IsVfpFeaturePresent (Compact 2013)
10/16/2014
This function queries the OEM adaptation layer (OAL) for information about vector floating-point (VFP) hardware.
Syntax
BOOL IsVfpFeaturePresent(
DWORD dwFeature
);
Parameters
dwFeature
Specifies the flag for the feature that you want to verify support for.The following table shows the flags for VFP processor features.
Flag
Description
PF_ARM_NEON
NEON Single Instruction Multiple Data (SIMD) hardware.
PF_ARM_NEON_HALF_PRECISION
NEON support for half-precision floating-point values.
PF_ARM_NEON_LOAD_STORE
NEON support for SIMD loads and stores.
PF_ARM_NEON_SINGLE_PRECISION
NEON support for 32-bit floating-point (float32) operations.
PF_ARM_VFP_ALL_ROUNDING_MODES
Supports all IEEE rounding modes.
PF_ARM_VFP_DENORMALS
Hardware support for denormalized number arithmetic.
PF_ARM_VFP_DIVIDE
Divide operations implemented in hardware.
PF_ARM_VFP_DOUBLE_PRECISION
VFP hardware support for 64-bit floating-point (float64) values.
PF_ARM_VFP_EXTENDED_REGISTERS
For VFP and NEON, 32x64-bit register bank.
PF_ARM_VFP_FP_EXCEPTIONS
Support for trapped exceptions.
PF_ARM_VFP_HALF_PRECISION
Support for half-precision floating-point values.
PF_ARM_VFP_HARDWARE
Any VFP hardware implementation.
PF_ARM_VFP_SHORT_VECTORS
Support for short vector operations.
PF_ARM_VFP_SINGLE_PRECISION
VFP hardware support for 32-bit floating-point (float32) values.
PF_ARM_VFP_SQUARE_ROOT
Square root implemented in hardware.
PF_ARM_VFP_SUPPORT
OS support for VFP (including OS VFP emulation; use PF_ARM_VFP_HARDWARE to query for hardware).
PF_ARM_VFP_V1
VFPv1 hardware.
PF_ARM_VFP_V2
VFPv2 hardware.
PF_ARM_VFP_V3
VFPv3 hardware.
Property Value/Return Value
Returns true if the feature is present; otherwise, returns false.
Remarks
This function is based on the ARM common VFP subarchitecture. If your platform uses a different VFP subarchitecture, you might have to implement your own function to query the OAL about VFP hardware.