to the following: 

 integer(4) function cublasZtbsv(h, u, t, d, n, k, a, lda, x, incx)
   type(cublasHandle) :: h
   integer :: u, t, d
   integer :: n, k, incx, lda
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(*) :: x

2.6.4.24. ztpmv

If you use the cublas_v2 module, the interface for cublasZtpmv is changed to the following: 

 integer(4) function cublasZtpmv(h, u, t, d, n, a, x, incx)
   type(cublasHandle) :: h
   integer :: u, t, d
   integer :: n, incx
   complex(8), device, dimension(*) :: a, x

2.6.4.25. ztpsv

If you use the cublas_v2 module, the interface for cublasZtpsv is changed to the following: 

 integer(4) function cublasZtpsv(h, u, t, d, n, a, x, incx)
   type(cublasHandle) :: h
   integer :: u, t, d
   integer :: n, incx
   complex(8), device, dimension(*) :: a, x

2.6.4.26. ztrmv

If you use the cublas_v2 module, the interface for cublasZtrmv is changed to the following: 

 integer(4) function cublasZtrmv(h, u, t, d, n, a, lda, x, incx)
   type(cublasHandle) :: h
   integer :: u, t, d
   integer :: n, incx, lda
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(*) :: x

2.6.4.27. ztrsv

If you use the cublas_v2 module, the interface for cublasZtrsv is changed to the following: 

 integer(4) function cublasZtrsv(h, u, t, d, n, a, lda, x, incx)
   type(cublasHandle) :: h
   integer :: u, t, d
   integer :: n, incx, lda
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(*) :: x

2.6.4.28. zhbmv

If you use the cublas_v2 module, the interface for cublasZhbmv is changed to the following: 

 integer(4) function cublasZhbmv(h, uplo, n, k, alpha, a, lda, x, incx, beta, y, incy)
   type(cublasHandle) :: h
   integer :: uplo
   integer :: k, n, lda, incx, incy
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(*) :: x, y
   complex(8), device :: alpha, beta ! device or host variable

2.6.4.29. zhemv

If you use the cublas_v2 module, the interface for cublasZhemv is changed to the following: 

 integer(4) function cublasZhemv(h, uplo, n, alpha, a, lda, x, incx, beta, y, incy)
   type(cublasHandle) :: h
   integer :: uplo
   integer :: n, lda, incx, incy
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(*) :: x, y
   complex(8), device :: alpha, beta ! device or host variable

2.6.4.30. zhpmv

If you use the cublas_v2 module, the interface for cublasZhpmv is changed to the following: 

 integer(4) function cublasZhpmv(h, uplo, n, alpha, a, x, incx, beta, y, incy)
   type(cublasHandle) :: h
   integer :: uplo
   integer :: n, incx, incy
   complex(8), device, dimension(*) :: a, x, y
   complex(8), device :: alpha, beta ! device or host variable

2.6.4.31. zher

If you use the cublas_v2 module, the interface for cublasZher is changed to the following: 

 integer(4) function cublasZher(h, t, n, alpha, x, incx, a, lda)
   type(cublasHandle) :: h
   integer :: t
   integer :: n, incx, lda
   complex(8), device, dimension(*) :: a, x
   real(8), device :: alpha ! device or host variable

2.6.4.32. zher2

If you use the cublas_v2 module, the interface for cublasZher2 is changed to the following: 

 integer(4) function cublasZher2(h, t, n, alpha, x, incx, y, incy, a, lda)
   type(cublasHandle) :: h
   integer :: t
   integer :: n, incx, incy, lda
   complex(8), device, dimension(*) :: a, x, y
   complex(8), device :: alpha ! device or host variable

2.6.4.33. zhpr

If you use the cublas_v2 module, the interface for cublasZhpr is changed to the following: 

 integer(4) function cublasZhpr(h, t, n, alpha, x, incx, a)
   type(cublasHandle) :: h
   integer :: t
   integer :: n, incx
   complex(8), device, dimension(*) :: a, x
   real(8), device :: alpha ! device or host variable

2.6.4.34. zhpr2

If you use the cublas_v2 module, the interface for cublasZhpr2 is changed to the following: 

 integer(4) function cublasZhpr2(h, t, n, alpha, x, incx, y, incy, a)
   type(cublasHandle) :: h
   integer :: t
   integer :: n, incx, incy
   complex(8), device, dimension(*) :: a, x, y
   complex(8), device :: alpha ! device or host variable

2.6.4.35. zgemm

If you use the cublas_v2 module, the interface for cublasZgemm is changed to the following: 

 integer(4) function cublasZgemm(h, transa, transb, m, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: transa, transb
   integer :: m, n, k, lda, ldb, ldc
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(ldb, *) :: b
   complex(8), device, dimension(ldc, *) :: c
   complex(8), device :: alpha, beta ! device or host variable

2.6.4.36. zsymm

If you use the cublas_v2 module, the interface for cublasZsymm is changed to the following: 

 integer(4) function cublasZsymm(h, side, uplo, m, n, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: side, uplo
   integer :: m, n, lda, ldb, ldc
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(ldb, *) :: b
   complex(8), device, dimension(ldc, *) :: c
   complex(8), device :: alpha, beta ! device or host variable

2.6.4.37. zsyrk

If you use the cublas_v2 module, the interface for cublasZsyrk is changed to the following: 

 integer(4) function cublasZsyrk(h, uplo, trans, n, k, alpha, a, lda, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: uplo, trans
   integer :: n, k, lda, ldc
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(ldc, *) :: c
   complex(8), device :: alpha, beta ! device or host variable

2.6.4.38. zsyr2k

If you use the cublas_v2 module, the interface for cublasZsyr2k is changed to the following: 

 integer(4) function cublasZsyr2k(h, uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: uplo, trans
   integer :: n, k, lda, ldb, ldc
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(ldb, *) :: b
   complex(8), device, dimension(ldc, *) :: c
   complex(8), device :: alpha, beta ! device or host variable

2.6.4.39. zsyrkx

If you use the cublas_v2 module, the interface for cublasZsyrkx is changed to the following: 

 integer(4) function cublasZsyrkx(h, uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: uplo, trans
   integer :: n, k, lda, ldb, ldc
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(ldb, *) :: b
   complex(8), device, dimension(ldc, *) :: c
   complex(8), device :: alpha, beta ! device or host variable

2.6.4.40. ztrmm

If you use the cublas_v2 module, the interface for cublasZtrmm is changed to the following: 

 integer(4) function cublasZtrmm(h, side, uplo, transa, diag, m, n, alpha, a, lda, b, ldb, c, ldc)
   type(cublasHandle) :: h
   integer :: side, uplo, transa, diag
   integer :: m, n, lda, ldb, ldc
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(ldb, *) :: b
   complex(8), device, dimension(ldc, *) :: c
   complex(8), device :: alpha ! device or host variable

2.6.4.41. ztrsm

If you use the cublas_v2 module, the interface for cublasZtrsm is changed to the following: 

 integer(4) function cublasZtrsm(h, side, uplo, transa, diag, m, n, alpha, a, lda, b, ldb)
   type(cublasHandle) :: h
   integer :: side, uplo, transa, diag
   integer :: m, n, lda, ldb
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(ldb, *) :: b
   complex(8), device :: alpha ! device or host variable

2.6.4.42. zhemm

If you use the cublas_v2 module, the interface for cublasZhemm is changed to the following: 

 integer(4) function cublasZhemm(h, side, uplo, m, n, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: side, uplo
   integer :: m, n, lda, ldb, ldc
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(ldb, *) :: b
   complex(8), device, dimension(ldc, *) :: c
   complex(8), device :: alpha, beta ! device or host variable

2.6.4.43. zherk

If you use the cublas_v2 module, the interface for cublasZherk is changed to the following: 

 integer(4) function cublasZherk(h, uplo, trans, n, k, alpha, a, lda, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: uplo, trans
   integer :: n, k, lda, ldc
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(ldc, *) :: c
   real(8), device :: alpha, beta ! device or host variable

2.6.4.44. zher2k

If you use the cublas_v2 module, the interface for cublasZher2k is changed to the following: 

 integer(4) function cublasZher2k(h, uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: uplo, trans
   integer :: n, k, lda, ldb, ldc
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(ldb, *) :: b
   complex(8), device, dimension(ldc, *) :: c
   complex(8), device :: alpha ! device or host variable
   real(8), device :: beta ! device or host variable

2.6.4.45. zherkx

If you use the cublas_v2 module, the interface for cublasZherkx is changed to the following: 

 integer(4) function cublasZherkx(h, uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasHandle) :: h
   integer :: uplo, trans
   integer :: n, k, lda, ldb, ldc
   complex(8), device, dimension(lda, *) :: a
   complex(8), device, dimension(ldb, *) :: b
   complex(8), device, dimension(ldc, *) :: c
   complex(8), device :: alpha ! device or host variable
   real(8), device :: beta ! device or host variable

2.7. CUBLAS XT Module Functions

This section contains interfaces to the cuBLAS XT Module Functions. Users can access this module by inserting the line use cublasXt into the program unit. The cublasXt library is a host-side library, which supports multiple GPUs. Here is an example: 

subroutine testxt(n)
use cublasXt
complex*16 :: a(n,n), b(n,n), c(n,n), alpha, beta
type(cublasXtHandle) :: h
integer ndevices(1)
a = cmplx(1.0d0,0.0d0)
b = cmplx(2.0d0,0.0d0)
c = cmplx(-1.0d0,0.0d0)
alpha = cmplx(1.0d0,0.0d0)
beta = cmplx(0.0d0,0.0d0)
istat = cublasXtCreate(h)
if (istat .ne. CUBLAS_STATUS_SUCCESS) print *,istat
ndevices(1) = 0
istat = cublasXtDeviceSelect(h, 1, ndevices)
if (istat .ne. CUBLAS_STATUS_SUCCESS) print *,istat
istat = cublasXtZgemm(h, CUBLAS_OP_N, CUBLAS_OP_N, &
                      n, n, n, &
                      alpha, A, n, B, n, beta, C, n)
if (istat .ne. CUBLAS_STATUS_SUCCESS) print *,istat
istat = cublasXtDestroy(h)
if (istat .ne. CUBLAS_STATUS_SUCCESS) print *,istat
if (all(dble(c).eq.2.0d0*n)) then
    print *,"Test PASSED"
else
    print *,"Test FAILED"
endif
end

The cublasXt module contains all the types and definitions from the cublas module, and these additional types and enumerations: 

TYPE cublasXtHandle
  TYPE(C_PTR)  :: handle
END TYPE
! Pinned memory mode
enum, bind(c)
    enumerator :: CUBLASXT_PINNING_DISABLED=0
    enumerator :: CUBLASXT_PINNING_ENABLED=1
end enum 
! cublasXtOpType
enum, bind(c)
    enumerator :: CUBLASXT_FLOAT=0
    enumerator :: CUBLASXT_DOUBLE=1
    enumerator :: CUBLASXT_COMPLEX=2
    enumerator :: CUBLASXT_DOUBLECOMPLEX=3
end enum 
! cublasXtBlasOp
enum, bind(c)
    enumerator :: CUBLASXT_GEMM=0
    enumerator :: CUBLASXT_SYRK=1
    enumerator :: CUBLASXT_HERK=2
    enumerator :: CUBLASXT_SYMM=3
    enumerator :: CUBLASXT_HEMM=4
    enumerator :: CUBLASXT_TRSM=5
    enumerator :: CUBLASXT_SYR2K=6
    enumerator :: CUBLASXT_HER2K=7
    enumerator :: CUBLASXT_SPMM=8
    enumerator :: CUBLASXT_SYRKX=9
    enumerator :: CUBLASXT_HERKX=10
    enumerator :: CUBLASXT_TRMM=11
    enumerator :: CUBLASXT_ROUTINE_MAX=12
end enum

2.7.1. cublasXtCreate

This function initializes the cublasXt API and creates a handle to an opaque structure holding the cublasXT library context. It allocates hardware resources on the host and device and must be called prior to making any other cublasXt API library calls. 

 integer(4) function cublasXtcreate(h)
   type(cublasXtHandle) :: h

2.7.2. cublasXtDestroy

This function releases hardware resources used by the cublasXt API context. This function is usually the last call with a particular handle to the cublasXt API. 

 integer(4) function cublasXtdestroy(h)
   type(cublasXtHandle) :: h

2.7.3. cublasXtDeviceSelect

This function allows the user to provide the number of GPU devices and their respective Ids that will participate to the subsequent cublasXt API math function calls. This function will create a cuBLAS context for every GPU provided in that list. Currently the device configuration is static and cannot be changed between math function calls. In that regard, this function should be called only once after cublasXtCreate. To be able to run multiple configurations, multiple cublasXt API contexts should be created. 

 integer(4) function cublasXtdeviceselect(h, ndevices, deviceid)
   type(cublasXtHandle) :: h
   integer :: ndevices
   integer, dimension(*) :: deviceid

2.7.4. cublasXtSetBlockDim

This function allows the user to set the block dimension used for the tiling of the matrices for the subsequent Math function calls. Matrices are split in square tiles of blockDim x blockDim dimension. This function can be called anytime and will take effect for the following math function calls. The block dimension should be chosen in a way to optimize the math operation and to make sure that the PCI transfers are well overlapped with the computation. 

 integer(4) function cublasXtsetblockdim(h, blockdim)
   type(cublasXtHandle) :: h
   integer :: blockdim

2.7.5. cublasXtGetBlockDim

This function allows the user to query the block dimension used for the tiling of the matrices. 

 integer(4) function cublasXtgetblockdim(h, blockdim)
   type(cublasXtHandle) :: h
   integer :: blockdim

2.7.6. cublasXtSetCpuRoutine

This function allows the user to provide a CPU implementation of the corresponding BLAS routine. This function can be used with the function cublasXtSetCpuRatio() to define an hybrid computation between the CPU and the GPUs. Currently the hybrid feature is only supported for the xGEMM routines. 

 integer(4) function cublasXtsetcpuroutine(h, blasop, blastype)
   type(cublasXtHandle) :: h
   integer :: blasop, blastype

2.7.7. cublasXtSetCpuRatio

This function allows the user to define the percentage of workload that should be done on a CPU in the context of an hybrid computation. This function can be used with the function cublasXtSetCpuRoutine() to define an hybrid computation between the CPU and the GPUs. Currently the hybrid feature is only supported for the xGEMM routines. 

 integer(4) function cublasXtsetcpuratio(h, blasop, blastype, ratio)
   type(cublasXtHandle) :: h
   integer :: blasop, blastype
   real(4) :: ratio

2.7.8. cublasXtSetPinningMemMode

This function allows the user to enable or disable the Pinning Memory mode. When enabled, the matrices passed in subsequent cublasXt API calls will be pinned/unpinned using the CUDART routine cudaHostRegister and cudaHostUnregister respectively if the matrices are not already pinned. If a matrix happened to be pinned partially, it will also not be pinned. Pinning the memory improve PCI transfer performace and allows to overlap PCI memory transfer with computation. However pinning/unpinning the memory takes some time which might not be amortized. It is advised that the user pins the memory on its own using cudaMallocHost or cudaHostRegister and unpins it when the computation sequence is completed. By default, the Pinning Memory mode is disabled. 

 integer(4) function cublasXtsetpinningmemmode(h, mode)
   type(cublasXtHandle) :: h
   integer :: mode

2.7.9. cublasXtGetPinningMemMode

This function allows the user to query the Pinning Memory mode. By default, the Pinning Memory mode is disabled. 

 integer(4) function cublasXtgetpinningmemmode(h, mode)
   type(cublasXtHandle) :: h
   integer :: mode

2.7.10. cublasXtSgemm

SGEMM performs one of the matrix-matrix operations C := alpha*op( A )*op( B ) + beta*C, where op( X ) is one of op( X ) = X or op( X ) = X**T, alpha and beta are scalars, and A, B and C are matrices, with op( A ) an m by k matrix, op( B ) a k by n matrix and C an m by n matrix. 

 integer(4) function cublasXtsgemm(h, transa, transb, m, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasXtHandle) :: h
   integer :: transa, transb
   integer(kind=c_intptr_t) :: m, n, k, lda, ldb, ldc
   real(4), dimension(lda, *) :: a
   real(4), dimension(ldb, *) :: b
   real(4), dimension(ldc, *) :: c
   real(4) :: alpha, beta

2.7.11. cublasXtSsymm

SSYMM performs one of the matrix-matrix operations C := alpha*A*B + beta*C, or C := alpha*B*A + beta*C, where alpha and beta are scalars, A is a symmetric matrix and B and C are m by n matrices. 

 integer(4) function cublasXtssymm(h, side, uplo, m, n, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasXtHandle) :: h
   integer :: side, uplo
   integer(kind=c_intptr_t) :: m, n, lda, ldb, ldc
   real(4), dimension(lda, *) :: a
   real(4), dimension(ldb, *) :: b
   real(4), dimension(ldc, *) :: c
   real(4) :: alpha, beta

2.7.12. cublasXtSsyrk

SSYRK performs one of the symmetric rank k operations C := alpha*A*A**T + beta*C, or C := alpha*A**T*A + beta*C, where alpha and beta are scalars, C is an n by n symmetric matrix and A is an n by k matrix in the first case and a k by n matrix in the second case. 

 integer(4) function cublasXtssyrk(h, uplo, trans, n, k, alpha, a, lda, beta, c, ldc)
   type(cublasXtHandle) :: h
   integer :: uplo, trans
   integer(kind=c_intptr_t) :: n, k, lda, ldc
   real(4), dimension(lda, *) :: a
   real(4), dimension(ldc, *) :: c
   real(4) :: alpha, beta

2.7.13. cublasXtSsyr2k

SSYR2K performs one of the symmetric rank 2k operations C := alpha*A*B**T + alpha*B*A**T + beta*C, or C := alpha*A**T*B + alpha*B**T*A + beta*C, where alpha and beta are scalars, C is an n by n symmetric matrix and A and B are n by k matrices in the first case and k by n matrices in the second case. 

 integer(4) function cublasXtssyr2k(h, uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasXtHandle) :: h
   integer :: uplo, trans
   integer(kind=c_intptr_t) :: n, k, lda, ldb, ldc
   real(4), dimension(lda, *) :: a
   real(4), dimension(ldb, *) :: b
   real(4), dimension(ldc, *) :: c
   real(4) :: alpha, beta

2.7.14. cublasXtSsyrkx

SSYRKX performs a variation of the symmetric rank k update C := alpha*A*B**T + beta*C, where alpha and beta are scalars, C is an n by n symmetric matrix stored in lower or upper mode, and A and B are n by k matrices. This routine can be used when B is in such a way that the result is guaranteed to be symmetric. See the CUBLAS documentation for more details. 

 integer(4) function cublasXtssyrkx(h, uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasXtHandle) :: h
   integer :: uplo, trans
   integer(kind=c_intptr_t) :: n, k, lda, ldb, ldc
   real(4), dimension(lda, *) :: a
   real(4), dimension(ldb, *) :: b
   real(4), dimension(ldc, *) :: c
   real(4) :: alpha, beta

2.7.15. cublasXtStrmm

STRMM performs one of the matrix-matrix operations B := alpha*op( A )*B, or B := alpha*B*op( A ), where alpha is a scalar, B is an m by n matrix, A is a unit, or non-unit, upper or lower triangular matrix and op( A ) is one of op( A ) = A or op( A ) = A**T. 

 integer(4) function cublasXtstrmm(h, side, uplo, transa, diag, m, n, alpha, a, lda, b, ldb, c, ldc)
   type(cublasXtHandle) :: h
   integer :: side, uplo, transa, diag
   integer(kind=c_intptr_t) :: m, n, lda, ldb, ldc
   real(4), dimension(lda, *) :: a
   real(4), dimension(ldb, *) :: b
   real(4), dimension(ldc, *) :: c
   real(4) :: alpha

2.7.16. cublasXtStrsm

STRSM solves one of the matrix equations op( A )*X = alpha*B, or X*op( A ) = alpha*B, where alpha is a scalar, X and B are m by n matrices, A is a unit, or non-unit, upper or lower triangular matrix and op( A ) is one of op( A ) = A or op( A ) = A**T. The matrix X is overwritten on B. 

 integer(4) function cublasXtstrsm(h, side, uplo, transa, diag, m, n, alpha, a, lda, b, ldb)
   type(cublasXtHandle) :: h
   integer :: side, uplo, transa, diag
   integer(kind=c_intptr_t) :: m, n, lda, ldb
   real(4), dimension(lda, *) :: a
   real(4), dimension(ldb, *) :: b
   real(4) :: alpha

2.7.17. cublasXtSspmm

SSPMM performs one of the symmetric packed matrix-matrix operations C := alpha*A*B + beta*C, or C := alpha*B*A + beta*C, where alpha and beta are scalars, A is a n by n symmetric matrix stored in packed format, and B and C are m by n matrices. 

 integer(4) function cublasXtsspmm(h, side, uplo, m, n, alpha, ap, b, ldb, beta, c, ldc)
   type(cublasXtHandle) :: h
   integer :: side, uplo
   integer(kind=c_intptr_t) :: m, n, ldb, ldc
   real(4), dimension(*) :: ap
   real(4), dimension(ldb, *) :: b
   real(4), dimension(ldc, *) :: c
   real(4) :: alpha, beta

2.7.18. cublasXtCgemm

CGEMM performs one of the matrix-matrix operations C := alpha*op( A )*op( B ) + beta*C, where op( X ) is one of op( X ) = X or op( X ) = X**T or op( X ) = X**H, alpha and beta are scalars, and A, B and C are matrices, with op( A ) an m by k matrix, op( B ) a k by n matrix and C an m by n matrix. 

 integer(4) function cublasXtcgemm(h, transa, transb, m, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasXtHandle) :: h
   integer :: transa, transb
   integer(kind=c_intptr_t) :: m, n, k, lda, ldb, ldc
   complex(4), dimension(lda, *) :: a
   complex(4), dimension(ldb, *) :: b
   complex(4), dimension(ldc, *) :: c
   complex(4) :: alpha, beta

2.7.19. cublasXtChemm

CHEMM performs one of the matrix-matrix operations C := alpha*A*B + beta*C, or C := alpha*B*A + beta*C, where alpha and beta are scalars, A is an hermitian matrix and B and C are m by n matrices. 

 integer(4) function cublasXtchemm(h, side, uplo, m, n, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasXtHandle) :: h
   integer :: side, uplo
   integer(kind=c_intptr_t) :: m, n, lda, ldb, ldc
   complex(4), dimension(lda, *) :: a
   complex(4), dimension(ldb, *) :: b
   complex(4), dimension(ldc, *) :: c
   complex(4) :: alpha, beta

2.7.20. cublasXtCherk

CHERK performs one of the hermitian rank k operations C := alpha*A*A**H + beta*C, or C := alpha*A**H*A + beta*C, where alpha and beta are real scalars, C is an n by n hermitian matrix and A is an n by k matrix in the first case and a k by n matrix in the second case. 

 integer(4) function cublasXtcherk(h, uplo, trans, n, k, alpha, a, lda, beta, c, ldc)
   type(cublasXtHandle) :: h
   integer :: uplo, trans
   integer(kind=c_intptr_t) :: n, k, lda, ldc
   complex(4), dimension(lda, *) :: a
   complex(4), dimension(ldc, *) :: c
   real(4) :: alpha, beta

2.7.21. cublasXtCher2k

CHER2K performs one of the hermitian rank 2k operations C := alpha*A*B**H + conjg( alpha )*B*A**H + beta*C, or C := alpha*A**H*B + conjg( alpha )*B**H*A + beta*C, where alpha and beta are scalars with beta real, C is an n by n hermitian matrix and A and B are n by k matrices in the first case and k by n matrices in the second case. 

 integer(4) function cublasXtcher2k(h, uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasXtHandle) :: h
   integer :: uplo, trans
   integer(kind=c_intptr_t) :: n, k, lda, ldb, ldc
   complex(4), dimension(lda, *) :: a
   complex(4), dimension(ldb, *) :: b
   complex(4), dimension(ldc, *) :: c
   complex(4) :: alpha
   real(4) :: beta

2.7.22. cublasXtCherkx

CHERKX performs a variation of the hermitian rank k operations C := alpha*A*B**H + beta*C, where alpha and beta are real scalars, C is an n by n hermitian matrix stored in lower or upper mode, and A and B are n by k matrices. See the CUBLAS documentation for more details. 

 integer(4) function cublasXtcherkx(h, uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasXtHandle) :: h
   integer :: uplo, trans
   integer(kind=c_intptr_t) :: n, k, lda, ldb, ldc
   complex(4), dimension(lda, *) :: a
   complex(4), dimension(ldb, *) :: b
   complex(4), dimension(ldc, *) :: c
   complex(4) :: alpha
   real(4) :: beta

2.7.23. cublasXtCsymm

CSYMM performs one of the matrix-matrix operations C := alpha*A*B + beta*C, or C := alpha*B*A + beta*C, where alpha and beta are scalars, A is a symmetric matrix and B and C are m by n matrices. 

 integer(4) function cublasXtcsymm(h, side, uplo, m, n, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasXtHandle) :: h
   integer :: side, uplo
   integer(kind=c_intptr_t) :: m, n, lda, ldb, ldc
   complex(4), dimension(lda, *) :: a
   complex(4), dimension(ldb, *) :: b
   complex(4), dimension(ldc, *) :: c
   complex(4) :: alpha, beta

2.7.24. cublasXtCsyrk

CSYRK performs one of the symmetric rank k operations C := alpha*A*A**T + beta*C, or C := alpha*A**T*A + beta*C, where alpha and beta are scalars, C is an n by n symmetric matrix and A is an n by k matrix in the first case and a k by n matrix in the second case. 

 integer(4) function cublasXtcsyrk(h, uplo, trans, n, k, alpha, a, lda, beta, c, ldc)
   type(cublasXtHandle) :: h
   integer :: uplo, trans
   integer(kind=c_intptr_t) :: n, k, lda, ldc
   complex(4), dimension(lda, *) :: a
   complex(4), dimension(ldc, *) :: c
   complex(4) :: alpha, beta

2.7.25. cublasXtCsyr2k

CSYR2K performs one of the symmetric rank 2k operations C := alpha*A*B**T + alpha*B*A**T + beta*C, or C := alpha*A**T*B + alpha*B**T*A + beta*C, where alpha and beta are scalars, C is an n by n symmetric matrix and A and B are n by k matrices in the first case and k by n matrices in the second case. 

 integer(4) function cublasXtcsyr2k(h, uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasXtHandle) :: h
   integer :: uplo, trans
   integer(kind=c_intptr_t) :: n, k, lda, ldb, ldc
   complex(4), dimension(lda, *) :: a
   complex(4), dimension(ldb, *) :: b
   complex(4), dimension(ldc, *) :: c
   complex(4) :: alpha, beta

2.7.26. cublasXtCsyrkx

CSYRKX performs a variation of the symmetric rank k update C := alpha*A*B**T + beta*C, where alpha and beta are scalars, C is an n by n symmetric matrix stored in lower or upper mode, and A and B are n by k matrices. This routine can be used when B is in such a way that the result is guaranteed to be symmetric. See the CUBLAS documentation for more details. 

 integer(4) function cublasXtcsyrkx(h, uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasXtHandle) :: h
   integer :: uplo, trans
   integer(kind=c_intptr_t) :: n, k, lda, ldb, ldc
   complex(4), dimension(lda, *) :: a
   complex(4), dimension(ldb, *) :: b
   complex(4), dimension(ldc, *) :: c
   complex(4) :: alpha, beta

2.7.27. cublasXtCtrmm

CTRMM performs one of the matrix-matrix operations B := alpha*op( A )*B, or B := alpha*B*op( A ) where alpha is a scalar, B is an m by n matrix, A is a unit, or non-unit, upper or lower triangular matrix and op( A ) is one of op( A ) = A or op( A ) = A**T or op( A ) = A**H. 

 integer(4) function cublasXtctrmm(h, side, uplo, transa, diag, m, n, alpha, a, lda, b, ldb, c, ldc)
   type(cublasXtHandle) :: h
   integer :: side, uplo, transa, diag
   integer(kind=c_intptr_t) :: m, n, lda, ldb, ldc
   complex(4), dimension(lda, *) :: a
   complex(4), dimension(ldb, *) :: b
   complex(4), dimension(ldc, *) :: c
   complex(4) :: alpha

2.7.28. cublasXtCtrsm

CTRSM solves one of the matrix equations op( A )*X = alpha*B, or X*op( A ) = alpha*B, where alpha is a scalar, X and B are m by n matrices, A is a unit, or non-unit, upper or lower triangular matrix and op( A ) is one of op( A ) = A or op( A ) = A**T or op( A ) = A**H. The matrix X is overwritten on B. 

 integer(4) function cublasXtctrsm(h, side, uplo, transa, diag, m, n, alpha, a, lda, b, ldb)
   type(cublasXtHandle) :: h
   integer :: side, uplo, transa, diag
   integer(kind=c_intptr_t) :: m, n, lda, ldb
   complex(4), dimension(lda, *) :: a
   complex(4), dimension(ldb, *) :: b
   complex(4) :: alpha

2.7.29. cublasXtCspmm

CSPMM performs one of the symmetric packed matrix-matrix operations C := alpha*A*B + beta*C, or C := alpha*B*A + beta*C, where alpha and beta are scalars, A is a n by n symmetric matrix stored in packed format, and B and C are m by n matrices. 

 integer(4) function cublasXtcspmm(h, side, uplo, m, n, alpha, ap, b, ldb, beta, c, ldc)
   type(cublasXtHandle) :: h
   integer :: side, uplo
   integer(kind=c_intptr_t) :: m, n, ldb, ldc
   complex(4), dimension(*) :: ap
   complex(4), dimension(ldb, *) :: b
   complex(4), dimension(ldc, *) :: c
   complex(4) :: alpha, beta

2.7.30. cublasXtDgemm

DGEMM performs one of the matrix-matrix operations C := alpha*op( A )*op( B ) + beta*C, where op( X ) is one of op( X ) = X or op( X ) = X**T, alpha and beta are scalars, and A, B and C are matrices, with op( A ) an m by k matrix, op( B ) a k by n matrix and C an m by n matrix. 

 integer(4) function cublasXtdgemm(h, transa, transb, m, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasXtHandle) :: h
   integer :: transa, transb
   integer(kind=c_intptr_t) :: m, n, k, lda, ldb, ldc
   real(8), dimension(lda, *) :: a
   real(8), dimension(ldb, *) :: b
   real(8), dimension(ldc, *) :: c
   real(8) :: alpha, beta

2.7.31. cublasXtDsymm

DSYMM performs one of the matrix-matrix operations C := alpha*A*B + beta*C, or C := alpha*B*A + beta*C, where alpha and beta are scalars, A is a symmetric matrix and B and C are m by n matrices. 

 integer(4) function cublasXtdsymm(h, side, uplo, m, n, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasXtHandle) :: h
   integer :: side, uplo
   integer(kind=c_intptr_t) :: m, n, lda, ldb, ldc
   real(8), dimension(lda, *) :: a
   real(8), dimension(ldb, *) :: b
   real(8), dimension(ldc, *) :: c
   real(8) :: alpha, beta

2.7.32. cublasXtDsyrk

DSYRK performs one of the symmetric rank k operations C := alpha*A*A**T + beta*C, or C := alpha*A**T*A + beta*C, where alpha and beta are scalars, C is an n by n symmetric matrix and A is an n by k matrix in the first case and a k by n matrix in the second case. 

 integer(4) function cublasXtdsyrk(h, uplo, trans, n, k, alpha, a, lda, beta, c, ldc)
   type(cublasXtHandle) :: h
   integer :: uplo, trans
   integer(kind=c_intptr_t) :: n, k, lda, ldc
   real(8), dimension(lda, *) :: a
   real(8), dimension(ldc, *) :: c
   real(8) :: alpha, beta

2.7.33. cublasXtDsyr2k

DSYR2K performs one of the symmetric rank 2k operations C := alpha*A*B**T + alpha*B*A**T + beta*C, or C := alpha*A**T*B + alpha*B**T*A + beta*C, where alpha and beta are scalars, C is an n by n symmetric matrix and A and B are n by k matrices in the first case and k by n matrices in the second case. 

 integer(4) function cublasXtdsyr2k(h, uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasXtHandle) :: h
   integer :: uplo, trans
   integer(kind=c_intptr_t) :: n, k, lda, ldb, ldc
   real(8), dimension(lda, *) :: a
   real(8), dimension(ldb, *) :: b
   real(8), dimension(ldc, *) :: c
   real(8) :: alpha, beta

2.7.34. cublasXtDsyrkx

DSYRKX performs a variation of the symmetric rank k update C := alpha*A*B**T + beta*C, where alpha and beta are scalars, C is an n by n symmetric matrix stored in lower or upper mode, and A and B are n by k matrices. This routine can be used when B is in such a way that the result is guaranteed to be symmetric. See the CUBLAS documentation for more details. 

 integer(4) function cublasXtdsyrkx(h, uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasXtHandle) :: h
   integer :: uplo, trans
   integer(kind=c_intptr_t) :: n, k, lda, ldb, ldc
   real(8), dimension(lda, *) :: a
   real(8), dimension(ldb, *) :: b
   real(8), dimension(ldc, *) :: c
   real(8) :: alpha, beta

2.7.35. cublasXtDtrmm

DTRMM performs one of the matrix-matrix operations B := alpha*op( A )*B, or B := alpha*B*op( A ), where alpha is a scalar, B is an m by n matrix, A is a unit, or non-unit, upper or lower triangular matrix and op( A ) is one of op( A ) = A or op( A ) = A**T. 

 integer(4) function cublasXtdtrmm(h, side, uplo, transa, diag, m, n, alpha, a, lda, b, ldb, c, ldc)
   type(cublasXtHandle) :: h
   integer :: side, uplo, transa, diag
   integer(kind=c_intptr_t) :: m, n, lda, ldb, ldc
   real(8), dimension(lda, *) :: a
   real(8), dimension(ldb, *) :: b
   real(8), dimension(ldc, *) :: c
   real(8) :: alpha

2.7.36. cublasXtDtrsm

DTRSM solves one of the matrix equations op( A )*X = alpha*B, or X*op( A ) = alpha*B, where alpha is a scalar, X and B are m by n matrices, A is a unit, or non-unit, upper or lower triangular matrix and op( A ) is one of op( A ) = A or op( A ) = A**T. The matrix X is overwritten on B. 

 integer(4) function cublasXtdtrsm(h, side, uplo, transa, diag, m, n, alpha, a, lda, b, ldb)
   type(cublasXtHandle) :: h
   integer :: side, uplo, transa, diag
   integer(kind=c_intptr_t) :: m, n, lda, ldb
   real(8), dimension(lda, *) :: a
   real(8), dimension(ldb, *) :: b
   real(8) :: alpha

2.7.37. cublasXtDspmm

DSPMM performs one of the symmetric packed matrix-matrix operations C := alpha*A*B + beta*C, or C := alpha*B*A + beta*C, where alpha and beta are scalars, A is a n by n symmetric matrix stored in packed format, and B and C are m by n matrices. 

 integer(4) function cublasXtdspmm(h, side, uplo, m, n, alpha, ap, b, ldb, beta, c, ldc)
   type(cublasXtHandle) :: h
   integer :: side, uplo
   integer(kind=c_intptr_t) :: m, n, ldb, ldc
   real(8), dimension(*) :: ap
   real(8), dimension(ldb, *) :: b
   real(8), dimension(ldc, *) :: c
   real(8) :: alpha, beta

2.7.38. cublasXtZgemm

ZGEMM performs one of the matrix-matrix operations C := alpha*op( A )*op( B ) + beta*C, where op( X ) is one of op( X ) = X or op( X ) = X**T or op( X ) = X**H, alpha and beta are scalars, and A, B and C are matrices, with op( A ) an m by k matrix, op( B ) a k by n matrix and C an m by n matrix. 

 integer(4) function cublasXtzgemm(h, transa, transb, m, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasXtHandle) :: h
   integer :: transa, transb
   integer(kind=c_intptr_t) :: m, n, k, lda, ldb, ldc
   complex(8), dimension(lda, *) :: a
   complex(8), dimension(ldb, *) :: b
   complex(8), dimension(ldc, *) :: c
   complex(8) :: alpha, beta

2.7.39. cublasXtZhemm

ZHEMM performs one of the matrix-matrix operations C := alpha*A*B + beta*C, or C := alpha*B*A + beta*C, where alpha and beta are scalars, A is an hermitian matrix and B and C are m by n matrices. 

 integer(4) function cublasXtzhemm(h, side, uplo, m, n, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasXtHandle) :: h
   integer :: side, uplo
   integer(kind=c_intptr_t) :: m, n, lda, ldb, ldc
   complex(8), dimension(lda, *) :: a
   complex(8), dimension(ldb, *) :: b
   complex(8), dimension(ldc, *) :: c
   complex(8) :: alpha, beta

2.7.40. cublasXtZherk

ZHERK performs one of the hermitian rank k operations C := alpha*A*A**H + beta*C, or C := alpha*A**H*A + beta*C, where alpha and beta are real scalars, C is an n by n hermitian matrix and A is an n by k matrix in the first case and a k by n matrix in the second case. 

 integer(4) function cublasXtzherk(h, uplo, trans, n, k, alpha, a, lda, beta, c, ldc)
   type(cublasXtHandle) :: h
   integer :: uplo, trans
   integer(kind=c_intptr_t) :: n, k, lda, ldc
   complex(8), dimension(lda, *) :: a
   complex(8), dimension(ldc, *) :: c
   real(8) :: alpha, beta

2.7.41. cublasXtZher2k

ZHER2K performs one of the hermitian rank 2k operations C := alpha*A*B**H + conjg( alpha )*B*A**H + beta*C, or C := alpha*A**H*B + conjg( alpha )*B**H*A + beta*C, where alpha and beta are scalars with beta real, C is an n by n hermitian matrix and A and B are n by k matrices in the first case and k by n matrices in the second case. 

 integer(4) function cublasXtzher2k(h, uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasXtHandle) :: h
   integer :: uplo, trans
   integer(kind=c_intptr_t) :: n, k, lda, ldb, ldc
   complex(8), dimension(lda, *) :: a
   complex(8), dimension(ldb, *) :: b
   complex(8), dimension(ldc, *) :: c
   complex(8) :: alpha
   real(8) :: beta

2.7.42. cublasXtZherkx

ZHERKX performs a variation of the hermitian rank k operations C := alpha*A*B**H + beta*C, where alpha and beta are real scalars, C is an n by n hermitian matrix stored in lower or upper mode, and A and B are n by k matrices. See the CUBLAS documentation for more details. 

 integer(4) function cublasXtzherkx(h, uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasXtHandle) :: h
   integer :: uplo, trans
   integer(kind=c_intptr_t) :: n, k, lda, ldb, ldc
   complex(8), dimension(lda, *) :: a
   complex(8), dimension(ldb, *) :: b
   complex(8), dimension(ldc, *) :: c
   complex(8) :: alpha
   real(8) :: beta

2.7.43. cublasXtZsymm

ZSYMM performs one of the matrix-matrix operations C := alpha*A*B + beta*C, or C := alpha*B*A + beta*C, where alpha and beta are scalars, A is a symmetric matrix and B and C are m by n matrices. 

 integer(4) function cublasXtzsymm(h, side, uplo, m, n, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasXtHandle) :: h
   integer :: side, uplo
   integer(kind=c_intptr_t) :: m, n, lda, ldb, ldc
   complex(8), dimension(lda, *) :: a
   complex(8), dimension(ldb, *) :: b
   complex(8), dimension(ldc, *) :: c
   complex(8) :: alpha, beta

2.7.44. cublasXtZsyrk

ZSYRK performs one of the symmetric rank k operations C := alpha*A*A**T + beta*C, or C := alpha*A**T*A + beta*C, where alpha and beta are scalars, C is an n by n symmetric matrix and A is an n by k matrix in the first case and a k by n matrix in the second case. 

 integer(4) function cublasXtzsyrk(h, uplo, trans, n, k, alpha, a, lda, beta, c, ldc)
   type(cublasXtHandle) :: h
   integer :: uplo, trans
   integer(kind=c_intptr_t) :: n, k, lda, ldc
   complex(8), dimension(lda, *) :: a
   complex(8), dimension(ldc, *) :: c
   complex(8) :: alpha, beta

2.7.45. cublasXtZsyr2k

ZSYR2K performs one of the symmetric rank 2k operations C := alpha*A*B**T + alpha*B*A**T + beta*C, or C := alpha*A**T*B + alpha*B**T*A + beta*C, where alpha and beta are scalars, C is an n by n symmetric matrix and A and B are n by k matrices in the first case and k by n matrices in the second case. 

 integer(4) function cublasXtzsyr2k(h, uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasXtHandle) :: h
   integer :: uplo, trans
   integer(kind=c_intptr_t) :: n, k, lda, ldb, ldc
   complex(8), dimension(lda, *) :: a
   complex(8), dimension(ldb, *) :: b
   complex(8), dimension(ldc, *) :: c
   complex(8) :: alpha, beta

2.7.46. cublasXtZsyrkx

ZSYRKX performs a variation of the symmetric rank k update C := alpha*A*B**T + beta*C, where alpha and beta are scalars, C is an n by n symmetric matrix stored in lower or upper mode, and A and B are n by k matrices. This routine can be used when B is in such a way that the result is guaranteed to be symmetric. See the CUBLAS documentation for more details. 

 integer(4) function cublasXtzsyrkx(h, uplo, trans, n, k, alpha, a, lda, b, ldb, beta, c, ldc)
   type(cublasXtHandle) :: h
   integer :: uplo, trans
   integer(kind=c_intptr_t) :: n, k, lda, ldb, ldc
   complex(8), dimension(lda, *) :: a
   complex(8), dimension(ldb, *) :: b
   complex(8), dimension(ldc, *) :: c
   complex(8) :: alpha, beta

2.7.47. cublasXtZtrmm

ZTRMM performs one of the matrix-matrix operations B := alpha*op( A )*B, or B := alpha*B*op( A ) where alpha is a scalar, B is an m by n matrix, A is a unit, or non-unit, upper or lower triangular matrix and op( A ) is one of op( A ) = A or op( A ) = A**T or op( A ) = A**H. 

 integer(4) function cublasXtztrmm(h, side, uplo, transa, diag, m, n, alpha, a, lda, b, ldb, c, ldc)
   type(cublasXtHandle) :: h
   integer :: side, uplo, transa, diag
   integer(kind=c_intptr_t) :: m, n, lda, ldb, ldc
   complex(8), dimension(lda, *) :: a
   complex(8), dimension(ldb, *) :: b
   complex(8), dimension(ldc, *) :: c
   complex(8) :: alpha

2.7.48. cublasXtZtrsm

ZTRSM solves one of the matrix equations op( A )*X = alpha*B, or X*op( A ) = alpha*B, where alpha is a scalar, X and B are m by n matrices, A is a unit, or non-unit, upper or lower triangular matrix and op( A ) is one of op( A ) = A or op( A ) = A**T or op( A ) = A**H. The matrix X is overwritten on B. 

 integer(4) function cublasXtztrsm(h, side, uplo, transa, diag, m, n, alpha, a, lda, b, ldb)
   type(cublasXtHandle) :: h
   integer :: side, uplo, transa, diag
   integer(kind=c_intptr_t) :: m, n, lda, ldb
   complex(8), dimension(lda, *) :: a
   complex(8), dimension(ldb, *) :: b
   complex(8) :: alpha

2.7.49. cublasXtZspmm

ZSPMM performs one of the symmetric packed matrix-matrix operations C := alpha*A*B + beta*C, or C := alpha*B*A + beta*C, where alpha and beta are scalars, A is a n by n symmetric matrix stored in packed format, and B and C are m by n matrices. 

 integer(4) function cublasXtzspmm(h, side, uplo, m, n, alpha, ap, b, ldb, beta, c, ldc)
   type(cublasXtHandle) :: h
   integer :: side, uplo
   integer(kind=c_intptr_t) :: m, n, ldb, ldc
   complex(8), dimension(*) :: ap
   complex(8), dimension(ldb, *) :: b
   complex(8), dimension(ldc, *) :: c
   complex(8) :: alpha, beta

3. FFT Runtime Library APIs

This section describes the Fortran interfaces to the cuFFT library. The FFT functions are only accessible from host code. All of the runtime API routines are integer functions that return an error code; they return a value of CUFFT_SUCCESS if the call was successful, or another cuFFT status return value if there was an error.

Chapter 10 contains examples of accessing the cuFFT library routines from OpenACC and CUDA Fortran. In both cases, the interfaces to the library can be exposed by adding the line 
use cufft
to your program unit.
Beginning with our 21.9 release, we also support a cufftXt module, which provides interfaces to the multi-gpu support available in the cuFFT library. These interfaces can be used within any Fortran program by adding the line 
use cufftxt
to your program unit. The cufftXt interfaces are documented beginning in section 4 of this chapter.

Unless a specific kind is provided in the following interfaces, the plain integer type implies integer(4) and the plain real type implies real(4).

3.1. CUFFT Definitions and Helper Functions

This section contains definitions and data types used in the cuFFT library and interfaces to the cuFFT helper functions.

The cuFFT module contains the following constants and enumerations:

integer, parameter :: CUFFT_FORWARD = -1
integer, parameter :: CUFFT_INVERSE = 1
! CUFFT Status
enum, bind(C)
    enumerator :: CUFFT_SUCCESS        = 0
    enumerator :: CUFFT_INVALID_PLAN   = 1
    enumerator :: CUFFT_ALLOC_FAILED   = 2
    enumerator :: CUFFT_INVALID_TYPE   = 3
    enumerator :: CUFFT_INVALID_VALUE  = 4
    enumerator :: CUFFT_INTERNAL_ERROR = 5
    enumerator :: CUFFT_EXEC_FAILED    = 6
    enumerator :: CUFFT_SETUP_FAILED   = 7
    enumerator :: CUFFT_INVALID_SIZE   = 8
    enumerator :: CUFFT_UNALIGNED_DATA = 9
end enum
! CUFFT Transform Types
enum, bind(C)
    enumerator :: CUFFT_R2C = z'2a'     ! Real to Complex (interleaved)
    enumerator :: CUFFT_C2R = z'2c'     ! Complex (interleaved) to Real
    enumerator :: CUFFT_C2C = z'29'     ! Complex to Complex, interleaved
    enumerator :: CUFFT_D2Z = z'6a'     ! Double to Double-Complex
    enumerator :: CUFFT_Z2D = z'6c'     ! Double-Complex to Double
    enumerator :: CUFFT_Z2Z = z'69'     ! Double-Complex to Double-Complex
end enum
! CUFFT Data Layouts
enum, bind(C)
    enumerator :: CUFFT_COMPATIBILITY_NATIVE          = 0
    enumerator :: CUFFT_COMPATIBILITY_FFTW_PADDING    = 1
    enumerator :: CUFFT_COMPATIBILITY_FFTW_ASYMMETRIC = 2
    enumerator :: CUFFT_COMPATIBILITY_FFTW_ALL        = 3
end enum
integer, parameter :: CUFFT_COMPATIBILITY_DEFAULT = CUFFT_COMPATIBILITY_FFTW_PADDING

3.1.1. cufftSetCompatibilityMode

This function configures the layout of cuFFT output in FFTW-compatible modes. 

 integer(4) function cufftSetCompatibilityMode( plan, mode )
   integer :: plan
   integer :: mode

3.1.2. cufftSetStream

This function sets the stream to be used by the cuFFT library to execute its routines. 

 integer(4) function cufftSetStream(plan, stream)
   integer :: plan
   integer(kind=cuda_stream_kind()) :: stream

3.1.3. cufftGetVersion

This function returns the version number of cuFFT. 

 integer(4) function cufftGetVersion( version )
   integer :: version

3.1.4. cufftSetAutoAllocation

This function indicates that the caller intends to allocate and manage work areas for plans that have been generated. cuFFT default behavior is to allocate the work area at plan generation time. If cufftSetAutoAllocation() has been called with autoAllocate set to 0 prior to one of the cufftMakePlan*() calls, cuFFT does not allocate the work area. This is the preferred sequence for callers wishing to manage work area allocation. 

 integer(4) function cufftSetAutoAllocation(plan, autoAllocate)
   integer(4) :: plan, autoallocate

3.1.5. cufftSetWorkArea

This function overrides the work area pointer associated with a plan. If the work area was auto-allocated, cuFFT frees the auto-allocated space. The cufftExecute*() calls assume that the work area pointer is valid and that it points to a contiguous region in device memory that does not overlap with any other work area. If this is not the case, results are indeterminate. 

 integer(4) function cufftSetWorkArea(plan, workArea)
   integer(4) :: plan
   integer, device :: workArea(*)

3.1.6. cufftDestroy

This function frees all GPU resources associated with a cuFFT plan and destroys the internal plan data structure. 

 integer(4) function cufftDestroy( plan )
   integer :: plan

3.2. CUFFT Plans and Estimated Size Functions

This section contains functions from the cuFFT library used to create plans and estimate work buffer size.

3.2.1. cufftPlan1d

This function creates a 1D FFT plan configuration for a specified signal size and data type. Nx is the size of the transform; batch is the number of transforms of size nx. 

 integer(4) function cufftPlan1d(plan, nx, ffttype, batch)
   integer :: plan
   integer :: nx
   integer :: ffttype
   integer :: batch

3.2.2. cufftPlan2d

This function creates a 2D FFT plan configuration according to a specified signal size and data type. For a Fortran array(nx,ny), nx is the size of the of the 1st dimension in the transform, but the 2nd size argument to the function; ny is the size of the 2nd dimension, and the 1st size argument to the function. 

 integer(4) function cufftPlan2d( plan, ny, nx, ffttype )
   integer :: plan
   integer :: ny, nx
   integer :: ffttype

3.2.3. cufftPlan3d

This function creates a 3D FFT plan configuration according to a specified signal size and data type. For a Fortran array(nx,ny,nz), nx is the size of the of the 1st dimension in the transform, but the 3rd size argument to the function; nz is the size of the 3rd dimension, and the 1st size argument to the function. 

 integer(4) function cufftPlan3d( plan, nz, ny, nx, ffttype )
   integer :: plan
   integer :: nz, ny, nx
   integer :: ffttype

3.2.4. cufftPlanMany

This function creates an FFT plan configuration of dimension rank, with sizes specified in the array n. Batch is the number of transforms to configure. This function supports more complicated input and output data layouts using the arguments inembed, istride, idist, onembed, ostride, and odist. In the C function, if inembed and onembed are set to NULL, all other stride information is ignored. Fortran programmers can pass NULL when using the NVIDIA cufft module by setting an F90 pointer to null(), either through direct assignment, using c_f_pointer() with c_null_ptr as the first argument, or the nullify statement, then passing the nullified F90 pointer as the actual argument for the inembed and onembed dummies. 

 integer(4) function cufftPlanMany(plan, rank, n, inembed, istride, idist, onembed, ostride, odist, ffttype, batch )
   integer :: plan
   integer :: rank
   integer :: n
   integer :: inembed, onembed
   integer :: istride, idist, ostride, odist
   integer :: ffttype, batch

3.2.5. cufftCreate

This function creates an opaque handle for further cuFFT calls and allocates some small data structures on the host. In C, the handle type is currently typedef'ed to an int, so in Fortran we use an integer*4 to hold the plan. 

 integer(4) function cufftCreate(plan)
   integer(4) :: plan

3.2.6. cufftMakePlan1d

Following a call to cufftCreate(), this function creates a 1D FFT plan configuration for a specified signal size and data type. Nx is the size of the transform; batch is the number of transforms of size nx. If cufftXtSetGPUs was called prior to this call with multiple GPUs, then workSize is an array containing multiple sizes. The workSize values are in bytes. 

 integer(4) function cufftMakePlan1d(plan, nx, ffttype, batch, worksize)
   integer(4) :: plan
   integer(4) :: nx
   integer(4) :: ffttype
   integer(4) :: batch
   integer(kind=int_ptr_kind()) :: workSize(*)

3.2.7. cufftMakePlan2d

Following a call to cufftCreate(), this function creates a 2D FFT plan configuration according to a specified signal size and data type. For a Fortran array(nx,ny), nx is the size of the of the 1st dimension in the transform, but the 2nd size argument to the function; ny is the size of the 2nd dimension, and the 1st size argument to the function. If cufftXtSetGPUs was called prior to this call with multiple GPUs, then workSize is an array containing multiple sizes. The workSize values are in bytes. 

 integer(4) function cufftMakePlan2d(plan, ny, nx, ffttype, workSize)
   integer(4) :: plan
   integer(4) :: ny, nx
   integer(4) :: ffttype
   integer(kind=int_ptr_kind()) :: workSize(*)

3.2.8. cufftMakePlan3d

Following a call to cufftCreate(), this function creates a 3D FFT plan configuration according to a specified signal size and data type. For a Fortran array(nx,ny,nz), nx is the size of the of the 1st dimension in the transform, but the 3rd size argument to the function; nz is the size of the 3rd dimension, and the 1st size argument to the function. If cufftXtSetGPUs was called prior to this call with multiple GPUs, then workSize is an array containing multiple sizes. The workSize values are in bytes. 

 integer(4) function cufftMakePlan3d(plan, nz, ny, nx, ffttype, workSize)
   integer(4) :: plan
   integer(4) :: nz, ny, nx
   integer(4) :: ffttype
   integer(kind=int_ptr_kind()) :: workSize(*)

3.2.9. cufftMakePlanMany

Following a call to cufftCreate(), this function creates an FFT plan configuration of dimension rank, with sizes specified in the array n. Batch is the number of transforms to configure. This function supports more complicated input and output data layouts using the arguments inembed, istride, idist, onembed, ostride, and odist.

In the C function, if inembed and onembed are set to NULL, all other stride information is ignored. Fortran programmers can pass NULL when using the NVIDIA cufft module by setting an F90 pointer to null(), either through direct assignment, using c_f_pointer() with c_null_ptr as the first argument, or the nullify statement, then passing the nullified F90 pointer as the actual argument for the inembed and onembed dummies.

If cufftXtSetGPUs was called prior to this call with multiple GPUs, then workSize is an array containing multiple sizes. The workSize values are in bytes. 

 integer(4) function cufftMakePlanMany(plan, rank, n, inembed, istride, idist, onembed, ostride, odist, ffttype, batch, workSize)
   integer(4) :: plan
   integer(4) :: rank
   integer :: n(rank)
   integer :: inembed(rank), onembed(rank)
   integer(4) :: istride, idist, ostride, odist
   integer(4) :: ffttype, batch
   integer(kind=int_ptr_kind()) :: workSize(*)

3.2.10. cufftEstimate1d

This function returns an estimate for the size of the work area required, in bytes, given the specified size and data type, and assuming default plan settings. 

 integer(4) function cufftEstimate1d(nx, ffttype, batch, workSize)
   integer(4) :: nx
   integer(4) :: ffttype
   integer(4) :: batch
   integer(kind=int_ptr_kind()) :: workSize(*)

3.2.11. cufftEstimate2d

This function returns an estimate for the size of the work area required, in bytes, given the specified size and data type, and assuming default plan settings. 

 integer(4) function cufftEstimate2d(ny, nx, ffttype, workSize)
   integer(4) :: ny, nx
   integer(4) :: ffttype
   integer(kind=int_ptr_kind()) :: workSize(*)

3.2.12. cufftEstimate3d

This function returns an estimate for the size of the work area required, in bytes, given the specified size and data type, and assuming default plan settings. 

 integer(4) function cufftEstimate3d(nz, ny, nx, ffttype, workSize)
   integer(4) :: nz, ny, nx
   integer(4) :: ffttype
   integer(kind=int_ptr_kind()) :: workSize(*)

3.2.13. cufftEstimateMany

This function returns an estimate for the size of the work area required, in bytes, given the specified size and data type, and assuming default plan settings. 

 integer(4) function cufftEstimateMany(rank, n, inembed, istride, idist, onembed, ostride, odist, ffttype, batch, workSize)
   integer(4) :: rank, istride, idist, ostride, odist
   integer(4), dimension(rank) :: n, inembed, onembed
   integer(4) :: ffttype
   integer(4) :: batch
   integer(kind=int_ptr_kind()) :: workSize(*)

3.2.14. cufftGetSize1d

This function gives a more accurate estimate than cufftEstimate1d() of the size of the work area required, in bytes, given the specified plan parameters and taking into account any plan settings which may have been made. 

 integer(4) function cufftGetSize1d(plan, nx, ffttype, batch, workSize)
   integer(4) :: plan, nx, ffttype, batch
   integer(kind=int_ptr_kind()) :: workSize(*)

3.2.15. cufftGetSize2d

This function gives a more accurate estimate than cufftEstimate2d() of the size of the work area required, in bytes, given the specified plan parameters and taking into account any plan settings which may have been made. 

 integer(4) function cufftGetSize2d(plan, ny, nx, ffttype, workSize)
   integer(4) :: plan, ny, nx, ffttype
   integer(kind=int_ptr_kind()) :: workSize(*)

3.2.16. cufftGetSize3d

This function gives a more accurate estimate than cufftEstimate3d() of the size of the work area required, in bytes, given the specified plan parameters and taking into account any plan settings which may have been made. 

 integer(4) function cufftGetSize3d(plan, nz, ny, nx, ffttype, workSize)
   integer(4) :: plan, nz, ny, nx, ffttype
   integer(kind=int_ptr_kind()) :: workSize(*)

3.2.17. cufftGetSizeMany

This function gives a more accurate estimate than cufftEstimateMany() of the size of the work area required, in bytes, given the specified plan parameters and taking into account any plan settings which may have been made. 

 integer(4) function cufftGetSizeMany(plan, rank, n, inembed, istride, idist, onembed, ostride, odist, ffttype, batch, workSize)
   integer(4) :: plan, rank, istride, idist, ostride, odist
   integer(4), dimension(rank) :: n, inembed, onembed
   integer(4) :: ffttype
   integer(4) :: batch
   integer(kind=int_ptr_kind()) :: workSize(*)

3.2.18. cufftGetSize

Once plan generation has been done, either with the original API or the extensible API, this call returns the actual size of the work area required, in bytes, to support the plan. Callers who choose to manage work area allocation within their application must use this call after plan generation, and after any cufftSet*() calls subsequent to plan generation, if those calls might alter the required work space size. 

 integer(4) function cufftGetSize(plan, workSize)
   integer(4) :: plan
   integer(kind=int_ptr_kind()) :: workSize(*)

3.3. CUFFT Execution Functions

This section contains the execution functions, which perform the actual Fourier transform, in the cuFFT library.

3.3.1. cufftExecC2C

This function executes a single precision complex-to-complex transform plan in the transform direction as specified by the direction parameter. If idata and odata are the same, this function does an in-place transform. 

 integer(4) function cufftExecC2C( plan, idata, odata, direction )
   integer :: plan
   complex(4), device, dimension(*) :: idata, odata
   integer :: direction

3.3.2. cufftExecR2C

This function executes a single precision real-to-complex, implicity forward, cuFFT transform plan. If idata and odata are the same, this function does an in-place transform, but note there are data layout differences between in-place and out-of-place transforms for real-to- complex FFTs in cuFFT. 

 integer(4) function cufftExecR2C( plan, idata, odata )
   integer :: plan
   real(4), device, dimension(*) :: idata
   complex(4), device, dimension(*) :: odata

3.3.3. cufftExecC2R

This function executes a single precision complex-to-real, implicity inverse, cuFFT transform plan. If idata and odata are the same, this function does an in-place transform. 

 integer(4) function cufftExecC2R( plan, idata, odata )
   integer :: plan
   complex(4), device, dimension(*) :: idata
   real(4), device, dimension(*) :: odata

3.3.4. cufftExecZ2Z

This function executes a double precision complex-to-complex transform plan in the transform direction as specified by the direction parameter. If idata and odata are the same, this function does an in-place transform. 

 integer(4) function cufftExecZ2Z( plan, idata, odata, direction )
   integer :: plan
   complex(8), device, dimension(*) :: idata, odata
   integer :: direction

3.3.5. cufftExecD2Z

This function executes a double precision real-to-complex, implicity forward, cuFFT transform plan. If idata and odata are the same, this function does an in-place transform, but note there are data layout differences between in-place and out-of-place transforms for real-to- complex FFTs in cuFFT. 

 integer(4) function cufftExecD2Z( plan, idata, odata )
   integer :: plan
   real(8), device, dimension(*) :: idata
   complex(8), device, dimension(*) :: odata

3.3.6. cufftExecZ2D

This function executes a double precision complex-to-real, implicity inverse, cuFFT transform plan. If idata and odata are the same, this function does an in-place transform. 

 integer(4) function cufftExecZ2D( plan, idata, odata )
   integer :: plan
   complex(8), device, dimension(*) :: idata
   real(8), device, dimension(*) :: odata

3.4. CUFFTXT Definitions and Helper Functions

This section contains definitions and data types used in the cufftXt library and interfaces to helper functions.

The cufftXt module contains the following constants and enumerations:

integer, parameter :: MAX_CUDA_DESCRIPTOR_GPUS = 64
! libFormat enum is used for the library member of cudaLibXtDesc
enum, bind(C)
    enumerator :: LIB_FORMAT_CUFFT     = 0
    enumerator :: LIB_FORMAT_UNDEFINED = 1
end enum
! cufftXtSubFormat identifies the data layout of a memory descriptor
enum, bind(C)
    ! by default input is in linear order across GPUs
    enumerator :: CUFFT_XT_FORMAT_INPUT = 0

    ! by default output is in scrambled order depending on transform
    enumerator :: CUFFT_XT_FORMAT_OUTPUT = 1

    ! by default inplace is input order, which is linear across GPUs
    enumerator :: CUFFT_XT_FORMAT_INPLACE = 2

    ! shuffled output order after execution of the transform
    enumerator :: CUFFT_XT_FORMAT_INPLACE_SHUFFLED = 3

    ! shuffled input order prior to execution of 1D transforms
    enumerator :: CUFFT_XT_FORMAT_1D_INPUT_SHUFFLED = 4

    enumerator :: CUFFT_FORMAT_UNDEFINED = 5
end enum
! cufftXtCopyType specifies the type of copy for cufftXtMemcpy
enum, bind(C)
    enumerator :: CUFFT_COPY_HOST_TO_DEVICE   = 0
    enumerator :: CUFFT_COPY_DEVICE_TO_HOST   = 1
    enumerator :: CUFFT_COPY_DEVICE_TO_DEVICE = 2
    enumerator :: CUFFT_COPY_UNDEFINED        = 3
end enum
! cufftXtQueryType specifies the type of query for cufftXtQueryPlan
enum, bind(c)
    enumerator :: CUFFT_QUERY_1D_FACTORS = 0
    enumerator :: CUFFT_QUERY_UNDEFINED  = 1
end enum
! cufftXtWorkAreaPolicy specifies the policy for cufftXtSetWorkAreaPolicy
enum, bind(c)
    enumerator :: CUFFT_WORKAREA_MINIMAL     = 0 ! maximum reduction
    enumerator :: CUFFT_WORKAREA_USER        = 1 ! use workSize parameter as limit
    enumerator :: CUFFT_WORKAREA_PERFORMANCE = 2 ! default - 1x overhead or more, max perf
end enum

The cufftXt module contains the following derived type definitions:

! cufftXt1dFactors type
type, bind(c) :: cufftXt1dFactors
    integer(8) :: size
    integer(8) :: stringCount
    integer(8) :: stringLength
    integer(8) :: subStringLength
    integer(8) :: factor1
    integer(8) :: factor2
    integer(8) :: stringMask
    integer(8) :: subStringMask
    integer(8) :: factor1Mask
    integer(8) :: factor2Mask
    integer(4) :: stringShift
    integer(4) :: subStringShift
    integer(4) :: factor1Shift
    integer(4) :: factor2Shift
end type cufftXt1dFactors
type, bind(C) :: cudaXtDesc
    integer(4) :: version
    integer(4) :: nGPUs
    integer(4) :: GPUs(MAX_CUDA_DESCRIPTOR_GPUS)
    type(c_devptr) :: data(MAX_CUDA_DESCRIPTOR_GPUS)
    integer(8) :: size(MAX_CUDA_DESCRIPTOR_GPUS)
    type(c_ptr) :: cudaXtState
end type cudaXtDesc
type, bind(C) :: cudaLibXtDesc
    integer(4) :: version
    type(c_ptr) :: descriptor     ! cudaXtDesc *descriptor
    integer(4) :: library         ! libFormat library
    integer(4) :: subFormat
    type(c_ptr) :: libDescriptor  ! void *libDescriptor
end type cudaLibXtDesc

3.4.1. cufftXtSetGPUs

This function identifies which GPUs are to be used with the plan. The call to cufftXtSetGPUs must occur after the call to cufftCreate but before the call to cufftMakePlan*. 

 integer(4) function cufftXtSetGPUs( plan, nGPUs, whichGPUs )
   integer(4) :: plan
   integer(4) :: nGPUs
   integer(4) :: whichGPUs(*)

3.4.2. cufftXtMalloc

This function allocates a cufftXt descriptor, and memory for data in the GPUs associated with the plan. The value of cufftXtSubFormat determines if the buffer will be used for input or output. Fortran programmers should declare and pass a pointer to a type(cudaLibXtDesc) variable so the entire information can be stored, and also freed in subsequent calls to cufftXtFree. For programmers comfortable with the C interface, a variant of this function can take a type(c_ptr) for the 2nd argument. 

 integer(4) function cufftXtMalloc( plan, descriptor, format )
   integer(4) :: plan
   type(cudaLibXtDesc), pointer :: descriptor  ! A type(c_ptr) is also accepted.
   integer(4) :: format ! cufftXtSubFormat value

3.4.3. cufftXtFree

This function frees the cufftXt descriptor, and all memory associated with it. The descriptor and memory must have been allocated by a previous call to cufftXtMalloc. Fortran programmers should declare and pass a pointer to a type(cudaLibXtDesc) variable. For programmers comfortable with the C interface, a variant of this function can take a type(c_ptr) as the only argument. 

 integer(4) function cufftXtFree( descriptor )
   type(cudaLibXtDesc), pointer :: descriptor  ! A type(c_ptr) is also accepted.

3.4.4. cufftXtMemcpy

This function copies data between buffers on the host and GPUs, or between GPUs. The value of the type argument determines the copy direction. In addition, this Fortran function is overloaded to take a type(cudaLibXtDesc) variable for the destination (H2D transfer), for the source (D2H transfer), or for both (D2D transfer), in which case the type argument is not required. 

 integer(4) function cufftXtMemcpy( plan, dst, src, type )
   integer(4) :: plan
   type(cudaLibXtDesc) :: dst  ! Or any host buffer, depending on the type
   type(cudaLibXtDesc) :: src  ! Or any host buffer, depending on the type
   integer(4) :: type          ! optional cufftXtCopyType value

3.5. CUFFTXT Plans and Work Area Functions

This section contains functions from the cufftXt library used to create plans and manage work buffers.

3.5.1. cufftXtMakePlanMany

Following a call to cufftCreate(), this function creates an FFT plan configuration of dimension rank, with sizes specified in the array n. Batch is the number of transforms to configure. This function supports more complicated input and output data layouts using the arguments inembed, istride, idist, onembed, ostride, and odist. In the C function, if inembed and onembed are set to NULL, all other stride information is ignored. Fortran programmers can pass NULL when using the NVIDIA cufft module by setting an F90 pointer to null(), either through direct assignment, using c_f_pointer() with c_null_ptr as the first argument, or the nullify statement, then passing the nullified F90 pointer as the actual argument for the inembed and onembed dummies. 

 integer(4) function cufftXtMakePlanMany(plan, rank, n, inembed, istride, &
     idist, inputType, onembed, ostride, odist, outputType, batch, workSize, &
     executionType)
   integer(4) :: plan
   integer(4) :: rank
   integer(8) :: n(*)
   integer(8) :: inembed(*), onembed(*)
   integer(8) :: istride, idist, ostride, odist
   type(cudaDataType) :: inputType, outputType, executionType
   integer(4) :: batch
   integer(8) :: workSize(*)

3.5.2. cufftXtQueryPlan

This function only supports multi-gpu 1D transforms. It returns a derived type, factors, which contains the number of strings, the decomposition of factors, and (in the case of power of 2 sizes) some other useful mask and shift elements, used in converting between permuted and linear indexes. 

 integer(4) function cufftXtQueryPlan(plan, factors, queryType)
   integer(4) :: plan
   type(cufftXt1DFactors) :: factors
   integer(4) :: queryType

3.5.3. cufftXtSetWorkAreaPolicy

This function overrides the work area associated with a plan. Currently, the workAreaPolicy can be specified as CUFFT_WORKAREA_MINIMAL and cuFFT will attempt to re-plan to use zero bytes of work area memory. See the CUFFT documentation for support of other features. 

 integer(4) function cufftXtSetWorkAreaPolicy(plan, workAreaPolicy, workSize)
   integer(4) :: plan
   integer(4) :: workAreaPolicy
   integer(8) :: workSize

3.5.4. cufftXtGetSizeMany

This function gives a more accurate estimate than cufftEstimateMany() of the size of the work area required, in bytes, given the specified plan parameters used for cufftXtMakePlanMany and taking into account any plan settings which may have been made. 

 integer(4) function cufftXtGetSizeMany(plan, rank, n, inembed, istride, &
     idist, inputType, onembed, ostride, odist, outputType, batch, workSize, &
     executionType)
   integer(4) :: plan
   integer(4) :: rank
   integer(8) :: n(*)
   integer(8) :: inembed(*), onembed(*)
   integer(8) :: istride, idist, ostride, odist
   type(cudaDataType) :: inputType, outputType, executionType
   integer(4) :: batch
   integer(8) :: workSize(*)

3.5.5. cufftXtSetWorkArea

This function overrides the work areas associated with a plan. If the work area was auto-allocated, cuFFT frees the auto-allocated space. The cufftExecute*() calls assume that the work area pointer is valid and that it points to a contiguous region in device memory that does not overlap with any other work area. If this is not the case, results are indeterminate. 

 integer(4) function cufftXtSetWorkArea(plan, workArea)
   integer(4) :: plan
   type(c_ptr) :: workArea

3.6. CUFFTXT Execution Functions

This section contains the execution functions, which perform the actual Fourier transform, in the cufftXt library.

3.6.1. cufftXtExec

This function executes any Fourier transform regardless of precision and type. In case of complex-to-real and real-to-complex transforms, the direction argument is ignored. Otherwise, the transform direction is specified by the direction parameter. This function uses the GPU memory pointed to by input as input data, and stores the computed Fourier coefficients in the output array. If those are the same, this method does an in-place transform. Any valid data type for the input and output arrays are accepted. 

 integer(4) function cufftXtExec( plan, input, output, direction )
   integer :: plan
   real, dimension(*) :: input, output  ! Any data type is allowed
   integer :: direction

3.6.2. cufftXtExecDescriptor

This function executes any Fourier transform regardless of precision and type. In case of complex-to-real and real-to-complex transforms, the direction argument is ignored. Otherwise, the transform direction is specified by the direction parameter. This function stores the result in the specified output arrays. 

 integer(4) function cufftXtExecDescriptor( plan, input, output, direction )
   integer :: plan
   type(cudaLibXtDesc) :: input, output
   integer :: direction

3.6.3. cufftXtExecDescriptorC2C

This function executes a single precision complex-to-complex transform plan in the transform direction as specified by the direction parameter. This multiple GPU function currently supports in-place transforms only; the result will be stored in the input arrays. 

 integer(4) function cufftXtExecDescriptorC2C( plan, input, output, direction )
   integer :: plan
   type(cudaLibXtDesc) :: input, output
   integer :: direction

3.6.4. cufftXtExecDescriptorZ2Z

This function executes a double precision complex-to-complex transform plan in the transform direction as specified by the direction parameter. This multiple GPU function currently supports in-place transforms only; the result will be stored in the input arrays. 

 integer(4) function cufftXtExecDescriptorZ2Z( plan, input, output, direction )
   integer :: plan
   type(cudaLibXtDesc) :: input, output
   integer :: direction

3.6.5. cufftXtExecDescriptorR2C

This function executes a single precision real-to-complex transform plan. This multiple GPU function currently supports in-place transforms only; the result will be stored in the input arrays. 

 integer(4) function cufftXtExecDescriptorR2C( plan, input, output )
   integer :: plan
   type(cudaLibXtDesc) :: input, output

3.6.6. cufftXtExecDescriptorD2Z

This function executes a double precision real-to-complex transform plan. This multiple GPU function currently supports in-place transforms only; the result will be stored in the input arrays. 

 integer(4) function cufftXtExecDescriptorD2Z( plan, input, output )
   integer :: plan
   type(cudaLibXtDesc) :: input, output

3.6.7. cufftXtExecDescriptorC2R

This function executes a single precision complex-to-real transform plan. This multiple GPU function currently supports in-place transforms only; the result will be stored in the input arrays. 

 integer(4) function cufftXtExecDescriptorC2R( plan, input, output )
   integer :: plan
   type(cudaLibXtDesc) :: input, output

3.6.8. cufftXtExecDescriptorZ2D

This function executes a double precision complex-to-real transform plan. This multiple GPU function currently supports in-place transforms only; the result will be stored in the input arrays. 

 integer(4) function cufftXtExecDescriptorZ2D( plan, input, output )
   integer :: plan
   type(cudaLibXtDesc) :: input, output

4. Random Number Runtime Library APIs

This section describes the Fortran interfaces to the CUDA cuRAND library. The cuRAND functionality is accessible from both host and device code. In the host library, all of the runtime API routines are integer functions that return an error code; they return a value of CURAND_STATUS_SUCCESS if the call was successful, or other cuRAND return status value if there was an error. The host library routines are meant to produce a series or array of random numbers. In the device library, the init routines are subroutines and the generator functions return the type of the value being generated. The device library routines are meant for producing a single value per thread per call.

Chapter 10 contains examples of accessing the cuRAND library routines from OpenACC and CUDA Fortran. In both cases, the interfaces to the library can be exposed in host code by adding the line 
use curand
to your program unit.

Unless a specific kind is provided, the plain integer type implies integer(4) and the plain real type implies real(4).

4.1. CURAND Definitions and Helper Functions

This section contains definitions and data types used in the cuRAND library and interfaces to the cuRAND helper functions.

The curand module contains the following derived type definitions:

TYPE curandGenerator
  TYPE(C_PTR)  :: handle
END TYPE

The curand module contains the following enumerations:

! CURAND Status
enum, bind(c)
    enumerator :: CURAND_STATUS_SUCCESS                   = 0
    enumerator :: CURAND_STATUS_VERSION_MISMATCH          = 100
    enumerator :: CURAND_STATUS_NOT_INITIALIZED           = 101
    enumerator :: CURAND_STATUS_ALLOCATION_FAILED         = 102
    enumerator :: CURAND_STATUS_TYPE_ERROR                = 103
    enumerator :: CURAND_STATUS_OUT_OF_RANGE              = 104
    enumerator :: CURAND_STATUS_LENGTH_NOT_MULTIPLE       = 105
    enumerator :: CURAND_STATUS_DOUBLE_PRECISION_REQUIRED = 106
    enumerator :: CURAND_STATUS_LAUNCH_FAILURE            = 201
    enumerator :: CURAND_STATUS_PREEXISTING_FAILURE       = 202
    enumerator :: CURAND_STATUS_INITIALIZATION_FAILED     = 203
    enumerator :: CURAND_STATUS_ARCH_MISMATCH             = 204
    enumerator :: CURAND_STATUS_INTERNAL_ERROR            = 999
end enum
! CURAND Generator Types
enum, bind(c)
    enumerator :: CURAND_RNG_TEST                    = 0
    enumerator :: CURAND_RNG_PSEUDO_DEFAULT          = 100
    enumerator :: CURAND_RNG_PSEUDO_XORWOW           = 101
    enumerator :: CURAND_RNG_PSEUDO_MRG32K3A         = 121
    enumerator :: CURAND_RNG_PSEUDO_MTGP32           = 141
    enumerator :: CURAND_RNG_PSEUDO_MT19937          = 142
    enumerator :: CURAND_RNG_PSEUDO_PHILOX4_32_10    = 161
    enumerator :: CURAND_RNG_QUASI_DEFAULT           = 200
    enumerator :: CURAND_RNG_QUASI_SOBOL32           = 201
    enumerator :: CURAND_RNG_QUASI_SCRAMBLED_SOBOL32 = 202
    enumerator :: CURAND_RNG_QUASI_SOBOL64           = 203
    enumerator :: CURAND_RNG_QUASI_SCRAMBLED_SOBOL64 = 204
end enum
! CURAND Memory Ordering
enum, bind(c)
    enumerator :: CURAND_ORDERING_PSEUDO_BEST    = 100
    enumerator :: CURAND_ORDERING_PSEUDO_DEFAULT = 101
    enumerator :: CURAND_ORDERING_PSEUDO_SEEDED  = 102
    enumerator :: CURAND_ORDERING_QUASI_DEFAULT  = 201
end enum
! CURAND Direction Vectors
enum, bind(c)
    enumerator :: CURAND_DIRECTION_VECTORS_32_JOEKUO6           = 101
    enumerator :: CURAND_SCRAMBLED_DIRECTION_VECTORS_32_JOEKUO6 = 102
    enumerator :: CURAND_DIRECTION_VECTORS_64_JOEKUO6           = 103
    enumerator :: CURAND_SCRAMBLED_DIRECTION_VECTORS_64_JOEKUO6 = 104
end enum
! CURAND Methods
enum, bind(c)
    enumerator :: CURAND_CHOOSE_BEST    = 0
    enumerator :: CURAND_ITR            = 1
    enumerator :: CURAND_KNUTH          = 2
    enumerator :: CURAND_HITR           = 3
    enumerator :: CURAND_M1             = 4
    enumerator :: CURAND_M2             = 5
    enumerator :: CURAND_BINARY_SEARCH  = 6
    enumerator :: CURAND_DISCRETE_GAUSS = 7
    enumerator :: CURAND_REJECTION      = 8
    enumerator :: CURAND_DEVICE_API     = 9
    enumerator :: CURAND_FAST_REJECTION = 10
    enumerator :: CURAND_3RD            = 11
    enumerator :: CURAND_DEFINITION     = 12
    enumerator :: CURAND_POISSON        = 13
end enum

4.1.1. curandCreateGenerator

This function creates a new random number generator of type rng. See the beginning of this section for valid values of rng. 

 integer(4) function curandCreateGenerator(generator, rng)
   type(curandGenerator) :: generator
   integer :: rng

4.1.2. curandCreateGeneratorHost

This function creates a new host CPU random number generator of type rng. See the beginning of this section for valid values of rng. 

 integer(4) function curandCreateGeneratorHost(generator, rng)
   type(curandGenerator) :: generator
   integer :: rng

4.1.3. curandDestroyGenerator

This function destroys an existing random number generator. 

 integer(4) function curandDestroyGenerator(generator)
   type(curandGenerator) :: generator

4.1.4. curandGetVersion

This function returns the version number of the cuRAND library. 

 integer(4) function curandGetVersion(version)
   integer(4) :: version

4.1.5. curandSetStream

This function sets the current stream for the cuRAND kernel launches. 

 integer(4) function curandSetStream(generator, stream)
   type(curandGenerator) :: generator
   integer(kind=c_intptr_t) :: stream

4.1.6. curandSetPseudoRandomGeneratorSeed

This function sets the seed value of the pseudo-random number generator. 

 integer(4) function curandSetPseudoRandomGeneratorSeed(generator, seed)
   type(curandGenerator) :: generator
   integer(8) :: seed

4.1.7. curandSetGeneratorOffset

This function sets the absolute offset of the pseudo or quasirandom number generator. 

 integer(4) function curandSetGeneratorOffset(generator, offset)
   type(curandGenerator) :: generator
   integer(8) :: offset

4.1.8. curandSetGeneratorOrdering

This function sets the ordering of results of the pseudo or quasirandom number generator. 

 integer(4) function curandSetGeneratorOrdering(generator, order)
   type(curandGenerator) :: generator
   integer(4) :: order

4.1.9. curandSetQuasiRandomGeneratorDimensions

This function sets number of dimensions of the quasirandom number generator. 

 integer(4) function curandSetQuasiRandomGeneratorDimensions(generator, num)
   type(curandGenerator) :: generator
   integer(4) :: num

4.2. CURAND Generator Functions

This section contains interfaces for the cuRAND generator functions.

4.2.1. curandGenerate

This function generates 32-bit pseudo or quasirandom numbers. 

 integer(4) function curandGenerate(generator, array, num )
   type(curandGenerator) :: generator
   integer(4), device :: array(*) ! Host or device depending on the generator
   integer(kind=c_intptr_t) :: num

4.2.2. curandGenerateLongLong

This function generates 64-bit integer quasirandom numbers. The function curandGenerate() has also been overloaded to accept these function arguments. 

 integer(4) function curandGenerateLongLong(generator, array, num )
   type(curandGenerator) :: generator
   integer(8), device :: array(*) ! Host or device depending on the generator
   integer(kind=c_intptr_t) :: num

4.2.3. curandGenerateUniform

This function generates 32-bit floating point uniformly distributed random numbers. The function curandGenerate() has also been overloaded to accept these function arguments. 

 integer(4) function curandGenerateUniform(generator, array, num )
   type(curandGenerator) :: generator
   real(4), device :: array(*) ! Host or device depending on the generator
   integer(kind=c_intptr_t) :: num

4.2.4. curandGenerateUniformDouble

This function generates 64-bit floating point uniformly distributed random numbers. The function curandGenerate() has also been overloaded to accept these function arguments. 

 integer(4) function curandGenerateUniformDouble(generator, array, num )
   type(curandGenerator) :: generator
   real(8), device :: array(*) ! Host or device depending on the generator
   integer(kind=c_intptr_t) :: num

4.2.5. curandGenerateNormal

This function generates 32-bit floating point normally distributed random numbers. The function curandGenerate() has also been overloaded to accept these function arguments. 

 integer(4) function curandGenerateNormal(generator, array, num, mean, stddev )
   type(curandGenerator) :: generator
   real(4), device :: array(*) ! Host or device depending on the generator
   integer(kind=c_intptr_t) :: num
   real(4) :: mean, stddev

4.2.6. curandGenerateNormalDouble

This function generates 64-bit floating point normally distributed random numbers. The function curandGenerate() has also been overloaded to accept these function arguments. 

 integer(4) function curandGenerateNormalDouble(generator, array, num, mean, stddev )
   type(curandGenerator) :: generator
   real(8), device :: array(*) ! Host or device depending on the generator
   integer(kind=c_intptr_t) :: num
   real(8) :: mean, stddev

4.2.7. curandGeneratePoisson

This function generates Poisson-distributed random numbers. The function curandGenerate() has also been overloaded to accept these function arguments. 

 integer(4) function curandGeneratePoisson(generator, array, num, lambda )
   type(curandGenerator) :: generator
   real(8), device :: array(*) ! Host or device depending on the generator
   integer(kind=c_intptr_t) :: num
   real(8) :: lambda

4.2.8. curandGenerateSeeds

This function sets the starting state of the generator. 

 integer(4) function curandGenerateSeeds(generator)
   type(curandGenerator) :: generator

4.2.9. curandGenerateLogNormal

This function generates 32-bit floating point log-normally distributed random numbers. 

 integer(4) function curandGenerateLogNormal(generator, array, num, mean, stddev )
   type(curandGenerator) :: generator
   real(4), device :: array(*) ! Host or device depending on the generator
   integer(kind=c_intptr_t) :: num
   real(4) :: mean, stddev

4.2.10. curandGenerateLogNormalDouble

This function generates 64-bit floating point log-normally distributed random numbers. 

 integer(4) function curandGenerateLogNormalDouble(generator, array, num, mean, stddev )
   type(curandGenerator) :: generator
   real(8), device :: array(*) ! Host or device depending on the generator
   integer(kind=c_intptr_t) :: num
   real(8) :: mean, stddev

4.3. CURAND Device Definitions and Functions

This section contains definitions and data types used in the cuRAND device library and interfaces to the cuRAND functions.

The curand device module contains the following derived type definitions:

TYPE curandStateXORWOW
    integer(4) :: d
    integer(4) :: v(5)
    integer(4) :: boxmuller_flag
    integer(4) :: boxmuller_flag_double
    real(4)    :: boxmuller_extra
    real(8)    :: boxmuller_extra_double
END TYPE curandStateXORWOW
TYPE curandStateMRG32k3a
    real(8)    :: s1(3)
    real(8)    :: s2(3)
    integer(4) :: boxmuller_flag
    integer(4) :: boxmuller_flag_double
    real(4)    :: boxmuller_extra
    real(8)    :: boxmuller_extra_double
END TYPE curandStateMRG32k3a
TYPE curandStateSobol32
    integer(4) :: d
    integer(4) :: x
    integer(4) :: c
    integer(4) :: direction_vectors(32)
END TYPE curandStateSobol32
TYPE curandStateScrambledSobol32
    integer(4) :: d
    integer(4) :: x
    integer(4) :: c
    integer(4) :: direction_vectors(32)
END TYPE curandStateScrambledSobol32
TYPE curandStateSobol64
    integer(8) :: d
    integer(8) :: x
    integer(8) :: c
    integer(8) :: direction_vectors(32)
END TYPE curandStateSobol64
TYPE curandStateScrambledSobol64
    integer(8) :: d
    integer(8) :: x
    integer(8) :: c
    integer(8) :: direction_vectors(32)
END TYPE curandStateScrambledSobol64
TYPE curandStateMtgp32
    integer(4) :: s(MTGP32_STATE_SIZE)
    integer(4) :: offset
    integer(4) :: pIdx
    integer(kind=int_ptr_kind()) :: k
    integer(4) :: precise_double_flag
END TYPE curandStateMtgp32
TYPE curandStatePhilox4_32_10
    integer(4) :: ctr
    integer(4) :: output
    integer(2) :: key
    integer(4) :: state
    integer(4) :: boxmuller_flag
    integer(4) :: boxmuller_flag_double
    real(4)    :: boxmuller_extra
    real(8)    :: boxmuller_extra_double
END TYPE curandStatePhilox4_32_10

4.3.1. curand_Init

This overloaded device subroutine initializes the state for the random number generator. These device subroutines are declared "attributes(device)" in CUDA Fortran and "!$acc routine() seq" in OpenACC.

4.3.1.1. curandInitXORWOW

This function initializes the state for the XORWOW random number generator. The function curand_init() has also been overloaded to accept these function arguments, as in CUDA C++. Device Functions are declared "attributes(device)" in CUDA Fortran and "!$acc routine() seq" in OpenACC. 

 subroutine curandInitXORWOW(seed, sequence, offset, state)
   integer(8) :: seed
   integer(8) :: sequence
   integer(8) :: offset
   TYPE(curandStateXORWOW) :: state

4.3.1.2. curandInitMRG32k3a

This function initializes the state for the MRG32k3a random number generator. The function curand_init() has also been overloaded to accept these function arguments, as in CUDA C++. Device Functions are declared "attributes(device)" in CUDA Fortran and "!$acc routine() seq" in OpenACC. 

 subroutine curandInitMRG32k3a(seed, sequence, offset, state)
   integer(8) :: seed
   integer(8) :: sequence
   integer(8) :: offset
   TYPE(curandStateMRG32k3a) :: state

4.3.1.3. curandInitPhilox4_32_10

This function initializes the state for the Philox4_32_10 random number generator. The function curand_init() has also been overloaded to accept these function arguments, as in CUDA C++. Device Functions are declared "attributes(device)" in CUDA