armfpu [ vfpv2 | softvfp ]
Use the armfpu
statement to specify how the compiler is to
deal with floating point instructions.
There are two ways in which floating point instructions can be handled on
an ARM processor:
-
through the use of floating point support functions.
-
directly on a hardware floating point co-processor, the Vector
Floating Point (VFP) unit.
softvfp
|
This keyword specifies that floating point instructions are handled
by floating point support functions.
On a device without a VFP unit, floating point support functions
are implemented in software and operate slowly. On a device with a VFP unit,
the floating point support functions are handled by the VFP unit, and operate
faster than when implemented in software.
Note that applications do not need to make any changes to take
advantage of VFP-enabled floating point support functions, if they are provided
in the device's ROM. Such functions are used automatically at run time.
If armfpu is not specified in a .mmp
file, then armfpu softvfp is assumed by default.
|
vfpv2
|
This keyword specifies that VFP specific code is to be generated at
compile time whenever floating point instructions are used.
On a device with a VFP unit, this option results in code that runs
faster than the floating point support functions generated by
softvfp . Note, however, that binaries produced with VFP specific
code do not work on devices without a VFP unit. Software Install
attempts to prevent their installation, but if any such binaries are installed
onto a device in some other way, for example by inclusion into the ROM at
device creation time, then a call to VFP specific code results in a KERN-EXEC 3
panic (see also Installation Guide).
|
|