[ARMedslack] mfloat-abi
Christophe Lyon
christophe.lyon at gmail.com
Thu Jan 27 21:45:54 UTC 2011
2011/1/27 Thorsten Mühlfelder <thenktor at gmx.de>:
> Hi,
>
> I did not really understand what's the difference between -mfloat-abi=soft
> and -mfloat-abi=softfp? Does "soft" always use software floating point
> and "softfp" could use hardware floating point if a correct -mfpu is given?
Exactly.
- soft: only software emulation
- softfp + mfpu=xxx: enables the compiler to use FP
instructions/registers inside functions, but not as function
parameters. This way, code remains compatible with 'soft' libraries
- hard: it's a different ABI, where parameters can be passed in FP
registers, avoiding expensive moves between core and FP registers. But
such code is not compatible with 'soft' libraries.
Christophe
More information about the ARMedslack
mailing list