FloatVectorOperations Class Reference

A collection of simple vector operations on arrays of floating point numbers, accelerated with SIMD instructions where possible and providing all methods from FloatVectorOperationsBase. More...

Inherits detail::NameForwarder< FloatVectorOperationsBase< float, int >, FloatVectorOperationsBase< float, size_t >, FloatVectorOperationsBase< double, int >, FloatVectorOperationsBase< double, size_t > >.

Static Public Member Functions
static void JUCE_CALLTYPE	convertFixedToFloat (float *dest, const int *src, float multiplier, int num) noexcept
static void JUCE_CALLTYPE	convertFixedToFloat (float *dest, const int *src, float multiplier, size_t num) noexcept
static void JUCE_CALLTYPE	enableFlushToZeroMode (bool shouldEnable) noexcept
	This method enables or disables the SSE/NEON flush-to-zero mode.
static void JUCE_CALLTYPE	disableDenormalisedNumberSupport (bool shouldDisable=true) noexcept
	On Intel CPUs, this method enables the SSE flush-to-zero and denormalised-are-zero modes.
static bool JUCE_CALLTYPE	areDenormalsDisabled () noexcept
	This method returns true if denormals are currently disabled.

Detailed Description

A collection of simple vector operations on arrays of floating point numbers, accelerated with SIMD instructions where possible and providing all methods from FloatVectorOperationsBase.

See also

FloatVectorOperationsBase

Member Function Documentation

convertFixedToFloat() [1/2]

static void JUCE_CALLTYPE FloatVectorOperations::convertFixedToFloat ( float * dest, const int * src, float multiplier, int num )

staticnoexcept

References JUCE_CALLTYPE.

convertFixedToFloat() [2/2]

static void JUCE_CALLTYPE FloatVectorOperations::convertFixedToFloat ( float * dest, const int * src, float multiplier, size_t num )

staticnoexcept

References JUCE_CALLTYPE.

enableFlushToZeroMode()

static void JUCE_CALLTYPE FloatVectorOperations::enableFlushToZeroMode ( bool shouldEnable )

staticnoexcept

This method enables or disables the SSE/NEON flush-to-zero mode.

References JUCE_CALLTYPE.

disableDenormalisedNumberSupport()

static void JUCE_CALLTYPE FloatVectorOperations::disableDenormalisedNumberSupport ( bool shouldDisable = true )

staticnoexcept

On Intel CPUs, this method enables the SSE flush-to-zero and denormalised-are-zero modes.

This effectively sets the DAZ and FZ bits of the MXCSR register. On arm CPUs this will enable flush to zero mode. It's a convenient thing to call before audio processing code where you really want to avoid denormalisation performance hits.

References JUCE_CALLTYPE.

areDenormalsDisabled()

static bool JUCE_CALLTYPE FloatVectorOperations::areDenormalsDisabled ( )

staticnoexcept

This method returns true if denormals are currently disabled.

References areDenormalsDisabled(), and JUCE_CALLTYPE.

Referenced by areDenormalsDisabled().