Largest atomic size: floating-point

R2026b

Largest floating-point data type selected hardware can atomically load and store

Model Configuration Pane: Hardware Implementation

Description

The Largest atomic size: floating-point parameter specifies the largest floating-point data type that can be atomically loaded and stored on the selected hardware. Use this parameter, where possible, to remove unnecessary double-buffering or unnecessary semaphore protection, based on data size, in generated multirate code.

Dependencies

This parameter is enabled only if you can modify it for the selected hardware.

Settings

Float (default) | Double | None

Selecting a device by using the Device vendor and Device type parameters sets a device-specific value for this parameter.

Float

Specifies that float is the largest floating-point data type that can be atomically loaded and stored on the hardware.

Double

Specifies that double is the largest floating-point data type that can be atomically loaded and stored on the hardware.

None

Specifies that there is no applicable setting or not to use this parameter in generating multirate code.

Recommended Settings

ApplicationSetting
DebuggingNo impact
TraceabilityNo impact
EfficiencyTarget specific
Safety precaution

No recommendation for simulation without code generation.
For simulation with code generation, select your Device vendor and Device type if they are available in the drop-down list. If your Device vendor and Device type are not available, set device-specific values by using Custom Processor.

Programmatic Use

Parameter: ProdLargestAtomicFloat
Type: string | character vector
Value: "Float" | "Double" | "None"
Default: 'Float'

Version History

Introduced in R2006b

See Also

Topics