org.bytedeco.cuda.cudart

Class __half

java.lang.Object

org.bytedeco.javacpp.Pointer

org.bytedeco.cuda.cudart.__half

All Implemented Interfaces:

AutoCloseable

@NoOffset
 @Properties(inherit=cudart.class)
public class __half
extends Pointer

\ingroup CUDA_MATH_INTRINSIC_HALF \brief __half data type \details This structure implements the datatype for storing half-precision floating-point numbers. The structure implements assignment, arithmetic and comparison operators, and type conversions. 16 bits are being used in total: 1 sign bit, 5 bits for the exponent, and the significand is being stored in 10 bits. The total precision is 11 bits. There are 15361 representable numbers within the interval [0.0, 1.0], endpoints included. On average we have log10(2**11) ~ 3.311 decimal digits. The objective here is to provide IEEE754-compliant implementation of \p binary16 type and arithmetic with limitations due to device HW not supporting floating-point exceptions.

Nested Class Summary

Nested classes/interfaces inherited from class org.bytedeco.javacpp.Pointer

Pointer.CustomDeallocator, Pointer.Deallocator, Pointer.NativeDeallocator, Pointer.ReferenceCounter

Field Summary

Fields inherited from class org.bytedeco.javacpp.Pointer

address, capacity, limit, position

Constructor Summary

Constructors

Constructor and Description
__half(__half_raw hr) Constructor from \p __half_raw.
__half(__nv_bfloat16 f) Construct \p __half from \p __nv_bfloat16 input using default round-to-nearest-even rounding mode.
__half() Constructor by default.
__half(double f) Construct \p __half from \p double input using default round-to-nearest-even rounding mode.
__half(float f) Construct \p __half from \p float input using default round-to-nearest-even rounding mode.
__half(int val) Construct \p __half from \p int input using default round-to-nearest-even rounding mode.
__half(long val) Construct \p __half from \p long input using default round-to-nearest-even rounding mode.
__half(Pointer p) Pointer cast constructor.
__half(short val) Construct \p __half from \p short integer input using default round-to-nearest-even rounding mode.

Method Summary

Modifier and Type	Method and Description
__half_raw	as__half_raw() Type cast to \p __half_raw operator.
boolean	asBoolean() Conversion operator to \p bool data type.
byte	asByte() Conversion operator to \p signed \p char data type.
float	asFloat() Type cast to \p float operator.
int	asInt() Conversion operator to \p int data type.
long	asLong() Conversion operator to \p long data type.
short	asShort() Conversion operator to \p short data type.
__half	put(__half_raw hr) Assignment operator from \p __half_raw.
__half	put(double f) Type cast to \p __half assignment operator from \p double input using default round-to-nearest-even rounding mode.
__half	put(float f) Type cast to \p __half assignment operator from \p float input using default round-to-nearest-even rounding mode.
__half	put(int val) Type cast from \p int assignment operator, using default round-to-nearest-even rounding mode.
__half	put(long val) Type cast from \p long \p long assignment operator, using default round-to-nearest-even rounding mode.
__half	put(short val) Type cast from \p short assignment operator, using default round-to-nearest-even rounding mode.

Methods inherited from class org.bytedeco.javacpp.Pointer

address, asBuffer, asByteBuffer, availablePhysicalBytes, calloc, capacity, capacity, close, deallocate, deallocate, deallocateReferences, deallocator, deallocator, equals, fill, formatBytes, free, getDirectBufferAddress, getPointer, getPointer, getPointer, getPointer, hashCode, interruptDeallocatorThread, isNull, isNull, limit, limit, malloc, maxBytes, maxPhysicalBytes, memchr, memcmp, memcpy, memmove, memset, offsetAddress, offsetof, offsetof, parseBytes, physicalBytes, physicalBytesInaccurate, position, position, put, realloc, referenceCount, releaseReference, retainReference, setNull, sizeof, sizeof, toString, totalBytes, totalCount, totalPhysicalBytes, withDeallocator, zero

Methods inherited from class java.lang.Object

clone, finalize, getClass, notify, notifyAll, wait, wait, wait

Constructor Detail

__half

public __half(Pointer p)

Pointer cast constructor. Invokes Pointer(Pointer).

__half

public __half()

Constructor by default.

__half

public __half(@Const @ByRef
              __half_raw hr)

Constructor from \p __half_raw.

__half

public __half(@Const @ByVal
              __nv_bfloat16 f)

Construct \p __half from \p __nv_bfloat16 input using default round-to-nearest-even rounding mode. Need to include the header file \p cuda_bf16.h

__half

public __half(float f)

Construct \p __half from \p float input using default round-to-nearest-even rounding mode.

__half

public __half(double f)

Construct \p __half from \p double input using default round-to-nearest-even rounding mode.

__half

public __half(short val)

Construct \p __half from \p short integer input using default round-to-nearest-even rounding mode.

__half

public __half(int val)

Construct \p __half from \p int input using default round-to-nearest-even rounding mode.

__half

public __half(long val)

Construct \p __half from \p long input using default round-to-nearest-even rounding mode.

Method Detail

put

@ByRef
 @Name(value="operator =")
public __half put(@Const @ByRef
                                                    __half_raw hr)

Assignment operator from \p __half_raw.

as__half_raw

@ByVal
 @Name(value="operator __half_raw")
public __half_raw as__half_raw()

Type cast to \p __half_raw operator.

asFloat

@Name(value="operator float")
public float asFloat()

Type cast to \p float operator.

put

@ByRef
 @Name(value="operator =")
public __half put(float f)

Type cast to \p __half assignment operator from \p float input using default round-to-nearest-even rounding mode.

put

@ByRef
 @Name(value="operator =")
public __half put(double f)

Type cast to \p __half assignment operator from \p double input using default round-to-nearest-even rounding mode.

asByte

@Name(value="operator signed char")
public byte asByte()

Conversion operator to \p signed \p char data type. Using round-toward-zero rounding mode. See __half2char_rz(__half) for further details

asShort

@Name(value="operator short")
public short asShort()

Conversion operator to \p short data type. Using round-toward-zero rounding mode. See __half2short_rz(__half) for further details

asInt

@Name(value="operator int")
public int asInt()

Conversion operator to \p int data type. Using round-toward-zero rounding mode. See __half2int_rz(__half) for further details

asLong

@Name(value="operator long")
public long asLong()

Conversion operator to \p long data type. Using round-toward-zero rounding mode.

put

@ByRef
 @Name(value="operator =")
public __half put(short val)

Type cast from \p short assignment operator, using default round-to-nearest-even rounding mode.

put

@ByRef
 @Name(value="operator =")
public __half put(int val)

Type cast from \p int assignment operator, using default round-to-nearest-even rounding mode.

put

@ByRef
 @Name(value="operator =")
public __half put(long val)

Type cast from \p long \p long assignment operator, using default round-to-nearest-even rounding mode.

asBoolean

@Cast(value="bool")
 @Name(value="operator bool")
public boolean asBoolean()

Conversion operator to \p bool data type. +0 and -0 inputs convert to \p false. Non-zero inputs convert to \p true.