Crate simple_soft_float

Soft-float library that intends to be a straightforward reference implementation of IEEE 754

Structs

DynamicFloat

Float with attached FPState and dynamically settable FloatProperties

F16Traits

FloatTraits where Bits = u16 and properties returns FloatProperties::STANDARD_16

F16WithPlatformPropertiesTraits

FloatTraits where Bits = u16 and properties returns FloatProperties::standard_16_with_platform_properties(self.0)

F32Traits

FloatTraits where Bits = u32 and properties returns FloatProperties::STANDARD_32

F32WithPlatformPropertiesTraits

FloatTraits where Bits = u32 and properties returns FloatProperties::standard_32_with_platform_properties(self.0)

F64Traits

FloatTraits where Bits = u64 and properties returns FloatProperties::STANDARD_64

F64WithPlatformPropertiesTraits

FloatTraits where Bits = u64 and properties returns FloatProperties::standard_64_with_platform_properties(self.0)

F128Traits

FloatTraits where Bits = u128 and properties returns FloatProperties::STANDARD_128

F128WithPlatformPropertiesTraits

FloatTraits where Bits = u128 and properties returns FloatProperties::standard_128_with_platform_properties(self.0)

FPState

The dynamic state of a floating-point implementation

FPStateMergeFailed

FPState merging failed due to incompatibility

Float

the floating-point type with the specified FloatTraits

FloatProperties

properties of a particular floating-point format

FloatPropertiesIncompatible

FloatProperties values incompatible: must be equal

PlatformProperties

properties of a floating-point implementation

StatusFlags

IEEE 754 status flags

Enums

BinaryNaNPropagationMode

Select how NaN payloads should be propagated

BinaryNaNPropagationResults

results of NaN propagation for binary operation

ExceptionHandlingMode

Select if the underflow exception should be signaled when the result is exact.

FMAInfZeroQNaNResult

select the result of fused Infinity * 0.0 + QNaN and 0.0 * Infinity + QNaN

FloatClass

float classification

FloatToFloatConversionNaNPropagationMode

select how NaN payloads are propagated in float -> float conversions

QuietNaNFormat

the format for quiet NaN values

RoundingMode

floating-point rounding mode

Sign

sign of floating-point number

TernaryNaNPropagationMode

Select how NaN payloads should be propagated

TernaryNaNPropagationResults

results of NaN propagation for ternary operation

TininessDetectionMode

IEEE 754 tininess detection mode

UnaryNaNPropagationMode

Select how NaN payloads should be propagated

UnaryNaNPropagationResults

results of NaN propagation for unary operation

UpOrDown

select Up or Down

Traits

FloatBitsType

type of underlying bits used to implement Float

FloatTraits

FloatProperties values along with the type used to represent bits for a floating-point format

Type Aliases

F16

standard 16-bit float

F32

standard 32-bit float

F64

standard 64-bit float

F16WithPlatformProperties

standard 16-bit float

F32WithPlatformProperties

standard 32-bit float

F64WithPlatformProperties

standard 64-bit float

F128

standard 128-bit float

F128WithPlatformProperties

standard 128-bit float