C/C++ API reference

Contents

C/C++ API reference#

This document describes the hipRAND APIs available in C and C++.

Device Functions#

group hipranddevice

Defines

HIPRAND_PHILOX4x32_DEFAULT_SEED#

Default seed for PHILOX4x32 PRNG.

HIPRAND_XORWOW_DEFAULT_SEED#

Default seed for XORWOW PRNG.

HIPRAND_MRG32K3A_DEFAULT_SEED#

Default seed for MRG32K3A PRNG.

HIPRAND_MTGP32_DEFAULT_SEED#

Default seed for MTGP32 PRNG.

HIPRAND_MT19937_DEFAULT_SEED#

Default seed for MT19937 PRNG.

Typedefs

typedef struct hiprandStateXORWOW hiprandStateXORWOW_t#

XORWOW PRNG state type.

typedef hiprandStateXORWOW_t hipRandState_t#

Deprecated alias of hiprandStateXORWOW_t.

Deprecated:

Please use hiprandStateXORWOW_t directly.

typedef struct hiprandStatePhilox4_32_10 hiprandStatePhilox4_32_10_t#

PHILOX PRNG state type.

typedef struct hiprandStateMRG32k3a hiprandStateMRG32k3a_t#

MRG32k3a PRNG state type.

typedef struct hiprandStateMtgp32 hiprandStateMtgp32_t#

MTGP32 PRNG state type.

typedef struct hiprandStateScrambledSobol32 hiprandStateScrambledSobol32_t#

Scrambled SOBOL32 QRNG state type.

typedef struct hiprandStateScrambledSobol64 hiprandStateScrambledSobol64_t#

Scrambled SOBOL64 QRNG state type.

typedef struct hiprandStateSobol32 hiprandStateSobol32_t#

SOBOL32 QRNG state type.

typedef struct hiprandStateSobol64 hiprandStateSobol64_t#

SOBOL64 QRNG state type.

typedef mtgp32_params mtgp32_kernel_params_t#

Opaque MTGP32 PRNG state parameters in the device format.

typedef struct mtgp32_params_fast mtgp32_params_fast_t#

Opaque MTGP32 PRNG state parameters in the host format.

typedef mtgp32_fast_params mtgp32_fast_param_t#

Deprecated alias of mtgp32_params_fast_t.

Deprecated:

please use mtgp32_params_fast_t directly.

typedef unsigned int hiprandDirectionVectors32_t[32]#

Convenience typedef for Sobol 32 direction vector array.

typedef unsigned long long hiprandDirectionVectors64_t[64]#

Convenience typedef for Sobol 64 direction vector array.

Functions

void hiprand_mtgp32_block_copy(hiprandStateMtgp32_t *src, hiprandStateMtgp32_t *dest)#

Copy MTGP32 state to another state using block of threads.

Copies a MTGP32 state src to dest using a block of threads efficiently. Example usage would be:

__global__
void generate_kernel(hiprandStateMtgp32_t * states, unsigned int * output, const size_t size)
{
     const unsigned int state_id = hipBlockIdx_x;
     unsigned int index = hipBlockIdx_x * hipBlockDim_x + hipThreadIdx_x;
     unsigned int stride = hipGridDim_x * hipBlockDim_x;

     __shared__ GeneratorState state;
     hiprand_mtgp32_block_copy(&states[state_id], &state);

     while(index < size)
     {
         output[index] = rocrand(&state);
         index += stride;
     }

     hiprand_mtgp32_block_copy(&state, &states[state_id]);
}
Parameters:
  • src – - Pointer to a state to copy from

  • dest – - Pointer to a state to copy to

void hiprand_mtgp32_set_params(hiprandStateMtgp32_t *state, mtgp32_kernel_params_t *params)#

Changes parameters of a MTGP32 state.

Parameters:
  • state – - Pointer to a MTGP32 state

  • params – - Pointer to new parameters

template<class StateType>
void hiprand_init(const unsigned long long seed, const unsigned long long subsequence, const unsigned long long offset, StateType *state)#

Initializes a PRNG state.

Note

MTGP32 must be initialized from the host using hiprandMakeMTGP32KernelState().

Template Parameters:

StateType – - Pseudorandom number generator state type. StateType type must be one of following types: hiprandStateXORWOW_t, hiprandStatePhilox4_32_10_t or hiprandStateMRG32k3a_t

Parameters:
  • seed – - Pseudorandom number generator’s seed

  • subsequence – - Number of subsequence to skipahead

  • offset – - Absolute subsequence offset, i.e. how many states from current subsequence should be skipped

  • state – - Pointer to a state to initialize

void hiprand_init(hiprandDirectionVectors32_t direction_vectors, unsigned int offset, hiprandStateSobol32_t *state)#

Initializes a Sobol32 state.

Parameters:
  • direction_vectors – - Pointer to array of 32 unsigned ints that represent the direction numbers.

  • offset – - Absolute subsequence offset, i.e. how many states should be skipped.

  • state – - Pointer to a state to initialize.

void hiprand_init(hiprandDirectionVectors32_t direction_vectors, unsigned int scramble_constant, unsigned int offset, hiprandStateScrambledSobol32_t *state)#

Initializes a ScrambledSobol32 state.

Parameters:
  • direction_vectors – - Pointer to array of 32 unsigned ints that represent the direction numbers.

  • scramble_constant – - Constant used for scrambling the sequence.

  • offset – - Absolute subsequence offset, i.e. how many states should be skipped.

  • state – - Pointer to a state to initialize.

void hiprand_init(hiprandDirectionVectors64_t direction_vectors, unsigned int offset, hiprandStateSobol64_t *state)#

Initializes a Sobol64 state.

Parameters:
  • direction_vectors – - Pointer to array of 64 unsigned long long ints that represent the direction numbers.

  • offset – - Absolute subsequence offset, i.e. how many states should be skipped.

  • state – - Pointer to a state to initialize.

void hiprand_init(hiprandDirectionVectors64_t direction_vectors, unsigned long long int scramble_constant, unsigned int offset, hiprandStateScrambledSobol64_t *state)#

Initializes a ScrambledSobol64 state.

Parameters:
  • direction_vectors – - Pointer to array of 64 unsigned long long ints that represent the direction numbers.

  • scramble_constant – - Constant used for scrambling the sequence.

  • offset – - Absolute subsequence offset, i.e. how many states should be skipped.

  • state – - Pointer to a state to initialize.

template<class StateType>
void skipahead(unsigned long long n, StateType *state)#

Updates RNG state skipping n states ahead.

Template Parameters:

StateType – - Random number generator state type. StateType type must be one of following types: hiprandStateXORWOW_t, hiprandStatePhilox4_32_10_t, hiprandStateMRG32k3a_t, hiprandStateSobol32_t, hiprandStateScrambledSobol32_t, hiprandStateSobol64_t or hiprandStateScrambledSobol64_t.

Parameters:
  • n – - Number of states to skipahead

  • state – - Pointer to a state to modify

template<class StateType>
void skipahead_sequence(unsigned long long n, StateType *state)#

Updates PRNG state skipping n sequences ahead.

PRNG

Sequence size [Number of elements]

XORWOW

\( 2^{67} \)

Philox

\( 4 \times 2^{64} \)

MRG32k3a

\( 2^{67} \)

Template Parameters:

StateType – - Random number generator state type. StateType type must be one of following types: hiprandStateXORWOW_t, hiprandStatePhilox4_32_10_t, or hiprandStateMRG32k3a_t

Parameters:
  • n – - Number of subsequences to skipahead

  • state – - Pointer to a state to update

template<class StateType>
void skipahead_subsequence(unsigned long long n, StateType *state)#

Updates PRNG state skipping n subsequences ahead.

PRNG

Subsequence size [Number of elements]

XORWOW

\( 2^{67} \)

Philox

\( 4 \times 2^{64} \)

MRG32k3a

\( 2^{127} \)

Template Parameters:

StateType – - Random number generator state type. StateType type must be one of following types: hiprandStateXORWOW_t, hiprandStatePhilox4_32_10_t or hiprandStateMRG32k3a_t

Parameters:
  • n – - Number of subsequences to skipahead

  • state – - Pointer to a state to update

template<class StateType>
unsigned int hiprand(StateType *state)#

Generates uniformly distributed random unsigned int from [0; 2^32 - 1] range.

Template Parameters:

StateType – - Random number generator state type. StateType type must be one of following types: hiprandStateXORWOW_t, hiprandStatePhilox4_32_10_t, hiprandStateMRG32k3a_t, hiprandStateMtgp32_t, hiprandStateSobol32_t, hiprandStateScrambledSobol32_t.

Parameters:

state – - Pointer to a RNG state to use

Returns:

Uniformly distributed random 32-bit unsigned int

uint4 hiprand4(hiprandStatePhilox4_32_10_t *state)#

Generates four uniformly distributed random unsigned ints from [0; 2^32 - 1] range.

Parameters:

state – - Pointer to a Philox state to use

Returns:

Four uniformly distributed random 32-bit unsigned ints as uint4

template<class StateType>
unsigned long long int hiprand_long_long(StateType *state)#

Generates uniformly distributed random unsigned long long int from [0; 2^64 - 1] range.

Template Parameters:

StateType – - Random number generator state type. StateType type must be one of the following types: hiprandStateSobol64_t or hiprandStateScrambledSobol64_t.

Parameters:

state – - Pointer to a RNG state to use

Returns:

Uniformly distributed random 64-bit unsigned long long int

template<class StateType>
float hiprand_uniform(StateType *state)#

Generates uniformly distributed random float value from (0; 1] range.

Template Parameters:

StateType – - Random number generator state type.

Parameters:

state – - Pointer to a RNG state to use

Returns:

Uniformly distributed random float value

float4 hiprand_uniform4(hiprandStatePhilox4_32_10_t *state)#

Generates four uniformly distributed random float value from (0; 1] range.

Parameters:

state – - Pointer to a Philox state to use

Returns:

Four uniformly distributed random float values as float4

template<class StateType>
double hiprand_uniform_double(StateType *state)#

Generates uniformly distributed random double value from (0; 1] range.

Note

When state is of type: hiprandStateMRG32k3a_t, hiprandStateMtgp32_t, hiprandStateSobol32_t or hiprandStateScrambledSobol32_t then the returned double value is generated using only 32 random bits (one unsigned int value). In case of the Sobol types, this is done to guarantee the quasi-random properties.

Template Parameters:

StateType – - Random number generator state type.

Parameters:

state – - Pointer to a RNG state to use

Returns:

Uniformly distributed random double value

double2 hiprand_uniform2_double(hiprandStatePhilox4_32_10_t *state)#

Generates two uniformly distributed random double values from (0; 1] range.

Parameters:

state – - Pointer to a Philox state to use

Returns:

Two uniformly distributed random double values as double2

double4 hiprand_uniform4_double(hiprandStatePhilox4_32_10_t *state)#

Generates four uniformly distributed random double values from (0; 1] range.

Parameters:

state – - Pointer to a Philox state to use

Returns:

Four uniformly distributed random double values as double4

template<class StateType>
float hiprand_normal(StateType *state)#

Generates normally distributed random float value.

Mean value of normal distribution is equal to 0.0, and standard deviation equals 1.0.

Template Parameters:

StateType – - Random number generator state type.

Parameters:

state – - Pointer to a RNG state to use

Returns:

Normally distributed random float value

template<class StateType>
float2 hiprand_normal2(StateType *state)#

Generates two normally distributed random float values.

Mean value of normal distribution is equal to 0.0, and standard deviation equals 1.0.

Template Parameters:

StateType – - Random number generator state type. StateType type must be one of following types: hiprandStateXORWOW_t, hiprandStatePhilox4_32_10_t, or hiprandStateMRG32k3a_t

Parameters:

state – - Pointer to a RNG state to use

Returns:

Two normally distributed random float values as float2

float4 hiprand_normal4(hiprandStatePhilox4_32_10_t *state)#

Generates four normally distributed random float values.

Mean value of normal distribution is equal to 0.0, and standard deviation equals 1.0.

Parameters:

state – - Pointer to a Philox state to use

Returns:

Four normally distributed random float values as float4

template<class StateType>
double hiprand_normal_double(StateType *state)#

Generates normally distributed random double value.

Mean value of normal distribution is equal to 0.0, and standard deviation equals 1.0.

Template Parameters:

StateType – - Random number generator state type.

Parameters:

state – - Pointer to a RNG state to use

Returns:

Normally distributed random double value

template<class StateType>
double2 hiprand_normal2_double(StateType *state)#

Generates two normally distributed random double values.

Mean value of normal distribution is equal to 0.0, and standard deviation equals 1.0.

Template Parameters:

StateType – - Random number generator state type. StateType type must be one of following types: hiprandStateXORWOW_t, hiprandStatePhilox4_32_10_t, or hiprandStateMRG32k3a_t

Parameters:

state – - Pointer to a RNG state to use

Returns:

Two normally distributed random double values as double2

double4 hiprand_normal4_double(hiprandStatePhilox4_32_10_t *state)#

Generates four normally distributed random double values.

Mean value of normal distribution is equal to 0.0, and standard deviation equals 1.0.

Parameters:

state – - Pointer to a Philox state to use

Returns:

Four normally distributed random double values as double4

template<class StateType>
float hiprand_log_normal(StateType *state, float mean, float stddev)#

Generates log-normally distributed random float value.

Template Parameters:

StateType – - Random number generator state type.

Parameters:
  • state – - Pointer to a RNG state to use

  • mean – - Mean value of log-normal distribution

  • stddev – - Standard deviation value of log-normal distribution

Returns:

Log-normally distributed random float value

template<class StateType>
float2 hiprand_log_normal2(StateType *state, float mean, float stddev)#

Generates two log-normally distributed random float values.

Template Parameters:

StateType – - Random number generator state type. StateType type must be one of following types: hiprandStateXORWOW_t, hiprandStatePhilox4_32_10_t, or hiprandStateMRG32k3a_t

Parameters:
  • state – - Pointer to a RNG state to use

  • mean – - Mean value of log-normal distribution

  • stddev – - Standard deviation value of log-normal distribution

Returns:

Two log-normally distributed random float values as float2

float4 hiprand_log_normal4(hiprandStatePhilox4_32_10_t *state, float mean, float stddev)#

Generates four log-normally distributed random float values.

Parameters:
  • state – - Pointer to a Philox state to use

  • mean – - Mean value of log-normal distribution

  • stddev – - Standard deviation value of log-normal distribution

Returns:

Four log-normally distributed random float values as float4

template<class StateType>
double hiprand_log_normal_double(StateType *state, double mean, double stddev)#

Generates log-normally distributed random double value.

Template Parameters:

StateType – - Random number generator state type.

Parameters:
  • state – - Pointer to a RNG state to use

  • mean – - Mean value of log-normal distribution

  • stddev – - Standard deviation value of log-normal distribution

Returns:

Log-normally distributed random double value

template<class StateType>
double2 hiprand_log_normal2_double(StateType *state, double mean, double stddev)#

Generates two log-normally distributed random double values.

Template Parameters:

StateType – - Random number generator state type. StateType type must be one of following types: hiprandStateXORWOW_t, hiprandStatePhilox4_32_10_t, hiprandStateMRG32k3a_t or hiprandStateMtgp32_t.

Parameters:
  • state – - Pointer to a RNG state to use

  • mean – - Mean value of log-normal distribution

  • stddev – - Standard deviation value of log-normal distribution

Returns:

Two log-normally distributed random double values as double2

double4 hiprand_log_normal4_double(hiprandStatePhilox4_32_10_t *state, double mean, double stddev)#

Generates four log-normally distributed random double values.

Parameters:
  • state – - Pointer to a Philox state to use

  • mean – - Mean value of log-normal distribution

  • stddev – - Standard deviation value of log-normal distribution

Returns:

Four log-normally distributed random double values as double4

template<class StateType>
unsigned int hiprand_poisson(StateType *state, double lambda)#

Generates Poisson-distributed random unsigned int value.

Template Parameters:

StateType – - Random number generator state type.

Parameters:
  • state – - Pointer to a RNG state to use

  • lambda – - Lambda (mean) parameter of Poisson distribution

Returns:

Poisson-distributed random unsigned int value

uint4 hiprand_poisson4(hiprandStatePhilox4_32_10_t *state, double lambda)#

Generates four Poisson-distributed random unsigned int values.

Parameters:
  • state – - Pointer to a Philox state to use

  • lambda – - Lambda (mean) parameter of Poisson distribution

Returns:

Four Poisson-distributed random unsigned int values as uint4

template<class StateType>
unsigned int hiprand_discrete(StateType *state, hiprandDiscreteDistribution_t discrete_distribution)#

Generates random unsigned int value according to given discrete distribution.

See also: hiprandCreatePoissonDistribution()

Template Parameters:

StateType – - Random number generator state type.

Parameters:
  • state – - Pointer to a RNG state to use

  • discrete_distribution – - Discrete distribution

Returns:

Random unsigned int value

uint4 hiprand_discrete4(hiprandStatePhilox4_32_10_t *state, hiprandDiscreteDistribution_t discrete_distribution)#

Generates four random unsigned int values according to given discrete distribution.

See also: hiprandCreatePoissonDistribution()

Parameters:
  • state – - Pointer to a Philox state to use

  • discrete_distribution – - Discrete distribution

Returns:

Four random unsigned int values as uint4

inline hiprandStatus_t hiprandMakeMTGP32Constants(const mtgp32_params_fast_t params[], mtgp32_kernel_params_t *p)#

Loads parameters for MTGP32.

Loads parameters for use by kernel functions on the host-side and copies the results to the specified location in device memory.

Parameters:
  • params[in] Pointer to an array of type mtgp32_params_fast_t allocated in host memory

  • p[out] Pointer to a mtgp32_kernel_params_t structure allocated in device memory

Returns:

  • HIPRAND_STATUS_ALLOCATION_FAILED if parameters could not be loaded

  • HIPRAND_STATUS_SUCCESS if parameters are loaded

inline hiprandStatus_t hiprandMakeMTGP32KernelState(hiprandStateMtgp32_t *s, mtgp32_params_fast_t params[], mtgp32_kernel_params_t *k, int n, unsigned long long seed)#

Initializes MTGP32 states.

Initializes MTGP32 states on the host-side by allocating a state array in host memory, initializes that array, and copies the result to device memory.

  • On the cuRAND backend k must be initialized by calling hiprandMakeMTGP32Constants() with the the same params

  • On the rocRAND backend k is ignored

For maximum portability application should initialize k as required by cuRAND, but when only targeting the rocRAND backend it may be left uninitialized.

Parameters:
  • s[out] Pointer to an array of states in device memory

  • params[in] Pointer to an array of type mtgp32_params_fast_t in host memory

  • k[in] Pointer to a mtgp32_kernel_params_t structure allocated in device memory

  • n – Number of states to initialize

  • seed – Seed value

Returns:

  • HIPRAND_STATUS_ALLOCATION_FAILED if states could not be initialized

  • HIPRAND_STATUS_SUCCESS if states are initialized

C Host API#

group hiprandhost

Defines

HIPRAND_VERSION#

hipRAND library version

Version number may not be visible in the documentation.

HIPRAND_VERSION % 100 is the patch level, HIPRAND_VERSION / 100 % 1000 is the minor version, HIPRAND_VERSION / 100000 is the major version.

For example, if HIPRAND_VERSION is 100500, then the major version is 1, the minor version is 5, and the patch level is 0.

HIPRAND_DEFAULT_MAX_BLOCK_SIZE#
HIPRAND_DEFAULT_MIN_WARPS_PER_EU#

Typedefs

typedef enum hiprandStatus hiprandStatus_t#

hipRAND function call status type

typedef enum hiprandRngType hiprandRngType_t#

hipRAND generator type

typedef enum hiprandOrdering hiprandOrdering_t#

hipRAND generator ordering

typedef enum hiprandDirectionVectorSet hiprandDirectionVectorSet_t#

hipRAND vector set for quasirandom generators.

Enums

enum hiprandStatus#

hipRAND function call status type

Values:

enumerator HIPRAND_STATUS_SUCCESS#

Success.

enumerator HIPRAND_STATUS_VERSION_MISMATCH#

Header file and linked library version do not match.

enumerator HIPRAND_STATUS_NOT_INITIALIZED#

Generator not created.

enumerator HIPRAND_STATUS_ALLOCATION_FAILED#

Memory allocation failed.

enumerator HIPRAND_STATUS_TYPE_ERROR#

Generator type is wrong.

enumerator HIPRAND_STATUS_OUT_OF_RANGE#

Argument out of range.

enumerator HIPRAND_STATUS_LENGTH_NOT_MULTIPLE#

Requested size is not a multiple of quasirandom generator’s dimension, or requested size is not even (see hiprandGenerateNormal()), or pointer is misaligned (see hiprandGenerateNormal())

enumerator HIPRAND_STATUS_DOUBLE_PRECISION_REQUIRED#

GPU does not have double precision.

enumerator HIPRAND_STATUS_LAUNCH_FAILURE#

Kernel launch failure.

enumerator HIPRAND_STATUS_PREEXISTING_FAILURE#

Preexisting failure on library entry.

enumerator HIPRAND_STATUS_INITIALIZATION_FAILED#

Initialization of HIP failed.

enumerator HIPRAND_STATUS_ARCH_MISMATCH#

Architecture mismatch, GPU does not support requested feature.

enumerator HIPRAND_STATUS_INTERNAL_ERROR#

Internal library error.

enumerator HIPRAND_STATUS_NOT_IMPLEMENTED#

Feature not implemented yet.

enum hiprandRngType#

hipRAND generator type

Values:

enumerator HIPRAND_RNG_PSEUDO_DEFAULT#

Default pseudorandom generator.

enumerator HIPRAND_RNG_PSEUDO_XORWOW#

XORWOW pseudorandom generator.

enumerator HIPRAND_RNG_PSEUDO_MRG32K3A#

MRG32k3a pseudorandom generator.

enumerator HIPRAND_RNG_PSEUDO_MTGP32#

Mersenne Twister MTGP32 pseudorandom generator.

enumerator HIPRAND_RNG_PSEUDO_MT19937#

Mersenne Twister 19937.

enumerator HIPRAND_RNG_PSEUDO_PHILOX4_32_10#

PHILOX_4x32 (10 rounds) pseudorandom generator.

enumerator HIPRAND_RNG_QUASI_DEFAULT#

Default quasirandom generator.

enumerator HIPRAND_RNG_QUASI_SOBOL32#

Sobol32 quasirandom generator.

enumerator HIPRAND_RNG_QUASI_SCRAMBLED_SOBOL32#

Scrambled Sobol32 quasirandom generator.

enumerator HIPRAND_RNG_QUASI_SOBOL64#

Sobol64 quasirandom generator.

enumerator HIPRAND_RNG_QUASI_SCRAMBLED_SOBOL64#

Scrambled Sobol64 quasirandom generator.

enum hiprandOrdering#

hipRAND generator ordering

Values:

enumerator HIPRAND_ORDERING_PSEUDO_BEST#

Best ordering for pseudorandom results.

enumerator HIPRAND_ORDERING_PSEUDO_DEFAULT#

Specific default thread sequence for pseudorandom results, same as HIPRAND_ORDERING_PSEUDO_BEST.

enumerator HIPRAND_ORDERING_PSEUDO_SEEDED#

Specific seeding pattern for fast lower quality pseudorandom results.

enumerator HIPRAND_ORDERING_PSEUDO_LEGACY#

Specific legacy sequence for pseudorandom results. This remains the same across releases, but not across backends.

enumerator HIPRAND_ORDERING_PSEUDO_DYNAMIC#

Specific ordering adjusted to the device it is being executed on, provides the best performance.

enumerator HIPRAND_ORDERING_QUASI_DEFAULT#

Specific n-dimensional ordering for quasirandom results.

enum hiprandDirectionVectorSet#

hipRAND vector set for quasirandom generators.

Values:

enumerator HIPRAND_DIRECTION_VECTORS_32_JOEKUO6#
enumerator HIPRAND_SCRAMBLED_DIRECTION_VECTORS_32_JOEKUO6#
enumerator HIPRAND_DIRECTION_VECTORS_64_JOEKUO6#
enumerator HIPRAND_SCRAMBLED_DIRECTION_VECTORS_64_JOEKUO6#

Functions

hiprandStatus_t hiprandCreateGenerator(hiprandGenerator_t *generator, hiprandRngType_t rng_type)#

Creates a new random number generator.

Creates a new random number generator of type rng_type, and returns it in generator. That generator will use GPU to create random numbers.

Values for rng_type are:

  • HIPRAND_RNG_PSEUDO_DEFAULT

  • HIPRAND_RNG_PSEUDO_XORWOW

  • HIPRAND_RNG_PSEUDO_MRG32K3A

  • HIPRAND_RNG_PSEUDO_MTGP32

  • HIPRAND_RNG_PSEUDO_MT19937

  • HIPRAND_RNG_PSEUDO_PHILOX4_32_10

  • HIPRAND_RNG_QUASI_DEFAULT

  • HIPRAND_RNG_QUASI_SOBOL32

  • HIPRAND_RNG_QUASI_SCRAMBLED_SOBOL32

  • HIPRAND_RNG_QUASI_SOBOL64

  • HIPRAND_RNG_QUASI_SCRAMBLED_SOBOL64

Parameters:
  • generator – - Pointer to generator

  • rng_type – - Type of random number generator to create

Returns:

  • HIPRAND_STATUS_ALLOCATION_FAILED, if memory allocation failed

  • HIPRAND_STATUS_INITIALIZATION_FAILED if there was a problem setting up the GPU

  • HIPRAND_STATUS_VERSION_MISMATCH if the header file version does not match the dynamically linked library version

  • HIPRAND_STATUS_TYPE_ERROR if the value for rng_type is invalid

  • HIPRAND_STATUS_NOT_IMPLEMENTED if generator of type rng_type is not implemented yet

  • HIPRAND_STATUS_SUCCESS if generator was created successfully

hiprandStatus_t hiprandCreateGeneratorHost(hiprandGenerator_t *generator, hiprandRngType_t rng_type)#

Creates a new random number generator on host.

Creates a new host random number generator of type rng_type and returns it in generator. Created generator will use host CPU to generate random numbers.

Values for rng_type are:

  • HIPRAND_RNG_PSEUDO_DEFAULT

  • HIPRAND_RNG_PSEUDO_XORWOW

  • HIPRAND_RNG_PSEUDO_MRG32K3A

  • HIPRAND_RNG_PSEUDO_MTGP32

  • HIPRAND_RNG_PSEUDO_MT19937

  • HIPRAND_RNG_PSEUDO_PHILOX4_32_10

  • HIPRAND_RNG_QUASI_DEFAULT

  • HIPRAND_RNG_QUASI_SOBOL32

  • HIPRAND_RNG_QUASI_SCRAMBLED_SOBOL32

  • HIPRAND_RNG_QUASI_SOBOL64

  • HIPRAND_RNG_QUASI_SCRAMBLED_SOBOL64

Parameters:
  • generator – - Pointer to generator

  • rng_type – - Type of random number generator to create

Returns:

  • HIPRAND_STATUS_ALLOCATION_FAILED, if memory allocation failed

  • HIPRAND_STATUS_VERSION_MISMATCH if the header file version does not match the dynamically linked library version

  • HIPRAND_STATUS_TYPE_ERROR if the value for rng_type is invalid

  • HIPRAND_STATUS_NOT_IMPLEMENTED if host generator of type rng_type is not implemented yet

  • HIPRAND_STATUS_SUCCESS if generator was created successfully

hiprandStatus_t hiprandDestroyGenerator(hiprandGenerator_t generator)#

Destroys random number generator.

Destroys random number generator and frees related memory.

Parameters:

generator – - Generator to be destroyed

Returns:

  • HIPRAND_STATUS_NOT_INITIALIZED if the generator was not initialized

  • HIPRAND_STATUS_SUCCESS if generator was destroyed successfully

hiprandStatus_t hiprandGenerate(hiprandGenerator_t generator, unsigned int *output_data, size_t n)#

Generates uniformly distributed 32-bit unsigned integers.

Generates n uniformly distributed 32-bit unsigned integers and saves them to output_data.

Generated numbers are between 0 and 2^32, including 0 and excluding 2^32.

Note: generator must be not be of type HIPRAND_RNG_QUASI_SOBOL64 or HIPRAND_RNG_QUASI_SCRAMBLED_SOBOL64.

Parameters:
  • generator – - Generator to use

  • output_data – - Pointer to memory to store generated numbers

  • n – - Number of 32-bit unsigned integers to generate

Returns:

  • HIPRAND_STATUS_NOT_INITIALIZED if the generator was not initialized

  • HIPRAND_STATUS_LAUNCH_FAILURE if generator failed to launch kernel

  • HIPRAND_STATUS_SUCCESS if random numbers were successfully generated

hiprandStatus_t hiprandGenerateChar(hiprandGenerator_t generator, unsigned char *output_data, size_t n)#

Generates uniformly distributed 8-bit unsigned integers.

Generates n uniformly distributed 8-bit unsigned integers and saves them to output_data.

Generated numbers are between 0 and 2^8, including 0 and excluding 2^8.

Parameters:
  • generator – - Generator to use

  • output_data – - Pointer to memory to store generated numbers

  • n – - Number of 8-bit unsigned integers to generate

Returns:

  • HIPRAND_STATUS_NOT_INITIALIZED if the generator was not initialized

  • HIPRAND_STATUS_LAUNCH_FAILURE if generator failed to launch kernel

  • HIPRAND_STATUS_SUCCESS if random numbers were successfully generated

hiprandStatus_t hiprandGenerateShort(hiprandGenerator_t generator, unsigned short *output_data, size_t n)#

Generates uniformly distributed 16-bit unsigned integers.

Generates n uniformly distributed 16-bit unsigned integers and saves them to output_data.

Generated numbers are between 0 and 2^16, including 0 and excluding 2^16.

Parameters:
  • generator – - Generator to use

  • output_data – - Pointer to memory to store generated numbers

  • n – - Number of 16-bit unsigned integers to generate

Returns:

  • HIPRAND_STATUS_NOT_INITIALIZED if the generator was not initialized

  • HIPRAND_STATUS_LAUNCH_FAILURE if generator failed to launch kernel

  • HIPRAND_STATUS_SUCCESS if random numbers were successfully generated

hiprandStatus_t hiprandGenerateLongLong(hiprandGenerator_t generator, unsigned long long *output_data, size_t n)#

Generates uniformly distributed 64-bit unsigned integers.

Generates n uniformly distributed 64-bit unsigned integers and saves them to output_data.

Generated numbers are between 0 and 2^64, including 0 and excluding 2^64.

Note: generator must be of type HIPRAND_RNG_QUASI_SOBOL64 or HIPRAND_RNG_QUASI_SCRAMBLED_SOBOL64.

Parameters:
  • generator – - Generator to use

  • output_data – - Pointer to memory to store generated numbers

  • n – - Number of 64-bit unsigned integers to generate

Returns:

  • HIPRAND_STATUS_NOT_INITIALIZED if the generator was not initialized

  • HIPRAND_STATUS_LAUNCH_FAILURE if generator failed to launch kernel

  • HIPRAND_STATUS_SUCCESS if random numbers were successfully generated

hiprandStatus_t hiprandGenerateUniform(hiprandGenerator_t generator, float *output_data, size_t n)#

Generates uniformly distributed floats.

Generates n uniformly distributed 32-bit floating-point values and saves them to output_data.

Generated numbers are between 0.0f and 1.0f, excluding 0.0f and including 1.0f.

Parameters:
  • generator – - Generator to use

  • output_data – - Pointer to memory to store generated numbers

  • n – - Number of floats to generate

Returns:

  • HIPRAND_STATUS_NOT_INITIALIZED if the generator was not initialized

  • HIPRAND_STATUS_LAUNCH_FAILURE if generator failed to launch kernel

  • HIPRAND_STATUS_LENGTH_NOT_MULTIPLE if n is not a multiple of the dimension of used quasi-random generator

  • HIPRAND_STATUS_SUCCESS if random numbers were successfully generated

hiprandStatus_t hiprandGenerateUniformDouble(hiprandGenerator_t generator, double *output_data, size_t n)#

Generates uniformly distributed double-precision floating-point values.

Generates n uniformly distributed 64-bit double-precision floating-point values and saves them to output_data.

Generated numbers are between 0.0 and 1.0, excluding 0.0 and including 1.0.

Note: When generator is of type: HIPRAND_RNG_PSEUDO_MRG32K3A, HIPRAND_RNG_PSEUDO_MTGP32, HIPRAND_RNG_QUASI_SOBOL32, or HIPRAND_RNG_QUASI_SCRAMBLED_SOBOL32 then the returned double values are generated from only 32 random bits each (one unsigned int value per one generated double).

Parameters:
  • generator – - Generator to use

  • output_data – - Pointer to memory to store generated numbers

  • n – - Number of floats to generate

Returns:

  • HIPRAND_STATUS_NOT_INITIALIZED if the generator was not initialized

  • HIPRAND_STATUS_LAUNCH_FAILURE if generator failed to launch kernel

  • HIPRAND_STATUS_LENGTH_NOT_MULTIPLE if n is not a multiple of the dimension of used quasi-random generator

  • HIPRAND_STATUS_SUCCESS if random numbers were successfully generated

hiprandStatus_t hiprandGenerateUniformHalf(hiprandGenerator_t generator, half *output_data, size_t n)#

Generates uniformly distributed half-precision floating-point values.

Generates n uniformly distributed 16-bit half-precision floating-point values and saves them to output_data.

Generated numbers are between 0.0 and 1.0, excluding 0.0 and including 1.0.

Parameters:
  • generator – - Generator to use

  • output_data – - Pointer to memory to store generated numbers

  • n – - Number of halfs to generate

Returns:

  • HIPRAND_STATUS_NOT_INITIALIZED if the generator was not initialized

  • HIPRAND_STATUS_LAUNCH_FAILURE if generator failed to launch kernel

  • HIPRAND_STATUS_LENGTH_NOT_MULTIPLE if n is not a multiple of the dimension of used quasi-random generator

  • HIPRAND_STATUS_SUCCESS if random numbers were successfully generated

hiprandStatus_t hiprandGenerateNormal(hiprandGenerator_t generator, float *output_data, size_t n, float mean, float stddev)#

Generates normally distributed floats.

Generates n normally distributed 32-bit floating-point values and saves them to output_data.

Parameters:
  • generator – - Generator to use

  • output_data – - Pointer to memory to store generated numbers

  • n – - Number of floats to generate

  • mean – - Mean value of normal distribution

  • stddev – - Standard deviation value of normal distribution

Returns:

  • HIPRAND_STATUS_NOT_INITIALIZED if the generator was not initialized

  • HIPRAND_STATUS_LAUNCH_FAILURE if generator failed to launch kernel

  • HIPRAND_STATUS_LENGTH_NOT_MULTIPLE if n is not even, output_data is not aligned to sizeof(float2) bytes, or n is not a multiple of the dimension of used quasi-random generator

  • HIPRAND_STATUS_SUCCESS if random numbers were successfully generated

hiprandStatus_t hiprandGenerateNormalDouble(hiprandGenerator_t generator, double *output_data, size_t n, double mean, double stddev)#

Generates normally distributed doubles.

Generates n normally distributed 64-bit double-precision floating-point numbers and saves them to output_data.

Parameters:
  • generator – - Generator to use

  • output_data – - Pointer to memory to store generated numbers

  • n – - Number of doubles to generate

  • mean – - Mean value of normal distribution

  • stddev – - Standard deviation value of normal distribution

Returns:

  • HIPRAND_STATUS_NOT_INITIALIZED if the generator was not initialized

  • HIPRAND_STATUS_LAUNCH_FAILURE if generator failed to launch kernel

  • HIPRAND_STATUS_LENGTH_NOT_MULTIPLE if n is not even, output_data is not aligned to sizeof(double2) bytes, or n is not a multiple of the dimension of used quasi-random generator

  • HIPRAND_STATUS_SUCCESS if random numbers were successfully generated

hiprandStatus_t hiprandGenerateNormalHalf(hiprandGenerator_t generator, half *output_data, size_t n, half mean, half stddev)#

Generates normally distributed halfs.

Generates n normally distributed 16-bit half-precision floating-point numbers and saves them to output_data.

Parameters:
  • generator – - Generator to use

  • output_data – - Pointer to memory to store generated numbers

  • n – - Number of halfs to generate

  • mean – - Mean value of normal distribution

  • stddev – - Standard deviation value of normal distribution

Returns:

  • HIPRAND_STATUS_NOT_INITIALIZED if the generator was not initialized

  • HIPRAND_STATUS_LAUNCH_FAILURE if generator failed to launch kernel

  • HIPRAND_STATUS_LENGTH_NOT_MULTIPLE if n is not even, output_data is not aligned to sizeof(half2) bytes, or n is not a multiple of the dimension of used quasi-random generator

  • HIPRAND_STATUS_SUCCESS if random numbers were successfully generated

hiprandStatus_t hiprandGenerateLogNormal(hiprandGenerator_t generator, float *output_data, size_t n, float mean, float stddev)#

Generates log-normally distributed floats.

Generates n log-normally distributed 32-bit floating-point values and saves them to output_data.

Parameters:
  • generator – - Generator to use

  • output_data – - Pointer to memory to store generated numbers

  • n – - Number of floats to generate

  • mean – - Mean value of log normal distribution

  • stddev – - Standard deviation value of log normal distribution

Returns:

  • HIPRAND_STATUS_NOT_INITIALIZED if the generator was not initialized

  • HIPRAND_STATUS_LAUNCH_FAILURE if generator failed to launch kernel

  • HIPRAND_STATUS_LENGTH_NOT_MULTIPLE if n is not even, output_data is not aligned to sizeof(float2) bytes, or n is not a multiple of the dimension of used quasi-random generator

  • HIPRAND_STATUS_SUCCESS if random numbers were successfully generated

hiprandStatus_t hiprandGenerateLogNormalDouble(hiprandGenerator_t generator, double *output_data, size_t n, double mean, double stddev)#

Generates log-normally distributed doubles.

Generates n log-normally distributed 64-bit double-precision floating-point values and saves them to output_data.

Parameters:
  • generator – - Generator to use

  • output_data – - Pointer to memory to store generated numbers

  • n – - Number of doubles to generate

  • mean – - Mean value of log normal distribution

  • stddev – - Standard deviation value of log normal distribution

Returns:

  • HIPRAND_STATUS_NOT_INITIALIZED if the generator was not initialized

  • HIPRAND_STATUS_LAUNCH_FAILURE if generator failed to launch kernel

  • HIPRAND_STATUS_LENGTH_NOT_MULTIPLE if n is not even, output_data is not aligned to sizeof(double2) bytes, or n is not a multiple of the dimension of used quasi-random generator

  • HIPRAND_STATUS_SUCCESS if random numbers were successfully generated

hiprandStatus_t hiprandGenerateLogNormalHalf(hiprandGenerator_t generator, half *output_data, size_t n, half mean, half stddev)#

Generates log-normally distributed halfs.

Generates n log-normally distributed 16-bit half-precision floating-point values and saves them to output_data.

Parameters:
  • generator – - Generator to use

  • output_data – - Pointer to memory to store generated numbers

  • n – - Number of halfs to generate

  • mean – - Mean value of log normal distribution

  • stddev – - Standard deviation value of log normal distribution

Returns:

  • HIPRAND_STATUS_NOT_INITIALIZED if the generator was not initialized

  • HIPRAND_STATUS_LAUNCH_FAILURE if generator failed to launch kernel

  • HIPRAND_STATUS_LENGTH_NOT_MULTIPLE if n is not even, output_data is not aligned to sizeof(half2) bytes, or n is not a multiple of the dimension of used quasi-random generator

  • HIPRAND_STATUS_SUCCESS if random numbers were successfully generated

hiprandStatus_t hiprandGeneratePoisson(hiprandGenerator_t generator, unsigned int *output_data, size_t n, double lambda)#

Generates Poisson-distributed 32-bit unsigned integers.

Generates n Poisson-distributed 32-bit unsigned integers and saves them to output_data.

Parameters:
  • generator – - Generator to use

  • output_data – - Pointer to memory to store generated numbers

  • n – - Number of 32-bit unsigned integers to generate

  • lambda – - lambda for the Poisson distribution

Returns:

  • HIPRAND_STATUS_NOT_INITIALIZED if the generator was not initialized

  • HIPRAND_STATUS_LAUNCH_FAILURE if generator failed to launch kernel

  • HIPRAND_STATUS_OUT_OF_RANGE if lambda is non-positive

  • HIPRAND_STATUS_LENGTH_NOT_MULTIPLE if n is not a multiple of the dimension of used quasi-random generator

  • HIPRAND_STATUS_SUCCESS if random numbers were successfully generated

hiprandStatus_t hiprandGenerateSeeds(hiprandGenerator_t generator)#

Initializes the generator’s state on GPU or host.

Initializes the generator’s state on GPU or host.

If hiprandGenerateSeeds() was not called for a generator, it will be automatically called by functions which generates random numbers like hiprandGenerate(), hiprandGenerateUniform(), hiprandGenerateNormal() etc.

Parameters:

generator – - Generator to initialize

Returns:

  • HIPRAND_STATUS_NOT_INITIALIZED if the generator was never created

  • HIPRAND_STATUS_PREEXISTING_FAILURE if there was an existing error from a previous kernel launch

  • HIPRAND_STATUS_LAUNCH_FAILURE if the kernel launch failed for any reason

  • HIPRAND_STATUS_SUCCESS if the seeds were generated successfully

hiprandStatus_t hiprandSetStream(hiprandGenerator_t generator, hipStream_t stream)#

Sets the current stream for kernel launches.

Sets the current stream for all kernel launches of the generator. All functions will use this stream.

Parameters:
  • generator – - Generator to modify

  • stream – - Stream to use or NULL for default stream

Returns:

  • HIPRAND_STATUS_NOT_INITIALIZED if the generator was not initialized

  • HIPRAND_STATUS_SUCCESS if stream was set successfully

hiprandStatus_t hiprandSetPseudoRandomGeneratorSeed(hiprandGenerator_t generator, unsigned long long seed)#

Sets the seed of a pseudo-random number generator.

Sets the seed of the pseudo-random number generator.

  • This operation resets the generator’s internal state.

  • This operation does not change the generator’s offset.

Parameters:
  • generator – - Pseudo-random number generator

  • seed – - New seed value

Returns:

  • HIPRAND_STATUS_NOT_INITIALIZED if the generator was not initialized

  • HIPRAND_STATUS_TYPE_ERROR if the generator is a quasi random number generator

  • HIPRAND_STATUS_SUCCESS if seed was set successfully

hiprandStatus_t hiprandSetGeneratorOffset(hiprandGenerator_t generator, unsigned long long offset)#

Sets the offset of a random number generator.

Sets the absolute offset of the random number generator.

  • This operation resets the generator’s internal state.

  • This operation does not change the generator’s seed.

Absolute offset cannot be set if generator’s type is HIPRAND_RNG_PSEUDO_MTGP32 or HIPRAND_RNG_PSEUDO_MT19937.

Parameters:
  • generator – - Random number generator

  • offset – - New absolute offset

Returns:

  • HIPRAND_STATUS_NOT_INITIALIZED if the generator was not initialized

  • HIPRAND_STATUS_SUCCESS if offset was successfully set

  • HIPRAND_STATUS_TYPE_ERROR if generator’s type is HIPRAND_RNG_PSEUDO_MTGP32 or HIPRAND_RNG_PSEUDO_MT19937

hiprandStatus_t hiprandSetGeneratorOrdering(hiprandGenerator_t generator, hiprandOrdering_t order)#

Sets the ordering of a random number generator.

Sets the ordering of the results of a random number generator.

  • This operation resets the generator’s internal state.

  • This operation does not change the generator’s seed.

The ordering choices for pseudorandom sequences are HIPRAND_ORDERING_PSEUDO_DEFAULT and HIPRAND_ORDERING_PSEUDO_LEGACY. The default ordering is HIPRAND_ORDERING_PSEUDO_DEFAULT, which is equal to HIPRAND_ORDERING_PSEUDO_LEGACY for now.

For quasirandom sequences there is only one ordering, HIPRAND_ORDERING_QUASI_DEFAULT.

Parameters:
  • generator – - Random number generator

  • order – - New ordering of results

Returns:

  • HIPRAND_STATUS_NOT_INITIALIZED if the generator was not initialized

  • HIPRAND_STATUS_OUT_OF_RANGE if the ordering is not valid

  • HIPRAND_STATUS_SUCCESS if the ordering was successfully set

  • HIPRAND_STATUS_TYPE_ERROR if generator’s type is not valid

hiprandStatus_t hiprandSetQuasiRandomGeneratorDimensions(hiprandGenerator_t generator, unsigned int dimensions)#

Set the number of dimensions of a quasi-random number generator.

Set the number of dimensions of a quasi-random number generator. Supported values of dimensions are 1 to 20000.

  • This operation resets the generator’s internal state.

  • This operation does not change the generator’s offset.

Parameters:
  • generator – - Quasi-random number generator

  • dimensions – - Number of dimensions

Returns:

  • HIPRAND_STATUS_NOT_CREATED if the generator wasn’t created

  • HIPRAND_STATUS_TYPE_ERROR if the generator is not a quasi-random number generator

  • HIPRAND_STATUS_OUT_OF_RANGE if dimensions is out of range

  • HIPRAND_STATUS_SUCCESS if the number of dimensions was set successfully

hiprandStatus_t hiprandGetVersion(int *version)#

Returns the version number of the cuRAND or rocRAND library.

Returns in version the version number of the underlying cuRAND or rocRAND library.

Parameters:

version – - Version of the library

Returns:

  • HIPRAND_STATUS_OUT_OF_RANGE if version is NULL

  • HIPRAND_STATUS_SUCCESS if the version number was successfully returned

hiprandStatus_t hiprandCreatePoissonDistribution(double lambda, hiprandDiscreteDistribution_t *discrete_distribution)#

Construct the histogram for a Poisson distribution.

Construct the histogram for the Poisson distribution with lambda lambda.

Parameters:
  • lambda – - lambda for the Poisson distribution

  • discrete_distribution – - pointer to the histogram in device memory

Returns:

  • HIPRAND_STATUS_ALLOCATION_FAILED if memory could not be allocated

  • HIPRAND_STATUS_OUT_OF_RANGE if discrete_distribution pointer was null

  • HIPRAND_STATUS_OUT_OF_RANGE if lambda is non-positive

  • HIPRAND_STATUS_SUCCESS if the histogram was constructed successfully

hiprandStatus_t hiprandDestroyDistribution(hiprandDiscreteDistribution_t discrete_distribution)#

Destroy the histogram array for a discrete distribution.

Destroy the histogram array for a discrete distribution created by hiprandCreatePoissonDistribution.

Parameters:

discrete_distribution – - pointer to the histogram in device memory

Returns:

  • HIPRAND_STATUS_OUT_OF_RANGE if discrete_distribution was null

  • HIPRAND_STATUS_SUCCESS if the histogram was destroyed successfully

hiprandStatus_t hiprandGetDirectionVectors32(hiprandDirectionVectors32_t **vectors, hiprandDirectionVectorSet_t set)#
hiprandStatus_t hiprandGetDirectionVectors64(hiprandDirectionVectors64_t **vectors, hiprandDirectionVectorSet_t set)#
hiprandStatus_t hiprandGetScrambleConstants32(const unsigned int **constants)#
hiprandStatus_t hiprandGetScrambleConstants64(const unsigned long long **constants)#

C++ Host API Wrapper#

group hiprandhostcpp

Typedefs

typedef philox4x32_10_engine philox4x32_10#

Typedef of hiprand_cpp::philox4x32_10_engine PRNG engine with default seed (HIPRAND_PHILOX4x32_DEFAULT_SEED).

typedef xorwow_engine xorwow#

Typedef of hiprand_cpp::xorwow_engine PRNG engine with default seed (HIPRAND_XORWOW_DEFAULT_SEED).

typedef mrg32k3a_engine mrg32k3a#

Typedef of hiprand_cpp::mrg32k3a_engine PRNG engine with default seed (HIPRAND_MRG32K3A_DEFAULT_SEED).

typedef mtgp32_engine mtgp32#

Typedef of hiprand_cpp::mtgp32_engine PRNG engine with default seed (HIPRAND_MTGP32_DEFAULT_SEED).

typedef mt19937_engine mt19937#

Typedef of hiprand_cpp::mt19937_engine PRNG engine with default seed (HIPRAND_MT19937_DEFAULT_SEED).

typedef sobol32_engine sobol32#

Typedef of hiprand_cpp::sobol32_engine QRNG engine with default number of dimensions (1).

typedef scrambled_sobol32_engine scrambled_sobol32#

Typedef of hiprand_cpp::scrambled_sobol32_engine QRNG engine with default number of dimensions (1).

typedef sobol64_engine sobol64#

Typedef of hiprand_cpp::sobol64_engine QRNG engine with default number of dimensions (1).

typedef scrambled_sobol64_engine scrambled_sobol64#

Typedef of hiprand_cpp::scrambled_sobol64_engine QRNG engine with default number of dimensions (1).

typedef xorwow default_random_engine#

Default random engine.

typedef std::random_device random_device#

A non-deterministic uniform random number generator.

hiprand_cpp::random_device is non-deterministic uniform random number generator, or a pseudo-random number engine if there is no support for non-deterministic random number generation. It’s implemented as a typedef of std::random_device.

For practical use hiprand_cpp::random_device is generally only used to seed a PRNG such as hiprand_cpp::mtgp32_engine.

Example:

#include <hiprand/hiprand.hpp>

int main()
{
    const size_t size = 8192;
    unsigned int * output;
    hipMalloc(&output, size * sizeof(unsigned int));

    hiprand_cpp::random_device rd;
    hiprand_cpp::mtgp32 engine(rd()); // seed engine with a real random value, if available
    hiprand_cpp::normal_distribution<float> dist(0.0, 1.5);
    dist(engine, output, size);
}

Functions

inline int version()#

Returns hipRAND version.

Returns:

hipRAND version number as an int value.

class error : public std::exception#
#include <hiprand.hpp>

A run-time hipRAND error.

The error class represents an error returned by a hipRAND function.

template<class IntType = unsigned int>
class uniform_int_distribution#
#include <hiprand.hpp>

Produces random integer values uniformly distributed on the interval [0, 2^(sizeof(IntType)*8) - 1].

Template Parameters:

IntType – - type of generated values. Only unsigned char, unsigned short, unsigned int, unsigned long long int are supported.

template<class RealType = float>
class uniform_real_distribution#
#include <hiprand.hpp>

Produces random floating-point values uniformly distributed on the interval (0, 1].

Template Parameters:

RealType – - type of generated values. Only float, double and half types are supported.

template<class RealType = float>
class normal_distribution#
#include <hiprand.hpp>

Produces random numbers according to a normal distribution.

Template Parameters:

RealType – - type of generated values. Only float, double and half types are supported.

template<class RealType = float>
class lognormal_distribution#
#include <hiprand.hpp>

Produces positive random numbers according to a log-normal distribution.

Template Parameters:

RealType – - type of generated values. Only float, double and half types are supported.

template<class IntType = unsigned int>
class poisson_distribution#
#include <hiprand.hpp>

Produces random non-negative integer values distributed according to Poisson distribution.

Template Parameters:

IntType – - type of generated values. Only unsinged int type is supported.

template<unsigned long long DefaultSeed = HIPRAND_PHILOX4x32_DEFAULT_SEED>
class philox4x32_10_engine#
#include <hiprand.hpp>

Pseudorandom number engine based Philox algorithm.

philox4x32_10_engine implements a Counter-based random number generator called Philox, which was developed by a group at D. E. Shaw Research. It generates random numbers of type unsigned int on the interval [0; 2^32 - 1]. Random numbers are generated in sets of four.

template<unsigned long long DefaultSeed = HIPRAND_XORWOW_DEFAULT_SEED>
class xorwow_engine#
#include <hiprand.hpp>

Pseudorandom number engine based XORWOW algorithm.

xorwow_engine is a xorshift pseudorandom number engine based on XORWOW algorithm, which was presented by George Marsaglia in “Xorshift RNGs” paper published in Journal of Statistical Software. It produces random numbers of type unsigned int on the interval [0; 2^32 - 1].

template<unsigned long long DefaultSeed = HIPRAND_MRG32K3A_DEFAULT_SEED>
class mrg32k3a_engine#
#include <hiprand.hpp>

Pseudorandom number engine based MRG32k3a CMRG.

mrg32k3a_engine is an implementation of MRG32k3a pseudorandom number generator, which is a Combined Multiple Recursive Generator (CMRG) created by Pierre L’Ecuyer. It produces random 32-bit unsigned int values on the interval [0; 2^32 - 1].

template<unsigned long long DefaultSeed = HIPRAND_MTGP32_DEFAULT_SEED>
class mtgp32_engine#
#include <hiprand.hpp>

Pseudorandom number engine based on Mersenne Twister for Graphic Processors algorithm.

mtgp32_engine is a random number engine based on Mersenne Twister for Graphic Processors algorithm, which is a version of well-known Mersenne Twister algorithm. It produces high quality random numbers of type unsigned int on the interval [0; 2^32 - 1].

template<unsigned long long DefaultSeed = HIPRAND_MT19937_DEFAULT_SEED>
class mt19937_engine#
#include <hiprand.hpp>

Pseudorandom number engine based on Mersenne Twister.

mt19937_engine is a random number engine based on the well-known Mersenne Twister algorithm. It produces high quality random numbers of type unsigned int on the interval [0; 2^32 - 1].

template<unsigned int DefaultNumDimensions = 1>
class sobol32_engine#
#include <hiprand.hpp>

Sobol’s quasi-random sequence generator.

sobol32_engine is quasi-random number engine which produced Sobol sequences. This implementation supports generating sequences in up to 20,000 dimensions. The engine produces random unsigned integers on the interval [0; 2^32 - 1].

template<unsigned int DefaultNumDimensions = 1>
class scrambled_sobol32_engine#
#include <hiprand.hpp>

Sobol’s quasi-random sequence generator.

scrambled_sobol32_engine is a quasi-random number engine which produces scrambled Sobol sequences. This implementation supports generating sequences in up to 20,000 dimensions. The engine produces random unsigned integers on the interval [0; 2^32 - 1].

template<unsigned int DefaultNumDimensions = 1>
class sobol64_engine#
#include <hiprand.hpp>

Sobol’s quasi-random sequence generator.

sobol64_engine is a quasi-random number engine which produced Sobol sequences. This implementation supports generating sequences in up to 20,000 dimensions. The engine produces random unsigned integers on the interval [0; 2^64 - 1].

template<unsigned int DefaultNumDimensions = 1>
class scrambled_sobol64_engine#
#include <hiprand.hpp>

Sobol’s quasi-random sequence generator.

scrambled_sobol64_engine is a quasi-random number engine which produces scrambled Sobol sequences. This implementation supports generating sequences in up to 20,000 dimensions. The engine produces random unsigned integers on the interval [0; 2^64 - 1].